Association between plasma 25-hydroxyvitamin D levels and fracture risk: the EPIC-Oxford study.

Journal article

The importance of vitamin D for bone health is well established, but few data exist on the relation between plasma levels of 25-hydroxyvitamin D and risk of fracture. The authors examined this association within the EPIC-Oxford (European Prospective Investigation into Cancer and Nutrition-Oxford cohort) study of men and women in the United Kingdom (1993-1999). Five years after recruitment, participants completed a follow-up questionnaire where fracture incidence was self-reported. Plasma 25-hydroxyvitamin D concentration was measured in 730 incident fracture cases and 1,445 matched controls. There was a clear association between plasma 25-hydroxyvitamin D concentration and month of blood draw, the highest values being during the summer months. Among women, there were significant relations between 25-hydroxyvitamin D levels and age, body mass index, marital status, use of hormone therapy, physical activity, diet group, dietary intake of vitamin D, and alcohol. Similar relations were seen among men, although often they were nonsignificant because of smaller numbers. There was no evidence of an association between plasma 25-hydroxyvitamin D and fracture risk for men or women; the relative risks associated with a doubling of plasma 25-hydroxyvitamin D were 1.15 (95% confidence interval: 0.82, 1.61) and 0.95 (95% confidence interval: 0.80, 1.13), respectively. These results were not affected by adjustment for potential confounders and were consistent across a number of subgroups.

Authors: Roddam, A; Neale, R; Appleby, P; Allen, N; Tipper, S

• Publication status: Published
• Journal: American journal of epidemiology
• Volume: 166
• Issue: 11
• Pages: 1327-1336
• Publication date: 2007-12-01
• DOI: 10.1093/aje/kwm210
• EISSN: 1476-6256
• ISSN: 0002-9262
• Language: English
• Keywords: Great Britain; Vitamin D; Incidence; Male; Chi-Square Distribution; Risk Factors; Adult; Female; Fractures, Bone; Case-Control Studies; Middle Aged; Linear Models; Humans; Questionnaires; Aged