Missing data are indicated by dashed lines and black dots. Findings are from modeling by piecewise linear regression of patient medical advice request trend using a spline for month with a single knot at March 2020 and Huber-White SEs.
eTable 1. Epic Inbox Message Categories
eTable 2. Physician Specialty Grouping
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Nath B, Williams B, Jeffery MM, et al. Trends in Electronic Health Record Inbox Messaging During the COVID-19 Pandemic in an Ambulatory Practice Network in New England. JAMA Netw Open. 2021;4(10):e2131490. doi:10.1001/jamanetworkopen.2021.31490
COVID-19–related disruptions have changed how patients access routine care. Anecdotal evidence indicates that, with the pandemic, ambulatory physicians experienced an increase in patient medical advice requests (PMARs). Therefore, we compared patient message volume during the pandemic with prepandemic levels.
This cross-sectional study analyzed deidentified electronic health record metadata (Signal, Epic Systems) from March 2018 to June 2021 in a large ambulatory practice network in New England. This study was deemed exempt for review and informed consent by Yale University’s institutional review board because it involves secondary research of deidentified electronic health record metadata that did not include any patient identifier or private health information. This study follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Trends in inbox message volume (categorized by message type and source; see eTable 1 in the Supplement), time in the inbox, visit volume and type (in-person vs telehealth), patient volume, and patient use of the patient portal were examined using descriptive statistics. Physician specialties were grouped into primary care, medical, and surgical specialties (eTable 2 in the Supplement). Variables were compared before (March 2018 to February 2020) and during (March 2020 to June 2021) the COVID-19 pandemic. To assess whether the onset of the pandemic was an inflection point in PMARs, message volume per physician per day was modeled by piecewise linear regression using a spline for month with a single knot at March 2020 and Huber-White SEs. Three months of inbox message data were missing (3 of 40 months [7.5%]) and excluded from the analysis. To test for significance (P < .05), we used a 2-sided Wald test for equivalence of the coefficients. We used Stata statistical software version 16 (StataCorp) for data analyses.
Forty months of inbox messages were analyzed, including 10 850 401 messages to 419 unique physicians from 38 specialties across 141 practice sites (Figure). Overall, primary care, medical, and surgical physicians received 49.3, 33.4, and 20.7 messages per day, respectively. Between March 2020 and June 2021, mean (SD) total messages per day increased from 45.0 (27.4) to 46.0 (27.4) messages per day for primary care physicians, from 29.3 (20.7) to 32.0 (20.8) messages per day for medical physicians, and from 16.6 (11.9) to 23.3 (17.9) messages per day for surgical physicians. Patient-originated messages also increased, including PMARs (from a mean [SD] of 1.8 [1.8] to 3.9 [3.2] messages per day for primary care physicians; from 1.0 [1.7] to 2.2 [2.9] messages per day for medical physicians; and from 0.4 [0.5] to 1.1 [1.3] messages per day for surgical physicians), patient calls, and time in the inbox (from 21.7 [12.7] to 25.1 [13.7] minutes per day for primary care physicians; from 13.4 [10.7] to 15.6 [9.2] minutes per day for medical physicians; and from 7.6 [7.2] to 11.1 [10.0] minutes per day for surgical physicians) (Table).
For primary care, before COVID, PMARs were increasing by 0.023 messages per physician per day each month (95% CI, 0.015 to 0.031 messages; P < .001). In March 2020, PMARs increased by 1.940 messages per physician per day (95% CI, 1.508 to 2.371 messages; P < .001). During COVID, the numbers of PMARs did not change significantly (slope, −0.020; 95% CI, 1.508 to 2.370; P = .40). The rate of change in PMARs per clinical day was not significantly different during COVID compared with before COVID (slope, −0.040; 95% CI, −0.089 to 0.005; P = .08) (Figure).
For medical specialties, before COVID, PMARs were increasing by 0.033 messages per physician per day each month (95% CI, 0.026 to 0.039 message; P < .001). In March 2020, PMARs increased by 0.686 messages per physician per day (95% CI, 0.442 to 0.931 message; P < .001). During COVID, PMARs were increasing by 0.030 messages per physician per day each month (95% CI, 0.008 to 0.052 message; P = .009). The rate of change in PMAR per clinical day was not significantly different during COVID compared with before COVID (Figure).
For surgical specialties, before COVID, PMARs were increasing by 0.018 messages per physician per day each month (95% CI, 0.013 to 0.023 message; P < .001). In March 2020, PMARs did not change by a statistically significant amount. During COVID, PMARs were increasing by 0.072 messages per physician per day each month (95% CI, 0.043 to 0.101 message; P < .001). The rate of change in PMAR per clinical day was faster during COVID compared with before COVID (slope, 0.05; 95% CI, 0.024 to 0.083; P = .001) (Figure).
Consistent with national trends,2 during 2020 COVID months, monthly in-person visits decreased for all specialties (primary care, 17.1%; medical, 18.5%; surgical, 7.7%), whereas telehealth visits increased. Surgical specialties, however, experienced an increase in in-person visits between January and June 2021. The number of unique patients seeking care during the pandemic decreased for primary care and medical specialties through June 2021 but increased for surgical specialties in 2021.
Findings from this cross-sectional study indicate that during the first 15 months of the COVID-19 pandemic there was a small but sustained increase in patient message volume, in particular PMARs, across all specialties, despite a decrease (for primary care and medical physicians) in the number of patients seeking care during the same period.3 The increase in surgical patients in 2021 could be associated with clearance of backlog of elective surgeries delayed in 2020.4 The increase in electronic messages from patients did not displace patient calls. Limitations of this study include systematic underestimation of inbox time based on how Signal defines and captures inbox activity, which is limited to the time in the inbox screen and is not inclusive of time spent on other tasks necessary to address an inbox message (eg, phone calls or medical record review).5 However, burnout related to inbox burden is established.1,6 Given the existing physician burnout crisis and the already known pandemic-related stressors and risks to the physician workforce, the additional inbox burden reported here warrants additional exploration to assess the nature of pandemic-related medical advice requests and the generalizability of these findings. With COVID-19 potentially remaining a long-term endemic threat to public health, the priority to systematically address inbox burden before the pandemic through workflow redesign, team-based inbox management, and consideration of reimbursement for inbox-related work remains.
Accepted for Publication: August 25, 2021.
Published: October 12, 2021. doi:10.1001/jamanetworkopen.2021.31490
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Nath B et al. JAMA Network Open.
Corresponding Author: Edward R. Melnick, MD, MHS, Department of Emergency Medicine, Yale School of Medicine, 464 Congress Ave, Ste 260, New Haven, CT 06519 (email@example.com).
Author Contributions: Dr Nath had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Nath, Williams, O’Connell, Goldstein, Melnick.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Nath, Melnick.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Nath, Jeffery, Melnick.
Obtained funding: Melnick.
Administrative, technical, or material support: Nath, Williams, O’Connell, Goldstein, Melnick.
Supervision: Williams, O’Connell, Goldstein, Melnick.
Conflict of Interest Disclosures: Dr Sinsky reported being employed by the American Medical Association. Dr Melnick reported receiving grants from the National Institute on Drug Abuse outside the submitted work. No other disclosures were reported.
Funding/Support: This work was supported by 2 American Medical Association Practice Transformation Initiatives (contract numbers 36648 and 36650).
Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the American Medical Association. The opinions expressed in this article are those of the authors and does not necessarily represent the official views of the American Medical Association.
Additional Contributions: We thank the physicians of the Yale-New Haven System. Prem Thomas, MD (Center for Medical Informatics, Yale School of Medicine), performed data visualization; Chris Gates, MS (Data Scientist, Signal Team, Epic), performed data collection; and John O’Bryan MA (Signal Director, Epic), performed data collection. These individuals were not compensated beyond their normal salaries.