All studies included adjusted for age; when studies reported more than one adjusted estimate, the multivariate adjusted estimate was selected (i.e., if a study reported either age adjusted or age, gender and smoking adjusted results, the latter was chosen).
Age-adjusted and multivariate-adjusted estimates are presented.
Generally, the multivariate-adjusted estimates were similar to the age-adjusted estimates, indicating that perhaps none of these factors is a significant confounder.
If publications were duplicated or shared in more than one study, either the former or the one without multivariate-adjusted estimates was excluded.
The first model showed crude odds ratio (OR) and 95% confidence interval (CI), whereas the second model showed multivariate-adjusted estimates controlling for age and sex.
Further, we found no evidence of substantial inflation in effect estimates when we compared the crude (data not shown) with the final multivariate-adjusted estimates.
Table 2 shows that the multivariate adjusted HR estimate for rectal cancer was similar for current and former smokers of both genders.
The literature review identified 83 full articles for detailed assessment, of which 53 were excluded for having no multivariate adjusted stroke estimate, six for being duplicated studies, and three for having a retrospective cohort design.
However, this may be due to the effect of pre-existing disease, as, after excluding breast cancer cases diagnosed during the first year of follow-up, the multivariate adjusted risk estimate for weight loss approached unity (HR for <−2 kg: 1.01, CI 0.59 1.76 based on 175 cases), while relative risks in the weight gain categories were not materially changed (data not shown).
We excluded 29 studies owing to lack of a multivariate adjusted stroke estimate, seven studies for providing only fatal endpoints; two studies for follow-up duration less than one year; eight studies for lacking proper impaired fasting glucose or impaired glucose tolerance categories at baseline; 13 14 15 16 17 18 19 20 and one study that was retrospective rather than prospective (fig 1).
