Abstract:
Background: In 2007, the International Agency for Research on Cancer classified shift work that involves circadian disruption as a probable human carcinogen. Nighttime chronodisruption has been shown to decrease concentrations of melatonin, a presumed antineoplastic hormone. Based on this biological plausibility, our objective was to investigate an association between night shift work and lung cancer risk among Chinese female textile workers.
Methods: we conducted a case-cohort study nested within a cohort of 267,400 female textile workers employed by the Shanghai Textile Industry Bureau. Participants were followed for lung cancer incidence from January 1, 1989 through December 31, 2006. The comparison group consisted of a Subcohort (n=4807) based on a 5-year frequency age-matched to all-cancer cases in cohort. We examined an association between exposure to cumulative night shift work and lung cancer outcomes (n=1423) adjusting for age, smoking, and parity. Hazard ratios (HRs) using Cox proportional hazards model with robust variance estimated relative risks with 95% confidence intervals (CIs).
Results: Surprisingly, cumulative years of working rotating night shifts and cumulative nights of rotating shiftwork were both associated with moderately reduced risk of lung cancer , although the trends were not statistically significant (Ptrend of 0.19 and 0.28, respectively). The hazard ratios, adjusted for age and smoking history, for the lowest categories of cumulative years working rotating night shifts (HR = 0.78, 95% CI = 0.64 to 0.95) and cumulative rotating nights worked (HR = 0.79, 95% CI = 0.65 to 0.96) were both statistically significant. Similar results were obtained when exposures were lagged by 20 years to account for disease latency. Parity was not included in the final model, as it did not significantly alter the relative risk estimates.
Conclusions: Long-term rotating nighttime shift work appears to demonstrate a decreased risk of lung cancer among Chinese female textile workers compared to non-exposed although trend was not statistically significant. Further investigation on potential factors not yet measured may elucidate a more definitive association of this effect.