Epi 712 – Intermediate Epidemiology Patty Kissinger, Ph.D. (Prof) Jeff Kopicko, MSPH (TA) Meg O’Brien, MPH (TA)
Jan 20, 2016
Epi 712 – Intermediate Epidemiology
Patty Kissinger, Ph.D. (Prof)
Jeff Kopicko, MSPH (TA)
Meg O’Brien, MPH (TA)
Objectives
• To discuss course logistics
• To review the basics of Epidemiology
• To describe the difference between descriptive and analytic epidemiology
• To discuss the basic study designs
• To discuss criteria for causality
• Readings: Szklo and Nieto Chapter 1
Definition of Epidemiology
• The study of the distribution and determinants of health-related states or events in specified populations and the application of this study to control health problems.
Classifications of Epidemiology
• Descriptive –– Used to describe person, place and time– Used to generate hypotheses
• Analytic– Used to test hypotheses
Uses of Epidemiology• Determine etiologic or causal factors• Describe factors associate with adverse
conditions• Community diagnosis of distribution of
diseases• Predicting disease occurrence, impact
and distribution• Estimating individuals risk of disease
Uses of Epidemiology (con’t)
• Evaluating therapeutic and intervention activities
• Measurement of efficacy of health measures
• Studying of historical disease• Identifying disease syndromes• Planning for current health needs• Predicting future needs.
Person, Place, Time
• Person - age, sex, race/ethnicity, martial status, occupation, education, socio-economic status
• Place - hospital-based, community-based, regional
• Time - seasonal, secular, am/pm/noc
Examples of descriptive epidemiologic findings
Gonorrhea (cases per 100,000)The Hidden Epidemic, Institute of Medicine, 1997
0 50 100 150 200
U.S.
England
Canada
Autstralia
Denmark
Germany
Sweden
Possible interpretations
• May be confounded by detection bias (i.e. More people get tested in U.S and policy in Europe is to just treat syndromically and not test)
• May be reporting bias (gonorrhea is not reported as well in countries other than the U.S.)
• The U.S. people are more likely to have unprotected sex.
Abnormal Pap results by age (U.S., 1991-1993, NBCCEDP)
0
2
4
6
8
10
12
14
< 30 30-39 40-49 50-59 60-69 >69
Possible interpretations
• Interpretation is difficult because we don’t know if it is a rate or just cases reported
• If it is cases, then it is not interpretable without know the population distribution (i.e. denominator information)
• If it is rates, then women < 30 are at highest risk (maybe due to Human Papillomavirus)
By 12th grade, nearly 70 percent of high school students have had sexual intercourse (YRBS,
1993)
3846
58
68
0
10
20
30
40
50
60
70
80
9th 10th 11th 12th
Possible Interpretation
• Again, we don’t know if it is rates or cases
• It is prevalence data, and we don’t know when the first sexual act occurred, but we know that 38% of the 9th graders already had sex so, sex education starting at high school is too late.
Definitions• Endemic - a persistent level of occurrence
with low to moderate disease level• Hyper-endemic - persistently high level of
occurrence• Sporadic - irregular • Epidemic (outbreak)- occurrence of disease
is in excess of expected levels• Pandemic - epidemic spreads over several
countries or continents
Role of an Epidemiologist
• Surveillance
• Outbreak investigation
• Hypothesis testing
• Evaluation
• Communication
Surveillance
The importance of surveillance for Management
Purposes of surveillance• setting of priorities• planning and allocating resources for service• defining population subgroups and risky behaviors for
targeted interventions• directing public health policy• informing diagnostic and therapeutic practice• evaluation of interventions• stimulating further research
Characteristics of Surveillance
• Purpose is to monitor trends of disease
• Usually large data sets
• Can be used to identify persons at risk for a disease
Possible Interpretations
• This is a better slide because we know it is rates.
• We can probably draw a conclusion that the risk of breast cancer increases with age.
Possible interpretations
• We know that the rates of diabetes is going up, we assume it is type II or adult onset.
• We don’t what factors are related but we could hypothesis that obesity is increasing or there is an aging of the cohort since these are not age adjusted
Possible interpretations
• If you look at the last two slides and imagine them being super-imposed upon one another you can see that physical inactivity and cardiovascular disease co-vary.
• You could hypothesis that physical inactivity may be associated with cardio-vascular disease
Possible interpretations
• There does seem to be a trend for an increase in smoking among teenagers
• A statistical test, such as a Chi-square test for trends would better help you to decide.
Possible interpretations
• We should probably be most worried about the heterosexual community because it is steadily increasing and it is a broader base of the population (~90% of the population)
• This does not tell us anything about incidence, it is only a description of prevalence.
Confounding
E D
C
E=exposure
C=confounder
D=disease
Birth Cohorts
• Age is a strong risk factor for many health outcomes
• For many diseases, exposures have a cumulative effect that are expressed over long periods of time.
Definitions
• Age effect – Change in the rate of a condition according to age irrespective of birth cohort and calendar time.
• Cohort effect – Change in the rate of a condition according to year of birth, irrespective of age and calendar time.
• Period effect – Change in the rate of a condition affecting an entire population at some point in time, irrespective of age and cohort effect
Surveillance Systems
• Are active or passive
• Can be for a whole population or for a selected group (sentinel)
Outbreak Investigation
John Snow (The Father of Field Epidemiology) Cholera Outbreak Investigation 1854
Possible Interpretations
• For both the community water sources (representing the poorer community) and individual home water sources (representing the higher socio-economic groups), Southwark rates were greater.
• Intervention should be started there.
Outbreak investigations
• A good idea to plot out the cases (better rates) and then determine when interventions were implemented to see if any helped.
Criteria for Causality• Strength of association - larger ratio between diseased and
non diseased greater the likelihood of causation• Biological credibility – the association has to make sense
biologically• Consistency of findings - association found in one study is
found in others.• Dose-response - with increasing levels of exposure to the
factor, a corresponding rise in occurrence of disease occurs• Specificity - The extent to which the occurrence of a variable
can be used to predict the occurrence of an outcome.• Time sequence- exposure precedes the disease
Study Designs
Descriptive Analytic Experimental
correlational
case report/case series
cross-sectional
case control
cohort
clinical trial
community trial