When the outcomes of health care are monitored, it’s usually the overall outcome that’s measured (overall death rates, for example, or blood stream infections). These rates are risk adjusted. However, it is only the preventable component of an overall rate that reflects what we are after – the quality of the service. We have constructed an equation to show how well adjusted overall rates reflect the preventable component. When the preventable component is over 30%, as in bloodstream infections, then the overall rate is a good diagnostic test for the preventable component. However, if less than 20% of outcomes can be prevented by optimal care, as in hospital deaths, then overall rates are rubbish. No doctor would use them as a diagnostic test for a disease because both false positive and the false negative rates exceed 70%. There’s always a cost to false positives and such a test is likely to do more harm than good. The equation shows that the ability of an overall rate to predict the preventable component is also sensitive to how much the preventable rate varies across the system. More studies into preventable rates and how they vary are urgently indicated. In the meantime, the equation provides a useful conceptual framework for quality improvement efforts, not just in healthcare, but in all services.
Girling AJ, Hofer TP, Wu J, Chilton PJ, Nicholl JP, Mohammed MA, Lilford RJ. Case-mix adjusted hospital mortality is a poor proxy for preventable mortality: a modelling study. BMJ Qual Saf. 2012 [accessed 19 Oct 2012]. Available from: http://qualitysafety.bmj.com/content/early/2012/10/12/bmjqs-2012-001202.full