Building interfaces for self-assessment and feedback in the EMR Andrew L. Yina,b, Inna Wanyin Linb and Pargol Gheissarib a Weill Cornell Medical College, 1300 York Avenue, New York, NY, USA b Cornell Tech, 2 W Loop Rd, New York, NY, USA Abstract Electronic medical records (EMRs) have played an increasingly significant role in healthcare. With these major advances, however, have come major pitfalls like the increase in physician burnout and stress. This position describes interventions that can help to address these pitfalls by supporting the desires and interests of the clinicians using the EMR. It presents three augmentations related to self-assessment and feedback to attempt to address these needs: a retrospective dashboard, a collaboration tool, and a research and note taking interface. Keywords 1 Self-assessment, feedback, self-learning, electronic medical record, computer-supported cooperative work, interfaces 1. Introduction clinicians appear to be uniquely burdened – receiving more messages, spending more time Since their widespread adoption, there has been after hours, and spending more time on clinical increasing understanding that electronic activities like note writing, ordering medical records (EMRs) play a large role in medications or tests, and reviewing patient physician stress and burnout. Physicians find charts [6]. With this said and all the themselves working more from home after complexities considered, there continue to be hours, answering more messages and emails both technical and political advancements that now sent and received at any time, and encourage and support growth of tools aimed at suffering through both major and minor improving the EMR for users [2,7]. usability issues [1,4,8]. Clinicians and This position will argue that developing and technologists alike cringe when reading Atul improving tools and interfaces to facilitate Gawande’s Why Doctor’s Hate Their hospital clinician self-assessment and feedback Computers or Schulte and Fry’s Death By can be one component to help address inpatient 1,000 Clicks – disturbed by the time demand of clinician burnout related to the EMR. Many the EMR, the legal-political complexities, the innovations focus on reducing the time using introductions of new errors, and the impersonal the EMR but few focus on increasing the feeling that the whole journey has caused personal value obtained from using the EMR. [5,10]. These systems appear to do everything Adjusting interfaces to include tools that to pull clinicians away from the work that they provide physicians the ability to holistically find most meaningful, a key feature in learn and grow could reduce the frustration and determining the likelihood of burnout [11]. One burnout associated with using the EMR, might reasonably think that this problem must improve patient care, and provide a feeling of be a global issue, impacting all health systems growth to each clinician. We propose features relying heavily on EMRs. Unfortunately, US for adoption and discuss how they fit into key Joint Proceedings of the ACM IUI 2021 Workshops, April 13-17, 2021, College Station, USA EMAIL: aly27@cornell.edu (A. 1); wl676@cornell.edu (A. 2); pg463@cornell.edu (A. 3) ORCID: 0000-0002-9560-0169 (A. 1); 0000-0003-2928-0096 (A. 2); 0000-0003-0048-559X (A. 3) ©️ 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). components of computer-supported individual cases as they deem necessary. With cooperative work (CSCW), improving the the added information, they could now have a alignment of the EMR with incentive way to roughly sense whether the changes they structures, workflow, and awareness [9]. make to their practice are making a difference CSCW has long had a role in shaping EMR for their patients, comparing themselves from development but the interplay between policy year to year or identifying trends or themes in makers, EMR vendors, hospitals, and care their patient population. Clinicians are already teams continues to make implementation spending large amounts of time finding ways to complex [3]. These suggested features include tally portions of this information themselves. ways to retrospectively present previous patient An interface providing this information in a information, improvements to collaborative readable and concise format will significantly tools, and tools for learning and research. Such improve the efficiency in their workflow and features could be integrated into the current allow them to spend time thinking of solutions workflow of clinicians and could help make the rather than merely tallying data [12]. Given the EMR a better tool for clinicians on the whole. amount and the scope of clinical data in the EMR, building such interfaces with existing 2. Discussion data could be an easy, user-focused implementation with tremendous value-add. 2.1 Improving Review of Previous Patient Information 2.2 Rekindling peer to peer, genuine collaborations and connections EMRs are primarily designed for real-time between clinicians clinical care as opposed to retrospective use. However, clinicians spend significant time, Clinicians have a common practice of often more than they wish, looking up previous discussing challenging or tricky cases with one patients and assessing themselves based on another, relying on friends or close colleagues what they find [12]. This aligns with both a in this process. At the same time, clinicians feel clinician’s personal incentive and interest to relatively isolated in a lot of their work and feel improve on their work as well as the hospital’s like they are left on their own to manage interest to improve the quality of care [9]. As patients that may benefit from multiple simple as this may seem, providing clinicians perspectives and opinions [12]. The EMR has with the opportunity to see what happens to the opportunity to build on this collaborative their patients after caring for them and get a nature of medicine. The current workings of sense of how their patients are doing EMR inboxes/emails have created a level of collectively is a function that is unavailable to noise and message fatigue that sometimes most clinicians. Simple information like the discourages genuine conversations between number of patients treated, number of patients providers [5]. These existing tools meant to discharged, number of patients readmitted, why improve collaboration and communication patients are getting readmitted, etc. are have become inundated with bureaucratic and relatively inaccessible to a clinician without system related messages, losing sight of their fairly significant additional work. Emergency original intentions. This leaves a gap where room and inpatient clinicians have little to no personal, formal peer-to-peer interaction is knowledge of what happens to a patient in the missing in the everyday use of the long run unless they spend the time technology—a part of a clinicians desired daily intentionally tracking these patients [12]. workflow that the technology could be designed to support [9]. We propose the potential value of a simple dashboard aimed at presenting this information, We propose a focus on using the EMR to focusing specifically on a clinician’s collective facilitate connections between clinicians rather patient panel and allowing them to review than isolating them further. It could kindle and start relationships with other providers to important cases to be reviewed again later, strengthen the healthcare community and allowing them to insert or connect comments or improve collaboration on patient care. Tools thoughts to a patient that may not be suitable for could aim at both communicating with the close the patient record but instead for the clinician’s friends that clinicians already have as well as future reflection. Tools could help coordinate finding new ones in different fields and searches among platforms, streamlining the specialties. These collaborations would be process for the clinician in finding answers to patient-based, driven by clinical questions and clinical questions. Second, the EMR or a tool curiosity about active care or retrospective within it has the potential to support the saving questions looking for feedback or teaching and organizing of these learning points, helping points. Integrating such interfaces in the current a clinician keep track of them and find them EMR systems also has the benefit of efficient again later on. These types of interventions communication that aligns with the privacy could reduce the frustrations associated with standards in the Health Information Portability inefficient learning and allow clinicians to more and Accountability Act (HIPPA). Again, the tangibly feel progression in the knowledge they EMR has an opportunity to reinforce and have achieved. encourage behaviors that clinicians actively perform as a way to feel more complementary 3. Conclusion to their work as opposed to antagonistic. Although the EMR has revolutionized the 2.3 Streamlining the process of healthcare industry, it has clearly been research and discovery accompanied by some frustrating side effects that are creating new problems such as One of the most enjoyable features of clinical physician burnout and stress. It is important that work is the continuous learning and discovery one recognizes the components that are that occurs from patients, other providers, and contributing to these issues, such as increased the literature. Although EMRs are designed to time using the EMR and reduction in time spent contain thousands of different alerts to their doing the things that are most important, among users, the platform does little to reinforce other things. Thus, augmenting the EMR education or provide the groundwork for new interface to support items that clinicians find learning to be made. Currently, many EMRs do interesting and help them do those things more link to open outside learning tools like efficiently is important to explore. UpToDate, Epocrates, or similar tools but leave This position describes behaviors and content the user alone in finding and coordinating an that clinicians are already engaging with while answer – something which might require using the EMR but with inefficient and ill- multiple resources, websites, and tools. In designed methods. It argues that designing addition, after finding the desired information, interfaces that intentionally address these areas there is nothing available for clinicians to more around self-assessment and feedback would be systematically keep track of the information beneficial in combating these EMR-related learned, opening the door for inefficiency as the issues, while also providing tools within the same search may be repeated many times. EMR that are designed for the clinician’s Clinicians try many different ways to keep track growth. Although there are policies and of or organize this information but are unable to structures that can make implementation find reliable and consistent methods [12]. challenging, we believe our suggestions do We propose that the EMR has an opportunity to little to go outside a clinician’s normal practice, support providers in these learning aspirations aiming to operationalize these practices. We by considering their workflow and helping acknowledge that different types of clinicians coordinate the tools related to them. Such may also experience these features differently things could allow providers to highlight and would advise a starting focus with emergency and inpatient physicians as these 12. Retrieved from groups have little to no systematic follow up https://www.newyorker.com/magazine/ with past patients. It is important to recognize 2018/11/12/why-doctors-hate-their- that the majority of the opportunity that exists computers is in the interface and intention of the system, [6] A. Jay Holmgren, N. Lance Downing, David W. Bates, Tait D. Shanafelt, without requiring new data, new variables, or Arnold Milstein, Christopher D. Sharp, new data collection infrastructure. This should David M. Cutler, Robert S. Huckman, make it all the more exciting to test and and Kevin A. Schulman. 2020. implement within the EMR. Assessment of electronic health record use between us and non-us health References systems. JAMA Intern. Med. 02163, [1] Julia Adler-Milstein, Wendi Zhao, (2020), 1–9. Rachel Willard-Grace, Margae Knox, DOI:https://doi.org/10.1001/jamaintern and Kevin Grumbach. 2020. Electronic med.2020.7071 health records and burnout: Time spent [7] Joshua C. Mandel, David A. Kreda, on the electronic health record after Kenneth D. Mandl, Isaac S. Kohane, hours and message volume associated and Rachel B. Ramoni. 2016. SMART with exhaustion but not with cynicism on FHIR: A standards-based, among primary care clinicians. J. Am. interoperable apps platform for Med. Inform. Assoc. 27, 4 (2020), 531– electronic health records. J. Am. Med. 538. Informatics Assoc. 23, 5 (2016), 899– DOI:https://doi.org/10.1093/jamia/ocz2 908. 20 DOI:https://doi.org/10.1093/jamia/ocv1 [2] Alex Azar II. 2020. 21st Century Cures 89 Act: Interoperability, Information [8] Edward R. Melnick, Liselotte N. Blocking, and the ONC Health IT Dyrbye, Christine A. Sinsky, Mickey Certification Program. Department of Trockel, Colin P. West, Laurence Health and Human Services. Retrieved Nedelec, Michael A. Tutty, and Tait from Shanafelt. 2020. The Association https://www.federalregister.gov/docum Between Perceived Electronic Health ents/2020/05/01/2020-07419/21st- Record Usability and Professional century-cures-act-interoperability- Burnout Among US Physicians. Mayo information-blocking-and-the-onc- Clin. Proc. 95, 3 (2020), 476–487. health-it-certification DOI:https://doi.org/10.1016/j.mayocp.2 [3] Geraldine Fitzpatrick and Gunnar 019.09.024 Ellingsen. 2013. A review of 25 years of [9] Wanda Pratt, Madhu C. Reddy, David CSCW research in healthcare: W. McDonald, Peter Tarczy-Hornoch, Contributions, challenges and future and John H. Gennari. 2004. agendas. Incorporating ideas from computer- DOI:https://doi.org/10.1007/s10606- supported cooperative work. J. Biomed. 012-9168-0 Inform. 37, 2 (April 2004), 128–137. [4] Rebekah L. Gardner, Emily Cooper, DOI:https://doi.org/10.1016/j.jbi.2004. Jacqueline Haskell, Daniel A. Harris, 04.001 Sara Poplau, Philip J. Kroth, and Mark [10] Fred Schulte and Erika Fry. 2019. Death Linzer. 2019. Physician stress and By 1 , 000 Clicks : Where Electronic burnout: the impact of health Health Records Went Wrong. Fortune. information technology. J. Am. Med. Retrieved from Informatics Assoc. 26, 2 (2019), 106– https://khn.org/news/death-by-a- 114. thousand-clicks/ DOI:https://doi.org/10.1093/jamia/ocy1 [11] C P West, L N Dyrbye, and T D 45 Shanafelt. 2018. Physician burnout : [5] Atul Gawande. 2018. Why Doctors contributors , consequences and Hate Their Computers. The New Yorker solutions. (2018). DOI:https://doi.org/10.1111/joim.1275 2 [12] Andrew Lukas Yin, Pargol Gheissari, Inna Wanyin Lin, Michael Sobolev, John P. Pollak, Curtis Cole, and Deborah Estrin. 2020. Role of technology in self-assessment and feedback among hospitalist physicians: semistructured interviews and thematic analysis. J. Med. Internet Res. 22, 11 (2020), 1–12. DOI:https://doi.org/10.2196/23299