Study: It's time to take a harder look at how hospital readmissions are defined as a quality measure

Patient and nurse in hospital
Hospitals are required to report their readmissions data by commercial payers and state Medicaid agencies. (Getty/monkeybusinessimages)

Healthcare regulators and insurers have long attached a high value to hospital readmissions rates as an indicator of hospital quality.

But a new study published Thursday in the New England Journal of Medicine offers more evidence that the quality calculation is getting less accurate over time. Why? Because it omits the increasing number of patients admitted for observational stays.

RELATED: 1 in 5 Medicare patients return to hospital after an observation stay

Amber Sabbatini, M.D.
(University of Washington
School of Medicine)

"Not only are readmissions going up for patients who are discharged from an observation stay, but the number of patients being placed in observation is going up each year," Amber Sabbatini, M.D., assistant professor of emergency medicine at the University of Washington School of Medicine and lead author of the study, told FierceHealthcare.

The study examined patient-level claims data from 2007 to 2015 from the Truven Health Analytics MarketScan Commercial Claims and Encounters Database.

In 2015, 57% of emergency department patients admitted to hospitals were inpatients, while 43% were admitted for observation stays. The data showed hospital readmissions within 30 days of an inpatient discharge dropped from 17.8% to 15.5% between 2007 and 2015. Hospital readmissions within 30 days of an observation stay increased from 10.9% to 14.8% in that time.

RELATED: As readmissions fall, observation status rises—even at hospitals outside readmissions program

This could obviously have a wide-reaching impact for hospitals.

Hospitals are required to report their readmissions data by commercial payers and state Medicaid agencies which occasionally link reimbursement and purchasing agreements to performance. In 2012, the Centers for Medicare & Medicaid Services introduced the Hospital Readmissions Reduction Program, penalizing hospitals with higher-than-expected 30-day readmission rates.

Sabbatini said she does not believe hospitals are gaming their admissions status in order to improve for data. But the phenomenon, which has been observed by other researchers in recent years, deserves a harder look and some thought about how to incorporate observation data into quality measures.

"The intent of readmission measures in the first place is to improve quality of care for patients with acute illnesses. But we're essentially ignoring a pretty large group of patients who are hospitalized with acute illnesses under the current framework," Sabbatini said.