Tuesday, February 06, 2007

Congratulations to our competitors

Congratulations to North Shore Medical Center, Newton-Wellesley Hospital, and South Shore Hospital for being included in the Fifth Annual Health Grades Hospital Quality and Clinical Excellence Study. This study identifies hospitals that have below-average mortality rates and complications across a wide range of procedures and diagnoses. Medicare administrative data covering 40 million discharges from 2003 through 2005 were used to make these calculations. Of 4,971 hospitals nationwide, only 266 received this designation.

3 comments:

Anonymous said...

OK so was anyone else surprised by the "demographic" of the hospitals making this list? No BIDMC, BWH or MGH but Newton Wellesley is on it. At first blush it looks like big city academic medical centers don't make the list. Is this because they take more difficult cases or there is some other instrumental variable that the study method doesn't adjust to compensate? No, wait - UCLA Medical Center makes it. So, Paul, or anyone else qualified to answer, what gives?

Anonymous said...

An interesting observation. I asked our folks. Here's what they say:

"We are not sure. Their risk adjustment methodology is proprietary. We have looked at our HealthGrades performance in the past, and the most concerning aspect was the use of coded diagnoses to determine complications. For example, atrial fibrillation is classified as a complication for several surgical procedures. But, in our review of cases, in the vast majority it has been coded because it is present on admission, and does not infer a complication. The “present on admission” modifier that has very recently been introduced to coding in the US should improve the accuracy of this type of analyses, but right now that is still a problem."

And Sir Brian Jarmin from IHI, offered a distinction between his work and theirs:

"The big differences between my approach and that adopted by Health Grades are:

"1) that [beyond single diagnoses analyses] I also do the Hospital Standardised Mortality Ratio (HSMR), which provides an overview of the hospital and has a number of advantages over single diagnoses (narrower confidence intervals, allowance for variations in coding practices, the fact that the case-mix acts as a proxy for social conditions in an area), and

"2)I adjust for the, mainly community, conditions in the area of the hospital such as the community morbidity, availability of intermediate care, availability of physicians and quality of community care; using these analyses I calculate the regression-adjusted HSMRs."

Anonymous said...

I live near South Shore Hospital but if I had a serious condition there is no way I would go to SSH to be treated. My father was a patient there five years ago and they ended up performing a procedure that was unnecessary and he wasn't getting much better under their care. We had him transferred here to BIDMC and his condition improved considerably very quickly.
I know other people who have been patients there and most were not particularly happy with the care there. So this report may look very favorably on them but my personal experience and that of those I know tell me something different.