a. Variations in health care (by place, race, class, gender)
Idea: Inequalities in how people are treated are associated with place, race, class, gender, even after conditioning on other relevant variables.
b. Heterogeneity within populations and subgroups
(continued in week 8)
Idea: How people respond to treatment may vary from one subgroup to another--When is this a matter of chance or of undetected additional variables? How do we delineate the boundaries between subgroups?
Initial notes on the Cases
From PT:
Inequalities Cases, a: Immunization levels (again), Access to cardiac procedures, Evaluation of angina, Variation in mortality of the elderly across communities
Readings: Egede 2003, Alter 1999, Roger 2000, Davey-Smith 2000, Bassuk 2002
(Notes on how to prepare these readings and prepare for class to be supplied by Sue.)
Heterogeneity Cases, b: Forms of breast cancer, Low dose aspirin and prevention of cardiovascular events, Statistical concerns
Readings: Regan 2005, Eikelboom 2003, Gum 2003, Nelson 2005, Lagakos 2006
Lagakos provides a statistician's cautions about the significance of results derived from subgroups of the whole population, especially if the subgroups were only defined after exploring the data.
The opposite caution is that treating everyone as if they were from the same population (for good statistical reasons) distracts our attention from the clues that might lead us to seeing that the population is not one uniform whole, but is a mixture of types. This can have significant health care implications -- see case studies about different kinds fo breast cancer and aspirin resistance.
As part of preparation, collect your thoughts so we can go around and have each person say how you would address both cautions in an ideal research program in your own area of interest.
Susan's Thoughts
This week’s lecture focuses on the inequalities of accessing health care depending on different variables such as gender, race, class, and location. Several articles focus on these inequalities, which makes one ask the question:
Can these scientific studies be accurate, concise, and helpful to the population as a whole if variables are omitted?
This week, we were to read and attempt to examine different variables which may or may not have an impact on adequate health care. For example, Bassuk (2000) examined mortality rates among the elderly in respect to their socioeconomic status, believing that other variables such as community attributes may influence health outcomes. Roger (2000) tested the hypothesis that there is a negative association of delivered care and outcomes of female sex diagnosed with unstable angina.
What comes to mind is that there is a lot of research stating that there is a problem with health care disparities whether its gender, age, race, culture, etc.
However, should researchers be now focusing on intervention strategies? What are the controllable causes of disparities? Are they political? Institutional? Regional? Nationwide?
Although these articles are not so much about epidemiology methods, but rather showing the existence of a problem, look at the articles and their descriptions of inequality.
How do the researchers measure this?
Examine the methods and determine if either the problem researched or the methods utilized could be applied to your own area of research. Come to class prepared to either find a reference in your area of research relating to this topic or design (briefly) an equivalent study. Due to the large amount of reading required for this week, I have only assigned the readings for my half of the class and paired us up into the following groups:
Louisa and Kaori; read Davey Smith, 2000;
Connie and Judy; read Alter, 1999;
Elizabeth and Jerrilyn; read Bassuk, 2000
Jill and Susan will read Roger, 2000.
Substantive statement
There continues to be problems in delivering quality health care and health services equally to all persons in today’s society including underuse, overuse and misuse of some services. Variations from regional vs. small areas reflect a pattern that has not kept pace with health care practices. Underuse of health services indicates that millions of people may not receive adequate health care. Ethnic and racial minority populations are often targeted for lack of quality health care. During our study of variations in health care, we examined articles reflecting these inequalities. Egede (2003), examined one type of health inequality by revealing health disparities with vaccination rates (influenza and pneumococcal). The aim of the study was to view differences in access to care and health care coverage which would then explain racial disparities in receiving these vaccinations. Data was collected on individuals > 18 years old with diabetes (excluding women with gestational diabetes). Stratification with a complex sampling design was utilized to give a representative sampling of US adults. Despite adjusting for access to health care, health care coverage, and SES, the investigators still noted racial and/or ethnic disparities in relation to receiving the vaccinations. The study concluded that further research is needed to collect data on social and cultural factors influencing health outcomes for this population.
Alter et.al.( 1999), devised a study which assessed effects of neighborhood income on access to invasive cardiac procedures of patients with an acute myocardial infarction (MI) and to examine the associations between socioeconomic status and mortality one year after acutemyocardial infarction. The researchers concluded that although their study was not to focus on clinical and social factors of adverse outcomes, it was still noted that poorer medium-term outcomes for patients in the lower-income groups were consistent with research that still revealed that populations with lower incomes lived shorter, less healthy lives. This may be due to their higher levels of risk factors which are further complicated by poorer compliance with medical therapy. The most interesting part of this study was that despite Universal Care (study performed in Ontario, Canada), people from lower-economic areas had reduced access to these cardiac procedures as compared to wealthier neighbors.
Bassuk (2002) believes that evidence relating to SES to mortality among the elderly population is less consistent. Four communities were examined by looking at the relation between SES and mortality. This study attempted to evaluate which SES indicator might be the strongest predictor of mortality: associations by gender, age, race, marital status, or urban/rural residence or SES differences in morbidity, behaviors, and/or other social conditions. The results indicated that despite higher SES, (whether measured by education, income, or occupational prestige) was associated with a decrease in mortality over 9 years.
Davey-Smith et.al (2000) addressed the difficulties of analyzing ethnic differentials in terms of cultural or social class differences. They believed that controlling for socio-economic differences is necessary to avoid conclusions which may be misleading in regards to causes of health variations between and within ethnic groups. The study also stated it was important that measures of socio-economic position were not misleading. This study was part of a larger study that combined qualitative and quantitative techniques by hoping to improve the use of socio-economic indicators in multi-ethnic population health studies. However, even though the qualitative findings revealed close observations of a variety of ways which ethnicity could change the meaning of indicators of socio-economic position, the amount of subjects studied were smaller than could be accessed by quantitative survey techniques. The conclusion was to utilize a combination of qualitative fieldwork and quantitative surveys in hopes of studying health and illness across ethnic groups.
Roger et.al (2000) tested the theory of a sexual bias in the delivery of cardiac care and outcomes. The results indicated that there was indeed an association of female sex and decrease use of cardiac procedures that wasn’t explained by measured sex differences in baseline characteristics. Of interest, men had worse outcomes than women despite the women having had less intervention. The researchers recommended further studies to evaluate if those adverse outcomes could be altered with interventions.
All these studies help health care practitioners to begin to address disparities in quality of care by improving clinician’s abilities to apply the results of previous research to ethnic/racial clients. The Agency for Healthcare Research and Quality (AHRQ) supports tools such as the National Guideline Clearinghouse (
http://www.guideline.gov) to give clinicians and other health decision makers better access to evidence-based information about diagnosis and treatment. Increasing evidence that cultural expectations, assumptions, and language can often affect the quality of care and outcome by the way clients interact with clinicians; the interpersonal processes. Research will need to be ongoing in an attempt to analyze health care disparities and strategies to overcome these differences in health care with the overall goal to improve access to quality healthcare for all populations (SA)
References per syllabus.
Annotated additions by students
Shaw, L., Miller, D., Romeis, J., et al. (1994). Gender differences in the noninvasive evaluation and management of patients with suspected coronary artery disease.
Annals of Internal Medicine 120 (7); 559-566.
This article’s objective was to determine if gender-based differences exist in the post-test management and clinical outcome of patients with clinically suspected coronary artery disease who have stress electrocardiographic or myocardial perfusion imaging evaluation. This study used a retrospective cohort study design and found that women with suspected coronary artery disease have fewer additional diagnostic tests than men. The retrospective design of this study has been called into question. (JN)
Transportation
When I was in Vancouver, Canada, about ten years ago, there was only one major line (like the T, red line here), which was operated by no one and no food was allowed in the train. The train was very convenient, but the line was not long enough for everyone to use. In suburb areas, we had to wait for a long time for a bus. Unlike here, people who buy a car need to buy/pay their auto insurance (only one kind, ICBC) in order to have their number plate and get registered at the same place. Also, people were encouraged to have more than two people in one car. Probably people were concerned about pollution and traffic. In the major high way, the very left lane was only for cars having two or more people inside.
So, even they have universal health insurance, people still need to pay for a small amount of fees for it. And even a student I knew around me didn’t pay for it.
Also, people were complaining about a low rate of employment/high rate of unemployment, and nurses were on strike.
So despite having a national health insurance policy, though I think it is a good idea, there is much going on in any country at one time of period, and it could be temporary and could be a long term issue.
Anyway, these were just my own experiences and thoughts, and I thought it might be good to share with you.
(km)
Rogers, Naomi (2007). Race and the Politics of Polio. American Journal of Public Health. Vol. 97, No. 5, p. 784.
This article is an historical account of the opening of the first polio treatment center for blacks in the U.S., (through the Tuskegee Institute in 1941) and how “medical segregation” prevented Blacks from getting necessary polio treatment from about he 1920’s through the 1940’s. While the article focuses more on the race and politics of polio treatment, it also highlights how epidemiological data was manipulated to maintain segregation policies. Early studies on outbreaks showed low prevalence among Blacks. This information was transformed into the
susceptibility argument, in which Blacks were considered less susceptible to polio than Whites. This was the eugenics era, so that the notion that it was an individual’s constitution (rather than an acquired immunity) that explained susceptible to polio was accepted and therefore helped sustain the policy of race segregated health facilities. However, leaders in the Black medical profession rightly argued that there were few cases of polio reported among Blacks (using the arguments similar to what we would see today as explaining the disparities in health outcomes) because of limited access to doctors and hospitals and inadequately trained Black health professionals. Actually, there WERE some epi studies of outbreaks where Blacks were infected, but these studies of these were being ignored.
The movement for polio treatment for Blacks coalesced against the Whites-only polio treatment center called Warm Springs started by F.D. Roosevelt in the 1920s. The Black medical establishment and others (including the NAACP) were outraged and this limited their support for Roosevelt during the 1936 election campaign. Warms Springs admission policy became a “potent symbol of medical racism”. An epidemiological analysis of outbreaks was undertaken by a Chicago public health official which selectively highlighted the outbreaks in which more whites were infected; there were other studies where the reverse was true but were cited as being unreliable because Negro reporting was considered not reliable. The study did mention that a lower prevalence among Blacks may be due to nonreporting rather than racial immunity, but with so much trepidation it did not convince. It wasn’t until the Tuskegee Institute opened a small unit for disabled Black children that Black researchers challenged previous epidemiological studies with new evidence showing Black patients unequal access to care and their physicians inadequate professional training. To sum, “epidemiological statistics that were believed to prove polio’s differential racial susceptibility were being used to justify medical segregation and to deny Blacks legitimacy as both patients and scientists.” (EB)
