Table 5 Observed power for different values of the treatment effect (β1) and of sample size, with a strong confounder effect (β2 = − 1), for 3 logistic regression models: unadjusted for confounder, adjusted in sample-based model, and adjusted in true model
Sample size | Unadjusted analysis | Adjusted, sample-based model | Adjusted, true model | Unadjusted analysis | Adjusted, sample-based model | Adjusted, true model |
|---|---|---|---|---|---|---|
Treatment effect β1 = 0.5, continuous confounder | Treatment effect β1 = 0.5, binary confounder | |||||
2 × 50 | 0.24 | 0.23 | 0.23 | 0.24 | 0.22 | 0.22 |
2 × 100 | 0.40 | 0.40 | 0.40 | 0.41 | 0.40 | 0.40 |
2 × 200 | 0.66 | 0.68 | 0.68 | 0.67 | 0.68 | 0.68 |
2 × 500 | 0.96 | 0.97 | 0.97 | 0.96 | 0.97 | 0.97 |
2 × 1000 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
Treatment effect β1 = 1, continuous confounder | Treatment effect β1 = 1, binary confounder | |||||
2 × 50 | 0.64 | 0.65 | 0.65 | 0.67 | 0.67 | 0.67 |
2 × 100 | 0.90 | 0.91 | 0.91 | 0.92 | 0.92 | 0.92 |
2 × 200 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
2 × 500 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
2 × 1000 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |