IN SEPTEMBER 2010, THE NATIONAL INSTITUTES OF HEALTH formalized a commitment to ending health disparities by establishing a 27th institute, the National Institute on Minority Health and Health Disparities, with priorities that include “improving the participation of health disparity populations in clinical research.” Accomplishment of this important goal will require recognition of the disproportionate incarceration of individuals from racial/ ethnic minority populations and reassessment of the federal near prohibition on the participation of current and former prisoners in clinical research. Currently, more than 2.3 million individuals are incarcerated in the United States. It is estimated that on any given day, 1 in 9 US black men aged 20 to 34 years is incarcerated, and 1 in 3 black men are expected to be imprisoned at some point in their lives if rates of incarceration stay the same. Ninety-five percent of these individuals are ultimately released back into society but most continue to cycle through the legal system throughout their lives. These numbers have direct bearing on the ability to design and implement studies of health disparity populations and thereby undermine the ability to understand and mitigate the racial/ ethnic health disparities such studies seek to diminish. For example, cardiovascular disease mortality is 20% higher in black men than in white men. Similarly, hypertension, a risk factor for cardiovascular disease, tends to be more common, develops at an earlier age, and is more severe for black than white individuals. Much of what is known about these disparities has been derived from federally funded cross-sectional studies such as the National Health and Nutrition Examination Studies, which sample noninstitutionalized, community-dwelling adults. However, if 1 in 9 young black men is currently incarcerated, how accurate can the surveillance of racial disparities be—especially since incarceration may be associated with an increased risk of developing cardiovascular disease? For comparison, if 1 in 9 of these black men were a current smoker (and given that such rates of smoking were exceptionally high), but these men could not be included in national surveys, estimates of smoking prevalence might be inaccurate. By not including the 2.3 million currently incarcerated individuals in national surveys, the surveillance systems might provide flawed estimates of racial/ethnic disparities in cardiovascular risk factors and miss opportunities to understand how incarceration and the correctional health care system affect cardiovascular risk factors and disease. Mass incarceration affects not only disease surveillance but also studies of risk factors for the development of cardiovascular disease or tests of interventions to reduce disease in minority populations. The National Heart, Lung, and Blood Institute spends hundreds of millions of dollars on prospective cohort studies and randomized controlled trials such as the Atherosclerosis Risk in Communities study, the Jackson Heart Study, the Multiethnic Study of Atherosclerosis, and the Coronary Artery Risk Development in Young Adults (CARDIA) study, all of which are focused on improving the health of minority populations. Of critical importance to the validity of these studies is participant follow-up over time. However, in all of the aforementioned studies, observing young black men over time has proven to be challenging. In CARDIA, for example, only 56% of black men remained in the study at the year 20 examination in 2005, compared with 75% of white men, 68% of black women, and 77% of white women. It is likely that at least some of the black men who dropped out of CARDIA did so because of incarceration. What are the potential implications if a disproportionate number of black men are being excluded from crosssectional and prospective cohort studies due to incarceration? Only 2 potential implications, power and precision, are considered in this Commentary. If a significant proportion of black men are lost to follow-up or excluded, statistical power for that group may be reduced, impeding efforts to identify significant associations. Furthermore, if data are not missing at random, which they are not if imprisonment disproportionately affects black men, the estimated effects may be biased for black men (assuming the dependent and explanatory variables are associated with incarceration). Thus, the loss of black men from medical research due to incarceration may produce biased, underpowered estimates in studies of health disparities.
Read full abstract