Systematic review of incremental non-vaccine cost estimates used in cost-effectiveness analysis on the introduction of rotavirus and pneumococcal vaccines

• We review CEA literature about pneumococcal and rotavirus vaccine introduction.
• Approaches used to consider costs other than vaccine procurement were evaluated.
• Many vaccine CEAs do not consider program costs or use poorly supported assumptions.

ObjectiveTo review the approaches used in the cost-effectiveness analysis (CEAs) literature to estimate the cost of expanded program on immunization (EPI) activities, other than vaccine purchase, for rotavirus and pneumococcal vaccines.MethodsA systematic review in PubMed and NHS EED databases of rotavirus and pneumococcal vaccines CEAs was done. Selected articles were read and information on how EPI costs were calculated was extracted. EPI costing approaches were classified according to the method or assumption used for estimation.ResultsSeventy-nine studies that evaluated cost effectiveness of rotavirus (n = 43) or pneumococcal (n = 36) vaccines were identified. In general, there are few details on how EPI costs other than vaccine procurement were estimated. While 30 studies used some measurement of that cost, only one study on pneumococcal vaccine used a primary cost evaluation (bottom-up costing analysis) and one study used a costing tool. Twenty-seven studies (17 on rotavirus and 10 on pneumococcal vaccine) assumed the non-vaccine costs. Five studies made no reference to additional costs. Fourteen studies (9 rotavirus and 5 pneumococcal) did not consider any additional EPI cost beyond vaccine procurement. For rotavirus studies, the median for non-vaccine cost per dose was US$0.74 in developing countries and US$6.39 in developed countries. For pneumococcal vaccines, the median for non-vaccine cost per dose was US$1.27 in developing countries and US$8.71 in developed countries.ConclusionsMany pneumococcal (52.8%) and rotavirus (60.4%) cost-effectiveness analyses did not consider additional EPI costs or used poorly supported assumptions. Ignoring EPI costs in addition to those for vaccine procurement in CEA analysis of new vaccines may lead to significant errors in the estimations of ICERs since several factors like personnel, cold chain, or social mobilization can be substantially affected by the introduction of new vaccines.