Skip to content Skip to navigation

Mostashari ‘Sets the Record Straight’ on Health Affairs Article

March 7, 2012
by David Raths
| Reprints
Article's focus on test ordering was too narrow, the ONC leader says

Kicking off the 33rd meeting of the federal Health IT Policy Committee on March 7, Farzad Mostashari, M.D., national coordinator for health IT, was eager to address the study published in the March issue of Health Affairs that questioned whether investment in health information technology would lead to cost savings.

Mostashari told the assembled audience that the only thing the study found was that clinicians with systems that let them view images ordered more images. “It is not a particularly surprising observation,” he said. What’s surprising, he added, is the conclusion that asks whether the federal government’s multibillion-dollar effort to promote health IT may not yield anticipated cost savings. Indeed, it suggests that it is possible that computerization will drive costs up, not down. “The interpretation obviously helps to grab some headlines,” he said.

But Mostashari added that he wanted to “set the record straight” on what the study actually found and what it actually did. The study was not about EHRs, much less about meaningful use of EHRs, he noted. He said the data was from 2008, and added that a lot has changed since then. The study didn't consider some of the aspects of meaningful use of EHRs, including decision support and information exchange, he added. He said that as an epidemiologist, he was surprised by the causality assigned by the study’s conclusions. “It could well be that ordering more tests leads to buying imaging systems, not the other way around,” he said.

He went on to say that this was an observational study that wasn't designed to answer questions about costs or about the relationship between EHRs and quality. It didn’t take into account factors such as the level of acuity of illness, physician training, the practice of defensive medicine and financial arrangements. From a clinical and patient point of view, the study didn’t answer questions about the appropriateness of the tests that were ordered.

“Finally, when we talk about health IT being the foundation for improving quality and safety and reducing costs, it is not going to come about by people ordering more or fewer lab tests,” Mostashari said. “The big savings are in improvements in coordination of care, and reducing unnecessary and harmful complications and hospitalizations.” Providers who are embracing new delivery systems such as accountable care organizations and patient-centered medical homes know they can’t succeed without health IT tools, he said.

“Despite the power of anecdote and headlines making an impact on our consciousness, we have to be careful to look at evidence systematically and not anecdotally,” he said.

A systematic review of the evidence shows that EHRs have the ability to give providers the tools and power they need to effectively improve quality and reduce costs, he said. “So this is one study, and it is a brief flurry of interest in the larger question. Yes we are succeeding in making progress on health IT. That part is no longer in question,” he said. “Appropriately, the conversation is now focused on what do we make of that progress. How will we as a country—providers and hospitals and vendors and academics and patients—use the progress? And that is where as least as great a challenge lies for us, in the effective use of these tools.”




...illustrates only that if a service becomes more useful (by providing added information) clinicians will use more of it. To draw any other inference regarding IT impact is extremely silly.

However, in my opinion, the greatest impact of full EHR implementation will be to increase value of interventions, in addition to eliminating waste and duplication, facillitating cost-effective care, and the other expected cost-saving effects that have been cited. I believe the estimates that each dollar spent on EHR will return three in efficiencies is probably conservative, BUT this depends on the quality of implementation. Key to this is user interface design: if EHR makes doctors, nurses and other caregivers more efficient, they will adopt it as fast as it can be provided. If it makes them less efficient, it may fail. As I write this, a combination of structured record keeping, discrete and continuous speech recognition, and haptic data entry looks most promising to me for the user interface, and this combination is now compatible with nearly all common hardware platforms (pad/tablet/smartphone/e-reader, notebook/laptop, desktop, workstation) except for Intel netbooks, which are not, at present, speech capable.

In addition, I strongly believe that EHR in order to succeed must promote use of best practices in healthcare wherever these can be identified (and be flexible enough to incorporate new best practices as they are later identified.

The exact hardware and software architecture of EHR will be unimportant as long as data interchange standards are well defined, well understood, and carefully adhered to. Though I have somewhat of a bias towards open source software solutions (due to the inherently vigorous Darwinian mechanism of software quality assurance, as well as the 'crowd sourcing' approach to design), I acknowledge that no single approach will get all of EHR correct the first time (or the second, third or fourth for that matter).

--Stuart A. Jones, MD

Dr. Jones, thank you very much for your very insightful comments. You are correct, I believe, that successful EHR implementations will increase the value of interventions, and that once physicians, nurses, and other clinicians find enhanced value in using EHRs, they will succeed. I also agree that a key element in the ultimate success of EHRs will be promoting the use of best practices, with flexibility.
Sometimes, studies can be of limited use in the broader understanding of a complex issue or set of issues. That seems to be the main point that Dr. Mostashari was trying to make in addressing the Health Affairs article.
Thank you again for your extremely thoughtful comments, Dr. Jones!
--Mark Hagland, Editor-in-Chief