University of Minnesota researchers found that the use of health information technology (IT), a popular health policy initiative, has had little or no effect on patient safety. Despite this, the researchers say it's still too early to judge the effectiveness of health IT. They employed innovative methods that are shedding entirely new light on the issue, necessitating extensive further examination. The study, funded by the Agency for Health Quality and Research (AHRQ) and the Changes in Healthcare Financing and Organization initiative, appears in the March 2009 issue of Health Affairs.
Researchers from the U's Carlson School of Management and the School of Public Health examined nationally representative Medicare data on more than 10 million hospital inpatient stays for the years 1999-2002. They looked at the effects of three different health IT applications on three separate patient safety indicators. The IT applications examined were electronic medical records, nurse charts and automated imaging. The patient safety indicators examined were infection, post-operative hemorrhage or hematoma and postoperative pulmonary embolism or deep vein thrombosis.
They found that electronic medical records are the only IT application to have a clear, though marginal, effect on patient safety. Electronic medical records reduced infections, but had no effect on the other patient safety indicators that were studied. The benefits of electronic medical records did grow with time.
The study authors wrote that their research suggested that early adopters of health IT were typically large academic medical centers that provide high quality care and likely have better-than-average outcomes. Consequently, generalizations from early health IT adopters could overstate health IT's value on a national scale.
"While this is an interesting result, it is clear that much more research is needed on this issue," said Stephen Parente, associate professor in the Carlson School of Management and a study author. Parente and his co-author Jeffrey McCullough of the School of Public Health agree that planned federal investment in health IT should be accompanied by investment in the evidence base needed to evaluate its effectiveness.