Table 3.
Results of Meta-analysis for All Studies and Subgroups of Studies
| Grouping | OR (95% CI) | Test for Heterogeneity (χ2); Estimated Random Effects Variance (REV) |
| All studies (n = 52) | 1.49 (1.31, 1.69) | 116.39, df = 51, P = 0; REV = 0.1 |
| Studies with at least 30 patients per arm | 1.46 (1.27, 1.66) | 106.02, df = 42, P = 0, REV = 0.1 |
| Studies with at least 50 patients per arm | 1.43 (1.24, 1.64) | 90.14, df = 32, P = 0, REV = 0.1 |
| Studies with at least 100 patients per arm | 1.38 (1.11, 1.71) | 90.14, df = 15, P = 0, REV = 0.15 |
| Studies not using excessive doses for first- and second-generation drugs (n = 21) | 1.34 (1.14, 1.57) | 30.72, df = 20, P = .059; REV = 0.04 |
| Clozapine (n = 3) | 0.97 (0.49, 1.92) | 3.81, df = 2, P = .149; REV = 0.17 |
| Olanzapine (n = 8) | 1.74 (1.38, 2.2) | 1.57, df = 7, P = .98; REV = 0 |
| Quetiapine (n = 3) | 1.04 (0.58, 1.87) | 6.84, df = 2, P = .03; REV = 0.17 |
| Risperidone (n = 5) | 1.30 (0.88, 1.93) | 15.3, df = 4, P = .004; REV = 0.13 |
| Ziprasidone (n = 2) | 0.93 (0.65, 1.34) | 1.26, df = 1, P = .26; REV = 0.01 |
| Flexible dose studies not using excessive dosing (n = 17) | 1.37 (1.12, 1.68) | 27.77, df = 16, P = .03; REV = 0.07 |
| Studies of patients with symptom exacerbation (n = 40) | 1.40 (1.26, 1.57) | 48.64, df = 39, P = .14; REV = 0.02 |
| Studies of nonresponder patients (n = 10) | 1.53 (1.01, 2.32) | 47.77, df = 9, P = 0; REV = 0.31 |
| Inpatient studies (n = 23) | 1.54 (1.3, 1.84) | 38.4, df = 22, P = .02; REV = 0.06 |
| Outpatient studies (n = 29) | 1.45 (1.2, 1.74) | 77.69, df = 28, P = 0; REV = 0.13 |