Overcoming The Penguin Problem: Setting Expectations for EHR Adoption










Economists call it “The Penguin Problem”  — No one moves unless everyone moves, so no one moves. 

The role of user expectations is crucial in getting penguins to move off of ice floes and in the successful adoption of new network technologies.  I’ll cover two main points in today’s essay:

  • How “The Penguin Problem” Helps Explain Low EHR (electronic health record) Adoption To-Date
  • How Recent Federal Actions Are Setting Higher Expectations for EHR Adoption

The Penguin Problem and Low EHR Adoption To-Date

While not the only factor, the role of user expectations is a crucial element in explaining the adoption of new network technologies. Harvard Business School Professor Tom Eisenmann explains:


In new markets, uncertainty about how technology, business models, and demand will evolve is often high. With fragmented demand—that is, with a very large number of users—it can be difficult for prospective users to communicate their expectations regarding the long-term outlook for a network and coordinate their behavior.

Paradoxically, a network with a fragmented base of potential users may stall, even if network effects are strong. I might be willing to pay for network access if I could be assured that many others would, too. Other parties may be in the same position—perhaps enough of them to satisfy mutual requirements for network scale, if everyone actually participated. Due to fragmentation, however, individuals cannot signal their intentions, so they cannot be assured that others will join the network. Facing this uncertainty, each isolated user may defer their purchase. No one moves unless everyone moves, so no one moves.

Economists Joseph Farrell and Garth Saloner labeled this scenario “excess inertia,” and more colloquially, the “penguin problem.” Hungry penguins gather at the edge of an ice floe, reluctant to dive into the water. There is food in the water, but a killer whale might be lurking, so no penguin wants to dive first. In such circumstances, individual rationality may lead a group to forfeit attractive opportunities, for example, a predator-free meal or an innovative new networked product. [Eisenmann, T. “Platform Mediated Networks: Definitions and Core Concepts”, Harvard Business School Module Note; October 2007]

How does this relate to the adoption of EHR technology?  The role of user expectations is crucial.  Prior to 2009, the U.S. had a major penguin problem — many physicians and hospitals have been reluctant to be first movers because they have doubted others will be following.  Much of the value of EHRs is dependent on achieving network effects —  the creation of a widely adopted network that allows for exchange of interoperable data and collaborative care management processes.

Recent Federal Actions Are Setting Higher Expectations for EHR Adoption

Recent Federal Government actions are setting the stage to overcome the penguin problem. The government is doing a good job — a very good job — in setting firm expectations for the adoption and meaningful use of EHR technology.

The passage of HITECH Act legislation comes with $34 B of EHR adoption incentives (carrots) through 2015, with the promise of reimbursement penalties (sticks) for non-adopters after 2015.

Agree or disagree — this is a major shift of Federal policy.  As opposed to the Bush philosophy of letting the free market address the challenge, what we’re seeing is the expression of Obama policy with strong intent to guide market development with a firm hand.  Personally, while my economics tend toward free market, HIT policy is one area where laissez faire hasn’t and isn’t going to work.

More recently, the recommendations of two workgroups of the Federal HIT Policy Committee in June and July turbocharge the changing Federal policy in HIT.  The recommendations are powerful — and address the two most significant to-be-defined terms in the HITECH Act — 1) what constitutes “meaningful use” of EHRs, and 2) Certification criteria and process for EHRs.

Meaningful Use.  In July and July the Meaningful Use Workgroup presented 1st and 2nd drafts of recommended criteria.  Physicians and hospitals who wish to receive funds subsidizing EHR purchases will have to demonstrate achievement of “meaningful use” criteria.  The Workgroup recommendations were approved by the HIT Policy Committee, and have been handed to HHS (Health and Human Services) to draft specific regulations by the end of 2009.

Some of the recommendations of the Meaningful Use Workgroup are exceptionally powerful in laying groundwork for network effects, and I’d like to highlight a few of these. These examples are from the latest Meaningful Use Matrix [MU Matrix] and Meaningful Use Workgroup Recommendations [MU Recommendations] for Eligible Providers (Physicians).   Note that some of the most significant criteria were added or moved up in timing in the July version of the recommendations:

  • Provide patients with timely electronic access to their health information by 2011 [Meaningful Use Matrix, p. 5–6; added in July]
  • Provide clinical summaries for patients for each encounter by 2011 [Meaningful Use Matrix, p. 6]
  • Capability to exchange key clinical information among providers of care and patient authorized entities electronically by 2011 [Meaningful Use Matrix, p. 5; added in July]
  • Access for all patients to PHR populated in real time with health data by 2013 [Meaningful Use Matrix, p. 5; moved up from 2015]
  • Upload data from home monitoring device by 2013 [Meaningful Use Matrix, p. 6]
  • Access comprehensive patient data from all available sources by 2015 [Meaningful Use Matrix, p. 7].  In the MU Workgroup Recommendations, this is described as “include required participation in nationwide HIE” [MU Recommendations, slide 18]

Certification. A rigid, complex and costly EHR certification process would be a barrier to adoption of new,  innovative EHR technologies, e.g., clinical groupware. The Certification/Adoption Workgroup Recommendations are very helpful in leveling the playing field between existing EMR technologies and new, innovative EHR technologies.

Here are some of the key recommendations:

  • Certification should support achievement of “meaningful use” objectives
  • The process of defining  Certification criteria should be performed by ONC, and separated from organizations that perform certification testing.. 
  • Multiple organizations should be allowed to perform Certification testing and provide certification
  • There should be multiple routes to certification, including certification of modular components of EHRs

Closing Thoughts


Will the recommended definitions of “meaningful use” and “certification” work toward overcoming the penguin problem of EHR adoption? I’m very optimistic.


But I’m also realistic — it might not. While I applaud the Government setting a high bar, it’s possible that the bar has been set too high. All the pushing being done by the Federal government might not be enough to get the penguins off the ice floe or to get doctors and hospitals to adopt EHRs. We need to monitor this closely.


Whether or not you share my optimism, the shift in Federal HIT policy is dramatic…and is a strong attempt to set expectations toward ending the EHR penguin problem.

8 thoughts on “Overcoming The Penguin Problem: Setting Expectations for EHR Adoption

  1. Pingback: Practice Fusion
  2. Pingback: arthurwlane
  3. The penguins are all on the ice because last time they dived in they got eaten and chased by killer whales aka EMR vendors. Even physicians can learn from pain. The truth is EMR’s are not new and have been around for 30 years or more. Most physicians have some experience with them. Most of the experience is bad for the simple reason they were not designed to aid in the care of patients. Until they are and user interface issues are adequately addressed, the piguins will rightfully stay on the ice.

  4. Bob T.

    You’re absolutely right…Difficult to use EMR 1.0 software is another factor that has limited adoption.

    I’m not suggesting that the penguin problem is the only factor to explain current low EMR adoption.

    I am suggesting that overcoming the penguin problem (i.e., the Govt. resetting expectations through HITECH) can be a strong contributing factor in future adoption.

  5. Pingback: racep
  6. Pingback: Bart Collet
  7. Your insight into this issue is spot on.
    At the heart of improving adoption of EHRs is people & process, not necessarily technology. It’s been my experience that some people are simply not ready for innovation and need time to adjust.
    Then there’s the process. In an ambulatory setting, process is everything. At the center of that process is a paper chart.
    Case in point – I’ve observed seasoned healthcare professionals shaking uncontrollably with fear and anger in their eyes when I tell them that I’m going to help replace their paper chart with an EHR.

    What they all seem realize at that very moment is that their entire world is about to change. They feel that everything they’ve accomplished in their career, the knowledge, the seniority, the respect, is now in jeopardy because it’s all predicated on them knowing the process of health care documentation and delivery, both of which dramatically, and even physically, change with and EHR.

    After eight years and 200+ implementations I’ve had only one clinical staff member quit on me during a project. And it’s something I will never forget. A nurse with 30 years in the industry simply got up from the table, walked out and never came back to work.

    My point is the EHR adoption issue always seems to be related to people & process, not necessarily the technology.

    Change leadership is needed.

  8. Pingback: Nina M Housel

Comments are closed.