Blog Home


Should Provider Performance Measures Be Risk-Adjusted For Sociodemographic Factors?

March 27th, 2014

The National Quality Forum released draft recommendations on March 18 to change the way we assess the care that doctors and hospitals provide, and they are sure to cause a buzz in and beyond the health care community. That’s a good thing, because reflection and conversation are vital pieces of ‘getting it right’ when determining how measures can be used to gauge healthcare performance.

The recommendations come from a panel of 26 national experts convened by NQF at the request of the federal government. The question before them:  Should the measures we use to assess providers’ performance be risk-adjusted to account for patients who are poor, homeless, illiterate, uneducated, or have other indicators of lower socioeconomic status? The panel’s recommendations are discussed below, and we encourage you to register your views by commenting on the report by April 16 and on this post.

The issue. NQF-endorsed measures are used to indicate strengths and weaknesses in the quality of care provided by clinicians and hospitals. NQF’s policy recommends adjusting some performance measures for clinical factors such as a patient’s severity of illness, recognizing that a patient who is sicker and has multiple conditions and comorbidities has a higher likelihood of worse outcomes, regardless of the care provided.

Increasingly, policymakers and researchers who study disparities have raised the question of whether performance measures would be even more accurate if they were adjusted for sociodemographic factors as well. Indeed, there is a growing understanding that social determinants significantly influence a person’s health. Factors far outside the control of a doctor or hospital—patients’ income, housing, education, even race—can significantly affect patient health, healthcare, and providers’ performance scores.

NQF’s current policy does not accept adjusting performance measures for any sociodemographic factors. We’ve always wanted to shine a light on any and all disparities in performance not attributable to health status because many believed it motivated providers to improve the care they delivered to disadvantaged populations.

What our expert panel recommends. But most of our expert panelists suggest otherwise.

A wide majority feel that not adjusting for patients’ sociodemographic factors might actually harm patients, exacerbate disparities in care, and produce misleading performance scores for a variety of providers, which means that no one has accurate information to use for comparison. They say that the effects of inaccurate assessments of quality are intensified when used in pay-for-performance programs. As providers are increasingly paid based on the quality of their care, they believe some will avoid serving disadvantaged populations altogether, which will then worsen access to care for the most vulnerable patients. They posit that performance-based payment incentives will shift from those who serve the disadvantaged to those who serve the affluent. Safety net providers may then have fewer resources to care for vulnerable populations and the array of additional services that they need.

These experts also say that consumers and payers will avoid providers who serve disadvantaged populations because they are labeled poor performers, which may not accurately reflect the actual quality of their care.  Panel members say that even with an adjustment for sociodemographic criteria, poor performance would still be transparent. An adjustment would enable all providers to be compared equally, with true gaps in quality made apparent.

A very small minority of the panelists feel there should be no adjustment to measures based on patient sociodemographics. They believe it could artificially raise the performance score of providers treating the most vulnerable patients, whereby two doctors might appear equal, even though one’s performance is considerably lower. They say it is not appropriate to have a different standard for providers who treat low socioeconomic status patients.

We want to hear what you think. Comments on the draft report are accepted until April 16 at 6:00 pm ET.

Getting this right is important, because reliable measures are critical for everyone. Patients and consumers need them to accurately choose high-performing doctors and hospitals. Providers of medical care need them to better understand and address gaps in their performance. Private and government payers need them to ensure the care they pay for is as good as it can be.

The National Quality Forum is, above all, a forum, so we welcome your comments and engagement.

Email This Post Email This Post Print This Post Print This Post

 to the #1 source of health policy research.

No Trackbacks for “Should Provider Performance Measures Be Risk-Adjusted For Sociodemographic Factors?”

2 Responses to “Should Provider Performance Measures Be Risk-Adjusted For Sociodemographic Factors?”

  1. Peter Glusker Says:

    Definition of illnesses must include socioeconomic factors if treatment outcomes is to be accurate. That is, not all MI’s are ‘equal’, despite all being MI’s. Size, location, etc AND socioeconomic factors ( with several subsets…nutritional, educational level, work exposure, bmi, etc) So, analysis of outcomes needs to be measured in terms of the total disease burden(s) being treated. However, accurate measurements of the various disease burdens and outcomes is lacking and thus creates inaccurate outcomes data.

    Restated, if outcomes measurement does not include an accurate assessment of total disease burden (including socioeconomic factors) then ‘pay for performance’ conflates true quality with incentive-motivation .

  2. JPMolnar,DO Says:

    Include in any deliberations available resources. In a rural area where I practice certain specialists are virtually not available for definitive services.
    Likewise, tying income to areas that we cannot control leads to cherry picking and lemon dropping. In England at the present time, the physicians have the right to exclude those patients who are non compliant,and have other conditions which cause them to lose income. Despite that, England is on the brink of collapse of their GP system.
    One other consideration and that is certain inherent conditions that change outcomes is hearing deficiency, visual difficulties, reading deficiencies, gait disturbances, dementia of any cause and patient accompanied by a caregiver.
    Thank you for the opportunity to comment.

Leave a Reply

Comment moderation is in use. Please do not submit your comment twice -- it will appear shortly.

Authors: Click here to submit a post.