In a 2011 JAMA article, Gil Welch and colleagues looked at how many chronic diseases Medicare enrollees had across different regions of the country. They came up with the following picture:
This picture may confirm some of your suspicions. For example, elderly in the north are sicker than ones in, say, Arizona or Southern Florida, possibly because they are too sick to move there in retirement.
But Welch’s study points toward another conclusion: doctors in some parts of the country are more inclined to diagnose their patients with lots of illnesses than doctors elsewhere, even taking into account how sick patients are. They conclude this, in part, because the mortality rate of people with, say, 2 chronic illnesses is lower in parts of the country that hand out lots of diagnoses:
This could mean that doctors in some places hand out too FEW diagnoses. Or that doctors in other areas hand out too MANY. Or that both things are going on.
One thing is clear from this study, however. Before Medicare can consider “pay for performance” schemes, that reward hospitals that achieve good outcomes for their patients, they need not only to adjust for how sick the patients are, but also how stingy the hospital’s doctors are in handing out diagnoses.