I've been closely following the Premier Hospital Quality Incentive Demonstration (HQID, or simply the "Premier Demo") ever since 2006, when the first results came out. They just released the results from year three and they are quite positive overall. More than a million patients at more than 250 hospitals have received improved care because of this innovative program that Premier launched with CMS in 2003. But for the first time in this project, hospitals that performed below established benchmarks were financially penalized.

From years one through three, quality of the 30 "nationally standardized and widely accepted care measures" has risen by nearly 16% on average for participating hospitals. These 30 measures are across five clinical areas: heart attack (AMI), heart bypass surgery (CABG), heart failure (HF), pneumonia (community-acquired, CAP), and hip and knee replacement. Approximately 2,500 heart attack patients' lives were saved during the first three years. That's progress.

The Premier Demo is unique, though, because year three marked the first time hospitals could feel financial pain if they didn't meet the benchmarks set in year one (for more detailed info, see here). In addition to the incentive (carrot) of earning more, they had the threat (stick) of earning less in year three if they didn't rise above the benchmark. They must beat the scores of the bottom fifth of year one.

A number of hospitals rose to the occasion; in fact, 15 hospitals went from "worst to first," moving from the bottom quintile to the top quintile in at least one clinical area between years one and three. These 15 improved by nearly a third in their quality score.

The carrot/stick approach also reduced clinical variation among the participating hospitals (which you know from reading this blog is a good thing):

The range of variance among participating hospitals also is closing, as those hospitals in the lower deciles continue to improve their quality scores and close the gap between themselves and the demonstration's top performers. This holds true for all types of participants—small and large, urban and rural, teaching and non-teaching.

Finally, the sticks. What about those that didn't make it above the benchmark in certain areas? There were only a few: 9 hospitals out of over 250 had at least one clinical area below the benchmark; another way to look at it, 11 hospital-clinical-areas were penalized out of 1,028 possible (hospitals multiplied by number of areas participating). That's a very, very small number, and more than half of the 11 were related to hip and knee replacement, which thankfully has the lowest mortality of the five conditions.

How much does the stick hurt? The 9 hospitals paid a total of $103,922 for their 11 under-performing conditions. That's pretty meager, especially compared to more than $7 million in bonuses paid this year by CMS in this project, and nearly $25 million over the three years, but the principle is the definition of value-based purchasing: provide high-quality care on standardized, nationally recognized benchmarks and you'll get paid more; if not, face a financial penalty.

Modern Healthcare has coverage of the Premier Demo here. I'll post soon with more reflections on this exciting topic.