Key points are not available for this paper at this time.
The COVID-19 pandemic triggered an unprecedented expansion of telehealth within the Medicare program, raising the question of whether this surge represented a temporary crisis response or a durable shift in care delivery. This study applied supervised machine learning and explainable artificial intelligence methods to national Medicare Telehealth Trends data to model nonlinear adoption dynamics and evaluate differential stabilization across demographic and geographic groups. Using state by quarter by demographic group observations from 2020 through 2025, random forest and gradient boosting models with temporally aware validation were implemented to characterize surge magnitude, peak timing, and post-surge equilibrium levels. Differential saturation was assessed using model predicted plateau estimates across age, race, dual eligibility status, and state. The results revealed a rapid and synchronous surge across demographic strata followed by stabilization into persistent but heterogeneous equilibrium levels. Age and dual eligibility status were associated with substantial divergence in plateau magnitude, while racial differences were comparatively modest after accounting for temporal and geographic factors. Geographic heterogeneity exceeded demographic variation, indicating that state level structural environments exert the strongest influence on long-term telehealth equilibrium. These findings suggest that telehealth has transitioned from emergency expansion to partial institutional embedding within Medicare preventive care delivery. However, uneven stabilization across states highlights structural disparities that may influence continuity of chronic disease management, behavioral health access, and routine preventive services. Strengthening digital infrastructure and policy alignment may therefore be necessary to ensure that telehealth functions as an equitable and durable component of preventive medicine beyond the pandemic period.
Omotoye et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: