Table 3.
Results of Logistic Repeated Measures Mixed Models for Differences in the Proportion of Individuals That Met Algorithm-Based Screening Criteria in the Pre- and Post-implementation Periods
| Model A (no interactions) N = 99,043 BIC = 119,319.8 |
Model B (with interactions) N = 99,043 BIC = 118,692.7 |
|
|---|---|---|
| Fixed effects* | ||
| Time (post- vs. pre-implementation) | 3.36 (3.23, 3.51) | 6.49 (5.97, 7.05) |
| Age | ||
| 40–49 | Reference | Reference |
| 50–69 | 6.10 (5.74, 6.47) | 9.50 (8.80, 10.26) |
| 70–75 | 3.72 (3.46, 4.00) | 5.67 (5.16, 6.27) |
| Race | ||
| Caucasian/White | Reference | Reference |
| Black/African American (AA) | 1.46 (1.38, 1.54) | 1.57 (1.47, 1.69) |
| Asian | 0.96 (0.86, 1.07) | 0.79 (0.68, 0.91) |
| Other | 0.93 (0.81, 1.06) | 0.93 (0.78, 1.12) |
| Not reported/declined/unavailable (NR) | 0.84 (0.73, 0.97) | 0.84 (0.69, 1.02) |
| Time (post) vs. Black/AA | -- | 0.87 (0.80, 0.95) |
| Time (post) vs. Asian | -- | 1.46 (1.22, 1.75) |
| Time (post) vs. other | -- | 0.99 (0.79, 1.25) |
| Time (post) vs. NR | -- | 0.99 0.79, 1.26) |
| Time (post) vs. age (50–69) | 0.43 (0.39, 0.46) | |
| Time (post) vs. age (70–75) | 0.45 (0.40, 0.51) | |
| Random effects† | ||
| Clinic (N = 26) | 0.37/0.61 | 0.39/0.62 |
| Subject within clinic (N = 64,634) | 2.43/1.56 | 2.55/1.60 |
| Likelihood ratio tests | ||
| Race | p < 0.0001 | -- |
| Age (40–59, 60–69, 70–75) | p < 0.0001 | -- |
| Time × Race | -- | p < 0.0001 |
| Time × Age (40–59, 60–69, 70–75) | -- | p < 0.0001 |
*Fixed effects presented as odds ratio (95% confidence interval)
†Random effects presented as variance/standard deviation