Outcome | Pre-APM | Post-APM | Post-ACA | Relative Rates/Difference in Difference | |
---|---|---|---|---|---|
Post-APM vs Pre-APM | Post-ACA vs Post-APM | ||||
Proportion of appointment searches indicating same day availability* | |||||
APM clinics | 0.25 (0.20, 0.31) | 0.33 (0.29,0.39) | 0.27 (0.24,0.31) | 1.33 (1.21,1.48) | 0.82 (0.71,0.94) |
Non-APM comparison clinics | 0.26 (0.23,0.30) | 0.29 (0.24,0.36) | 0.25 (0.22,0.29) | 1.11 (0.97, 1.27) | 0.85 (0.70,1.03) |
APM vs non-APM Comparison | 0.95 | 1.14 | 1.10 | 1.20 | 0.96 |
Relative ratio | (0.74,1.22) | (0.88,1.49) | (0.88,1.37) | (1.02, 1.42) | (0.76, 1.22) |
Mean days to 3rd next available appointment‡ | |||||
APM clinics | 2.96 (2.36, 3.55) | 2.24 (1.61, 2.86) | 2.87 (2.2, 3.53) | −0.72 (−1.34, −0.11) | 0.63 (0.27, 0.99) |
Non-APM comparison clinics | 3.96 (2.86, 5.07) | 4.23 (3.05, 5.41) | 4.66 (3.12, 6.2) | 0.27 (−0.51, 1.05) | 0.43 (−1.07, 1.93) |
APM vs non-APM comparison difference in means | −1.00 (−2.26, 0.25) | −2.00 (−3.39, −0.61) | −1.79 (−3.48, −0.1) | − (−1.99, 0) | 0.21 (−1.33, 1.75) |
Overbooks per month/per 100 appointment slots | |||||
APM clinics | 5.29 (4.43, 6.31) | 5.55 (4.41, 6.99) | 5.83 (4, 8.48) | 1.05 (0.81, 1.36) | 1.05 (0.8, 1.37) |
Non-APM comparison clinics | 5.96 (4.13, 8.59) | 6.59 (4.88, 8.89) | 5.90 (4.17, 8.35) | 1.11 (0.83, 1.47) | 0.90 (0.72, 1.12) |
APM vs non-APM comparison relative rate | 0.89 (0.6, 1.32) | 0.84 (0.57, 1.25) | 0.99 (0.58, 1.67) | 0.95 (0.65, 1.39) | 1.17 (0.83, 1.66) |
No-shows/late cancellation per month per 100 appointment slots*[1] | |||||
APM clinics | 7.04 (6.41, 7.73) | 6.69 (6.27, 7.14) | 6.57 (6.19, 6.97) | 0.95 (0.84, 1.07) | 0.98 (0.93, 1.04) |
Non-APM comparison clinics | 5.80 (4.95, 6.8) | 5.68 (5.19, 6.23) | 5.70 (5.3, 6.13) | 0.98 (0.87, 1.1) | 1.00 (0.95, 1.06) |
APM vs non-APM comparison | 1.21 | 1.18 | 1.15 | 0.97 | 0.98 |
Relative rate | (1.03, 1.43) | (1.06, 1.31) | (1.06, 1.25) | (0.82, 1.15) | (0.91, 1.06) |
ACA, Affordable Care Act; APM, Alternative Payment Methodology.
↵* Proportions were determined from logistic regression models of APM status × month, utilizing a robust sandwich estimator to account for correlations within clinics and were adjusted for percent Hispanic, percent Medicaid insurance and percent of panel in a high-needs age group. There were no significant trends in proportion within the pre-period.
† Rates were determined from Poisson regression models of APM status × month, utilizing a robust sandwich estimator to account for correlations within clinics and were adjusted for percent Hispanic, percent Medicaid insurance and percent of panel in a high-needs age group. There were no significant trends in rates within the pre-period.
↵‡ There was a positive trend in outcome within the pre-period in APM clinic and a downward trends within the pre-period in comparison clinics.