MMS 2014 Public Health Leadership Forum: The Impact of Health Care Reform on Health Care Disparities

The key to lessening health care disparities lies in better data collection, pay-for-performance systems that properly measure and reward improvement, and technology that engages patients in their own treatments, according to according to a panel of experts featured at the 2014 MMS Public Health Leadership Forum.

The presentation, “The Impact of Health Care Reform on Health Care Disparities,” was hosted by MMS April 4 in collaboration with the national Commission to End Health Care Disparities.

As one of the first states to pioneer universal health coverage, the nation is looking to Massachusetts for ideas and solutions as this year’s implementation of the Affordable Care Act is expected to bring millions of previously uninsured patients into doctors’ offices, said MMS President Ronald Dunlap.

Massachusetts has lower-than-average rates of disparities in key health areas such as infant mortality, hypertension, obesity and adult diabetes.  But access to primary care physicians in certain regions of the state remains a problem, as do Medicaid payment models that dis-incentivize physicians, said Dr. Dunlap.

Dr. Joel Weissman

Joel Weissman, PhD

Can Pay-for-Performance Create Equity?

Among the most promising tools for bridging the gaps are new payment models that measure and reward reductions in disparities, said Joel Weissman, PhD, Deputy Director and Chief Scientific Officer Center for Surgery and Public Health at Brigham and Women’s Hospital.  “No information means no improvement,” he said.

But most current pay-for-performance models are not effectively addressing disparities and creating incentives that could reduce them.  “Not only do we need to know more about measures that are “disparities-sensitive”, but how to select measures that are ready to have an impact on clinical practice, and how to represent differences in a statistically meaningful and policy-relevant way,” Weissman said.

Dr. John Moore

John Moore, MD

Patient Empowerment Through Technology

Grassroots approaches to health, including personalized patient engagement and “navigators,” who help patents cut through red tape to get social services are already helping reduce disparities in some areas.

John Moore, M.D., CEO and co-founder of Twine Health, said the new health models must also include the patient as “an active participant.”  The old-fashioned paternalistic doctor-patient relationship is fading away, he said. Patients of the future will set their own health care goals and meet them using technology and peer support.

The approach has already worked, he said, citing his study published in 2013 in the Journal of Clinical Outcomes Management that found hypertension controlled in a group of patients for less than 30 percent of the average annual Medicare cost for the same outcome.

Sonia Sarkar

Sonia Sarkar, MPH

Making Physician Advice Actionable in the Moment

Another effective disparities-reducing program has been Boston-based Health Leads, which connects patients to advisors who will coordinate the nitty-gritty details of social services and enter the information on a patient’s EMR for physicians to track and follow-up, said Sonia Sarkar, the company’s chief of staff to the CEO.

The program has partnered with major medical centers in Boston, Providence, Baltimore, Chicago, New York and Washington D.C. and helps them close disparity gaps for patients without resources to get or remain healthy. Connecting patients at risk of disparities to needed food, heat, child care, transportation or other services makes “the doctor’s advice actionable in the moment,” Sarkar said.  “It insures health care delivery is centered around health.”

See the full forum agenda and download the presentations here.

–Erica Noonan

  1. The presentations at the leadership conference are reflective of a problem with all health and healthcare disparities research conducted without recognition of the patterns by which standard measures of differences between outcome rates tend to be systematically affected by the prevalence of an outcome.

    The problem is highlighted by the statement attributed to Professor Joel Weissman to the effect that “among the most promising tools for bridging the gaps are new payment models that measure and reward reductions in disparities.” The main (or only) existing effort to tie payment models to disparities reductions is in the Massachusetts Medicaid pay-for performance (P4P) program. But as a result of a failure to understand the ways that measures employed in the program tend to be affected by the prevalence of an outcome, the disparities element of the program is more likely to increase than decrease healthcare disparities. See pages 30 to 32 of my Federal Committee on Statistical Methodology 2013 Research Conference paper “Measuring Health and Healthcare Disparities” (FCSM Paper). See also the Between Group Variance subpage of the Measuring Health Disparities page of jpscanlan.com.

    Professor Weissman’s presentation itself (slide 7) shows a situation where one would get reach different conclusions about changes in the size of disparities between 2000 and 2010 depending on whether one examined relative differences in failure to receive an appropriate test or absolute differences between rates of receiving the test. (It also shows, for those willing to perform the calculations, that relative differences in receipt of the test would yield an opposite conclusion from that yielded by the relative difference in failure to receive the test and the same conclusion as that yielded by the absolute difference.) But the presentation – like the measurement guide co-authored by Professor Weissman titled “Commissioned Paper: Healthcare Disparities Measurement” and discussed at pages 30 to 32 of the FCSM Paper – crucially fails to reflect an understanding (a) that each standard measure of differences between outcome rates tends to be systematically affected by the prevalence of an outcome; (b) that such fact makes it impossible to employ such measures to quantify the strength of the forces causing appropriate healthcare rates of advantaged and disadvantaged groups to differ without consideration of the way the measures are affected by the prevalence of an outcome; and (c) that there exists an underlying reality, and only one such reality, respecting whether the strength of those forces has increased or decreased over time or is otherwise larger in one setting than another.

    As a result of such failure in all of the conference presentations, the Commonwealth of Massachusetts will reasonably read those presentations as suggesting that the disparities element of its Medicaid P4P program is a step in the right direction rather than something that needs (urgently) to be materially modified or eliminated.

  1. There are no trackbacks for this post yet.