A link between patient safety indicators (PSIs) and patient outcomes can be found—but only after researchers went deeper than claims data would allow and questioned the accuracy of billing based on PSIs.

The study, published in the American Journal of Medical Quality, was written by a team of Ohio State University researchers led by Darrell Gray, MD, MPH.

Looking at patient discharges at six hospitals between 2012 and 2014, the researchers used “clinically validated” PSIs which had been reviewed by an internal quality team. The goal was to see whether outcomes like length of stay, readmission and mortality changed compared to cases not flagged for a PSI after controlling for the “the poor validity of the current PSI algorithm.”

“Cases flagged with a clinically validated PSI are associated with a statistically greater length of stay, 30-day unplanned readmission, and mortality as compared to cases without a PSI,” Gray and his coauthors wrote. “This study demonstrates a strong association between clinically validated PSIs and patient outcomes.”

Patient flagged with clinically validated PSIs were more likely to be readmitted in every PSI type. For example, 16 percent of patients flagged for postoperative respiratory failure were readmitted in 30 days, compared to 6 percent of those not flagged.

That same PSI type saw a great difference in mortality: those flagged with a clinically validated case had a nearly 20 percent mortality rate compared to 0.1 percent rate among patients not flagged.

Changes are likely to be made in the near future. The National Quality Forum is due to submit a report evaluating to use of claims data to measure PSIs before the end of its next patient safety cycle.