Tuesday Poster Session - Oxford Academic

Tuesday Poster Session

Poster Number: T12 Abstract #: 2840Improving our Understanding of MycoplasmaGenitalium Epidemiology: A Re-Analysis of TwoCohort Studies.

T. Smieszek, PhD1,2 and P. J. White, PhD1,2

1Imperial College, London, United Kingdom, 2Public Health England,

London, United Kingdom

INTRODUCTION: Sexually-transmitted M. genitalium has

recently gained increasing attention as a major cause of urethritis,

cervicitis, pelvic inflammatory disease, infertility, and increased HIV

risk. Unravelling the natural history of M. genitalium is crucial to

understand its transmission dynamics to inform control policy.

METHODS: We fitted a transmission model to infection data col-

lected among a cohort of female students from London (2378 base-

line and 900 follow-up samples; Oakeshott, 2010). We determined

maximum likelihood estimates for the force of infection of a low-

risk group (<2 partners last year), klow, recovery rate, c, and ratio

between the force of infection of the high- and low-risk groups. We

also determined if c from the London student data was consistent

with infection data from a cohort of Ugandan sex workers (1027

participants, five measurements; Vandepitte, 2013).

RESULTS: From the London student data, we estimated

klow¼0.0147/year, khigh/klow¼3.0, and c¼0.87/year. The 95% confi-

dence interval (CI) for c was [0.52/year, 2.23/year]. Neither the esti-

mated recovery rate (0.87/year) nor the upper CI was consistent with

the Ugandan data: unless we assume a test sensitivity of maximally 0.56

(c¼0.87/year) or 0.80 (c¼2.23/year) for the Ugandan data, which

seem unlikely, we cannot explain the Ugandan data with that recovery

rate.

CONCLUSIONS: Both cohort studies suffer from biases that poten-

tially have affected our estimates. More information (e.g. partner-

ship duration) would help to design more accurate models. The

difference in recovery rate estimates is probably partly due to the

sex workers having more frequent antibiotic treatment or perhaps

developing immunity promoting faster clearance. Additionally, for

some of the students, apparently persistent infection might have

been due to reinfection occurring from long-term partners. Further

data controlling for confounders are needed.

Poster Number: T74 Abstract #: 3231Social Relationships and Mental Health amongthe Victims of The Great East Japan Earthquake;A Multilevel Longitudinal Study.

Y. Matsuyama, J. Aida, PhD, A. Hase, Y. Sato, T. Tsuboya, PhD,K. Ito, MS, S. Koyama and K. Osaka, PhD

Tohoku University Graduate School of Dentistry, Sendai, Japan

INTRODUCTION: Three years have passed since The Great East

Japan Earthquake (2011). Over 43,000 victims still live in tempo-

rary housings, and they tend to have psychological distress. There

are a few studies focusing on the association between social support

and long-term mental health after disaster. We analyzed large-scale

longitudinal data to investigate the association between individual

social support, social participation and long-term mental health

among adults living in temporary housings.

METHODS: We used two-wave panel data from the Health Survey

of Residents in Temporary Housing in Miyagi prefecture, which

was conducted in 2012 and 2013. Environmental variables of tem-

porary housing community at baseline were also used. More than

12 point of K6 score was defined as having psychological distress.

Using three-level multilevel logistic regression (level 1; individual,

level 2; household, and level 3; temporary housing community), we

analyzed the association between individual social support, social

participation, and psychological distress in 2013 with adjusting indi-

vidual demographic and local environmental factors including com-

munity level social support and social participation.

RESULTS: Baseline survey included 21,450 respondents (response

rate¼58.6%), and 10,850 of them also responded in 2013 (follow-up

rate¼50.6%). We used 6,310 data without missing value. At base-

line, 581 respondents had distress while 5,729 respondents didn’t

have distress. Among these 5729 respondents, 260 newly had distress

in follow-up. After adjusted for covariates, factors associated with

higher risk of onset of psychological distress were no social support

(odds ratio (OR)¼2.27, 95% confidence interval (CI); 1.67–3.09),

lack of social participation (OR¼2.54, 95% CI 1.45–4.46). On the

other hand, among the 581 respondents with distress at baseline, 247

still had distress. After adjusted covariates, no social support signifi-

cantly associated with distress (OR¼1.91, 95% CI 1.25–2.93).

CONCLUSIONS: This study suggests social support and social par-

ticipation may protect long-term mental health among victims in

temporary housing.

Poster Number: T92 Abstract #: 3774Changes in Quality Indicators Relating to

Organization of Primary Care Services between

2007 and 2010 in Sao Paulo State, Brazil.

T. F. T. Zarili, MS1, E. R. L. Castanheira, PhD1, A. Dias, DrPH2, S.

Alves Rocha, MS3, L. O. Nunes4 and M. I. B. Nemes, PhD5

1Faculdade de Medicina de Botucatu UNESP Univ Estadual Paulista,

Botucatu, Brazil, 2Botucatu Medical School, Botucatu, Brazil, 3Faculdade

de medicina de Botucatu—UNESP, Botucatu, Brazil, 4Faculdade de

Medicina de Botucatu/UNESP, Botucatu, Brazil, 5Faculdade de Medicina

da Universidade de Sao Paulo FMUSP, Sao Paulo, Brazil

i196 International Journal of Epidemiology, 2015, Vol. 44, Supplement 1

Dow

nloaded from https://academ

ic.oup.com/ije/article/44/suppl_1/dyv277/2573981 by guest on 12 July 2022

INTRODUCTION: The expansion of Brazilian Public Primary

Health Care (BPPHC) must to be preceded by continuous quality

improvement. Nowadays, several policies and training of health

workforce are undertaken to ensure universal access, equity and

comprehensiveness in the health system. The self-assessment tool

QualiAB (‘Evaluation of the Quality of Primary Care Services’)

addresses characteristics of care organization and local manage-

ment. The managers of 350 Primary Care Units and their teams

filled the QualiAB in 2007 and 2010 from 85 cities of the Sao Paulo

State – Brazil. To determine the trends of the quality indicators

used, we compared 2007 outcomes with 2010.

METHODS: We used 60 indicators, which were categorized into

three grades: 0 (poor), 1 (acceptable) or 2 (standard expected). We

used the chi-square test. Statistical significance considered when

P�0.05 with a confidence interval of 95%. Data analysis was car-

ried out with SPSS (version 21.0).

RESULTS: 34 indicators showed significant changes (% of units

with progression to the expected standard compared 2007). There

are improve on 10 indicators of the managerial dimension, espe-

cially those related to the coordination of work and access to basic

inputs, as tests and drugs. On the assistance dimension, 19 indica-

tors showed positive progression especially provision of nursing and

oral health appointments. Had poor performance: 1 indicator man-

agement and 7 assistance.

CONCLUSIONS: The results indicate improved quality of BPPHC,

while identifying weaknesses to overcome. However the number of

indicators in standard expected may indicate a need ever more

demanding of rigor in the assessment, from the conception of quality

as incremental attribute.

Poster Number: T148 Abstract #: 3162Association of Smoking with IntraocularPressure in Middle-Aged and Older JapaneseResidents.

M. Yoshida1, M. Ishikawa1,2, S. Take1,3, A. Kokaze1,4, M.

Harada1,5 and K. Karita1

1Kyorin University School of Medicine, Tokyo, Japan, 2Genkiplaza

Medical Center for Health Care, Tokyo, Japan, 3Saitama Kaisei Hospital,

Saitama, Japan, 4Showa University School of Medicine, Tokyo, Japan,5Teikyo Junior College, Tokyo, Japan

INTRODUCTION: There have been no studies investigating the

interrelationship among intraocular pressure (IOP), blood viscosity,

and smoking. Therefore, we attempted to clarify whether smoking

was associated with elevated IOP, and to evaluate the interrelation-

ship among IOP, blood viscosity, and smoking. The results of this

study might contribute to preventing the progression of IOP-related

optic nerve damage and visual-field defects in glaucoma patients.

METHODS: This cross-sectional study analyzed health examination

data obtained between 2001 and 2004 from 1113 individuals (829

men and 284 women), ranging in age from 28 to 79 years, who had

not undergone any ocular surgery or medical treatment for hyper-

tension, ocular hypertension, and glaucoma

RESULTS: Multiple-regression analysis showed that systolic blood

pressure (SBP), diastolic blood pressure (DBP), body mass index

(BMI), and number of cigarette smoking per day had a significantly

positive association with IOP in men (P < 0.05). In women also,

SBP, DBP, BMI was positively related to IOP (P < 0.05). On the

contrary, age had a significant negative association with IOP in both

sexes (P < 0.01). Analysis of covariance and multiple logistic regres-

sion analyses showed that the adjusted mean IOP and the multivari-

ate odds ratios for IOP increased with increasing cigarette

consumption in men (P for trend¼0.01 and 0.06, respectively).

Analysis of covariance found that smoking was significantly associ-

ated with both high IOP and high hematocrit in men (P for trend

< 0.05); however, the adjusted mean IOP values were higher in

smokers than in nonsmokers, regardless of the hematocrit level.

CONCLUSIONS: The results of this study suggested that the IOP

level may be substantially affected by smoking habit in middle-aged

and older Japanese men.

Poster Number: T19 Abstract #: 3038An Insight into Factors Responsible for AlcoholDependency among People with AlcoholDependent Syndrome (Ads) – ACross-Sectional Study.

R. N. Hiremath, MD

Central Govt, Bagalkot, India

INTRODUCTION: Alcohol consumption has been steadily increas-

ing in developing countries including India since 1980s. Recently

among various strata of society there is increasing social acceptance

of alcohol intake. In India, no reliable data about the extent of the

public health problems due to alcoholism and various socio-demo-

graphic factors responsible are available.

AIM: To study the Socio-demographic factors among people with

Alcohol Dependence Syndrome (ADS) undergoing treatment at large

tertiary Hospital of India

METHODS: A cross-sectional hospital based descriptive study was

carried out for duration of six months and all the admitted patients

with ADS were taken into the study. A total of 75 people with ADS

consented to be part of the study and were included in the study.

Data was collected by means of personal interview method with the

help of pre-tested questionnaire.

RESULTS: 30.6% of persons with ADS were of age group 35–39

years, while 29% were of 30–34 Years. 86.6% had family income

of <20,000/- while 12% had between 20,000–30,000/- and only

1.3% had family income above 30,000/-.As per education status is

considered 34.6% had primary education, 36% secondary educa-

tion, 17.3% had higher secondary and 11.9% had undergone gradu-

ation.82.6% belong to rural areas. 28% had up to three siblings,

29% had 4–5 siblings and 42% had more than six siblings.15.6%

had history of alcohol intake in their family. 30% say that they

started drinking when away from family. 57.3% said that they

started drinking at the age of 20–24 years while 25% told that they

started at 25–29 years. The main reasons for starting alcohol were

to be social (66.6%), to relieve stress (30.6%) and to improve work

(2.6%). 76% said that they usually drink outside i.e., hotels while

24% drink at home.

CONCLUSIONS: This study clearly shows the effects of family

history, age of onset, loneliness, socialism, low income, education

and effect of family size on the severity of alcoholism. Intensive edu-

cational programs for school children and campaign for general

public through most common mass-media technology would be

most effective. An proactive approach by motivating alcohol

addicted parents to join drug de-addiction center for de-addiction

will go a long way to save their children from becoming alcohol

dependent.

International Journal of Epidemiology, 2015, Vol. 44, Supplement 1 i197

Dow



Poster Number: T1 Abstract #: 3258Asymptomatic Cerebral White Matter LesionsPredict Future Cognitive Decline in JapaneseElderly.

F. Ando, PhD1, Y. Nishita, MA2, C. Tange, PhD2, R. Otsuka, PhD2

and H. Shimokata, PhD3

1Aichi Shukutoku University, Nagakute, Japan, 2National Center forGeriatrics and Gerontology, Obu, Japan, 3Nagoya University of Arts andSciences, Nissin, Japan

INTRODUCTION: Cerebral white matter lesions (WMLs) are

sometimes with cerebrovascular disease (CVD) but usually asympto-

matic, and yet not so rare in the elderly. Recently, WMLs are of con-

siderable interest in cognitive decline researches.This study aims to

clarify the influence of asymptomatic WMLs on future cognitive

decline for 10 years in Japanese elderly.

METHODS: The data were derived from the National Institute for

Longevity Sciences – Longitudinal Study of Aging (NILS-LSA). The

subjects were 597 men and women, who were the participants in the

2nd wave examination (baseline), aged 65 and over, did not have

dementia (Mini-mental State Examination, MMSE>¼24) nor cere-

brovascular diseases at that time, and attended at the sequence of

biannual examinations (the 3rd, 4th, 5th, 6th and 7th wave exami-

nations), at least one time. WMLs with no obvious CVD on brain

MRI were classified into three categories (normal/mild/moderate or

severe). The influences of WMLs at baseline and time from baseline

on cognitive dysfunction (MMSE <¼23) for maximum 10 years

were estimated by generalized linear model, adjusted for age, sex

(Model 1), plus education, smoking habit and diseases (diabetes,

hypertension, dyslipidemia and heart diseases) (Model 2). A P value

<0.05 was considered as statistically significant.

RESULTS: The average number of attendance at the follow-up

examinations was 3.40 (s.d. 1.57) and the latest average of follow-

up periods was 7.20 (s.d. 3.16) years. In the model 1, the odds ratios

of moderate or severe WMLs by normal were 2.77 (95% confidence

interval: 1.39—5.51). As for mild WMLs by normal, the odds ratio

was 1.92 (0.99—3.72), that was not significant but marginal. After

adjusting for more covariates (Model 2), the odds ratios were 2.88

(1.42—5.82) and 2.05 (1.02—4.11), respectively.

CONCLUSIONS: Asymptomatic WMLs, including initial mild

lesions, might predict future cognitive decline in the elderly.

Poster Number: T21 Abstract #: 3366Incidence of Ovarian and Endometrial Cancer byEthnicity in the Million Women Study.

T. Gathani, MD1,2, A. Balkwill, MS2, K. A. Moser, MS2, G. K.Reeves, PhD2, J. Green, MD2 and V. Beral, MD2

1Oxford University Hospitals NHS Trust, Oxford, United Kingdom,2University of Oxford, Oxford, United Kingdom

INTRODUCTION: In the UK some have reported differences in the

incidence of ovarian and endometrial cancer by ethnicity, but the

extent to which these differences may be due to known risk factors

is unknown. In a large prospective study, we describe ovarian and

endometrial cancer incidence by ethnicity, before and after adjust-

ment for known risk factors for these cancers.

METHODS: Women were recruited into the study in 1996–2001,

when information on reproductive and lifestyle factors known to

influence the risk of ovarian (C56) and endometrial (C54) cancer

was obtained. They were followed up for an average of 12 years for

cancer incidence. Cox regression models were used to calculate

adjusted relative risks (RR) for incident ovarian and endometrial

cancer among South Asians and Blacks compared to Whites.

RESULTS: Analyses included 5,853 South Asian, 4,883 Black and

1,031,869 White women resident in England at recruitment. During

follow up, incident ovarian cancer occurred in 24 South Asians, 36

Blacks and 6,521 Whites, and corresponding numbers for endome-

trial cancer were 52, 37 and 7,693. The relative risk of ovarian can-

cer was significantly lower in South Asians (RR 0.64 (0.43–0.95))

but not in Blacks (RR 1.16 (0.84–1.62)), compared to Whites after

adjustment for known risk factors. For endometrial cancer, the cor-

responding relative risks were 1.13 (0.86–1.50) for South Asians

and 0.88 (0.64–1.22) for Blacks.

CONCLUSIONS: South Asian women in England have significantly

lower incidence rates of ovarian cancer than White women, which

persists after adjustment for known risk factors for the cancer. The

incidence of endometrial cancer does not vary significantly in South

Asians, Blacks and Whites.

Poster Number: T143 Abstract #: 2844Atopy and Helicobacter Pylori (H, Pylori)infection: A Systematic Review andMeta-Analysis.

B. Taye, MPH

Addis Ababa University, College of Health science, Addis

Ababa, Ethiopia

INTRODUCTION: The extent to which H pylori infection reduced

the risk of atopy is still inconclusive.This review aimed to synthesize

evidence from studies that evaluated the association between atopy

and H pylori infection.

METHODS: A comprehensive literature search in MEDLINE/

PUBMED and, EMBASE (up to August 2013) and published in

English language was done to identify observational epidemiological

studies (cross-sectional, cohort, case–control) that evaluated the

association between atopy and H pylori infection. Study quality of

included studies was assessed by the Newcastle-Ottawa scale.

Random-effects meta-analyses were used for pooled analysis

RESULTS: Twenty-three observational studies involving 22,716

participants were identified. Of these, only 17 met our inclusion cri-

teria for quantitative analysis. Compared H pylori negative, the

overall combined odds ratios (ORs) for developing atopy were 0.82

(95% CI 0.74–0.91; P<0.01) for H pylori infected, the odds of

developing atopy was 18% lower in H pylori infected than

non infected. There was no evidence of heterogeneity in

estimated effect size between the studies (Heterogeneity chi-

squared¼12.29 (d.f.¼16) P¼0.724 and I2¼0.0%,) using random

effects model.

CONCLUSIONS: Evidence from epidemiological studies suggests

that, H pylori infection is inversely associated with the risk of devel-

oping atopy.

Poster Number: T30 Abstract #: 3073Incidence Rates of Myeloproliferative Neoplasmsin India- A Hospital Based Study.

S. Varma, MD, S. Naseem, MD, P. Malhotra, MD, J. Binota, MA,

M. U. S. Sachdeva, MD, A. Sood, MA and N. Varma, MD


Dow



Post-Graduate Institute of Medical Education & Research (PGIMER),

Chandigarh, India

INTRODUCTION: Myeloproliferative neoplasms (MPN) are clonal

disorders of hematopoietic stem cells characterized by proliferation of

one or more myeloid lineages. These include BCR-ABL positive

chronic myelogenous leukemia (CML), polycythemia vera (PV), pri-

mary myelofibrosis (PMF) and essential thrombocythemia (ET).

Published epidemiological data regarding incidence rates for MPN are

scarce for Indian population as there is no well defined cancer registry

for these neoplasms and molecular diagnosis is not available to many

clinicians. However, it is important to know their frequency in order

to assess the burden of illness for public health planning. Also with

the advent of specific targeted therapies for CML (tyrosine kinase

inhibitors) and other MPN (JAK-2 inhibitors), it is even more impor-

tant to know their incidences separately for future drug development.

METHODS: All cases diagnosed according to WHO criteria as

MPN based upon clinical features, complete blood counts, bone

marrow morphology and molecular genetic studies (for BCR-ABL

fusion gene and JAK2V617F mutation) were enrolled in the study,

from August 2012 to October 2013. Following which the incidence

rates of various MPN was determined.

RESULTS: During the study period, a total of 18,14,298 patients

attended the out-patient department of our institute, and 231

(0.0127%) were diagnosed as MPN. Of the 231 MPN patients, 207

(89.6%) were BCR-ABL positive CML. The remaining 24 (10.4%)

patients were BCR-ABL negative MPN; most common was PV diag-

nosed in 11 (4.7%), followed by PMF in 7 (3%) and ET in 6 (2.6%)

cases. Molecular analysis of BCR-ABL negative MPN revealed

JAK2V617F positivity in 9/11 (81.8%) cases of PV, 7/7 (100%)

cases of PMF and 2/6 (33.3%) cases of ET.

CONCLUSIONS: We observed that incidence of BCR-ABL positive

CML is much higher than other MPN, in contrast to reports from

the west, where PV is the most common MPN. JAK2V617F muta-

tional frequency in our BCR-ABL negative MPN was 81.8% for PV,

100% for PMF and 33.3% for ET, which is also different from that

reported in western literature as more than 95% for PV, and �60%

for ET and PMF.

Additional research with carefully designed studies and generation

of database is needed to estimate the national incidence rates of

MPN. This will provide valuable information about the burden of

illness to policy makers, healthcare insurers and pharmaceutical

manufacturers. Because of availability of targeted therapy for MPN

and their chronic disease course, it is especially important to study

their epidemiology for judicious utilization of health care resources.

Poster Number: T13 Abstract #: 3458Sociodemographic Characteristics and RiskBehaviors in Patients Recently DiagnosedPositive to HIV-1 in the Northwest Mexico-USABorder.

D. V. Salas, MPH1, N. C. Martınez, MS1, A. O. Vargas, PhD2,R. D. Menchaca, PhD2, A. R. Gonzalez, PhD2, A. M. Valles, PhD2

and R. S. Muniz, PhD1

1Universidad Autonoma de Baja California, Ensenada, Mexico,2Universidad Autonoma de Baja California, Tijuana, Mexico

INTRODUCTION: The Tijuana-San Diego border is an interna-

tional metropolitan area in the northwest of Mexico -USA border.

Recently there have been many migrants from the southeast states of

Mexico and people deported from the U.S. jails who arrived in

Tijuana and remain in poor socioeconomic conditions. We consid-

ered important to determine the sociodemographic, behavioral char-

acteristics and access to health services of patients newly diagnosed

positive for Human Immunodeficiency Virus Type 1 (HIV-1) who

live in this complex and multicultural population of Tijuana, Baja

California, Mexico.

METHODS: This is a descriptive, cross-sectional study in a non

randomized sample of 46 HIV-positive patients who had not

received antiretroviral treatment in the twelve months prior to the

blood sampling who attend the Ambulatory Center for prevention

and treatment of patients with AIDS and Sexually Transmitted

Diseases (CAPACITS) in Tijuana from January, 2013 to June 2013.

A structured survey for monitoring 2013 HIV/STD program (Centro

Nacional para la prevencion y el control del VIH/SIDA), was

applied to each patient to explore and statistically analyze the socio-

demographic and behavioral variables.

RESULTS: 46 patients infected with HIV-1 were included in the

study, mean age 35 years, 74% were man,18% had recently pay for

sex and did not use condom, 88% who recently did not pay for sex

did not use condom.60% of men who recently charged for anal sex

did not use condom, 28% of man who did not charged for anal sex

did not use condom. 15% used drugs intravenously and 10% used

several times the same syringe.! 0% reported abuse by health

personnel.

CONCLUSIONS: Although the sample is small, it appears that risk

behaviors for HIV/AIDS, poor sociodemographic conditions and

poor access to health services continue to occur in the border city of

Tijuana. We need to approach the underlying problem more

effectively.

Poster Number: T56 Abstract #: 2968Blood on the Ice: the Need for Culturally InclusiveOne Health Surveillance of Anthropozoonoses inthe Arctic.

S. J. Romain, BS1, C. M. Nelson, BS2 and M. F. Davis, PhD3

1University of Toronto Scarborough, Toronto, ON, Canada, 2University of

Alaska Anchorage, Odense, Denmark, 3Johns Hopkins University,

Baltimore, MD

INTRODUCTION: A zoonotic disease research focus on tropical

and temperate climates often overlooks the arctic regions that also

host diverse animal-borne pathogens. Indigenous populations in the

arctic have close connections with both the land and animals which

can put them at risk. Inherent in spiritual, cultural, social, and sub-

sistence activities, time on the land is essential to definitions of

health and wellness, a connectedness so elemental that it has been

recognized in the UN Declaration on the Rights of Indigenous

Peoples(2008).

Subsistence hunting, fishing, herding, and butchering of animals

takes place in conditions that are suboptimal for the prevention of

zoonotic infection. Given the Arctic’s small, remote populations and

often substandard medical care, cases of infection can be overlooked

as a consequence (Indigenous One Health in the Arctic: A systematic

Literature Review of Circumpolar Zoonoses, Nelson et al., 2014).

METHODS: The One Health Initiative seeks to build multidiscipli-

nary collaborations for the purposes of controlling zoonotic diseases


Dow



that include veterinary and medical professionals. By first examining

the Initiative through a medical anthropology framework, then sub-

sequently discussing and considering traditional indigenous knowl-

edge sources on animal behavior and human health, an inclusive

model that would respect and incorporate elements of both models

is developed.

RESULTS: The analysis of the One Health Initiative model shows

areas for improvement in cultural sensitivity. By utilizing both the

One Health Initiative and the traditional indigenous methods, a

more successful model can be implemented.

CONCLUSIONS: The One Health model of integrated care encom-

passing both human and animal health has real value in these

unusual circumstances and would best be implemented in arctic

communities with the inclusion of indigenous knowledge sources.

New studies would merit consideration to address these disparities

in indigenous health care surveillance, diagnosis, and treatment in

the arctic.

Poster Number: T170 Abstract #: 3235Influence of Healthcare Facility Utilized onPulmonary Tuberculosis Treatment Delay inSouth West Nigeria.

M. A. Alagi, MD, E. Owoaje, MD, A. O. Adebiyi, MD and O.Uchendu, MPH

University College Hospital (UCH), Ibadan, Nigeria

INTRODUCTION: Delayed tuberculosis (TB) therapy may increase

the risk of disease progression, inappropriate treatment, develop-

ment of multi-drug resistant tuberculosis (MDR-TB) and poor prog-

nosis. Delayed DOTS treatment may be due to healthcare facilities

utilized by patients. Nigeria has one of the highest national burdens

of pulmonary tuberculosis (PTB) in Africa but there is limited infor-

mation on treatment delays and the associated factors among TB

patients. The objectives of the study were to estimate treatment

delay among TB patients and the influence of healthcare facilities

(HCF) utilized.

METHODS: This was a cross-sectional study of new TB patients

within intensive treatment phase at DOTS centres in Oyo State

south west Nigeria. Multistage sampling technique was used to sam-

ple the 403 estimated respondents from urban and rural DOTS

centres. Treatment delay was defined as time from onset of symp-

toms to commencement of DOTS therapy. Kaplan-Meier survival

analysis was used to compare treatment delay, while Cox regression

analysis was used to assess influence of HCF utilized on DOTS treat-

ment delay.

RESULTS: In all, 403 respondents were studied 204 urban and 199

rural. Mean age of respondents was 37.8 þ 12.3 years. The median

number of HCF utilized before accessing DOTS centres was 2 (range

1–6). The Kaplan-Meier estimation of median treatment delay was

45 days (range 2–363 days) overall: rural 40 days (range 6–344

days) versus urban 49 days (range 2–363 days); P¼0.001.

However, median patient time to first HCF was four days. The Cox

regression model showed significant risk in treatment delay associ-

ated with utilization of patent medicine vendors, traditional

medicine, private hospitals and government secondary healthcare

facilities but, utilization of PHCs reduced the risk of delay

treatment.

CONCLUSIONS: Tuberculosis patients seek care early but, pro-

longed treatment delay could be attributed to influence of non-

DOTS HCF.

Poster Number: T139 Abstract #: 3434Socioeconomic Factors, Chronic Diseases andLifestyle Habits Associated with CardiovascularDisease in Southern Brazil: A Population-BasedStudy with the Elderly.

I. J. Schneider, PhD1, M. W. Giehl, MS2, D. L. Antes, MS2,

L. P. Marques2 and E. d’Orsi, PhD3

1Federal University of Santa Catarina, Florianopolis, Brazil, 2UFSC,

Florianopolis, Brazil, 3Universidade Federal de Santa Catarina,

Florianopolis, Brazil

INTRODUCTION: Chronic noncommunicable diseases are respon-

sible for 70% of mortality in Brazil. The aim of this study was to

investigate the prevalence and socioeconomic factors, chronic dis-

eases and lifestyle habits associated cardiovascular disease in the eld-

erly living in Florianopolis (SC), South Brazil.

METHODS: Data were obtained from a population survey with

1705 people over 60 years, in the years 2009–10. It was considered

as cardiovascular disease when the elderly reported to have heart

disease, heart attack, stroke or cardiovascular disease. The inde-

pendent variables were gender, age, marital status, ethnicity, educa-

tion, per capita income in tertiles, private health insurance,

rheumatologic disease, diabetes, pulmonary disease, hypertension,

alcohol consumption, smoking, physical activity, body mass index

for elderly, waist circumference, intake of fruits and vegetables rec-

ommended by WHO. Descriptive, bivariate, and crude and adjusted

Poisson regression was performed to identify independent factors.

RESULTS: Among the elderly respondents, 63.9% were female,

50.1% aged between 60 and 69 years, 58.2% were married, 85.5%

were white. The prevalence of cardiovascular disease was 32.4%.

The independent factors were being male, having more than 70

years, have rheumatologic disease, diabetes, pulmonary disease,

hypertension and to be insufficiently active.

CONCLUSIONS: In this population, the economic characteristics

and lifestyle habits, except physical activity, were not associated

with cardiovascular disease, however all chronic diseases studied

increased prevalence of cardiovascular disease. The lack of associa-

tion between economic factors and some lifestyle habits could be

due to improved access to health and public policies to reduce these

diseases.

Poster Number: T201 Abstract #: 3298International Comparative Research of OralHealth Inequality between Japan and England.

K. Ito, MS1, J. Aida, PhD1, N. Cable, PhD2, T. Yamamoto, PhD3,

K. Suzuki, PhD4, K. Kondo, PhD4 and K. Osaka, PhD1

1Tohoku University Graduate School of Dentistry, Sendai, Japan,2University College London, London, United Kingdom, 3Kanagawa Dental

University Graduate School of Dentistry, Yokosuka, Japan, 4Nihon

Fukushi University, Nagoya, Japan


Dow



INTRODUCTION: Socioeconomic inequalities have been reported

to have effects on oral health. Our aim is to examine the association

between income and oral health among older adults in England and

Japan.

METHODS: We used the data from the Japan Gerontological

Evaluation Study (JAGES) and from the English Longitudinal Study

of Ageing (ELSA), both collected during 2010 and 2011.

Information obtained from the old adults aged 65 years or over was

used for this study. Quartile equivalised income was used as an indi-

cator for social gradients. Outcomes are edentulousness (zero

remaining teeth), which obtained from the reported numbers of

remaining teeth, and self-reported oral health status (SROH). Log-

binomial regression model were applied to determine the independ-

ent association between income and outcomes (SROH and edentu-

lousness) after controlling for the effects from sex, age and

educational attainment of the participants in each country.

RESULTS: Of 19,726 Japanese participants and 4,876 English par-

ticipants, 15.0% and 20.9% were edentulous and very few felt that

their SROH was poor in either country (JAGES¼5.5%;

ELSA¼3.5%). In both Japan and England, lower income was sig-

nificantly and independently associated with both higher risk of

edentulousness and poorer SROH. Income gradient was appeared to

be steeper in England than in Japan in both edentulousness (ELSA:

PR¼1.69, 95% CI¼1.38, 2.07; JAGES: PR¼1.37, 95%

CI¼1.28, 1.48) and SROH (ELSA: PR¼2.46, 95% CI¼1.27,

4.76; JAGES: PR¼1.81, 95% CI¼1.50, 2.19).

CONCLUSIONS: Oral health of old adults is likely to be deter-

mined by their income in both Japan and England and the gradient

appeared to be steeper in England than in Japan. Future interna-

tional comparative studies should examine social gradient of dental

care access and preventive service, which would explain this country

difference in social gradients in oral health.

Poster Number: T175 Abstract #: 3307Spatial Distribution of Tuberculosis in a MidsizeBrazilian Town.

P. Camargos, PhD1, C. L. Oliveira, PhD2 andM. de Abreue Silva2

1Federal University of Sao Joao del-Rei, Belo Horizonte, Brazil, 2Federal

University of Sao Joao del-Rei, Divinopolis, Brazil

INTRODUCTION: The health-disease process involves several fac-

tors, including the geographical space where people live and about

which some collective actions for control and prevention were done.

It is well known that the occurrence of tuberculosis (TB) is linked to

both, the socio-demographic factors and the physical environment.

The objective of this work was to assess the spatial distribution of

TB in Divinopolis, a midsize (Southeastern) Brazil town, from 2002

to 2012.

METHODS: All new reported TB cases were included in the

SINAN, the Brazilian System for notifiable diseases. The spatial dis-

tribution of cases was done according to the home address contained

in that database. The latitudes and longitudes, in the coordinate sys-

tem UTM were identified by Google EarthTM and the ArcGIS 10.1

software. All geocoded cases were located according to the Brazilian

2010 census data. We used censitary information to calculate the

crude incidence rates.

RESULTS: 326 cases were reported and we was able to map 309

(94.7%) of them. The disease occurred predominantly (99.3%) in

the urban area. There was a predominance of males (67.8%) and

individuals aged 15–39 years old (� 40%). Case distribution ranged

from zero to six individuals per census sector area and were concen-

trated in the town central area. The crude incidence rate by sector

ranged from 0.06% to 1.1%. Among the 295 existing sectors, 163

(55.2%) had at least one case of TB during the period of study, indi-

cating a wide geographical distribution of cases. 30% of the house-

hold head earned less than one Brazilian minimum wage/month

(US$280,00), demonstrating a low socioeconomic status of the

studied population.

CONCLUSIONS: The disease is spatially distributed throughout the

urban area, without a sharp cluster specific of area, but with both a

light inter-urban aggregation, and with higher number of cases in

some sectors.

Poster Number: T176 Abstract #: 2850Calling all Campy – How Routine Investigationand Molecular Characterization Impacts theUnderstanding of CampylobacteriosisEpidemiology – Alaska, 2004–13.

L. J. Castrodale, G. M. Provo, C. M. Xavier andJ. B. McLaughlin, MD

Alaska Division of Public Health, Anchorage, AK

INTRODUCTION: In Alaska, campylobacteriosis has been report-

able to public health since 1984. Campylobacter is the most fre-

quently reported bacterial enteric pathogen, averaging 90 cases

annually. Routine pulsed-field gel electrophoresis (PFGE) of

Campylobacter isolates began in 2002 at the Alaska State Public

Health Laboratory (ASPHL); however, Campylobacter isolates are

neither nationally reportable nor routinely subject to PFGE.

We summarized Alaska’s experience in universally investigating

Campylobacter isolates to demonstrate the value in more fully char-

acterizing the public health burden.

METHODS: ASPHL classified all Campylobacter PFGE data by fre-

quency of pattern matches during 2004–13. Alaska data submitted

to the National Outbreak Reporting System (NORS) and individual

case/cluster reports collected through mandatory reporting were

also summarized.

RESULTS: From 2004 to 13, 669 Campylobacter isolates were

processed at ASPHL; 45 were from environmental sources.

Excluding environmental isolates, 268 (43%) isolates comprised 23

outbreaks logged in NORS. Additionally, 82 isolates (13%) com-

prising 30 clusters (two or more isolates with a matching pattern

and a plausible epidemiologic link) prompted extra attention includ-

ing follow-up interviews, field investigations, or public facility/water

source inspections. The remaining 274 isolates (44%) were classified

as sporadic. PFGE coupled with epidemiologic data was critical in

early identification of two larger smoldering Campylobacter out-

breaks because the food sources (raw peas and raw milk) were con-

sumed for many weeks and a typical point source distribution was

not readily appreciated before PFGE data were available.

CONCLUSIONS: Over half of Alaska Campylobacter cases were

linked to an outbreak/cluster, leaving 44% classified as truly spora-

dic. Although that estimate is vastly different than the>90%

national estimate, it is not surprising because uninvestigated cases

are unlikely to be linked to other illness. Routine reporting and

PFGE of Campylobacter isolates in Alaska has contributed to more

fully documenting the infectious gastrointestinal disease burden and


Dow



prompting more public health investigations and subsequent

interventions.

Poster Number: T179 Abstract #: 2884Internet-Based Infectious Disease surVeillance:Assessing the Merits and ComparingPerformance Across Diseases.

G. J. Milinovich, PhD1, A. C. Clements, PhD1 and W. Hu, PhD2

1The University of Queensland, Brisbane, Australia, 2Queensland

University of Technology, Brisbane, Australia

INTRODUCTION: Internet-based surveillance systems provide a

novel approach to monitoring infectious diseases. Internet-based

systems are economically, logistically and epidemiologically appeal-

ing and have shown significant promise. The potential for these sys-

tems has increased with growing global internet penetration and

shifts in health-related information seeking behaviour. Whilst moni-

toring infectious diseases using internet-based data has been applied

to single or small groups of infectious diseases, no study has system-

atically assessed the suitability of this approach for a wide range of

infectious diseases of high public health importance and ranked

them according to their suitability for monitoring using this

approach. This study aims to assess correlations between a wide

spectrum of infectious diseases and Internet metrics for related

search terms and to identify diseases for which internet-based data

could be used to support early warning systems.

METHODS: Official monthly notifications for 64 infectious dis-

eases were correlated with Google Trends metrics for 164 search

terms using data from Australia. Spearman’s rank correlations were

performed on both national and state data and results used to assess

performance of search terms for estimating disease notifications.

Time series cross correlations were also performed on national data.

RESULTS: Notifications for 17 infectious diseases (26.6%) were

found to be significantly correlated with a selected search term. The

use of internet metrics as a means of surveillance has not previously

been described for 13 (76.5%) of these diseases. The majority of dis-

eases identified were vaccine-preventable, vector-borne or sexually

transmissible; cross correlations, however, indicated that vector-

borne and vaccine preventable diseases are best suited for develop-

ment of early warning systems.

CONCLUSIONS: The findings of this study suggest that internet-

based surveillance systems have broader applicability to monitoring

infectious diseases than has previously been recognised. Furthermore,

these internet-based surveillance systems have a potential role in

forecasting emerging infectious disease events, especially for vaccine-

preventable and vector-borne diseases.

Poster Number: T115 Abstract #: 2973Cardiovascular Risk Assessment Among UrbanPopulation of Bangladesh Using WHO/ISH RiskPrediction Chart.

M. S. A. M. Ahmed, MPH1, M. Moniruzzaman, MPH2, S.Chowdhury, MPH3 and P. C. Banik, MPH4

1Bangladesh University of Health Sciences (BUHS), Dhaka, Bangladesh,2Bangladesh University of Health Sciences (BUHS), Dhaka, Bangladesh,3Bangladesh University of Health Sciences, Dhaka, Bangladesh,4Bangladesh Institute of Health Sciences, Dhaka, Bangladesh

INTRODUCTION: Globally, noncommunicable diseases (NCDs)

are increasingly recognized as a major cause of morbidity and mor-

tality. It is noteworthy from the evidence that Cardiovascular dis-

ease is now a major public health problem in Bangladesh in term of

morbidity and mortality and it is logical, considering NCD the risk

profile of Bangladesh to predict that there is a large segment of adult

population at risk for being ideal victims of cardiovascular events

which is still undiagnosed. The Objective of this study is to estimate

the 10 years cardiovascular risk of the study population who has not

yet developed clinically manifest cardiovascular disease.

METHODS: A cross-sectional study was conducted among 150

respondents with age ranged from 40–70 years who have not yet

developed clinically manifest cardiovascular disease. The respond-

ents were selected randomly from urban primary health care centers

in Dhaka city.Pre tested questionnaire was used for data collection.

The WHO/ISH risk prediction chart for South-East Asian Region D

(SEAR D) was used to predict the risk of developing CVD event in

next 10 years. Data were analysed by SPSS version 16.0.

RESULTS: Regarding the risk categories of developing CVDs, 122

(81.3%) of the total respondents were found to have low (<10%)

risk of developing CVDs in next 10 years, where 23(15.3%) have

moderate (�10% to 20%) risk, 5(3.4%) have high (�20%) risk.

Females showed the higher proportion of moderate (17.2%) and

high (4.0%) risk of developing cardiovascular event in compared to

males (moderate-11.8%, high risk-2.0%).

CONCLUSIONS: About 19% of the study populations are at mod-

erate to high risk of developing CVD events in next 10 years which

is quite high in numbers and should be addressed for preventive

interventions. Moreover, this cardiovascular risk scores may offer

useful insights that can assist health service delivery planning.

Poster Number: T93 Abstract #: 2976Structure of and Fatality from Stroke in 2001–12in Semey, East Kazakhstan: A Descriptive Study.

T. Khaibullin1, A. Akimzhanova1, Y. Semenova1, R. Bikbaev1,

A. Tusupbekova1, D. Bryzhakhin1, E. Kirillova1,

A. M. Grjibovski, PhD2,3 and O. Dolmatova, MD4

1Semey State Medical University, Semey, Kazakhstan, 2Northern State

Medical University, Arkhangelsk, Russia, 3Norwegian Institute of Public

Health, Oslo, Norway, 4Kazakh Academy of Nutrition, Almaty, Kazakhstan

INTRODUCTION: Stroke is one of the most common cause of dis-

ability in Kazakhstan. Although it is known that the incidence of and

mortality from stroke in Kazakhstan is among the highest in the

European WHO region, little information on the structure and case

fatality of stroke is available in international literature. The aim of this

descriptive study is to summarize the evidence on the structure and case

fatality among the patients with stroke in Semey, East Kazakhstan.

METHODS: We analyzed case histories of all patients hospitalized

in municipal emergency hospital where all stroke patients get hospi-

talized during the period 2001–12. The population served by his

hospital increased from 296037–341429 during the study period.

We present the incidence of stroke using hospital data assuming that

all cases are hospitalized to this hospital, proportions of different

types of stroke among all cases and case fatality among hospitalized

patients. The population served by his hospital increased from

296,037–341,429 with changes in structure, therefore we calculated

standardized incidence.

RESULTS: Altogether, there were 10,294 patients during the study

period. Cerebral infarction constituted 77.8% of all cases followed


Dow



by intracerebral hemorrhage (18.0%) and subarachnoid hemorrhage

(4.2%). The population served by his hospital increased from

296037 to 341429. Standardized incidence of stroke increased from

2.1 to 3.0 per 1000. This increase was mainly attributed to the 80%

increase in cerebral infarction. The overall case fatality increased

from 15.1% in 2001 to 18.2% in 2012. This increase was observed

for all categories of stroke.

CONCLUSIONS: An increase in both incidence and case fatality of

stroke warrants urgent preventive measures and improvements in

treatment of stroke patients. We also suggest establishment of the

stroke registry with detailed information on risk factors, circumstan-

ces of stroke as well as treatment which can be a valuable tool for

further research.

Poster Number: T189 Abstract #: 3041Challenges of Malaria Diagnosis in PaediatricPatients at a Nigerian Hospital.

O. K. Ige, MPH1 and R. E. Oladokun, MPH2

1Malaria Action Program for States, Ibadan, Nigeria, 2University College

Hospital, Ibadan, Nigeria

INTRODUCTION: Errors in routine microscopy make it difficult

to assess the performance of new malaria diagnostic tests in routine

settings. This study compares the performance of routine malaria

diagnostic tests, and explores the challenges of malaria diagnosis in

paediatric patients in an endemic setting in South West Nigeria.

METHODS: The proportion of children classified as having malaria

by clinical diagnosis, Rapid Diagnostic Test (RDT) and blood smear

microscopy were compared. Test efficiency, the proportion of all

tests that gave a correct result was determined relative to routine

microscopy. Possible inaccuracies were explored

RESULTS: By clinical diagnosis 45.2% of children were diagnosed

as having malaria, 37.6% tested positive to malaria parasite on

RDT and 19.3% had positive blood smears on microscopy. Test

efficiency of clinical diagnosis when compared to RDT and routine

microscopy was 61.5% and 60.5%, respectively. RDTs had a test

efficiency of 80.7% when compared with routine microscopy. Based

on expected malaria prevalence of 50.3% in South West Nigeria,

clinical diagnosis of malaria was found to have a probability of a

false positive/negative of 0.51 and 0.52, respectively. RDTs had a

probability of a false positive result of 0.19 and a probability of a

false negative result of 0.026.

CONCLUSIONS: With an expected malaria positivity of 50.3% for

South west Nigeria, the true malaria morbidity among these paediat-

ric patients remains questionable due to the differences in the results

produced by the different diagnostic methods. The clinical implica-

tion of RDT-positive but microscopy-negative samples may be grave

if microscopy results are erroneous. Quality control systems and sur-

veillance of routine malaria diagnostics are imperative to limit mis-

diagnosis of malaria.

Poster Number: T213 Abstract #: 3083Is Measuring Social Capital Culturally- AndGroup-Specific? Psychometric Validation of theNew South Wales Social Capital QuestionnaireAcross Distinct Population Groups in Cyprus.

C. Nicolaou, BS1, P. Andreou, PhD1, E. Papastavrou, PhD1, C.Kouta, PhD1, M. Karanikola, PhD1, O. Kolokotroni, MD2, P.

Yiallouros, PhD3, E. Papathanassoglou, PhD1 andN. Middleton, PhD1

1Cyprus University of Technology, Limassol, Cyprus, 2University of

Nicosia, Nicosia, Cyprus, 3Cyprus International Institute for

Environmental and Public Health, Cyprus University of Technology,

Limassol, Cyprus

INTRODUCTION: Despite over a decade of research on social cap-

ital, there is no consensus regarding its measurement. In the fore-

ground of a continuing debate on whether an ecological or also an

individual property, most studies make opportunistic use of survey

data. While a number of purpose tools have been developed, valida-

tion of the same tool in different settings and across population

groups has been rare.

METHODS: Data collected in three studies – a sample of

Alzheimer’s’ caregivers and their age-matched neighbours (n¼225),

mothers of children with cancer and hospital controls (n¼260) and

professional nurses (n¼362), were pooled in order to assess the con-

struct validity of the Greek-version of a 36-item tool (SCQ), origi-

nally developed in Australia and subsequently used in the USA and

Greece.

RESULTS: Exploratory factor analysis yielded a similar 7–8 struc-

ture across all three Cypriot samples (48%–50% of the variance).

While the factor configuration was not dissimilar from the postu-

lated structure elsewhere, important differences were observed.

‘Participation in the local community’ was the most robust factors

across samples while, similarly to Greece, ‘social agency’ was the

most controversial. Identified as a small subset of the original items

only among mothers, this did not reflect a generalised ‘pro-activity

in a social context’ as intended by culturally variant items such as

‘picking up other people’s garbage’. Also, ‘trust’ (e.g. most people

can be trusted) and ‘safety’ (e.g. safe walking down your street after

dark) were not always captured together. Finally, ‘neighbourhood’

appears to take a different meaning (geographical construct Vs safe

environment/sense of belonging) among an elderly and a younger

population.

CONCLUSIONS: SCQ generally performs well in a different con-

text; however several items appear culturally-sensitive and not of

generic value across all population groups. Cognitive validation

studies using qualitative methods are needed for developing cross

culturally-appropriate or adapting existing tools.

Poster Number: T226 Abstract #: 3183Surgery and Risk of Venous Thromboembolismin Women with Cancer: A UK-based ProspectiveCohort Study.

S. Sweetland, PhD, A. Balkwill, MS, V. Beral, MD, T. Gathani,MD, J. Green, MD and G. K. Reeves, PhD

University of Oxford, Oxford, United Kingdom

INTRODUCTION: It is known that the incidence of venous throm-

boembolism (VTE) is increased in people with cancer and that sur-

gery is also a major risk factor. However, less is known about how

the risk varies by cancer type or how much of this increased risk is

due to cancer-related surgery. This study aims to describe the inci-

dence of VTE in relation to cancer type, taking different types of sur-

gery into account.

METHODS: 1.3 million UK women were recruited into the Million

Women Study through the NHS Breast Screening Programme in


Dow



1996–2001. These women are being followed by record linkage to

routinely collected inpatient and day-case NHS hospital admissions

data, cancer registrations and death records. The relative risks (RRs)

and 95% confidence intervals (CIs) of an inpatient/day-case hospital

admission or death record for VTE (ICD10 codes I26,I80–I82) were

estimated in relation to a variety of malignancies. Further analyses

will investigate how different surgical procedures can affect these

risks.

RESULTS: Among 1,198,848 women with no previous VTE or can-

cer and an average of 11.7 years of follow-up per woman, 108,607

(9%) had a hospital admission for incident cancer and 21,501

(1.8%) had a hospital admission or death from VTE, including 5251

women whose VTE followed cancer. Overall, women were more

than eight times more likely to have a hospital admission for VTE

after cancer than without (or before) any cancer diagnosis (adjusted

RR¼8.7, 95% CI 8.4–9.0). A preliminary analysis by cancer type

found the highest relative risks (over 60-fold) associated with can-

cers of the brain and pancreas. Results will also be presented taking

surgical procedures into account.

CONCLUSIONS: The risk of VTE is significantly increased follow-

ing a cancer diagnosis. This study has sufficient power to investigate

in detail how these risks differ by type of cancer, whilst allowing for

the risks associated with surgery.

Poster Number: T39 Abstract #: 3071Is Sexual Health a Matter of Concern among LateAdolescent Girls in Rural Areas: ACross-Sectional study in Northern India.

S. Kansal, MD and A. Kumar, PhD

Banaras Hindu University, Varanasi, India

INTRODUCTION: Adolescence is shrouded in myths & miscon-

ception about sexual health & sexuality in India, especially in rural

areas. Talking about sex is considered as taboo and consequently lit-

tle information is provided to adolescent about sexual health. In

India, about 20 per cent of all young women are thought to be sexu-

ally active by the time they are 15. Providing reproductive & sexual

health education during adolescence is an important but challenging

part of ARSH

METHODS: A descriptive, cross-sectional study using a mix

method approach was conducted among 650 adolescent girls of

15–19 years age group. Quantitative data were collected on various

components of sexual health along with socio-economic & demo-

graphic characteristics at individual (adolescent concerned) and

household level. Focus group discussions were also conducted to

substantiate the survey findings

RESULTS: Awareness about RTI/STI was also very low (4%) in

contrast to HIV/AIDS (70%) among adolescents in study areas. 23

percent were aware about different family planning methods &

about 80 percent were not aware about safe abortion. About fifty

per cent were not responded about their views on sexual relations &

to one third stated that it was irrelevant. Almost all of them were

not aware about safe sex and more than one-third considered only

rape as sexual abuse. Legal age of marriage & childbearing was not

known to more than 50 & 85 percent, respectively. Only 10 per

cent were aware about Govt. health programs specifically for

adolescents.

CONCLUSIONS: Overall sexual health awareness was very poor

among respondents. Since majority were school going, so information

on all issues to improve their sexual health, consequences of RTI &

unplanned pregnancy should be explained in school settings. Govt.

health workers at community level can also play an important role in

creating the awareness about various health programs for adolescents

Poster Number: T61 Abstract #: 3345Isolation of Leptospira spp. in Small Farm

Populations in Nine States of Brazil.

F. Silva, MS1, L. Mathias, PhD1, S. Loffer, MS2, B. Brihuega,

PhD2, L. Samartino, PhD2, C. Santos, DO3, G. Silva, MS1 and

M. Alarcon, MS1

1State University of Sao Paulo, Campus of Jaboticabal, Jaboticabal,

Brazil, 2National Institute of Agricultural Technology, Castelar, Buenos

Aires, Buenos Aires, Argentina, 3Federal University of Mato Grosso

(UFMT), Cuiaba, Cuiaba, Brazil

INTRODUCTION: Leptospirosis is a worldwide occurrence infec-

tious zoonosis of bacterial etiology whose epidemiological chain is

related to poor sanitary practices. The present study aimed to

describe the epidemiology of leptospirosis in populations of small

farms in several Brazilian states.

METHODS: Urine samples were randomly collected from wild and

domestic mammals and humans that coexist in these rural areas.

Of each urine sample, an aliquot was sown in Fletcher and

Ellinghausen—McCullough—Johnson—Harris (EMJH) media. The

samples in which there was growth of leptospires were forwarded to

the Leptospirosis Laboratory of the Institute of Pathobiology in the

National Institute of Agricultural Technology, Buenos Aires,

Argentina and were genotyped using Multiple Locus Variable num-

ber tandem repeat Analysis (MLVA). MLVA strain typing procedure

was performed using primers flanking the VNTRs: 4, 7, 9, 10, 19,

23, 31, Lb4 and Lb5 were used to discriminate strains of

L. Interrogans and L. borgpetersenii.

RESULTS: A total of 20 strains of Leptospira spp. in nine Brazilian

States were isolated and genotyped. Leptospira spp. was found in

wild mammals of the species Cavea aperea, Cebus apella, Cerdocyon

thous, Dasypus septemcinctus, Didelphis albiventris, Didelphis aurita,

Leopardus tigrinus, Myocastor coypus and Nasua Nasua. There were

also isolation findings in one horse, one sheep, a cow, a dog and two

humans. The MLVA profiles obtained were identical to the profiles of

Leptospira interrogans Icterohaemorrhagiae Copenhageni Fiocruz

L1-130, Icterohaemorrhagiae RGA and Leptospira interrogans

Canicola Canicola Hond Utrecht IV. It is noteworthy that, inside one

farm, the pathogenic Leptospira interrogans Canicola Canicola Hond

Utrecht IV strain was identified from different sources including

domestic animals, wildlife and humans.

CONCLUSIONS: These findings are very important because they

draw attention to the participation of a large number of species of

wild and domestic mammals in the transmission of pathogenic lep-

tospires to human populations.

Poster Number: T199 Abstract #: 3423Characteristics of the Educational Setting and

Self-rated Health at Age 40 in the USA.

A. K. Cohen, MPH, E. J. Ozer, PhD and B. Abrams, DrPH

University of California Berkeley School of Public Health, Berkeley, CA


Dow



INTRODUCTION: Educational is a key social determinant of

health; it is strongly associated with morbidity and mortality.

Specifically, Americans who attained more education rate their over-

all health as better. The educational experience, though, includes

not only educational attainment but also the educational setting.

Less is known about the contributions of the school setting, which

we seek to elucidate here.

METHODS: We used the USA National Longitudinal Survey of

Youth 1979 cohort, which followed a nationally representative sam-

ple of individuals from adolescence through adulthood (n¼7354).

School staff reported high school characteristics (e.g. student demo-

graphics, student-teacher ratio, drop-out rate). At age 40, partici-

pants rated their health as excellent, very good, good, fair, or poor.

We calculated odds ratios using ordered logistic regression, adjust-

ing for race/ethnicity, gender, childhood geography, and parental

socioeconomic position, and accounting for sampling design with

survey weights.

RESULTS: For each five percentage point increase in the proportion

of the high school student body classified as disadvantaged, individ-

uals had a lower ordered odds of better health at age 40 after adjust-

ing for confounders (OR¼0.97, 95% CI 0.96, 0.99). Student-

teacher ratio (OR¼0.99, 95% CI 0.99, 1.00), student racial/ethnic

composition (OR¼1.00, 95% CI 1.00, 1.00), and high school drop-

out proportion (OR¼1.00, 95% CI 1.00, 1.00) were not associated

with self-rated health.

CONCLUSIONS: In general, high school characteristics are weakly

associated with self-rated health at age 40, and only proportion of

students who were disadvantaged was statistically significant.

American education policymakers already prioritize schools with

large proportions of disadvantaged students for funding through

Title I; this research suggests that health policymakers may want to

as well. As we seek to better understand the nuances of education as

a health determinant, future researchers should examine if findings

are similar for health outcomes assessed earlier in the life course or

for other measures of the high school experience.

Poster Number: T7 Abstract #: 3432The Association between Participation inCognitive Activities and Incident FunctionalDisability in Elderly Japanese: the Ohsaki Cohort2006 Study.

K. Sugiyama, MD, Y. Kaiho, MD, K. Honkura, MD, Y. Sugawara,MS, Y. Tomata, PhD, M. Kakizaki, PhD and I. Tsuji, PhD

Tohoku University Graduate School of Medicine, Sendai, Miyagi, Japan

INTRODUCTION: Previous reports have indicated that

participation in cognitively stimulating activities is related to lower

risk of cognitive impairment among older persons. However, it has

never been investigated whether participation in cognitive

activities is associated with incident risk of functional disability.

This study examined the association between participation in

cognitive activities and incident functional disability in a Japanese

population.

METHODS: We conducted a questionnaire survey of 31,694 sub-

jects aged �65 living in Ohsaki City, northeastern Japan, in

December 2006, and 23,091 of the subjects responded. After

excluding those who were either disabled at the time of the survey

or for whom data were missing, 14,204 formed the study cohort.

For the survey, we inquired about the frequency of participation

in cognitive activities (watching television, listening to the radio,

reading newspapers, reading magazines, reading books, playing

games, or going to museums), adopted by Wilson et al (JAMA

2002). We then calculated the cognitive activity scores (CAS), higher

scores indicating more frequent cognitive activity. Records of func-

tional disability were extracted from the Long-term Care Insurance

database of Ohsaki City. We used the Cox proportional hazard

regression model for analysis, after classifying the subjects into four

groups according to CAS.

RESULTS: Baseline CAS ranged from 0.14 to 5.00 (mean; 2.68, SD;

0.80). During the 5.7-year follow-up, we identified 2,822 individu-

als who were newly certified as functionally disabled. There was an

inverse association between CAS and the risk of disability. The mul-

tiple-adjusted hazard ratios (95% confidence interval) of incident

functional disability were 0.87 (0.79–0.96), 0.75 (0.67–0.83),

and 0.63 (0.55–0.71) among subjects with CASs of 2.28–2.84,

2.85–3.41, and �3.42, respectively, as compared with those whose

CASs were <2.28 (P for trend <.0001).

CONCLUSIONS: Participation in cognitive activities is significantly

associated with a lower risk of incident functional disability.

Poster Number: T126 Abstract #: 3181The Bidirectional Association between Insomnia

and Gastroesophageal Reflux Symptoms, the

HUNT Study, Norway.

A. Lindam, MA1, E. Ness-Jensen, MD2, C. Jansson, PhD1, H.

Nordenstedt, PhD1, K. Hveem, PhD2 and J. Lagergren, PhD1,3

1Karolinska Institutet, Stockholm, Sweden, 2Norwegian University of

Science and Technology, Levanger, Norway, 3Kings’s College, London,

United Kingdom

INTRODUCTION: Gastroesophageal reflux symptoms (GERS)

and insomnia are associated, but although a bidirectional associa-

tion has been suggested, no studies have addressed this issue. Our

aim was to investigate this potential bidirectional association in a

large population-based cohort study.

METHODS: The study is based on a longitudinal cohort study

(HUNT) conducted in the county of Nord-Trøndelag, Norway,

including HUNT2 (1995–97) and HUNT3 (2006–08). All adult

county residents were invited to answer questionnaires about their

health and diseases, including insomnia and GERS. First, we studied

incidence of GERS in HUNT3 among participants without GERS in

HUNT2 but exposed to insomnia (n¼15,310). Second, we investi-

gated the incidence of insomnia in HUNT3 among participants

without insomnia in HUNT2 but with GERS (n¼18,520). Logistic

regression was used to estimate odds ratios (ORs) with 95% confi-

dence intervals (CIs) adjusted for age, sex, tobacco smoking, educa-

tion and body mass index (BMI).

RESULTS: During a mean of 11 years study period 4,566 (30%)

participants reported GERS and 501 (3%) insomnia. The adjusted

OR of incident GERS was 2.14 (95% CI 1.46–3.14) among partici-

pants with persistent insomnia (i.e. in both HUNT2 and HUNT3),

1.56 (95% CI 1.27–1.91) among participants with insomnia in

HUNT2 only, and 1.70 (95% CI 1.39–2.07) among those with

insomnia in HUNT3 only, compared to those without insomnia.

The adjusted OR of insomnia was 2.14 (95% CI 1.70–2. 70) among

participants with persistent GERS, 1.53 (95% CI 1.07–2.19) among

those with GERS in HUNT2 only, and 1.63 (95% CI 1.31–2.12)


Dow



among those with GERS in HUNT3 only, compared with those

without GERS.

CONCLUSIONS: There seems to be bidirectional associations of

similar strengths between insomnia symptoms and GERS and GERS

and insomnia. Persistent presence of the other disorder increases the

risk about 2-fold.

Poster Number: T89 Abstract #: 3395ITN Ownership and Usage among HouseholdMembers in Okposi and UmuoghanaCommunities in Ezza North LGA, Ebonyi State,Nigeria.

O. C. Nwaorgu, PhD

Nnamdi Azikiwe University, Awka, Anambra State, Nigeria

INTRODUCTION: Malaria is an important cause of illness and

death in many parts of the world especially in sub-Saharan Africa

including Nigeria. Insecticide Treated Nets (ITNs) are the most

malaria preventive measure for large scale control in highly endemic

areas. The effectiveness of ITNs in reducing prevalence of malaria

among household members following free distribution of two nets

per household in communities by State Ministry of Health and Roll

Back Malaria awareness campaign was assessed.

METHODS: A total of 14201 ITNs were deployed in the two com-

munities. Both structured questionnaire and blood examinations

were carried out among household members visited to determine the

malaria pointers. Data was analyzed using appropriate statistical

packages

RESULTS: Results from the survey showed malaria parasitaemia in

175 (24.5%) out of 715 individuals examined, highest prevalence

was among individuals 5–10 years old (33.7%). Females (52.6%)

were more infected than males (47.4%) even though not statistically

significant (P>0.05). one hundred and fifty six (156) households

(61.4%) owned ITNs but only 116 (74.4%) slept under net the pre-

vious night. Household members who slept under ITNs were less

likely to have malaria parasite in their blood (6.7% prevalence)

compared with household who did not (17.8% prevalence)

P<0.001. These households also have 6.12 chance of being infected

with malaria based on 95% confidence interval odd ratio

calculations.

CONCLUSIONS: Sleeping under ITNs as shown in the study signif-

icantly reduced the prevalence of malaria among household mem-

bers. Campaign on ITN usage should be intensified after increase in

ITN ownership in communities for reduction of malaria parasitae-

mina among community members

Poster Number: T10 Abstract #: 2962Knowledge and Attitude Towards Vaccine TrialConcepts and the Functional Status andDisability of Patients with HIV/AIDS.

A. Sha, MPH

Mother Hospital, Thrissur, India

INTRODUCTION: AIDS vaccine is seen as the ultimate prevention

tool that will complement the existing prevention strategies in place.

Patients participate in HIV vaccine trials with hope that

developing a safe and effective AIDS vaccine is possible. HIV

infected individuals live with various functional disabilities which

need lifelong continuous care. Also, the data about the functional

status of AIDS patients are rare. On this background, the study was

conducted to evaluate knowledge of vaccine trial concepts and atti-

tude of HIV infected populations to participate in HIV vaccine trials

and also assess their functional and disability status.

METHODS: A Cross-sectional study was done among 90 male and

20 female HIV positive patients attending a tertiary care hospital,

and two non-governmental organizations at Mangalore, India, from

November 2010 to May 2011after approval from the institutional

ethical committee. A pretested structured questionnaire was used to

measure the participants’ knowledge and attitudes about HIV vac-

cine trials and also their functional status and disability was graded

according to WHO criteria after obtaining written informed

consent.

RESULTS: Knowledge of concept of vaccine trial was low and mis-

conceptions about the concepts were common. 69% of study popu-

lation will be ready to be a part of a randomized vaccine trial. Only

52% were able to lead an economically productive life. The func-

tional status of the patients positively correlated with WHO disease

stage (P < 0.0001) and CD4 count.

CONCLUSIONS: Volunteers will be possibly ready to enrol in HIV

vaccine trials with hope that developing a safe and effective AIDS

vaccine is possible and more than fifty per cent of the AIDS patients

are disabled and need support and care. Functional status of patients

would be helpful in making important policy decisions and health

care interventions strengthening community-based palliative care

for AIDS patients.

Poster Number: T146 Abstract #: 3230The Community Prevalence of Early RheumatoidArthritis and Health Seeking Behavior of AffectedIndividuals.

I. Atukorala, MD, P. WIjewickrama, N. Gunawardena, MPH, K.Atukorala, D. Weerathunga and D. Dharmasena

Faculty of Medicine, University of Colombo, Colombo, Sri Lanka

INTRODUCTION: Early diagnosis of rheumatoid arthritis (RA)

and prompt treatment will minimize irreversible joint damage.

Detection of early rheumatoid arthritis (ERA) in the community

poses challenges. This study aims to establish the prevalence of ERA

in a community setting in Sri Lanka and to describe health seeking

behavior of the affected.It employed an innovative strategy of data

collection and utilizesa validated early inflammatory arthritis

screening tool (EIA-3) and 53 joint count of the Ritchie Articular

Index (RAI).

METHODS: This cross-sectional household study conducted in a

Health Division in a suburb of Colombo included persons aged

16–60 years (n¼1022) without previous rheumatologic disease

using multistage cluster sampling. A health sub-division was a clus-

ter and study was conducted in 7/12 randomly selected clusters.

Four physiotherapists trained on data collection and in joint exami-

nation screened persons and positives by either EIA-3 or RAI were

re-examined by a consultant rheumatologist and were investigated

to confirm the diagnosis of ERA based on ACR/EULAR 2010

criteria.

RESULTS: 51.8% were females. Mean age was 48.3 (6 4.3)

years.Self reported joint complaints revealed: 32.5% hand-wrist

pain, 10.3% hand-wrist swelling, 9.6% early morning stiffness>1


Dow



hour and 13.2% functional difficulty. 3.9% persons had at least one

tender joint. Of those identified by the EIA-3 and RAI, 7 were diag-

nosed to have RA based on ACR/EULAR 2010 criteria,establishing

a community prevalence of 0.7% for ERA. Amongst 346 persons

with joint symptoms, only 8.3% sought treatment with others

delaying treatment mainly due to lack of awareness of disease

(6.9%). 32.1% used over-the-counter analgesics, whilst 5.8% used

Ayurvedic/traditional treatments.

CONCLUSIONS: The community prevalence of ERA was low in

this suburban setting.Triangulation of epidemiological methods was

feasible in assessing community prevalence of this uncommon condi-

tion. Though a significant number suffered from joint symptoms,

many didn’t seek treatment due to unawareness.

Poster Number: T47 Abstract #: 3254Levels, Trends and Projection of Under-fiveMortality Rates and Prospects of Achievement ofMillennium Development Goal Four in India.

D. Sahu, PhD1, A. Pandey, PhD2 and P. De, PhD3

1National Institute of Medical Statistics, New Delhi, India, 2Indian

Council of Medical Research, New Delhi, India, 3Indian Statistical

Institute, Kolkata, India

INTRODUCTION: India has made significant strides in reducing

under-five mortality from 204 per 1000 live births in 1976 to 55 in

2011. However, India is a federation of 35 States/UTs, each of

which has different rates of decline in infant and child mortality and

hence different chances of reaching the goal. The present study

firstly provides levels and trends, then projected estimates for the

year 2015–21 of infant and under-five mortality rates of India and

its states. By doing so, it hopes to bring in a sense of urgency among

those who are not likely to achieve the MDG-4, and a sense of

renewed commitment among those who are likely to.

METHODS: Trends in infant and under-five mortality rates are esti-

mated by fitting a regression line (Hill et al. 1997) to the relation

between observations and time. A time series structure analysis is

carried out by applying the auto regressive integrated moving aver-

ages (ARIMA) model to the IMR and Under-five mortality to fore-

cast beyond the series up to year 2021.

RESULTS: Comparison of the levels and trends of components of

under-five mortality rates from 1968 to 2011 reveals that the level

of U5MR in India was about 190 per 1000 live births while IMR

was around 130 per 1000 live births during 1968–70. The U5MR

started declining in the late 1970s and until 1993 the rate of decline

was substantial. The decline, however, slowed during 1993–98.

Between 1981–86 and 1999–2005 the late neonatal, post-neonatal

and child mortality (age 1–4 years) reduced by around 50 percent.

Encouragingly, the decline in neonatal (both early and late) was sub-

stantial in Haryana, Uttaranchal, Uttar-Pradesh, Bihar, Orissa,

Karnataka, Kerala and Tamil Nadu. The percent decline in non-

southern states was higher in states with relatively higher levels of

neonatal mortality during 1981–86. Rates of decline were higher for

child and toddler than infant and within infant it was higher for

post-neonatal and early neonatal.

Projections based on Autoregressive integrated moving averages

(ARIMA) model confirms that without further intervention, India will

not be able to achieve the set target of an U5MR of less than 39 in

MDG4 goal by 2015. Only six states, namely, Kerala and Tamil

Nadu in South, Maharashtra in West, Punjab and Himachal Pradesh

in North and West Bengal in East will be able to achieve the reduction

of the U5MR below 39 by 2015. Further, the projections reveal that

the states of Assam (74), Bihar (55), Haryana (53), Madhya Pradesh

(75), Orissa (72), Rajasthan (68) and Uttar Pradesh (70) will continue

to have high U5MR of over 50. Though the projections show U5MR

of less than 50 for the states of Gujarat (49) and Karnataka (46) the

levels in these states suggest need for focused efforts to achieve MDG4

target in these states as well.

CONCLUSIONS: Except six Indian states, India will not be able to

achieve the set target of an U5MR of less than 39 in MDG4 goal by

2015. It suggest urgent attention to improve all components of

maternal and child survival.

Poster Number: T48 Abstract #: 3430The Cost Effectiveness Analysis ofElectrocardiogram Screening Program forPrevention of Sudden Cardiac Death in Japan.

H. Takahashi, PhD1, M. Yoshinaga, PhD2 and E. Ma, PhD3

1Faculty of Medicine, University of Tsukuba, Tsukuba, Japan, 2National

Hospital Organization Kagoshima Medical Center, Kagoshima, Japan,3University of Tsukuba, Tsukuba, Japan

INTRODUCTION: To prevent sudden cardiac death (SCD) of hyper-

trophic cardiomyopathy (HCM) and long QT syndrome (LQTS) in

youth is an important public health initiative. In Japan, students were

examined by electrocardiograms (ECG) at 6, 12, and 15 years to detect

high risk ones. But we have not the evidence of cost effectiveness of this

examination. The objective of this study is to clarify it in Japan.

METHOD: For the life parameters, the number of six year people

(1,061,000) was obtained by the estimated population of Japan at

Oct 1st in 2012, and the death rate of each age-stratum of each year

was obtained by vital statistics, respectively. The prevalence of

HCM was estimated 1/68,000 (six-year children), 1/20,000 (twelve-

year children) by the averaged values in medical examination

records obtained by six prefectures in Japan, and the prevalence of

LQTS was obtained by Kagoshima medical examination records.

We assumed the precision parameters of the examination (sensitiv-

ity, specificity, etc) were same as a previous work (Anderson BR,

et al. Pediatr Cardiol, 2013). For cost parameters, the examinations’

were obtained by Kagoshima medical examination records. Because

there was a report which showed the death declined 60% by the

comparison in the period between pre-examination and post-exami-

nation, we assumed it for the effectiveness for preventing death rate

by treatment. We applied Markov transition model for the analysis.

RESULTS: The life-years saved by the both examinations was

66,879 [person-years], and the cost was 23,743,000,000 [yen]

($232,000,000). Incremental cost-effectiveness ratio (ICER) was

355,020 [yen/person-years] ($3,481 [1/person-years]), which was

less than $41,400 (Anderson BR, et al.).

CONCLUSION: The cost effectiveness of Electrocardiogram

Screening Program was firstly estimated in Japan as $3,481 per per-

son-years. ($1�102 yen)

Poster Number: T105 Abstract #: 3135Long Term Effects of a Perinatal DepressionIntervention on Child Development Outcomes.

J. Maselko, PhD1, S. Sikander, PhD2, S. Bhalotra, PhD3, O.Bangash, MD2, N. Ganga, PhD4, S. Mukherjee, PhD5 and A.Rahman, PhD6


Dow



1Duke University, Durham, NC, 2Human Development ResearchFoundation, Islamabad, Pakistan, 3Univesity of Essex, Colchester, UnitedKingdom, 4Duke University, Durham, NC, 5University of Essex, Colchester,United Kingdom, 6University of Liverpool, Liverpool, United Kingdom

INTRODUCTION: Ameliorating the impact of maternal depres-

sion on child development is a significant public health priority.

Short term evidence from trials of perinatal depression treatment

suggests that treatment of depression may have beneficial effects on

child development. However, longer term studies are lacking and it

remains unknown whether the observed improvements in child

developmental outcomes immediately post intervention represent a

shift in the developmental trajectory or a temporary improvement.

We present results from a seven year follow-up of the Thinking

Healthy Programme (THP), one of the first very successful commun-

ity based perinatal depression Randomized Controlled Trial (RCT)

delivered by lay health workers to show meaningful improvements

in depression in a low resource setting (Pakistan).

METHODS: The original clustered RCT began in the third trimester

in pregnancy and lasted through 10 months post-partum in

2005–06. In 2013, the mother-child dyads in the original control

and experimental arm clusters were re-enrolled and interviewed

(LTFU¼15%). We also interviewed a random sample of children

whose mothers screened out of the original intervention because

they were not depressed. Outcomes of interest include cognitive,

socio-emotional and physical development outcomes. An intention-

to-treat random effects modelling strategy was used as the main

method of analysis.

RESULTS: Preliminary results (n¼738) indicate a lack of meaning-

ful differences in cognitive, socio-emotional or physical developmen-

tal outcomes between children born to mothers in the control and

intervention arms. Differences between children of non-depressed

mothers at the time of screening for the RCT and those of mothers

who participated in the RCT were more robust, with children of

depressed mothers having consistently worse developmental

outcomes.

CONCLUSIONS: A single intervention in the perinatal period is not

sufficient to meaningfully improve the compromised developmental

trajectory of children exposed to maternal depression.

Poster Number: T128 Abstract #: 2937Longitudinal Association Between Body MassIndex and Health-Related Quality of Life.

J. Muller-Nordhorn, MD1, R. Muckelbauer, PhD1, H. Englert,PhD2, U. Grittner, PhD3, H. Berger, PhD4, C. Prugger, MD5, K.Wegscheider, PhD6 and S. N. Willich, MD7

1Berlin School of Public Health, Berlin, Germany, 2University of AppliedSciences, Munster, Germany, 3Institute of Biometry and ClinicalEpidemiology, Berlin, Germany, 4Faculty of Applied Sciences, Osnabruck,Germany, 5Paris Cardiovascular Research Centre, Paris, France,6Department of Medical Biometry and Epidemiology, Hamburg, Germany,7Institute for Social Medicine, Epidemiology and Health Economics,Berlin, Germany

INTRODUCTION: Health-related quality of life (HRQoL) is an

important outcome in individuals with a high risk for cardiovascular

diseases. We investigated the association of HRQoL and body mass

index (BMI) as an indicator for obesity.

METHODS: We included high-risk cardiovascular primary care

patients with hypercholesterolemia and an indication for

statin therapy. HRQoL was determined with the generic Short Form

(SF)-12 health status instrument. Body weight and height were

assessed at baseline and at months 6, 12, 18, 24, 30, and 36. We

used a linear and a linear mixed-effects regression model to investi-

gate the association between BMI and SF-12 summary scores at

baseline as well as between change in BMI and SF-12 summary

scores over three years. We adjusted for age, sex, smoking status,

and in the longitudinal analysis also for the study arm.

RESULTS: Of the 7640 participants who completed the baseline

questionnaire, 6726 participants (mean age: 61 years) were ana-

lyzed. The baseline BMI was inversely associated with physical and

mental SF-12 summary scores (b [95% CI] per 1 kg/m2: �0.36

[�0.41; �0.30] and �0.05 [�0.11; �0.00], respectively). A signifi-

cant association between the change in BMI and physical SF-12

summary scores over time was only present in women (�0.18

[�0.27; �0.09]) and only in obese participants (�0.19 [�0.29;

�0.10]). A change in BMI was directly associated with mental SF-12

summary scores (0.12 [0.06; 0.19]) in the total population.

CONCLUSIONS: Increases in BMI were associated with decreases

in physical HRQoL, particularly in obese persons and in women. In

contrast, the mental HRQoL seemed to increase with increasing

BMI over time. Thus, body weight management with respect to the

HRQoL should be evaluated differentially by sex and body weight

status.

Poster Number: T197 Abstract #: 3032Contextual Barriers and Facilitators for thePrevention of Dengue in Affected HumanSettlement in Lima-Peru: A Qualitative Study.

G. M. Carmona, MS1,2, R. A. Tejada, MD1 and L.Garcia-Fernandez, MD1,3

1Instituto Nacional de Salud, Lima, Peru, 2Universidad San Martın de

Porras, Lima, Peru, 3Institute of Tropical Medicine, Antwerp, Belgium

INTRODUCTION: Dengue is a vector borne disease, with an

increasing incidence in the last years. Its transmission is related to

social determinants of health, mainly in poor urban settings. The

objective of this study was to identify contextual barriers and facili-

tators for dengue prevention in affected shantytowns, from 2010 to

2013 in Lima, Peru.

METHODS: Semi-structured interviews with residents, community

leaders and health workers were conducted in order to explore their

perceptions from a socio-ecological approach, based on the health

belief model.

RESULTS: For each of the six shantytowns, two residents, one com-

munity leader and one health professional were selected to partici-

pate in the study. Total of 24 individuals interviewed. The main

barriers identified were lack of continuous access to piped water

resulting in the necessity of storing water. Also, the inadequate

household practices favoring vector breeding conditions and expo-

sure to vectors (for example, the custom of wearing light summer

clothes that do not protect from mosquito bites) were identified

among the barriers, as well as, a scarce participation in the commun-

ity dengue control activities and the prioritization of other type of

problems. Among the facilitators mentioned were the activities

developed by the community health workers such as educational

conferences to the community, distribution of larvicide and outdoor

space spraying. The identification of conditions for vector breeding,


Dow



location of breeding places and the acknowledgement of community

participation in control activities were also stated as facilitators as

well as an adequate knowledge of the consequences of the disease.

CONCLUSIONS: The barriers and facilitators for dengue preven-

tion identified in this study represent potential action scenarios for

designing control interventions in poor developed urban settings of

Lima.

Poster Number: T107 Abstract #: 3023Coping Behaviors and Suicide in theMiddle-Aged and Older Japanese GeneralPopulation: the Japan Public HealthCenter-based Prospective Study.

T. Svensson, MD1,2, M. Inoue, PhD1,3, H. Charvat, PhD1, N.Sawada, PhD1, M. Iwasaki, PhD1, S. Sasazuki, PhD1, T.Shimazu, PhD1, T. Yamaji, PhD1, A. Ikeda, PhD1, N. Kawamura,PhD4, M. Mimura, PhD2 and S. Tsugane, PhD1

1Research Center for Cancer Prevention and Screening, National CancerCenter, Tokyo, Japan, 2Keio University School of Medicine, Tokyo, Japan,3The University of Tokyo, Tokyo, Japan, 4Gyokikai Medical Corporation,Tokyo, Japan

INTRODUCTION: Cross-sectional studies have shown an associa-

tion between different coping styles and suicidal behavior. It is

unknown whether there is any prospective association between cop-

ing behaviors and suicide in the general population.

METHODS: The study population consisted of participants of the

Japanese Public Health Center-based prospective study (JPHC

Study). In the 10-year follow-up questionnaire, subjects aged 50–79

were asked how they handle daily problems. Coping behaviors were

used to determine two coping strategies (approach coping and

avoidance coping). Out of 99,439 subjects that returned the 10-year

follow-up questionnaire, 70,213 subjects provided complete

answers on coping and were included in our analyses. Cox regres-

sion models, adjusted for confounders, were used to determine the

risk of committing suicide according to coping style. Mean follow-

up time was 8.8 years.

RESULTS: During 620,026 person-years of follow-up, 172 suicides

occurred. Age-standardized suicide rates (as per 100 000 individu-

als) of our entire study population, males and females were 26.8,

38.8 and 16.2, respectively.

The coping behaviors adopted by the highest and lowest proportion

of participants were positive reappraisal (22.9%) and avoidance

(4.0%). Planning was used mostly by men, with the remaining cop-

ing behaviors used predominantly by women. The avoidance coping

strategy was significantly more likely to be adopted by women, and

the approach coping strategy used to an equal extent by both men

and women.

Two coping behaviors were significantly associated with suicide

over time: planning (HR¼0.64, 95% CI 0.41, 0.98), and self-blame

(HR¼2.20 95% CI, 1.29, 3.76).

For the two coping strategies, the association with suicide was

inverse for approach oriented coping (HR¼0.64, 95% CI, 0.38,

1.06), and significant positive for avoidance coping (HR¼2.45,

95% CI 1.24, 4.85).

CONCLUSIONS: For the first time two coping behaviors and one

coping strategy have been shown to have a significant prospective

association with suicide in a general population.

Poster Number: T31 Abstract #: 3144The Effect of Test Kit Provision, and Individualand Family Education on the Uptake Rates ofFaecal Occult Blood Test in an Asian population:a Randomised Controlled Trial.

T. C. Ha, PhD1, S. K. Yong, MPH2, K. Kamberakis, MPH2, K. W.Yeoh, MD2, M. C. R. Yeo, MD2 and C. H. G. Koh, PhD3

1DUKE-NUS Graduate Medical School, Singapore, Singapore, 2National

Cancer Centre, Singapore, Singapore, 3National University of Singapore

(NUS), Singapore, Singapore

INTRODUCTION: Screening for colorectal cancer (CRC) using

Faecal Occult Blood Test (FOBT) has not reached sufficient levels in

Singapore. This study aims to investigate whether FOBT home-

delivery and individual education or combined with family educa-

tion increases FOBT uptake rates.

METHODS: This is a randomised-controlled intervention study of

Singaporean residents aged 50 years and above. The study was con-

ducted over 2100 household units in May 2012 till May

2013.Eligible individuals in randomly selected households were

screened and one member was randomly selected and allocated to

one of the four arms: Group A (individual and family education,

FOBT kits provided), Group B (individual education only, FOBT

kits provided) and group C (no education, FOBT kits provided) and

Group D (no education or FOBT kits provided). The primary out-

come measure was FOBT return rates which was defined as the pro-

portion of all those in each group who returned at least one

completed FOBT kit to Singapore Cancer Society (SCS) within three

months of interview.

RESULTS: Overall response rate was 74.7%. The FOBT return

rates for Group A, B, C and D were 24.5%, 25.3% 10.7% and

2.2%, respectively. Respondents who were provided education and

home-delivered FOBT kits were 15 times more likely to return

FOBT kits (Group A: OR¼15.0 [3.4–66.2]; Group B: OR¼15.5

[3.5–68.8]) and those provided with home-delivered FOBT without

education were five times more likely to return FOBT kits (Group

C: OR¼5.8 [1.2–28.3]) than those without education and FOBT

kits (Group D). There was no significant difference in return of

FOBT kits whether education was provided to subject with or with-

out a family member.

CONCLUSIONS: Home-delivery of FOBT kits increased FOBT

return rates and individual education combined with home-delivered

FOBT increased FOBT return rates even further. However, addi-

tional combination with family education did not increase FOBT

rates further.

Poster Number: T120 Abstract #: 3112Marked Gender Differences in the Prevalence ofDiabetes Mellitus among Xavante Indians fromCentral Brazil.

L. J. Franco, MD1, A. L. Dal Fabbro, MD1, D. S. Sartorelli, PhD1,A. S. Silva, MD1, L. P. Soares, MPH1, L. F. Franco, MD2, P. C.Kuhn, MD2, R. S. Moises, MD2 and J. P. B. Vieira-Filho, MD2

1Ribeirao Preto Medical School—University of Sao Paulo, Ribeirao

Preto, Brazil, 2Escola Paulista de Medicina—Federal University of Sao

Paulo, Sao Paulo, Brazil


Dow



INTRODUCTION: Xavante Indians live in the scrubland of

Central Brazil, state of Mato Grosso, and have a low degree of

admixture confirmed by genome-wide analysis. Traditionally

hunter-gatherers, and due to conflicts with newcomer farmers,

started to be settled in delimitated areas in 1957. This new condition

produced important changes in their traditional way of life. In spite

of these changes, some traditional habits are still preserved, like fes-

tivals, seclusion of adolescent boys, young age at marriage, large

number of pregnancies and prolonged breast feeding. Aim: To ana-

lyze gender differences in diabetes prevalence, by age, nutritional

status and leptin levels.

METHODS: Field work was carried out in the Sangradouro and

Sao Marcos reservations from October 2009 to January 2013. Basal

capillary glycemia and 2-hour after 75g of glucose load were meas-

ured by a portable HemoCueVR 201 glucose analyzer. Blood samples

were collected for other laboratory tests. Leptin was measured by

commercial ELISA kits. Body fat was measured by bioelectric

impedance.

RESULTS: From 948 participants (463 men and 485 women) aged

�20 years, overall age-adjusted prevalence of diabetes (WHO crite-

ria), with 95% confidence intervals was 28.2% (25.3–31.1), being

18.4% (14.9–22.2) in men and 40.6% (36.2–45.1) in women

(P<0.001). Prevalence of diabetes was higher among women in all

age-groups (20–29 yrs: 14.1 vs 4.2%; 30–39 yrs: 31.1 vs 15.2%;

40–49 yrs: 55.5 vs 32.5%; 50–59 yrs: 63.6 vs 27.3% and 60 yrs and

more: 46.4 vs 17.8%). The amount of body fat (%) was 32.6 6 6.0

in women and 22.7 6 6.0 in men (P<0.001). Leptin levels (ng/ml)

were higher in women (24.0 6 10.5 vs 7.4 6 5.9; P<0.05).

Xavante women are having less physical activity to carry out their

daily activities.

CONCLUSION: The Xavante women have less physical activity to

carry out their daily activities and still keeping high fertility and pro-

longed breast feedings, conditions of leptin resistance that are associ-

ated with higher amount of body fat and insulin resistance. As result,

women have higher rates of diabetes than men, even at younger ages.

Poster Number: T8 Abstract #: 3194The Influence of Migration on the High RiskSexual Behavior of Polish Migrants in theUnited Kingdom.

M. Ganczak, PhD1, G. Czubinska, MS2 and Z. Szych, PhD1

1Pomeranian Medical University, Szczecin, Poland, 2Polish University

Abroad, London, United Kingdom

INTRODUCTION: Objective: to assess the influence of migration

on the high risk sexual behavior of Polish migrants in the United

Kingdom after 2004.

METHODS: The cross-sectional study was conducted between

March-August 2013 of Polish migrants in the UK with the use of a

Computer-Assisted Web Interview method. A self-administered

questionnaire which contained DMIS (Bennett) and Self Esteem

(Rosenberg) scales was used as the study instrument. Significance

was tested by chi-square and McNemar tests.

RESULTS: Among 408 respondents (56.9% women), median age

32 years, 30.4% admitted to have unprotected sexual contact with a

casual partner while in the UK, a greater (P<0.001) frequency than

while in Poland (18.8%); more were engaged in unprotected sex

after the use of illegal drugs in the UK than in Poland (25.9% vs

10.7%; P<0.001). Unprotected sex with a casual partner while in

the UK was reported more often by those with a lower rather than

with high self-esteem (54.6% vs 20.5%; P<0.001), being in the eth-

nocentric stages of the sensitivity to difference on Bennett’s scale

rather than those in the ethnorelative stages (44.2% vs 28.5%;

P¼0.003), and by those who did not receive sexual education at

school (40.9% vs 20%; P¼0.01). One in three (31.7%) of those

who were engaged in unprotected sex with a casual partner had a

history of STI versus 7% in the group who did not report it

(P<0.001).

CONCLUSIONS: Polish migrants in the UK are more involved in

high risk sexual behavior than when in Poland which may result in

contracting a STI.

The process of acculturation, together with low self-esteem can exert

a marked impact on high risk sexual behavior.

The results point to the establishment of greater access to specialists

for Polish migrants in the UK to promote and maintain sexual health

in this group.

Poster Number: T221 Abstract #: 3219Coverage of the Pap Testing, Breast Ultrasoundand Mammography in Women Residents in aLow-Middle Income City in Brazil: Challengingfor Reducing Inequalities.

A. Bousquat, PhD1, K. Almeida, BA1, C. Barros, PhD2, B. Reno,MD2, M. Andrade, BA1, S. Restrepo, DO2, F. Filgueiras, MD2

and M. C. Alves, PhD3

1Universidade Catolica de Santos, Santos, Brazil, 2Universidade Catolicade Santos, Santos, Brazil, 3State Department of Health of Sao Paulo,Institute of Health, Sao Apulo, Brazil

INTRODUCTION: The cervical and breast screening tests for the

detection of cancer is recognized as essential to the improvement of

the mortality indicators of the female population. The aim of this

study was to analyze the coverage of the Pap testing breast ultra-

sound and mammography in public or private services.

METHODS: Household survey conducted with a representative

sample in two stages of 515 women aged 19–102 years in 2013 in

the city of Praia Grande, Brazil that has good coverage of primary

care services. Independent variables included sociodemographic

characteristics, health insurance and medical specialty. Dependents

variables were breast ultrasound and mammography with cut point

age in 40 years and Pap testing (cytology) with age categorized in

‘25–64’; ‘>64 and <25’ years. Pearson Qui-squared test or the

Fisher exact test was applied for differences between proportions.

The significance level was 5%.

RESULTS: The majority (73,8%) was aged between 19–59 years

and 66% of them have low middle income. 21% had some health

insurance positive associated with their income. The majority of the

women did not run neither Pap Testing (70,3%) nor mammography

(86%) in the last year, without statically differences between ages

recommended for the screening tests by Brazilian Health Ministry or

the medical specialty used (family doctors or gynecologist). Among

all women that had run the tests most of them attended the public

sector Pap Testing (65%) and mammography (63%). The breast

ultrasound was run for only 7% of the women, most of them

were conducted in the private health sector. Women with health

insurance have higher covers than the others, although the rates are

still low

CONCLUSIONS: The results show that guaranteeing the access to

a good complementary test net is essential for an appropriate


Dow



primary care, inducing health improvement and contributing to

reduce health inequalities. (Sponsor by CNPq-Brasil)

Poster Number: T38 Abstract #: 2970Maternal Antenatal Bereavement and

Hypospadias and Cryptorchidism: A Follow-up

Study of 898,961 Boys in Denmark.

K. G. Ingstrup, MS1, M. S. Jensen, PhD2, J. Olsen, PhD1, C. S.

Wu, PhD1, E. A. Nohr, PhD3, B. H. Bech, PhD1, J. Li, PhD1 and

E. Susser, DrPH4

1Aarhus University, Aarhus, Denmark, 2Perinatal Epidemiology Research

Unit, Aarhus N, Denmark, 3Odense University Hospital, Odense,

Denmark, 4Columbia University, New York, NY

INTRODUCTION: Normal descent of the testis and function of the

penis is essential to the reproductive health of man. Cryptorchidism and

hypospadias are common male genital malformations associated with

increased risk of low semen quality and testis cancer. Etiology is mostly

unknown. Endocrine disruptors have been associated with cryptorchid-

ism and hypospadias. Little is known about prenatal emotional stress

possibly affecting prenatal hormones such as cortisol. Our aim was to

investigate the associations between maternal bereavement up to one

year before and during pregnancy and the development of cryptorchid-

ism and hypospadias in the male offspring. We hypothesized bereave-

ment during first trimester to have the largest effect.

METHODS: We conducted a large population-based cohort study

(1978–2008) with data from the Danish national registries on

898,961 boys. Exposure was maternal bereavement due to death of

a close relative and outcome was cryptorchidism and hypospadias

according to the international classification system of disease. We

used cox regression taking delayed diagnosis into consideration

adjusting for parental age, birth year, family history of the outcome

and stratifying on common risk factors; birth weight, gestational age

and size for gestational age.

RESULTS: We found an increased risk for hypospadias when com-

paring bereaved mothers to non-bereaved mothers, hazard ratio (HR)

1.20 95% CI (1.00;1.44), especially mothers bereaved during the first

trimester of pregnancy HR 1.63 95% CI (1.04;2.56). No statistically

significant associations were found for cryptorchidism HR 1.01 95%

CI (0.92;1.13) or during first trimester HR 1.24 95% CI (0,94;1.64).

CONCLUSION: We had a large sample, validated outcome and

adjusted for common risk factors making our results robust.

However registration of family history of the outcome and possible

lack of registration of testis descended to a low scrotal position

should be considered. Severe emotional stress may possibly be added

to the risk factors for hypospadias, but not cryptorchidism.

Poster Number: T214 Abstract #: 3102Cross-Cultural Applicability of the Medical

Outcomes Study – Social Support Survey as a

Measure of Perceived Social Support among

Greek-Cypriot Mothers.

C. Nicolaou, BS, C. Kouta, PhD, E. Papathanassoglou, PhD and

N. Middleton, PhD

Cyprus University of Technology, Limassol, Cyprus

INTRODUCTION: Several studies link social support with psycho-

social and physical well-being; nevertheless, the way social support

is conceptualized and operationalized differs widely between stud-

ies. Some tools focus on structural (social network) or proxy meas-

ures such marital status while others on functional aspects of

support, such as the Medical Outcomes Study—Social Support

Survey (MOS-SSS) questionnaire. Furthermore, the cross-cultural

applicability of such measures has not always been established.

METHODS: A descriptive comparative study was performed with a

purposive sample of 52 mothers of children with cancer and 208

mothers of age- and gender-matched children without chronic illness

in order to assess the psychometric properties of the Greek transla-

tion of the 19-item MOS-SSS. Differences in perceived social sup-

port (overall and by type) between the study groups and by

socio-demographic characteristics was investigated in regression

models.

RESULTS: A clear structure was observed explaining 79% of the

total variance – tangible (4 items, Cronbach’s a¼0.92), emotional/

informational (8 items, a¼0.95) and positive social interaction/

affectionate support (7 items, a¼0.96). Consistently lower social

support was reported by unmarried/divorced mothers, those with

lower income, financial difficulties, not owning their house and

higher residential mobility, a further support of the known-group

validity of the tool. While no significant differences were observed

in terms of overall perceived social support between the study

groups, caregivers of children with cancer tended to report lower

emotional/informational support but higher tangible support, lend-

ing support to the tool’s scale discriminant validity. Family was

identified as the main source of support by both groups, followed by

friends (in controls) and other parents of children with cancer (in

cases).

CONCLUSIONS: The MOS-SSS appears to be a valid and reliable

measure of functional aspects of social support among Greek-speak-

ing Cypriot mothers, further supporting the cross-cultural applic-

ability of the tool and its use in cross-national epidemiological

studies.

Poster Number: T129 Abstract #: 3248Cross-Sectional Association between SedentaryTime and Body Mass Index in JapanesePopulation: the NIPPON DATA 2010.

M. Ohashi, BS1, N. Miyagawa, MS1, Y. Nakamura, PhD2, M.Nagai, PhD1, M. Yanagita, PhD3, Y. Miyamoto, MD4, N. Okuda,MD5, H. Ueshima, MD1, A. Okayama, MD6 and K. Miura, MD1

1Shiga University of Medical Science, Otsu, Japan, 2Jichi MedicalUniversity, Tochigi, Japan, 3Doshisha University, Kyotanabe, Japan,4National Cerebral and Cardiovascular Center, Suita, Japan, 5NationalInstitute of Health and Nutrition, Tokyo, Japan, 6Japan Anti-TuberculosisAssociation, Tokyo, Japan

INTRODUCTION: Recent studies show the association between

sedentary time and obesity; however, few studies have examined the

association in Japanese population where obesity is less prevalent

compared with Western populations. The purpose of this study is to

examine the association between sedentary time and body mass

index (BMI) in a representative Japanese population who partici-

pated in a national survey.

METHODS: This is a cross-sectional study of 2829 general popula-

tions (1,201 men, 1,628 women, aged 20 years and older)

(the NIPPON DATA2010), who participated in the National Health


Dow



and Nutrition Survey of Japan conducted in 2010 by the Ministry of

Health, Labour and Welfare. Participants were collected from 300

randomly selected districts from all over Japan. Total sedentary time

including TV-viewing and sitting time per day was measured by self-

report and the following interview by trained interviewers. We exam-

ined the association between sedentary time and BMI using multiple

linear regression. Since age has an interaction with the association

between sedentary time and BMI, we examined the association in par-

ticipants aged less than 50 years and 50 years or over, separately.

RESULTS: The average sedentary time was longer in men (5.8

hours/day) than in women (5.2 hours/day). Mean BMI was 23.9 in

men and 22.7 in women. Sedentary time was significantly and posi-

tively associated with BMI in women after adjustment for age, edu-

cation, total energy intake, hours of heavy or moderate physical

activity, smoking status, and drinking status. The association was

significant both in women aged less than 50 years (ß¼0.184,

P¼0.001) and in those aged 50 years or over (ß¼0.117,

P¼0.002). Meanwhile, the association between sedentary time and

BMI was not significant in younger and older men.

CONCLUSIONS: Sedentary time was independently and positively

associated with BMI in Japanese women. Reducing sedentary behav-

ior may have health benefits by lowering BMI.

Poster Number: T194 Abstract #: 3378The Physical Activity Profiles of Adult EthnicGroups in England.

P. Bhatnagar, MPH, N. Townsend, PhD and C. Foster, PhD


INTRODUCTION: We aimed to identify which types of activity

contribute to overall physical activity in ethnic groups who achieve

the UK physical activity recommendations. We investigated how

this varies according to sex and age within each ethnic group.

METHODS: The Health Survey for England is a nationally repre-

sentative, cross-sectional survey which is conducted annually. In

1999 and 2004 the samples for ethnic minority populations were

boosted, allowing for more detailed analysis on ethnic minority

physical activity. We merged the two survey years and analysed the

self-reported physical activity measures available. We analysed data

from over 7000 people aged 16 and above years who met the UK

physical activity recommendations that were published in 1999 and

2004. The proportions of total physical activity achieved through

walking, sports, housework and manual DIY were calculated. We

stratified by age group and examined associations between ethnic

group, sex and socioeconomic status.

RESULTS: In all age and ethnic groups, including the White British

population, all women had more activity through housework in

comparison to men. Among women in the 16 to 34 age-group, more

Caribbean and Chinese women were being active through sports

(43% and 50%) than the other ethnic groups. In the 35 to 54 age-

group, 41% of Bangladeshi men’s activity came from walking,

which was the highest percentage of walking for any male ethnic

group in this age category. The percentage that walking and sports

contributed to activity declined with age for both men and women,

for all ethnic groups.

CONCLUSIONS: Ethnic minority groups in the UK are active in

different ways, and this varies by age, sex and socioeconomic status.

Walking was significant for all age groups and ethnicities and is an

accessible activity. These results can be used to appropriately target

population level interventions for increasing physical activity levels

in adult ethnic minority populations.

Poster Number: T209 Abstract #: 3011Measuring Equity and Mutuality: Reflections onthe Epidemiology of Social Justice.

P. C. Mackie, BA

NHS Health Scotland, Edinburgh, United Kingdom; University ofEdinburgh, Edinburgh, United Kingdom

INTRODUCTION: Epidemiology has provided a fruitful mecha-

nism for exploring inequalities in health and their origins in forms of

social injustice. However, how valid this approach remains is open

to discussion. The evidence underpinning this debate and possible

ways forward are explored in this presentation.

METHODS: Literature review and synthesis.

RESULTS: Recent public health analyses have begun to highlight the

limitations of the epidemiological approach in relation to existing and

emerging public health challenges. Work on the ‘Glasgow Effect’

(Walsh et. al. 2010) has questioned the degree to which health

inequalities may be attributed to economic inequality and the growth

of an ecological discourse in explaining possible future health experi-

ences highlight these limitations (Lang & Rayner 2012a & 2012b).

There is also a concern that the focus on equity has resulted in a mar-

ginalisation of mutuality as an underlying requirement for socially

just outcomes in population health (Mackie P 2010).

CONCLUSIONS: The evidence suggests that current epidemiologi-

cal approaches may not be sufficient to capturing the nuances of

socially just, population health in the 21st century. What factors are

required to be part of necessary developments of epidemiological

methods and analysis must be debated as a matter of urgency.

Poster Number: T160 Abstract #: 3414Dangers of Commercial Fishing in the Arctic,Alaska 2000–12.

J. M. Lincoln, PhD, M. Davis, MPH and D. Lucas, PhD

National Institute for Occupational Safety and Health, Anchorage, AK

INTRODUCTION: Commercial fishing is the most hazardous

occupation in the United States. The National Institute for

Occupational Safety and Health (NIOSH) recently completed a

study to identify the most hazardous fisheries in Alaska and the in

the rest of the U.S. to describe the unique safety issues in each

region. The cold water and remote fishing grounds found in Alaska

make it a dangerous place to work. However, significant improve-

ments in safety have been made due to focused prevention efforts on

high risk groups of fishing vessels.

METHODS: The NIOSH Commercial Fishing Incident Database

was used to identify all work-related fatalities that occurred in

Alaska during 2000–12. Data are available on person, vessel and

environmental conditions at the time of the event.

RESULTS: During 2000–12, 158 fishermen died in Alaska. Half

(79) of the deaths were caused by drowning (71) or other events

(e.g. asphyxiation, hypothermia, and smoke inhalation) following

vessel disasters (e.g. sinking, capsizing, fire). Another 31% (49) of

fatalities were the result of falls overboard. Only one of the 49 vic-

tims who died from falls overboard was wearing a personal flotation

device. In addition there were 30 fatalities resulting from other

causes (e.g. asphyxiation, fall from height, poisoning, and struck by


Dow



gear). Fatalities occurred most frequently while fishing for salmon

(47, 30%) after the victim fell overboard or after a skiff capsizing.

CONCLUSIONS: Tailored prevention programs for specific high

risk fisheries in Alaska have resulted in safety improvements. In

addition, NIOSH has developed interventions focusing on creating

incentives for fishermen to wear personal flotation devices since

there is no regulation for them to do so. Safety training is also avail-

able and very affordable in Alaska. Progress has been made, but

there remains a need for continued safety interventions to combat

the unique work hazards found in Alaska.

Poster Number: T118 Abstract #: 3123Metabolic Syndrome (Ms) among XavanteIndians from Central Brazil.

A. L. Dal Fabbro, MD1, L. J. Franco, MD1, D. S. Sartorelli, PhD1,A. S. Silva, MD1, L. P. Soares, MPH1, L. F. Franco, MD2, P. C.Kuhn, MD2, R. S. Moises, MD2 and J. P. B. Vieira-Filho, MD2

1Ribeirao Preto Medical School—University of Sao Paulo, Ribeirao

Preto, Brazil, 2Escola Paulista de Medicina—Federal University of Sao

Paulo, Sao Paulo, Brazil

INTRODUCTION: Xavante Indians live in an area from Central

Brazil. After contact with members of the Brazilian Society, this new

condition produced important changes in their traditional way of life,

incorporating new foods such as rice, resulting in obesity (50.8%) and

diabetes mellitus (28.2%), especially among women. Aims: To analyze

the prevalence of Ms among adults aged 20 years or more according to

gender and two diagnostic criteria of Ms: IDF 2005 and ATPIII 2001.

METHODS: population-based survey was carried out among 948

Xavante. Fasting and 2-hour after 75g glucose capillary were meas-

ured by a portable glucometer (HemoCueVR Glucose201) (WHO cri-

teria) in the Sangradouro and Sao Marcos Reservations, Mato

Grosso State, Brazil. Anthropometric data and medical characteris-

tics were measured, like waist circumference. Blood pressure was

measured by an automated device (OMRON 7421INTCVR ), and

hypertension was defined according WHO criteria. Blood samples

were collected for laboratory tests of HDL-c and Triglycerides (TG).

RESULTS: according to IDF criteria the prevalence of Ms was:

64.2% in general (61.2-67.3), 52.6% (48–57.1) in men and 75.3%

(71.5-79.2) in women (P<0.05); among individuals with Ms,

86.1% had reduced HDL-c, 70.1% had raised TG, 24.1% had

hypertension and 78% had diabetes mellitus (DM) or Impaired

Glucose Tolerance (IGT). According to ATPIII criteria, prevalence

of Ms was: 45% in general (43.7-50.1), 23.9% (20.0-27.8) in men

and 69.1% (65.0-73.2) in women (P<0.05); among individuals

with Ms, 91.9% had reduced HDL-c, 70.0% had raised TG, 23.7%

had hypertension and 80.6% had DM or IGT.

CONCLUSIONS: IDF and ATPIII criteria produced different results

in the prevalence of Ms, especially in men. Ms is an important risk

factor to cardiovascular disease and high prevalence rates of Ms

were observed among Xavante Indians. Preventive and control

measures need to be implemented in this population.

Poster Number: T88 Abstract #: 3296Defining Reference Range for Blood CultureRates – an Analysis within the German HospitalInfection Surveillance System (KISS).

A. Karch1, S. Castell1, C. Geffers2, F. Schwab2, F. M.Brunkhorst3, P. Gastmeier2 and R. T. Mikolajczyk1

1Helmholtz Centre for Infection Research, Braunschweig, Germany,2Charite University Medicine Berlin, Berlin, Germany, 3Jena UniversityHospital, Jena, Germany

INTRODUCTION: Early and appropriate blood culture diagnos-

tics is associated with both, a reduction of sepsis mortality and a

reduction of duration of antibiotic treatment. To date, 100–200

blood culture sets per 1000 patient-days are recommended as the

target range. However, the empirical basis of this recommendation

is not clear. A previous study observed lower sepsis rates in institu-

tions with low blood culture rates, suggesting that some sepsis cases

remain undetected. The aim of this study was to analyze the form of

the association between blood culture rates and sepsis rates and to

derive a reference range for blood culture rates in Intensive Care

Units (ICUs).

METHODS: The present study is based on data from 223 ICUs tak-

ing part in the German hospital infection surveillance system, which

captures all cases of primary sepsis in the participating institutions.

We applied locally weighted regression in order to assess the form of

association between blood culture rates and sepsis rates.

Subsequently, we applied segmented Poisson regression to estimate

the detection threshold.

RESULTS: Studied by means of nonparametric additive models,

sepsis rates increased when the number of blood cultures was below

80–90 per 1000 patient days, but stabilized for higher blood culture

rates. A threshold of 87 (95% confidence interval 54–120) per 1000

patient days was identified. Sixty-five percent of the ICUs participat-

ing in this study showed blood culture rates below this threshold.

These ICUs were more likely to represent interdisciplinary ICUs

(P¼0.048) in smaller (P<0.001), non-academic hospitals

(P<0.001) with a remote microbiology lab (P<0.001), a high pro-

portion of short stay patients (P¼0.003) and showed lower ventila-

tion (P<0.001) and CVC rates (P¼0.006).

CONCLUSIONS: We provided empirical justification for a target

range for blood culture rates in ICUs. In a large majority of the

studied ICUs, blood culture rates were far below this threshold.

Poster Number: T37 Abstract #: 3003The Smoking Actual Situation of JapaneseJunior High School Students and the Effects ofEducation for Smoking Prevention.

T. Eto, MD1 and K. Aoishi, PhD2

1Miyazaki Prefectural Nursing University, Miyazaki, Japan, 2Miyazaki

University, Miyazaki, Japan

INTRODUCTION: We performed smoking prevention education

to Japanese junior high school students for the purpose of motivat-

ing smoking prevention and the non-smoking support to the family

who is smoking. We carried out the questionnaire about smoking

before and after the lecture to the students and judged the educa-

tional effects.

METHODS: The object number was 4,876. They were divided into

smoker group and nonsmoker group, the questionnaire items about

smoking were evaluated, and the t test was performed between two

groups and before and after education. Moreover, in order to clarify

the factor of smoking, logistic regression analysis was conducted.

RESULTS: Male students’ smoking rate was 4.5% and female’ was

1.8%. The smoker students had the interest to smoking intentionally

compared with the non-smoker (OR, 6.01; 95% CI, 4.06–8.89).

As for the non-smoker students, the numerical value changed


Dow



intentionally after education by all the questionnaire items of

‘Interest to smoking’, ‘Recognition to others’ smoking’, ‘Desire to

lead others to prohibition of smoking’, ‘The troublesome feeling to

others’ smoking’ and ‘the reason for the ability not to give up smok-

ing’. In the group of smoker students, only three items such as

‘Recognition to others’ smoking’, ‘Desire to lead others to prohibi-

tion of smoking’, ‘The troublesome feeling to others’ smoking’

changed intentionally after education. The education reduced the

percentage of risk of ‘the interest to smoking’ over smoking from

‘OR, 3.85; 95% CI 2.50–5.94’ to ‘OR, 2.88; 95% CI 1.89–4.37’.

CONCLUSIONS: Education changed the all student’s view over

smoking dramatically and suppressed the interest to smoking for

non-smoker students. However, the smoker students were difficult

to suppress their interests to smoking by only one lecture. The pro-

gram for smoker students was required for behavior modification,

and the necessity of considering the intervention which piled up the

number of times was indicated.

Poster Number: T206 Abstract #: 3333Micro-Business and Occupational Stress Process:Occupational Demands, Job Autonomy, andDepression in Korean Immigrant Micro-businessOwners and Paid Employees in Toronto, Canada.

I. H. Kim, PhD1 and S. Noh, PhD1,2

1Centre for Addiction and Mental Health, Toronto, ON, Canada,2University of Toronto, Toronto, ON, Canada

INTRODUCTION: Immigrant status is a pivotal determinant of

access to employment and job-related physical and psychological

health in North America. Due to labor market challenges, large pro-

portions of Asian immigrants turn to self-employment in low-yield-

ing service sectors. Although occupational stress has long been a

central focus of psychological research, few studies investigate how

immigrant micro-business owners (MBOs) respond to their unusu-

ally demanding occupation, or how their unresolved occupational

stress manifests in psychological distress or disorders. Based on a

Job Demands-Control (JD-C) model, this study compares MBOs to

paid employees on depressive symptoms, occupational demands,

and job autonomy.

METHODS: Data were derived from a cross-sectional survey of

1,288 Korean immigrant workers (MBOs, professionals, office

workers, and manual workers) aged 20 to 64, living in Toronto and

surrounding areas. Person to person interviews were conducted

between March and November 2013.

RESULTS: Among the four occupational groups, MBOs reported

the greatest physical, psychological, and emotional demands, as well

as job autonomy. MBOs also had a higher level of depression than

professional and office workers. While all three types of occupa-

tional demands were significantly associated with depressive symp-

toms, the influence of emotional demand was greater for MBOs

than professional employees. An inspection of autonomy-stress

interactions suggested autonomy significantly moderated against the

impact of emotional demand (P<.022) and psychological demand

(P<.006) on depressive symptoms. We also found the stress-moder-

ating impact of job autonomy was more clearly highlighted among

MBOs than paid employees.

CONCLUSIONS: One in two Korean immigrants choose self-

employment, most typically in family-owned micro-businesses,

which require long hours of physically demanding work and

emotionally taxing dealings with clients and suppliers. However, the

mental health of MBOs may be protected from the adverse effects of

job stress by the benefits of having more control and autonomy at

work compared to paid employees.

Poster Number: T191 Abstract #: 3381The Social Context and Perception of Poor Healthin Colombia: A Multilevel Analysis.

A. A. Agudelo-Suarez, PhD1,2, E. Martinez, PhD3 and J.Fernandez, MS4

1University of Alicante, Alicante, Spain, 2University of Antioquia,

Medellın, Colombia, 3University of Antioquia. National School of PublicHealth, Medellin, Colombia, 4Escuela de Salud Publica de Mexico.,Cuernavaca, Mexico

INTRODUCTION: Self perceived health of a society is built on cul-

ture and making analysis by regions and local political and adminis-

trative division in Colombia could assess differences by social

context in the perception of health

METHODS: A multilevel analysis using data from the National

Survey of Health and Quality of Life of 2009 was performed. In

relation with survey details, the perceived health was dichotomized

and individual level variables related were extracted. Information on

HR, ICV and GDP per capita (GDP pc) at the departmental level

was taken from DANE. Prior to the completion of the multivariate

model, t test were performed and chi-square, to explore the associa-

tion of each independent variable and the presence of poor perceived

health. Logistic mixed effects models with random intercept a

municipality and department level were performed. The existence of

random parameters was ruled out. Each statistical model evaluation

was performed using the Akaike criteria. Subsequently logistic

mixed effects models with random intercept were prepared by

municipality level and department, including intercepts sequentially

by town and municipality.

RESULTS: A statistically significant association between the Index

of Quality of Life at the departmental level and the presence of poor

perceived health was found (OR 0.96; 95% CI 0.94–0.97), this

effect was independent of the different variables and adjusted for

departmental GDPpc and the percentage of explained variance was

1% for department level and 4% for the municipality level.

CONCLUSIONS: There is an effect of the departmental and munici-

pal level that clearly outlined the differences in context. It confirms

the importance of the social context of these levels on the determina-

tion and social construction of perceived health and could enhance

the comprehension on inequities because of the location where peo-

ple live.

Poster Number: T196 Abstract #: 3406Minority Status as a Determinant of Tobacco-Caused Health Inequalities: A Global Perspective.

S. E. Cox1, D. C. Li1, A. Hulbert2, M. D. Teran1, C. M. Hooker1, N.Wang3, A. Memon4, J. M. Samet5, A. Alberg6 and M. Brock1

1The Johns Hopkins University School of Medicine, Baltimore, MD,2Johns Hopkins Sydney Kimmel Cancer Center, Baltimore, MD, 3Peking

University Cancer Hospital, Beijing, China, 4Brighton and Sussex MedicalSchool, Sussex, United Kingdom, 5University of Southern CaliforniaInstitute for Global Health, Los Angeles, CA, 6Medical University of SouthCarolina, Charleston, SC


Dow



INTRODUCTION: Tobacco is a leading global cause of premature

morbidity and mortality producing worldwide health disparities.

These disparities are often attributed to differences in tobacco use

due to socioeconomic inequality, but in many countries, wealthier

and more educated minorities inexplicably smoke more than the less

fortunate majority. We hypothesized that the embodiment of high

levels of perceived and/or experienced psychosocial distress as a

member of a minority group drives increased tobacco smoking, and

therefore increased risk of lung cancer.

METHODS: We collected age-adjusted lung cancer incidence rates

and smoking prevalence for minority and majority groups from 32

countries, representing 60% of the world’s population; and esti-

mated relative risk (RR) of lung cancer and smoking for minority

versus majority groups.

RESULTS: The minority versus majority RRs were in the direction

of significantly increased risk in 80% of the comparisons for lung

cancer incidence rates (P<0.0001; 90% for the largest minority

group; 37% of the RRs were �1.5), and in 85% of the comparisons

for smoking prevalence (P<0.0001; 93% for the largest minority

group; 29% of the RRs were �1.5).This association was consis-

tently observed whether minority status was defined by race/ethnic-

ity, religion, language spoken, indigenous affiliation, immigrant

status, or whether the minority population had lower or higher

socioeconomic status than the majority.

CONCLUSIONS: Minority status is a strong determinant of smok-

ing prevalence and risk of lung cancer. This is a global phenomenon

and not limited to minority status defined by race/ethnicity, but also

generalizes to minority status defined by religious, linguistic or

nativity criteria; and it is applicable regardless of the economic or

educational status of the minority group. The vulnerability of

minority groups to increased tobacco smoking warrants specifically

targeted strategies to reduce smoking in these populations.

Poster Number: T95 Abstract #: 2847Mortality Trends and Disparities in the AlaskaNative and American Indian Population Living inAlaska, 1980–2011.

R. Mohelsky, MPH1 and P. Holck, PhD2

1Alaska Native Tribal Health Consortium, Anchorage, AK, 2PeakEnterprises, Falmouth, MA

INTRODUCTION: Alaska Native/American Indian (AN/AI) peo-

ple living in Alaska have experienced differences in mortality rates

compared to Alaska non-Natives and U.S. Whites (USW).

METHODS: Using data from the Alaska Bureau of Vital Statistics,

we examined trends in mortality of AN/AI people over a 32 year

period. Four-year average annual age-adjusted mortality rates were

computed by cause, gender, and age group.

RESULTS: While the AN/AI all-cause mortality rate decreased on

average 3% per decade during 1980–2011 (P<0.01), sharper

declines were observed among non-Natives and USW. During

2008–11, AN/AI rates were 67% (RR¼1.67) higher than non-

Natives and 51% higher (RR¼1.51) than USW.

Cancer, heart disease, unintentional injury, and suicide were leading

causes of death in every time period. Declines occurred in many

cause-specific death rates including unintentional injury and heart

disease. Due to a slower rate of decrease, the AN/AI heart disease

mortality rate surpassed that of non-Natives and USW in 2008–11

despite having been lower in previous years. No significant change

occurred in AN/AI cancer mortality rates, in contrast to non-Natives

and USW. AN/AI suicide mortality rates did not change signifi-

cantly, and COPD mortality rates increased on average 2.0% per

year (P<0.05).

Mortality rates of AN/AI males were higher than females for all-

causes, heart disease, unintentional injury and suicide, in all time

periods. Female all-cause mortality rates increased an average of

0.4% per year (P<0.01) while decreasing an average 0.7% per year

(P<0.01) among males. For each age-group examined, AN/AI all-

cause mortality rates were higher than rates of non-Natives and

USW.

CONCLUSIONS: There have been significant improvements in AN/

AI all-cause mortality rates since 1980. Despite these improvements,

cancer and suicide mortality rates haven’t changed significantly, and

the COPD mortality rate has increased. Differences in rates between

AN/AI and both non-Natives and USW have increased for all-

causes, cancer, COPD, and heart disease.

Poster Number: T215 Abstract #: 3236The Social Epidemiology of Hypertension amongResidents of Ilishan-Remo, Nigeria.

V. Norman, MA

Walden University, Fontana, CA

INTRODUCTION: There is high incidence and prevalence of

hypertension and other chronic diseases in sub-Saharan Africa.

Nigeria is a sub-Saharan nation with a high prevalence of hyperten-

sion. Researchers report low knowledge of hypertension, its preven-

tion, and outcomes, and rapid urbanization with the adoption of

Western eating habits as major causes. This is an important issue

because chronic disease is placing a huge economic, social, and dis-

ease burden on Nigerian society and government. The purpose of

this mixed methods study was to investigate quantitatively lifestyle

practices and qualitatively what knowledge, attitudes, and practices

Nigerians hold on risks, and prevention of hypertension to discover

the practices used to prevent or minimize risk for chronic disease.

METHODS: Participants were recruited from Ilishan-Remo, Ogun,

Nigeria. Blood pressure, height, and weight were measured and par-

ticipants completed the WHO STEPS lifestyle survey. Some partici-

pants were invited to remain for a focus group discussion. The

general linear regression model with sums of squares and Pearson’s

product correlation were used to analyze quantitative data. By the

general linear regression model and Pearson’s product correlation

analysis it was found lifestyle can be predictive for hypertension.

Qualitative data were analyzed deductively using the PEN-3 model.

RESULTS: The greatest predictors of elevated blood pressure were

BMI (P <.024), and alcohol use (P <.007). Qualitative data

revealed the matriarchs of the family were primary sources of health

knowledge. A major impact on community health in Ilishan-Remo

has been occasional community health intervention programs spon-

sored by Babcock University (BU).

CONCLUSIONS: One of the big challenges in low and mid resource

countries is educating the population of chronic disease risks.

Local universities can assist ministries of health and local govern-

ments in providing much needed health education. Public health

campaigns should continue to encourage healthier diets and active

lifestyles.


Dow



Poster Number: T76 Abstract #: 3223The use of the International Classification ofFunctioning, Disability and Health (ICF) toClassify the Degree of Deficiency in Victims ofTraffic Accidents by Motorcycles.

L. Oliveira, PhD1, J. H. Scatena, PhD1 and A. P. Macedo2

1Universidade Federal de Mato Grosso, Cuiaba, Brazil, 2Secretaria deEstado da Saude de Mato Grosso, Cuiaba, Brazil

INTRODUCTION: Nowadays, traffic accidents are one of the

most important public health problems. Although the magnitude of

the morbidity and mortality of these events is well established, little

is known about their consequences. To better understand the dis-

abilities and standardize a language on them the World Health

Organization developed the ICF. The aim of the study was to ana-

lyze simultaneously the International Statistical Classification of

Diseases and Health Related Problems (ICD-10), focusing the injury

resulting from traffic accidents, with the components of ICF, related

to qualification of disability, structure and function of the body.

METHODS: This is an exploratory, prospective study with 49

patients in a rehabilitation unit, victims of motorcycle accident; defi-

ciencies function and body structure were identified from analysis of

physical therapy and coded according to the health dimensions of

the ICF. Data regarding injury were coded by ICD-10.

RESULTS: It was found a mean number of 2.3 lesions per individ-

ual. The lesions on the lower limbs and upper limbs appeared in

greater percentage: 41% and 37%, respectively. There was a wide

gradation of impairment of the structures of the body injured, with

a predominance of fractures (57%) and multiple trauma (14%). The

main functional deficiencies were related to muscle strength (94%),

joint mobility (92%) and pain (84%). Concerning the degree of

these deficiencies, severe (30%) or complete (44%) lesions predomi-

nated. As regards the structure of the body, high percentages of

severe or complete impairment occurred among the victims who had

fractures (83%) and polytrauma injuries (80%).

CONCLUSIONS: The use of ICF has highlighted aspects that ICD-

10 does not show and the concomitant analysis of these two instru-

ments resulted in important information, evidencing that such com-

plementarity extends and improves the measure of the health status

of individuals.

Poster Number: T145 Abstract #: 3052Neurophysiological Effects of Heart RateVariability Biofeedback Training in Adolescents: aRussian Study.

D. Demin1,2, L. Poskotinova1,2, A. M. Grjibovski, PhD3,4 and Z.Varakina, PhD3

1Institute of Environmental Physiology, Russian Acad. Sci., Ural Branch,Arkhangelsk, Russia, 2Northern Arctic Federal University, Arkhangelsk,Russia, 3Northern State Medical University, Arkhangelsk, Russia,4Norwegian Institute of Public Health, Oslo, Norway

INTRODUCTION: Earlier studies suggest that heart rate

variability biofeedback (HRVB) training is a promising method for

non-pharmacological correction of vascular dystonia. The aim of this

study was to assess the effect of HRVB sessions on the functional

brain activity and hemodynamic parameters in Russian adolescents.

METHODS: A group of 15–16 years old adolescents with high sym-

pathetic tone (Group I, n¼24) and a group of adolescents of similar

age, but with a normal sympathetic tone (n¼22) participated in the

study. HRVB sessions were carried out as specified in the patent N

2317771 RU). Every adolescent held 10 HRVB sessions.

Electroencephalogram (EEG) was recorded at the first and at the

tenth session of HRVB. Maximal amplitude and index in each fre-

quency EEG-ranges (4–7 Hz, 8–12 Hz, 13–25 Hz) were performed.

Paired data were compared using Wilcoxon sign rank test while

independent data were compared by Mann-Whitney tests.

RESULTS: Higher baseline levels of the studied EEG parameters in

adolescents with high sympathetic tone compared to adolescents

with normal sympathetic tone (P for all tests <0.05) were observed

at the first HRVB session. After the first session of HRVB, EEG

alpha-activity increased in both groups (P for both tests <0.05)

while systolic blood pressure decreased (P for both tests <0.05) as

compared with baseline values. Adolescents with high sympathetic

tone showed a reduction in theta-activity (P<0.05) while no

changes were found in adolescents with normal sympathetic tone.

After the tenth HRVB session all studied indices of EEG-activity in

the group of adolescents with high sympathetic tone (all P<0.01)

became comparable with the corresponding indices observed in the

group of adolescents with normal sympathetic tone.

CONCLUSIONS: HRVB sessions have a potential for reduction of

blood pressure and for optimization of the functional brain activity

and increase of the stability of a subcortical brain structures regula-

tion among adolescents with high sympathetic tone.

Poster Number: T210 Abstract #: 3376Three-year Change in the Wellbeing of Orphanedand Abandoned Children in Institutional andFamily-Based Care Settings in Five Low andMiddle Income Countries.

L. C. Messer, PhD1, K. Whetten, PhD2, J. Ostermann, PhD2, B.W. Pence, PhD3, S. Ariely, PhD2, K. O’Donnell, PhD2, R.Whetten, MPH2, A. Wasonga, MA4, V. Vann, MA5,M. Eticha, MA6 and I. Madan, MS7

1Portland State University, Portland, OR, 2Duke University, Durham, NC,3University of North Carolina, Chapel Hill, NC, 4ACE-Africa, Bongoma,Kenya, 5Homeland, Battambang, Cambodia, 6Stand for VulnerableOrganization, Addis Ababa, Ethiopia, 7Sahara Centre for Residential Careand Rehabilitation, New Delhi, India

INTRODUCTION: With an estimated 132 million orphans and

more than two million children living in group homes, or ‘institu-

tions’ worldwide, the extent to which institution-based caregiving

negatively affects development and wellbeing has become a central

question for international policymakers.

METHODS: A two-stage random sampling methodology identified

community representative samples of 1357 institution-dwelling

orphaned and abandoned children (OAC) and 1480 family-dwelling

OAC aged 6–12 from five low and middle income countries. Data

were collected from the children and their primary caregivers at

baseline and twice annually for three years. Survey-analytic techni-

ques and linear mixed effects models describe child wellbeing at

baseline and at 36 months and the variation in outcomes between

children, care settings, and study sites.

RESULTS: At the 36-month follow-up, institution-dwelling OAC

had statistically significantly higher mean height for age and health;

family-dwelling OAC did not have statistically significantly better

outcomes on any of the measures. At both baseline and follow-up,

the magnitude of the differences between the institution- and


Dow



family-dwelling groups was small. Relatively little variation in out-

comes was attributable to differences between sites (11–27% of

total variation) or care settings within sites (8–14%), with most var-

iation attributable to differences between children within settings

(60–75%). The percent of variation in outcomes attributable to the

care setting type, institution- versus family-based care, ranged from

0–4% at baseline, 0–3% at 36-month follow-up, and 0–4% for

changes between baseline and 36 months.

CONCLUSIONS: These findings contradict the hypothesis that

group home placement universally adversely affects child wellbeing.

Without substantial improvements in and support for family set-

tings, the removal of institutions, broadly defined, would not signifi-

cantly improve child wellbeing and could worsen outcomes of

children who are moved from a setting where they are doing rela-

tively well to a more deprived setting.

Poster Number: T190 Abstract #: 3005Time-to-event Analysis for CER in ProgressiveDiseases: The Case of Parkinson’s Disease.

J. P. Reese, PhD1, A. Werdecker, PhD1, R. Westerman, PhD1, R.Dodel, MD1, U. Mueller, MD1 and K. Kostev, PhD2

1Philipps-University Marburg, Marburg, Germany, 2IMS Health,Frankfurt, Germany

INTRODUCTION: Comparative long-term health effects of

treatment alternatives can hardly be covered by experimental

RCT designs. This is a pivotal problem in slow degenerative disorders

like Parkinson’s Disease (PD). We used PD as paragon to investigate

the effect of different initial therapies using time-to-levodopa as

primary outcome. Levodopa is the most effective drug and a lead

substance in PD, but it may cause severe complications in long-term

use. Therefore the initial treatment in many PD patients starts with

other substances like MAO-B inhibitors, dopamine agonists or anti-

cholinergics. No head-to-head comparisons of the alternative treat-

ments exist.

METHODS: We used the Disease Analyzer database (IMS HEALTH),

containing basic medical data from approximately 20 million patients

in Germany. The primary outcome was the therapy change rate from

initial treatment to levodopa estimated by Kaplan-Meier analyses. A

Cox proportional hazards model was used to estimate the relationship

between time-to-levodopa and confounders for a maximum follow-up

of 10 years (until Dec 2011). Adjusted hazard ratios(HR) and 95% con-

fidence intervals(CI) are presented for change-to-levodopa rate. In addi-

tion to our empirical data we discuss applications of time-to-event

analyses in PD using other databases and additional outcomes relevant

for CER and health service research.

RESULTS: A representative sample of de-novo patients diagnosed

with PD was drawn (n¼108,885). 71.8% of patients received levo-

dopa as a first-line treatment. 29,708 patients started with other

anti-PD substances: 13.3% with dopamine agonists (DA), 3.6%

with amantadine, 5.9% with anticholinergics, and 0.8% with

MAO-B inhibitors. 29.0% of patients not starting with levodopa

switched to levodopa within five years. Compared with PD patients

starting with anticholinergics, patients starting with MAO-B-

Inhibitors or DAs showed significantly lower proportions of levo-

dopa free patients after five years (35% and 55%, respectively).

Compared to MAO-B inhibitors, the HR for switching to levodopa

was 0.38 (CI 0.34–0.43; P <.001) for anticholinergics and 0.74 (CI

0.67–0.83; P<.001) for non-ergot DA.

CONCLUSIONS: Comparative time-to-event analysis using admin-

istrative data is a fruitful approach for CER in health service

research and pharmacoepidemiology. A systematic screening for its

application in progressive chronic diseases may be of great practical

value especially where head-to-head comparisons are unavailable.

Poster Number: T228 Abstract #: 3131Do Women with Coeliac Disease ClinicallyPresent with Fertility Problems more often thanWomen in the General Population.

N. N. Dhalwani, MS, J. West, PhD, A. Abdul Sultan, PhD,

L. Ban, PhD and L. J. Tata, PhD

University of Nottingham, Nottingham, United Kingdom

INTRODUCTION: Several studies have raised the spectre of infertility

being associated with coeliac disease (CD). These have primarily been

carried out in small groups of women attending infertility specialist serv-

ices and subsequently screened for CD. We carried out a large popula-

tion-based study to compare clinical reporting of fertility problems in

women with and without CD in the United Kingdom (UK).

METHODS: Using a large primary care database of patients from

across the UK, we estimated age-specific rates of new clinically

recorded fertility problems in women with and without CD during

their childbearing years between 1990 and 2013. These rates were

stratified by whether the fertility problem was recorded before or

after the CD diagnosis and were compared to the general population

rates using Poisson regression with adjustments for socioeconomic

status and calendar time.

RESULTS: Out of 2,426,225 women, 6506 (0.3%) had a diagnosis

of CD. The age specific-rates of new clinically recorded fertility

problems in women with CD were very similar to the general popu-

lation rates (11.6/1,000 person-years in women with CD and 12.6/

1,000 person-years in the general population in the 30–34 year age

group). When stratified by the time of CD diagnosis the rates of clin-

ically recorded fertility problems were very similar to the general

population in both the time before and after CD diagnosis. However

these rates were 50% higher after CD diagnosis in women aged

25–29 compared to the same age group in the general population

(IRR 1.50, 95% CI 1.10–2.04).

CONCLUSIONS: Women with CD do not have a substantially

greater likelihood of reporting fertility problems than women with-

out the condition. Rates were slightly higher in younger women

with diagnosed CD, however as this effect did not hold for women

Poster Number: T46 Abstract #: 3370To Investigate the Infant Feeding PracticesPrevalent Among Mothers of Chandigarh, India.

R. Pradhan, PhD and A. Arora

Government Home Science College, Chandigarh, India

INTRODUCTION: Poor feeding practices in infancy and early

child hood results in malnutrition.

OBJECTIVES: To investigate the infant feeding practices prevalent

among mothers of infants a community-based, cross-sectional

descriptive study was conducted in Chandigarh, India.

METHODS: A total of 300 mothers, with infants aged 6–24 months,

who visited the Pediatric O.P.D. at hospitals of Chandigarh, were


Dow



randomly selected and interviewed. The self developed, pre tested

questionnaire was used for data collection

RESULTS: Out of 300 mothers interviewed, 87.6% and 40.7%

mothers knew about exclusive breastfeeding and weaning foods,

respectively. The proportions of infants with exclusive breastfeeding

under six months (78%) and infants who received complementary

feeding at the age of 6–8 months (81.7%) were high. 86.7% and

60% infants were continuing with breast feeding till the age of 6–12

months and 12–24 months, respectively. Mother was the primary

care giver (73%). Doctors (66.7%) were major source of informa-

tion for mothers regarding weaning foods. Homemade weaning

foods were commonly given to infants. 25% mothers included green

leafy vegetables like spinach in weaning foods. Weight monitoring

was done for all infants. 11.33% infants were found under weight

(WHO, 2008). 56.7% followed a time table for giving food to their

children. 81.7% respondents included fat and sugar to make wean-

ing foods energy dense. 28.7% infants were given nutritional supple-

ments and tonics.

CONCLUSIONS: Nutrition education regarding exclusive breast

feeding and healthy complementary feeding will go a long way in

improving and achieving optimal nutritional status and reducing

malnutrition.

Poster Number: T157 Abstract #: 3297Dysphagia and Dietary Intake of JapaneseAdults: the Shizuoka Area in a Japanese Multi-Institutional Collaborative Cohort Study.

M. Naito, PhD1, K. Wakai, PhD1, S. Kawai, PhD1, H. Nakagawa,

PhD1, T. Tamura, PhD1, S. Suma1, Y. Sugimoto, MS1, N.

Fukuda, MS1, T. Sasakabe1, R. Okada, PhD1,

E. Morita, PhD1, Y. Asai, MD2, A. Mori, MD3 and

N. Hamajima, PhD1

1Nagoya University Graduate School of Medicine, Nagoya, Japan,2Mikatahara, Hamamatsu, Japan, 3Seirei, Hamamatsu, Japan

INTRODUCTION: Dysphagia is an indicator of nutritional risk in

older adults. Few studies have reported the incidence of dysphagia

and its relationship with food intake in young and middle-aged

adults. We aimed to clarify the association of dysphagia with

nutrient intake in people in the local community.

METHODS: We analyzed data from 4,647 Japanese males and

females (mean age 6 SD, 52 6 12 years; range 35–69 years) who

participated in a cohort study (J-MICC) from 2006 to 2007. They

completed a validated food frequency questionnaire to estimate their

nutrient intake levels. Dysphagia and number of teeth lost were self-

reported and functional difficulty was categorized as always/often,

sometimes, seldom, or none. We calculated mean daily intake

according to dysphagia level, adjusting for sex and age by analysis

of covariance and p values for trend were calculated.

RESULTS: Overall, 19% of the subjects had dysphagia, and the

proportion increased with age (P < .001) from 13% (30s) to 23%

(60s). There were no statistical differences between males and

females. The mean intake of carotene (P < .001), soluble dietary

fiber (P = .008), and insoluble dietary fiber (P = .005) decreased

with an increasing level of dysphagia. Even after statistical adjust-

ment for the number of teeth lost, these associations remained

significant.

CONCLUSIONS: Swallowing problems were associated with an

impaired intake of some nutrients. Dysphagia may affect the choice

of nutritious foods, such as vegetables and fruits, among Japanese

adults. Additional studies are needed to confirm this association and

to determine whether the association relates to risk of future health

problems.

Poster Number: T166 Abstract #: 3321Oral health and Nutritional Status in Patients

Affected with Parkinson’s Disease.

A. Shyam, MD

Meghna institute of dental sciences, Nizamabad, India

INTRODUCTION: The prevalence of Parkinson’s disease (PD) is

expected to increase over the next decade. The disease has several

implications relating to oral and general health, and is manageable

with proper awareness and knowledge. The aim of the present study

was to assess oral health status and nutritional status which may

influence oral and general health among patients with Parkinson

disease.

METHODS: A cross sectional study was conducted to assess the

Oral Health and Nutritional Status of Parkinson’s disease affected

patients. The study population consisted of 75 subjects in the range

of 60–76 years. Oral health status was assessed using the WHO

Oral Health Assessment Form 1997, Nutritional status using

anthropometric measurement of Body Mass Index after obtaining

informed consent. Data was analyzed using SPSS 17 software and

the statistical tests performed were Chi-square and ANOVA.

RESULTS: In the present study, the frequency of untreated caries,

periodontal diseases, and missing teeth was significantly high. The

proportion of filled teeth was insignificant while the need for pros-

thesis was largely unmet. Patients had more complaints of chewing

difficulties and denture discomfort than controls. More than half of

the patients had problems with swallowing. Intra oral mucosal

lesion was more prevalent on the buccal mucosa region. (61.5%) of

subjects were found to be underweight while (38.5%) subjects were

considered as having normal nutritional status.

CONCLUSIONS: The affected patients had poor oral health and

general status and required restorative and prosthetic therapy. These

findings may help as adjunct for the personal care attendant of

Parkinson’s disease patients for improving their oral and general

health which would be helpful in making health care interventions

strengthening community-based palliative care for Parkinsonian

patients.

Poster Number: T97 Abstract #: 3359Tracking MPOWER across Asian Countries –

Results from the Global Adult Tobacco Survey,

2008–11.

M. Nguyen, MPH, M. Husain, PhD, S. Asma, K. M. Palipudi and

E. Ridgway

US Centers for Disease Control and Prevention, Atlanta, GA

INTRODUCTION: The World Health Organization (WHO)

MPOWER technical package consists of six proven strategies to

reduce tobacco use and assist countries in fulfilling their obligations

to the WHO Framework Convention on Tobacco Control (FCTC).

The Global Adult Tobacco Survey (GATS) provides the means by


Dow



which a country can produce national and subnational estimates to

assess the impact of tobacco control and prevention initiatives.

METHODS: GATS is a nationally representative household survey

of adults aged 15 years and older chosen through a stratified multi-

stage cluster sampling approach. It includes measures of WHO

MPOWER indicators and has a standard and consistent protocol

allowing for cross-country comparisons. Data was analyzed using

SPSS version 17 and Sudaan version 10. GATS has been conducted

in eight countries in Asia, and we selected one key indicator to repre-

sent each of the six MPOWER measures and compared findings

across the continent.

RESULTS: GATS results show that smoking rates ranged from

14.0% in India to 34.8% in Indonesia. Secondhand smoke exposure

varied greatly by indoor public place; generally exposure was high-

est in bars/nightclubs and lowest in healthcare facilities. The

Philippines reported the highest proportion of smokers who were

advised to quit smoking by healthcare professionals (76.5%), while

only 26.7% of smokers in India were advised to quit. Point-of-sale

tobacco advertisement exposure rates ranged from 4.1% in China to

53.7% in the Philippines. The percentage of smokers who thought

about quitting because of health warning labels on cigarette pack-

ages was lowest in Indonesia (27.1%) and highest in Bangladesh

(74.4%).

CONCLUSIONS: As tobacco use continues to be a leading cause of

preventable death, there is a demonstrated need for the continued

monitoring of tobacco use behaviors and control policies. GATS

provides key data for monitoring the implementation and impact of

MPOWER interventions.

Poster Number: T112 Abstract #: 3095Early Life Risk Factors for Childhood Adiposity –a Comparison of Marginal Growth Models UsingHong Kong’s ‘Children of 1997’ Birth Cohort.

S. Lin, PhD1, E. Corpeleijn, PhD2, C. M. Schooling, PhD1 andE. van den Heuvel, PhD2

1The University of Hong Kong, Hong Kong, Hong Kong, 2University

Medical Centre Groningen, Groningen, Netherlands

INTRODUCTION: Risk factors for childhood obesity identified

were largely based on cross-sectional data analysis considering obe-

sity at one single time point. Recent development of growth model-

ing techniques has facilitated the investigation from a longitudinal

perspective, but no systematic comparison of marginal models has

been performed. The main purpose is to determine risk factors for

growth at a population average level and the influence of different

statistical approaches.

METHODS: Body mass index (BMI) from a large population-repre-

sentative Chinese birth cohort, Hong Kong’s ‘Children of 1997’ was

transformed to sex- and age- specific z-scores. Maximum likelihood

estimation (MLE) under the assumption of normally distributed

data, generalized estimating equations (GEE), and quantile regres-

sion (QR) for estimation of the median growth in relation to

early life risk factors for childhood adiposity were compared. The

models were compared with and without multiple imputation (MI)

for missing BMI and risk factors under the assumption of missing at

random.

RESULTS: The distribution of BMI z-scores seemed symmetric and

did not strongly deviate from normality. However, the three meth-

ods differed in the estimates, standard errors, significance levels, and

interpretation. Most notably MLE seemed to deviate from GEE and

QR, while GEE and QR seemed more similar, irrespective of the use

of imputation or not. For example, a subgroup of children was iden-

tified for which the MLE demonstrated a decline in BMI z-scores

over time of 0.15, while GEE and QR showed an increase of 0.2.

CONCLUSIONS: The results did not indicate one optimal method

for a population-averaged interpretation. Given the lack of a ‘gold

standard’ statistical method, we suggest assessing the consistency of

conclusions with all these methods when analyzing longitudinal

data to eliminate possible discrepancies due to statistical methods.

Further studies will be needed to determine the best method for ana-

lyzing this type of longitudinal growth data.

Poster Number: T26 Abstract #: 3338Ovarian and Uterine Cancer Incidence andMortality in Non-Hispanic American Indians andAlaska Native Women, United States, 1999–09.

S. D. Singh, MD1, A. B. Ryerson, PhD1, M. Wu, MD1 and

J. S. Kaur, MD2

1Centers for Disease Control and Prevention, Atlanta, GA, 2Mayo Clinic

College of Medicine, Rochester, MN

INTRODUCTION: Limited information exists on ovarian and ute-

rine cancer among American Indian/Alaska Native (AI/AN) women

and misclassification of race in these populations can lead to

underestimations.

METHODS: We linked mortality data (1990–2009) from National

Center for Health Statistics (NCHS) and incidence data

(1999–2009) from National Program of Cancer Registries (NPCR)

and Surveillance, Epidemiology, and End Results (SEER) program

to the Indian Health Services (IHS). We compared ovarian and ute-

rine cancer incidence and mortality for non-Hispanic AI/AN popula-

tions in CHSDA counties to non-Hispanic White (NHW) women.

RESULTS: Ovarian and uterine cancer death rates were similar in

non-Hispanic AI/AN women and NHW women. Ovarian and ute-

rine cancer incidence and death rates were higher for non-Hispanic

AI/AN residing in CHSDA counties as compared to those residing in

all US counties. We also observed geographic differences, regardless

of CHSDA residence, in incidence and death rates of ovarian and

uterine cancer in non-Hispanic AI/AN women by IHS region, with

women in Pacific Coast (IR 10.2)and Southern Plains (IR 9.7) hav-

ing higher death rates from ovarian cancer and women in Northern

Plains (IR 5.4) having higher uterine cancer death rates.

CONCLUSIONS: There are significant regional differences in the

incidence and mortality of ovarian and uterine cancers among AI/

AN women in the US. More research among correctly classified AI/

AN women is needed to understand these observations.

Poster Number: T195 Abstract #: 2204Providing National Vital Health Statistics byEthnic Group at Low Cost: SHELS.

R. Bhopal, MD

University of Edinburgh, Edinburgh, Scotland

On behalf of the Scottish Health and Ethnicity Linkage Study Team

INTRODUCTION: increasing international migration is producing

ethnically diverse societies, with population groups exhibiting

important differences in health status and health care utilisation.


Dow



Legislation, policy, professional standards and business efficiency

requires data by ethnic group but these are sparse.

METHODS: The Scottish Health and Ethnicity Linkage Study

(SHELS) is a retrospective cohort of 4.65 million people created

using name, address, sex and date of birth for probability-linkage of

the 2001 Census population for Scotland (reporting 14 ethnic group

categories) to the National Health Service Scotland identification

number (CHI). Subsequently, CHI permitted linkage to hospitalisa-

tion, mortality, compulsory psychiatric detention, breast cancer

screening and other databases. Eight indicators of socio-economic

position were available as potential confounding variables.

RESULTS: Ethical and privacy advisory committee permissions were

achieved after resolving technical problems especially on minimising

risk of inadvertent disclosure of identity. Analyses include those on

cardiovascular diseases, cancers, breast cancer screening, maternal

and child health, mental health including compulsory detention (all

published or about to be). For example, compared to the White

Scottish population (reference value 100) men in the Pakistani ethnic

group had the following age-adjusted risks (95% confidence intervals

exclude 100): colorectal cancer 45.6 and myocardial infarction 164.1.

For Pakistani women the risk of non-attendance for breast cancer

screening was 181.7. Analyses on gastrointestinal disorders and respi-

ratory disorders are completed (papers under preparation). Analysis

are planned on all-cause mortality, all-cause hospitalisation, all infec-

tions, accidents & poisoning, bowel cancer screening, hepatitis C & B

and HIV. SHELS has cost about 20–30 pence (US 30–45 cents) per

person in Scotland.

CONCLUSIONS: Data by ethnic group (or like-variables) can be

obtained at low cost through data linkage methods wherever census

(or population register) records ethnicity. SHELS offers a model for

use internationally.

Poster Number: T130 Abstract #: 3339Early Mortality in Patients after Hospitalizationfor Diabetic Care Sensitive Conditions Primary ina Midle City of Brazil: A Worrying Reality.

C. L. Oliveira, PhD1, A. C. S. Lopes, PhD2, C. M. Padua, PhD2

and C. S. Cardoso, PhD3

1Federal University of Sao Joao del-Rei, Divinopolis, Brazil,2Universidade Federal de Minas Gerais, Belo Horizonte, Brazil, 3Federal

University of Sao Joao del-Rei, Divinopolis, Brazil

INTRODUCTION: Hospitalizations related to diabetes mellitus

(DM) and its complications are considered Ambulatory Care Sensitive

Conditions (ACSC) and represent about 4.3% of hospital admissions.

METHODS: A prospective cohort study that included all diabetic

patients older than 18 years, hospitalized in the period July-

October, 2011 in Divinopolis, MG, Brazil. After 18–24 months of

discharged, we conducted the first visit of follow up. Data were col-

lected containing sociodemographic, lifestyle, and food and medica-

tions, social support networks and quality of life, clinical and

anthropometric examination, ECG and blood tests

RESULTS: 92 patients were eligible for the study and 72 have

already been approached by phone or home visit. After about 20

months, there were 22 deaths, with a mortality rate of 30.6% in the

study population. The mortality rate in patients 50–69 years was

36.4% and above 70 years was 43.5%. Approximately 20.4% of

patients had poor metabolic control, 18.2% had high blood pressure

and 29.5% had signs of diabetic foot, 25% of patients were unable

to walk to the nearest health Unit. In this study, diabetes contributed

126 years of potential life lost, which equates to an average of 10.5

years of life less per individual.

CONCLUSIONS: Around 50% of patients remained without

adequate treatment after discharge. For those seeking health care,

the costs of medication and maintaining a proper diet can also make

it difficult for low-income patients benefit from health care. The per-

centage of complications observed in the study group points to a

potential seriousness of the situation. Brazil advanced in building a

comprehensive health system, but because of its complexity improve

their quality and resolution is still a major challenge. The partner-

ship between health services and universities are a viable possibility

for diagnosis and intervention in these scenarios

Poster Number: T188 Abstract #: 3367Pattern and Safety of Drug Prescriptions amongAntenatal Women in Nigeria.

N. U. Igboeli1 and M. O. Adibe2

1University of Nigeria, Nsukka., Nsukka, Nigeria, 2University of Nigeria

Nsukka., Nsukka, Nigeria

INTRODUCTION: Drug use in pregnancy is always a concern

owing to the potential risk(s), both to the mother and to the foetus.

Thus pregnant women are advised to consult their doctor or phar-

macist before taking any drug, including over the counter medicines.

The study aimed at evaluating the pattern and safety of drug pre-

scriptions at antenatal clinics.

METHODS: The study involved retrospective assessment of patients’

antenatal folders selected from two public hospitals in Enugu State,

Nigeria. Relevant information was extracted using a proforma.

RESULTS: The average number of drugs per prescription was 4.97.

Among the 815 women whose folders were analysed, 270 (33.1%)

had at least one of the FDA category C, D or X drug prescribed dur-

ing pregnancy. The exposure rate was higher for category C drugs

(31.5%) than for category D (6.7%) or X (0.1%) drugs. Antenatal

routine drugs, comprising of multivitamins and iron, were pre-

scribed for all women. The second most prescribed drug group was

antimalarials (82.8%), followed by analgesics (25.6%). Total num-

ber of drugs prescribed per woman [aOR¼38.37, 95%

CI¼14.84–99.21], prescription of an analgesics [aOR ¼79.44,

95% CI¼13.21–481.07] or an antibiotics [aOR¼0.03, 95%

CI¼0.01–0.18] were significantly associated with the chance of

women receiving at least a drug from category C, D or X.

CONCLUSIONS: Exposure to drugs with recognized potential preg-

nancy risk was high—occurring in at least three out of every 10 preg-

nant women in this urban area. However, prescription of antenatal

routine drugs, which included folic acid and haematinic supplements,

was very commendable. These findings emphasized the need for regu-

lar prescription audit and continuing education for physicians in order

to promote rational and safer drug prescribing in pregnancy.

Poster Number: T116 Abstract #: 2842Typology of Families Living in Urban areas andObesity: One Family does not Fit all.

R. Andrade, PhD1,2, D. Costa, BA2, S. Bispo, MD2, C. Ramos, BA2,

P. Reis, BA2, O. Chaves, MD2, C. Xavier, PhD2, F. Proietti, PhD2,

M. I. Correia, PhD1,2, A. Friche, PhD1,2 and W. Caiaffa, PhD1,2


Dow



1Federal University of Minas Gerais, Belo Horizonte, Brazil, 2Belo

Horizonte Observatory for Urban Health/UFMG, Belo Horizonte, Brazil

INTRODUCTION: The prevalence of overweight is increasing

alarmingly among adolescents and may cluster in families due to

social and lifestyle characteristics. This study aims to build a score

to identify family profiles that may be related to obesity.

METHODS: Using data from a population-based household survey

data randomly clustered in three stages in Belo Horizonte, Brazil,

(census tracts, households and residents). We created eleven

domains to characterize profiles of healthy and unhealthy family.

Information obtained from interviews from one adult (�18 yo) and

one adolescent (11–17yo) at the same household were paired (1042

pairs of observations). The following domains were created accord-

ing to a theoretical model of obesity: D1. Pattern of family meal,

D2. Relationships between adolescent and family, D3. High risk

behavior of family, D4. Wellbeing satisfaction of the family, D5.

Family self perception of health D6. Physical active household, D7.

Sedentary behavior of family, D8. Educational level of family, D9.

Family with preventive health attitudes, D10. Chronic conditions in

the family and D11. Food insecurity in the household. Eleven scores

were constructed and evaluated by Cronbach’s a.

RESULTS: Each domain had 1–10 itens, on average of five itens

and Cronbach’s a ranged from 0.30 to 0.76. Discriminant analysis

using Body Mass Index-BMI (normal weight -BMI between 3 and

85 percentile and obese -BMI �percentile 97) correctly classify

90.2% of adolescents in two groups of families: healthy and unheal-

thy with type I error of 2.2%. The domains of D10 (Chronic condi-

tions in the family), D5 (Family self perception of health), D8

(Educational level of family) and D3 (High risk behavior of family)

showed to discriminate better nutritional status.

CONCLUSIONS: The score constructed seems to discriminate pro-

files of ‘obesogenic’ family, those with unhealthy profile related to

adolescent overweight suggesting that family typology may help to

direct specific interventions.

Poster Number: T96 Abstract #: 3046Understanding Barriers to Delayed-StageDiagnosis of Cervical Cancer.

J. W. Morgan, DrPH1, S. Saghari, MPH2, M. Ghamsary, PhD2

and A. Marie-Mitchell, MD3

1Loma Linda University School of Public Health and SEER Cancer

Registry of Greater California, Loma Linda, CA, 2Loma Linda University

School of Public Health, Loma Linda, CA, 3Loma Linda University School

of Medicine, Loma Linda, CA

INTRODUCTION: California Hispanic women experience highest

cervical adeno- and epidermoid- carcinomas(CACX) risk, followed

by non-Hispanic blacks (NHB), Asian-others and Non-Hispanic

whites (NHW). The California Cancer Registry (CCR) is the state-

wide population-based cancer reporting system and consists of the

three largest SEER registries.

METHODS: We conducted a non-concurrent cohort investigation

assessing independent roles of age, race/ethnicity, socioeconomic

status (SES), and marital status as independent predictors of stage II-

IV (delayed-) vs stage I (early-stage) CACX.

FINDINGS: Among 5,324 early- and 5,222 delayed-stage CACX

cases in the CCR from 1996 to 2005, 61.8% were age 50þ years.

Logistic regression adjusted odds ratios (OR) with 95% confidence

intervals (OR, 95% CI) contrasting delayed- vs early-stage CACX

for age-categories were: OR20-34/50-69¼0.29, 0.26–0.33; OR35-49/50-

69¼0.50, 0.46–0.54 and OR70þ/50-69¼1.45, 1.29–1.62; Trend

P<0.0002. Similar contrasts for race/ethnic groups with NHWs

showed: ORAsian-other/NHW¼1.04, 0.94–1.15; ORNHB/NHW¼1.27,

1.11–1.46; and ORHispanic/NHW¼1.06, 0.99–1.14. Contrasts of

lower SES-categories with highest showed: ORSES1/5¼1.53,

1.38–1.71; ORSES2/5¼1.40, 1.25–1.56; ORSES3/5¼1.23, 1.09–1.38;

and ORSES4/5¼1.04, 0.92–1.17 (Trend P<0.0001). Marital status

contrasts of single (S); separated, divorced, or widowed (SDW); and

unknown with married (M) were: ORS/M¼1.27, 1.18–1.38;

ORSDW/M¼1.97, 1.81–2.14; and ORUnknown/M¼0.84, 0.71–1.01.

DISCUSSION: Lower ratio of delayed- vs early-stage CACX during

reproductive years, with increased odds for age 50þ is consistent

with poorer screening after menopause. Absence of differences in

odds of delayed- vs early-stage cervical cancer among Asian/other

and Hispanic women versus NHW suggests similar success screen-

ing in these race/ethnic groups, while a higher odds of delayed- vs

early-stage among NHB women persists in the adjusted model. An

inverse dose-gradient for delayed- versus early-stage CACX with

progressively higher SES was evident, with markedly higher odds of

delayed-stage among SDW compared to married women.

CONCLUSIONS: Progressively lower SES showed an inverse dose-

gradient for delayed- versus early-stage CACX that is more robust

than race/ethnicity, while SDW versus married was the strongest

independent predictor of delayed-stage CACX.

Poster Number: T133 Abstract #: 3147Pattern of Areca (betel) Nut Chewing andObesity Measures in Guam and Saipan,Mariana Islands.

Y. C. Paulino, PhD1, E. L. Hurwitz, PhD2, L. R. Wilkens, DrPH3,R. Novotny, PhD2, M. J. Miller1 and K. M. G. Quinata1

1University of Guam, Mangilao, Guam, 2University of Hawaii-Manoa,

Honolulu, HI, 3University of Hawaii Cancer Center, Honolulu, HI

INTRODUCTION: Areca (betel) nut is chewed by 600 million peo-

ple worldwide, and is the fourth most frequently used substance fol-

lowing nicotine, ethanol, and caffeine. Chewing patterns vary

among populations. The practice has been shown to be associated

with chronic diseases, including obesity. The purpose of this study is

to estimate associations between chewing patterns and obesity meas-

ures among residents of the Mariana Islands.

METHODS: A cross-section of 300 betel nut chewers (157 males;

143 females) �18 years old were recruited in Guam and Saipan

between January 2011and June 2012. Information on demographics,

betel nut use and related behaviors, and anthropometry were col-

lected. Height and weight measurements were used to calculate body

mass index (BMI) for weight status, and waist and hip measurements

were used to calculate waist-to-hip ratio (WHR) for central adiposity.

Sex-specific adjusted associations using both the logistic and log-bino-

mial models were used to estimate the prevalence of obesity (BMI

�30 kg/m2) and high central adiposity (WHR>1.0 in males,>0.85

in females) and their association with betel nut chewing pattern,

which was classified into two patterns by cluster analysis.

RESULTS: The majority (n¼222; 74%) of the participants were

Class 2 chewers who chewed with tobacco; the remainder were Class

1 chewers who chewed the areca nut without tobacco. Compared to

male and female Class 2 chewers, obesity [males: prevalence ratio

(PR)¼1.12; confidence interval (CI)¼0.78–1.62/females: PR¼1.09;


Dow



CI¼0.84–1.43] and high central adiposity [males: PR¼1.68;

CI¼1.07–2.65/females: PR¼1.00; CI¼0.80–1.23] were more preva-

lent in Class 1 chewers. The adjusted associations were stronger in

males than in females for Class and obesity [males: odds ratio

(OR)¼1.76; CI¼0.75–4.09/females: OR¼1.31; CI¼0.43–4.01]

and Class and high central adiposity [males: OR¼2.92;

CI¼1.21–7.07/females: OR¼0.90; CI¼0.24–3.41].

CONCLUSIONS: The findings suggest that Class 1 chewing pattern

is associated with high central adiposity among males, but not

among females in the Mariana Islands.

Poster Number: T80 Abstract #: 2929Unintentional Injury Mortality among AlaskaNative People.

E. Provost, DO1, J. A. Craig, MPH1, H. D. Strayer, MPH1, E.

Asay, MS1, G. Day, MPH1 and P. Holck, PhD2

1Alaska Native Tribal Health Consortium, Anchorage, AK, 2Peak

Enterprises, Falmouth, MA

INTRODUCTION: Injury has been the leading cause of death

among Alaska Native people (AN). AN life expectancy, a major

population health status indicator, is 7.2 years lower than US

Whites (2004–08), most of which can be attributed to injury mortal-

ity. Many injury deaths are preventable. Findings from analyses of

AN injury mortality (2004–08 and 2002–11) will be presented with

a focus on the impact of unintentional injuries.

METHODS: Data were obtained from the Alaska Bureau of Vital

Statistics and the Alaska Department of Labor and Workforce

Development. Age-adjusted rates standardized to the 2000 US

standard population and trends were calculated.

RESULTS: Injury remains the leading cause of death among AN

people. Intentional and unintentional (UI) injury deaths were the top

causes of death for persons less than 45 years of age. UI deaths

accounted for 25% of years of life lost before age 75 (2004–08). UI

deaths represented 60.3% of all injury deaths (1,718). Poisoning

(26.6%), motor and off-road vehicle (24.7%) related injuries, and

drowning (16.3%) comprised 67.6% of all UI deaths (1037). UI

death rates have declined 47% between 1980–83 and 2004–08, and

13.2% between 1992–95 and 2008–11. A disparity in UI death rates

of 2.1 times persists for AN people compared to Alaska non-Natives

(105.1 and 49.5 per 100,000, P<0.05).

CONCLUSION: Injury is a leading cause of AN mortality.

Unintentional mortality rates have decreased for Alaska Native

people over the past three decades however the rate of decrease

has slowed and a disparity still exists. Despite increased levels

of exposure to injury risk, further improvements in prevention of

unintentional injuries are needed. With these improvements, the gap

in life expectancies for AN people can be significantly decreased.

Poster Number: T216 Abstract #: 3360Effect of Socioeconomic Vulnerability, FamilyStructure Patterns and Non-Acceptance ofPregnancy on Preterm Birth: A StructuralEquation Model. Londrina, Brazil.

A. Oliveira, MPH1, G. P. Alencar, PhD1, P. L. Assuncao, PhD2,A. M. Rigo Silva, PhD3, H. M. Novaes, PhD4 and M. F. Almeida,

PhD1

1Faculdade de Saude Publica/School of Public Health, Sao Paulo, Brazil,2Faculdade de Saude Publica/School of Public Health, Sao Paulo, Brazil,3Universidade Estadual de Londrina (UEL), Londrina, Brazil, 4University of

Sao Paulo, School of Medicine, Sao Paulo, Brazil

INTRODUCTION: Preterm (PT) birth is associated to perinatal

morbidity and mortality. Its prevalence is increasing in many coun-

tries and the complexity of factors associated has been investigated.

METHODS: To analyze the effects of observed variables and latent

variables was used a conceptual framework via exploratory struc-

tural equations modeling and the confirmatory approaches in a

case-control study in Londrina/PR, Brazil (Jun/06 to Mar/07).

Continuous latent variables generated represent the constructs soci-

oeconomic vulnerability(SEV), family structure pattern(FSP) and

non-acceptance of pregnancy(NAP). The classification of prenatal

care was defined according to the time of the 1st appointment and

tests performed. PT births were evaluated through gestational age

(GA) as a continuous variable. The latent variables involved and the

full model generated were built and validated with the weighted

least square estimator in the software MPlus.

RESULTS: SEV comprises number of residents per rooms

(std.coef.:0.66; P<0.001), income per capita (�0.87;<0.001),

maternal education (0.84;<0.001), education of household head

(0.73;<0.001), place of residence (0.65;<0.001). FSP was formed by

family type (0.91;<0.001), having a partner <2 years

(0.72;<0.001), presence of elderly (0.76;<0.001), mother relation-

ship with household head (0.82;<0.001). NAP considers negative

reactions to the pregnancy of mother (0.81;<0.001), father

(0.84;<0.001), family (0.83;<0.001).

The effects on GA included direct effects of prenatal care

(�0.32;<0.001), pregnancy complications (�0.50;<0.001); NAP

(0.19;<0.03), alcohol use (�0.11;<0.01); multiparity (�0.25;

<0.01). There is also an indirect effect of NAP (0.20;<0.02) and

multiparity (�0.28;<0.001) was mediated through prenatal care.

SEV (0.12;<0.05) and FSP (0.17;<0.05) showed only an indirect

effect on GA mediated through prenatal care. It was found a direct

effect of NAP (0.11;0.05) and SEV (0.20;0.04) on the prenatal care.

CONCLUSIONS: SEM allowed dealing with latent variables and

also observed variables to understand the effects on the outcome.

FSP and SEV only express its effect mediated by prenatal care.

Poster Number: T82 Abstract #: 3293Effects of the Great East Japan Earthquake onPerinatal Outcomes.

K. Suzuki, PhD1, Z. Yamagata, PhD1, M. Kawado, PhD2 and S.

Hashimoto, PhD2

1University of Yamanashi, Chuo, Japan, 2Fujita health University School

of Medicine, Toyoake, Aichi, Japan

INTRODUCTION: Preterm birth rates, especially of female infants,

might increase among pregnant women exposed to strong stressors,

such as natural disasters, during early pregnancy. Additionally, male-

to-female birth ratio decreases among such pregnant women.

Therefore, this study aimed to examine effects of the Great East Japan

Earthquake on perinatal outcomes using vital statistics of Japan.

METHODS: Birth registration data from vital statistics of Japan

between March 2010 and March 2012 were used for analysis.

Pregnant women were categorized according to their gestational

period as of 11 March 2011, as follows: gestational months 2–3,

4–5, 6–7, and 8–9 (2011 group, n¼679,131). Similarly, pregnant


Dow



women categorized according to their gestational period as of 11

March 2010 were used as controls (2010 group, n¼688,479).

Moreover, we categorized prefectures as ‘extremely affected’ (Iwate,

Miyagi, and Fukushima), ‘moderately affected,’ and ‘slightly or

unaffected’ regions. Secondary sex ratio, birth weight, and gesta-

tional period were compared between both groups in each category

and region.

RESULTS: In the extremely affected region, male-to-female ratio

among the women in their 2–3 months of gestation significantly

declined in the 2011 group compared with the 2010 group (49.8%

vs 52.1%, P¼0.006). In the same region, the gestational periods

among the women in their 2–3 months of gestation were signifi-

cantly shorter in the 2011 group than in the 2010 group (274.6 days

vs 275.1 days, P¼0.02). Moreover, these effects were especially

greater in the Iwate prefecture than in the other 2 prefectures.

However, the differences between the 2010 and 2011 groups were

not significant in other regions.

CONCLUSION: This is the first study to analyze the effects of the

Great East Japan Earthquake on perinatal outcomes giving consider-

ation to seasonal patterns of perinatal indicators, and suggests that

these effects might be observed only in extremely affected areas.

Poster Number: T59 Abstract #: 3380Urbanity and Indicators of Mortality in Brazil’sCounties.

J. Moreira, MS, R. Luiz, PhD, J. Cerqueira, BA, P. Castro, BA,C. Muzi, MS, C. L. Cunha, MS and A. Brito, PhD

Federal University of Rio de Janeiro—UFRJ, Rio de Janeiro, Brazil

INTRODUCTION: Urbanity in Brazil often doesn’t express specif-

icities of each environment, whether considering the political-

administrative definition or dichotomous classification: urban and

rural. This study aims to present general mortality and the five lead-

ing causes of death in Brazil rates (cardiovascular diseases, neo-

plasms, external causes, respiratory, infectious and parasitic

diseases) by three levels of urbanity: small, medium-sized and large

counties.

METHODS: County level ecological study classified into three

groups considering population size and population density (PD):

small (<50.000 inhabitants [inh.] and PD <80 inh./km2), medium-

sized (between 50 and 100.000 inh. or PD>80 inh./km2 and

<100.000 inh.), and large (>100.000 inh.). We calculated counties

groups’ general and specific mortality rates. Data from the Health

System Computer Department (DATASUS – 2010).

RESULTS: As results, 80.1% of the counties were small counties,

14.8% medium-sized counties and 5.1% large counties. General

mortality rates were 581.5 [standard deviation – SD¼191,5]

(small), 591.3 [SD¼142,9] (medium-sized) and 596.4 [SD¼125, 5]

(large) per 100.000 inh. Circulatory and neoplasms diseases, the

two biggest causes of death in Brazil, showed rates of 172.2

[SD¼46,9] (large), 182.1 [SD¼63,7] (medium-sized), 184.3

[SD¼85,5] (small) per 100.000 inh.; and 98.8 [SD¼31,0] (large),

86.8 [SD¼37,1] (medium-sized), 91.7 [SD¼55,4] (small) per

100.000 inh., respectively. External causes mortality rates were 78.3

[SD¼26,5] (large), 72.9 [SD¼27,2] (medium-sized), 69.4

[SD¼36,4] (small) per 100.000 inh. The rates for respiratory

diseases were 64.9 [SD¼24.3], 57.9 [SD¼28,7], 62.6 [SD¼42.3]

per 100.000 inh., for counties large, medium and small sized,

respectively. Infectious and parasitic diseases had rates of 26.5

[SD¼9.5] (large), 23.7 [SD¼11,8] (medium), 28.5 [SD¼20.0]

(small).

CONCLUSIONS: Mortality caused by cardiovascular diseases, and

infectious and parasitic diseases were higher in counties classified as

small. In general, a gradient can be observed between the three levels

of urbanity in cardiovascular disease and external causes mortality.

The exception is the lower rates in circulatory disease among large

counties.

Poster Number: T81 Abstract #: 3118Poisoning Injuries among Alaska Native Peoplein Alaska, 2002–11.

H. D. Strayer, MPH, J. A. Craig, MPH, E. Provost, MD and E.Asay, MS

Alaska Native Tribal Health Consortium, Anchorage, AK

INTRODUCTION: During 1999–2005, poisoning was the third

leading cause of unintentional injury death for Alaska Native People

(AN) and suicide attempts by intentional poisoning remains a lead-

ing cause of injury hospitalizations. Current findings on poisoning

for AN and the associated data challenges are reported.

METHODS: Data were obtained from the Alaska State Trauma

Registry (ATR), the Alaska Bureau of Vital Statistics, and the Alaska

Department of Labor and Workforce Development for 1992–2011.

Injury frequencies, age-adjusted rates, and trends were calculated.

Significance was tested at the 95% confidence level using the z test.

RESULTS: From 2002 to 2011, 276 AN died from unintentional

poisoning: 51.4% from medications and drugs, 41.3% from alco-

hol, and 7.2% from other poisons. The AN statewide rate for poi-

soning deaths was 2.4 times higher than non-Natives (27.2 and 11.5

per 100,000, respectively, P<0.05). Reported rates for AN poison-

ing deaths were 4.6 times higher in 2008–11 than they were in

1992–95, a significant increase (8.6 and 39.6 per 100,000, respec-

tively, P<0.05). Intentional poisoning-related suicide attempts were

the second leading cause of AN injury hospitalizations (2,253) state-

wide. The rate of AN intentional poisoning-related suicide attempt

hospitalizations doubled between 1992–95 to 2008–10 (10.4 and

20.7 per 10,000, respectively, P<0.05).

CONCLUSIONS: Poisoning is a leading cause of AN injury morbid-

ity and mortality. Unintentional poisoning mortality rates and inten-

tional suicide-related poisoning hospitalization rates have both

increased for Alaska Native people over the past two decades.

Recent changes in diagnostic coding and reduced ATR reporting for

poisoning have altered data availability and comparability. Despite

data challenges, it was determined that poisoning causes a large pro-

portion of Alaska Native injuries. In order to monitor this issue and

evaluate interventions, it will be important to have surveillance sys-

tems that provide adequate data.

Poster Number: T75 Abstract #: 3287Use of Electronic Medical Records of theEmergency Department for an AutomatedEpidemiological Surveillance of AttemptedSuicide: Pilot Study in a French UniversityHospital.

N. Tvardik, MPH1, Q. Gicquel, MS1, T. Durand, MS2,V. Potinet-Pagliaroli, MD3 and M. H. Metzger, PhD1


Dow



1University of Lyon, Lyon 1, Villeurbanne, France, 2GCS SISRA, Lyon,

France, 3Hospices Civils de Lyon, Lyon, France

INTRODUCTION: The aim of our study was to assess whether an

extraction and automated processing of the computerized emer-

gency record would improve the estimate of the annual rate of emer-

gency department visits for attempted suicide, compared with the

annual rate produced in the framework of a national surveillance

currently carried out by manual coding of emergency physicians and

underestimating it.

METHODS: A feasibility study was conducted in the emergency

department of the Lyon University hospital (France) on the popula-

tion of patients admitted in 2011 and 2012 in the ward. After auto-

matic extraction and data preprocessing, including automatic

normalization of textual medical data through Unique Concept

Identifiers of the Unified Medical Language System, predictive asso-

ciation rules to classify the reason of the visit into « suicide attempts

» vs « other reasons » were developed. The performance of these

rules was evaluated by comparison with a gold standard (reading

the medical documents by medical practitioners).

RESULTS: In a test sample of 339 patients admitted to the emer-

gency in 2012 (99 admitted for attempted suicide, 14 with suicidal

ideation and 226 without any of these non fatal suicidal behavior),

the sensitivity for automatic detection varied from 94.9% [95% CI:

88.6%–98.3%] to 95.9% [95% CI: 90%–98.9%] and the specificity

between 96.5% [95% CI 93.1%–98.5%] and 97.8% [95% CI

94.9%–99.3%].

CONCLUSIONS: This study demonstrates the usefulness of devel-

oping semantic extraction and data mining methods to improve the

quality of epidemiological indicators produced as part of national

surveillance of suicide attempts.

Poster Number: T211 Abstract #: 3385Effects of the SAVA Syndemic-Related Factors onViral Suppression of HIV Women of Color.

L. C. Messer, PhD1, K. Sullivan, PhD2, E. B. Quinlivan, MD3, A.

E. Blank, PhD4, N. Verdecias, MPH4 and J. E. Fletcher, PhD4

1Portland State University, Portland, OR, 2Duke University, Durham, NC,3University of North Carolina, Chapel Hill, NC, 4Albert Einstein College of

Medicine, Bronx, NY

INTRODUCTION: Research suggests the interacting epidemics of

substance abuse, violence, HIV/AIDS and related health and psycho-

social factors create an excess burden in marginalized communities

(the SAVA syndemic). This study examines the relationships

between SAVA syndemic-related factors and viral suppression

among HIV-positive women of color (WoC) who have received

recent HIV care.

METHODS: Data are from the cross-site evaluation of the Health

Resources and Services Administration-funded Special Programs of

National Significance initiative designed to engage and retain WoC

in HIV care. A standardized multi-site baseline survey was employed

and matched with chart abstraction data. Women were included if

they had a viral load test within 90 days of the baseline survey

(implying the receipt of some level of HIV care). Generalized

estimating equations were used to explore associations between the

psychosocial/health variables on viral suppression (y/n) while

accounting for covariates and clustering of women by site. Models

included the dichotomous predictor variables frequent mental dis-

tress (�14 days of symptoms/month), substance abuse, binge alco-

hol use, sexual risk-taking behaviors, and intimate partner violence,

controlling for demographic variables.

RESULTS: Data for 563 HIVþ WoC were analyzed and just under

half (n¼260) were virally suppressed. Higher values on the SAVA

score (0 to 6) were associated with reduced risk of viral suppression;

risk ratio (RR)¼0.88, 95% confidence interval (95% CI): 0.81,

0.96. The observed association was slightly attenuated following

adjustment for covariates. Results from models including SAVA var-

iable interactions indicate several of the relationships are on the

multiplicative scale (P<0.10).

CONCLUSIONS: SAVA-related factors, in isolation and as part of

a syndemic scale, were negatively associated with viral suppression.

The presence of multiplicative/synergistic effects suggest that the

syndemic approach may be a viable framework for predicting HIV

clinical outcomes among this population.

Poster Number: T69 Abstract #: 2988Efficacy of Lung Cancer Screening: ACase-Control Study in Atomic Bomb Survivors.

K. Yokota, PhD1, M. Mine, PhD1 and Y. Shibata, PhD2

1Atomic Bomb Disease Institute, Nagasaki University, Nagasaki, Japan,2Fukushima Medical University, Fukushima, Japan

INTRODUCTION: While the randomized trial of the Prostate,

Lung, Colorectal, and Ovarian (PLCO) cancer screening trial in

United States suggested that annual lung cancer screening would not

decrease lung cancer mortality, case-control studies conducted in

several regions of Japan indicated the efficacy of annual screening.

We conducted a case-control study in the atomic bomb survivor

cohort to reduce the self-selection bias.

METHODS: A total of 448 lung cancer deaths were observed in the

cohort members who were in Nagasaki city during the observation

period, i.e. from 1 January 1989 to 31 Dec. 2007, who had opportu-

nities to undergo screenings during the period and who were alive as

of 1 Jan. 2000. Excluding 209 subjects whose age at death was 80

years or over and who had no information on smoking status, 239

subjects were identified as cases. For each case an attempt was made

to select three controls matched by sex, year of birth, smoking status

and radiation exposure status, and 713 subjects were selected as

controls. The odds ratios and 95% confidence intervals (CI) of death

for screened individuals versus unscreened before diagnosis of lung

cancer were calculated by the conditional logistic regression analyses

with adjustment by smoking index using PROC PHREG of the

SASVC .

RESULTS: The proportions of those screened within 12 months

before diagnosis were 12.6% and 23.6% in cases and controls,

respectively. Adjusted odds ratio was 0.56 (95% CI 0.35–0.88,

P¼0.012). The lung cancer mortality was significantly lower in

screened individuals than those unscreened. Furthermore, the pro-

portions of screened almost every year were 6.3% and 12.3% in

case and control, respectively. Adjusted odds ratio was 0.52 (95%

CI 0.27–1.00, P¼0.050).

CONCLUSIONS: The present case-control study in the Nagasaki

atomic bomb survivors cohort indicated reduction of lung cancer

mortality by 44% by annual lung cancer screening.


Dow



Poster Number: T163 Abstract #: 3097Using Theory to Develop a CulturallyAppropriate Family-Based Oral HealthIntervention in Brazil: the ‘Boca bocasaudavel’ Study.

A. M. Cascaes, PhD1, A. J. Barros, PhD2 and G. X. Ayala, PhD3

1Federal University of Pelotas, Pelotas, Brazil, 2Universidade Federal de

Pelotas, Pelotas, Brazil, 3San Diego State University, San Diego, CA

INTRODUCTION: Early childhood caries impacts on negative

quality of life in children and families, and it is a strong predictor of

caries in permanent dentition. In Brazil, a slightly reduction of caries

prevalence was observed in 5-years old children and 80% of their

teeth remain with cavities, which indicates oral health promotion

interventions. Our purpose was to design an oral health intervention

to prevent early childhood dental caries through family behavior

change that would be sustained by Primary Health Care, in Brazil.

METHODS: Intervention Mapping and Theoretical Domain

Framework were used to guide the intervention planning process.

Systematic literature reviews assessed current oral health interventions.

A population-based oral health survey with 1123 mother-child dyads

provided data on family oral health behaviors and children’s prevalence

of caries and dental plaque. In-depth interviews with 32 families eval-

uated their understanding of oral health meaning, beliefs, motivations

and barriers to adopt preventive behaviors. Six focus groups with health

providers, community health workers and families provided insights

into challenges and opportunities in reaching the audience and inputs

on program logistic aspects and opinions on educational materials in

order to increase program feasibility and acceptability.

RESULTS: Results from different research methods were combined

to identify relevant behavioral determinants and define the interven-

tion methods and strategies according to target population needs.

The intervention includes two components: a) seven group-based

wellness sessions, using a video series as the main communication

strategy, b) oral health campaign and printed educational materials

for parents and children.

CONCLUSIONS: We demonstrated how the planning protocol is a

helpful method for developing a feasible and theory-based oral

health intervention. This turns into a clear framework for process

evaluation, improving the intervention potential effectiveness. Our

study provides further evidence that can be used by others working

to reach this population in Latino communities.

Poster Number: T98 Abstract #: 3336Population-Based Surveillance and Prevalence ofDuchenne/Becker Muscular Dystrophies (DBMD)in the United States (U.S.).

P. A. Romitti, PhD

The University of Iowa, Iowa City, IA

Additional author list: S Puzhankara, G Zamba, Y Zhu, K James, J

Andrews, C Cunniff, E Ciafaloni, K Conway, C Druschel, K Mathews,

D Matthews, F Meaney, L Miller, S Pandya, S Au, S Scollon, M

Adams, N Street, and the Muscular Dystrophy Surveillance, Tracking,

and Research Network (MD STARnet)

INTRODUCTION: DBMD have an estimated prevalence of 1/3500

male births. Worldwide, estimates of DBMD vary, likely due to

differences in diagnostic criteria, ascertainment, and survival.

To date, no U.S., population-based prevalence data for DBMD by

race/ethnicity have been published, leaving potential health dispar-

ities in diagnosis and treatment of these disorders undetected.

METHODS: In 2002, the Centers for Disease Control and

Prevention established the MD STARnet to conduct longitudinal,

population-based surveillance of diagnosis and life course of DBMD

in the U.S. Six sites conducted active case finding and record

abstraction to identify males with DBMD born from January 1,

1982 through 31 December 2011. Using these data, we calculated

prevalence of DBMD by race/ethnicity and birth quinquennium

(1986–90; 1991–95; 1996–2000; 2001–05). Prevalence was calcu-

lated as (number of affected males age 5–9 years/number of male

residents age 5–9 years); the lower bound of this age range corre-

sponded to the average age at diagnosis. As such, census data used

to identify male residents per birth quinquennium were those avail-

able in a subsequent data release (e.g. 2010 estimates used for

2001–05 births).

RESULTS: Overall, 667 males with DBMD resided in an MD

STARnet site during one or more quinquennia from 1991 to 2010,

producing a prevalence of 1.8/10,000 males. For this time period,

racial/ethnic-specific prevalence was higher for Hispanics than non-

Hispanic Whites or non-Hispanic Blacks. A similar pattern for prev-

alence was found for each quinquennium examined.

CONCLUSIONS: We present the first U.S., population-based report

of racial/ethnic-specific prevalence of DBMD. Prevalence estimates

differed by race/ethnicity and persisted across quinquennia, suggest-

ing potential, persistent cultural and socioeconomic influences in

diagnosis of DBMD that require more detailed examination. Use of

our longitudinal, population-based surveillance approach permits

comprehensive ascertainment and follow-up of individuals with

DBMD to examine racial/ethnic differences in prevalence, treat-

ment, and outcomes for these individuals.

Poster Number: T144 Abstract #: 2986Positive Association between Breakfast Skippingand Incidence of type 2 Diabetes Mellitus:Evidence from a Japanese Worksite-BasedCohort.

M. Uemura, MS1, H. Yatsuya, PhD1, Y. Li, PhD1, C. Wang, MS1,

E. H. Hilawe, MPH1, C. Chiang, PhD1, H. Toyoshima, PhD2, K.

Tamakoshi, PhD3, Y. Zhang, MS1 and A. Aoyama, PhD1

1Nagoya University Graduate School of Medicine, Nagoya, Japan, 2Anjo

Kosei Hospital, Anjo-shi, Aichi-prefecture, Japan, 3Nagoya University

School of Health Science, Nagoya, Japan

INTRODUCTION: We examined the association between break-

fast skipping and type 2 diabetes mellitus (T2DM) incidence. We

explored whether the association would be independent of baseline

body mass index (BMI) and fasting blood glucose (FBG) levels, as

these could influence subjects’ dietary habits including breakfast eat-

ing behavior.

METHODS: Between 2002 and 2011, we followed 4,672 subjects

(3626 men, 1046 women) aged 35–66 years enrolled in the Aichi

workers’ cohort study. According to self-reported information, sub-

jects were classified into five categories of breakfast-eating fre-

quency: everyday, almost every day with occasional skips, 3-5 days,

1-2 days, and never, per week. Using multivariable adjusted Cox

proportional hazards models, we examined the association between

breakfast-eating frequency and T2DM incidence. Possible effect

modifications by sex, and baseline values of BMI (<25 kg/m2, 25


Dow



kg/m2 or more) and FBG (<110 mg/dL, 110 mg/dL or more) were

assessed by adding multiplicative interaction terms to the models.

RESULTS: In comparison to that of those who eat breakfast

‘everyday’, hazard ratios (HRs) and respective 95% confidence

intervals (95% CIs) for the risk of T2DM among subjects who eat

‘almost everyday with occasional skips’, ‘3–5 days’, ‘1–2 days’, and

‘never’ were 1.10 (0.76–1.60), 2.20 (1.26–3.83), 1.51 (0.89–2.55)

and 2.15 (1.20–3.87), respectively, in a maximally adjusted model

(P for linear trend <0.001). The association was attenuated to a

level of insignificance after further inclusion of insulin concentra-

tions in the model. Nonetheless, the association was not modified by

sex, and baseline values of BMI and FBG (P for interaction>0.05

for all).

CONCLUSIONS: We found a positive association of breakfast skip-

ping with T2DM incidence independent of dietary intake, lifestyle

factors, baseline levels of BMI and FBG, and other confounding fac-

tors in middle-aged Japanese workers. However, the association was

not independent of baseline serum insulin levels.

Poster Number: T70 Abstract #: 3228Encouraging Alternative Transport Use inAdelaide: Public Perception of Traffic,Environment and Health.

T. Xia, PhD1 and Y. Zhang, PhD2

1University of Adelaide, Adelaide, Australia, 2University of Sydney,

Sydney, Australia

INTRODUCTION: This study aimed to investigate the relationship

between participant’s attitudinal factors and their travel behaviors,

and to explore the predictors of participant’s intension to reduce car

use.

METHOD: A cross-sectional survey was conducted in metropolitan

Adelaide. Total 381 residents were interviewed using a computer-

assisted telephone interviewing (CATI) system. Statistical analyses

including Factor analysis, Pearson’s correlation, and multiple logis-

tic regressions were performed.

RESULTS: Four factors were yielded from the analysis, including:

(1) benefit awareness of alternative transport use; (2) problem

awareness of traffic; (3) safety and comfort; (4) negative emotion

towards public transport. We found factor 3 and 4 were positively

related to annual driving distance of the participants (r¼0.29,

r¼0.14). Both factor 1 and 2 were positively associated with accept-

ance of transport measures (r¼0.35, r¼0.13), whilst factor 3 and

factor 4 was negatively associated with it (r=�0.13, r=�0.12).

Participants who had high scores on benefit awareness (OR¼2.29)

and problem awareness (OR¼1.87) were more likely to shift travel

mode towards alternative transport, and those who had high score

on safety and comfort (OR¼0.41) would be less likely to change.

Furthermore, participants who drove over than 10,000 km would

be less likely to reduce private car use, compared to those who

drove less than 10,000 km. Moreover, bicycle usage was found as a

significant predictor of intension to reduce private car use

(OR¼6.46). Gender, age, annual house income and employment

status were not found to be significant predictors.

CONCLUSION: This study indicates that it is important to focus

on increasing public awareness of the health and environmental

co-benefits in promoting environmental travel behaviors. It also

suggests that among the set of available measures to reduce car use,

education approach would be more accepted among the public.

Poster Number: T159 Abstract #: 3398Validation of a Constraint Questionnaire againstwork Incident by Biological Fluid among theNursing Staff of a Public Hospital, Sao Paulo,Brazil.

K. dos Santos, MS1, R. D. Vilela, DrPH2 and M. R. A. Cardoso,

DrPH3

1Universidade de Sao Paulo, Sao Paulo, Brazil, 2Universidade de Sao

Paulo, Campinas, Brazil, 3Faculdade de Saude Publica—USP, Sao Paulo,

Brazil

INTRODUCTION: French-speaking ergonomists conceive constraint

as a work situation in which the worker has his power of acting, think-

ing and deciding reduced or hampered. This may facilitate the occur-

rence of work incidents or accidents, being an important issue among

nursing staff due to the risk of biological fluid exposure. This study

aimed at validating a constraint questionnaire among nursing staff.

METHODS: From February to September 2013, a nursing staff cen-

sus was carried out in a public hospital in Sao Paulo, Brazil. Four

items composed the constraint questionnaire: if the nurse in emer-

gency situation can carry out his/hers tasks according to technical

guidelines; if the nurse’s production is delayed by his/hers execution

of the assigned tasks with the necessary care according to technical

guidelines; if personal protection equipment (PPE) hampers the exe-

cution of a good job; and how often the nurse stopped using PPE.

The first question was dichotomous and the others included a Likert

scale, never (zero score) up to always (four scores). Work incident

by biological fluid was the gold standard, and its frequency was

categorized as: always, frequently, sometimes, seldom, never. Zero

value was assigned to the seldom and never categories, and one to

the others. R software was used to analyze the validity, using logistic

regression, gold standard versus the total sum of the four questions.

Cutoff, sensitivity and specificity were also analyzed.

RESULTS: Approximately 92%(n¼1648) of nurses were inter-

viewed. Incident’s prevalence was 35%(95% CI 33%-37%). The

mean for the constraint questionnaire was 3,25(95% CI 3,14–3,36).

OR was 1,20(95% CI 1,15–1,26), a one increase in the score of feel-

ing constraint was associated with an increase of 20% in the chance

of the nurse suffering an incident. Cutoff was seven scores, sensibil-

ity 14% and specificity 93%.

CONCLUSIONS: Constraint questionnaire for nursing staff was

valid, but its sharpness was better to identify nurses with low con-

straint that the inverse.

Poster Number: S67 Abstract #: 1562Homicidal Violence Related to Labor Activity:Unveiling the Silence of Statistics on Mortality inBelo Horizonte, Brazil.

T. Hang-Costa, BS1, H. N. F. Souza, BS1, J. Medeiros Silva, BA2

and E. Drumond, DrPH2

1Federal University of Minas Gerais, Belo Horizonte, Brazil, 2City

Department of Health and Welfare of Belo Horizonte, Belo Horizonte,

Brazil


Dow



INTRODUCTION: Homicide related to labor activity are the most

hidden of mortality statistics in Brazil. In Information System (SIM)

the most important Brazilian source of mortality data, it is impossi-

ble to notify the occurrence of work-related homicides. Objective:

Deterministic linkage of two data sources (SIM and Disease

Reporting System – SINAN, with 105 million records from SIM and

6.893 million records SINAN) are done to describe the profile of

workers victims of murder related to the work reported in both

information systems.

METHODS: Cross-sectional study of 47 homicides occurred among

workers in the period from 2008 to 2012 reported to (SINAN) from

Belo Horizonte, Brazil. Characteristics of the victims of the accident

and were analyzed: sex, age, race, education, place of occurrence,

causes accident, type of accident, treatment regimen and occupation.

RESULTS: The homicide victims were men (94%), aged>25 years

(85%), blacks (53%), �8 years of education (55%), dead in the

street (43%) especially for gun fire (75%). In 62% the homicide was

directly related to the labor activity of merchants, guards, bus driv-

ers, taxi drivers, company directors and officers. None of these cases

was reported in SIM as an occupational homicide.

CONCLUSIONS: The results raise discussions on the impact of

homicidal violence in worker health. Furthermore, draws attention

to the need to notify these cases the SIM as events related to the

work. The routine integration of different data sources can contrib-

ute to reduce the undercount of such occurrence and qualify infor-

mation regarding the risks associated to labor activity in Brazil and

conduct preventive approaches.

Poster Number: T58 Abstract #: 3226Warning from Fukushima? Radiation Dose andPrevalence of Thyroid Nodules has Positive andSignificant Correlation.

Y. Hamaoka

Keio University, Tokyo, Japan

INTRODUCTION: After the Fukushima–Daiichi nuclear power

plant disaster, thyroid screening was performed in subjects who

were aged �18 years. This study examined the relationship between

the number of participants with thyroid nodules and radiation dose,

using publicly available municipality level data (n¼25).

METHODS: Poisson regression was applied to that data. The num-

bers of participants with smaller nodule (<5 mm), with larger nod-

ule (>5.1 mm), and sum of them are explained by WHO thyroid

dose, year of screening dummy, average age of participants (or per-

centage of age group), fraction of early evacuation, and whether sta-

ble iodine was distributed.

RESULTS: The WHO thyroid dose had positive and significant

coefficients for three dependent variables. The z-value was larger for

smaller nodules (b¼0.017, z-value¼5.75, P¼0.000) than for

larger nodules (b¼0.009, z-value¼2.71, P¼0.007). Malignancy

(including suspected malignancy) was analyzed in the same manner,

the dose was insignificant. However, the number of nodules and

that of malignancy, including suspected malignancies, were positive

and significantly correlated.

CONCLUSIONS: Although this is an ecological study based on

aggregated secondary data, and causality should not be inferred, the

results might indicate an early warning for future incidence of thy-

roid cancer. A follow-up study is therefore necessary.

Poster Number: T178 Abstract #: 3396Epidemiology of Meningitis Caused byHaemophilus influenzae, Neisseria meningitidis,and Streptococcus pneumoniae in the NorthAmerican Arctic, 2000a 10.

P. Gounder, MD1, T. Zulz, MPH2, S. Desai, MD3, F. Stenz, MD4,K. Rudolph, PhD2, A. Li, MD5 and M. G. Bruce, MD6

1Centers for Disease Control and Prevention, Anchorage, AK, 2CDC,

Anchorage, AK, 3Public Health Agency of Canada, Ottawa, ON, Canada,4Government of Greenland, Nuuk, Greenland, 5Public Health Agency of

Canada, Ottawa, ON, Canada, 6CDC Arctic Investigations Program,

Anchorage, AK

INTRODUCTION: In the 1970s, the annual incidence/100,000

persons of meningitis caused by Haemophilus influenzae, Neisseria

meningitidis, and Streptococcus pneumoniae in indigenous Alaskans

was 62.8, 2.4, and 19.3, respectively, and in indigenous Canadians

was 46.3, 42.8, and 7.1, respectively; updated estimates for these

populations are unavailable. Therefore, we estimated the incidence

of H. influenzae, N. meningitidis, and S. pneumoniae meningitis in

the North American Arctic during 2000–10.

METHODS: Surveillance data for Alaska, Northern Canada, and

Greenland were obtained from International Circumpolar

Surveillance (ICS). We defined a case of bacterial meningitis caused

by H. influenzae, N. meningitidis, or S. pneumoniae as a culture-

positive isolate obtained from a normally sterile site in a resident

with a meningitis diagnosis. Population estimates were obtained

from national census statistics (Greenland does not record residents’

indigenous status). We assumed ICS identified all cases so 95% con-

fidence intervals (CI) are not presented.

RESULTS: The mean population in Alaska, Northern Canada, and

Greenland during the study period was 668,662 (19% indigenous),

136,921 (58% indigenous), and 56,550 (% indigenous unknown),

respectively. The annual incidence/100,000 persons for meningitis

caused by H. influenzae, N. meningitidis, and S. pneumoniae among

all North American Arctic residents was: 0.6, 0.5, and 1.5, respec-

tively; among North American Arctic children aged <2 years was:

15.4, 5.1, and14.4, respectively; among indigenous persons (in

Alaska and Canada) was: 2.1, 0.8, and 2.4, respectively.

CONCLUSIONS: The incidence of meningitis caused by H. influen-

zae, N. meningitidis, and S. pneumoniae among indigenous Alaskans

and Canadians declined since the 1970s, but remains higher compared

with all North American Arctic residents. Compared with published

reports on the incidence of meningitis caused by H. influenzae, N.

meningitidis, and S. pneumoniae in the overall US population during

2006–07, the incidence in North American Arctic residents was 7.5,

2.6, and 1.9-times higher, respectively.

Poster Number: T219 Abstract #: 1484Inequities in co-coverage of PreventiveIntervention for Children: Analyses of DHS datafrom low- and Middle Income Countries.

F. C. Wehrmeister, PhD1, M. C. Restrepo-Mendez, PhD2,G. V. Franca, MS1, A. D. Bertoldi, PhD1, A. J. D. Barros, PhD1

and C. G. Victora, PhD2

1Federal University of Pelotas, Pelotas, Brazil, 2Universidade Federal de

Pelotas, Pelotas, Brazil


Dow



INTRODUCTION: Surveys in low- and middle income countries

(LMIC) present separate estimates for coverage of preventive inter-

ventions for child survival, but it is also important to know how

many essential interventions each child is receiving, or ‘co-coverage’.

We aimed to assess inequities in co-coverage of key child survival

interventions.

METHODS: We analyzed 66 LMIC for which nationally represen-

tative data were available from Demographic and Health Surveys

(DHS). Only the most recent year for each country was considered

from 1994 to 2011. Adequate (but not perfect) co-coverage was

defined as having received at least six out of nine interventions

(BCG, polio and measles vaccination, vitamin A supplementation

and use of insecticide-treated bednet for children; tetanus toxoid for

mothers, antenatal care; skilled delivery and safe water supply), for

children aged 12–59 months. Socioeconomic inequality was assessed

through wealth quintiles based on asset indices.

RESULTS: The average number of interventions among all coun-

tries was 4.6 (3.8 in the poorest and 5.6 in the richest quintile). The

number of interventions increased according to the survey year

(from 2.7 in 1994–97 to 5.4 in 2009–11). The overall percentage of

children receiving 6þ interventions was 43%. Countries from

Central and Eastern Europe and the Commonwealth of Independent

States (34.2%) have lower coverage of 6þ interventions compared

to South Asia (51.5%). Analyses by wealth quintiles showed a linear

increase in percentage receiving 6þ interventions (mean co-coverage

was only 26.3% in the poorest and 63.8% in richest quintile).

CONCLUSIONS: Although many of the countries studied managed

to achieve high coverage level for isolated interventions, their per-

formance in delivering all – or at least most – preventive interven-

tions to all children was poor. Socioeconomic inequities in co-

coverage are evident. Public policies are needed to improve co-cov-

erage, especially in the poorer population.

Poster Number: T187 Abstract #: 3025Prevalence and Pattern of Drug-Drug Interactionsin the Critical Care Units of a Tertiary Hospital inAlexandria, Egypt.

S. M. Mohamed, MPH1, Z. M. Gad, DrPH2 andN. A. El-Nimr, DrPH2

1Alexandria Main University Hospital, Alexandria, Egypt, 2High Instituteof Public Health, Alexandria University, Alexandria, Egypt

INTRODUCTION: The complexity of the pharmacotherapy

involved in the simultaneous use of several drugs and various thera-

peutic classes makes critically ill patients at an increased risk for

DDIs. The objectives were to estimate the prevalence of DDIs in the

critical care units (CCUs) at a main tertiary hospital, to analyze their

clinical significance, onset, documentation and severity and to iden-

tify their possible determinants.

METHODS: Using a cross sectional design, 750 patients, admitted

to the CCUs, whose medical prescriptions contain four or more

drugs were included. A pre-designed structured questionnaire and a

record review sheet were used to collect the following data: sociode-

mographic, smoking habits, medical history, long term used medica-

tions, the presence of hospital-aquired infections, APAHE II score,

length of stay, organ impairment, number of drugs per prescription

and the number of prescribing physicians. Calculating the number

of interactions for each patient was performed. The list of drugs for

each prescription was analyzed using different software.

RESULTS: The prevalence of DDIs among patients admitted to

CCUs was 53.07%. The mean number of interactions occurred per

patient was 2.9861.91 interactions. The highest proportion of inter-

actions had a significance number 1.0, possible and suspected docu-

mentation, delayed onset and moderate severity. Age of the patient

and the number of prescribed drugs were the two independent

factors found to be significantly affecting the prevalence of DDIs.

CONCLUSIONS: Critically ill patients are at risk of DDIs and the

patients’ age and the number of drugs prescribed increases this

possibility.

Poster Number: T20 Abstract #: 3074Estimating the Crude Probability of Death due toCancer using French Registry Data.

N. Bossard, PhD1,2,3, L. Remontet, MS1,2,3, F. Binder-Foucard,PhD4,5, L. Daubisse, PhD5,6, A. Belot, PhD1,2,3,7 and H. Charvat,PhD1,2,3

1Hospices Civils de Lyon, Lyon, France, 2Universite Claude Bernard Lyon1, Villeurbanne, France, 3CNRS, UMR5558, Villeurbanne, France,4Registre des Cancers du Bas Rhin, Strasbourg, France, 5Reseau desRegistres Francais des Cancers Francim, Toulouse, France, 6CRLCCInstitut Claudius Regaud, Toulouse, France, 7Institut de Veille Sanitaire,Saint Maurice, France

INTRODUCTION: Though net survival is an important epidemio-

logical tool allowing temporal or geographical comparisons, it can-

not give information on the ‘real’ survival of cancer patients. Indeed,

net survival is the survival that would be observed if cancer were the

only possible cause of death. In this work, we provide estimates of

the ‘real’, or ‘crude’, probabilities of death due to cancer and due to

other causes as well as the probability of being alive up to ten years

after diagnosis according to the age and year of diagnosis.

METHODS: We used data from the French network of cancer regis-

tries and studied five common cancers: head and neck, breast, prostate,

lung and colorectal cancers.Crude probabilities for a particular cause

were obtained by combining the hazard specific of that cause and the

overall survival. Cancer-specific hazard was estimated from a flexible

excess hazard model, also known as relative survival model, with

departement, age and year of diagnosis as covariates. Other-cause-spe-

cific hazard was obtained from general population life tables.

RESULTS: For breast, prostate and colorectal cancer, the impact of

other causes of death on the total probability of death becomes

more important as the age at diagnosis increases while it stays negli-

gible for lung and head and neck cancer whatever the age at diagno-

sis. For breast, prostate and colorectal cancer, there is a strong

decrease in the probability of death due to cancer for patients diag-

nosed more recently.In elderly patients with prostate cancer, few

patients die from their cancer although the cancer-specific hazard is

high; indeed, they die from other causes.

CONCLUSIONS: The crude probability of death is an intuitive con-

cept that may prove particularly useful in the appraisal of the real

impact of the cancer as well as in the choice of an appropriate treat-

ment or in the indications of a screening strategy by allowing the

clinician to estimate the proportion of cancer patients who really die

from their disease.


Dow



Poster Number: T158 Abstract #: 1658Characterization of Industrial Dusts to EstablishOccupational Health Hazards on Exposure in TwoIndustries in Nigeria.

I. A. Atting, PhD

University of Uyo/University of Uyo Teaching Hospital, Uyo, Akwa Ibom

State—Nigeria, Uyo, Nigeria

INTRODUCTION: Characterization of selected industrial dusts in

terms of physico-chemical and microbial pollutants to establish

occupational health hazards on exposure was carried out. The level

of conformity to Occupational Health Hazards Management has

been determined for Niger Flour Mill, Calabar (41.2%) and

Champion Breweries PLC, Uyo (25.5%).

METHODS: The total concentrations of inhalable dust were meas-

ured using PM10 particulate monitor as preliminary hazardous

index. The exposure sampling was conducted for duration of eight

hours shift and concentrations of respirable dust were determined

using SKC Sidekick pump aspirated at rate of 2.2 l/min.

RESULTS: The total concentrations of inhalable dust measured at

the two industries were 2.030 mg/m3 for Niger Flour Mill, Calabar

and 1.530 mg/m3 for Champion Breweries, Uyo. The concentrations

of respirable dust determined for Champion Breweries PLC, Uyo

(0.7mg/m3) and Niger Flour Mill, Calabar (1.1 mg/m3). The total

concentration of detected heavy metals was 7.37 mg/m3 for the first

and 4.70mg/m3 for the second against FEPA (1991) limit of 10.0

mg/m3. The cumulative exposure to heavy metals could result in

renal vascular disease, diabetic nephropathy, glomerulonephritis,

pneumoconiosis, skin and lung cancers, immune suppression and

other respiratory illness. The common bacteria isolates were

Staphylococcus aureus, 2(14.2) and Esherichia coli, 2(14.2) known

for blood poisoning and immune suppression while fungi isolate,

Aspergillus spp, indicates possibility of mycotoxin infection, myco-

toxicosis and respiratory diseases on prolonged exposure.

CONCLUSIONS: The contaminants detected were below

Occupational Exposure Limit for full blown hazards. The conse-

quences of accumulative effects in 10–60 years cannot be ruled out

for respiratory, dermal and optical (especially ptyregium) complica-

tions to occur. The research results have shown that no industrial

operation can exist at zero hazards unless employees are off the site

and operations shut—down.

Poster Number: T127 Abstract #: 3328Who are Obese? Prevalence and LifestyleCharacteristics Associated with Normal-WeightObesity in the National FINRISK 2012 Study(Finland).

S. Mannisto, PhD, K. Harald, N. Kanerva, K. Borodulin, PhD, N.

Kaartinen and P. Jousilahti, PhD

National Institute for Health and Welfare, Helsinki, Finland

INTRODUCTION: A syndrome called normal-weight obesity

(NWO) has recently been identified among normal weight subjects

(BMI<25 kg/m2) whose body composition is not favorable to

health. The aim was to compare the lifestyle characteristics (age,

education, leisure-time physical activity, smoking habits and alcohol

consumption) of NWO subjects with lean and overweight subjects.

METHODS: The cross-sectional National FINRISK 2012 Study

with a health examination and questionnaires included 5827 sub-

jects (25–74 years). The participation rate was 59% for the health

examination including, e.g. anthropometric measures. Together,

390 participants (7%) were considered as non-eligible because of

missing measures or pregnancy. Consequently, the final data

included 2617 men and 2820 women. NWO was defined as

BMI<25 kg/m2 linked with different cut-points of body fat by gen-

ders, for men (>20%) and women (>30%). Three weight catego-

ries were obtained called as lean, NWO syndrome and at least

overweight. The multivariate models were adjusted for obesity

related factors (F-test, Wald Chi-Square, p-value<0.05).

RESULTS: The prevalence of normal weight subjects was 33% in

men and 47% in women. Of those, 29% of men and 40% of women

were NWO (among all subjects: 10% and 19%, respectively). In all,

0.2% of NWO men and 9% of women had abdominal obesity

(waist>102 cm in men and>88 cm in women). NWO men were as

physically inactive as overweight men whereas their smoking habits

and alcohol consumption reminded more lean men. In NWO

women, the prevalence of being physically inactive fell between lean

and overweight women. They also consumed the most alcohol com-

pared with the other women.

CONCLUSIONS: From the public health perspective, the propor-

tion of NWO subjects was surprisingly high. The identification of

them may be of importance because although they appear lean, they

have some unhealthy lifestyle related to obesity and overall health.

Poster Number: T217 Abstract #: 3082European Comparative Cohort Study of SocialInequalities in Child Health and Development:Findings from the DRIVERS Birth Cohorts.

M. Ruiz, MS, M. G. Marmot, PhD, P. Goldblatt, PhD, J.

Morrison, MPH and H. Pikhart, PhD

University College London, London, United Kingdom

INTRODUCTION: Healthy growth in early childhood has become

a major priority to reduce health inequalities across Europe. It is

now increasingly recognised that the first years of life may provide

children with lasting benefits throughout the life course.

Investigation of social inequalities in early childhood health has

important implications for the health of future generations, but it is

not readily known how inequalities in early childhood health and

development vary across the region of Europe.

METHODS: Longitudinal data from thirteen birth cohorts across

the European region participating in the DRIVERS for Health

Equity – Early Childhood Development research programme were

used. Mother’s level of education was collected at the time of birth

and health and confounding data were measured prospectively at

subsequent follow-ups to the age of 8. Regression models were esti-

mated within each cohort and meta-analyses were conducted to

assess the extent of heterogeneity between regions of Europe.

RESULTS: The association between low maternal education and

increased risk of poor child health were not entirely consistent

across Europe. Social inequalities in preterm births, estimated by the

Relative Index of Inequality (RII), were generally unobserved in

Southern and Central Europe, but were strong in the Netherlands,


Dow



Sweden and the UK: 1.72 (1.15–2.60), 1.57 (1.14–2.15), and 1.49

(1.07–2.01). A similar pattern arose for children born small for ges-

tational age. Social inequalities in asthma were identified in the

Western countries alone, namely France, 2.01 (1.09–3.71), the

Netherlands, 1.65 (1.01–2.72) and the UK, 1.71 (1.42–2.06).

Overweight and obesity among preschool children were inversely

associated with high maternal education in Northern and Western

Europe, but gradients were steeper in the West.

CONCLUSIONS: This work highlights the value of comparative

cohort analysis in early childhood to better understand how social

inequalities in health emerge in different settings across Europe.

Poster Number: T162 Abstract #: 3274Work Limitation due to Low Back Pain amongTobacco Farmers in Southern Brazil.

R. D. Meucci, MS, A. G. Fassa, PhD, N. M. Faria, PhD,

N. S. Fiori, MS, V. I. Miranda, MS and D. P. Resende, BS

Federal University of Pelotas, Pelotas, Brazil

INTRODUCTION: Farm work require high physical effort and

awkward postures, which are risk factors for low back pain and

work limitation. Low back pain, particularly chronic pain, is

responsible for a high number of sick leaves. In Brazil, there are no

studies about work limitation due to low back pain among farmers.

METHODS: This paper, through a cross-sectional study carried out

in a random sample of Brazilian tobacco farmers, assesses the preva-

lence of work limitations due to chronic low back pain (CLBP), low

back pain in the last month (LBPLM) and acute low back pain

(ALBP), as well as the tasks the farmers have to stop doing.

RESULTS: The prevalence of work limitation due to CLBP, LBPLM

and ALBP were, respectively 37,6%, 14,4,% e 7,8%. Carrying

leaves, stacking firewood and harvesting bottom leaves were the

tasks that tobacco farmers have to stop doing more frequently. In

the adjusted analysis, age, road tax expenses, difficulty in paying

debts and minor psychiatric disorders were associated to work limi-

tation due to LBPLM.

CONCLUSIONS: LBPLM is inappropriate for acute pain measure-

ment, as 1/5 of the individuals with LBPLM have CLBP. Individuals

with ALBP stopped doing the same tasks than CLBP individuals, but

CLBP individuals have higher work limitations. Low back work lim-

itation studies should discriminate acute from chronic pain.

Poster Number: T22 Abstract #: 3728Trend of Lung Cancer Mortality in Brazil,1980–2011.

R. M. Guimaraes, PhD1, C. Muzi, MS2, R. D. Santos, BS1 and

F. L. Gomes, BS1

1Federal University of Rio de Janeiro, Rio de Janeiro, Brazil, 2Federal

University of Rio de Janeiro—UFRJ, Rio de Janeiro, Brazil

INTRODUCTION: Given the major threat of smoking on public

health, due to the large number of smokers and the growth of free

marketing of tobacco, Brazil, in 1996, enacted the Federal Law

9.294/96, which restricts the tobacco using and advertising products

in collective, public or private places.

OBJECTIVES: This study aimed to describe the pattern of distribu-

tion of mortality from lung, bronchial tubes and trachea cancer, and

assess the evolution of this trend, stratified by sex, in Brazil, regions

and federal states in the period 1980–2011.

METHODS: A descriptive, time series, based on secondary data

obtained from the Mortality Information System (SIM) was per-

formed. We calculated the standardized mortality rate for both sexes

and to minimize the difficulty in performing the trend analysis

caused by the presence of cyclical fluctuations, we used two techni-

ques to reduce variability shown in the series: a polynomial regres-

sion and joinpoint.

RESULTS: By analyzing the results, we observed a decreasing trend

among males in Brazil (fall 0.6% a year from 1995, P<0.001), in

most states and regions, a few growth however, most to fluctua-

tions. Among women, an increasing trend in Brazil (2.4% increase

per year since 1980), was observed, indicating a homogeneous

growth.

CONCLUSION: It is known that tobacco is the major risk factor

for lung cancer and, therefore, the high prevalence of smoking is

strongly associated with high mortality rates from cervical cancer.

Effective to minimize smoking and to reduce initiation among

youth, with emphasis on women, actions should be seen as a priority

and are actually necessary to reduce the incidence and mortality of

lung cancer.

Poster Number: T168 Abstract #: 2941Prevalence of Fissure Sealants in PortugueseAdolescents.

N. J. Veiga, MPH1, C. M. Pereira, PhD2, O. P. Amaral, PhD2,C. Chaves, PhD2, P. Nelas, PhD2, M. Ferreira, PhD2, I. J. Correia,PhD3, P. C. Ferreira, PhD4, J. Rocha, MS5 and S. Teixeira, MS5

1Universidade Catolica Portuguesa, Viseu, Portugal, 2CI&DETS, Viseu,Portugal, 3Research Centre – Health Sciences Faculty. Beira InteriorUniversity, Covilha, Portugal, 4CIEPQPF, Chemical EngineeringDepartment, University of Coimbra, Coimbra, Portugal, 5School Group ofSatao, Satao, Portugal

INTRODUCTION: Fissure sealants is an effective intervention for

the primary prevention of dental caries among children and adoles-

cents. The aim of this study was to assess the prevalence of fissure

sealants in Portuguese adolescents.

METHODS: A cross-sectional study was designed with a sample of

447 adolescents aged 12 to 19 years old, attending a public school

in Satao, Portugal. A self-administered questionnaire with questions

about oral health behaviours, knowledge and socio-economic status

was answered by the adolescents in the classroom. Clinical examina-

tion of oral health status and assessment of fissure sealants was

accomplished by a trained and calibrated research team. Prevalence

was expressed in proportions and compared by the chi-square test.

Crude odds ratio (OR) with 95% confidence intervals (CI) were

used to measure the strength of association between the presence of

fissure sealants and the independent variables.

RESULTS: We verified that 58.8% of adolescents had at least one

fissure sealant applied. From our total sample, we observed 830

teeth sealed, in which 63.3% were completely intact, 11.3% infil-

trated but still without dental caries, 1.6% infiltrated but with the

presence of dental caries, 23.5% with a partial fissure sealant but

without dental caries and only 0.3% with a partial fissure sealant

with dental caries present on the tooth surface. Significant statistical

differences were found when analyzing the presence of fissure seal-

ants related with parents’ educational level (>9th grade, OR¼1.56

95% CI¼1.05–2.54), gender (female, OR¼1.86 95%


Dow



CI¼1.19–2.98), experience of dental pain (yes, OR¼0.62 95%

CI¼0.39–0.97) and presence of dental caries (yes, OR¼0.35 95%

CI¼0.19–0.65).

CONCLUSIONS: It is essential the establishment of a more targeted

primary preventive program with better and more effective oral

health education, which will increase the application of fissure

sealants. Oral health interventions need to have in account socio-

demographic aspects, having special attention on adolescents and

families with a lower socio-economic status.

Poster Number: T83 Abstract #: 2947Epidemiology of Injuries among Elderly in anUrban Kazakhstani Setting: A Descriptive Study.

A. Tlemissov, MD1, T. Bulegenov, PhD1, A. Myssayev, PhD1, Y.Manarbekov, MD1, A. Bukatov, MD1 andA. M. Grjibovski, PhD2,3

1Semey State Medical University, Semey, Kazakhstan, 2Northern State

Medical University, Arkhangelsk, Russia, 3Norwegian Institute of Public

Health, Oslo, Norway

INTRODUCTION: The incidence of injuries in Kazakhstan greatly

exceeds the corresponding numbers from most developed countries

representing a serious public health problem. Elderly in Kazakhstan

face increased risk of injuries due to increasingly active life style, more

traffic, less social support during the period of transition combined

with gradual decrease in motor and cognitive functions. However, lit-

tle is known about the epidemiology of injuries in Kazakhstan. The

aim of this descriptive study is to present epidemiology of injuries

among elderly in a typical urban Kazakhstani setting.

METHODS: All injuries occurred in individuals aged 60 years or

older between 1 January 2010 and 31 December 2012 and regis-

tered in the medical facilities of the Semey municipality were

included. We studied distribution of injuries by causes, localization,

nationality, gender, and where the patients got injured

RESULTS: Altogether, there were 6065 injuries (4065 or 67.0%

among women) during the three years. The age ranged from 60 to

108 years. By cause, most injuries were a result of falls (W00–W19,

82.2%) followed by exposure to inanimate mechanical forces

(W20–W49, 7.7%) and injuries of pedestrians in traffic accidents

(1.5%). Categories S50–S59 (injuries to the elbow and forearm),

S40–S49 (injuries to the shoulder and upper arm) and S80–S89 (inju-

ries to the knee and lower leg) comprised 17.3%, 15.5% and 14.6%

of all injuries, respectively. Fifty percent of all injuries occurred

among ethnic Russians while 47.4% of injuries occurred in ethnic

Kazakhs. Most injuries (51.9%) occurred at home, on the street

(39.4%) and in yards next to home (5.2%).

CONCLUSIONS: This descriptive study is among the first present-

ing the epidemiology of injuries in urban Kazakhstani setting. The

results can be used for development of preventive strategies particu-

larly addressing the risk of falls with special emphasis on women

and home environment.

Poster Number: T25 Abstract #: 3729Trends of Colorectal Cancer Mortality in Braziland Regions According to Sex, 1980–2011.

R. M. Guimaraes, PhD1, V. G. Parreira, MS1 andO. G. Cruz, PhD2

1Federal University of Rio de Janeiro, Rio de Janeiro, Brazil, 2OswaldoCruz Foundation, Rio de Janeiro, Brazil

INTRODUCTION: Colorectal Cancer is among the four most com-

mon cancers in Brazil, being, therefore, a serious public health prob-

lem. The study of their tendency helps to understand its magnitude,

vulnerability and transcendence.

Objective: The aim of this study is to describe the distribution pat-

tern of mortality from colorectal cancer in Brazil and regions, by

gender, in the period between 1980–2011. I

METHODS: t was used, for trend analysis, the values of the stand-

ardized mortality rates stratified by sex between 1980 and 2011,

through polynomial regression models.

RESULTS: It was observed that there are for trends in mortality

rates for both sexes, a statistically significant linear trend of increase

in both models, with the highest rates and trends for men than for

women, respectively, 10.5% year (R2¼0.91, P<0.001) and 6.8%

per year (R2¼0.83, P<0.001). Still, we observed a difference in

the magnitude and trend of behavior between regions, with the

Northeast and North systematically higher rates in females, in South

and Southeast systematically higher rates for males, and for Brazil

and Midwest are no fees higher among women than among men

until 1996, when there is a change in this relationship for sex.

CONCLUSION: It is understood, therefore, that it is important to

know the trend of rates, especially by sex and age group, in order to

define priority populations for early interventions that improve sur-

vival and reduce mortality.

Poster Number: T114 Abstract #: 3372A Brief Assessment of Neighbourhood Effect onNeonatal Mortality: Translation of Area LevelVariance in the Odds Ratio Scale in MultilevelLogistic Regression.

B. Thakur, PhD1, V. Sreenivas, PhD2, S. Dwivedi, PhD2 andA. Pandey, PhD3

1All India Institute of Medical Sciences, New Delhi, India, 2All IndiaInstitute of Medical Science, New Delhi, India, 3Indian Council ofMedical Research, New Delhi, India

INTRODUCTION: Logistic regression is frequently used in epide-

miological and public health research to measure the binary out-

come. Unlike linear regression analysis, logistic regression does not

agree the interpretational attribute of the normal model. It is always

contend the dissatisfactory of existing measures while quantifying

results from multilevel logistic regression model. The variability at

different levels is not directly comparable. Quantifying area-level

variance in a meaningful way is a challenge in multilevel logistic

regression

METHODS: We obtained individual and district level information

on the binary outcome neo-natal mortality from District Level

Household Survey-3. The exploration of data structure confesses the

consideration of only two-level structure in analysis, conceptualized

as children nested within districts. Estimations of Variance

Component Model (empty model) and Random Intercept Model in

multilevel logistic analysis were carried out. The latent variable

method converts the individual level variance from the probability

scale to the logistic scale to compute the intra-class correlation. The

median odds ratio translates the area level variance on the odds ratio

scale.


Dow



RESULTS: The median odds ratio was equal to 1.60, in the empty

model which shows if a person moves from one district to another

district with a higher probability of neonatal mortality, their risk of

mortality (in median) will increase by 1.6 times, when randomly

picking out two persons in different districts. After adjusting the

individual effect in the random intercept model, this ratio reduced to

1.54. Area level variance and Intra-class correlation were 0.246 and

0.067 in the empty model as well as 0.210 and 0.059 in the subse-

quent model, respectively.

CONCLUSIONS: The usual odds ratio are not proper interpretable

for district-level covariates because it is impossible to make compari-

son within district. As MOR quantifies cluster variance in terms of

odds ratios, it is comparable to the fixed effects odds ratio and can

be useful in epidemiological studies.

Poster Number: T40 Abstract #: 3369A Case-Control Study to Detect Genetic andAcquired Risk Factors for Pediatric InflammatoryBowel Disease.

G. Kobashi, MD1,2, A. Hata, MD2, K. Uchida, MD3, T. Ishige,MD4, D. Abukawa, MD5, H. Tajiri, MD6, K. Uchiyama, MD7, Y.Hirota, MD8, M. Nagai, MD9 and T. J. P. I. B. D. researchGroup2

1National Institute of Radiological Sciences, Chiba, Japan, 2ChibaUniversity Graduate School of Medicine, Chiba, Japan, 3Mie UniversityGraduate School of Medicine, Tsu, Japan, 4Gunma University GraduateSchool of Medicine, Maebashi, Japan, 5Miyagi Children’s Hospital,Sendai, Japan, 6Osaka General Medical Center, Osaka, Japan, 7The JikeiUniversity Kashiwa Hospital, Kashiwa, Japan, 8Osaka City UniversityFaculty of Medicine, Osaka, Japan, 9Saitama Medical University,Moroyamamachi, Japan

INTRODUCTION: Pediatric inflammatory bowel disease (PIBD) is

considered to be a multifactorial disease with both genetic and

acquired factors involved in its etiology. Proposed acquired factors

include lifestyles and environmental factors of both patients in child-

hood and their mothers in perinatal period. However, to date, few

studies have been examined these factors simultaneously and clari-

fied their confounding factors. This paper reports preliminary results

of a case-control study which aims to elucidate genetic and acquired

risk factors for PIBD and their confounding factors.

METHODS: PIBD cases and controls were recruited from affiliated

hospitals of the Japan pediatric inflammatory bowel disease research

group. Saliva sample of patients for genotyping and self-adminis-

trated questionnaire for their mothers were obtained with written

informed-consent.

RESULTS: Sixty-five PIBD patients, including 43 ulcerative colitis

(UC) and 22 Crohn’s disease (CD), and 62 controls were recruited

to the study. An analysis of family history, gestational factors and

past history revealed that family history of IBD and past history of

surgical operation showed a positive association with CD, while no

significant associations were found in parental smoking, birth

weight and breast-feeding.

CONCLUSIONS: The present study is expected to develop early

and individualized measures to prevent PIBD, intervention for life-

styles and environmental factors of expectant mothers possessing

genetic risk factors for baby’s future PIBD manifestation.

Furthermore, the results may contribute to clarify new pathogenesis

of PIBD manifestation and to produce more useful disease classifica-

tion of PIBD.

Poster Number: T151 Abstract #: 3386Prevalence of Inadequate Dietary Intake for OlderPeople in a Medium Size City of Sao Paulo State,Brazil.

J. E. Corrente, PhD1, L. B. de Souza, MS2 and S. J. Papini, PhD3

1Bioscience Institute—University of Sao Paulo State—UNESP, Botucatu,Brazil, 2Medicine Faculty—Federal University of Uberlandia, Uberlandia,Brazil, 3Botucatu Medical School—University of Sao Paulo State—UNESP, Botucatu, Brazil

INTRODUCTION: A purpose of dietary assessment is to evaluate

the dietary intake of a group or a population. In Brazil few studies

have been carried out to identify dietary intake for older people pop-

ulation. Then, the aim of this work is to estimate the distribution as

well as estimate the prevalence of inadequacy of the usual intake in

a representative sample of older population from Botucatu city, Sao

Paulo.

METHODS: A sample of 365 older was used and some instruments

to evaluate quality of life, activities of daily living and instrumental

of daily living were applied as well as three 24-hours recall. Data

from the recalls were transformed in consumption of macro and

micronutrients using Nutrition Data System for Research (NDSR)

software and were analyzed using National Cancer Institute method

through the developed macros MIXTRAN and DISTRIB for

Statistical Analysis System (SAS) v.9.3 in order to estimate the inad-

equacy prevalence. The values of Estimate Average Reference (EAR)

or Adequate Intake (AI) were used as a cut-off to estimate de

inadequacy.

RESULTS: The energy and macronutrients intake of the studied

population agree with their needs, however the consumption of min-

erals and vitamins were below form the recommendation even after

including the covariates.

CONCLUSIONS: This can reflects a monotone intake that is char-

acteristic of this age and these inadequateness can be a serious public

health problem that can carry on development of chronic diseases.

Also, It is important to highlight that NCI method provide a good

estimate of the usual intake.

Poster Number: T150 Abstract #: 3416Evaluation of Sodium in Adolescents: APopulation-Based Study in Campinas, SaoPaulo, Brazil.

M. Messaggi Dias, DO1, A. A. Barros Filho, PhD2, D. deAssumpcao, MD3, M. Barros, PhD4, R. M. Fisberg, PhD5 andM. C. Alves, PhD6

1State University of Campinas—UNICAMP, Brazil., Campinas, Brazil,2State University of Campinas—UNICAMP, Brazil, Campinas, Brazil,3State University of Campinas—UNICAMP, Brazil., Campinas, Brazil,4State University of Campinas—UNICAMP, Campinas, Brazil, 5Universityof Sao Paulo, Sao Paulo, Brazil, 6State Department of Health of SaoPaulo, Institute of Health, Sao Paulo, Brazil

INTRODUCTION: The excess of sodium chloride in food is related

to hypertension and cardiovascular diseases, which are among the

leading causes of death in the adult Brazilian population. The aim was

to evaluate the association between the 4th quartile of the distribution

of sodium intake compared to the 1st quartile of the distribution of

this nutrient, in adolescents aged from 10 to 19 years-old, according

to demographic, socioeconomic and health-related behaviors.


Dow



METHODS: The data obtained from the ‘Health Inquiry of

Campinas’, a population-based survey. The study analyzed 924 ado-

lescents of both sexes—obtained by the population sample stratified

by clusters. The 24-hour dietary recall provided the information nec-

essary to evaluate the intake of sodium. We used linear regression to

evaluate the association between the 4th quartile of the distribution

of sodium intake compared to the 1st quartile of the distribution of

this nutrient and independent variables.

RESULTS: The sample studied the highest proportion in males

(68.5%) and aged from 15–19 years (55.2%). The prevalence of

inadequacy according to the Maximum Tolerable Unit for the popu-

lation was 81.65% (95% CI 78.32–84.57%). The average nutrient

intake was 3495.13 mg (95% CI 3357.75–3632.5). The main find-

ings were: a prevalence of inadequate sodium intake significantly

higher in male adolescents (P¼0.000), in which reported that they

had up to five numbers of equipment at home (P¼0.03), possess

shack/room as characterization of domicile (P¼0.02) and adoles-

cents working (P¼0.01). Furthermore, we found a higher preva-

lence of lower sodium intake among adolescents who reported

consumption of soft drinks seven days a week (P¼0.01), from 1-4

times a week consumption of alcoholic beverages (P¼0.0009).

CONCLUSIONS: The findings revealed a high sodium intake

according to socioeconomic variables, pointing to a need for atten-

tion to socially vulnerable segments and signaling the relevance of

global orientation.

Poster Number: T102 Abstract #: 3271Prevalence of Internet Use and InternetAddiction Disorder among Medical Students:A Case from Low Income Country.

S. Awan, MS

Aga Khan University hospital, Karachi, Pakistan

INTRODUCTION: Prolonged use of internet caused a series of

problems such as internet dependence, problematic internet use,

compulsive internet use, internet abuse and Internet addiction disor-

der (IAD), This has aroused attentions of researchers all over the

world, with IAD being formally recognized as a mental disorder.

The aim of this study is to investigate the prevalence and risk of

Internet Addiction Disorder (IAD) and the associated predictive fac-

tors amongst medical students in Karachi, Pakistan This was a pro-

spective study.

METHODS: A cross-sectional study was conducted among 412 stu-

dents at different medical colleges in Karachi. For gathering data a

questionnaire was used. For measuring internet addiction the

Internet Addiction Test tool was applied. This test was developed by

Yung (1998). Young defined 20–39 points as an average user,

40–69 points as a possible or intermittent user, and more than 70

points as an addicted user. The subjects’ psychological well-being

were assessed by self-reporting questionaire-20 (SRQ20).

RESULTS: Internet Addiction test showed that 74% of the sample

could be classified as average internet users, 24% as problem over-

users and 2% as addicted to the internet. The score of psychological

morbidity was higher in internet addicted or problem over-users as

compare to average user (P value<0.001).over-users and addicts

spending increasingly more time in online activities, being more

socially anxious, and gaining greater support from internet social

networks (P value¼0.003) and online chatting (P value¼0.01)

than average internet users.

CONCLUSIONS: Internet addiction is prevalence among students

and Key health messages and interventions to reduce stress and anxi-

ety among students may help in curtailing the burden of this disease

which has serious adverse consequences.

Poster Number: T109 Abstract #: 3292Evaluation of the Modified HIV ProximateDeterminant Framework in Zimbabwe usingStatistical and Geo-Statistical Methods.

G. O. Atilola, MS1 and P. Nyasulu, MS2

1Redeemer’s University, Ogun, Nigeria, 2Monash University, South Africa

INTRODUCTION: The HIV proximate determinant framework is

a conceptual framework for the study of distribution and determi-

nant of infection in population. A modification to this framework

was developed by Till and Tanser taking into account the role of the

community. This study aimed to evaluate the modified proximate

determinant framework in Zimbabwe using statistical and geo-

statistical methods.

METHODS: This study used the Zimbabwe demographic and

health survey of 2010/2011 which included a representative sample

of 9171 women and 7104 men- aged 15 to 49. Following the proxi-

mate determinant framework, mixed effect binary logistic regression

models were fitted. Spatial clustering was assessed using Kuldorrff

spatial scan technique. Spatial and non-spatial random geoadditive

Bayesian models were then fitted. Model fit was assessed using

the deviance information criterion and Bayesian information

criterion.

RESULTS: Prominent significant underlying risk factors of HIV

infection were age group and widowhood adjusting for proximate

determinants. Total lifetime partners and symptoms of sexually

transmitted infections emerged as significant proximate determi-

nants of infection. Evidence of global and local spatial autocorrela-

tion was found. Kulldorff spatial scan technique identified three

hotspots including Matabeleland South province. No evidence of

significant spatial heterogeneity was found with respect to underly-

ing risk factors.

CONCLUSIONS: A combined application of statistical and geo-

statistical modeling procedures using the modified proximate deter-

minant framework provides a robust approach to investigate HIV

spread in generalized epidemic settings.

Poster Number: T225 Abstract #: 3036Prevalence of Obesity, Diabetes, andHypertension among Household Women, India.

R. N. Hiremath, MD

Central Govt, Bagalkot, India

INTRODUCTION: Diseases linked to our way of life are growing,

in every country of the world. Lack of proper knowledge, inad-

equate time, faulty eating habits, no exercise, smoking are all to

blame for lifestyle diseases particularly for household women who

are neglected the most. This study was carried out mainly to esti-

mate prevalence of lifestyle diseases for predicting the future burden

of cardiovascular, cerebrovascular diseases etc as well as to under-

take preventive measures among women where there is hardly any

data regarding lifestyle diseases.


Dow



OBJECTIVES: (i) To assess the prevalence of obesity, diabetes

(DM) and hypertension and to co-relate risk factors associated if

any among women of Ranchi, Jharkhand. (ii) To carry out lecture

cum discussion to increase their knowledge and teach them simple

preventive measures based on the prevalence of lifestyle diseases.

METHODS: A cross-sectional study was carried out among all the

women staying in a particular community of Ranchi city. Data was

collected among 1373 women who consented for the study using a

pre-tested self-administered structured questionnaire and relevant

investigations were carried out.

RESULTS: The mean age of the participants was 29.01 with SD of

5.71.Majority (63.4%) of them basically belonged to urban areas.

65% of them had a family monthly income of less than 20,000 while

2.5% had above 40,000/-. 65% were having mild anemia, while 9%

and 1% had moderate and severe anemia, respectively. The preva-

lence of DM was 0.30% while 2% were pre-diabetic. 16% were

obese and 38% were overweight. 6% were found to be hypertension

that needed further evaluation. Obesity was significantly associated

with DM (P¼0.041) and hypertension (P¼0.000). Similarly

income was significantly associated with obesity (P¼0.00), DM

(P¼0.010) and hypertension (P¼0.000). Chi-square trend analysis

also showed increasing trend of lifestyle diseases with income and

DM, hypertension with obesity. An intensive lecture cum discussion

and video display was carried out with regards to lifestyle diseases

preventive measures which were appreciated by all.

CONCLUSIONS: Prevalence of lifestyle diseases was high even

though no baseline data exists. Intensive information, education,

communication (IEC) activities involving simple preventive meas-

ures targeted to household women and frequent check on lifestyle

diseases is the need of the hour. Lifestyle pattern and local environ-

mental factors may play an important role which needs to be studied

in detail.

Poster Number: T77 Abstract #: 3317Evolution of Disability in Victims of TrafficAccidents Estimated from the Application of theInternational Classification of Functioning,Disability and Health (ICF).

L. Oliveira, PhD1, J. H. Scatena, PhD1 and A. P. Macedo2

1Universidade Federal de Mato Grosso, Cuiaba, Brazil, 2Secretaria deEstado da Saude de Mato Grosso, Cuiaba, Brazil

INTRODUCTION: Each year, about 1.3 million deaths and 20–50

million injuries are due to traffic accidents, concentrated in the

countries of middle and low income. Traffic accidents as a cause of

morbidity and mortality have reached almost epidemic proportions

in Brazil, but little is known about sequelae and disabilities resulting

from them. The CIF enables monitoring of clinical status of the

patient, at different times, which is possible due to the qualifiers that

measure the magnitude of disability. The goal of this study was to

apply the ICF to characterize the progression of physical disability

in victims of traffic accidents.

METHODS: It was conducted an exploratory case series study, pro-

spective, with 68 victims of traffic accidents attended at a rehabilita-

tion unit reference in Mato Grosso, Brazil in 2010. Data were

obtained from physiotherapy assessment at two different times (first

day of treatment and discharge), being coded according to the health

dimensions of the ICF.

RESULTS: The average time between the evaluation and revalua-

tion date was 76.4 days (SD¼23.6 days). More positive functional

evolution occurred in the categories: muscle strength (41%), pain

sensation (40%) and gait pattern (35%). At the first evaluation

63% of patients had severe or complete disability of joint mobility,

whereas this percentage decreased to 23% at the revaluation.

Regarding muscular strength, while severe and complete disability

predominated at the evaluation (48%), over 50% showed mild dis-

ability or absence of it at the revaluation. There was a reduction

(70%) in levels of severe and complete disabilities related to pain.

There was also a reduction of deficiencies in the structure of the

body, with 27% of patients without them.

CONCLUSIONS: The use of ICF allowed to qualify the information

about physical disabilities, as well as to monitor the evolution of

these victims, most of which showed improvement in functionality.

Poster Number: T29 Abstract #: 3035A Study of Incidence of Recurrent GeneticTranslocations in Adult and Pediatric AcuteLymphoblastic Leukemia (ALL) Patients inNorth India.

N. Varma, MD, S. Naseem, MD, J. Binota, MD, S. Varma, MD,P. Malhotra, MD and R. K. Marwaha, MD

Post-Graduate Institute of Medical Education & Research (PGIMER),

Chandigarh, India

INTRODUCTION: Acute lymphoblastic leukemias (ALL) are char-

acterized by various distinctive chromosomal/molecular abnormal-

ities, having important diagnostic and prognostic implications.

Many western studies quote the incidence of fusion transcripts to be

30–35% in ALL. However the data from Indian sub-continent is

scarce. Studies in different population groups are needed to better

characterize/categorize these abnormalities and to design appropri-

ate treatment protocols. Strategies for molecular monitoring of min-

imal residual disease (MRD) in ALL cases are also designed

according to these abnormalities.

METHODS: Present study was undertaken from May 2010 to July

2013, enrolling all cases diagnosed as ALL by morphology and flow

cytometry immunophenotyping. Common fusion transcripts, includ-

ing TEL-AML1 [t(12;21)]; BCR-ABL [t(9;22)]; E2A-PBX1 [t(1;19)]

and MLL-AF4 [t(4;11)] were tested with a single multiplex RT-PCR

assay. Incidence of translocations was determined in ALL cases.

RESULTS: 397 cases of ALL were tested for above mentioned

fusion transcripts, of which 127 (32%) were adults and 270 (68%)

children, with a male:female ratio of 1.9:1. Overall, 70 (17.6%)

cases were positive for any of the fusion transcripts; of these, 25

were adult and 45 pediatric patients. The most common fusion tran-

script was TEL-AML1, detected in 28 (7.1%) cases, followed by

BCR-ABL in 27 (6.8%) cases. Adult ALL- Of the 127 cases, 25

(19.7%) were positive for any one fusion transcripts, with BCR-

ABL being most common, detected in 14/127 (13.7%) cases.

Pediatric ALL- Of the 270 cases, 45 (20%) were positive for any

one fusion transcripts, TEL-AML1 was most common detected in

22 (8.2%) cases. Overall reported incidence of transcripts in adult

and pediatric ALL from studies in different parts of world ranges

from 16.7% to 50% and 8.6% to 33.35, respectively.

CONCLUSIONS: The overall incidence of fusion transcript was

17.6% which was lower than that reported from western countries.

This observation could result from certain ethnic, genetic and envi-

ronmental factors. However, there was no significant difference in

the incidence of occurrence of recurrent genetic translocations


Dow



between adult and pediatric ALL cases, which is comparable to

studies from western countries which have also not reported differ-

ences between adult and pediatric patients. TEL-AML1 [t(12;21)],

indicating good prognosis, was the most frequent fusion transcript.

Rest of the fusion transcripts are associated with worse prognosis.

We are in the process of generating a database of hematological

malignancies, incorporating demographic, hematological, cytoge-

netic/molecular genetic features in north India.

Poster Number: T222 Abstract #Prevention of Breast and Cervical Cancer among

Brazilian Women: A Comparison between Urban

and Rural Areas.

J. Cerqueira, BA, R. R. Luiz, PhD, A. Brito, PhD, J. Moreira, MS,

C. Muzi, MS, C. L. Cunha, MS and P. Castro, BA

Federal University of Rio de Janeiro—UFRJ, Rio de Janeiro, Brazil

INTRODUCTION: Breast cancer is the most common cancer in

Brazilian and worldwide women. Cervical cancer is the fourth most

common cancer among worldwide women and the third among

Brazilian women. The fight against breast cancer occurs mainly by

mammography. In cervical cancer case, it’s recommended perform-

ing regular Pap test to detect HPV. This study aims to analyze the

pattern of actualization of these two exams among Brazilian from

25 years and comparing it between urban and rural areas. Data

from the National Survey Household Sample of 2008.

METHODS: The analysis was performed by crossing the elapsed

time of the completion of the last cervical exam and mammography,

and the results of this cross was divided into four hierarchical levels.

Level 1: women completely appropriate. Level 2: women partially

inadequate, level 1 closest. Level 3: women partially inadequate,

level 4 closest. Level 4 women totally inadequate. Hierarchical lev-

el’s boundaries were different for the three age groups considered.

RESULTS: In the range of 25 to 39 years old, 79.2% of women were

in level 1 and prevalence was 1.16 times higher in urban than in rural

areas. Among women 40–59 years old, in level 2, 7% of women were

in urban areas and 6.2% in rural areas. In level 3 these percentages in

urban and rural areas were 18.8% and 36.8%, respectively. Among

women with 60 years old or more 41,8% never performed any of this

2 tests, and the prevalence was 2.79 times higher in rural areas.

CONCLUSIONS: Data show that younger women tend to perform

preventive exams more frequently compared to older women. And

women in urban areas perform preventive exams more frequently

than women in rural areas. This data suggest the need of improve-

ment in Public policies to reduce those inequalities.

Poster Number: T15 Abstract #: 3392Factors Associated with HIV-infected Mothers’

Perceptions of and Reactions to, Infant Feeding

Choices in South Africa.

M. Zunza, MA1, M. Esser, MD2, J. A. Bettinger, PhD3 and M. F.

Cotton, PhD2

1Stellenbosch University, Stellenbosch, South Africa, 2Stellenbosch

University, Tygerberg, South Africa, 3University of British Columbia,

Vancouver, BC, Canada

INTRODUCTION: As Prevention of mother-to-child transmission

transitions towards universal combination antiretroviral therapy

(cART), the incidence of breastfeeding is increasing and formula feeding

decreasing for HIV-exposed infants in the Western Cape province of

South Africa. In a prospective cohort study (the Mother-Infant Health

Study) we are assessing infant feeding practices in HIV exposed infants.

This sub-study assessed HIV infected mothers’ perceptions of and reac-

tions to feeding their infants. The aim of this study was to capture, from

a phenomenological perspective, a socially constructed view of HIV

infected mothers’ lived experiences regarding feeding their infants.

METHODS: Through application of Interactive Qualitative Analysis,

two separate focus group discussions, a) breastfeeding mothers (n¼7)

and b) formula feeding mothers (n ¼8) were conducted to determine

factors associated with their feeding experiences. The focus group dis-

cussions were conducted in November 2013. Mothers reflected on

their thoughts, beliefs, and experiences regarding feeding their infants.

In depth individual interviews (n¼24) based on focus group

responses were then conducted to explore the personal meaning, rele-

vance, and life history examples of the themes with the mother.

RESULTS: Preliminary findings revealed that the main factors com-

prising experiences of breastfeeding HIV positive mothers were:

advice on breastfeeding, reasons for wanting to breastfeed, choosing a

feeding option that could make a baby sick (HIV infected), discour-

agement of breastfeeding, and challenges of breastfeeding. Main fac-

tors that described HIV positive formula feeding mothers experiences

were: advice on formula feeding, consideration of baby’s health, social

expectations, disclosure of HIV status, making decisions about their

own situation and fear of HIV transmission to their infant.

CONCLUSIONS: Breastfeeding and formula feeding HIV positive

mothers had different feeding experiences. Factors that describe

HIV positive mothers’ infant feeding experiences should be explored

on how they relate to each other to understand the meaning mothers

attach to their feeding experiences.

Poster Number: T65 Abstract #: 2861A Systematic Review of Qualitative Study inMultidisciplinary Teamwork in Healthcare.

F. Liu, BS and J. M. Johnston, PhD

The University of Hong Kong, Hong Kong, Hong Kong

INTRODUCTION: The world is getting old and Hong Kong is expe-

riencing an ageing population. In Hong Kong, there are various types of

multidisciplinary elderly healthcare teams working to meet needs of the

elderly. Efficient teamwork is thought to produce high quality of care.

However, the factors that influence teamwork efficiency are poorly

understood. The aim of this study is to describe the factors that influ-

ence teamwork efficiency in multidisciplinary heath service teams.

METHODS: Literature published in PubMed, ProQuest, ScienceDirect

and CNKI from 2000 to 2013 were searched using ‘[patient care team

(MeSH) OR multidisciplinary team (in title/abstract)] AND [coopera-

tive behavior (Majr) OR inter-professional relations (Majr)]’ as search

terms. Selection criteria included full-text English or Chinese language

qualitative study publications related to teamwork efficiency, involved

at least two types of medical workers and elderly patients with chronic

disease. Articles which were reviews, letters, comments, editorials,

news, not related to teamwork efficiency or non-qualitative studies

were excluded. Quality assessment of publications was carried out using

framework from BSA Medical Sociology Group.

RESULTS: Eight publications met the criteria. Four studies used

semi-structured interviews, one used focus groups and three used


Dow



observation. Three key themes related to team efficiency were iden-

tified: i) interdisciplinary communication. Team members experi-

enced efficient teamwork when interdisciplinary communication

happened frequently in their teams. ii) Role boundaries. Role boun-

daries were reported to be a reason of weak interdisciplinary cooper-

ation between team members. iii) Information flow and interaction.

Information flow and interactions often happened in multidiscipli-

nary teamwork and team members felt that they needed more advice

in dealing with information interactions.

CONCLUSIONS: Some factors that influence teamwork efficiency

have been indicated but the internal nature and process of healthcare

teamwork needs more understanding and exploration.

Poster Number: T208 Abstract #: 3158Family Income and Adolescent Blood Pressure ina Chinese Birth Cohort: ‘Children of 1997’.

M. K. Kwok, PhD1, G. M. Leung, MD2 andC. M. Schooling, PhD1

1The University of Hong Kong, Hong Kong, China, 2The University ofHong Kong, Hong Kong, Hong Kong

Lack of inconsistency in the relation of national income and popula-

tion mean blood pressure (BP) highlights the importance of contex-

tually specific understanding on whether and how individual income

affects blood pressure. Income confers not only the ability to pur-

chase material resources but also the capability to possess symbolic

resources. Social comparisons of income (relative income) or posi-

tion (relative rank) may be more relevant than income per se (abso-

lute income), particularly in settings with higher income disparities.

In a developed non-Western setting with a high Gini co-efficient, we

prospectively examined the associations of family income (account-

ing for family size) at birth with BP z-score at �13 years relative to

sex-, age- and height-specific references for contemporary US chil-

dren in a population-representative Chinese birth cohort, Hong

Kong’s ‘Children of 1997’ (n¼4,925). Higher Yitzhaki index of rel-

ative deprivation in income in units of HKD 1,000 (USD 129)

(income difference between own and others with higher income), i.e.

more deprivation was associated with higher diastolic BP (0.02

mmHg, 95% CI 0.001, 0.03), but not systolic BP. Higher relative

rank in income (percentile position in the income distribution) was

associated with lower systolic BP (-0.01 mmHg, 95% CI �0.20,

�0.002) and with lower diastolic BP (�0.09 mmHg, 95% CI

�0.14, 0.04). Absolute income (continuous) was not associated

with BP. In a developed non-Western setting, relative income and

relative rank, rather than absolute income, were associated with

higher adolescent BP, suggesting limited access to socially desirable

resources or psychological stress arising from social comparisons

could have implications for cardiovascular health.

Poster Number: T55 Abstract #: 3209Acceptability of Primary and SecondaryPrevention of Cervical Cancer among MedicalStudents in West China.

X. F. Pan1, Z. M. Zhao1, Y. Wen1, S. Q. Li1, Y. Zhao1, H. Chang1,Q. P. Xue1, C. J. Fu2 and C. X. Yang1

1Sichuan University, Chengdu, China, 2Chongqing Medical University,Chongqing, China

INTRODUCTION: The study aimed to understand the knowledge

and acceptability of cervical cancer screening and HPV vaccine

among medical students, and explore potential factors that influ-

enced their acceptability.

METHODS: We conducted a survey among medical students

among six universities across west China using a validated question-

naire on knowledge and perceptions of HPV, cervical cancer, and

HPV vaccine.

RESULTS: We surveyed 1878 medical students aged 20.8 (standard

deviation: 1.3) years. 67.9% were females, 95.3% were ethnic Han

Chinese, and 43.8% majored in clinical medicine. 49% and 80%

believed cervical cancer can be prevented by HPV vaccine and

screening, respectively, while 60.2% and 71.2% would like to

receive/advise HPV vaccine and screening. 35.4% thought HPV vac-

cine should be given to adolescents or young people aged 13–18

years. 41.9% stated women should start screening from 20 years,

and 49.2% considered women should receive screening every year.

The worry about complications (38.3% and 39.8%) and inadequate

information (42.4% and 35.0%) were the mostly cited obstacles for

receiving/advising HPV vaccination and cervical cancer screening.

Females were more likely to accept HPV vaccine (OR, 1.86; 95% CI

1.47–2.35) or cervical cancer screening (OR, 3.69; 95% CI

2.88–4.74). Students who showed negative attitude towards premar-

ital sex were less like to accept either HPV vaccine (OR, 0.67; 95%

CI 0.47–0.96) or screening (OR, 0.68; 0.47–0.10). Students with a

higher level of related knowledge were much more willing to receive

or advise vaccine (P trend<0.001) or screening (P trend<0.001).

Nonclinical students showed lower acceptability of cervical screen-

ing compared to students in clinical medicine (OR, 0.74; 95% CI

0.56–1.00).

CONCLUSIONS: The acceptability of HPV vaccine and cervical

cancer screening is low among medical students in west China. It is

important to improve the knowledge of HPV and cervical cancer

among medical students in order to supply well informed and

trained health care providers.

Poster Number: T104 Abstract #: 3374Prospective Investigation of Video Game Use inAdolescents and Subsequent Conduct Disorderand Depression.

P. J. Etchells, PhD1, S. H. Gage, MS2, A. D. Rutherford, PhD3

and M. R. Munafo, PhD2

1Bath Spa University, Bath, United Kingdom, 2University of Bristol,

Bristol, United Kingdom, 3BBC, London, United Kingdom

INTRODUCTION: There is increasing public and scientific con-

cern regarding long-term behavioural effects of violent video game

use in children, but there is currently little consensus in the literature

as to the nature of any such relationships. The aim of the present

study was to investigate violent game use in children, degree of vio-

lence in games, and subsequent relationship to depression or aggres-

sive behaviour.

METHODS: Data from 2661 children from the Avon Longitudinal

Study of Parents and Children were used. A stratified measure of

game use at age 8/9 years was developed, taking into account degree

of violence. Associations with conduct disorder and depression,

measured at age 15, were investigated using ordinal logistic regres-

sion, adjusted for a number of potential confounders.

RESULTS: Violent video game use was associated with increased

risk of being categorised in a higher conduct disorder band. Those

who played non-violent games had the lowest risk, followed by


Dow



those who played no video games, those who played moderately vio-

lent games, and those who played very violent games. No associa-

tion was found between number of games played and conduct

disorder, nor between video game use and depression.

CONCLUSIONS: There is a positive association between violent

video game use at age 8/9 years and risk of conduct disorder at age

15 years. However, violent video game use does not appear to be

associated with depression.

Poster Number: T212 Abstract #: 2980First-Time Exploration into Breast CancerIncidence and Mortality Across Small-Areas inCyprus: Spatial Patterning and Associations withRurality Indicators.

G. Zannoupas, MS1, A. Charalambous, PhD1, C. Kouta, PhD1,P. Pavlou, MD2, E. Papathanassoglou, PhD1, C. Nicolaou, BS1,O. Kolokotroni, MD3, D. Lamnisos, PhD1, P. Yiallouros, PhD4

and N. Middleton, PhD1

1Cyprus University of Technology, Limassol, Cyprus, 2Cyprus Ministry ofHealth, Nicosia, Cyprus, 3University of Nicosia, Nicosia, Cyprus, 4CyprusInternational Institute for Environmental and Public Health, CyprusUniversity of Technology, Limassol, Cyprus

INTRODUCTION: At 77 per 100,000, female breast cancer inci-

dence rates in Cyprus are consistent with other countries in

Southern Europe. However, with complete absence of GIS from the

Public Health arena, the extent of geographic inequalities in mortal-

ity and incidence across communities on the island, if any, remain

unknown. We investigated the geographical patterning of breast

cancer across small-areas in Cyprus and its association with rurality

indicators.

METHODS: Standardised Mortality and Incidence Ratios (SMRs/

SIRs) across 370 communities were calculated based on latest avail-

able registry data for period 2004–11 and 2003–08, respectively to

ensure sufficient numbers. Bayesian hierarchical Poisson models

with spatially unstructured and/or structured random effects were

used to smooth maps and investigate the association with popula-

tion density (rurality), population potential (remoteness from major

population centers) and percentage of retired population.

RESULTS: SMRs (range 0–6) and SIRs (range 0–4) were unreliable

at such a small level of aggregation (median population 129, IQR:

47–416, 10% of areas>1500). Nevertheless, up to two-fold differ-

ences remained across smoothed maps, with as much as 80% of the

variation explained locally. Mortality and incidence demonstrated a

similar geographic pattern with higher rates in and around metro-

politan areas and lower rates in rural and mountainous areas.

Population density (1.13 95% CI¼1.04–1.23, per SD increase) and

population potential (1.09 95% CI¼1.01–1.19) were both signifi-

cantly associated with increased mortality rates. Similar associations

were observed with incidence rates which were also inversely associ-

ated with the percentage of retired population (0.92, 95%

CI¼0.84–0.99). Interestingly, more than three-fold differences were

observed in Mortality-to-Incidence ratios, suggesting differences in

survival and/or registration quality.

CONCLUSIONS: Even on a small island like Cyprus there is sub-

stantial small-area variation in breast cancer mortality and incidence

rates, more likely to suggest urban-rural differences in reproduction-

related factors, even though the influence of other lifestyle and envi-

ronmental factors cannot be ruled out.

Poster Number: T3 Abstract #: 3129Recording of Dementia among UK Women:A Comparison of Primary Care and HospitalAdmission Records.

A. Brown, MS, M. E. Kroll, PhD, O. Kirichek, MS, G. K. Reeves,PhD, J. Green, MD and V. Beral, MD


INTRODUCTION: There is little reliable evidence on modifiable

(lifestyle) risk factors for dementia; epidemiological studies are typi-

cally small and ascertainment of outcomes may be incomplete.

METHODS: The Million Women Study is a cohort of 1.3 million

UK women recruited through the NHS Breast Screening Program

during 1996–2001, and offers the opportunity to study dementia in

a large prospective study with long and virtually complete follow-up

through record linkage to National Health Service medical records.

The cohort has linkage to death and hospital admissions records in

England (Hospital Episode Statistics), which may be incomplete,

since some women with dementia may not be admitted to hospital

or die from the condition. A sample of the cohort has recently been

electronically linked to primary care data using the Clinical Practice

Research Datalink (CPRD). This provides primary care records for

about 8% of the population of England. In this linked dataset we

have compared records of dementia in hospital admissions and in

primary care data. Dementia diagnosis codes were taken to be ICD-

10 F00, F01, F02, F03, G30 in hospital records, and the equivalent

READ/OXMIS codes in CPRD.

RESULTS: A random sample of 10,000 participants in the cohort

were matched to CPRD data and for 821 (8.2%) women primary

care records were obtained. A preliminary analysis of the data shows

that the proportion of women with any mention of dementia was

1.1% in primary care records, and 0.8% in hospital admissions. For

women with a record of dementia in both primary care and hospital

records, the first mention of dementia occurred in their primary care

records and was on average approximately two years earlier. The

entire cohort is now being linked to CPRD data and comparisons of

the recording of different subtypes of dementia will be presented.

CONCLUSIONS: Linkage of primary care and hospital admission

data is feasible in England and should provide important informa-

tion for epidemiological studies of dementia.

Poster Number: T33 Abstract #: 2869Relationship between HouseholdSocio-Economic Status and Under- FiveMortality in Rufiji DSS, TANZANIA.

C. Nattey, PhD1,2 and H. Masanja, PhD3

1National Institute for Occupational Health/National Cancer Registry,

Johannesburg, South Africa, 2Ifakara Health Institute, Dares Salaam,

Tanzania, 3Ifakara Health Institute, DaresSalaam, Tanzania

INTRODUCTION: Disparities in health outcomes between the

poor and the better off are increasingly attracting attention from

researchers and policy makers. However, policies aimed at reducing

inequity need to be based on evidence of their nature, magnitude,

and determinants. The study aims to investigate the relationship

between household socio-economic status (SES) and under-five mor-

tality, and to measure health inequality by comparing poorest/least


Dow



poor quintile mortality rate ratio and the use of a mortality concen-

tration index. It also aims to describe the risk factors associated

with under-five mortality at Rufiji Demographic Surveillance Site

(RDSS), Tanzania.

METHODS: This analytical cross sectional study included 11,189

children under-five residing in 7,298 households in RDSS in 2008.

Principal component analysis was used to construct household SES.

Kaplan-Meier survival incidence estimates were used for mortality

rates. Health inequality was measured by calculating and comparing

mortality rates between the poorest and least poor wealth quintile.

We also computed a mortality concentration index. Risk factors of

child mortality were assessed using Poisson regression taking into

account potential confounders.

RESULTS: Under-five mortality was 26.9 per 1000 person-years

[95% confidence interval (CI) (23.7–30.4)].The poorest were 2.4

times more likely to die compared to the least poor. Our mortality

concentration index [�0.16; 95% CI (�0.24, �0.08)] indicated con-

siderable health inequality. Least poor households had a 52%

reduced mortality risk [incidence rate ratio (IRR) ¼0.48; 95% CI

0.30–0.80]. Furthermore, children with mothers who had attained

secondary education had a 70% reduced risk of dying compared to

mothers with no education [IRR ¼0.30; 95% CI (0.22–0.88)].

CONCLUSIONS: Household socio-economic inequality and maternal

education were associated with under-five mortality in the RDSS.

Targeted interventions to address these factors may contribute

towards accelerating the reduction of child mortality in rural Tanzania

Poster Number: T141 Abstract #: 2985Alcohol Consumption and Mortality from AorticDissection and Aneurysm among Middle-agedJapanese Men and Women: The JapaneseCollaborative Cohort Study (JACC).

T. Shirakawa, BA1, K. Yamagishi, PhD2, H. Iso, PhD1 andA. Tamakoshi, PhD3

1Osaka University, Suita, Japan, 2University of Tsukuba, Tsukuba, Japan,3Hokkaido University, Sapporo, Japan

INTRODUCTION: Few studies have examined the association

between alcohol consumption and mortality from aortic diseases

and the results have been inconsistent.

METHODS: The JACC Study is a nation-wide community-based

cohort study involving 45 Japanese communities. Between 1988 and

1990, 110 585 men and women, aged 40 to 79 years, completed a

self-administrated questionnaire including information about alco-

hol consumption and were followed up for a median duration of

17.9 years. Subjects who reported cancer or cardiovascular disease

history, or did not provide information about drinking habits, were

excluded. Mortality from thoracic aortic disease (aortic dissection

and thoracic aortic aneurysm) and abdominal aortic aneurysm was

identified by ICD-10 codes on the death certificate.

RESULTS: Of the 75 556 responders (32 032 men and 43 523

women), 104 men and 73 women died from aortic disease. For men,

compared with nondrinkers, light-to-moderate alcohol consumption

(less than 46 g of ethanol intake per day) was associated with

reduced risk of mortality from thoracic aortic disease, whereas

heavy alcohol consumption (more than 46 g of ethanol intake per

day) was not. The respective multivariable hazard ratios (95 percent

confidence interval) were 0.32 (0.14–0.72) and 0.99 (0.51–1.94).

Current male drinkers had reduced risk of mortality from abdominal

aortic aneurysm with the multivariable hazard ratio of 0.46

(0.23–0.91). For women, drinkers seemed to have reduced risk of

mortality from thoracic aortic disease with the multivariable hazard

ratio of 0.22 (0.05–0.92) compared with nondrinkers.

CONCLUSIONS: There was a U-shaped relationship between alco-

hol consumption and risk of mortality from thoracic aortic disease.

Poster Number: T110 Abstract #: 3305Flexible Modelling of the Cumulative Effect ofSmoking on Lung Cancer.

A. Gasparrini, PhD1, L. Richiardi2, D. Mirabelli2, L. Simonato3,F. Forastiere4 and F. Merletti2

1London School of Hygiene and Tropical Medicine, London, United

Kingdom, 2University of Turin and CPO-Piemonte, Turin, Italy, 3University

of Padova, Padova, Italy, 4Lazio Regional Health Service, Rome, Italy

INTRODUCTION: the risk of lung cancer due to tobacco smoking

is the results of exposures sustained in previous years and possibly

decades. The effect has been modelled using several time-varying

factors, such as intensity, duration, time since first exposure and/or

cessation. Here we propose an alternative model providing a flexi-

ble, easy-to-interpret yet sophisticated approach.

METHODS: the modelling framework is based on the

combination of two functions describing the potentially non-linear

dose-response, and the weights of contributions from past exposures,

defined here as lag-response. The model can systematically describe

complex associations changing dynamically in time, easily allowing

for instance for a delay in the increase in risk after smoking initiation

and a decrease after cessation. The method is implemented in the

freely available package dlnm within R. The framework is applied to

pooled data from three case-control studies with complete exposure

histories for 1,479 cases and 1,918 controls.

RESULTS: the logistic model using spline functions to describe

the exposure and lag-responses indicates a sub-linear exposure-

response, with lung cancer risk flattening out at high tobacco con-

sumption. Smoking 20 cigarettes/day in a given year produces a

risk peaking after eight years, with an odds ratio (OR) of 1.12 (95%

CI 1.09–1.15), then decreasing to 1.05 (1.03–1.07) after 25 years,

and disappearing only after 50 years. After 20 years of smoking

20 cigarettes/day the cumulative OR is 4.68 (3.82–5.74), and if then

smoking ceases it decreases to 3.56 (2.66–4.78) after 10 more years.

CONCLUSIONS: the proposed method offers a flexible alternative

to models including several time-varying variables for describing the

complex temporal dependency between smoking and lung cancer.

The framework can be used to produce dynamic predictions of risk

based on specific exposure histories, and it can be extended to any

association where the risk is assumed to depend on the cumulative

effect of time-varying exposures.

Poster Number: T2 Abstract #: 3085Foot Pain among the Elderly Subjects ofDhaka City.

P. C. Banik, MPH1, M. Das, PhD1,2, S. K. Paul, BS3, S. R. Sultana,BS3, M. Moniruzzaman, MPH1 and L. Ali, PhD1


Dow



1Bangladesh University of Health Sciences (BUHS), Dhaka, Bangladesh,2University of Rajshahi, Rajshahi, Bangladesh, 3State College of Health

Sciences (SCHS), Dhaka, Bangladesh

INTRODUCTION: Foot pain is a common condition reported by

at least one in every five among general population. Existing papers

indicate that the prevalence of foot pain increases with age and, it is

higher in female gender which suggested that intensity of pain is

higher in overweight and obese population. Hitherto, actual causa-

tive factors related to foot pain is unknown. The objectives of this

study were to determine the prevalence of foot pain and its associ-

ated factors among older subjects of Dhaka city.

METHODS: Under a cross-sectional study design a total number of

200 subjects from different selected areas in Dhaka city were

screened purposively in the mid July 2013. According to WHO crite-

ria 65-years were considered as elderly. Data was collected by face-

to-face interview through authenticate questionnaire. Appropriate

univariate, bivariate and multivariate analyses were done accord-

ingly by SPSS software Windows version.

RESULTS: The mean(6SD) age of the total respondents was

67(65.5) (95% CI �54.9, 79.8) and 80.5% (95% CI �75, 85.9)

were male. The mean(6SD) BMI was 26.3(65.6) (CI �12.4, 40.2)

and 59.5% (CI �52.7, 66.3) were overweight. The proportion of

subjects with foot pain was 65.5% (CI �58.9, 72). The nature of

foot pain was temporary and 41.98% (CI �35.1, 48.8) had moder-

ate pain. Significant association for foot pain was seen with female

gender (P<0.001), senior age group (P<0.001), married subjects

(P<0.001), long standing working subjects (P¼0.002), subject

with large family (P¼0.009) and high BMI (P¼0.002). Footwear

associated discomfort (P<0.001), prolonged footwear wearing

(P¼0.005), footwear size (P¼0.016), poor body balance

(P<0.001) and tremor during walking/standing (P<0.001) were

showed significant association with foot pain.

CONCLUSIONS: Above half of elderly subjects were sufferings

from foot pain. Female gender, senior citizens, married subjects,

prolonged standing subjects, big family’s subjects, elevated BMI,

footwear associated discomfort, prolonged footwear wearing, foot-

wear size, poor body balance and tremor at walking/standing were

the major predictors for foot pain.

Poster Number: T155 Abstract #: 3240Relationship of 24-hr Urinary Na/K ratio to 24-hrUrinary Na and K Excretion in Men and Womenfrom Multi-Ethnic General Populations: theINTERSALT Study.

T. Iwahori, MS1,2, K. Miura, MD2, H. Ueshima, MD2, Q. Chan,

PhD3, P. Elliott, PhD3, A. R. Dyer, PhD4 and J. Stamler, MD4

1OMRON Healthcare Co., Ltd., Kyoto, Japan, 2Shiga University of

Medical Science, Otsu, Japan, 3Imperial College London, London, United

Kingdom, 4Northwestern University, Chicago, IL

INTRODUCTION: High dietary sodium (Na) and low dietary

potassium (K) intakes are associated with adverse blood pressure

levels and excess risks of cardiovascular diseases. Na/K ratio in

24-hr urine has been reported to be a useful index in relation to

blood pressure and cardiovascular disease risk. Our aim was to clar-

ify the associations of 24-hr urinary Na/K ratio with 24-hr urinary

Na and K excretion in 52 diverse populations in the world using

data from the International Cooperative Study on Salt, Other

Factors, and Blood Pressure (INTERSALT).

METHODS: INTERSALT collected standardized data on 24-hr uri-

nary Na excretion, K excretion and Na/K ratio in 10,079 men and

women ages 20–59 years from 52 population samples in 32 coun-

tries, in 1980s. The associations between Na/K ratio and Na and K

excretion in 24-hour urine of individuals were analyzed by correla-

tion analysis. Na/K ratio of 24-hr urine stratified in one unit inter-

vals was compared with the Na excretion and K excretion.

RESULTS: Mean 24-hour Na excretion was 156.0 mmol/24h and

mean 24-hour K excretion was 55.2 mmol/24h; Na/K ratio was

3.24. Correlations between Na/K ratio and Na and K excretions

were 0.57 and �0.48, respectively. The proportion of participants

with estimated salt (NaCl) intake less than 5 g/day was 94% in those

with urinary Na/K ratio less than 1, and 5% in those with urinary

Na/K ratio 3 or over. Mean K excretions in participants with urinary

Na/K ratio less than 1 and 3 or over were 74 mmol/24h and 44

mmol/24h, respectively.

CONCLUSIONS: To keep urinary Na/K ratio low, e.g. less than 1,

dietary improvements are needed, to reduce individual Na intake

and increase individual K intake.

Poster Number: T35 Abstract #: 2913Respiratory Infectious Disease Hospitalizationsin American Indian Children Presenting to theEmergency Department in the Upper Midwest.

K. M. Burgess, MPH1, N. R. Payne, MD2, A. B. Kharbanda, MD2

and S. E. Puumala, PhD1

1Sanford Research, Sioux Falls, SD, 2Children’s Hospitals and Clinics ofMinnesota, Minneapolis, MN

INTRODUCTION: American Indian (AI) children are hospitalized

at almost twice the rate of the general population for lower respira-

tory infections. This variation in care may be secondary to severity

of illness, access to primary care or disparity in care provided in the

emergency department (ED). We examined differences in hospital

admissions in AI children compared to White children presenting to

the ED with a respiratory infection in a large multi-center cohort to

elucidate possible disparities.

METHODS: We collected data on all children 0-17 years presenting

to one of six geographically diverse EDs in the upper Midwest from

June 2011 to May 2012. Our analyses were limited to patients who

self –identified as White or AI and were diagnosed with any respira-

tory infection (based on the Diagnostic Grouping System for Child

ED Visits). We explored differences in admission by race after

adjusting for age, insurance type, triage level, and rural/urban set-

ting using multiple logistic regression. We also explored possible

interactions with race, triage level, and insurance status.

RESULTS: A total of 4,329 encounters in the ED resulted in a pri-

mary diagnosis of a respiratory infection with 1301 (30%) resulting

in hospitalization. Children that were either<1 year of age, at a

high triage level, or from an urban setting were more likely to be

admitted. After controlling for confounders, AI children presenting

to the ED were less likely to be admitted for a respiratory infection

than White children [OR¼0.40, 95% CI (0.28, 0.58)]. Similar

results were obtained when limiting to only lower respiratory-tract

infections [OR¼0.36, 95% CI (0.23, 0.57)].


Dow



CONCLUSIONS: We found a lower odds of admission for AI

children presenting to the ED with a respiratory infection compared

to White children after adjusting for confounders. Future efforts will

work to address and improve this potential disparity.

Poster Number: T36 Abstract #: 2921Gender Differences in Objectively AssessedPhysical Activity Between Asthmatic andNon-Asthmatic Children: A DescriptiveComparative Study among CypriotSchoolchildren.

M. Economou, MS1, N. Middleton, PhD1, O. Kolokotroni, MD2,S. Savva, PhD3, M. Gavatha, MD4, P. Ioannou, BS5,T. Karpathios, PhD6 and P. Yiallouros, PhD7

1Cyprus University of Technology, Limassol, Cyprus, 2University ofNicosia, Nicosia, Cyprus, 3Research and Education Foundation of ChildHealth, Nicosia, Cyprus, 4Hospital ‘Archbishop Makarios III, Nicosia,Cyprus, 5Hospital ‘Archbishop Makarios III’, Nicosia, Cyprus, 6AthensMedical Centre, Athens, Greece, 7Cyprus International Institute forEnvironmental and Public Health, Cyprus University of Technology,Limassol, Cyprus

INTRODUCTION: While asthma may be viewed as a barrier to

exercise, physical activity has been proposed to protect against

asthma development. Commonly using self-reported measures, stud-

ies have reported comparable or even higher physical activity in

asthmatic compared to non-asthmatic children. With the use of

accelerometers, the aim here was to compare and assess any gender

differences in physical activity between asthmatic children and their

peers.

METHODS: A sample of 104 children aged 8–9 years whose

parents reported ever having asthma in a community-based study

were further categorized into active and inactive asthma based on

report of at least one episode of wheezing in the last 12 months and

were compared to a control group of 98 randomly selected children

matched for gender. Logistic regression was used to examine gender

differences in the association between asthma status and physical

activity measured for seven consecutive days with the use of an

accelerometer.

RESULTS: While no overall difference in physical activity was

observed between controls and children with active or inactive

asthma, there appeared to be important gender-differences,

Specifically, girls with inactive and active asthma had significantly

lower moderate-to-vigorous physical activity (MVPA) than their

healthy peers with geometric means ratios of activity (min/day) of

0.73 (95% CI 0.503, 1.048; P-value¼0.086) and 0.59 (95% CI

0.369, 0.929; P-value¼0.024), respectively. In contrast, no differ-

ences were observed between boys with either active or inactive

asthma and their healthy peers; P-value of likelihood ratio test for

effect modification by gender ¼0.034.

CONCLUSIONS: This gender difference may mask a true difference

in levels of physical activity, at least between asthmatic and non-

asthmatic girls, in an overall comparison (i.e. both gender com-

bined) and it can at least partly explain the existing confusion in the

literature as to whether asthmatic children are as physically active as

healthy children.

Poster Number: T140 Abstract #: 3154Aldehyde Dehydrogenase 2—a Potential GeneticRisk Factor for Lung Function among SouthernChinese in China: the Guangzhou BiobankCohort Study.

S. L. Au Yeung, PhD1, C. Jiang, MD2, K. K. Cheng, PhD3,P. Adab, MD3, K. B. Lam, PhD3, B. Liu, PhD2, W. Zhang, PhD2,T. H. Lam, MD4, G. M. Leung, MD4 and C. M. Schooling, PhD4,5

1The University of Hong Kong, Hong Kong, China, 2Guangzhou Number12 Hospital, Guangzhou, China, 3University of Birmingham, Birmingham,United Kingdom, 4The University of Hong Kong, Hong Kong, Hong Kong,5Hunter College, New York, NY

INTRODUCTION: In East Asia moderate alcohol users have better

lung function, but never users may have more inactive aldehyde

dehydrogenase 2 (ALDH2) alleles (A) potentially generating con-

founding, because inactive alleles may increase acetaldehyde expo-

sure and reduce lung function.

METHODS: We examined the association of ALDH2 genotypes

with percentage predicted lung function for age, sex and height,

assessed from forced expiratory volume in one second (FEV1), and

forced vital capacity (FVC), among 5,641 older Chinese using multi-

variable linear regression.

RESULTS: ALDH2 genotypes were associated with alcohol use and

height but not other attributes. Inactive alleles were inversely associ-

ated with lung function (% predicted FEV1 �1.52%, 95% confi-

dence interval (CI) �2.52%, �0.51% for one inactive allele and

�2.05%, 95% CI �3.85%, �0.26% for two inactive alleles com-

pared to two active alleles; and for % predicted FVC �1.25%, 95%

CI �2.15%, �0.35% and �1.65%, 95% CI �3.25%, �0.04%).

The association of moderate alcohol use with lung function was

attenuated after adjusting for ALDH2, in addition to other potential

confounders.

CONCLUSIONS: Previous findings in East Asians may be con-

founded by ALDH2. High frequency of inactive ALDH2 alleles in

East Asia may exacerbate the effect of environmental acetaldehyde

exposure on lung function, and potentially on chronic obstructive

pulmonary disease.

Poster Number: T63 Abstract #: 3327An Analysis of CLINICAL Signs and symptomswhich best Predict the Need for Haart Initiationin HIV Infected South African Women.

P. G. Horumpende, MS

Lugalo Military Medical School, Dar es Salaam, Tanzania

INTRODUCTION: South Africa is currently experiencing one

of the most severe AIDS epidemics in the world. The major

challenge lies in prompt identification and early initiation of

treatment in those eligible for HAART. Clinical staging has previously

been recommended for use in settings where CD4 þ count testing is

not available. We conducted secondary data analysis to determine

whether clinical symptoms and signs are useful in predicting the

need for HAART initiation (CD4 þ count<200 cells/lL) in South

Africa.


Dow



METHODS: Screening data from a randomized controlled trial in

women who were HIV positive were analysed. All participants

were interviewed using a structured questionnaire to elicit symptom

history and then physical examination was done. Participants were

staged using WHO criteria. Blood was drawn for CD4 þ testing.

The association between signs and symptoms and a CD4 þ<200

cells/lL was assessed using logistic regression.

RESULTS: Among 589 HIV infected women aged between 18 and 58

years, 90% were assessed as WHO clinical stages I/II. The median

CD4 þ count was 403 cells/lL (IQR: 273–586). Among women who

were WHO stage I/II, 13% had CD4 þ count<200 cells/lL and

required HAART. The WHO clinical staging had a low sensitivity

(4%) but high specificity for detecting those that require treatment.

CONCLUSIONS: In a setting where asymptomatic patients are

diagnosed with HIV, clinical assessment cannot replace CD4 þcount testing as a method of identifying those that need treatment.

Poster Number: T207 Abstract #: 3346Gender, Occupational Stress, Resources, andSleep Deprivation among Immigrant Micro-business Owners: Cross-sectional Survey ofKorean Immigrants in Toronto, Canada.

I. H. Kim, PhD1 and S. Noh, PhD1,2

1Centre for Addiction and Mental Health, Toronto, ON, Canada,2University of Toronto, Toronto, ON, Canada

INTRODUCTION: Literature highlights occupational stressors as

critical determinants of psychological well-being. Although immi-

grant micro-business owners (MBOs) tend to suffer from long work

hours with heavy workloads in low-yielding service sectors, studies

have rarely investigated their experiences. This study compares a

sample of immigrant MBOs with matched (sex, age) samples of paid

employees on: 1) occupational stressors (i.e. physical, psychological,

and emotional demands), 2) occupational resources (i.e. autonomy,

job satisfaction), and 3) sleep deprivation.

METHODS: Between March and November 2013, 1288 Korean

immigrant workers (51.4% men, 48.6% women, aged 20 to 64) living

in and around the Greater Toronto Area were interviewed in person.

Participants included owners of micro-businesses such as convenience

stores, laundromats, and small restaurants. Paid employees consisted

of professionals, office workers, and manual workers. The health out-

come, sleep deprivation, was assessed by a standardized 9-item scale

tapping sleeplessness, disturbed sleep, fatigue, and work dysfunction.

RESULTS: Compared to paid employees, MBOs reported a higher

degree of occupational demands (physical, psychological, and emo-

tional) and job autonomy. Among males, job satisfaction was higher

for MBOs; female MBOs expressed less satisfaction than male

MBOs and female professional and office workers. Among male

samples, the effects of physical and emotional demands on sleep

deprivation were greatest for MBOs; however, effect differences

were not apparent within the female samples. Job satisfaction signif-

icantly mitigated the associations of sleep deprivation with physical

demand (P=.011) for female MBOs and emotional demands

(P=.053) for male MBOs, but not paid employees. Autonomy also

seemed to benefit male MBOs in buffering the negative health effects

of emotional demand (P=.0221).

CONCLUSIONS: Immigrant MBOs experience elevated levels of

occupational stress, but enjoy greater autonomy, which may buffer

the impact of occupational stress on sleep deprivation. The stress

buffering effects of occupational resources seem greater for male

MBOs than for female MBOs.

Poster Number: T147 Abstract #: 3050Risk Factors for Obesity at Age 3 in AlaskanChildren, Including the Role of BeverageConsumption: Results from Alaska PRAMS2005–06 and its Three-Year Follow-up Survey,CUBS, 2008–09.

M. B. Young, MPH1, J. M. Wojcicki, PhD2, K. Perham-Hester,

MS3, P. DeSchweinitz, MD4 and B. Gessner, MD5

1Alaska Division of Public Health, Anchorage, AK, 2UCSF, San Francisco,

CA, 3Alaska Department of Health and Social Services, Anchorage, AK,4Tanana Chiefs Department, Fairbanks, AK, 5Agence de Medecine

Preventive, Paris, France

BACKGROUND: Prenatal and early life risk factors are associated

with childhood obesity in the preschool years. American Indian/

Alaska Native children have one of the highest prevalences of child-

hood obesity of all US racial/ethnic groups. Methods: Using the

Pregnancy Risk Assessment Monitoring System (PRAMS) and its 3-

year follow-up, the Childhood Understanding Behaviors Survey

(CUBS), we evaluated health, behavioral, lifestyle and nutritional

variables during the pre-pregnancy, prenatal, and postpartum peri-

ods and first few years of life in relation to risk for obesity (95th per-

centile for body mass index (BMI)) at three years of age among

Alaska Native and non-Native children. Multivariate logistic regres-

sion modeling was conducted using Stata 12.0 to evaluate independ-

ent risk factors for obesity in non-Native and Alaska Native

children. Results: Longer duration of breastfeeding was independ-

ently protective against obesity at age 3(OR 0.95, 95% CI

0.92–0.99) in non-Native children. Among Alaska Native children,

increased paternal education was protective, (OR 0.42, 95% CI

0.19–0.95) as was increasing income/dependent ratio. Alaska

Native race interacted with maternal use of chew/spit tobacco dur-

ing pregnancy, whether the mother was trying to get pregnant and

consumption of fried potatoes in relation to child obesity at age 3.

Among Alaska Native children, residents of the Northern/Southwest

region had an increased daily consumption of beverages including

water, milk, 100% fruit juice, soda and sweetened/fruit drinks com-

pared to residents of other regions. Conclusions: Risk factors for

obesity at age three differed for Alaska Native and non-Native chil-

dren. Among both populations, pre-pregnancy and prenatal factors

appeared to have less of an association with obesity at age three

than current lifestyle. The higher prevalence of obesity in Alaska

Native children may be explained by lifestyle patterns and food con-

sumption in certain parts of the state, specifically the Northern/

Southwest region.

Poster Number: T53 Abstract #: 3160Anemia in Children Residing in ChecherskDistrict of Belarus.

T. Zhunussova, PhD

Norwegian Radiation Protection Authority, Osteras, Norway

INTRODUCTION: The accident at the Chernobyl nuclear power

plant in Ukraine (April 26, 1986) has contaminated not only neigh-

boring countries but the whole Europe. However, the most


Dow



contaminated regions were in southern Belarus, northern Ukraine

and the Bryansk and Kaluga regions of Russia by radionuclides such

as Cs, Sr and other trans-uranium elements. According to some pub-

lications the children (including in utero) of the Chechersk district of

the Gomel region and the Mtsensk and Bolhov districts of the Orel

region had various levels of exposure. Previous studies found anemia

in 11.5% among 1926 surveyed children during 1986–98.

METHODS: A total of 2395 children aged 3–15 years old at the

baseline checkup (2005) in Chechersk district were investigated for

anemia prevalence. Data on the following plausible predictors were

recorded: age, sex, living conditions, type of residency, food con-

sumption, zone of residency, level of internal contamination.

Anemia diagnosis was based on the hemoglobin level, hematocrit,

number of erythrocytes and serum iron. Cut off criteria (140 g/l) for

children set up in Belarus was compared to WHO cut off criteria

(120 g/l) to look for the differences in the prevalence of anemia in

the checkup periods during the years 2005, 2007 and 2009. The two

criteria of anemia were compared using two statistical methods:

NcNemar’s test for marginal homogeneity and the Kappa statistic to

measure agreement. The prevalence of anemia by predictor variables

was determined using descriptive statistics. The statistical analysis

was performed using SAS for Windows (version 9.3). Multivariate

logistic regression modeling was used to estimate the relative risk

and 95% confidence intervals of anemic children compared with

non-anemic.

RESULTS: Anemia was correlated with hemoglobin level, serum

iron, height, globular value, erythrocytes, unsatisfactory social life,

and zone of residence 5 compared to other residence zone (Pr <

0.05 for all). Nutrition variables were not consistent across the

exams. OR ¼3.7 for anemia if the person does not have potatoes in

the diet. The OR is slightly less than one if they have dairy in the

diet.

CONCLUSIONS: Prevalence of anemia in children was correlated

with hemoglobin level, serum iron, height, globular value, erythro-

cytes, unsatisfactory social life, and zone of residence.

Poster Number: T177 Abstract #: 3192HBV Vaccination uptake at Family MedicineClinics in Poland: A Cross-Sectional Sero-Survey.

M. Ganczak, PhD1, G. Dmytrzyk-Danilow, MS1, M. Korzen,PhD2 and M. Drozd-Dabrowska, PhD3

1Pomeranian Medical University, Szczecin, Poland, 2WestpomeranianUniversity of Technology, Szczecin, Poland, 3Primary Care Clinic, Gryfino,Poland

INTRODUCTION: Hepatitis B is a significant health burden in

Poland with nosocomial transmission being the main source of

infection. Therefore, HBV vaccination before surgery is widely rec-

ommended. Objective: To assess the determinants of HBV vaccina-

tion uptake among patients attending family medicine clinics (FMC)

and to establish their serological status in terms of HBV infection.

METHODS: In March 2013 consecutive patients attending three

randomly selected FMCs located in Zgorzelec region, Poland, com-

pleted an anonymous questionnaire on HBV vaccination history.

Immunization status was based on self-reports, checked by

the results of vaccination cards and anti-HBs titer. In those

unvaccinated, serum samples were assayed for anti-HBc total with

the use of third-generation testing methods.

RESULTS: Response rate: 99.3%. Of 410 participants (66.1%

females, median age 56 years), 55.4% (95% CI 50.5–60.1%) were

previously vaccinated against HBV; in those 11.5%—with two doses

of vaccine, 66.1%—with three doses,18.1% – with four doses.

Elective surgery was the main reason (57.0%) for HBV immunization,

4.9%—were vaccinated due to recommendations by GPs. The multi-

variable regression model revealed that high educational status (OR

1.35), living in a city (OR 1.75), as well as having a surgery in the

past (OR 2.93) were each associated with greater odds of being vacci-

nated. Anti-HBc total prevalence among those unvaccinated was

13.6% (24/176 95% CI 9.3%–19,5%). None of the patients had the

history of clinical hepatitis B, none was aware of an infection.

CONCLUSIONS: Low HBV immunization coverage among adult

patients and the presence of serological markers of HBV infection

among those unvaccinated call for comprehensive preventative

measures against infection, including greater FMCs involvement.

Although interventions should cover the whole population, low edu-

cated inhabitants living in the rural areas should be a group of spe-

cial interest. Preoperative immunization against HBV seems to be an

efficient tool to increase the vaccination uptake.

Poster Number: T49 Abstract #: 3019Risk of Aggression and Criminal Tendenciesamong Adolescents Living in AlexandriaGovernorate, Egypt.

I. M. Wahdan, DrPH, N. A. El-Nimr, DrPH, R. E. Kotb, DrPH andA. M. Wahdan, DrPH

High Institute of Public Health, Alexandria University, Alexandria, Egypt

INTRODUCTION: Adolescence has been regarded as a time in

which risk taking behaviors are common. Adolescent aggression and

violent behaviors are among the most visible forms of violence in

society, often emanating from multiple risk factors. The current

study was carried out to identify risk factors responsible for aggres-

sion, violence and criminal tendency and assess the magnitude of

these risks among adolescents in Alexandria, Egypt.

METHODS: Using a multistage cluster sampling technique, a repre-

sentative sample of families from all health districts in Alexandria

was selected and all adolescents (11–19 years) living within these

families were included. A predesigned interview questionnaire was

used to collect socio-demographic data and an Arabic translation of

the screening questionnaire developed by the Mentor Research

Institute was used to assess various behaviors of the adolescents and

determine the level of risk for aggression and criminal tendency.

RESULTS: As much as 26.9% of adolescents in this study were

assessed to be at high risk and another 20.2% were at extremely

high risk of aggression and of criminal tendency. Logistic regression

analysis showed that the adolescent’s sex, the presence of behavioral

changes in the family, violence against brothers and sisters of the

adolescent and drug addiction among any member in the family

were the risk factors significantly associated with the risk of aggres-

sion and violence among adolescents.

CONCLUSIONS: It is concluded that the risk of aggression and

criminal tendencies among adolescents in Alexandria is significant.

There is also an increasing pattern of exposure to violence. It is

hoped that the responsible authorities would begin to give special

attention to reduce the impacts of these exposures on the prevalence

of aggression and criminal tendency in Egypt.


Dow



Poster Number: T90 Abstract #: 3069Health Information Seeking Behaviour andNeeds of Grass Root Healthcare Providers inNorthern City of India.

S. Raj, MPH1, A. Singh, MD2 and V. L. Sharma1

1Panjab University, Chandigarh, India, 2PGIMER, Chandigarh, India

INTRODUCTION: Free universal access to information for health-

care providers is a need to achieve Health for All. Providing access

to reliable information for healthcare workers in developing coun-

tries is potentially the most cost-effective strategy for improving the

quality of healthcare delivery.So the present study was conducted to

ascertain the health information seeking behaviour and needs of

basic level healthcare workers in Chandigarh.

METHODS: A cross-sectional study was conducted among 150

Anganwadi Workers (AWW) and Auxiliary Nurse Midwives

(ANM) in November-December, 2013.Pretested self-administered

questionnaire was used to collect data on information seeking

behaviour, needs, attitudes and access to internet. Data analysis was

done using SPSS-16. Descriptive statistics and chi square test was

used to analyze the study variables.

RESULTS: Majority of the respondents (92%) acknowledged the

need of health information in their daily routine. Most preferred

source for health information was Training/Workshops (80%) fol-

lowed by communication with seniors/colleagues (12%) and televi-

sion (8%). Majority (95%) preferred print format in local language.

Only 20% have access to internet sources.75% had encountered

information gaps on their work due to lack of information.

Important barriers to access of information were technical language,

lack of practical information and time. Age, educational status and

computer literacy were found to be significantly associated with

health information seeking behaviour.(P<0.05).

CONCLUSIONS: Health information seeking among grass root

workers is passive rather than active. Efforts are needed to provide

grass root healthcare workers with necessary health information

resources and infrastructure that provide quality health care to a

large segment of population.

Poster Number: T171 Abstract #: 3030Rocky Mountain Spotted Fever Reemergence inchildren from Sonora, Mexico. 2004–12.

G. Alvarez-Hernandez, PhD1, C. Murillo-Benitez, MD2, M. C.Candia-Plata, PhD3 and M. Moro, PhD4

1Universidad de Sonora, Hermosillo, Mexico, 2Instituto Tecnologico y deEstudios Superiores de Monterrey, Monterrey, Mexico, 3Universidad deSonora, Hermosillo, Mexico, 4National Institutes of Health,Bethesda, MD

INTRODUCTION: Rocky Mountain spotted fever (RMSF) is a re-

emerging disease in Sonora, Mexico, spreading from the south of the

state to the counties bordering the United States. The disease remains a

life-threatening condition, particularly for children living in social dep-

rivation. This study aims to describe RMSF’s epidemiologic and clini-

cal profiles and the medical factors associated with mortality.

METHODS: We carried out a retrospective analysis of 90 children

consecutively hospitalized at the major public pediatric hospital of

Sonora, between January 2004 and March 2012. Descriptive statis-

tics and multiple logistic regression were used.

RESULTS: Over the study period, an increasing trend in the inci-

dence and the case fatality ratio (CFR) was observed. Twenty-six

patients died due to RMSF. The overall CFR was 28.9% for the

entire period, while the highest CFR (41.9%) occurred in patients

less than four years of age. Children were hospitalized after a

median of six days with symptoms, mainly fever (100%), rash

involving palms and soles (89%) and headache (75%). A higher pro-

portion (88%) of fatal cases had very low platelet counts (29,403/

mm3), and 73% showed serum sodium concentrations <135 mEq/

dL. Acute kidney injury was significantly higher in the fatal group

than in survivors (69.0% vs 6.0%, respectively, P<0.001) and

remained as the only risk factor that was independently associated

with mortality (OR a¼28.08, 95% CI (6.2, 127.5)).

CONCLUSIONS: RMSF is a public health problem in Sonora,

Mexico; the timely diagnosis and initiation of disease-specific treat-

ment remains as a challenge for the medical community. Acute kid-

ney injury is a life-threatening condition associated with the high

fatality rate observed in children

Poster Number: T142 Abstract #: 3146Significant Associations of Abdominal Obesitywith LYPLAL1 Polymorphism (rs4846567)among a General Japanese Population:A Cross-Sectional Study.

T. Tamura, PhD, E. Morita, PhD, S. Kawai, PhD, T. Sasakabe,

N. Fukuda, MS, Y. Sugimoto, MS, S. Suma, J. Mokuno, MD,

H. Nakagawa, PhD, A. Hishida, PhD, R. Okada, PhD, M. Naito,

PhD, N. Hamajima, PhD and K. Wakai, PhD

Nagoya University Graduate School of Medicine, Nagoya, Japan

INTRODUCTION: Obesity is one of the most important health

concerns in the world. To date, a lot of studies including genome

wide association studies (GWAS) have reported that single nucleo-

tide polymorphisms (SNPs) were clearly associated with obesity.

Among them, lysophospholipase-like-1 (LYPLAL1, rs4846567) is

reported to be associated with female-biased waist circumference

(WC). We therefore investigated whether this SNP is associated with

WC, and confirmed an interaction with sex.

METHODS: Study subjects were 5131 participants aged 35–69

years (1458 males and 3673 females) in the Daiko Study, a part of

the Japan Multi-Institutional Collaborative Cohort Study (J-MICC

Study). WC was classified into two groups,<or �85 cm (in male)

and<or �90 cm (in female), respectively, based on the criteria of

international diabetes federation (IDF) 2005 for Japanese.

RESULTS: Genotype frequencies of LYPLAL1 (rs4846567) were

39.9% for GG, 46.2% for GT and 13.9% for TT, and were in Hardy-

Weinberg equilibrium (P¼0.513). The allele frequencies were 0.630

for G and 0.370 for T. The multivariate-adjusted OR and 95% confi-

dence interval (CI) for high WC was 1.63 (95% CI 1.03–2.58) for those

with TT relative to those with G allele in men, while the OR was 0.67

(95% CI 0.41–1.10) in women. The crude means 6 standard deviation

(SD) of WC were significantly different between genotypes as a

whole (P¼0.02), and the interaction with sex was also highly signifi-

cant (P<0.001). The means 6 SD in men were 83.4 6 8.4 for GG,

84.5 6 8.3 for GT and 85.3 6 9.2 for TT, while those in women were

78.0 6 8.7 for GG, 78.6 6 9.0 for GT and 77.9 6 8.5 for TT.

CONCLUSIONS: Our study showed that LYPLAL1 (rs4846567)

was significantly associated with female-biased waist circumference


Dow



in a general Japanese population. Stratified analyses including other

LYPLAL1 SNPs related with obesity were required.

Poster Number: T27 Abstract #: 3075Assessment of the Effect of the National Project‘Health’ on Survival among Patients with RectalCancer in the Arkhangelsk Region,Northwest Russia.

D. Dubovichenko1, L. Valkova1, M. Levit, PhD1, T. Izmailov,PhD2, D. Korotov1, A. M. Grjibovski, PhD1,3, Z. Manambaeva,PhD4, L. Pak4, A. Masadykov4 and M. Valkov, PhD1

1Northern State Medical University, Arkhangelsk, Russia, 2Russian

Scientific Center of Roentgenoradiology, Moscow, Russia, 3NorwegianInstitute of Public Health, Oslo, Norway, 4Semey State MedicalUniversity, Semey, Kazakhstan

INTRODUCTION: The incidence of and mortality from rectal can-

cer (RC) in Russia is higher than in most countries in Europe and

North America. Moreover, most cases are diagnosed at late stages.

The National project ‘Health’ was initiated in 2007 to improve pri-

mary health care and could therefore increase the rate of early detec-

tion of RC in Russia. The aim of the study is to assess whether of

RC patients increased in Russia after initiation of this project.

METHODS: Data on all cases of rectal cancer (ICD codes:

C20–C21) were obtained from the Arkhangelsk Regional Cancer

Registry for the period 2000–11. One- and five-year survival was

calculated using life-tables. The effect of the implementation of the

National project ‘Health’ on survival was assessed by comparing

survival in time-periods before and after the initiation of the project

adjusted for the stage at diagnosis, age, gender and place of resi-

dence using multivariable Cox proportional hazard analysis.

RESULTS: Altogether, there were 2440 cases of and 1385 deaths

from RC during the study period. One- and five-year survival

increased considerably from 66.2% and 29.8% in 2000–06 to

71.4% and 47.3% in 2007–11, respectively. Crude analysis revealed

significant difference in survival before and after initiation of the

National project ‘Health’ (HR¼0.65, 95% CI 0.58–0.73).

Adjustment for other factors did not change the association

(HR¼0.64, 95% CI 0.57–0.72). Men (HR¼0.87, 95% CI

0.77–0.97) and urban residents (HR¼0.86, 95% CI 0.77–0.96) had

better survival than women and rural residents. No interaction

between time-periods and other variables was observed.

CONCLUSIONS: After initiation of the project, the survival of

patients with RC increased by 56% on average. Adjustment for stage

did not alleviate the association between the project and survival sug-

gesting that better treatment, but not earlier diagnosis was the main

contributing factor to the improvement in survival of patients with RC.

Poster Number: T72 Abstract #: 3331Home Fall Injuries in SouthWestern Nigeria.

E. Owoaje, MD1, O. Uchendu, MPH1 and O. Iken, BS2

1University College Hospital (UCH), Ibadan, Nigeria, 2University CollegeHospital, Ibadan, Nigeria

INTRODUCTION: Unintentional home injuries constitute signifi-

cantly to the public health burden of injuries in all age groups especially

in children and the elderly in developing countries. Fall-related injuries

are among the most serious and most common medical problems

experienced by older people. This study compared fall injuries among

household members in rural and urban areas of south western Nigeria

METHODOLOGY: A community based cross sectional survey

using a multi-stage sampling technique was used to select 4433 indi-

viduals from 1015 households. Information was obtained on socio-

demographic characteristics and unintentional home injury events.

Multivariate analysis was done to determine factors associated with

unintentional home injuries.

RESULTS: There were 2340 (52.8%) and 2093 (47.2%) respond-

ents from the rural and urban areas, respectively. The urban fall

incidence of 195 injuries/1000 persons per year was significantly

higher than the 150 injuries/1000 persons per year in the rural popu-

lation (P¼0.044).The incidence of fall related disability was 11.9/

1000 persons per year.

Overall, children aged 0–4 years, 64 (13.2%) experienced falls injury

compared to all age groups (P<0.001), with a higher proportion in

the rural area 51 (16.2%) than urban (7.6%).In the rural area,those

having some education (OR¼0.4; 95% CI¼0.27–0.59)and currently

married (OR¼0.39; 95% CI¼0.15–0.97) were less likely to have

fall injuries. In the urban area, age, and occupational grade of

respondents were statistically significant with the occurrence of injury;

overall,180 (71.1%) people had falls that required treatment.

CONCLUSION: Fall injury in the home environment is significant

as it affects the youngest age group the most. Further studies are

needed to provide evidence based public enlightenment and develop

policies that will reduce the burden of fall injuries in the home.

Poster Number: T18 Abstract #: 2862Smoking among Teenage Children AttendingSchool: A Systematic Review of ObservationalStudies with Assessment of Quality of Studies.

D. S. Rozi, PhD1, D. G. Lancaster, PhD2 andD. S. Mahmud, PhD1

1Aga Khan University, Karachi, Pakistan, 2Lancaster University,

Lancashire, United Kingdom

INTRODUCTION: This systematic review describe the methods of

design and analysis currently adopted in survey research published

in the literature of school based observational studies for smoking,

and identify the common pitfalls made by researchers.

METHODS: This systematic review followed Cochrane style meth-

odology in five steps: i) setting selection criteria for studies and con-

ducting a literature search, ii) review of the abstracts, iii) review of

the complete articles, iv) data extraction and quality assessment of

included studies, and v) synthesis of studies.

RESULTS: Out of the 292 abstracts retrieved, 45 articles were

selected for the final review. We found inconsistencies in the defini-

tion of smoking behavior or the tool for measuring adolescent smok-

ing was not standardized across studies which impedes

generalizability. Despite the different study designs, methods of

analysis and variety of covariates our results indicate that individual

level factors have particular importance, but environmental level

factors are also not negligible in studying the etiology of smoking.

Findings showed that parents, siblings and peers are powerful influ-

ences for adolescent smoking. The exclusion of adolescents who did

not attend school on the day of the survey or who dropped out of

school may produce biased results and also have an effect on

generalizability. Many studies failed to measure important con-

founding factors and the sample size was often improperly reported.


Dow



Hierarchical linear modeling, random effects modeling and struc-

tural equation modeling were employed in relatively few studies.

CONCLUSIONS: Our results highlight concerns regarding the anal-

ysis of data from complex surveys. Many studies ignored the envi-

ronmental (e.g. retailer near to school, social environment of school

and implementation of smoking policies at school) effects in their

studies which may produce misleading inferences. The results of

multilevel analysis help us to understand school level effects and

between schools variance provides more insightful information on

school level covariates that influence adolescents smoking habits.

Poster Number: T68 Abstract #: 3429Hospitalization Trends among Adults withHypertension and Diabetes: Population-BasedHealth Surveys—Sao Paulo, Brazil.

S. R. Stopa, MS1, C. Monteiro, MS2, N. Segri, PhD3, R. Gianini,DO2, M. Goldbaum, PhD4, M. Barros, PhD5 and C. Cesar, PhD1

1School of Public Health, University of Sao Paulo, Sao Paulo, Brazil,2University of Sao Paulo, Sao Paulo, Brazil, 3Federal University of MatoGrosso—UFMT, Cuiaba, Brazil, 4University of Sao Paulo, Medical School,Sao Paulo, Brazil, 5State University of Campinas—UNICAMP, Campinas,Brazil

INTRODUCTION: Non-communicable chronic diseases, particu-

larly hypertension and diabetes mellitus, are the major cause of mor-

bimortality among adults in Brazil. Such diseases are public health

priorities, since they impact on a higher demand in health services

with hospitalizations for complications control.

METHODS: Data from two cross-sectional population-based

Health Surveys (ISA-Capital 2003 and ISA-Capital 2008), carried

out by home interviews. Target population were from both sexes,

20 years old and greater, who reported suffering from hypertension

and diabetes mellitus. Using a combined dataset and Stata 11, preva-

lence estimates were performed and prevalence ratios were calcu-

lated by Poisson regression considering the complex sample design

of the surveys.

RESULTS: In 2003, there was no significant difference in hospital-

ization proportion according to age and schooling. Although, a stat-

istically significant difference (P¼0.016) was found in

hospitalization percentages according to sex in population with self-

reported hypertension (men: 17.6% versus women: 8.6%). In 2008,

there was no significant difference in hospitalization proportions.

Also, there was a significant increase in hospitalizations proportion

covered by private health plans in the period 2003–08: for hyperten-

sive and/or diabetics, the percentage in 2008 (26.6%) was 3.06

times the percentage in 2003 (8.7%) (P¼0.020). Furthermore, there

was an increase in hospitalizations proportion covered by health

insurance companies in hypertensive women, from 5.6% in 2003 to

30.1% in 2008 (P¼0.008) and in hypertensive/diabetics women,

from 4.5% to 33.0% (P<0.001).

CONCLUSIONS: The difference in hospitalization proportion in

women may be explained by coverage expansion of private health

plans in Brazil. Recently, there has been an increase in companies

offering private health plans to employees, which may have contrib-

uted to our findings. Brazilian health system has seen an increasing

number of private services hiring to comply with health demands.

Public service cannot fully attend population health demands, which

violates the principle of universality proposed by the Brazilian

Unified Health System.

Poster Number: T198 Abstract #: 2863How Types of Ageism Affect the Health of OlderKoreans?

H. Chun, PhD1 and I. H. KIM, PhD2

1Jungwon University, Chung-buk, South Korea, 2Social and EpidemiologicalResearch Centre for Addiction and Mental Health, Toronto, ON, Canada

INTRODUCTION: Although respecting older people was tradi-

tionally considered commonplace in family and society, getting

older in modern Korea is often associated with unpleasant experien-

ces of personal and institutional age-related discrimination, com-

monly known as ‘ageism’. This study aims to examine how Korean

elders experienced types of Ageism and how this conversely relates

to mental, physical, and self-rated health.

METHODS: Data was gathered from a clustered sample of 638 peo-

ple aged 60–89 via face-to-face interviews. Ageism was measured

using a 20-item questionnaire from ‘The Ageism Survey’ by

Palmore. An exploratory factor analysis was performed to classify

various types of Ageism. Health outcomes included depressive

symptoms for mental health, using CES-D 20 Questions. Physician-

diagnosed hypertension and self-rated health were also recorded.

RESULTS: Of the 20 Ageism items, the results yielded four factors:

ignorance (a=.76), stereotype (a=.80),employment (a=.74), and

healthcare (a=.63). Gender, marital status, education, and residence

were found to account for some variances in types of Ageism. In the

effect of four types of ageism on health, three types— ignorance,

stereotype, and health care—were all positively related to depressive

symptoms. Only employment-related ageism increased the risk of

developing hypertension, whereas healthcare-related ageism was sig-

nificantly associated with poor self-rated health.

CONCLUSIONS: This study illustrated the differential effects that

types of ageism have on the poorer health amongst older Koreans.

The harmful impact of ageism on health needs to be better under-

stood with an in depth study on the types of Ageism and the differ-

ent health outcomes that arise as a result of this.

Poster Number: T193 Abstract #: 3204Smoking Cessation and Gender: APopulation-Based Study Conducted in Campinas,Sao Paulo, Brazil.

T. Bastos, MD and M. Barros, PhD

State University of Campinas—UNICAMP, Campinas, Brazil

INTRODUCTION: The aim was to evaluate the smoking cessation

rate between men and women and identify its association with soci-

odemographic characteristics, health-related behaviors, health status

and use of health services, according to gender, in Campinas, Sao

Paulo, Brazil.

METHODS: Cross-sectional study, whose data from the health sur-

vey conducted in the city of Campinas, Sao Paulo, Brazil, in 2008

(ISACamp 2008). Records of a representative sample of 1,118

respondents, 18 to 59 years were included. The dependent variable

was smoking, which the smoking cessation rate was obtained

through the ratio between the number of former smokers and the

number of current and former smokers. The magnitude of the asso-

ciations between the dependent and independent variables were

assessed using multivariate Poisson regression.


Dow



RESULTS: The smoking cessation rate in the male population was

32.7% (95% CI 26.8-39.1) and 39.0% (95% CI 30.0-49.0) in the

female population. In the segment of men Poisson regression, adjusted

for age and education, showed an inverse association between alcohol

consumption and smoking cessation in two or more times a week

(PR: 0.53; 95% CI 0.30–0.94) and direct association of smoking ces-

sation with the presence of at least one chronic disease (PR:2.29; 95%

CI 1.49–3.54), morbidity in the last 15 days (PR: 1.63; 95% CI

1.07–2.47) and drug use in the past three days (PR: 1.81; 95% CI

1.08–3.04). Among women, the cessation rate was directly associated

with obesity (PR: 1.93; CI 1.19–3.15), presence of two or more

chronic diseases (PR: 2.15; 95% CI 1.29–3.58), hypertension (PR:

1.94; 95% CI 1.18–3.17), allergy (PR: 1.67, 95% CI 1.03–2.69).

CONCLUSIONS: There was no statistically significant difference

between the genders, but in each subgroup analysis showed that,

except for number of chronic diseases, different factors were associ-

ated with smoking cessation rate in the analysis by gender. This indi-

cates that the particularities of each gender should be considered in

planning policies and actions that encourage cessation, in order to

achieve increasingly satisfactory levels in tobacco control.

Poster Number: T54 Abstract #: 2990Smoking Kills at Least One Out of Two

Smokers?–A 10-year Follow-up Study in Hong

Kong, China.

T. H. Lam, MD1, L. Xu, PhD1, C. M. Schooling, PhD2,

W. M. Chan, BS3, R. S. Y. Lee, BS3 and G. M. Leung, MD1

1The University of Hong Kong, Hong Kong, Hong Kong, 2The Universityof Hong Kong, Hong Kong, China, 3Department of Health, Wanchai,Hong Kong

INTRODUCTION: World Health Organization states ‘Tobacco

kills up to one in every two users’. This absolute risk of 1/2 was

based on a relative risk (RR) of 2 and the attributable fraction (AF)

in the exposed of 50% ((RR-1)/RR)) for total mortality due to

smoking. Recent large UK and US cohort studies have shown an RR

of 3, meaning two out of three smokers will be killed by smoking.

RRs could be under-estimated partly because few studies examined

oldest old smokers separately. We examined RR and AF of total

mortality due to smoking in the older (65–84y) and oldest old

(85þy) people in Hong Kong, China.

METHODS: Multivariable Cox regression was used to assess the

risks of total mortality from smoking using a population-based pro-

spective cohort of 65,510 Chinese aged 65þ years enrolled from

1998 to 2001 and followed until May 2012.

RESULTS: For participants aged 65–84 years, after adjustment for

sex, age, education, social security assistance, housing type, monthly

expenditure, alcohol use and health status, the RR was 1.93 (95%

confidence interval (CI) 1.84–2.03), corresponding to an AF of

about 50%. For 85þ years, the adjusted RR was 1.29 (1.05–1.58),

corresponding to an AF of about 25%.

CONCLUSIONS: Our study has shown that smoking kills one

out of two older smokers and one out of four oldest old smokers

in Hong Kong which is at an earlier stage of epidemic (by about

20 years) than the West. Global disease burden estimates

due to smoking based on an RR of two may need to be revised

upwards.


Dow



Tuesday Poster Session - Oxford Academic

Documents