Highlights
- •
Gender differences exist in ophthalmologist communication patterns.
- •
Female ophthalmologists seem to use a more patient-centered communication style.
- •
The COVID-19 pandemic impacted ophthalmologist communication patterns.
- •
Burnout may have impacted ophthalmologist communication patterns.
- •
Artificial intelligence–based telehealth algorithms must sample a diverse physician population.
Purpose
Physician communication patterns can increase patient satisfaction and adherence to therapy in the primary care setting. This study investigated gender differences in ophthalmologist communication patterns before and after the COVID-19 pandemic.
Design
Retrospective cohort study.
Methods
Messages sent by ophthalmic patients at Penn Medicine from 2017 to 2022 were collected. Differences in the number of physician messages sent for a given patient, median response length, and response time to patient inquiries and messages were examined based on year and physician gender.
Results
Female ophthalmologists sent longer response messages to their patients (median [25th, 75th percentiles] response length for women vs men: 672 [492-965] characters vs 637 [460, 918] characters; P < .0001) and a higher number of response messages per patient than their male counterparts (mean [SD] for women vs men: 5.5 [2.9] vs 3.0 [1.5]; P = .04). There was an increase in this gender difference in the peri- and post-COVID-19 period (ie, 2020-2022) ( P = .007). Male ophthalmologists sent a higher percentage of same-day responses from 2017 to 2020 ( P < .0001), whereas female ophthalmologists sent a higher percentage of same-day responses from 2021 to 2022 ( P < .0001). The largest gender difference in same-day responses occurred in 2020 (34% for men vs 30% for women; P < .0001).
Conclusions
Gender differences exist in ophthalmologist communication patterns, and the COVID-19 pandemic impacted these differences. Future studies will be helpful in determining the potential association of these specific communication patterns with patient satisfaction assessments, eye health outcomes, and physician burnout.
Effective interactions between patients and physicians are critical to the healing relationship. Gender differences in communication with patients have been noted in multiple disciplines. , Studies showed that female physicians use a more patient-centered communication style, including being more attentive, providing more information, and showing greater empathy. , Patient-centered communication is associated with increased patient satisfaction and adherence to therapy. Given the accelerated expansion of telemedicine due to COVID-19 and the shift toward artificial intelligence (AI)-based communication, there is an increasing need to advance our understanding of communication patterns. ,
Effective patient-centered communication , is critical for ophthalmic patients because it can convey the importance of the treatment plan and enhancing adherence. , Gender differences in communication has increasing relevance within ophthalmology given the rising representation of women within ophthalmology over the past 50 years (from 4% in 1969 to 27% in 2019). , Previous studies have investigated virtual forms of patient-physician communication within ophthalmology including patient education, medication reminders, and visit reminders. However, these studies have not examined gender differences in patient-physician communication. In addition, the impact of major disruptions to health care systems’ workflow and patient-physician communication, as experienced during COVID-19, is largely unknown.
Therefore, the purpose of this study is to determine whether there is a gender difference in virtual communication patterns, including response time and response length, among ophthalmologists and to establish whether factors related to COVID-19 impacted these ophthalmologists’ communication patterns.
METHODS
After approval by the University of Pennsylvania institutional review board, a retrospective chart review was performed on existing data from Penn Medicine’s electronic message system from January 2017 to December 2022. Patients were included in this study if they were assigned a physician in Penn’s ophthalmology department and were enrolled in the health system’s electronic message system. The study population included patients seen at the Scheie Eye Institute and all affiliated clinics that were located in Philadelphia and surrounding suburbs, as described previously. Informed consent was not required as this was a retrospective study. Gender was defined as one’s personal identity as male, female, a blend of both or neither. Physician gender was determined based on gender-specific pronouns used on the institutional website (women = 24 and men = 21) ( Table 1 ).
| Female Ophthalmologists | Male Ophthalmologists | Gender Difference (Female – Male) | P Value (Comparing Female vs Male) | |
|---|---|---|---|---|
| No. of ophthalmologists (%) | 24 (53.3) | 21 (46.7) | 3 (6.6) | NA |
| No. of responses (%) | 23 846 (52.2) | 21 846 (47.8) | 2000 (4.4) | NA |
| No. of “other” responses (%) | 6944 (48.0) | 7522 (52.0) | –578 (–4.0) | NA |
| Mean no. of responses/patient, 2017-2022 (SD) | 5.5 (2.9) | 3.0 (1.5) | 2.6 (1.4) | .04 |
| No. of responses, 2017-2019 (%) | 7613 (52.1) | 7001 (47.9) | 612 (4.2) | NA |
| No. of responses, 2020-2022 (%) | 16 233 (52.2) | 14 845 (47.8) | 1388 (4.5) | NA |
| NA | NA | NA | NA | NA |
| Mean no. of responses/patient, 2017-2019 (SD) | 3.0 (0.9) | 1.6 (0.5) | 1.4 (0.3) | .04 |
| Mean no. of responses/patient, 2020-2022 (SD) | 8.0 (1.0) | 4.3 (0.1) | 3.7 (0.9) | .001 |
| P value (comparing 2017-2019 vs 2020-2022) | .001 | .0006 | .007 | NA |
Differences in the number of physician messages sent for a given patient, median response length (characters), and response time (same day, 1-7 days, 8+ days, other response) to patient inquiries and messages (n = 45 692) were examined based on year and physician gender using t tests, Wilcoxon rank sum tests, and χ 2 tests, respectively. Response time characterized as “other” included the following: 46% marked as “no further action required,” 17% represented a physician calling the patient, 16% represented a message reply from the clinic’s visit navigator to schedule appointments, and 21.1% were marked in the portal system as “not handled” ( Table 2 ). Logistic regression was used to evaluate associations between physician gender and response lengths greater than the median (>656 characters), adjusting for patient age, sex, race, and ethnicity. The content of subject lines from physician messages was also assessed.
| “Other” Responses | Percentage |
|---|---|
| No further action required | 45.7 |
| Physician called patient | 17.2 |
| Message reply from visit navigator | 16.0 |
| Not handled | 21.1 |
RESULTS
A total of 45 692 physician messages sent to patients on the health care portal between 2017 and 2022 were evaluated ( Table 1 ). During this period, female ophthalmologists sent 52% of the messages to patients, whereas male ophthalmologists sent 48%. Of all the messages sent during this study period, 31.6% (14 466 total messages) were categorized as “other.” Among the messages in the “other” category, 48% were sent by female ophthalmologists and 52% by male ophthalmologists ( Table 1 ). There was a wide range of message subject lines sent by physicians from 2020 to 2022 ( Table 3 ). The most common subject lines in the “other” category were as follows: RE: Appointment Question (18.2%), RE: Nonurgent Medical Question (16.1%), RE: Prescription Question (14.7%), RE: Visit Follow-up Question (11.5%), and RE: Test Results Question (1.8%).
| Physician Message Subject Line | Percentage |
|---|---|
| RE: Appointment Question | 18.2 |
| RE: Nonurgent Medical Question | 16.1 |
| RE: Prescription Question | 14.7 |
| RE: Visit Follow-up Question | 11.5 |
| RE: Test Results Question | 1.8 |
| Appointment | 0.9 |
| RE: Attachment: Medical Records Release | 0.5 |
| RE: Attachment: Photos | 0.5 |
| RE: Attachment: Forms (FMLA, Disability) | 0.4 |
| Combination of remaining subject lines | 35.4 |
Table 1 demonstrates that there were more than 2 times as many messages sent from 2020 to 2022 as from 2017 to 2019 for both female ophthalmologists (16 233 response messages vs 7613 response messages) and male ophthalmologists (14 845 response messages vs 7001 response messages). Female ophthalmologists sent significantly more response messages per patients than male ophthalmologists (mean ± SD: 5.5 ± 2.9 vs 3.0 ± 1.5; P = .04). There was a significant increase in the number of physician messages per patient after the start of 2020 compared with before 2020 for both female ( P = .001) and male ophthalmologists ( P = .0006). There was also a significant increase in the difference in the number of messages per patient based on physician gender ( P = .007). Before 2020, female ophthalmologists sent an average of 1.4 more messages per patient than male ophthalmologists. From 2020 to 2022, female ophthalmologists sent an average of 3.7 more messages per patient.
From 2017 to 2020, the median response length sent by female ophthalmologists was longer than that by male ophthalmologists (672 vs 637 characters; P < .0001) ( Table 4 ). These results were similar when adjusting for patient sex, age, race, and ethnicity ( Table 5 ). The gender difference in median response length ranged from 22 to 58 characters by year, with the largest difference observed in 2020. We were not able to extract response length data from the Penn’s message portal for 2021 and 2022.
