Joint tobacco smoking and alcohol intake exacerbates cancer risk in women– the Danish Nurse Cohort

Abstract Purpose To explore separate and combined tobacco and alcohol use and risk of overall, smoking-related, alcohol-related, breast and gynecological cancers in women. Method Questionnaires from 19,898 women in The Danish Nurse Cohort in 1993 were linked to registries on hospitalizations, death causes and migration until Dec 2016. Cancer risk by tobacco and alcohol was estimated using Cox proportional hazards models. Results 16,106 nurses, aged >44 years (mean = 56), were eligible for analysis. Throughout 23 years (mean follow-up = 18.8 years) overall cancers counted 4,968. Of these, 1,897, 2,231, 1,407 and 579 events were smoking-related, alcohol-related, breast cancers and gynecological cancers. Increased risks of overall, smoking-related, and breast cancer were observed for current smoking and excess alcohol intake (>14 units/week), separately, compared to never smoking and light drinking (1–7 units/week) respectively. Moderate drinking (8–14 units/week) increased the risk of alcohol-related and breast cancer. Additional risk increases were observed among smokers drinking alcohol above light levels for overall, smoking-related, alcohol-related and breast cancer (HR = 1.40, 95% CI:1.30–1.51, HR = 1.72, 95% CI:1.52–1.94, HR = 1.33, 95% CI:1.26–1.40, HR = 1.32, 95% CI:1.15–1.53, respectively), compared to non-smokers drinking lightly. These risks increased further for smokers drinking above moderate levels (HR = 1.49, 95% CI:1.36–1.63, HR = 1.97, 95% CI:171.-2.26, HR = 1.40, 95% CI:1.22–1.60, HR = 1.33, 95% CI:1.12–1.57, respectively). No significant associations were found for gynecological cancer. Conclusions Smoking and alcohol, both separately and combined, increased risks of overall, smoking-related, alcohol-related and breast cancer; combined use resulted in incremental risk increases. Co-use of smoking and alcohol represent an extensive threat to public health; thus, prevention could benefit from combined targeting.