Adjusting Outcomes

I wrote recently about the need to take into account patient characteristics when using patient outcomes to compare the quality of care provided by different physicians. That is a well-accepted principle, and the need for so-called “risk-adjustment” applies not only to evaluating physicians, but also to evaluating hospitals and larger care delivery systems. There has been a smoldering controversy, however, about which patient characteristics to consider and, in particular, the implications of including socioeconomic factors in such comparisons. This controversy played out again in a recent issue of the Annals of Internal Medicine.

Here is the core of the issue.

It is well known that “health” of an individual or population depends on a lot more than “healthcare.” Good health outcomes are much more dependent on the collective impact of the physical environment, genetic predisposition, individual behavior (e.g., smoking) and socioeconomic factors such as education level, income and social support. On the face of it then, it would seem “unfair” to evaluate care providers and, by extension, compare them with one another, without considering all of these factors. We might question, for example, an assessment of the quality of asthma care based on hospitalization rates in the local population that did not take into account the fact that some communities have a much higher prevalence of asthma than other. And yet, if the outcome (in this example, rates of hospitalization) are adjusted to reflect the local prevalence of disease, is this essentially the same as saying that the healthcare system has no responsibility for addressing the greater burden of disease in some communities? Is “adjusting” for these differences tantamount to accepting disparities in health outcomes that we should be striving to eliminate?

In the Annals, a paper by Kind et. al. looked at differences in 30-day rehospitalization rates in the Medicare population, and found that “residence within a disadvantaged U.S. neighborhood is a rehospitalization predictor of magnitude similar to chronic pulmonary disease.” While some may read that and conclude that comparisons of rehospitalization rates should therefore take this into account, an accompanying editorial by Krumholz and Bernheim – who developed the Medicare rehospitalization measures – argued otherwise. They wrote: “To address disparities in outcomes, existing disparities should be fully transparent. Measures that obscure disparities [i.e., by adjusting for socioeconomic status or SES] can contribute to complacency in addressing differences in outcomes by SES and race. “

I agree with Krumholz and Bernheim. What do you think?

1 thought on “Adjusting Outcomes

  1. A perfect example of Occam’s Razor- that an outcome with fewer assumptions is usually correct. Further assumptions can change the outcome and be incorrect too.

    A poor person in a minority population could be receiving better physical, emotional, and financial support from their family and would have a better outcome than somebody that has $$$. The $$$ person could have the family taking advantage of the physical condition to empty the coffers. Such people would also ignore, neglect the family member.

    A person could go on forever about assumptions and accurate results and glaring mistakes.

Join the Discussion! Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s