Peters 2007

Methods	Matched, nested case‐control study Country: US
Participants	Participants: 26,975 white non‐Hispanic men Inclusion criteria: 55 to 74 years of age; excluded: no baseline questionnaire/informed consent/blood sample, no further contact after screening Name of parent cohort: Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial Recruitment: September 1993 to June 2001 Outcome assessment: 1 October 2001 Number of cases: Prostate cancer: 724 (male/female: 724/0) Case definition: incidence Years of follow‐up: 0.3 to 8.0 years Type of selenium marker: serum
Interventions	d.n.a.
Outcomes	Analysed cases: 724 of 803 cases included in analysis (reason for non‐inclusion: no selenium measurement available) Statistical methods: n.r. Variables controlled in analysis: age, time since initial screening, year of blood collection, study centre Variables controlled by matching: age, month of sample collection, time since initial screening
Risk estimates [95% CI]	Reference category: lowest quartile Results: Prostate cancer              highest quartile: OR 0.84 (95% CI 0.62 to 1.14)
Selenium levels in exposure categories	lowest quartile: 50.5 to 126.7 ng/ml highest quartile: 158.0 to 253.0 ng/ml
Notes