Abstract： Objective To explore the inputs and outputs of areas with different anti-HAY prevalence rates on universal childhood vaccination,and to provide a scientific basis for the formulation of the immunization strategy.Methods Since hepatitis A vaccination was scheduled at 12 and 18 months of age for all the healthy children,a single cohort including 1 000 000 individuals was formed in 2009,using the Chinese inactivated vaccine.Decision analysis was used to build Markov-decision tree model.The universal childhood hepatitis A vaccination was compared with nonvaccination group to evaluate the number of symptomatic infection,hospitalization,death,qualityadjusted life years (QALYs) lost,and the incremental cost-utility from the health system and the societal perspectives.Outcomes of the vaccination for the next 70 years were also predicted.The process of analysis was run separately in five regions defined by the anti-HAV prevalence rates (around 50％,50％-69％,70％-79％,80％-89％ and ＞90％ ).Sensitivity analysis was performed to test the stability or reliability of the results,and to identify sensitive variables.Results The study projected that,in the lowest,lower,and intermediate infection regions,the cost and output indicators of universal childhood hepatitis A vaccination were all lower than non-vaccinated group.Universal vaccination could gain QALYs and save both costs from the health systen or the society.In the regions with higher infection rate,the output indicators of universal childhood hepatitis A vaccination were lower than in those non-vaccinated groups,except for the number of death due to hepatitis A,which had a 20 cases of increase.The model also predicted that in the highest infected region,universal vaccination would increase 4 560 814 and 5 840 430 RMB Yuan in the total costs from both the health system and the societies,respectively,when compared to the non-vaccination groups.Universal vaccination would also decrease the numbers of symptomatic infection,hospitalization,and QALYs lost,but would increase 51 deaths due to hepatitis A,and 1507,1929 more RMB Yuan for each QALY gained from the health system and societal respectively,in the regions with highest infection rate.Sensitivity analyses discovered that the infection rate among those susceptible population and the proportion of those who initially under protection but subsequently lost their immunity every year,were the two main sensitive variables in the model.Conclusion Our research discovered that the universal vaccination strategy should be based on the protective period of the vaccine and the anti-HAV prevalence in different endemic areas.