Where Is the $81 Billion?

News flash: EHRs don’t save money. Surprised? Well, it’s big news. In a New York Times (NYT) article last week, Reed Abelson and Julie Creswell focus on a Health Affairs (HA) article by Arthur Kellermann and Spencer Jones from RAND entitled, “What It Will Take To Achieve The As-Yet-Unfulfilled Promises of Health Information Technology.” The HA article is a retrospective look at a 2005 article by Richard Hillestad et al (also from RAND) which concluded that rapid adoption of health IT could save the United States more than $81 billion annually.

The NYT reporters suggest that some of the exuberance for health IT was stimulated by the over-stated potential economic benefit of health IT. Abelson and Creswell make a point that the 2005 article was sponsored by industry. The disclosure at the end of the HA article states that it was sponsored by Cerner, General Electric, Hewlett-Packard, Johnson and Johnson, and Xerox — and guided by an independent Steering Committee.

With that as background, here are some key points from the NYT article:

  • The updated RAND report concludes that the $81B a year figure was overstated.
  • Evidence to date for savings from the implementation of EHRs is “scant.”
  • EHRs may actually add to costs by making it easier to bill for more services (see AmericanEHR blog Oct 3, 2012).
  • Currently available EHRs are not as usable as they should or could be.
  • The 2005 RAND report helped industry sell new systems, “… despite criticism at the time that the analysis was too rosy.”

Kellermann and Spencer do a nice job reviewing the the 2005 report and, while acknowledging that, “Seven years later, critics of the RAND team’s analysis can claim a measure of vindication,” they identify some reasons why the United States hasn’t seen the anticipated financial savings. According to the authors, the failure to achieve results is…

  • “… not due its lack of potential but to shortcomings in the design and implementation of health IT systems.”
  • Related to slow adoption.
  • A result of clinicians’ reluctance to invest the time and effort to learn “difficult-to-use technology.”
  • Due the failure of of health care systems to modify processes.

In addition, Kellermann reviews the assumptions made in the 2005 article, including:

  • The presence of interoperability
  • A 90% penetration of EHR use
  • Usable EHRs
  • Changes in payment to shift the focus away from volume

None of these assumptions have been met since 2005. They conclude that until the preconditions of, “Fully interoperable, patient-centered, and easy-to-use systems…” are in place, and providers re-engineer processes of care, and payment models change, it is unlikely that savings will be realized from health IT. Interestingly, while the authors refer to the federal incentives for EHR adoption, they do not specifically address the positive (or negative) consequences of Stages 1 and 2 of Meaningful Use — or consider the potential for Stage 3, currently slated for 2016, to advance the United States closer to the point when we might expect some savings.

Regarding Meaningful Use, most of AmericanEHR visitors would likely agree that the push towards certifying products based on desired features/functionality, and the incentives for clinicians to capture information in structured format so that data can be re-purposed (e.g., for clinical decision support systems, population management, and quality improvement), are important features of the federal incentive program. But could the EHR Incentive Program be improved and move the U.S. even more quickly towards the time when significant savings could be achieved? Since the authors don’t address whether the current and proposed Meaningful Use objectives and metrics are helpful (or are a hindrance), consider the following questions:

  • Are the requirements for certification of EHR technology resulting in less complex and easier to use EHRs?
  • Are vendors able to address the pace and number of certification requirements?
  • Is the MU program committing vendors to specific deliverables in lieu of addressing more fundamental usability issues?
  • Are the increasing number of processes required to meet Stage 2 and proposed for Stage 3 allowing practices and hospitals the flexibility to innovate around better care delivery or are they just focusing on achieving the MU metrics?
  • Is it possible that some Meaningful Use measures are locking in potentially inefficient work flows?
  • Should the certification process be on a different time line than the introduction of new features/functions into practice in order to allow testing and validation of new measures?

An incredible amount of time, energy, and money (public and private) has gone into moving the state of health IT in the U.S. to our current status, imperfect as it may be. It is too early to expect that simply (and rapidly) implementing certified EHR technology will result in better care at lower cost. Products have a long way to go. Practices and health systems must also complete an arduous journey to revise care delivery processes so that they can leverage the full capabilities of health IT. Expectations are high — especially given the federal investment and the economic challenges facing the United States. Impatience is understandable. The calculations of the 2005 study were based on assumptions that have not been achieved — yet. Rather than question the whether the EHR Incentive Program is achieving its goals, we would be better served by:

  • Acknowledging the complexity of this long term endeavor.
  • Evaluating, based on existing evidence, the positive and potentially negative consequences of the EHR Incentive Program.
  • Adjusting the goals and measures for Stage 3 (and beyond) based on the evaluation.
  • Focusing on attaining the important health system goals articulated in the 2005 Hillestad article.

What do you think about the NYT article and HA articles? What can/should be done to improve the ability of EHRs to support better care and reduced costs?

Remember, today (11:59 pm ET) is the deadline for submitting comments to the Health Information Technology Policy Committee on the draft Stage 3 measures.

This post is the personal opinion of the author and does not necessarily reflect the official policy or position of the American College of Physicians (ACP).


2 responses to "Where Is the $81 Billion?"
  • January 15, 2013
    Arvind Cavale

    It is safe to say that the stated objective of reducing costs will never be realized, mainly because all involved with this process are viewing the process upside down.

    For most successful industries/businesses that integrate IT into their processes, value and ROI is achieved by engineering IT to suit their process. These businesses adopt IT based on their own requirements and implement IT based on their “meaningful use”, not to satisfy some 3rd party’s mandate. Successful businesses spend on IT because they can pass it on to their customers, rather than wait for the government’s gravy train to arrive. Finally, in most service-oriented businesses (except health care) quality and value of service is determined by the user of service, not some uninterested 3rd party.

    If we need real answers, we must ask the correct questions. Do we have the courage to do that? I doubt that.

  • January 18, 2013
    Mike Soppet

    Not surprised. Did not know the original Rand Study was sponsored by Industry however. The apparent windfall of profits to Certified EHR systems has led to Meaningful Use functionalities being cobbled on to (at least according to our vendor’s representative trainer)very old EHR software. Thus, even within the software, entering discrete data in one field does not automatically satisfy that data point in all fields within the EHR. This is basic stuff and the fact that an EHR has been able to achieve “certification” without this basic functionality and then receive the pass-through Meaningful use money from the government will further delay our ability to achieve meaningful use for anyone. Obviously just one physician’s opinion.

Leave a Reply

Your email address will not be published. Required fields are marked *