Researchers examined mortality data for 1,984 hospitals nationwide from 2008 to 2012. During surprise inspections, 7.03% of patients died within 30 days of being admitted to the hospital, the study found. At other times, the 30-day mortality rate was 7.21%.
The difference was more pronounced at major teaching hospitals, where mortality dropped to 5.93% during inspection weeks from 6.41% at other times.
“Our findings are surprising because they highlight how increased focus, attention, and cognitive bandwidth, all of which happen when people are being monitored, could lead to measurable improvements in patient outcomes,” says senior study author Dr. Anupam Jena, a health policy researcher at the Harvard T. H. Chan School of Public Health in Boston.
The study focused on inspections by the Joint Commision, an independent nonprofit group that evaluates hospitals at least once every three years to see how well they follow patient safety guidelines designed to avoid preventable deaths, infections and errors like medication mistakes. Hospitals typically work hard during inspections to make sure they pass. Failures, which are rare, can result in a halt to payments from Medicare and Medicaid, which often make up at least half of hospitals’ revenue.
Jena and colleagues examined data on patients insured by Medicare, the U.S. health program for the elderly. The study included 244,787 patients hospitalized during inspection weeks and another 1,462,339 patients hospitalized during the three weeks before and after inspection periods.
Patients were typically about 73 years old and roughly 56% were female. Many of them had chronic health problems like high blood pressure, heart disease, elevated cholesterol or diabetes.
The researchers didn’t find any meaningful differences in how long patients were hospitalized, how many people were admitted to the hospital, reasons for diagnosis, patient characteristics or procedures performed based on whether or not it was an inspection week.
Across all the hospitals, however, they calculated that there was a 1.5% decrease in mortality during inspection weeks. For major teaching hospitals, researchers found a 5.9 percent decrease in deaths during inspection periods.
Major teaching hospitals admitted an average of 900,000 Medicare patients a year during the study period, researchers note in JAMA Internal Medicine, online on March 20.1
The absolute reduction in 30-day mortality rates of 0.39% during inspection weeks at major teaching hospitals suggests there may be some room to improve quality throughout the year, Jena says.
“Nearly 3600 deaths could be avoided each year, or 10 deaths per day, if care patterns that are observed during hospital inspection weeks were the same year-round,” Jena adds.
It’s possible that during inspection periods, doctors and nurses pay closer attention to things that the inspectors check like hand-washing to prevent the spread of infections. But there were not fewer infections or adverse safety events during inspection weeks than at other times, Jena says.
Another possibility is that shifts in how doctors and nurses make diagnoses and treatment decisions during inspection weeks might lead to better outcomes, Jena says.
One limitation of the study is that researchers weren’t able to identify what might have caused the lower mortality rates during inspection periods.
Patients admitted during inspection weeks might also be different from patients seen at other times in ways the study didn’t capture, notes Dr. Vineet Arora, a researcher at University of Chicago Medicine who wasn’t involved in the study.
Hospitals might also respond to the arrival of inspectors with heightened vigilance that translates into better quality care, Arora says by email. For example, checking prescriptions before people leave the hospital might improve mortality rates by reducing medication errors.
“A modest reduction in mortality is certainly still a positive thing,” Arora says.
“This study highlights that there is potential for us to learn what is going on during those weeks that is associated with better patient outcomes,” Arora adds. “The question is whether it is due to a concerted effort on the part of the hospitals to follow safe practices or whether there is something else going on.”
- Barnett ML, Olenski AR, Jena AB. Patient mortality during unannounced accreditation surveys at U.S. hospitals. JAMA Intern Med. 2017 Mar 20. doi:10.1001/jamainternmed.2016.9685