PPACA has been attempting to bend the cost curve by penalizing stupid and avoidable errors. One class of errors that has been amenable to reduction has been Medicare beneficiaries getting re-admitted to hospitals after their initial admission for a set of circumstances. As soon as data started to be collected and before penalties started to be imposed, the readmission rate crashed. Since penalties have been imposed, the rate is still going down but at a slower pace. This does two things to the cost curve. First, it reduces direct Medicare expenses as Medicare is not paying for another hospital day. Secondly, changes in practices and procedures that result in lower Medicare readmission rates tend to diffuse throughout a hospital and all of its patients so people who are not covered by Medicare also benefit from the improvement in practice.
This sounds great. We save money, save Grandma as hospitals are where old people die, and get better quality care.
However, wonks have worried that any quality measure that has real money attached to it can and will be gamed.
The easiest way to game a re-admission measure is to redefine admissions. Hospitals have the ability to put people on “observation” status where to anyone but the billing and quality metrics department, the person looks like they are admitted. They get the uncomfortably flimsy robe, they get the wrist band, they get poked and prodded and monitored just like an admitted patient. They can stay in observation status for a time period including one midnight. Yet these individuals are not part of the “admitted” or “re-admitted” population universes.
There was a possibility that a significant chunk of the seemingly great decline in readmission rates was really a bureaucratic shift of people getting moved from short term admissions to observation status.
That is not the case.
The New England Journal of Medicine * has an interesting study on this matter: