Telemedicine impact on patient disparities and physician practice patterns in cancer rehabilitation: A multicenter retrospective study.

Objective: To examine the impact of telemedicine on demographic and practice patterns between outpatients receiving virtual versus in-person cancer rehabilitation physiatry care.

Design: Multicenter retrospective study.

Setting: Outpatient cancer rehabilitation physiatry clinics at four academic medical centers in the United States.

Patients: Patients with cancer diagnoses or history of cancer diagnosis.

Interventions: Cancer rehabilitation physiatry encounters.

Main outcome measures: Visit mode (in-person, telemedicine); disparities variables (age, race, and gender) by visit mode, and practice interventions (imaging, medications, procedures, other orders, and orders of any type) by visit mode.

Results: Among a total of 7004 encounters, 2687 unique patients were found. In-person participants were significantly older than the average telemedicine participant (mean 62.9 vs. 60.7 years; p < .001). A race effect was seen (p = .037) with individuals reporting as Asian or other being more likely to have telemedicine encounters. No gender disparities were seen. Using a random visit analysis model to compare populations receiving in-person versus telemedicine care, a slight majority (53%) of follow-up visits were via telemedicine, versus 40% of new patient visits (p < .001). No significant differences were seen in medication prescribing frequency (38.9% telemedicine vs. 36.7% in-person, adjusted relative risk [RR]: 0.988, confidence interval [CI]: 0.73-1.34; p = .988) or imaging frequency (2.4% telemedicine vs. 7.6%; adjusted RR: 0.784, CI: 0.44-1.39; p = .408) between telemedicine versus in-person visit types. Other orders were significantly less likely to be placed during telemedicine than in-person visits (19.9% telemedicine vs. 28.6% in-person; adjusted RR: 0.623, CI: 0.45-0.86, p = .004). Order(s) of any type were placed in 54% of visits (52% telemedicine vs. 56% in-person; adjusted RR: 0.92 for telemedicine, CI: 0.83-1.01, p = .082).

Conclusions: Telemedicine has been integrated into cancer rehabilitation physiatry practices and appears to be conducive for placing many types of orders, especially medications. Age was found to be the only major demographic difference between in-person and telehealth patients.