Understanding and measuring the disparities present in health care are key first steps in the ongoing attempts to eliminate those disparities. This requires looking at the issue from a variety of angles, including racial and ethnic discrimination as well as less obvious potential disparities including geographic, educational and insurance-related issues.
Benjamin Cook, MPH, PhD, director of the Health Equity Research Lab at the Cambridge Health Alliance and Harvard Medical School, spoke about some important considerations for measuring health disparities in general, and specifically with regard to liver disease during yesterday’s Public Health/Health Care Delivery SIG program. He described a useful image in considering disparities involving a cliff, representing a serious medical issue: some people exist far back from the cliff with a fence representing primary prevention keeping them away from it and a strong safety net and good medical care awaiting them if they do fall off.
Others, meanwhile, live much closer to the cliff with only questionable prevention and care at the edge. It is important to remember that there is a host of social determinants pushing them closer to that edge, including outright discrimination along with food security, housing, insurance status and allostatic load associated with the piling up of these issues.
In 2002, a document put forth by the Institute of Medicine (IOM) attempted to define some of those disparities. Dr. Cook noted that there are policies such as one in Massachusetts that specifically reward or penalize hospitals based on their progress toward addressing disparities, and to do so it becomes increasingly important to define and track those in a quantitative fashion over time.
The question then becomes, what exactly should count as a disparity? Studies have consistently shown poorer quality of care delivered to minorities than to white patients, and that difference is due to several categories of factors: clinical need, appropriateness and patient preferences; health care systems and legal or regulatory issues such as insurance status; and discrimination. The IOM concluded that some differences, including variations in need for care as well as patient preferences, are allowable; others, including outright discrimination as well as insurance, are not allowable and would be considered disparities.
Dr. Cook noted, though, that what we think of as patient preferences often have origins in discrimination or other issues that we would consider unfair. Mirroring earlier work in renal transplant, a 2016 study in liver transplant found that black-white disparities in referral for transplant evaluation could not be explained by patient preferences; in fact, all patients reported wanting a transplant if it was recommended. Also, Dr. Cook noted that “discrimination from past experiences may inform your preferences,” and that using preference as an allowable source of difference in treatment is “fraught with difficulty.”
Access to liver transplant represents a disparity that already has been tracked carefully over time, in particular regarding the geographic differences in transplant rates and waitlist times. It is important, Dr. Cook said, to consider the intersectional nature of racial and geographic differences when conducting transplant studies. “How do we factor racial and ethnic disparities into geographic disparities, and vice versa?”
He noted that assessing access to liver transplant by MELD strata follows the spirit of the IOM report, in that it takes need into account. A full picture of racial and ethnic disparities in liver transplant, though, will require consideration and careful measurement of disparities in detection, referral, follow-up and other steps, before those disparities can begin to be eliminated.