Gender (in)equality in Human Papilloma Virus (HPV) vaccinations and treatment

The protective (herd) effect of the selective vaccination of pubertal girls against human papillomavirus (HPV) implies a high probability that one of the two partners involved in intercourse is immunised, hence preventing the other from this sexually transmitted infection. The dynamic transmission models used to inform immunisation policy should include consideration of sexual behaviours and population mixing in order to demonstrate an ecological validity, whereby the scenarios modelled remain faithful to the real-life social and cultural context. The primary aim of this review is to test the ecological validity of the universal HPV vaccination cost-effectiveness modelling available in the published literature. The research protocol related to this systematic review has been registered in the International Prospective Register of Systematic Reviews (PROSPERO: CRD42016034145). Our results are consistent with the outcomes of recently published reviews of HPV vaccination modelling approaches. The heterogeneity of outcomes observed in the reviewed studies results from the high degree of sensitivity to boundary conditions and the choice of inputs. In more general terms, sensitivity analyses showed that the vaccine price per vial is one of the factors most relevant to the determination of the incremental cost-effectiveness ratio, and hence of the cost-effectiveness of universal vaccination. All else equal, a vaccine price per vial of €28 would drive all the base-case incremental cost-effectiveness ratio (ICER) values reported by the studies included in the review below an acceptability threshold of $50,000 (in $2015 values). The main original contribution of our review is that all the included models might be affected by a different degree of ecological bias, which implies an inability to reflect the natural demographic and behavioural trends in their outcomes and, consequently, to accurately inform public healthcare policy. In particular, ecological bias have the effect to over-estimate the preference-based outcomes of selective immunisation. A relatively small (15 to 20%) over-estimation of quality-adjusted life years (QALYs) gained with selective immunisation programmes could induce a significant error in the estimate of cost-effectiveness of universal immunisation, by inflating its incremental cost effectiveness ratio (ICER) beyond the acceptability threshold