In A, other specialties comprise 10% of total volume. B, Unadjusted data stratified by county-level primary care physician (PCP) supply. Quartile 1 includes 10% of enrollees; quartile 2, 39%; quartile 3, 33%; quartile 4, 18%. C, Unadjusted data stratified by county-level psychiatrist supply. Counties with no psychiatrists included 7% of enrollees; counties with low supply (≤6.2 psychiatrists per 100 000 population [below average among counties with any supply]), 25%; counties with high supply (>6.2 psychiatrists per 100 000 population [above average among counties with any supply]), 67%. D and E, Unadjusted data stratified by state-level telemedicine parity and coverage laws. No mandate for coverage or parity included 68% of enrollees; any mandate for insurer coverage of telemedicine, 26%; full mandate of coverage and reimbursement parity, 6%.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Barnett ML, Ray KN, Souza J, Mehrotra A. Trends in Telemedicine Use in a Large Commercially Insured Population, 2005-2017. JAMA. 2018;320(20):2147–2149. doi:10.1001/jama.2018.12354
Telemedicine may improve access to specialty care, particularly in underserved, rural areas.1 To promote telemedicine adoption, “parity” laws, which mandate coverage and reimbursement for telemedicine, have passed in 32 US states (64%) as of 2016.2 However, little is known about telemedicine adoption nationally among the commercially insured. To address this gap, we examined trends in telemedicine use and its association with regional factors (parity legislation and physician supply) within a large commercial health plan.
We used 2005-2017 data from OptumLabs Data Warehouse, a deidentified claims database for privately insured and Medicare Advantage enrollees in a large, private US health plan. Database enrollees are younger and more concentrated in the South compared with the overall US population. Telemedicine visits were identified using Medicare criteria3 and grouped into 3 categories: telemental health (visits with mental health clinicians, mental health–specific Current Procedural Terminology codes, or primary mental health diagnoses), primary care telemedicine (non–telemental health visits with primary care clinicians), and other telemedicine (all remaining specialist visits).
We estimated growth in telemedicine use from 2005 to 2017 using a regression model with a linear variable for time and log-transformed telemedicine visit volume as the dependent variable. We examined trends in telemedicine use stratified by state parity laws4 and county-level physician supply. Using separate multivariable regression with the dependent variable of telemedicine visits per 1000 enrollees, we examined linear indicators for time interacted with presence of parity law and level of physician supply to estimate differences in trends between categories.
All statistical analyses were performed in R version 3.5.0 (R Project for Statistical Computing), with 2-sided P < .05 considered significant. The Harvard Medical School institutional review board exempted this study from review.
From 2005-2017, there were 383 565 telemedicine visits by 217 851 patients. From 2015-2017, the mean age of telemedicine users was 38.3 years, 63.0% were female, and 83.3% resided in urban areas. Users of primary care telemedicine were younger on average than users of telemental health (38.0 vs 39.7 years, respectively) and were more likely to reside in urban areas (87.1% vs 75.2%, respectively).
Annual telemedicine visits among all members increased from 206 in 2005 (0.020 per 1000 [95% CI, 0.018-0.021]) to 202 374 in 2017 (6.57 per 1000 [95% CI, 6.54-6.60]), an average annual compound growth rate of 52% (95% CI, 45%-61%) (43% [95% CI, 37%-49%] from 2005 to 2014 and 261% [95% CI, 206%-331%] from 2015 to 2017; Figure). Most telemedicine visits were either telemental health (53%) or primary care telemedicine (39%). Primary care telemedicine visits grew 36% annually before 2016 and then increased sharply to 136 366 visits in 2017, while telemental health grew 56% annually to 57 095 visits in 2017. By 2017, primary care telemedicine was the most frequently used form of telemedicine.
Use of telemental health increased significantly faster in counties with no psychiatrists (P < .001 for interaction) (Figure) and in states with comprehensive parity mandates (P = .02 for interaction). In contrast, growth of primary care telemedicine was not associated with primary care physician supply (P = .76 for interaction), and there was a small negative association with comprehensive parity laws (P = .04 for interaction).
Although telemedicine use increased substantially from 2005 to 2017, use was still uncommon by 2017. Use of telemental health grew steadily over this period. In contrast, there was a rapid increase in growth for primary care telemedicine in 2016 and 2017 after coverage for direct-to-consumer telemedicine expanded.5,6 An important limitation is that these data are from a single insurer whose population and policies may not generalize to other populations.
In this sample, telemedicine for subspecialty care beyond mental health was uncommon, and despite the attention given to telemedicine for rural settings, most telemedicine users lived in urban areas. Physician supply appears to be influential for telemental health but not for primary care telemedicine, the brisk adoption of which may reflect consumers seeking convenience rather than reflecting low primary care supply. This evidence suggests that local coverage and reimbursement regulations may have influenced growth of telemental care but not primary care telemedicine.
Accepted for Publication: August 1, 2018.
Corresponding Author: Michael L. Barnett, MD, MS, Department of Health Policy and Management, Harvard T. H. Chan School of Public Health, 677 Huntington Ave, Kresge 411, Boston, MA 02115 (email@example.com).
Author Contributions: Dr Barnett had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Barnett, Ray, Mehrotra.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Barnett.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Barnett, Souza.
Obtained funding: Mehrotra.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Funding/Support: This study was supported by a gift from Melvin Hall and a grant from the National Institutes of Health (NIH) (R01MH112829). Dr Barnett was also supported by a career development award from the National Institute on Aging (K23AG058806). Dr Ray was also supported by a career development award from the NIH (NICHD K23HD088642).
Role of the Funder/Sponsor: The NIH had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Create a personal account or sign in to: