Table 3. Difference-in-Differences Estimate for the Association of High Telemedicine Adoption With Rates of Acute, Chronic, and Aggregate ACSC Visits per 1000 Patients per Yeara.
| Tertile of telehealth use | Visits per 1000 patients per year | Difference in differences (95% CI)b | ||
|---|---|---|---|---|
| Prepandemic period (June to September 2019) | Postpandemic period (June to September 2020) | Difference | ||
| All ACSC visitsc | ||||
| Low (reference) | 24.30 | 14.93 | −9.37 | NA |
| Medium | 23.94 | 15.26 | −8.68 | 0.69 (−0.93 to 2.31) |
| High | 27.48 | 20.21 | −7.28 | 2.10 (0.22 to 3.97) |
| Acute ACSC composite | ||||
| Low (reference) | 16.28 | 10.48 | −5.79 | NA |
| Medium | 15.91 | 10.63 | −5.27 | 0.52 (−0.85 to 1.89) |
| High | 17.90 | 12.98 | −4.92 | 0.87 (−0.50 to 2.24) |
| Chronic ACSC composite | ||||
| Low (reference) | 8.57 | 4.67 | −3.90 | NA |
| Medium | 8.55 | 4.86 | −3.69 | 0.21 (−0.64 to 1.07) |
| High | 10.14 | 7.44 | −2.69 | 1.21 (−0.22 to 2.64) |
Abbreviations: ACSC, ambulatory care–sensitive condition; NA, not applicable.
a
Data are from an analysis of Blue Cross Blue Shield of Michigan claims for 2019 to 2020.
b
Estimated using average marginal effects.
c
Ambulatory care–sensitive condition visits per 1000 patients per year estimated using estimated margins.