Fig. 3:

Sensitivity analysis for association between exposure to ASHA and utilization of maternity services using a linear probability model. **p < 0.05; **p < 0.01. For each outcome, the point estimates and 95% confidence intervals (indicated by the whiskers) correspond to program effect sizes estimated using difference-in-differences models fitted using ordinary least squares regression, controlling for a range of potentially confounding variables. Model 1: S1 exposure definition (cluster-level intensity of ASHA exposure measured as the number of women who reported exposure to an ASHA in a cluster during the last birth divided by the total number of eligible women who had a birth in 6 years preceding the survey in the cluster); Model 2: S1, for clusters > 2; Model 3: S1, for clusters > 3; Model 4: S3 exposure definition (Exposure is coded based on individual women’s response to whether they received ASHA services—0 for No, 1 for Yes); Model 5: S2 exposure definition (All women in clusters in which at least one woman reports seeing an ASHA received a value of “exposed” (i.e., 1)). All models control for cluster-level fixed effects