Abstract
Aims
Known racial, ethnic, age, and socioeconomic disparities in video telemedicine engagement may widen existing health inequities. We assessed if telemedicine disparities were alleviated among patients of high-video-use providers at a large cardiovascular practice.
Methods and results
All telemedicine visits from 16 March to 31 October 2020 and patient demographics were collected from an administrative database. Providers in the upper quintile of video use were classified as high-video-use providers. Descriptive statistics and a multivariable logistic model were calculated to determine the distribution and predictors of a patient ever having a video visit vs. only phone visits. A total of 24 470 telemedicine visits were conducted among 18 950 patients by 169 providers. Video visits accounted for 48% of visits (52% phone). Among telemedicine visits conducted by high-video-use providers (n = 33), ever video patients were younger (P < 0.001) and included 78% of Black patients vs. 86% of White patients (P < 0.001), 74% of Hispanic patients vs. 86% of non-Hispanic patients (P < 0.001), and 79% of public insurance patients vs. 91% of private insurance patients (P < 0.001). High-video-use provider patients had 9.4 (95% confidence interval 8.4–10.4) times the odds of having video visit compared to low-video-use provider patients.
Conclusion
These results suggest that provider-focused solutions alone, including promoting provider adoption of video visits, may not adequately reduce disparities in telemedicine engagement. Even in the presence of successful clinical infrastructure for telemedicine, individuals of Black race, Hispanic ethnicity, older age, and with public insurance continue to have decreased engagement. To achieve equity in telemedicine, patient-focused design is needed.
Keywords: Telemedicine, Health equity, Healthcare access, Video virtual care
Graphical Abstract
Introduction
Telemedicine offset disruptions to outpatient care during the COVID-19 pandemic.1,2 Studies have identified decreased engagement with video visits among patients of Black race, Hispanic ethnicity, older age, and with public insurance.3–5 We assessed if similar patterns were seen among providers who predominantly adopted video telemedicine during the COVID-19 pandemic.
Methods
We abstracted patient data for all telemedicine visits between 16 March and 31 October 2020 with providers at the Massachusetts General Hospital Corrigan Minehan Heart Center. Providers were split into even quintiles according to their video use (number of video visits among total telemedicine visits). Providers in the upper quintile of video use were classified as high-video-use providers; all other clinicians were classified as low-video-use providers. Patients who saw high- and low-video-use providers were excluded (580/18 950, 3%). The primary outcome was whether a patient had one or more video visits (ever video) or only phone visits (phone) during the study period.
Patient characteristics were compared by provider video-use and outcome using Chi-square and Student’s t-tests. Difference in percent video use was calculated between patient subgroups with respect to a reference group across quintiles of provider video use to identify potential dose–response relationships. To identify predictors of ever video (vs. phone) visits among patients, we calculated a logistic model including patients’ age, sex, insurance, race and ethnicity (self-reported), residence type (urban vs. suburban-rural) by ZIP code, activation of MyChart—our institution’s online patient portal—and provider video-use (high vs. low).
All analyses were conducted using R.6 The Mass General Brigham and Yale University Institutional Review Boards exempted this study from review.
Results
Our centre’s 169 providers conducted 33 650 visits for 24 562 patients. Telemedicine visits accounted for 73% of visits of which 48% were video visits (52% phone). Video use was highly variable among providers with a mean use of 44% (standard deviation ±29%). Thirty-three clinicians (24%) were high-video-use providers, conducting ≥70% of their visits via video. High-video-use providers saw 4228 patients during 5084 telemedicine visits with a mean video use of 87% (±10%). Low-video-use providers (n = 136) had a mean video use of 34% (±22%) seeing 14 722 patients during 19 386 telemedicine visits.
Ever video use was higher among patients of high-video-use providers vs. low-video-use providers (86% vs. 40%) (Table 1). High-video-use providers saw more female and White patients with private insurance and an activated MyChart (all P < 0.01). Among high-video-use provider patients, those with ever having a video visit were younger (P < 0.001) and included 78% of Black patients vs. 86% of White patients (P < 0.001), 74% of Hispanic patients vs. 86% of non-Hispanic patients (P < 0.001), and 79% of public insurance patients vs. 91% of private insurance patients (P < 0.001). Patients with a video visit more often had an activated MyChart (91% vs. 76% without an activated account, P < 0.001). Similar patterns were seen among low-video-use provider patients except for male patients and suburban-rural patients having greater video use.
Table 1.
Sample characteristics by provider video-use level and outcomea,b
| High-video-use providers |
Low-video-use providers |
|||||
|---|---|---|---|---|---|---|
| Ever video (n = 3135) | Phone (n = 513) | P-valuec | Ever video (n = 5925) | Phone (n = 8797) | P-valuec | |
| Age (mean ±SD) | 60 (±17) | 69 (±16) | <0.001 | 63 (±16) | 70 (±14) | <0.001 |
| Sex | 0.93 | 0.02 | ||||
| Male | 86% (1660) | 14% (270) | 41% (3507) | 59% (5037) | ||
| Female | 86% (1475) | 14% (243) | 39% (2418) | 61% (3760) | ||
| Race | 0.01 | <0.001 | ||||
| White | 86% (2815) | 14% (443) | 41% (5239) | 59% (7643) | ||
| Black | 78% (60) | 22% (17) | 35% (165) | 65% (312) | ||
| Asian | 88% (98) | 11% (13) | 44% (205) | 56% (265) | ||
| Other or missing | 80% (162) | 20% (40) | 35% (316) | 65% (577) | ||
| Ethnicity | <0.01 | <0.001 | ||||
| Non-Hispanic | 86% (2711) | 14% (440) | 41% (5287) | 59% (7684) | ||
| Hispanic | 74% (69) | 26% (24) | 30% (150) | 70% (351) | ||
| Other or missing | 88% (352) | 12% (49) | 39% (487) | 61% (760) | ||
| Payor | <0.001 | <0.001 | ||||
| Private | 91% (1850) | 8% (171) | 50% (3215) | 50% (3221) | ||
| Public | 79% (1263) | 21% (339) | 32% (2683) | 67% (5542) | ||
| None | 88% (22) | 12% (3) | 44% (27) | 56% (34) | ||
| MyChart | <0.001 | <0.001 | ||||
| Activated | 91% (2215) | 9% (227) | 50% (4619) | 50% (4671) | ||
| Not activated | 76% (920) | 24% (286) | 24% (1306) | 76% (4126) | ||
| Residence | 0.98 | 0.02 | ||||
| Urban | 86% (2984) | 14% (489) | 40% (5646) | 60% (8457) | ||
| Suburban-rural | 86% (145) | 14% (23) | 45% (270) | 55% (334) | ||
Row percent (count) may not sum to 100% due to rounding.
Five hundred and eighty patients were seen by both types of providers and were excluded.
χ2 and Student’s t-tests.
After controlling for patient characteristics, high-video-use provider patients had 9.4 (95% confidence interval 8.4–10.4) times the odds of having a video visit compared to low-video-use provider patients.
The disparity in video use among Black patients compared to White patients followed an overall dose–response where increasingly higher quintiles of provider video use corresponded with a larger gap in video use (Figure 1). All other subgroups demonstrated general heterogeneity across quintiles.
Figure 1.
Difference in ever video use by patient characteristics with respect to reference groups across quintiles of provider video use. Points closest to the dashed red line indicate more similarity in video use to the reference group. Patients having visits with multiple providers were randomized to a single quintile; therefore, all 18 950 patients were included. Quintile 1 (Q1) contained 34 providers with a combined video use of 6% (151 video visits of 2602 telemedicine visits), Q2 had 34 providers with 25% video use (1454/5713), Q3 had 34 providers with 43% video use (2175/5054), Q4 had 33 providers with 61% video use (3615/5928), and Q5 had 34 providers with 85% video use (4394/5173).
Discussion
We found that even among providers with high video proficiency, the digital access divide persisted and even widened for Black patients compared to White patients. Suboptimal and disparate video engagement in subgroups identified in other studies3,4 may not be surmountable by increased provider adoption of telemedicine alone. While providers who largely adopted video virtual care had increased video engagement among all patient groups, patient selection or structural determinants may have limited parity in video use across subgroups.7 Patients with an activated MyChart were more likely to engage in video visits, suggesting that digital literacy influences video use. As telemedicine becomes more accessible and available from providers, certain populations will require personalized health and digital literacy education to achieve equitable access. Limitations of this study include generalizability to other patient populations or regions as we studied a medical subspecialty in a state with high overall rates of telemedicine use.2
Acknowledgements
The authors would like to thank Jiye Kwon at Yale University for programming consultation and Theresa Mills at Massachusetts General Hospital for curating the data.
Data availability
Data will be made available upon reasonable request.
Conflict of interest: none declared.
References
- 1. Koonin LM, Hoots B, Tsang CA, Leroy Z, Farris K, Jolly T, Antall P, McCabe B, Zelis CBR, Tong I, Harris AM. Trends in the use of telehealth during the emergence of the COVID-19 pandemic—United States, January-March 2020. MMWR Morb Mortal Wkly Rep 2020;69:1595–1599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Patel SY, Mehrotra A, Huskamp HA, Uscher-Pines L, Ganguli I, Barnett ML. Trends in outpatient care delivery and telemedicine during the COVID-19 pandemic in the US. JAMA Intern Med 2020;181:388–391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Ye S, Kronish I, Fleck E, Fleischut P, Homma S, Masini D, Moise N. Telemedicine expansion during the COVID-19 pandemic and the potential for technology-driven disparities. J Gen Intern Med 2021;36:256–258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Eberly LA, Kallan MJ, Julien HM, Haynes N, Khatana SAM, Nathan AS, Snider C, Chokshi NP, Eneanya ND, Takvorian SU, Anastos-Wallen R, Chaiyachati K, Ambrose M, O'Quinn R, Seigerman M, Goldberg LR, Leri D, Choi K, Gitelman Y, Kolansky DM, Cappola TP, Ferrari VA, Hanson CW, Deleener ME, Adusumalli S. Patient characteristics associated with telemedicine access for primary and specialty ambulatory care during the COVID-19 pandemic. JAMA Netw Open 2020;3:e2031640. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Rodriguez JA, Betancourt JR, Sequist TD, Ganguli I. Differences in the use of telephone and video telemedicine visits during the COVID-19 pandemic. Am J Manag Care 2021;27:21–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2020. [Google Scholar]
- 7. Reed ME, Huang J, Graetz I, Lee C, Muelly E, Kennedy C, Kim E. Patient characteristics associated with choosing a telemedicine visit vs office visit with the same primary care clinicians. JAMA Netw Open 2020;3:e205873. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data will be made available upon reasonable request.
Conflict of interest: none declared.