Primary Care—Leading the HIT Revolution, Part I: Supporting Clinical Processes

No, the title is not a mistake. It simply reflects what I have come to realize over the last month or so. Innovation is about solving problems and, as the saying goes, “Necessity is the mother of invention.”   Misery is a source of innovation, and the loudest cries about EHR systems with poor usability have come from primary care providers and their professional organizations.   Having practiced primary care internal medicine, I understand the misery well. Primary care clinicians have been cast in the role of gatekeepers and monitors of their patients’ health, but have been given tools that mostly provide them data. The fact is clinicians are drowning in data because they do not have the tools required to make proper use of it. Here are a few clinical process examples.

Results management
When results appear on a daily basis, few systems provide a means of managing those results, tracking them over time, or stratifying them by patient type or required follow-up.   EHR systems provide test results, but lack features that make asking questions—searching, sorting, tagging—easy (1,2,3).

Population management
Many primary care practices have both registry software and an EHR. EHR systems with built-in registries usually offer fewer features than standalone products because those registries are after-the-fact add-ons to what is essentially an electronic paper chart (1,2,3).

Care Coordination
Communication is one area where current EHR systems are much improved. Even so, many EHRs have only basic messaging capability. Managing messages and tying them to specific patient interventions and interactions are features rarely offered in typical EHR systems (1, 2, 3).

Patient Engagement
Patient portals have limited functionality, and having experienced firsthand a major vendor’s — NO. Just, no. After using the portal, I have found that I still have to make a phone call. The idea that patient outcomes surveys could be tied to care plans and managed through current portal offerings, a CPC+ Track 2 requirement, is close to science fiction.

The above clinical process categories are essential to primary care, and any electronic system that purports to help clinicians in their daily work must offer robust support for each of them.   Face-to-face patient encounters constitute only part of the clinical work expenditure required to deliver quality primary care. Yet, current EHR systems are designed primarily with direct encounters in mind.   Clinical process management features, when present, are after thoughts.

Interoperability is important, but so is clinical process management.   While interoperability has many champions, processes are being lost amid the cries for interoperability. If FHIR works perfectly, it will not solve the process support problem.   Moving more data from place to place without better process management tools will simply worsen the lot of already harried primary care clinicians.

Lately, user-centered design has gained a higher profile. Unfortunately, the solutions offered via usability testing and UCD rely on data-centric systems. Current EHR systems do not sport architectures that acknowledge clinical processes; rather, they focus on patient data (see Is the Electronic Health Record Defunct? ). Usability/UCD cannot overcome architectures that are process-agnostic.

Along with research on usability and UCD, clinical workflow research has caught on. Both NIST and AHRQ have funded clinical workflow research (see NIST and AHRQ Workflow Reports: A Few Observations). While these reports offer interesting information, they assume software workflows are static and hardcoded. In other words, they assume that EHR systems will have built-in workflows and that clinicians must necessarily adapt to the software.   The problem with this way of thinking (and with applying usability/UCD approaches to systems with static workflows) is that there is no such thing as an average or standard primary care workflows (4, 5, 6), whether doing face-to-face encounters or when doing results management. If there is no such thing as a standard workflow, then any system that has static, hardcoded workflows will not be suitable for all users. No, the key is providing practices with systems that have flexible workflows that are explicit and adjustable by local users.

To the above recounting of EHR issues primary care clinicians must deal with, I could add alert fatigue and a few words about medical scribes—all of which point to the need for a new approach to clinical care system design. Decision support cannot work properly if the person entering the data is not making the decisions.

So, you may be asking, exactly how is primary care leading the HIT revolution? Some of the best—no, make that THE best–clinical software design guidance I have seen has come from researchers looking at primary care processes and patient-centered medical homes. This research both acknowledges the array of patient care activities (within and outside of the direct encounter) that primary care clinicians engage in AND, this is critical, ties those actions back to specific features (or lack thereof) in current HIT offerings. What may seem like complaining to some sounds to me like: “Here is what is making me miserable; care to provide a solution?” The one great thing about MU is that it drove EHR adoption to the point where clinicians had to become active in HIT design or suffer. Happily, they have chosen to get involved!

EHR functionality issues have been taken up by professional organizations representing primary care clinicians (1, 3).   These documents are essentially feature requests, and they are surprisingly detailed in what they ask for and in the explanations offered for why the features are needed. Before MU, one would be hard-pressed to get clinicians to offer such detailed guidance on something as esoteric as software design—but here we have it. Adding to this wealth of design input is the increasing body of research focused on results management (see Diagnostic Error, Results Management, and Software Design) and care coordination (see Care Coordination and Clinical Care Systems: A Look at Clinical Work Support Needs).   With this research, we finally have actionable design information that can be used to build clinical care systems that support primary care in all of its glory!

But that’s not all… For the first time ever, the technology needed to render these feature requests into real systems exists!   FHIR is promising; high-speed networking is widely-available; cloud technology has matured significantly; relational databases have been joined by more flexible data stores, and the first generation of mobile platforms suitable for clinical use appeared last year.  The clinical care systems envisioned by the consensus report is finally possible.   The only question is who will be the first to build one. The time has come for the EHR to be replaced by a true clinical care system.

The next post in this series will take a closer look at three articles (Krist, O’Malley, Sinsky) that focus on EHR functionality with the goal of discovering common themes regarding features, clinical processes, and potential software requirements. Until next time…

  1. Krist AH, Beasley JW, Crosson JC, Kibbe DC, et al.Electronic health record functionality needed to better support primary care. J Am Med Inform Assoc. 2014 Sep-Oct;21(5):764-71.
  1. O’Malley AS, Draper K, Gourevitch R, Cross DA, Scholle SH. Electronic health records and support for primary care teamwork. J Am Med Inform Assoc. 2015 Mar;22(2):426-34.
  1. Sinsky CA, Beasley JW, Simmons GE, Baron RJ.Electronic health records: design, implementation, and policy for higher-value primary care. Ann Intern Med. 2014 May 20;160(10):727-8.
  1. Holman GT, Beasley JW, Karsh BT, Stone JA, Smith PD, Wetterneck TB. The mythof standardized workflow in primary care. J Am Med Inform Assoc. 2016 Jan;23(1):29-37.
  1. Militello LG, Arbuckle NB, Saleem JJ, Patterson E, Flanagan M, Haggstrom D, Doebbeling BN. Sources of variation in primary care clinical workflow: Implications for the design of cognitive support.Health Informatics J. 2014 Mar;20(1):35-49.
  1. Friedman A, Crosson JC, Howard J, Clark EC, Pellerano M, Karsh BT, Crabtree B, Jaén CR, Cohen DJ. A typology of electronic health record workarounds insmall-to-medium size primary care practices. J Am Med Inform Assoc. 2014 Feb;21(e1):e78-83.

Suggested Reading
Clinical Process Management – The Patient in the Age of Clinical Processes
Clinical Process Management, Next Steps
Clinical Software Design Principles



  1. there was an interesting article in HBR on how IoT needs design (and a focus on user experience) not just technology.

    I have a sense that this is what EHR researchers are beginning to realize too.

    Interestingly a recently developed system to support CTAS had as its primary – almost exclusive – design goal user experience. Users of CTAS (triage nurses) don’t have the time to work around a messy process or unfriendly set of input screens and simply won’t use a bad system.

    1. Author

      Hi Edwin, thanks for your comment! I agree, design is an important element. Design encompasses at least two components: appearance and process. Good systems match the processes of the user and good interfaces make optimal use easier and faster to attain.

Leave a Reply

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