Bad Research Shouldnt Affect Good Policy
Professor Stephen Soumerai of Harvard Medical School and Professor Ross Koppel of the University of Pennsylvania have authored or co-authored multiple articles questioning the functionality, safety and efficiency of electronic health records.
Their voices need to be heard, especially in the era of meaningful use, paid for by your government with borrowed money. But I fear a new opinion piece they published in the Wall Street Journal does a disservice to the industry, policymakers and themselves.
They contend the EHR meaningful use program in the stimulus law was the work of lobbyists promising medical cost savings of up to $100 billion a year, and note that many in the industry are skeptical of such savings. That’s a fair point.
But Soumerai and Koppel then go a step further in the WSJ piece. “An extensive new study indicates that the caution is justified: The savings turn out to be chimerical.” They go on to explain that the study confirmed what many researchers suspected: “The savings claimed by government agencies and vendors of health I.T. are little more than hype.”
But the study is flawed, hopelessly flawed, and two distinguished researchers at two major universities should have known that, if they were being objective researchers to any degree. The study first sifted through about 36,000 studies of health I.T. over the past four or five decades. Let’s flag one word: “decades”. Researchers then identified 31 studies out of the sample of 36,000 “that specifically examined the outcomes in light of the technology’s cost-savings claims.”
When were the 31 studies conducted; how many of them came after the stimulus? The study doesn’t say and neither to Soumerai and Koppel. And since the meaningful use program is less than two years old, it’s too early for a “meaningful” study of the impact of meaningful use. Nonetheless, Soumerai and Koppel contend that study authors found no evidence--leading as far back as 50 years--that health I.T. reduces overall costs.
How many providers had an EHR in 1962? How about 1972? How about 2002? Say what you will about the merits of programs funded by stimulus dollars, but how much longer was the nation going to wait until its largest industry--and the one slurping up personal and tax dollars at an unsustainable rate --was going to automate? Forced automation with a financial carrot is what the meaningful use incentive program is all about.
Having given no evidence of the meaningful use program being a failure less than two years after inception, Soumerai and Koppel in the WSJ article make another claim that simply can’t objectively stand up at this point in time. “It is already common knowledge in the health care industry that a central component of the proposed health I.T. system--the ability to share patients’ health records among doctors, hospitals and labs--has largely failed. The industry could not agree on data standards--for instance, on how to record blood pressure or list patients’ problems.”
Do they understand anything about the meaningful use program? Do they understand that Stage 1 is designed to help providers understand and use the functions of an EHR, including accessing vital signs and problems lists? Do they understand that Stage 2 has a strong focus on health information exchange that will only get bigger in Stage 3, when metadata could be among what is exchanged?
Health information exchange has been tried to some degree for at least a couple decades and never took off until meaningful use. Look at the HIE activity since 2009 and especially during the past 18 months, still in the early stages, but with tons of work being done.
Look at this list of meaningful HIE initiatives or achievements announced just during the first couple weeks of September, 2012: ONC takes a softer approach to HIE, HL7 to offer free licenses for its standards; Epic will use Surescripts’ network to aid HIE; federal agencies are testing transmission of metadata; eClinicalWorks launches a national network for clinicians; ONC unveils free online privacy/security training games; more rural providers get grants for broadband connectivity; and Availity and Medecision integrate their products to facilitate transitions of care.
If this isn’t proof that the goals of meaningful use already are starting to bear real fruit, then what is?
Are today’s EHRs optimally functional? Have the expected cost benefits and improved quality of care arrived? No. Is it way, way too early to make the claims that Soumerai and Koppel make? Yes. Should they know better? Yes.