Specialty and Operator Status Influence Electronic Health Record Use Variation

ABSTRACT

Objectives: Electronic health record (EHR) systems are central to modern practice yet contribute to physician workload and burnout. Metrics such as documentation timeliness, efficiency, and after-hours “pajama time” are increasingly used to assess provider performance, but variation across procedural roles and specialties remains understudied.

Study Design: Cross-sectional study.

Methods: We analyzed 23 months of provider-level EHR data from a single academic health system. Providers with more than 0 operative notes were classified as operators. Six standardized metrics—delayed visit closure, delayed discharge note signing, delayed cosign and verbal order completion, delayed review of high-priority results, mean pajama time, and an institutional proficiency score—were compared between operators and nonoperators using Welch t tests. Subgroup analyses were performed within medicine, obstetrics-gynecology (ob-gyn), and pediatrics. Additional operator comparisons across 6 procedural specialties used analysis of variance.

Results: Of 2516 providers, 724 (28.8%) were operators. Operators had higher rates of delayed cosign (9.5% vs 5.7%; P < .001) and verbal order completion (20.5% vs 17.1%; P = .003) but similar pajama time and proficiency compared with nonoperators. In medicine, operators had lower pajama time than nonoperators (21.8 vs 29.2 minutes; P = .006). In pediatrics, operators had fewer delayed discharge notes (P = .035). In ob-gyn, operators showed fewer verbal order and result review delays but higher proficiency and more delayed discharge notes (all P < .05). Among procedural specialties, ophthalmology operators had the highest proficiency yet greater delays across timeliness metrics (all P < .05).

Conclusions: EHR utilization varies by procedural status and specialty, underscoring the need for workflow-specific optimization rather than uniform performance benchmarks.

Am J Manag Care. 2026;32(1):In Press

Takeaway Points

Our study shows that physician training and specialty influence how efficiently clinicians use electronic health records (EHRs). Understanding these differences can guide better resource allocation, workflow design, and technology support.

• Targeted EHR training can improve efficiency and reduce after-hours “pajama time.”
• Specialty-specific workflows should inform EHR design and staffing decisions.
• Measuring EHR use can identify burnout risks and guide wellness initiatives.
• Data-driven policies can align technology use with clinician needs, improving both productivity and patient care quality.

Electronic health record (EHR) systems are integral to clinical practice but contribute to physician workload and burnout.1 Measures such as documentation timeliness, efficiency, and after-hours “pajama time” are used to gauge provider EHR performance.1,2 The consistency and specialty-specific nature of EHR utilization differences are unknown.3,4 This study compared EHR utilization and documentation performance between operators and nonoperators, with subgroup analyses to identify specialty-specific variation.

METHODS

This cross-sectional study analyzed provider EHR data from March 2023 through January 2025, including documentation timeliness, after-hours activity, and proficiency scores across a single academic health system using an enterprise Epic EHR. Providers with more than 0 operative notes were classified as operators, whereas those with 0 operative notes were defined as nonoperators. This categorization reflects procedural vs nonprocedural practice patterns in EHR workflow studies.

Seven standardized metrics were examined: delayed visit closure (> 14 days), delayed discharge note signing (> 48 hours), delayed cosign and verbal order completion (> 48 hours), delayed review of high-priority results (> 48 hours), mean pajama time, and an institutionally derived proficiency score.

Operators and nonoperators were compared using Welch t tests. Subgroup analyses were performed within medicine, obstetrics-gynecology (ob-gyn), and pediatrics. Additional comparisons among operators in 6 procedural specialties (surgery, orthopedics, otolaryngology, neurosurgery, ophthalmology, and interventional radiology) used analysis of variance with post hoc Tukey testing. Analyses were conducted in Python 3.12.2 (Python Software Foundation).

RESULTS

Of 2516 providers, 724 (28.8%) were operators. Operator proportions varied by specialty, with procedural fields contributing the most operators. Compared with nonoperators, operators had higher rates of delayed cosign (9.5% vs 5.7%; P < .001) and verbal order completion (20.5% vs 17.1%; P = .003). Pajama time (P = .26) and proficiency scores (P = .87) were similar.

In medicine, operators had lower pajama time than nonoperators (21.8 vs 29.2 minutes; P = .006) (Figure). In pediatrics, operators had fewer delayed discharge notes (0.00% vs 0.34%; P = .035). In ob-gyn, operators had fewer delays in verbal order completion (13.5% vs 34.7%; P = .010) and high-priority result review (17.5% vs 34.9%; P = .024), but more delayed discharge notes (0.35% vs 0.00%; P = .004) and higher proficiency scores (5.26 vs 4.46; P = .040).

Among operators, significant between-specialty differences were observed in visit closure, verbal order delay, high-priority result review, and proficiency. Ophthalmology operators had greater delays in visit closure (vs otolaryngology, P = .025), verbal orders (vs orthopedics, otolaryngology, and surgery; all P < .05), and result review (vs orthopedics, P = .027), yet the highest proficiency scores (vs surgery, P = .009; vs neurosurgery, P < .001).

Emergency medicine operators had fewer order delays (P = .003), whereas family medicine operators had more order delays (P = .048) but less pajama time (P = .002).

DISCUSSION

EHR performance varied by both operator status and specialty. Operators demonstrated more delays in cosign and verbal order completion, likely reflecting team-based perioperative workflows in which delegated orders are signed later. Proficiency and pajama time did not differ, likely because these metrics emphasize ambulatory activity, which is less relevant to procedural practice.

Within departments, proceduralists in medicine and pediatrics had lower pajama time and fewer delayed discharge summaries, whereas even within a single specialty, ob-gyn operators showed a mixed pattern of higher proficiency but more delayed discharge documentation. Across procedural specialties, ophthalmology stood out for high proficiency but poor timeliness, highlighting workflow-specific trade-offs.

EHR utilization differed both between and within specialties, despite physicians often being held to a universal utilization standard. These findings suggest that system-level comparisons, incentives, and penalties should account for specialty-specific workflows and support targeted, context-aware optimization. Interpretation should also consider the single-center setting and potential misclassification of operator status, which may limit generalizability.

Author Affiliations: Oregon Health & Science University (PDJ, GG, AD, RT, JAG, JD), Portland, OR; University of Pennsylvania (EW), Philadelphia, PA.

Source of Funding: None.

Author Disclosures: Dr Gold has received a grant from the Agency for Healthcare Research and Quality on analyzing electronic health record use. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (PDJ, EW, RT, JD); acquisition of data (AD, JAG, JD); analysis and interpretation of data (PDJ, GG, AD, RT, JAG, JD); drafting of the manuscript (PDJ, EW, RT, JD); critical revision of the manuscript for important intellectual content (PDJ, GG, EW, RT, JAG, JD); statistical analysis (PDJ, GG, JD); administrative, technical, or logistic support (AD); and supervision (JD).

Address Correspondence to: Julie Doberne, MD, PhD, Oregon Health & Science University, 3181 Sam Jackson Park Rd, Portland, OR 97239. Email: doberne@ohsu.edu.

REFERENCES

1. Khairat S, Burke G, Archambault H, Schwartz T, Larson J, Ratwani RM. Perceived burden of EHRs on physicians at different stages of their career. Appl Clin Inform. 2018;9(2):336-347. doi:10.1055/s-0038-1648222

2. Arndt BG, Beasley JW, Watkinson MD, et al. Tethered to the EHR: primary care physician workload assessment using EHR event log data and time-motion observations. Ann Fam Med. 2017;15(5):419-426. doi:10.1370/afm.2121

3. Melnick ER, Dyrbye LN, Sinsky CA, et al. The association between perceived electronic health record usability and professional burnout among US physicians. Mayo Clin Proc. 2020;95(3):476-487. doi:10.1016/j.mayocp.2019.09.024

4. Al-Rayes SA, Alumran A, AlFayez W. The adoption of the electronic health record by physicians. Methods Inf Med. 2019;58(2-03):63-70. doi:10.1055/s-0039-1695006