Analysis approaches to address treatment nonadherence in pragmatic trials with point-treatment settings: a simulation study

Table 4 Estimated effect of vitamin A supplementation on childhood mortality in a two-arm randomized control trial with 450 villages in northern Sumatra

Method	RD	SE	95% CI
ITT	-2.58	0.93	-4.40, -0.76
Naïve PP	-5.15	0.82	-6.76, -3.53
Naïve AT	-6.47	0.82	-8.08, -4.86
Naïve 2SLS	-3.23	1.16	-5.50, -0.95
Naïve 2SRI	-3.23	1.16	-5.50, -0.96
NPCB ¹	-	-	-5.39, 194.62

Abbreviations: RD: risk difference per 1,000 in the treatment group than the control; SE: standard error, CI: confidence interval; ITT: intention-to-treat; PP: per-protocol; AT: as-treated; 2SLS: two-stage least square; 2SRI: two-stage residual inclusion; NPCB: nonparametric causal bound;
¹The interval estimate from the NPCB is the bound for the average causal estimate per 1,000 than a 95% CI.

ISSN: 1471-2288