Ringstad 1988

Methods	Matched, nested case‐control study Country: Norway
Participants	Participants: 9364 men and women Inclusion criteria: 20 to 54 years of age (men), 20 to 49 years of age (women); inhabitants of Tromso; blood sample provided in 1979; no history of cancer at baseline Name of parent cohort: Tromso Heart Study II Recruitment: 1979 to 1980 Outcome assessment: 1985 Number of cases: Any cancer: 60 (male/female: 26/34) Case definition: incidence Years of follow‐up: 5.0 to 7.0 years Type of selenium marker: serum
Interventions	d.n.a.
Outcomes	Analysed cases: 60 of 72 cases analysed (reason for non‐inclusion: no sample available) Statistical methods: n.r. Variables controlled by matching: age, gender, smoking status, month of sample collection, place of residence (district of Tromso)
Risk estimates [95% CI]	Reference category: highest three quartiles Results: Any cancer    both genders: lowest quartile: OR 1.4 (95% CI 0.6 to 3.5)
Selenium levels in exposure categories	lowest quartile: ≤ 114.49 µg/l highest three quartiles: 114.50 to 114.51 µg/l
Notes