New pay-for-performance rate disagrees with physicians 90% of the time

Using claims data, the Agency of Healthcare Research and Quality (AHRQ) developed the preventable quality indicators (PQI) rate. CMS has begun adopting the PQI rate into its incentive and penalty program as a way to measure the number of preventable hospital admissions. The hope is that by using the PQI rate between 9% to 36% out of the 40 million annual admissions could be eliminated.

The problem? The PQI wasn’t meant to be used that way. The AHRQ developed the PQI to assess population-level access to primary care across a geographical region.

"The PQI cannot identify an appropriate rate for hospital admissions within a given population," said Krishna Patel, MD, and colleagues in a report. "Therefore, it cannot serve as a tool to help hospitals focus their prevention efforts."

Patel’s team ran a study to see how often a physician’s assessment of a case’s preventability agreed with the PQI. When asked to rate 322 admissions on if they were preventable, physicians said that 38% were preventable, while PQI ruled 23% of admissions as preventable. However, physicians and the PQI only agreed on the preventability of 10% of admissions.

In their conclusion, the researchers said that physicians were better able to identify more preventable admissions along with methods of prevention and tools for local quality improvement. They also wrote that the PQI tool was unable to identify a large proportion of preventable admissions and more study is needed on its use in pay-for-performance programs.

"For healthcare providers to reduce admission rates, they must understand why preventable admissions occur: 52% of the preventable admissions in our study were considered very or somewhat easy to prevent, and these were more likely to be due to clinician factors such as inadequate follow-up, no contact between the admitting MD and primary care provider and the low threshold for admission," the authors wrote.

Found in Categories: 
Quality & Errors

More Like This