Can higher readmission rates ever be better?

Here's a wrinkle in the idea of disclosing quality metrics: what if one of the metrics is misleading? The rate of hospital readmissions is a commonly accepted quality metric: hospitals that have more risk-adjusted readmissions are typically deemed to provide worse care. Medicare even ties reimbursement to readmissions, penalizing hospitals that have readmissions that are too high. Some doctors from Cleveland Clinic have argued that not all readmissions are bad. Their angle is provocative, showing that hospitals that had a higher readmissions rate for heart failure tend to have a lower mortality rate. The authors posit that the higher readmissions rate is because hospitals operated sufficiently well to avoid patients' deaths, although those cases were complicated enough to warrant a second visit to the hospital. The argument goes that the hospitals that could not prevent death had fewer readmissions as an unintended side effect.

I'm not familiar enough with the issue to understand how plausible this argument is, but I can very well see the possibility that some metrics need refinement. If this is a serious issue, then the caveat should be noted when patients are viewing such data. Nevertheless, I consider the collection and dissemination of this data to be a win for transparency, even if it comes with a counterintuitive disclaimer. Starting with this data and then understanding its limitations moves the conversation further along much more than if the data were never collected to begin with. Transparency gives people visibility; sometimes, though, the image can be hard to interpret.