Parent versus child reporting of tobacco smoke exposure at home and in the car.

AIMS: To compare self-reported exposure to tobacco smoke in the home or in cars between parents and their pre-adolescent children.
METHODS: We analysed data on self-reported exposure to secondhand smoke from 3,645 matched pairs of children at baseline (aged between 10 and 13 years) and their parents whether smokers or not, who were participants in Keeping Kids Smokefree (KKS), a community-based study in South Auckland, New Zealand from 2007-2009. The study aimed to reduce children's smoking initiation through parental behaviour change. The responses of the parent-child pairs were analysed using proportions, Kappa scores, and McNemar's Chi-squared test. Additionally, 679 children were biochemically tested for smoking exposure using exhaled carbon monoxide.
RESULTS: There was approximately a 30% discordance between the self-reports of children and their parents, with parents reporting less smoking in homes or cars than their children. Kappa scores for parent-child agreement by ethnicity ranged from 0.15 to 0.41 for smoking at home and 0.17 to 0.54 for smoking in cars. Biochemical testing suggested that around 30% of children had been exposed to secondhand smoke, corroborating their self-reported proportion of 37% (baseline in the home) whereas few parents (11%) reported smoking in home or cars.
CONCLUSION: Parents were significantly less likely than children to report smoking inside the home or car. Biochemical testing indicated that children's reporting is more accurate. This has implications for future studies relying on self-reporting by children and/or their caregivers.