Table 3. A Comparison of Bleomycin/Vincristine versus Vincristine Monotherapy on the Probability of Tumor Improvement at 40–42 Weeks*.
Risk Ratio | 95% Confidence Interval | ||
A. Comparison of Tumor Improvement versus non-improvement (stable, progressive) (N = 346) | |||
Unadjusted | 2.25 | (1.47, 3.44) | |
Adjusted | 2.46 | (1.57, 3.86) | |
B. Comparison of tumor improvement versus non-improvement (stable, progressive, dead, default) (N = 354) | |||
Unadjusted | 1.88 | (1.16, 3.03) | |
Adjusted | 2.06 | (1.24, 3.41) | |
C. Comparison of tumor improvement versus non-improvement (stable, progressive, dead, default, missing) (N = 449) | |||
Unadjusted | 1.42 | (0.81, 2.51) | |
Adjusted | 1.48 | (0.85, 2.60) |
*Comparison of the probability of tumor improvement versus tumor non-improvement among patients on Bleomycin/Vincristine versus those on Vincristine monotherapy. In model (A), only patients with recorded tumor information are included. In model (B), all patients who were dead or lost/defaulted were also included and categorized with tumor non-improvement. In model C, all patients who were dead, lost/defaulted, or missing were classified as having tumor non-improvement. All adjusted models control for age and baseline CD4 count.