AMA Call To Action On Health Records Should Tell Doctors To Heal Themselves

Andy OramThe American Medical Association (AMA) is one of the most powerful and well-known institutions in this country. Opposition from the AMA helped to bury hopes for universal healthcare back in the Harry Truman presidency, and now the AMA maintains a stranglehold on Current Procedural Terminology (CPT) codes and therefore on any innovation in reporting medical services. So when the AMA puts out a press release titled AMA Calls for Design Overhaul of Electronic Health Records to Improve Usability, describing the serious usability problems of EHRs, and announcing the release of their “solution” white paper titled Improving Care: Priorities to Improve Electronic Health Record Usability, headlines get made and policy-makers start to stir. Can a snap of the fingers by the AMA bring the EHR industry in line?

I'd love it for the AMA to contact mobile phone manufacturers and tell them their voice recognition systems should respond to all my commands and their apps should pop up at useful times without my pressing on them. These enhancements lie beyond currently known technologies and the companies are working hard on them already, but for the AMA to weigh in couldn't hurt. And the AMA could tell my grocer to stock all the items I like, along with clear definitions of ingredients. The markings on the packages are outside the grocer's control, but the grocer should at least understand that these are good ideas.

You may have gotten the message by now that I don't find much reason for the 12-page call for action by the AMA on electronic records. The sentiments expressed are fine, but the question is how, when, and by whom they can be carried out.

Let's start with recommendation 7, Facilitate Digital and Mobile Patient Engagement. True, most EHRs are not designed to accept sensor data from patients--but that's the least of the barriers to data use. Doctors don't know what to do with the stuff even if patients give it to them. Will the AMA sponsor CME and other initiatives to spread the practices developed by Geisinger, Kaiser, and a few other institutions using patient-generated data?

Recommendation 3, Promote Care Coordination, is also more a call to change the field of medicine than a request for enhanced EHRs. It envisions doctors who can longitudinally view a patient's entire "continuum of care," which passes through multiple institutions.

Recommendation 6, Promote Data Liquidity, is welcome because it puts the AMA on record opposing the habit of withholding information on patients to make it hard for them to get care elsewhere. With initiatives such as FHIR and Blue Button, interoperability of records is coming closer to fruition--will health providers allow those fruits to fall to the ground and rot?

The AMA also tries to shift responsibility in recommendation 5, Reduce Cognitive Workload, which contains this bit of wishful thinking: "EHRs should support medical decision-making by providing concise, context-sensitive and real-time data uncluttered by extraneous information. EHRs should manage information flow and adjust for context, environment and user preferences." (p. 6)

What the AMA is asking is probably the most difficult research area in computing: knowing what to place before the user, and how to move the user expeditiously through steps that can very widely from one use case to another. Although they claim to recognize "that many of the recommendations can only be implemented in the long term," they give no indication that they understand the tough job of making systems that anticipate user information needs and actions.

EHR vendors can't do it on their own. They can start the process by offering hooks that clinicians can use to customize the product (Recommendation 4, Offer Product Modularity and Configurability) and by incorporating user feedback (Recommendation 8, Expedite User Input into Product Design and Post-Implementation Feedback). But then each physician must determine his or her workflow and create a system that meets those needs. Customization goes far beyond "reminders and alerts".

As Roger A. Maduro from Open Health News explains in his comment to the AMA press release, the flexibility requested by the AMA requires open source EHRs. Even with open source, delicate decisions need to be made for customization to become real and useful. Programmers must balance the simplicity of coding in immediate fixes with the benefits of finding more general solutions that adapt well over time. And they will want to avoid making so many changes that they create an actual fork of the original project, because that leads to a second independent code based and doubles the maintenance effort.

Maduro's sentiments were echoed by Dr. Seong K, Mun, President and CEO of the Open Source Electronic Health Record Alliance (OSEHRA), who said that  “AMA’s call for action on EHR usability is long overdue,” pointing out that.  “Many clinicians are burdened by data entry clerical activities as EHR’s evolved from legacy systems designed for documentation, not patient management. The optimal solution to this problem is to adopt an open source approach based on transparent collaboration and agile development.”

The AMA call for action shows the organization joining health care reformers in support of robust usability criteria for EHRs, patient engagement, interoperable records, APIs for third-party innovation, and better coordinated care. Doctors can do much more than the document suggests to bring these things to pass.