Ontology highlight
ABSTRACT: Purpose
Adult secondhand tobacco smoke (SHS) exposure is related to stroke and coronary heart disease (CHD) risk, but long-term effects are less clear. We evaluated whether childhood SHS exposure affects subsequent stroke or CHD risk among adult black and white never-smokers followed for stroke and CHD.Methods
In this prospective cohort study, inverse probability weights were calculated to correct for bias due to attrition and survey nonresponse. Cox proportional hazards models were used to assess hazard ratios and 95% confidence intervals for stroke or CHD, separately, by number of childhood household smokers.Results
Of 13,142 eligible participants, 6136 had childhood SHS exposure assessed. Baseline mean (SD) age was 63.5 (9.0), 65% were female, 30% black, 46% reported 0 childhood household smokers, 36% reported 1, and 18% reported 2+. In 60,649 person-years, 174 strokes were observed (2.9% of participants), and in 45,195 person-years, 114 CHD events were observed (2.1% of participants). The weighted and adjusted hazard ratios (95% confidence intervals) of stroke for 2+ versus 0 childhood household smokers was 1.66 (1.29-2.13) and was 1.15 (0.82-1.59) for CHD.Conclusions
We observed a significant association between childhood SHS exposure and stroke, but not CHD, after age 45 years and adjusting for missing information.
SUBMITTER: Pistilli M
PROVIDER: S-EPMC6441374 | biostudies-literature | 2019 Apr
REPOSITORIES: biostudies-literature
Pistilli Maxwell M Howard Virginia J VJ Safford Monika M MM Lee Brian K BK Lovasi Gina S GS Cushman Mary M Malek Angela M AM McClure Leslie A LA
Annals of epidemiology 20190205
<h4>Purpose</h4>Adult secondhand tobacco smoke (SHS) exposure is related to stroke and coronary heart disease (CHD) risk, but long-term effects are less clear. We evaluated whether childhood SHS exposure affects subsequent stroke or CHD risk among adult black and white never-smokers followed for stroke and CHD.<h4>Methods</h4>In this prospective cohort study, inverse probability weights were calculated to correct for bias due to attrition and survey nonresponse. Cox proportional hazards models w ...[more]