Patient Portal Messaging in Shoulder and Elbow Surgery Patients: A Retrospective Cohort Study

Katelyn E Parsons; Devika A Shenoy; Alyssa R Henriquez; Kevin A Wu; Eoghan T Hurley; Oke A Anakwenze; Kevin P Speer; Yaw D Boachie-Adjei; Michael Moverman; Jay M Levin; Jonathan F Dickens; Christopher S Klifto

doi:10.1016/j.jse.2026.05.002

Patient Portal Messaging in Shoulder and Elbow Surgery Patients: A Retrospective Cohort Study

J Shoulder Elbow Surg. 2026 May 19:S1058-2746(26)00295-8. doi: 10.1016/j.jse.2026.05.002. Online ahead of print.

Affiliations

¹ Department of Orthopaedic Surgery, Duke University Health System, Durham, NC. Electronic address: katie.parsons@duke.edu.
² Department of Orthopaedic Surgery, Duke University Health System, Durham, NC.
³ Department of Orthopaedic Surgery, Mount Sinai Hospital, New York, NY.

PMID: 42162893
DOI: 10.1016/j.jse.2026.05.002

Abstract

Background: Patient portals have become a primary mechanism for patient-provider communication, offering convenience for patients while also creating an uncompensated workload burden for providers. This issue is underexplored in shoulder and elbow surgery, where pain, quality of life burden, and complex rehabilitation plans may drive frequent messaging. This study evaluated trends in portal messaging within a shoulder and elbow orthopedic cohort.

Methods: This study was a retrospective review of all patients evaluated in a tertiary center's shoulder and elbow division between January 1^st, 2022, and December 31^st, 2025. Five fellowship-trained shoulder and elbow surgeons and four advanced practice providers (APPs) were included. Demographic data (age, sex, race, ethnicity, preferred language, surgical status) were collected from the medical record. Associations between patient characteristics and sending ≥1 medical advice message were assessed using logistic regressions, with results reported as adjusted odds ratios (aORs). Patient characteristics associated with the number of messages was characterized using a Poisson regression, with results reported as adjusted incidence rate ratios (aIRRs).

Results: The cohort included 23,643 patients; 21% (n=4,953) sent ≥1 medical advice message. Medical advice requests represented 41,542 (30%) of 137,960 total messages. The number of medical advice messages steadily grew by nearly 400% between 2022 and 2026. Surgical patients were more likely to send messages (aOR 5.93, p<0.001) and had a higher message count (aIRR 23.3, p<0.001) than non-surgical patients. Black/African American patients had a lower odd of sending a medical advice message (aOR 0.55, p<0.001) and lower message counts (aIRR 0.58, p<0.001) compared to White patients. Spanish-speaking patients had a lower odds of sending a medical advice message (aOR 0.21, p<0.001) and lower message counts (aIRR 0.17, p<0.001) compared to English-speaking patients.

Discussion: This study identified a growing volume of medical advice requests in shoulder and elbow surgery by nearly 400%, representing an expanding administrative concern. The independent negative predictive value of Black race and Spanish language for low messaging use suggests that current portal structures may create or exacerbate disparities in digital health access. Future implementation must address uncompensated provider workload while ensuring patients from all groups have equitable access to portal tools.

Level of evidence: IV, Descriptive Epidemiology Study.

Keywords: Digital Healthcare; Electronic Medical Record; Messaging; Orthopaedic Surgery; Patient Portal; Shoulder and Elbow Surgery.