Alicia K Heath1, Elizabeth J Williamson2, Allison M Hodge3, Peter R Ebeling4, Darryl W Eyles5, David Kvaskoff6, Kerin O'Dea7, Graham G Giles3, Dallas R English8. 1. Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, University of Melbourne, 207 Bouverie St, Melbourne, Victoria 3010, Australia; Cancer Epidemiology & Intelligence Division, Cancer Council Victoria, 615 St Kilda Rd, Melbourne, Victoria 3004, Australia; Nuffield Department of Population Health, University of Oxford, Roosevelt Drive, Oxford OX3 7LF, UK. 2. Farr Institute of Health Informatics Research, 222 Euston Rd, Kings Cross, London NW1 2DA, UK; Department of Medical Statistics, London School of Hygiene & Tropical Medicine, Keppel St, Bloomsbury, London WC1E 7HT, UK. 3. Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, University of Melbourne, 207 Bouverie St, Melbourne, Victoria 3010, Australia; Cancer Epidemiology & Intelligence Division, Cancer Council Victoria, 615 St Kilda Rd, Melbourne, Victoria 3004, Australia. 4. Department of Medicine, School of Clinical Sciences, Monash University, Monash Medical Centre, Clayton, Victoria 3168, Australia. 5. Queensland Brain Institute, University of Queensland, St Lucia, Queensland 4072, Australia; Queensland Centre for Mental Health Research, The Park Centre for Mental Health, Wacol, Queensland 4076, Australia. 6. Queensland Brain Institute, University of Queensland, St Lucia, Queensland 4072, Australia. 7. Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, University of Melbourne, 207 Bouverie St, Melbourne, Victoria 3010, Australia. 8. Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, University of Melbourne, 207 Bouverie St, Melbourne, Victoria 3010, Australia; Cancer Epidemiology & Intelligence Division, Cancer Council Victoria, 615 St Kilda Rd, Melbourne, Victoria 3004, Australia. Electronic address: d.english@unimelb.edu.au.
Abstract
AIMS: Inverse associations between vitamin D status and risk of type 2 diabetes observed in epidemiological studies could be biased by confounding and reverse causality. We investigated the prospective association between vitamin D status and type 2 diabetes and the possible role of reverse causality. METHODS: We conducted a case-cohort study within the Melbourne Collaborative Cohort Study (MCCS), including a random sample of 628 participants who developed diabetes and a sex-stratified random sample of the cohort (n = 1884). Concentration of 25-hydroxyvitamin D (25(OH)D) was measured using liquid chromatography-tandem mass spectrometry in samples collected at recruitment. Logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for the risk of type 2 diabetes for quartiles of 25(OH)D relative to the lowest quartile and per 25 nmol/L increase in 25(OH)D, adjusting for confounding variables. RESULTS: The ORs for the highest versus lowest 25(OH)D quartile and per 25 nmol/L increase in 25(OH)D were 0.60 (95% CI: 0.44, 0.81) and 0.76 (95% CI: 0.63, 0.92; p = 0.004), respectively. In participants who reported being in good/very good/excellent health approximately four years after recruitment, ORs for the highest versus lowest 25(OH)D quartile and per 25 nmol/L increase in 25(OH)D were 0.46 (95% CI: 0.29, 0.72) and 0.71 (95% CI: 0.56, 0.89; p = 0.003), respectively. CONCLUSIONS: In this sample of middle-aged Australians, vitamin D status was inversely associated with the risk of type 2 diabetes, and this association did not appear to be explained by reverse causality.
AIMS: Inverse associations between vitamin D status and risk of type 2 diabetes observed in epidemiological studies could be biased by confounding and reverse causality. We investigated the prospective association between vitamin D status and type 2 diabetes and the possible role of reverse causality. METHODS: We conducted a case-cohort study within the Melbourne Collaborative Cohort Study (MCCS), including a random sample of 628 participants who developed diabetes and a sex-stratified random sample of the cohort (n = 1884). Concentration of 25-hydroxyvitamin D (25(OH)D) was measured using liquid chromatography-tandem mass spectrometry in samples collected at recruitment. Logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for the risk of type 2 diabetes for quartiles of 25(OH)D relative to the lowest quartile and per 25 nmol/L increase in 25(OH)D, adjusting for confounding variables. RESULTS: The ORs for the highest versus lowest 25(OH)D quartile and per 25 nmol/L increase in 25(OH)D were 0.60 (95% CI: 0.44, 0.81) and 0.76 (95% CI: 0.63, 0.92; p = 0.004), respectively. In participants who reported being in good/very good/excellent health approximately four years after recruitment, ORs for the highest versus lowest 25(OH)D quartile and per 25 nmol/L increase in 25(OH)D were 0.46 (95% CI: 0.29, 0.72) and 0.71 (95% CI: 0.56, 0.89; p = 0.003), respectively. CONCLUSIONS: In this sample of middle-aged Australians, vitamin D status was inversely associated with the risk of type 2 diabetes, and this association did not appear to be explained by reverse causality.
Authors: A Giustina; R A Adler; N Binkley; J Bollerslev; R Bouillon; B Dawson-Hughes; P R Ebeling; D Feldman; A M Formenti; M Lazaretti-Castro; C Marcocci; R Rizzoli; C T Sempos; J P Bilezikian Journal: Rev Endocr Metab Disord Date: 2020-03 Impact factor: 6.514