Table 1.

Age-standardized ^a characteristics of the study population at baseline in 1995 according to quartiles of predicted vitamin D score.

	Predicted vitamin D score at baseline
	Quartile 1 (n=12,383)	Quartile 2 (n=12,384)	Quartile 3 (n=12,384)	Quartile 4 (n=12,383)
Factors included in the vitamin D prediction model
Body mass index, kg/m²	32.0 (7.4)	28.3 (6.5)	26.3 (5.2)	25.0 (4.9)
Smoking status: Current smoker, %	26	15	12	7
Alcohol consumption: Current drinker, ≥7 drinks/week, %	13	7	4	3
Vigorous exercise: ≥5 hours/week, %	8	11	15	21
Vitamin D supplementation, %	0	0	1	29
Multivitamin use, %	14	52	80	93
Dietary vitamin D intake, kcal-mcg/day	2.4 (1.3)	2.6 (1.4)	3.0 (1.7)	3.1 (1.9)
Potential confounders
Age	40.3 (10.1)	39.1 (10.6)	38.2 (10.6)	36.1 (10.3)
Years of education: ≤12 years, %	25	18	15	12
Family history of colorectal cancer, %	9	9	10	9
Calcium supplementation, %	9	16	22	46
Folate supplementation, %	3	4	6	8
Current aspirin use, %	10	9	9	8
Red meat consumption: Highest quartile, %	31	26	22	19
Processed meat consumption: Highest quartile, %	30	26	22	20
Total fiber intake: Highest quartile, %	18	23	28	32
Geographic region:
Northeast, %	25	28	28	28
South, %	31	31	30	31
Midwest, %	22	23	23	24
West, %	21	18	19	17

^aValues are means (SD) or percentages and are standardized to the age distribution of the study population.