Contents lists available at ScienceDirect International Journal of Medical Informatics journal homepage: www.elsevier.com/locate/ijmedinf SEWA: A framework for sociotechnical analysis of electronic health record system workarounds Vincent Blijleven a,b, ⁎ , Kitty Koelemeijer a , Monique Jaspers b a Center for Marketing and Supply Chain Management, Nyenrode Business University, Straatweg 25, 3621 BG, Breukelen, the Netherlands b Department of Medical Informatics, Academic Medical Center, Meibergdreef 15, 1105 AZ, Amsterdam, the Netherlands ARTICLEINFO Keywords: Electronic health records Framework Patient safety Unintended consequences Workarounds Workfow ABSTRACT Objective: To develop a conceptual framework, SEWA, to address challenges of studying workarounds emerging from Electronic Health Record (EHR) system usage. Materials and methods: SEWA is based on direct observations and follow-up interviews with physicians, nurses and clerks using their EHR at a large academic hospital. SEWA was developed by an iterative process: each new version was reviewed by experts (case study participants, hospital management, EHR developers) and refned accordingly till deemed fnal. Results: SEWA defnes the work system and its fve components constituting the context in which EHR work- arounds are created. It also contains 15 rationales for creating EHR workarounds. Furthermore, four attributes are included that defne EHR workarounds: cascadedness, anticipatedness, avoidability, and repetitiveness. Finally, SEWA lists the possible efects of workarounds on outcomes of clinical processes in terms of scope and impact. Discussion: SEWA provides a grounded foundation for performing sociotechnical analyses of EHR workarounds based on components of the work system. SEWA can likewise be supportive in planning redesign eforts of the work system. Finally, workarounds are subject to gradual change caused by e.g. changes in one’s knowledge of the EHR, hospital policies, care directives, and system updates. Snapshots of SEWA can be taken over time and compared to gain insights into the evolution of workarounds. Conclusion: Given the absence of a sociotechnical framework to study EHR workarounds, SEWA could aid re- searchers and practitioners to identify, analyze and resolve workarounds, and thereby contribute to improved patient safety, efectiveness of care and efciency of care. 1. Background and significance Electronic health record systems (EHRs) are increasingly being adopted by healthcare organizations worldwide [1–3], spurred by government-led incentives [4] and in pursuit of favorable outcomes related to patient safety [5–7], quality of care [7,8], efciency of care [7,9,10], and reduced costs [11,12]. However, realizing the expected benefts of adopting EHRs remains challenging. Examples are EHR users experiencing ‘alert fatigue’ [13], not all clinical work being supported by EHRs [14], unavailability of complete clinical information at the point of care [15], difculties in fnding the right information in the EHR [16], and signifcantly disrupted workfows due to modifed timing, sequence of work practices and revised professional responsi- bilities [17–19]. Many causes of unintended consequences of EHRs can be traced back to discrepancies between the behavior, intentions and expecta- tions of EHR users, and the workfows dictated by EHRs [18,20–23]. When users experience workfow mismatches they create workarounds [24]. Workarounds have been defned as “informal temporary practices for handling exceptions to normal workfow” [25] that “do not follow explicit or implicit rules, assumptions, workfow regulations, or inten- tions of systems designers” [26]. The impact of workarounds may extend far beyond solely allowing EHR users to proceed with their workfow to get their clinical tasks done. Existing studies on EHR workarounds tend to report their fndings from a single out of four angles we identifed. First, EHR users have distinct rationales for creating workarounds such as a lack of declara- tive knowledge (i.e. not knowing how to use (a part of) the EHR), memory aids (e.g. writing patient data down on paper during an out- patient consultation session and entering this data into the EHR later https://doi.org/10.1016/j.ijmedinf.2019.02.012 Received 11 May 2018; Received in revised form 28 September 2018; Accepted 28 February 2019 ⁎ Corresponding author at: Straatweg 25, 3621 BG, Breukelen, the Netherlands. E-mail addresses: v.blijleven@nyenrode.nl, v.b.blijleven@amc.uva.nl (V. Blijleven). International Journal of Medical Informatics 125 (2019) 71–78 1386-5056/ © 2019 Elsevier B.V. All rights reserved. T