Ann S. O’Malley, M.D., M.P.H., a senior researcher at the Washington, D.C.-based Center for Studying Health System Change, on March 24, published an article in the New England Journal of Medicine in which she discussed some of what she called the “daunting task” of fully leveraging information technology to improve healthcare delivery over time. In the article, titled “Tapping the Unmet Potential of Health Information Technology,” O’Malley wrote that “Although EHRs laudably provide immediate access to patient data and electronic messaging functions, clinicians have been frustrated by the difficulty of using them to support care delivery and coordination. Transforming EHRs [electronic health records] into effective clinical tools rather than a means of capturing information primarily for documentation and billing purposes will require progress on multiple fronts,” including moving forward strategically to coordinate and integrate care, particularly for patients with multiple chronic conditions,” she argued in the NEJM article.
Most importantly, O’Malley stated, “HIT, especially if widely implemented, can facilitate coordination by making information electronically at the point of care. As clinical care processes become more effective and efficient, they can inform new HIT cap abilities that will better support coordination,” including, for example, developing consistent notification processes around communication about care transitions, she added. The conclusions she shared with the NEJM’s readers were in turn based on research she and her colleagues had done earlier that she highlighted in an article in the Journal of Internal Medicine.
O’Malley spoke recently to HCI Editor-in-Chief Mark Hagland regarding her NEJM Perspective article. Below are excerpts from that interview.
You wrote in the NEJM article back in March about one of the core problems others have identified as well, which is the current lack of interoperability impeding the potential for better coordinated care. Can you speak a bit about that?
Yes, and just the core design of EHRs is itself is an inhibitor. They’re designed right now to help clinicians document what goes on in the clinical encounter, but of course, there are a lot of requirements for documentation, for billing purposes. And until we realign payment incentives around doing what’s right for the patient, providers will probably not demand the types of EHR functionalities that are necessary to do the kinds of tasks necessary to coordinate patient care. EHRs could provide great tools if they provided, for example, better tools for tracking referrals between primary care physicians and specialists; if they were more interoperable, obviously—because right now, one EMR from one vendor doesn’t talk to an EMR from another.
Obviously, the Office of the National Coordinator for Health IT is working hard on this. And advances are taking place through HIE [health information exchange] development, and even shared servers, to standardize data so that it can be shared from one place to another.
Ann S. O'Malley, M.D., M.P.H.
What is your perspective on the development of accountable care organizations, and their care coordination potential?
I think, in the sense that they try to realign payment incentives around quality of care for the patient, and try to shift them away from volume-based fee-for-service incentives, they have strong potential.
And won’t ACO development spur forward advances in interoperability and EHR development?
Yes, that’s the hope. And if they can revise payments towards greater coordination of care, then the hope is certainly that they’ll help spur advances in EHR interoperability and improved functionality.
What kinds of comments did physicians give you in your original research?
First of all, there’s a lot of repetition, because we now have these electronic text pieces in the record, so there’s a lot of temptation to cut and paste a lot of pieces of text, so while you used to have very concrete pieces on paper, now we have 10 pages of output and redundancy, and it makes it very difficult for the physician on the receiving end of the note to separate the wheat from the chaff, and to get to the core of what you need to do for the patient. And that’s a human frailty; people are busy and they’re getting lazy.
The fact that the note is legible, etc., is a huge advance. But the tension between free text and templates—what is the patient’s chief complaint, and what is the history of the present illness? Those are two elements you don’t want in templates, those elements need to be free-text-entered to capture the narrative of the patient. But there’s a lot of disagreement about what things to template and what things to free-text.
What are your core recommendations to CIOs, CMIOs, and vendors, based on what you’ve learned in your research?
My core recommendation is to work with clinicians to develop care processes in the clinical setting that help improve the coordination of care for patients, and to do that in tandem with how they use electronic health records, and how they design functionalities for EHRs—the two have to happen side by side. You can’t just plop an EHR down among clinicians and expect them to immediately change behaviors. At the same time, clinical processes have to evolve as we develop EHRs, and will be important to help achieve some higher-function tasks, such as population-based management.
Will you continue research in this area?
My research focuses on primary care and tools to support it. So, to the extent that HIT supports those processes, I probably will.