Table 2.

ATT Estimates of the Placebo Treatment Effect (Versus Other Placebo Treatment (i.e., Expected True Treatment Effect 0)) on EDSS Score at 36 Weeks in Patients With Relapsing-Remitting Multiple Sclerosis, on an Absolute Difference Scale, Derived Using Synthetic Data

Estimator	Estimation Time, seconds a	Point Estimate	95% CI a
Naive	2	−0.33	−0.57, −0.30
Nearest-neighbor matching	29	0.02	−0.12, 0.10
Optimal full matching	1,897	0.07	−0.11, 0.13
Inverse probability weighting	6	0.06	−0.08, 0.13
Target distribution weighting	4	0.04	−0.09, 0.10

Abbreviations: ATT, average treatment effect in the treated population; CI, confidence interval; EDSS, Expanded Disability Status Scale.

^a Results are based on 1,000 bootstrap samples (nonparametric 95% CI: 2.5th and 97.5th percentiles).