A Sample Set of Sketches
Personal and Professional Development Plan: Sketch Week 1 – September 16th
a) When I entertain the possibility of becoming literate in the methods used to reconstruct pathways of development of behaviors and diseases, the questions/ideas/experiences that come to mind include…
My background in public health, specifically my work over the past decade in HIV/AIDS (prevention, education, counseling, testing, and referral), Reproductive Health (family planning, STIs and abortion – advocacy, prevention, education, counseling, and medical services), Domestic Violence (intervention and research), Substance Abuse (advocacy, prevention, and research). In looking over this list, I see how much I have focused on health issues that rest heavily on individual behavioral choices, and controversial issues at that. I’ve tended to focus on the host and the environment, rather than on the agent. In the case of each, there are complex forces at play, and these health issues are very social in nature. This last sentence now makes me question whether or not that is the case for any or all health issues…
I recall the State-of the Science conference I attended on preventing youth violence, where I first had a chance to hear Robert Sampson speak about social capital and neighborhood effects. The degree to which residing in a poor, racially segregated neighborhoods, isolated from public transportation and good jobs contributed to violence, substance abuse, and other health issues was alarming to me then and now. I also listened to other researchers describe a nursing intervention program focused on pre and post-natal visits as an opportunity to improve mothers’ health and education and capacity in the present – so as to prevent health and behavioral problems among their children, in adolescence and adulthood in the future. This conference planted the seed for my pursuing a doctoral degree.
Although my current job within a program situated at a community non-profit is based on theories of risk & protective factors and healthy communities, I find we have not done much to address social and economic determinants of health. This is partly due to constraints imposed by funders, but it also speaks to the limitations of employing a programmatic approach to address structural issues such as poverty, academic failure, unemployment, and racism.
To some extent I feel like the pendulum has swung from one extreme to another, in terms of my perspectives on public health. Infectious agents and chronic diseases seem almost incidental in the face of these larger determinants. I liked the video presentation assigned the first week, which I watched in its entirety, as it brought me back to the realization that even if poverty were greatly diminished, practitioners would still be faced with cancer and heart disease. It’s helpful to keep in mind that determinants foster certain conditions, predispose one to, and/or exacerbate, a health problem – but act in concert with genetics and agents.
I hope that this course in epidemiological thinking will help me to broaden my thinking about determinants and health outcomes, enabling me to remain open to critical perspectives as I narrow my dissertation focus.
b) Particular epidemiological/health research or policy question in my own area of interest – to be used each week to sketch out my ideas about how the concepts, methods and problems of the week might be applied to it.
Disparities in health status persist in the U.S., leading to negative differences in physical and mental health outcomes for one group, as compared to a reference group. While disparities may be measured along lines of age, gender, ability and other characteristics, the most notable disparities in the U.S. have been found along racial and ethnic group lines. Indicators of health status include self-reported health status, disability, and weight (and height where available). Increasingly, health researchers are looking to overweight and obese as a proxy for a number of health outcomes, particularly diabetes, cardiovascular disease and cancer.
Employment-based non-mandated benefits (EBNMB) may be considered an economic determinant of health status and outcomes, in conjunction with wage income and social insurance programs. In the U.S., EBNMB refers to benefits such as health insurance, paid leave, and retirement pensions that employers may or may not choose to provide to employees. Apart from some regulations regarding provision of certain benefits, according to size of firm and worker status, EBNMB are not mandated by government legislation or ordinance.
The U.S. is unique among industrialized countries in providing very limited universal social insurance benefits to citizens. Virtually all benefits are tied to employment, including Social Security, as workers contribute funds that provide benefits to retired or disabled workers. There are few mandated employment-based benefits, limited to Social Security, unpaid Family and Medical Leave, Unemployment Insurance and Workers’ Compensation. Concerning the latter two, there is great of variation in eligibility and benefits among U.S. states. These social insurance programs are intended to pool and manage risk among U.S. workers and their families. By tying social insurance to employment in the form of EBNMB, U.S. society has further heightened the vulnerability of workers by relegating social risk to the private sphere, both markets and individuals.
1) Is there a disparate distribution of employment-based non-mandated benefits among disparate groups of workers in the U.S.? Variation by work sector? By job status? By firm size?
2) What is the distribution of physical and mental health status and outcomes among U.S. workers?
3) How does the distribution of employment-based non-mandated benefits contribute to distribution of health status and outcomes among workers in the U.S.?
4) Is there a disparate distribution of employment-based non-mandated benefits within and among disparate groups of women workers in the service industry in the U.S?
5) What is the distribution of physical and mental health status and outcomes within and among disparate groups of women workers in the service industry in the U.S.?and their children?
6) How does the distribution of employment-based non-mandated benefits contribute to distribution of health status within and among disparate groups of women workers (and their children) in the service industry in the U.S.?
7) How does variation in model of provision relate to variance in health status among countries? Distribution of health status among groups within countries?
8) How does variation in model of provision relate to variance in cost of social risk benefits? and in cost of health burden? To individuals, employers, and government?
9) What are the opportunities for converting private risk to shared social risk programs in the U.S.?
Sketch Week 4 – September 30th
In preparing my dissertation proposal, I find the issue of categories to be particularly salient. I am considering the following categories for inclusion.
Categories of workers: personal characteristics (age, gender, race/ethnicity); household size and composition; human capital characteristics (education); industry/sector and job characteristics (occupation, white/blue collar/service, public vs. private sector, union/non-union, tenure); and job status (full-time, part-time, temporary)
Categories of employment based benefits (non-mandatory): health insurance (eligibility, coverage such as physical, mental, dental, eye care, and out of pocket expenses such as deductibles, co-pays, prescriptions); pensions (eligibility, coverage); paid leave (maternity/paternity leave, sick time, vacation time, holidays, civic/legal leave)
And categories of health outcomes and status: self-reported health status, disability, weight and height, diagnosis and limitations of cardiovascular disease, diabetes, respiratory disease, asthma, arthritis, cancer, and nine psychiatric disorders
A challenge in working with all of these categories is that they are closely interconnected and fluid. For example, a white-collar worker in a unionized, public sector position today may well find themselves in a white-collar non-unionized private sector position next year. Moreover, these categories may be defined broadly or narrowly. A case in point is health insurance – should coverage be measured along binary terms, does one have insurance? or not? Or is it more illustrative to look at issues of employer match, deductible, co-pays, access to wide range of providers, etc. The latter approach is more comprehensive and yet harder to model in quantitative terms. In the case of health outcomes and status, the categories are not discreet. An individual worker may report very good health status, alongside a height and weight that indicates obesity, and diseases such as asthma and diabetes may co-occur.
Sketch Week 5 – October 7th
Associations, Predictions, Causes, and Interventions
It will be particularly challenging to establish an association between the presence or absence of a combination of employment based benefits, given the wide range of other health determinants. These include congenital health problems, environmental conditions (toxins, substandard housing), inequalities in education, inequalities in wage income, racism, immigration, neighborhood effects, cohort effects, stress, nutrition, fitness, tobacco and alcohol use, family violence and so on. In no way am I trying to establish causation, just correlation. I’m approaching the research topic and questions re: link between inequality in employment based benefits and disparities in health outcomes and status as just one piece of a larger puzzle
Sketch Week 6 - October 14th
Confounders and Conditioning
The articles this week are closely related to my research interests, in that they attempt to establish correlation and causality among variation in SES and racial/ethnic differences to inequities or disparities in health status and outcomes.
Confounding factors that I am very clear that I will be controlling for include cohort effects, environmental conditions (such as substandard housing, toxins), and individual risk factors such as poor nutrition, physical inactivity, tobacco and alcohol use.
Some other considerations include race/ethnicity as a proxy for SES. The likelihood that having no or poor employment benefits serves as a proxy for also having a low-income job, which may have a stronger influence on health status and outcomes. I’ll have to consider the issue of absolute versus relative inequality in both wage and non-wage income.
Sketch Week 11 - November 18th
This approach holds tremendous promise for the quantitative models I need to develop for my dissertation, to address my research questions. As I intend to design a longitudinal study, looking at the cumulative effects of the presence and absence of employment-based benefits on health status and outcomes, it will be necessary to include a range of complex and interconnected independent variables. Specifically, multivariable structural models for change will be useful in my analysis of PSID data on workers and their households over time. I have also recently begun to consider approaching the issue of disparities in health status and outcomes as the dependent variable for two of the three essays, followed by a third essay, in which health status and outcomes become yet another independent variable pertaining to the dependent variable of economic mobility. Health is widely understood as both an outcome and a determinant. A longitudinal study of the type I am proposing could afford a unique opportunity to explore the stronger effect, in the context of socio-economic issue of non-wage income, i.e. employment-based benefits.
Sketch Week 12 – November 23rd
Nature versus nurture is particularly difficult to disentangle, as most individuals are raised in their family of origin. Moreover, the U.S. offers limited mobility to families, with the majority of individuals ending up in the same socio-economic strata in which they began life. So when a particular health advantage or disadvantage is observed across generations, it’s difficult to attribute its occurrence and severity to either genetic predisposition or environmental expression exclusively. Given that heritability is not restricted to families, but has also been observed among age cohorts in countries, as in the case of IQ levels, this further complicates the issue of heritability, while also offering hope. The possibility of effecting a widespread change among an age cohort, suggests that it may be possible to address other intellectual, behavioral and health issues, i.e. to shift the bell curve within a generation. This recalls Lynch’s video when he raises the controversial question as to whether we would prefer to rectify inequalities, improving the ‘balance’ of health outcomes, or whether it would be acceptable, or even preferable, to improve health status and outcomes across a population. In the latter scenario, all individuals would experience substantial gains, yet at the cost of preserving inequities.
Sketch Week 13 – December 2nd
Genetic Diagnosis, etc.
The connection between this week’s readings and my research interests are a bit tenuous, given the emphasis placed on the genetic basis for diseases. Expanding on my annotation for the Bowcock article, I do have concerns that increased focus on genetics will result in further disparities in health status and outcomes, as well as medical care. For example, if it is determined that a particular gene or combination of genes is determined to be the basis for diabetes, it’s conceivable that research and resources would be dedicated to finding a way to ‘turn off’ this gene, i.e. prevent it’s environmental expression. Additional research and resources would likely be devoted to prophylactic pharmaceuticals to also prevent onset of Type II diabetes. This would shift the focus away from addressing population level issues of nutrition and physical activity, as well as structural and policy issues related to our food industries and sedentary lifestyles. Those with the means to purchase the pharmaceuticals could avoid developing diabetes, while those with fewer resources would be left even further behind. Over-reliance on pharmaceuticals could also have unintended consequences.