Skip to main content
. 2018 Nov;62:147–164. doi: 10.1016/j.jhealeco.2018.09.010

Fig. 6.

Fig. 6

Sensitivity analysis of main treatment effects of low price (phase 1) and loyalty card (phase 3) with respect to the sufficient supplementation threshold (right axis). Randomization Inference (RI) p-values for each treatment effect estimate are depicted as bars (left axis). The vertical dashed lined indicates the lower bound threshold of 0.13 sachets per day (Suchdev et al., 2016) used to generate the hazard model results reported in the tables.