Objectives We investigated hypotheses for the association between shiftwork and breast cancer based on our a priori theoretical framework of five biological mechanisms which might be operational in shiftwork: light at night; phase shift (when central cycles have adjusted to night work, but peripheral cycles have not); sleep disruption; lifestyle factors (diet, physical activity and alcohol intake) and low vitamin D.

Methods We conducted a population-based case-control study with 1205 breast cancer cases, identified from 2009 to 2011 identified through the Western Australian Cancer Registry, and 1789 age-matched controls from the Western Australian electoral roll. An occupational history was collected by self-completed questionnaire for every job a woman had held for at least six months (job title, main tasks, year started, duration, hours per week and weeks per year worked, and whether the job involved night work, shift work, or work at unusual hours). Using the web application OccIDEAS, we obtained further details about the shiftwork by telephone interview. Automatic assessments with manual reviews were used to assess occupational exposure to the hypothesized factors.

Results We found a 22% increase in breast cancer risk (OR 1·22, 95% CI 1·01–1·47) for phase shift with a statistically significant dose response relationship (p = 0.04). For the other hypothesized mechanisms, risks were marginally elevated and not statistically significant. No association was significant in Bayesian analyses.

Conclusions We suggest that future studies use similar biologically-based exposure assessments in order for us to be sure what advice we should give to the millions of women around the world who work at night.