Telemedicine Adoption Before, During, and After COVID-19

The COVID-19 pandemic dramatically accelerated the adoption of telemedicine, also referred to as telehealth, as a vehicle to maintain access to healthcare during lockdowns. However, with the relaxation of lockdown measures and social distancing, telemedicine use appears to have tapered down some, though still definitely much higher compared to pre-COVID levels, indicating that telemedicine has become an integral component of the healthcare landscape. Earlier related research undertaken during the pandemic and the recovery period has examined telemedicine trends using descriptive statistics at the aggregate level, occasionally also using descriptive statistics or bivariate statistical methods to study the difference in telemedicine adoption between univariate segmentations of the population (such as segmentation based solely on gender or solely on age or solely on income). A limited number of studies have estimated multivariate statistical models that control simultaneously for a suite of individual/household sociodemographic determinant variables at once.

Motivated by this discussion, our study aims to contribute to the telemedicine adoption literature by using multivariate econometric models that can identify the nature of (and the factors behind) the shift in telemedicine-associated attitudes in the after-COVID period relative to before- and during-COVID periods. The express interest in this study is to investigate the individual-level factors that affect telemedicine adoption in the after-COVID period. But we use adoption during the before- and during-COVID periods within a longitudinal setting (that is, observing individuals’ telemedicine adoption over time) as a means to control for unobserved individual-level factors that affect telework adoption during the after-COVID period. Doing so lends efficiency in our estimation, as well as allows us to accurately trace the evolution patterns (that is, disentangling state dependence from unobserved individual-level heterogeneity) of telemedicine adoption between the before-COVID and after-COVID periods. In this regard, while the during-COVID adoption tendencies are not really of much interest here (because adoption in this period was significantly impacted by external lockdown regulations), this time point in our longitudinal analysis still contributes in an important way to controlling for unobserved individual-level heterogeneity. In addition to investigating telemedicine adoption tendencies, our methodology also incorporates a multivariate probit model to examine the underlying reasons for both adopting and not adopting telemedicine in the after-COVID period. While existing literature has explored telemedicine adoption motivators and non-adoption concerns using descriptive approaches, limited research has investigated consumer perceptions of telemedicine service. The models proposed in this study are estimated using data from the COVID Future Survey Wave 3 conducted across the U.S. during the period spanning from October and November of 2021.

       The results from the study will contribute significantly to our understanding of telemedicine adoption and its implications, and will provide important insights for multiple sectors, including healthcare, telecommunication, and land use-transportation planning. Regarding land use-transportation planning, by eliminating travel needs for healthcare, the rise of telehealth could influence urban mobility patterns and long-term transportation planning.