• The difference between high-quality, low-cost and low-quality, high-cost hospitals

    There’s a new study in Health Affairs that examines differences between “high” and “low” quality hospitals:

    As policy makers design national programs aimed at managing the quality and costs of health care, it is important to understand the potential impact on minority and poor patients and the hospitals that provide most of their care. We analyzed a range of hospital data and assigned hospitals to various categories, including “best”—high-quality, low-cost institutions—and “worst”—where quality is low and costs high. We found that the “worst” hospitals—typically small public or for-profit institutions in the South—care for double the proportion (15 percent versus 7 percent) of elderly black patients as the “best” hospitals—typically nonprofit institutions in the Northeast. Similarly, elderly Hispanic and Medicaid patients accounted for 1 percent and 15 percent, respectively, of the patient population at the best hospitals, while at the worst hospitals, these groups represented 4 percent and 23 percent of the patients. Patients with acute myocardial infarction at the worst hospitals had 7–10 percent higher odds of death compared to patients with those conditions admitted to the best hospitals. Our findings have important implications for Medicare’s forthcoming value-based purchasing program. The worst institutions in particular will have to improve on both costs and quality to avoid incurring financial penalties and exacerbating disparities in care.

    Some days you can’t turn on the TV without some politician proclaiming that we need to pay for “quality”, not “quantity”. Hell, I’m sure I’ve said it here at least a few times. But I’ve also been skeptical about the idea that “pay for performance” will be as easy as everyone thinks. This study is one of the reasons why.

    One of the ways you can have really good stats is to treat healthier people in general. It may be difficult to do that on a person by person basis, but in the aggregate it’s not as hard as you might think. Wealthy people tend to be healthier than poorer people, in general, so focusing your practice in wealthier neighborhoods, and limiting your Medicaid exposure, may tend to improve your outcomes. Minority populations tend to be unhealthier than non-minority populations as well. Therefore, hospitals that tend to have lower numbers of black or Hispanic patients might look better, too.

    It gets worse. Because the hospitals that cover more poor people and more minorities tend to cover more unhealthy people, they may spend more money as well. So it’s possible for them to get the worst of both worlds. Their quality looks worse and their costs look higher.

    Please understand, I’m not taking back what I said before. I really do think we need to start paying for quality and stop paying for quantity. But when politicians do health policy, they like to reduce everything to sound bites. They like to make things simple. There will be a strong impulse to create straightforward metrics like this study used to grade which hospitals are doing “well” and which are “wasting money”. When they do that, and decide to reward those that come out on top, hospitals that need the resources the most may be deprived of them first.

    That’s backwards, and avoidable. Disparities will worsen, not improve. You can’t say you haven’t been warned.

    Share
    Comments closed
     
    • There is some dissonance:

      High spending generally has no, if not lower associations with quality metrics, This is familar theme from Dartmouth.

      The association between process measures and outcomes–a work in progress:
      http://jama.ama-assn.org/content/296/22/2694.abstract

      Square the circle.

      The paper’s findings are interesting, altogether not surprising, but my bet is cost, quality, demographics, are confounded to a degree, and the attn the press gives the paper will be overhyped; there is more under the hood and the devil is in the details.

      brad

    • Great post!

      I have bashed a lot of Aaron’s prior posts but a post like this one is why I read this blog. It really got to the heart of a really complex problem.
      We (our firm) are trying to design a P4P program for an entire state right now and running into very similar issues.

      There are some workarounds to this issue but in practice, they are extremely complex and people tend to game them. Ultimately, a system based on measuring inputs is going to be far easier and appear to be far more “fair.” This is extremely unfortunate…

    • Very much reminds me of some of the problems in evaluating schools for performance. Is performance evaluation of teachers going to get rid of bad teachers, or is it going to punish teachers who work with the toughest populations…

      Just anecdotaly, here are some of the ways that I’ve seen “Wealth” or “Education-level” play into health-outcomes (and I’d love to know if there is data on any of this) and all of these are outside of things like prior health status, smoking use, etc.:
      1) more family support in hospital (family can take time off?), great for encouraging ambulation or incentive spirometry use post-surgery, seems to reduce post-surgical opiate use (patient is distracted, doesn’t want to be zonked), reduces ICU related delirium and in-hospital falls, increases patient mood, facilitates good and accurate communication with doctors/nurses
      2) better compliance with discharge instructions (PCP follow-up, prescription filling/use, physical therapy) which I would hope reduces readmission rates
      3) fewer non-medical barriers to discharge (where do you discharge a homeless person to? how about a new single mom who just had a c-section?)
      4) wealthy/educated individuals seem to have quicker recognition of serious conditions (or more willing to come in because they have insurance…) always feel like they come in at hour one of a stroke or MI, rather than hour 12….

      take these with a grain of salt, like i said I have no idea if there is evidence of these effects they are just observations.
      anyways, this post just struck a chord with me, as someone who works at a public hospital and lives with a public school teacher…

    • Minority populations tend to be unhealthier than non-minority populations as well. Therefore, hospitals that tend to have lower numbers of black or Hispanic patients might look better, too.

      Hispanic and Asian minorities tend to healthier than whites and blacks but your main point is still good. The reason that I post this correction is that people should know that Hispanics are healthier despite being poorer more likely to be obese and more likely to have type 2 diabetes which points to genetics. On factor might be that they like Asians have fewer multiple low weight births.