Introduction
An admission at 2:00 am, a stack of paper charts to shuffle through (at least they were not lost), illegible handwriting to decipher, a trip to radiology to try and find the chest film and hoping it was not at the back of the roller board, a critical laboratory test that had to be repeated because it was lost. The medical record has come a long way over the past few decades as it has transitioned from a paper-based system to an electronic one. Electronic health records (EHRs)* have been widely adopted by health systems, allowing information to be readily accessible from different sites, radiographs and laboratory results to be easily viewed and trended, and order entry to become safer and more standardized.
Yet We Can and Must Do Better
The nostalgia imperfecta about the pre-EHR era is reflective of the shortcomings of the present EHR rather than the laudable attributes of the pre-EHR era. There is great potential to increase the efficiency and effectiveness of the EHR and to enhance its use as an educational tool. It is important to note that most of today’s United States trainees have never practiced in a non-EHR system, and perhaps, may be better equipped to improve the system rather than yearning for the “good old days.”
Physicians spend half of their working hours using the EHR (1), but in its current form, the EHR has unintended consequences (2). The EHR is a major contributor to increased provider burnout and frustration due to usability challenges, “alert overload,” and data entry demands. Patient interactions may be negatively affected; information is missing and/or erroneously propagated due to copy-forward and prepopulation errors. In addition, lack of interoperability between health systems is associated with poor health outcomes (3). Notes become bloated with redundant information for billing and coding purposes only, leaving a superficial discussion and analysis of patient problems and inadequate documentation of clinical reasoning. We have traded one problem (lack of information) for another (information overload).
These EHR challenges and their effect on nephrology fellowship education are clearly demonstrated in the study by Yuan et al. (4) in this issue of CJASN. In the authors’ survey of nephrology program directors, faculty, and fellows, the EHR was viewed as reducing engagement in educational activities, hampering direct patient interactions, and contributing to work-hour violations. Despite a relatively low response rate (data available from 72 of 827 nephrology fellows in United States nephrology fellowship programs), the survey data are valuable in providing the trainee perspective. Of course, the low response rate engenders a substantial risk of bias to those who have strong feelings about the topic. Only 51% of responding fellows agreed that the EHR contributed positively to their education. The most highly rated benefits of the EHR were accessibility from home or a mobile device, efficient laboratory result trending and medication reconciliation, and ensuring correct kidney replacement orders.
Negative aspects of the EHR included “excessive and/or irrelevant EHR documentation,” less time spent on direct patient interactions, and copy-forward and prepopulation errors (4). In fact, 91% of faculty members responded that they have observed fellows making copy-forward errors; there is little reason to expect that faculty members would make fewer. Most concerning was the perception that the EHR negatively affected important aspects of their education, including performing procedures, attending conferences, and doing self-directed reading. Although the EHR facilitated and ensured accuracy in dialysis and continuous renal replacement therapy order entry, concerns were raised by program directors that fellows would not be able to write orders without prompts. That said, because they are very unlikely to do so in the modern era, it is not clear whether this is a meaningful deskilling or simply a workflow modernization.
EHR usability (efficiency, effectiveness, and ease of use) needs to be improved to make data entry and information retrieval and synthesis less burdensome. The EHR is, in many ways, one of the core characteristics of the modern clinical ecosystem. It is important then to envision how to optimally integrate the EHR into the learning environment to enhance its effect across the medical education continuum. We highlight some of the ways this can be accomplished.
Learning Tool
The EHR can be an effective learning tool for both the trainee and the health system. Clinical decision support systems can be integrated into the EHR to facilitate evidence-based decision making that is updated continuously. This can be in the form of drug reference libraries, links to clinical practice guidelines, or other forms of context-based microlearning that divide information into small chunks and deliver it to learners at a time that is most relevant to the clinical situation. This “just-in-time learning” has the potential to both educate the learner and improve the quality of care.
The EHR contains a vast amount of health data that can be analyzed to generate real-time actionable health system knowledge that can be used to improve clinical care, track health outcomes, and guide decision making. The process of turning data into knowledge and then, into patient and system improvement is the foundation of the Learning Health System (LHS) (5). The ability to harness large amounts of health data contained in the EHR is a powerful means to engage learners in the LHS model and to address the Accreditation Council for Graduate Medical Education competencies of practice-based learning and improvement and systems-based care.
Outcomes Tracking
The EHR can be used by trainees to manage their patient panel and track the outcomes of patients under their care. Results can be compared with benchmarks and with the outcomes of patient panels managed by other trainees. Long-term follow-up of patients can be tracked, and alerts can be placed in the EHR notifying trainees when one of their panel patients is receiving care within the health system. This is a particularly valuable in longitudinal educational experiences. Program directors can use data from the EHR to track the patient mix and procedures of learners to identify and address gaps in training.
On a larger scale, training programs can use the EHR to link educational outcomes to clinical outcomes to evaluate the quality of their educational programs by the quality of care delivered by their graduates (6). This will facilitate the design of educational programs that positively affect patient outcomes, a link of utmost importance that has yet to be broadly made.
Enable Better Feedback
One of the most important methods by which learners (and practicing clinicians) improve their practice is by receiving feedback. Performance and confidence are often not well correlated, overconfidence is common, and diagnostic errors cause substantial harm (7). The EHR is a largely untapped tool that can be used to enable precise feedback about clinical reasoning; the EHR should, ideally, allow an accurate and easily accessible means to track decision making and patient outcomes over time in a given case. Tools, such as chart-stimulated recall, coupled with the information in the EHR can allow effective feedback to be delivered about specific cases.
Predictive Analytics
The EHR provides a rich data source (“big data”) for machine learning and artificial intelligence. A goal of this work is to develop predictive algorithms on the basis of a patient’s unique characteristics to forecast outcomes and guide clinical decision making. An example of this approach is the collaboration between Google, DeepMind, and the Veterans Administration Health Care System to apply artificial intelligence to the prediction of AKI (8). The modeling used in this study involved >700,000 patient records and 6 billion independent data elements. Data from the EHR can also be used to identify specific patients for clinical trials on the basis of their unique characteristics. This is especially important as nephrology and other specialties develop an “on study” culture.
Making Transitions in Care Safer
Transitions in care are a vulnerability that can often lead to patient safety events due to communication breakdowns. Although there are few downsides to modern duty-hour restrictions, clinical handovers are now more ubiquitous than ever. The EHR provides tools to standardize handovers between providers and systems (9,10). The EHR can also be used to enhance communication with other health care providers to promote interprofessional practice and education.
In summary, the EHR is a powerful tool to access health information from any site and improve the efficiency, appropriateness, and safety of health care. Despite these potential benefits, usability issues remain a major barrier, contributing to frustration and burnout. The EHR can provide “just-in-time” context-based learning, improve and teach about health system performance, track patient and learner outcomes, improve and document clinical reasoning, enable effective feedback, be a source of “big data” for predictive analytic modeling, and improve transitions in care. The present nihilism about the EHR must soon give way to a period of pragmatic optimism in which we begin to realize its full potential.
Disclosures
M. Rosenberg reports honorarium support from Wolters Kluwer. A. Olson has nothing to disclose.
Funding
None.
Acknowledgments
Dr. Mark Rosenberg is Past President of the American Society of Nephrology (ASN). The views expressed are his own and not those of the ASN.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
*The Office of the National Coordinator for Health Information Technology recommends using “electronic health record” over “electronic medical record” as it is a broader term focusing on the total health of the patient and contains information from all providers involved in the car of the patient.
See related article, “The Electronic Medical Record and Nephrology Fellowship Education in the United States: An Opinion Survey,” on pages 949–956.
References
- 1.Sinsky C, Colligan L, Li L, Prgomet M, Reynolds S, Goeders L, Westbrook J, Tutty M, Blike G: Allocation of physician time in ambulatory practice: A time and motion study in 4 specialties. Ann Intern Med 165: 753–760, 2016. [DOI] [PubMed] [Google Scholar]
- 2.Colicchio TK, Cimino JJ, Del Fiol G: Unintended consequences of nationwide electronic health record adoption: Challenges and opportunities in the post-meaningful use era. J Med Internet Res 21: e13313, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Usher M, Sahni N, Herrigel D, Simon G, Melton GB, Joseph A, Olson A: Diagnostic discordance, health information exchange, and inter-hospital transfer outcomes: A population study. J Gen Intern Med 33: 1447–1453, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Yuan CM, Little DJ, Marks ES, Watson MA, Raghavan R, Nee R; Nephrology Education Research and Development Consortium—NERDC: The electronic medical record and nephrology fellowship education in the United States: An opinion survey. Clin J Am Soc Nephrol 15: 949–956, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Platt JE, Raj M, Wienroth M: An analysis of the Learning Health System in its first decade in practice: Scoping review. J Med Internet Res 22: e17026, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rosenberg ME, Gauer JL, Smith B, Calhoun A, Olson APJ, Melcher E: Building a medical education outcomes center. JMIR Med Educ 5: e14651, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Meyer AN, Payne VL, Meeks DW, Rao R, Singh H: Physicians’ diagnostic accuracy, confidence, and resource requests: A vignette study. JAMA Intern Med 173: 1952–1958, 2013. [DOI] [PubMed] [Google Scholar]
- 8.Tomašev N, Glorot X, Rae JW, Zielinski M, Askham H, Saraiva A, Mottram A, Meyer C, Ravuri S, Protsyuk I, Connell A, Hughes CO, Karthikesalingam A, Cornebise J, Montgomery H, Rees G, Laing C, Baker CR, Peterson K, Reeves R, Hassabis D, King D, Suleyman M, Back T, Nielson C, Ledsam JR, Mohamed S: A clinically applicable approach to continuous prediction of future acute kidney injury. Nature 572: 116–119, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tisdale RL, Eggers Z, Shieh L: EMR-based handoff tool improves completeness of internal medicine residents’ handoffs. BMJ Open Qual 7: e000188, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Olson A, Rencic J, Cosby K, Rusz D, Papa F, Croskerry P, Zierler B, Harkless G, Giuliano MA, Schoenbaum S, Colford C, Cahill M, Gerstner L, Grice GR, Graber ML: Competencies for improving diagnosis: An interprofessional framework for education and training in health care. Diagnosis (Berl) 6: 335–341, 2019. [DOI] [PubMed] [Google Scholar]