BACKGROUND: Smoking has been associated with low back pain (LBP) in several epidemiologic studies but the results are contradictory. Despite this, smoking is often assumed to be a causative factor of LBP.

STUDY OBJECTIVE: To appraise the epidemiologic literature to establish whether there is evidence for a causal link between smoking and LBP.

DATA SOURCES: Twenty-two original research reports published between 1974 and 1993 were reviewed and a systematic investigation was made of eight of these, which were retained because they included study samples representative of their target populations.

DATA SYNTHESIS: To uncover any evidence for a causal relationship between smoking and LBP, these eight reports were examined in detail for strength of association, dose-response effect, temporality and consistency of findings.

RESULTS: Some studies found a positive association between smoking and LBP; when present, the strength of this association was generally small. Some associations remained unchanged after multivariate analysis, whereas others became statistically nonsignificant. Contradictory results were also noted in studies which reported on the dose-response relationship and time of exposure in relation to time of onset of LBP. There was inconsistency of findings within and between studies relating to LBP. However, the evidence was consistently against a causal association between smoking and sciatica/discal hernia.

CONCLUSION: It cannot be clearly deduced whether smoking has a causal effect on LBP or whether the positive findings sometimes noted are linked to some other, still unidentified factor. However, there is clearly no proof supporting a causal association between smoking and sciatica/discal hernia.

This abstract is reproduced with the permission of the publisher. Article only available in print.