Objectives: Life course exposure to passive smoke may predict health, but there are few validated measures. We tested the reliability and validity of a retrospective life course passive smoking questionnaire.

Methods: Participants from the third follow-up of the Childhood Determinants of Adult Health study (2014-2019, ages 36-49 years) retrospectively reported mother/father/other household member smoking when living at home during childhood, including duration (years) and smoking location (never/sometimes/always inside house). The severity of exposure index (SEI; sum of mother/father/other years smoked multiplied by smoking location), cumulative years of exposure (CYE; sum of mother/father/other years), and total household smokers (THS) were derived. The reliability of retrospective passive smoking reports was examined with intraclass correlation coefficients (ICCs) using household smoking reported 34 years earlier in 1985 by participants when aged 7-15 years. Construct validity was examined by correlating retrospective passive smoking with participants' smoking in adulthood and lung function in childhood and adulthood.

Results: Among 2082 participants (mean±standard deviation [SD], 45.0±2.5 years; 55.2% females), THS ranged from 0 to 5 (mean± SD, 0.9±1.0), CYE ranged from 0 to 106 (mean±SD, 10.5±13.9), and SEI ranged from 0 to 318 (mean±SD, 24.4±36.0). Retrospective measures showed moderate agreement with total household smokers reported in childhood (ICC, 0.58 to 0.62). The retrospective measures were weakly but significantly (p<0.05) correlated with participants' smoking (r=0.13 to 0.15) and lung function (r= -0.05 to -0.06).

Conclusions: The retrospective passive smoking questionnaire showed reasonable reliability and validity. This measure may be useful for epidemiological studies.