Tobacco use and incident sleep parameters among a rural ageing population in South Africa

INTRODUCTION Tobacco use may be associated with incident insomnia. The aim of the study was to investigate the association between tobacco use and incident sleep parameters in a longitudinal study in South Africa. METHODS Longitudinal data from two consecutive waves of middle-aged and older adults in 2014–2015 (n=5059) and 2018–2019 (n=4176) in rural South Africa were analyzed. Tobacco use and sleep parameters were assessed by self-report. The associations between tobacco use and incident sleep parameters were estimated with multivariable logistic regression. RESULTS The prevalence of baseline sleep parameters was poor sleep quality 6.5%, sleep disturbance 13.6%, restless sleep 32.9%, and breathing stops 7.0%. In the fully adjusted model for people without poor sleep quality at baseline, daily tobacco smoking, smoking ≥10 units of tobacco products, current tobacco use and current smokeless tobacco use did not increase the odds of incident poor sleep quality. Smoking ≥10 units of tobacco products in a day (AOR=3.83; 95% CI: 1.77–8.28), current tobacco use (AOR=1.65; 95% CI: 1.09–2.51), and daily tobacco smoking (AOR=2.16; 95% CI: 1.15–4.07), were significantly positively associated with incident sleep disturbance. Furthermore, incident restless sleep was significantly positively associated with smoking ≥10 units of tobacco products in a day (AOR=3.97; 95% CI: 1.18–13.37), current smokeless tobacco use (AOR=2.78; 95% CI: 1.17–6.62) and current tobacco use (AOR=2.00; 95% CI: 1.00–4.00). Incident breathing stops were significantly positively associated with daily smoking tobacco (AOR=2.08; 95% CI: 1.11–3.34), smoking 1–9 units of tobacco products in a day (AOR=2.17; 95% CI: 1.20–3.94), and current tobacco use (AOR=1.77; 95% CI: 1.16–2.72). CONCLUSIONS Higher tobacco use was independently associated with incident sleep disturbance, incident restless sleep, and incident breathing stops, but not with incident poor sleep quality.