Social disparities in cancer among Mexican women living in Mexico and in the United States Gloria Giraldo, MPH DrPH Program School of Public Health UCLA.

Social disparities in cancer among Mexican women living in Mexico and in the United States

Gloria Giraldo, MPHDrPH ProgramSchool of Public HealthUCLA

Proposed “cancer disparity” definition by Nancy Krieger (2005)

'Social disparities in cancer refer to health inequities spanning the full cancer continuum, across the lifecourse.

These cancer disparities involve social inequalities in the prevention, incidence, prevalence, detection and treatment, survival, mortality, and burden of cancer and other cancer related health conditions and behaviors.

They arise from inequities involving, singly and in combination, adverse working and living conditions and inadequate health care, as linked to experiences and policies involving socioeconomic position (e.g., occupation, income, wealth, poverty, debt, and education) and discrimination. This discrimination, both institutional and interpersonal, can be based on race/ethnicity, socioeconomic position, gender, sexuality, age, language, literacy, disability, immigrant status, insurance status, geographic location, housing status, and other relevant social categories.'

Cancer disparities in the U.S.Since the early 20th century,

substantial socioeconomic and racial/ethnic disparities in breast and cervical cancer incidence, survival and mortality have been documented in the United States (US) (Krieger, 2005).

Latinas are diagnosed with cervical cancer almost twice as often as non-Latina White women.

Saraiya and colleagues (2007) reported that the incidence rate declined from 10.2 cases in 1998 to 8.5 in 2002 for the entire population, the average annual incidence rate was highest among Latinas (14.8).

The incidence of cervical cancer for all groups between 1998 and 2003 was 8.9.

Furthermore, the overall cervical cancer mortality rate was 2.7 and the median age of death was 57; for Latinas the rate was 3.4 and the median age was 51.

Mexican immigrant women A study that examined nativity status and prevalence of

cervical cancer screening found that 17% of immigrant women from Mexico who have lived more than 25% of their lives in the US have never been screened for cervical cancer.

32% of those who have lived in the US less than 25% of their lives have never been screened (4).

Remarkably, 55% to 60% of cervical cancer cases arise in women that are never or rarely screened (5).

To put these numbers in perspective, the Healthy People 2010 goal for women over 18 for ever having a Pap test is 97%. Currently, the rate of ever having a Pap test for Latinas in general is 88%; however, low-income immigrant Mexican women lag behind the national screening goal potentially by as much as 29 percentage points (6).

Screening plays a pivotal roleA robust body of literature has

established that advanced stage diagnosis, shorter survival and higher mortality due to cervical cancer among Latinas are largely due to inadequate or lack of screening.

The picture in MexicoInternational data from the 1980s

showed that Mexico had the highest cervical cancer mortality rates among fifty countries from three continents (Europe, the Americas and Asia) (Boring, 1992). Mexico’s mortality rate in 1986 was 14.7.

In 1989 it climbed to its highest level 16.1 per 100,000.

In 2006, cancer mortality rate had decreased to 9.9.

The picture in MexicoCervical cancer mortality risk is

three times higher in rural areas, as compared to urban zones, and women living in states where socio-economic development is lower have the highest mortality risks, as compared with women living in Mexico City (Lazcano-Ponce,2003)

Breast Cancer in MexicoBreast cancer mortality rates in Mexico

show a substantial increase in the last five decades.

Between 1955 and 1960, data on breast cancer show 4 deaths per 100,000 women. It then increased to approximately 9 per 100,000 and has remained somewhat stable since then.

Breast cancer is responsible for a high number of premature deaths since 60% of deaths are in women between the ages of 30 and 59.

Breast cancer in U.S. Latinas

Although breast cancer rates are lower in Latinas than in Non-Latina White women in the U.S., published data indicate that the disease presentation among Latinas includes: Earlier age at diagnosis, larger tumor size, more advanced stage, higher proportion of adverse prognostic indicators, co-morbidities, poorer overall survival (Howe HL, Wu X, Ries LA, et al, 2006).

ObjectiveMy dissertation will explore social

disparities in cancer screening rates and cancer-related behaviors among Mexican women living in Mexico and in the United States taking into consideration health system variables in both countries.

ImplicationUnderstanding cancer screening

trends and cancer related behaviors of Mexican women in the larger binational context may  shed new light on the social processes that influence cancer screening disparities in the US with implications for public health practice and theory refinement in the field of cancer control.

Studying breast and cervical cancer screening binationally is of paramount importance because Mexico is currently experiencing a cancer epidemiological transition

RationaleIn the realm of cancer research, the type

of cancer screening programs in the place of origin, cancer knowledge or educational campaigns, the type of access to screening and to cancer care, and preventive health behaviors are all contextual variables that will impact the type of knowledge, attitudes and experiences with which the immigrant woman arrives to her new destination and influence her ability to navigate her new healthcare context.

AimTo compare cancer screening

rates and prevalence of cancer-related behaviors and health services variables (healthcare coverage and access) of Mexican women of different socioeconomic backgrounds living in Mexico and in the United States.

DataEncuesta Nacional de Salud y

Nutrición (ENSANUT 2006) (Mexico’s National Health and Nutrition Survey). Comprehensive health and nutrition surveys have been conducted in Mexico in the last 20 years. ENSANUT 2006 is the third national survey of this nature.

CaliforniaCalifornia Health Interview Survey

2007 – The California Health Interview Survey (CHIS) is a population-based telephone survey of California’s population conducted every other year since 2001. CHIS is the largest health survey conducted in any state and one of the largest health surveys in the nation. This survey contains the largest sample of Mexican-origin individuals in the United States.

Variables

Independent variables:Socioeconomic related questions

(education, income, poverty level, rural vs. urban)

Language use and health literacy questions (and length of residence in the US for CHIS only)

Health coverageReproductive health-related

questions

Dependent variables

General health conditionPap-smear related questionsMammogram-related questionsSmoking and drinking related

questionsFruit and vegetable consumption

questionsExercise related questions

Statistical AnalysesI will initially estimate the

reported Pap test and mammography screening rates by age, socioeconomic status, educational attainment as well as other demographic, and health-related individual-level covariates in bivariate analyses.

Statistical Analyses A test of association will assess

the overall statistical significance of each potential factor, such as the association of employment or health insurance status with Pap test and mammography use, after adjusting for age

Statistical analyses Conduct multivariate logistic regression analyses to

adjust for factors indicated from the bivariate analysis to have a significant association with Pap test or mammography screening rates, and use adjusted Wald F tests to determine the significance of the model covariates.

At first, I expect to obtain intermediate reduced models that will include significant individual-level covariates or factors related to the Pap test and mammography screening outcomes.

I will test for potential two-way interaction terms. Finally, reduced models will include all socioeconomic covariates, and all other significant individual-level main effects, to provide fully adjusted screening estimates for the Pap test and mammography screening outcomes and other cancer-related health behaviors.

LimitationsThe main limitations of this study are

inherent to international comparisons utilizing two different sources of data with different data collection methodologies. However, at this point in time, there are no binational studies on cancer screening. Therefore, this study will highlight the need to not only conduct a binational cancer study but will promote an agenda of binational collaboration in cancer control.