Supplementary Table 1. characteristics of 21 independent studies (extracted from 1 eligible articles) of blood 2(OH)D Levels and incident type 2 Author, Publicatio n year Source populatio n Design N Follow -up years Ascertainment Main Outcome (Highest vs. Lowest Category Comparison) Knekt P et al, 2008 Knekt P et al, 2008 Finnish Mobile Clinic Examinatio n Survey, Mini- Survey, Age, yea r 69 (cases/controls or participants, gender) Cases: 10 men and 12 Controls: 206 men and 246 Cases: 83 men and 99 Controls: 24 men and 289 22 egister health information 17 egister health information 2(OH)D ranges dian or mean); Relative Risk (9% CI) and P for linear trend* Men:. vs. 23. 0.49 (0.1-1.64), P=0.06 Women: 62. vs. 22.6 0.91 (0.37-2.23), P=0.66 Men: 7.6 vs. 23.9 0.17 (0.0-0.2), P<0.001 Women: 62.4 vs. 20.4 1.4 (0.8-3.62), P=0.83 ariates adjusted in the model Age BMI, physical activity, smoking status, and education. Age BMI, physical activity, smoking status, and education. 2013 American Association. Published online at http://care.journals.org/lookup/suppl/doi:10.2337/dc12-0962/-/dc1
Pittas AG, Grimnes, G Nurses, TromsØ Norway U.S. female nurses 6.4 an) 43-70 Cohort 0-608 cases and 9 controls 4,17 nonsmokers (183 cases) and 1,962 smokers (64 cases) 14 elf-reported 3. vs. 36.0 nmol/l.2 (0.33-0.83), P=0.008 11 elf-reported questionnaire,g lycemic status, and hospitalconfirmed diagnosis Non-smokers: 72.8 vs. 34. 0.73 (0.48-1.12), P<0.0 Age, race, fasting status, month of blood draw, and laboratory batch for plasma 2OHD, latitude, history of hypercholesterolemia, history of hypertension, smoking status, physical activity, alcohol consumption, multivitamin use, and dietary intakes of caffeine, trans fat, cereal fiber, heme iron, magnesium, fish, and calcium. Age, sex, BMI, physical activity, number of cigarettes smoked and years of smoking (in current smokers). Anderson JL et al, 2010 Intermount ain care system, care population Cohort ± 21 913 cases from 41,04 and men 1.3 an), 9.3 Clinical diagnosis Smokers: 98.6 vs. 73.0 0.68 (0.29-1.61), P<0.0 92. vs. 20 0.3 (0.43-0.6); P, NA Age, gender, history of hypertension, hyperlipidemia, mellitus, peripheral vascular disease, and other clinical conditions. 2013 American Association. Published online at http://care.journals.org/lookup/suppl/doi:10.2337/dc12-0962/-/dc1
Bolland MJ Gagnon C et al, 2011 Robinson, JG et al, 2011 Thorand, B et al, 2011 Pittas, AG et al, A randomized trial of calcium supplement s, New Zealand AusDiab Australia Women s Initiative, MONICA/ KORA Augsburg German Prevention Program, y communit y-dwelling Postmeno pausal Women Adults at high risk for type 2 Cohort > y y) Cohort 0.6 ±12. Casecohort ve cohort from a randomiz ed trial 0-79 66y) 3- >=2 1y) 29 cases from 1,471 postmenopausa l 199 cases from,200 men and 317 cases and 4,823 controls Cases: 231 men and 18 Noncases: 67 men and 610 426 cases from 2039 participants (1,022 in the combined placebo group and 1,017 in intensive lifestyle group) elf-reported, 6 vs. 38 verified using.9 (0.4-1.9); P, NA medical records reatment, fasting glucose, or 2-h OGTT 7.3 Newly treated an) clinical 11.0 Questionnaires or an) interviews validated by physician or medical chart review 2.7 OGTT an) 6 vs. 28..6 (0.36-0.86); P=0.001 64.2 vs. <34.7 1.0 (0.62-1.76); P=0.93 Men: 68.0 vs. 27.7 Women: 8.0 vs. 27.0 nmol/l Total: 0.63 (0.44-0.90); P=0.010.3 vs. 32 0.72 (0.6-0.90); P=0.00 Age, body weight, smoking status, treatment allocation (calcium or placebo), and season. Age, ethnicity, waist circumference, family history of, smoking status, physical activity, and season and latitude. Age, ethnicity, latitude, month of blood draw, WHI ancillary indicators, BMI, hypertension, fiber intake, magnesium intake, and physical activity. Age, sex, survey, season, BMI, smoking status, alcohol intake, physical activity, systolic BP, total cholesterol/hdl cholesterol, parental history of. Recruitment location, age, sex, BMI, race, physical activity, family history of, history of hypertension, smoking status, alcohol consumption, multivitamin use, CRP, kidney function, calcium intake, and treatment arm (intensive lifestyle or placebo). 2013 American Association. Published online at http://care.journals.org/lookup/suppl/doi:10.2337/dc12-0962/-/dc1
Gonzalez- Molero, I et al, Deleskog, A et al, Forouhi, NG, Pizarra (Southern Spain) Stockholm Prevention Program, Sweden European ve Investigatio n into Cancer (EPIC)- Norfolk UK Medical Research Council (MRC) Ely ve casecohort ve 0.3 an) 3-6 7 69 26 new cases among 412 individuals Cases: 179 men and 100 Controls: 607 men and 404 Cases:621; non-case control subcohort:826 37 cases among 0 nondiabetic participants OGTT and glycosylated hemoglobin 46.2 vs. <46.2 0.17 (0.0-0.61); P=0.007 8-10 OGTT >71.0 vs. <43.2 Men: 0.2 (0.30-0.90); P=0.02; Women: 0.79 (0.36-1.73); P=0.23 9-13 Medical record nkage with general ractice,hospital and eath registries. 10 WHO criteria dia (OGTT) n) 80.0 vs. <48.8 0.0 (0.32-0.76); P<0.001 NA; 0.69 (0.17-2.91); P, NA Age, sex, obesity (BMI>30), smoking, outdoor activity, alcohol intake, month of blood sampling, PTH, phosphorus and creatinine. Age, BMI, family history of, physical activity, and blood pressure. smoking, physical activity, education, alcohol intake, and supplement and/or cod liver oil use. alcohol intake, smoking, socioeconomic status, and physical activity. 2013 American Association. Published online at http://care.journals.org/lookup/suppl/doi:10.2337/dc12-0962/-/dc1
Husemoen, LLN et al, Husemoen, LLN et al, Pilz, S et al, Inter99 MONICA 1 population survey Hoorn ve cohort from a randomiz ed trial ve ve 30-6 4 6.3y ) 41-71 0-7 3,79 men and 133 incident cases 1,276 men and 1,29 288 incident cases 280 men and 4 incident cases OGTT, glycosylated hemoglobin, known history of, and/or use of medication 16.4 nformation on hospitalization, registration, blood glucose levels, or antidiabetic medication use 7. OGTT, an) glycosylated hemoglobin, and/or hypoglycemic drugs 7 vs. <2 0.61 (0.28-1.33); P=0.18 142.3 vs. 28.7 0.7 (0.38-0.8); P=0.006 7 vs. <0 0.2 (0.13-2.10); P, NA. Age, sex, BMI, season of blood collection, family history of, change in weight during the followup, physical activity, dietary habits, alcohol intake, smoking status, total energy intake, social class, and randomization group and self-reported changes in dietary habits, physical activity, smoking status, and alcohol intake during the follow-up Age, sex, season of blood collection, history of CVD, waist circumference, physical activity, healthy food index, fish intake, supplement use, smoking status, alcohol intake and educational level. physical activity, hypertension, fasting glucose, HDL-C and triglycerides. * NA, not available; OGTT, oral glucose tolerance test; MONICA/KORA, the Monitoring of Trends and Determinants in Cardiovascular Disease /Cooperative Research in the Region of Augsburg cohort. results were extracted from multivariate-adjusted models without variables that could be intermediate phenotypes between vitamin D and type 2. 2013 American Association. Published online at http://care.journals.org/lookup/suppl/doi:10.2337/dc12-0962/-/dc1