The relative cost measure was then applied to the estimated national GSK-3 cancer mean direct medical cost of rotavirus [41] to calculate a mean rotavirus cost by geographic and socio-economic setting. Averted medical costs (AvertCostr,q,s) were then estimated for each subpopulation by combining information on the coverage and efficacy of each dose by time period with information on the expected medical cost over time. All costs were adjusted to 2013 US$ (1US$ = 61.8 Indian rupees, INR). equation(6) AvertCostq,r,s=∑d,tCovd,r,q,s,t⋅VacEffd,t⋅MedCostq,r,s,t

The incremental cost of the intervention (IntCostq,r,g) includes vaccine and administration costs. Intervention costs were estimated assuming a baseline vaccine price of $1.25 (77.3 INR) per dose, wastage of 10% and an incremental administration cost of $1.25 per dose [8]. The cost parameters were varied in the sensitivity analysis ( Table 1). The main outcome measure was the incremental cost-effectiveness ratio (ICERq,r), which was estimated for each geographic and economic subpopulation. equation(7) ICERq,r,s=IntCost−AvertCostq,r,sVacBenefitq,r,s A series of analyses were conducted to assess the impact of uncertainty to predicted outcomes. One-way sensitivity analyses were

used to estimate the effect of changes in individual input variables (ranges listed in Table 1). A probabilistic sensitivity analysis (PSA) using Monte Carlo analysis was used to assess the effect of simultaneous changes in multiple input variables. Key input variables were characterized as distributions (Table 1) and a simulation procedure using 10,000 Afatinib iterations was conducted in Crystal Ball [43] to develop a distribution of estimated impact and cost-effectiveness by region. Lastly, specific scenarios were examined including on-time vaccination, equitable coverage, and full coverage. In addition,

we developed an “Equal risk” scenario where we assumed homogeneous RV mortality risk and treatment costs. We used this scenario to approximate the estimated crotamiton benefits and cost-effectiveness ratio if inter and intra region disparities were not considered. Estimated mortality and direct medical costs are shown for each region-quintile sub-group (Fig. 1a) and state-quintile sub-group (Fig. 1b). In the figures, each line represents a different region or state and each of the dots represent different wealth quintiles. Difference in mortality among regions reflects the differences estimated by Morris and colleagues [14]. Within all of the regions, children in poorer households had higher risk of mortality, due to reduced nutritional status and reduced likelihood of receiving rehydration. Conversely, within all regions children in richer households had a higher estimated direct medical cost burden ( Fig. 1a and b). This difference is driven by an increased likelihood of treatment and in particular increased utilization of private hospitals ( Table 2).