Center for the Elimination of Minority Health Disparities

Current Research Projects

Overcoming Barriers to African American Women's Reproductive Healthcare Seeking

Annis Golden and Anita Pomerantz

African American women suffer significant disparities in disease incidence and health outcomes in relation to reproductive health, including HIV/AIDS, STIs, breast and cervical cancer. African American women who live in small towns and cities encounter barriers in obtaining adequate reproductive healthcare that are specific to their social environments: limited number of reproductive healthcare providers, difficulty in traveling to and from provider's locations, and privacy concerns, in addition to lack of knowledge about preventive reproductive healthcare and concerns about obtaining follow-up healthcare services in the event that cancer is detected, which are faced by low income African American women more generally. Publicly funded providers, whose missions explicitly encompass providing education and services to the underserved, are an important source of comprehensive reproductive healthcare services in both urban settings and smaller towns and cities. In the latter setting, however, these organizations face challenges in carrying out their mission that are directly linked to the social environment. These challenges include addressing patients' privacy concerns and problems with organizational image created by their association with stigmatized reproductive healthcare services, such as STI treatment and abortion. The goal of the proposedstudy is to identify successful strategies that can be applied in similar settings for overcoming barriers that contribute to African American women in smaller towns and cities not obtaining adequate reproductive healthcare services and to improve African American women's reproductive health. THe proposed study's location is Hudson, New York, a small city in New York state with a 25% African American population, which served as the setting for a pilot project sponsored by the EXPORT Center at the University of Albany in 2006. The study will be undertaken in collaboration with Upper Hudson Planned Parenthood (UHPP) and local community organizations, and will utilize an innovative multi-method design, informed by principles of community-based participatory research design and grounded theory methodology, to conduct and evaluate the efficacy of interventions providing community health education and transportation. UHPP will collect quantitative data on clinical participation and its relationship to the interventions; qualitative interview data will be collected to evaluate the interventions' impact on healthcare seeking behavior.

Reproductive healthcare seeking has implications for disparities in HIV/AIDS, STIs, and breast and cervical cancer. Small towns and cities, home to increasing numbers of African American women, present unique contextual challenges for reproductive health promotion. This study will evaluate the impact of communicty-based education and transportation interventions on healthcare seeking, with the goal of identifying effective health promotion strategies that can be reproduced in similar settings and improving women's health.

Environmental Contaminants and Reproductive Health of Akwesasne Mohawk Women

Lawrence Schell, Ph.D ; Mia Gallo, Ph.D and David Carpenter, M.D.

Considerable concern exists over possible effects of endocrine compounds such as polychlorinated biphenyls (PCBs) and lead (Pb) on human reproductive health. Effects on mammalian reproduction are known, but evidence regarding non-occupational exposure to environmental toxicants on parameters related to human fertility is lacking. Reproductive health is of great concern to the Akwesasne Mohawk Nation which is located on the St. Lawrence River. Like many minority communities, Akwesasne is located near a site of major environmental contamination; it is adjacent to a federal and two state Superfund sites. Important local food sources are contaminated, and exposure is perceived as a threat to the community already under pressure of assimilation and dispersal. Good reproductive health is seen as essential to the survival of the nation as a culture and a people. The project builds on a 12 year collaborative relationship between the Akwesasne Mohawk Nation and the University at Albany. This study aims to determine the effects of PCBs, other persistent organic pollutants, and lead (understudied for its endocrine disrupting properties) on characteristics of the menstrual cycle among Mohawk women between the ages of 20 and 35. It will follow 180 women through one menstrual cycle collecting blood, urine and daily saliva samples to investigate the relationship of PCB congeners and other toxicants to:

  1. Gonadal function as indexed by steroid estradiol and progesterone levels throughout the cycle measured in saliva
  2. Pituitary function indexed by gonadotropin levels
  3. Other characteristics of the menstrual cycle reported through diary and questionnaire

The project is innovative in its' simultaneous consideration of multiple toxicants with endocrine disrupting properties, the congener specific analysis of PCBs that allows testing structure-function relationships with regard to reproductive effects and the consideration of these effects within the context of other common influences on reproductive parameters (measures of thyroid function, autoimmune disease, overweight, and of physical activity) using standard statistical methods and growth curve analysis to take full advantage of the density of data obtained through the cycle and which have not been applied before in studies of this matter.

This study will determine the effects of specific toxicants on menstrual cycle characteristics which impacts fertility, and thereby, address public health concerns regarding effects of toxicants on reproduction and diseases of the reproductive system.

Decomposing Racial/Ethnic Disparities in Health

Kajal Lahiri, Ph.D and Pinka Chatterji, Ph.D

The overarching goal of the proposed study is to examine the nature and determinants of health disparities not only between but also within racial/ethnic groups (non-Latino whites, Latinos, African-Americans, Asians, American Indians/Alaskan Natives) in the US. We focus on community size (e.g., urban vs. rural) as a potential determinant of disparities. Data for the study will come from the 1994-2008 waves of the national Behavioral Risk Factor Surveillance System (BRFSS). To measure health using the BRFSS, we propose to construct indices of individual quality of health. These indices will be based on a 5-category self-assessed health status measure and other binary ill-health indicators (e.g., diabetes, obesity, asthma etc.), supplemented by several objective determinants of overall health including different diseases/risk factors and socio-demographic characteristics. The first goal of the study is to estimate and compare quality of health by racial/ethnic group, by state, and by community size. Our second goal is to estimate three types of health inequality-total health inequality, income related health inequality, and racial/ethnic inequality in health- using the Gini coefficient, concentration index, and disparity index, respectively. We also will examine these comparisons by year, to determine whether racial/ethnic and urban/rural differences in quality of health are worsening or improving over time. Finally, our third aim is to decompose both total and income related health inequalities into socio-demographic and community-level factors using Oaxaca-Blinder type decompositions for non-linear models. This approach will be used to quantify the separate contributions of group differences in measurable characteristics such as education, income, work status and disability on health disparity (which we call indirect discrimination), and also due to the difference in the health generating processes of different groups (which we call direct discrimination).

The proposed project is intended to generate new, policy-relevant information on health disparities based on the most recent national data available. Rather than focusing public health initiatives towards specific minority groups or regions broadly, the proposed study will inform policy designs to target subgroups within such groups and regions for more fine-tuned public policy interventions.

Pilot Projects- Final Report


The Establishment of a Minority Health Disparities Database for the Communities of Upstate New York
Richard Alba and David Strogatz
Progress report:

a. Specific aims
Little is known about health disparities among racial/ethnic groups in smaller cities, such as those that predominate in upstate New York. The state of knowledge is quite different from that for larger metropolitan regions, which have been the main focus of research to date.

The primary aim of this project is to create a data base about disparities involving major health-impacting conditions and diseases for the cities of upstate New York with significant minority populations. The creation of the data base will exploit the publicly available bodies of data about health incidents and conditions among New Yorkers—in particular, the SPARCS data on hospitalizations (i.e., Statewide Planning and Research Cooperative System), created by the New York State Department of Health and an annual survey of health-risk factors, the BRFSS (the Behavioral Risk Factor Surveillance System), which is mandated in every state by the Centers for Disease Control. These data will be compiled and merged with the basic demographic data for upstate communities from Census 2000 in order to create rates of occurrence of health conditions and diseases, such as diabetes and hypertension, for majority and minority populations.

A goal of the project is to disseminate subsequently this newly created information to major “publics,” who can use it. Such publics would include community groups as well as health scientists and organizations concerned with health-related policies. The web is envisioned as the main mechanism of dissemination, although occasional reports on specific subjects such as diabetes will also be issued. The CEMHD website as well as those of related university centers, such as the Lewis Mumford Center, will provide the platforms for the web-based dissemination.

In the long run, the intent of the analysis is to be able to relate health disparities in the different upstate New York cities to the demographic and sociological realities of minority communities, as well as to local health infrastructures and the access to them and use made of them by majority and minority groups. Such an analysis will ultimately reveal whether the extent, nature and sources of health disparities differ between the nation’s major metropolitan areas and smaller places.

b. Studies and results
To date, our efforts have been focused on establishing the framework for developing the data base and especially on making the methodological decisions necessary to meld effectively the hospitalization data (the SPARCS data) with census data. We are very near the end of this preliminary process and poised to generate a range of data for health-disparity indicators.

We have accomplished the following:

  1. Set up a working group that, in addition to Alba and Strogatz, includes Michael Zdeb of the School of Public Health, an expert on Department of Health data, Dr. Ruby Wang, a statistician/data analyst at the Center for Social and Demographic Analysis and the School of Public Health, and Hyoung-jin Shin, a graduate student attached to the project.
  2. Chosen to focus on the measurement of health disparities for the Prevention Quality Indicators (PQI) developed by the Agency for Healthcare Research and Quality (AHRQ), because minority disparities for any of these indicators are likely to signal inequalities in the access to and use of medical services. The PQI primarily refer to hospitalizations for conditions, such as diabetes complications and dehydration, that can usually be treated successfully by appropriate interventions before the need for hospitalization arises.
  3. Resolved to our satisfaction the data problems that complicate the merger of administrative with census data and thus create difficulties for the precise measurement of health disparity rates. These include: a substantial amount of missing race/ethnicity information in the hospitalization data, and problems in matching the geographies of the different data sets.

We have now generated preliminary estimates of health disparities for the PQI indicators for 20 incorporated cities of upstate New York with populations of 20 thousand or more (we have also included Amsterdam, though it does not meet this criterion). We are in the process of checking these results before making them available.

c. Significance
It is apparent from our preliminary results that health disparities in upstate New York cities rival those in New York City metropolitan region. Recently, The New York Times gave a great deal of attention to the epidemic of diabetes in the city’s minority communities. A brief glance at the data for upstate New York demonstrates very large differentials, especially between blacks and whites, in hospitalizations for diabetes complications. According to our still provisional estimates, the differential is four-fold for short-term complications in the city of Albany and more than three-fold for long-term complications; while in Buffalo, the differentials are three-fold and more than two-fold, respectively. Similar disparities are found in other upstate cities with significant black populations.

d. Plans
Our plans for the near future are as follows:

1. Once we have thoroughly verified our preliminary results, we will begin to make them available through the CEMHD website.

2. We will issue reports on major findings, beginning with a report on diabetes complications.

e. Publications
See above.

f. Project-generated Resources
See above.

Monitoring the Trends in Health Disparity in Upstate New York

Kajal Lahiri and Zulkainain Pulungan
Progress Report:
link to full paper

The aims of the project have not changed: This study estimates inequality in health for different ethnic/racial groups in the New York adult population and decomposes the inequality into its components. In order to accomplish these aims, we estimate a continuous measure of health for each person that is used to calculate health inequalities between and within various ethnic/racial groups as well as across 10 regions of New York State . We decompose the inequality into its components and estimate two types of health inequality: total health inequality and health inequality related to socioeconomic status.

The data come from Behavioral Risk Factor Surveillance System (BRFSS), an annual survey of the adult New York State population. To optimize the number of observations and number of variables to be analyzed, we use data from 1999 through 2004. Missing values in the dataset are imputed using latest multiple-imputation techniques. The total sample size is 22,083 individuals.

Preliminary analysis of self-assessed health status (SAH) variable shows that 23 percent of respondents consider themselves in excellent health, 33 percent in very good health, 30 percent in good health, 11 percent in fair health, and only 3 percent in poor health. The distribution of the SAH varies across ethnic/racial groups as well as across regions. We estimate an ordered Probit regression model of SAH with various reported diseases and individual characteristics as covariates. The estimated coefficients for Blacks, Hispanics, and other races are negative and statistically significant relative to Whites. This means that quality of health of the minority racial and ethnic populations in New York is lower than that of Whites.

After re-scaling the predicted values of the underlying variable of SAH, we calculate average quality of health for each ethnic/racial group as well as for each region. The average quality of health of American Indians (AIAN) is the lowest, followed by Hispanics. Across regions, the North Country has the lowest average quality of health, followed by New York City . In addition, since the age distribution varies considerably among ethnic/racial groups, we adjust the quality of health for age and find that Hispanics have the lowest age-adjusted quality of health, followed by AIAN.

American Indians have the highest total health inequality, followed by Hispanics, while the highest health inequality related to income is found within Hispanic group, followed by AIAN group. Across regions, the highest total health inequality and health inequality related to income are found in the North Country , Mohawk, Sothern Tier, and New York City regions. We found that groups with lower quality of health tend to have higher total health inequality as well as higher health inequality related to income.

From decomposition analysis, we found that the contribution of each component to health inequality varies considerably across groups. For example, the contribution of education to total health inequality in the Hispanic groups is the highest, while the contribution of age is the lowest. This suggests that eliminating the health disparity across education would have the highest impact on the Hispanic group in reducing total health inequality.

Most health information is available at an ordinal or binary level. For example, one of the most commonly used indicators of overall health in a general population survey is self-assessed health (SAH) status, which is defined by the ordinal response to a survey question on self-perceived health. To analyze health inequality among population groups, we need to measure health with a continuous variable. This study tries to transforms the categorical measures of health into continuous measures that can be used to measure the burden of diseases, inequalities, and disparities of health conditions among groups in the population. In addition, we decompose the inequalities into their components that are meaningful for policy purposes.

Our plan for the next phase is to extend the analysis into the national level. Such analysis entails measuring inequality between and within groups of U.S. population: ethnic/racial groups as well as across 51 states. At the same time, we are continuing to find more robust estimation methods for estimating of the parameters and the continuous measure of health based on categorical measure of health.

None to date.

None to date.

Overcoming Cultural Barriers to Hispanic Use of Health Education Services

Blanca Ramos and Anne Fortune

Progress Report:

a. Background
Latino older adults tend to underutilize health education services. Data collected primarily with Mexican Americans in large urban settings indicate this may be due to inaccessibility, discrimination, limited availability of services, and language and cultural incongruities. Less is known about service utilization among older Latinos in smaller cities, particularly those of Puerto Rican and other Latino heritages. This study examines the structural and cultural factors that may contribute to service use by Hispanic older adults and their families in a small rural community. The results will inform the development of a culturally relevant health education program that will address the identified structural barriers to service utilization. The study is based on ecological theory, which emphasizes person-environment interactions recognizing contextual factors such as culture and ethnicity as they influence service use and intervention.

b. Proposed Activities
The goal of this study will be achieved through the following activities carried out during a period of three years:

Collect qualitative data to identify structural and cultural factors that may contribute to service utilization

Review of the literature on empirically tested health education programs

Assemble a task force comprised of UAlbany faculty and health service agencies and community representatives

Develop a culturally relevant health education program targeting a specific health disparity area

c. Progress
To date, the bulk of the time has been spent organizing and planning the proposed activities and in data collection.

The collection of data on structural and cultural factors contributing to service use is almost completed. Five focus groups with Hispanic older adults and their families have been conducted.

The literature review on health education programs is underway. A number of sources have been identified and are being compiled.

The assembling of a task force has been initiated. This will include the two primary investigators, two UAlbany faculty from the School of Public Health, and representatives from the Internships in Aging Project, Montgomery County Office for Aging, St Mary’s Hospital, and Centro Civico in Amsterdam.

d. Next Steps
Complete data collection and begin conducting analyses.

Complete gathering and begin reviewing the literature on health education programs.

Begin holding meetings of the task force

Begin developing the health education program.

Social Networks, Culture, and Chronic Illness Management

Mary P. Gallant and Glenna Spitze
Progress Report:

a. Specific Aims
African American and Latino older adults shoulder a disproportionate burden from chronic illnesses, such as diabetes, arthritis, and heart disease. Many of these illnesses have a significant self-management component, including regimens of medication-taking, physical activity, dietary and weight management, and specific disease-related behaviors. Substantial evidence demonstrates that successful self-management of chronic illness is related to better overall physical and psychological health outcomes. The social environment is an important influence on self-management behavior, yet little is known about the social context of chronic illness management, particularly among older minority populations.

The original objective of this pilot study was to examine the influence of family members and friends on chronic illness management behaviors among Latino and African American adults, with a special emphasis on exploring cultural factors. Our long-term goal is to design and evaluate a self-management intervention. The pilot study had the following specific aims:

Develop a collaborative team of University at Albany researchers (Gallant, Spitze) and members of selected African American and Latino communities, including representatives of the health care and community services communities;

Conduct a literature review of chronic illness management among Latino and African American populations, and of family and social networks among older Latinos and African Americans.

Define the study population according to disease status (e.g. diabetes, heart disease, etc.) based on the literature review, input from community collaborators and perhaps informal interviews with community members;

Develop a plan for pilot data collection (quantitative and qualitative) that would focus on beliefs and attitudes about chronic illness management, chronic illness management behaviors, social network influences on illness management;

Conduct pilot studies; and

Develop and pilot test an intervention.

Based on our work during this past year (see Section B), the aims of this project have been broadened and modified. We now plan to examine a variety of social influences on self-management behavior among Latino and African American older adults, instead of restricting our focus to family and friends. We are also tentatively planning to spend the third year writing and submitting a proposal for grant funding which will allow us to conduct the final aim (above) on a larger scale with adequate resources.

b. Studies and Results
We have been engaged in three sets of activities to date: conducting a comprehensive literature review of chronic illness self-management and the social networks of Latino and African-American older adults; planning for focus groups; and making contacts with representatives of the Amsterdam and downtown Albany communities.

We are in the process of preparing a comprehensive literature review for publication that will synthesize literature on chronic illness self-management with literature on family and social networks among Latino and African-American older adults. This review is unique in that it brings together two bodies of literature that have not previously been looked at together, and the resulting publication will contribute to knowledge about chronic illness management in minority populations. During the summer of 2005, we employed two part-time graduate research assistants to search the empirical literature on (1) chronic illness self-management among Latino and African-American older adults and (2) social networks among Latino and African-American older adults and categorize the results. Their work resulted in a detailed outline with supporting publications of the relevant work in these two areas that has been published in the last 10 years. We are now in the process (albeit without the help of graduate assistants during the academic year) of reviewing this literature and preparing a manuscript for publication.

We have also been further refining our data collection methodology for this pilot study. In summer 2006, we plan to conduct 6 focus group interviews with adults 55 years old and older with one or more common chronic illness (e.g. diabetes, arthritis, heart disease). Four of these focus groups will be conducted with members of the Latino community in Amsterdam and 2 will be conducted with members of the African-American community in downtown Albany. We plan to conduct more focus groups with Latino individuals because this pilot study builds on an earlier study in which we conducted focus groups with African-American men and women on the same topic, and because the literature review indicates that there is much less known about chronic illness management in Latino populations.

Finally, we have made contacts with representatives of both the Albany and Amsterdam communities to identify individuals who will advise us with respect to our pilot data collection plans, recruitment materials, and who will work with us in recruiting focus group participants. In February 2006 we plan to meet with these community representatives in Amsterdam to learn more about the community and to further refine our data collection plans based on their input and advice.

c. Significance
African American and Latino older adults are disproportionately burdened with chronic illnesses, yet they have lower levels of the socioeconomic resources that might facilitate self-management of those illnesses. Our exploratory research will contribute to an understanding not only of the barriers they perceive in maintaining their health, but also factors in their social network and living environment that may facilitate their self-management. This knowledge can be used to conduct more comprehensive quantitative research in this area as well as to design a self-management intervention for these older persons, potentially involving not only the older persons themselves but also their network members and the health care professionals who serve them.

d. Plans
Our plans for the second year include:

Complete a draft of the synthetic literature review of chronic illness management among Latino and African American populations, and of family and social networks among older Latinos and African Americans;

Continue to make contacts in both the African American and Latino communities and develop a collaborative team that includes health care and community service representatives in both communities;

Plan and carry out (during Summer 2006) six focus group interviews with older men and women with one or more chronic illnesses, two in the African-American community in Albany and four in the Latino community in Amsterdam;

Make plans for transcribing and analyzing focus group data;

Plan for writing and submitting a grant proposal based on this pilot study and our previous work, which will provide adequate funding for our planned intervention.

e. Publications
There are no publications to date. We are in the process of drafting a synthetic literature review that will be submitted for publication.

f. Project-Generated Resources

Physical activity: Environmental, social, and familial influences

Kirsten Davison, Catherine Lawson, & Jeffrey Olson
Progress Report:

The specific aims of the study are unchanged and are as follows:

(1) To examine familial characteristics that impact on children’s physical activity including parent characteristics, parenting behaviors and the family context;

(2) To examine community and environmental factors which constrain children’s physical activity behaviors either directly or indirectly through parents’ ability to support their child’s activity; and

(3) To examine the moderating effect of race/ethnicity on the associations assessed under specific aims (1) and (2).

Pilot study: Constraints on parents’ efforts to encourage their children to be physically active

During the 2004-2005 funding cycle, pilot data were collected to examine (a) parents’ perceived constraints to their ability to promote active lifestyles among their children and (b) associations between such constraints and parents’ support of their children’s physical activity. A 72-item questionnaire was developed, mailed to 82 parents, and returned with responses from 76 parents. Exploratory factor analysis, guided by Leisure Constraints Theory, was used during the past year to identify the factors present in the scale. Results showed that most parents reported constraints, but the types of constraints differed based on their circumstances. For example, Black parents reported that accessibility (i.e., cost) and neighborhood safety constrained their ability to encourage their children to be physical activity, while White parents were more likely to report constraints resulting from a lack of similar-aged children in the area and their children not being able to walk/cycle to a play area. Constraints were more often associated with lower levels of support among African American parents than White parents. This finding may reflect the possibility that African American parents have fewer resources to negotiate such constraints.

Review paper on associations between the physical environment and children’s physical activity

In a separate but related endeavor, Kirsten Davison and Catherine Lawson compiled a review of the literature on associations between the physical environment and children’s physical activity. This paper was submitted March 2006 to the International Journal of Behavioral Nutrition and Physical Activity. This review is a preliminary step in addressing specific aim 2, which focuses on community and environmental factors influencing children’s physical activity.

Resource Gap for Representing Environmental Factors

The literature review revealed a number of unresolved issues with respect to the appropriate representation of the various environmental factors. A technical review was conducted on literatures that included the use of geographic information systems (GIS) to gather and test environmental variables. The review revealed the following issues:

  • A lack of meta data (data about the data) with respect to the GIS data used;
  • The use of inappropriate geographic scale (i.e., average characteristics of the environment extracted from large zones rather than within the walking areas relevant to children); and
  • A lack of information on the steps or processes (documentation) used reducing the ability of others to replicate or standardize the development of data.

A series of exploratory studies were conducted by the project graduate research assistant to determine and evaluate approaches for the appropriate use of GIS. These approaches included the use of remote sensing and aerial photo techniques to efficiently develop necessary line work. Another strategy used the emerging technology of ARCGIS server to prepare a set of web-based GIS datasets. Additional work will include the documentation of the most cost/time-effective methodologies to generate a variety of constructs (e.g. distance to nearest park, traffic speed on a child’s street) for future research on the environment and activity of children.

R21 grant submission: The Context of African American Parents’ Support for Children’s Physical Activity

Kirsten Davison, in conjunction with Glenn Deane and Catherine Lawson, submitted an R21 grant application entitled The Context of African American Parents’ Support for Children’s Physical Activity to the NIH (NICHD) on February 1 st 2006 . In response to PA announcement PA-04-121, the proposed exploratory study will examine links between parental support of physical activity and physical activity levels of elementary-school-aged African American children, while taking into consideration the context, or the ecology, in which parenting occurs.

As outlined in Healthy People 2010 and the Youth Risk Surveillance Survey, 35% of youth fail to meet the minimum physical activity guidelines and another 14% are completely inactive. Consequently, children and adolescents are at heightened risk of the negative health outcomes associated with low levels of physical activity including obesity, type II diabetes mellitus, hypertension, osteoporosis and depression. Low levels of physical activity are more pronounced among minority youth. As a result, African American youth disproportionately experience the negative health outcomes associated with insufficient levels of physical activity. Two accumulating, yet independent bodies of research, highlight the importance of families and communities in shaping physical activity behaviors among children and adolescents. Research, however, has not examined the interplay of the family and the community as contexts in which youth activity patterns evolve or how these associations may vary for children of different ethnic/racial backgrounds. The planned research will help guide the development of realistic activity promotion programs for youth that families can incorporate into their lifestyles and which can be tailored to meet the unique circumstances of White and African American families.

During the coming year, the investigators will largely prepare for the R21 project in anticipation of possible funding. Activities will include: (1) Conducting a small qualitative study using focus groups to explore the validity of the Activity Support Scale (a key measure in the R21 proposal) with African American parents; (2) Develop a code book outlining specific methods to quantify the physical environment (e.g., distance to the nearest playground, presence of sidewalks) using geographic information systems (GIS) and where necessary, pilot test the methods using on aerial photography (this assessment protocol is a central component of the grant submitted); and (3) revising and resubmitting the grant application if it is not funded.

None to date.


Links to further Research Activities and Training Summaries