Abstract
Many systematic reviews have summarized evidence on the association between behavioural factors and incident cancers. To date, there has been little synthesis of heterogeneity by sex/gender of this evidence.An umbrella review was conducted of systematic reviews with quantitative synthesis (meta-analysis, meta-regression) examining the exposures of body size; physical activity; wholegrains, vegetables, fruit and beans; "fast foods"; red and processed meat; sugar sweetened drinks; dietary supplements; alcohol; tobacco; and sun exposure with incident non-sex-specific cancers. A search of Ovid MEDLINE, Ovid Embase, and Cochrane library from database inception to May 2023 was conducted. We calculated the proportion of systematic reviews that provided quantitative sex/gender findings (e.g., subgroup analyses) and summarized findings narratively. Methodological quality was appraised with the AMSTAR-2 tool.From 13,227 records, 705 full-text systematic reviews were identified as meeting inclusion criteria. Of these, 361 (51.2%) reported quantitative sex/gender findings. The terms "sex" and "gender" were used interchangeably by 36.3% of the 361 systematic reviews and none reported findings for transgender, gender-diverse, or non-binary individuals. Overall, 98.6% (356/361) of systematic reviews were rated "critically low" with the AMSTAR-2 tool. Most of the 361 systematic reviews with quantitative sex/gender findings reported no statistically significant differences by sex/gender.This umbrella review found conflation of sex and gender in systematic reviews of behavioural factors and non-sex-specific cancers and a lack of research among non-cisgender individuals. The existing evidence base is of critically low quality and our findings of no sex/gender-specific trends must be interpreted with caution.