Skip to main content

Health IT savings estimates are 'wishful thinking,' say Harvard researchers

By Chelsey Ledue

Harvard researchers say an increase in the use of IT in hospitals hasn't lowered costs or made hospitals more efficient.

Those findings, published in the online edition of The American Journal of Medicine, indicate that health IT may have “modestly” improved the quality of care for incidents such as heart attacks. They fly in the face of claims by President Barak Obama and many lawmakers that health IT, including electronic medical records, saves billions and makes reform affordable.

"Our study finds that hospital computerization hasn't saved a dime, nor has it improved administrative efficiency," said David Himmelstein, an associate professor at Harvard Medical School and former director of clinical computing at Cambridge Hospital, who authored the report. "Claims that health IT will slash costs and help pay for the reforms being debated in Congress are wishful thinking."

Himmelstein and others analyzed data from approximately 4,000 hospitals, including those on a list of the "100 Most Wired," between 2003 and 2007 for evidence of increased quality, cost savings or improvements in administrative efficiency.

The data was compiled from The HIMSS (Healthcare Information and Management Systems Society) Analytics annual survey of hospital computerization, Medicare cost reports submitted annually by hospitals to the Centers for Medicare and Medicaid Services and the 2008 Dartmouth Health Atlas, which compiles CMS data on costs and quality of care.

Researchers found that U.S. hospitals increased their use of IT between 2003 and 2007, but there was no indication that it lowered costs or streamlined administration, even in the "most wired" institutions.

While U.S. hospital administrative costs increased from 24.4 percent in 2003 to 24.9 percent in 2007, the study said, hospitals that computerized most rapidly had the largest increases in administrative costs.

The study also found no evidence of delayed effects of lower costs in 2007 resulting from information technology introduced in 2003.

Modest quality gains were noted in the treatment of heart attacks (acute myocardial infarction) in more-computerized hospitals, but the study's authors said those improvements may have been caued by better documentation rather than an increase in IT use.

Himmelstein said a report from the Congressional Budget Office in 2008 signed by Peter Orszag, now Obama's budget director, expressed skepticism about claims by the RAND Corp. and others that health IT could generate $80 billion annually in savings.

"Part of the CBO's skepticism was based on the limited information available to the RAND study and similar studies," Himmelstein said. "But this new, detailed, national survey of diverse hospitals shows such doubts are well-founded. Information technology can't rescue us from our national healthcare crisis."

"Any savings may have been offset by the costs of purchasing and running new computer systems," said Steffie Woolhandler, a professor of medicine at Harvard and the study's co-author. "In addition, most software is designed around the accounting and billing needs of hospitals, not the clinical side."

She offered the Veterans Administration, where financial issues are separated from the clinical side and doctors focus on practicing healthcare, as an example of IT success.

"The VA system now has our nation's highest quality and patient approval ratings," Woolhandler said. "Congress should take note: To get the most benefit from our healthcare dollars and from health IT, we should adopt a single-payer, Medicare-for-all program. Nothing short of that will allow us to reap the full potential of computerization or to provide comprehensive, quality and affordable care to all."