Health care is the most information-intensive industry in the economy, but it uses information technology less intensively and is organized less effectively than almost any other sector of the economy. This fact is a major contributor to the enormous divergence between what is possible with modern medical care and what is actually delivered to patients.
How can we improve our health care infrastructure to generate savings? Investments in health information technology are the first need. In a summary of the potential of health IT, the RAND Corporation estimated $77 billion per year net savings if electronic medical records were fully adopted. We also take that as our goal, assuming that potential can be realized over the course of a decade.
We start by backing out areas where subsequent studies suggest the RAND estimates were too optimistic. We then add in additional savings from more recent studies showing possible savings in interactions between physicians and insurers, and in the reduced time spent on documentation that could come with voice recognition software and links between medical equipment and electronic medical records, all which we will detail below. We assume that 90 percent of these savings are realized by 2019, consistent with the CBO estimate. Finally, we net out the savings from health IT that the Congressional Budget Office has already accounted for.
The result of these infrastructure changes is a federal saving of $196 billion in the next decade. Some savings will flow into practices and hospitals directly from IT investments. These include allowing a physician to enter notes about a patient’s condition and care directly into a computerized record, thus eliminating or substantially reducing the need for clerical staff to physically pull medical charts from office files. Health IT and administrative simplification will also free up physician and nurse time for more productive tasks.
For more on this topic, please see:
- The Two Trillion Dollar Solution, by Melinda Beeuwkes Buntin and David M. Cutler.