New indicators for the evaluation of community policies based on period and cohort effects in cerebrovascular disease mortality rates.

Introduction: Countermeasures against cerebrovascular diseases (CVD) are one of the important health policies in Japan. This study proposes new indicators that are based on period and cohort effects in CVD mortality rates. The main aim of the study is to contribute to community diagnosis with the existing policies.

Methods: CVD mortality rates for all prefectures in Japan were analyzed according to age, time period, and cohort effects, using the Bayesian Poisson age-period-cohort model. Several indicators were extracted based on the principal component analysis of the estimates of the effects.

Results: Two indicators named the change-in-magnitude and the time-of-decrease were extracted from estimates of the period effects, and three indicators named the change-cohort-of-improvement were extracted from estimates of the cohort effects. These were considered to be related to the countermeasures against CVD.

Conclusion: Under the assumption that the new indicators reflect the result of past policies, the new indicators allow us to evaluate the validity of past policies, and to suggest the necessity for improvement in the conventional policies.