Study design: A cross-sectional postal survey of 29,424 people aged 12-41 years obtained from a population-generated panel of twin individuals.
Objectives: To study whether smoking causes low back pain.
Summary of background data: Despite insufficient evidence in the epidemiologic literature, it has become increasingly accepted that smoking causes low back pain and that discontinuation of smoking is a suitable means of secondary prevention.
Methods: Dose-response was examined for smoking (daily use, number of years smoked, and total cigarette use during the years of smoking) in correlation with low back pain (occurring 1-7 days, 8-30 days, and > 30 days in the past year). A possible modifying effect was studied for age, gender, and body mass index. A negative gradient was sought in relation to the time since smoking was discontinued. The prevalence of low back pain was studied in monozygotic twin pairs, only one of whom smoked.
Results: There was a significant positive association between smoking and low back pain that increased with the duration of low back pain: occurring 1-7 days (odds ratio, 1.4), 8-30 days (odds ratio, 2.1), and more than 30 days (odds ratio, 3) in the past year. However, these differences in reports of low back pain disappeared in monozygotic twin pairs discordant on present smoke status. There was no biologic gradient for any of the low back pain definitions or measures of smoking-dose, and the prevalence of low back pain did not decrease with the number of years since smoking was stopped. Smaller people (youngsters, women, people with low body mass index) were not more vulnerable to development of low back pain with smoking.
Conclusions: There is a definite link between smoking and low back pain that increases with the duration and frequency of the low back pain problem, but this link is unlikely to be causal.