Age, period, and cohort effects in adult lifetime asthma prevalence in California: an application of hierarchical age-period-cohort analysis

Purpose: Using 27 years of survey data, the contributions of age, period, and cohort effects on the increase in adult lifetime asthma prevalence in California were examined. Methods: Lifetime asthma diagnosis for adults was assessed in 1984-1992 and 1995-2011 through the California Behavioral Risk Factor Surveillance System, an annual, cross-sectional, population-based survey (n = 144,100). Using date of survey and date of birth, we classified 18,305 adult respondents with lifetime asthma into 7 age groups, 6 periods, and 17 cohorts. Using hierarchical; cross-classified random effects models, birth cohort, period, and age patterns in adult lifetime asthma prevalence were analyzed. Results: After adjusting for sex, ethnicity, education, and smoking, age effects peak in young adulthood, flatten from 40 to 60 years old, and then decrease in older adulthood. A significant positive trend in asthma prevalence was observed in the two earliest survey periods (1984-1993; P value < .0001). Survey period trends appear to flatten beginning in 2004. Although the overall birth cohort effect was statistically significant, the magnitude of the effect for each birth cohort category was small (P value = .0005). Conclusions: We observed that strong age and period effects have been driving the increase in lifetime asthma prevalence in California over the past 3 decades. (C) 2016 Elsevier Inc. All rights reserved.