Why Hasn't IT Transformed U.S. Health Care? And Will It?

By TAP Staff Blogger

Posted on September 7, 2010


The following article, by Professor Jonathan Levin, was published in the June 2010 issue of the Toulouse Network for Information Technology (TNIT) newsletter. It is re-published with permission.

Health care, unlike other major sectors of the U.S. economy such as retailing, manufacturing, and financial services, has yet to experience an “IT revolution”. Although hospitals and physicians use computerized billing, and insurers compile data on administrative claims and drug prescriptions, information technology plays a relatively small role in health care delivery. According to the 2009 National Ambulatory Medical Care Survey, less than half of physician offices maintain patient records electronically, and fewer than 10% have what the Department of Health and Human Services defines as a fully functioning electronic health record system.

It is easy to think of ways in which electronic health records (EHRs) might lead to benefits in clinical practice. Nowadays, patients frequently receive care from a number of physicians, specialists, and other professionals. An EHR can permit all of these providers to access complete health information on patients rather than having to transmit information back and forth using phone calls and faxes. A well-designed HER system can provide reminders about preventive care, information about clinical guidelines, drug interactions, and other decision support.

Storing data electronically can also allow a physician or healthcare organization to track patients at a population level, and more easily measure performance and clinical outcomes. In principle, data can be transmitted and aggregated to allow tracking of epidemics, postapproval monitoring of drugs and devices, and a wide range of clinical and health systems research. In fact, many of these benefits seem so obvious it is hard to believe they are not possible today.

So why hasn’t IT been adopted and utilized more widely? One answer lies in the organizational and economic structure of the health care system. Health care delivery in the United States is highly fragmented. More than half of U.S. physicians work in groups of fewer than five physicians. A typical Medicare enrollee might see physicians from six or seven organizations in a given year. This structure makes it difficult to internalize the data-sharing and coordination benefits promised by EHRs.

The payment system also plays an important role. Most physicians are compensated on a fee-for-service basis, which creates little incentive to invest in technologies that would reduce duplication or increase efficiency, and also supports the fragmented organizational structure we currently have. It is hardly surprising that the leading users of IT are large integrated organizations such as Kaiser Permanente, which are paid on a capitated basis.

Recent federal policy initiatives could begin to change this. In 2009, as part of the stimulus bill, Congress allocated roughly twenty billion dollars to promote health IT. Physicians who adopt EHRs and satisfy a “meaningful use” requirement can receive subsidy payments from Medicare. Starting in 2015, nonadopters will have their Medicare reimbursements reduced. Already, there has been a proliferation of cloud-based EHR systems aimed at small physician groups, and a large fraction of physicians are likely to adopt EHRs in the not-too-distant future.

Of course, the mere adoption of electronic records may not have a major effect. In fact, the first draft of the meaningful use criteria spells out tasks so basic that one worries what physicians would be tempted to do with their new EHR systems absent the regulatory requirements. Still, it seems possible that EHR adoption will be complementary to other reforms. Improvements in measurement frequently enable improvements in contracting, and a great deal of economics research supports the claim that IT can facilitate increases in organizational efficiency.

The stimulus bill also contains another, and possibility even more far-reaching idea, which is to link patient records into a national health information network. Such a project faces substantial challenges, including concerns about privacy, difficulties in agreeing to technological standards, and the likely resistance of hospitals and other health organizations to share proprietary data. The federal government has also outlined a plan that envisions health organizations exchanging documents, but does not require data to travel with patients or be readily assembled and aggregated for public health uses.

These more ambitious steps are probably needed to realize the benefits described above, and certainly if we learn anything from the experience of technology industries over the last fifteen years, it is the power of creating a flexible communication platform and unleashing market innovation. If these things happen, who know? Maybe in ten years some of the engineers developing iPad applications will be trying to improve the healthcare system.