Associations between behavioural risk factors and smoking, heavy smoking and future smoking among an Australian population-based sample

Abstract

Introduction Tobacco smoking co-occurs with behavioural risk factors including diet, alcohol use and obesity. However, the association between behavioural risk factors and heavy smoking (> 20 cig/day) compared to light–moderate smoking is unknown. The link between behavioural risk factors and future smoking for both ex and current smokers is also unknown. This study sought to examine these relationships. It is hypothesised that behavioural risk factors will be more strongly associated with heavy smoking.
Method Data from Wave 7 (2007) of the Household and Labour Dynamics in Australia (HILDA) survey was analysed using logistic regression to determine relationships between diet (fruit and vegetable consumption, and unhealthy diet choices), alcohol consumption, obesity and physical activity with light–moderate smoking and heavy smoking. The association between these risk factors and future smoking (2008) was assessed for current and ex-smokers (2007).
Results Obese respondents were less likely to be light/moderate smokers (RRR: 0.53; 95% CI: 0.43, 0.66) but not heavy smokers. Those who consume confectionary weekly were less likely to be light/moderate smokers (RRR: 0.73; 95% CI: 0.61, 0.87), but not heavy smokers. Smokers in 2007 were more likely to continue smoking in 2008 if they consumed 1–4 drinks per occasion (OR: 2.52; 95% CI: 1.13, 5.62). Ex-smokers in 2007 were less likely to relapse in 2008 if they consumed recommended levels of both fruit and vegetables (OR: 0.31; CI: 0.10, 0.91).
Conclusion The relationships between heavy smoking and behavioural risk factors differ from moderate–light smoking. Future primary care interventions would benefit from targeting multiple risk factors, particularly for heavy smokers.