Table 2

Multivariate-adjusted odds ratios (95% CIs)a for associations between total intakes of vitamin D and magnesium and risk of vitamin D deficiency and insufficiency, National Health and Nutrition Examination Survey (NHANES) 2001 to 2006
Nutrient intakes Controls, N Deficiency Insufficiency
N OR (95% CI) N OR (95% CI)
Total vitamin D intake (IU/day)
 Q1 <114 1,741 1,005 1.00 1,450 1.00
 Q2 114 to 307 1,740 446 0.55 (0.40 to 0.75) 1,086 0.75 (0.62 to 0.91)
 Q3 308 to 539 1,741 152 0.22 (0.15 to 0.34) 665 0.55 (0.44 to 0.69)
 Q4 ≥540 1,740 72 0.10 (0.06 to 0.17) 419 0.37 (0.27 to 0.51)
 P for trend <0.001 <0.001
Dietary vitamin D intake (IU/day) 
 Q1 <57.8 1,741 764 1.00 1,183 1.00
 Q2 57.8 to 137 1,740 468 0.71 (0.55 to 0.91) 997 0.86 (0.71 to 1.05)
 Q3 138 to 261 1,741 279 0.54 (0.40 to 0.72) 807 0.72 (0.59 to 0.88)
 Q4 ≥262 1,740 164 0.29 (0.20 to 0.41) 633 0.69 (0.56 to 0.86)
P for trend <0.001 <0.001
Supplemental vitamin D intake (IU/day) 
 Non-user 3,704 1,473 1.00 2,602 1.00
 1 to 239 809 93 0.32 (0.20 to 0.50) 377 0.84 (0.68 to 1.03)
 240 to 399 331 19 0.29 (0.13 to 0.64) 109 0.57 (0.38 to 0.87)
 ≥400 2,118 90 0.19 (0.11 to 0.33) 532 0.40 (0.29 to 0.53)
P for trend <0.001 <0.001
Total magnesium intake (mg/day)
 Q1 <225 1,749 876 1.00 1,362 1.00
 Q2 225 to 310 1,733 392 0.67 (0.50 to 0.90) 907 0.78 (0.61 to 1.00)
 Q3 311 to 419 1,737 245 0.43 (0.30 to 0.63) 758 0.73 (0.59 to 0.89)
 Q4 ≥420 1,743 162 0.34 (0.21 to 0.56) 593 0.62 (0.46 to 0.82)
P for trend <0.001 <0.001
Dietary magnesium intake (mg/day) 
 Q1 <195 1,754 726 1.00 1,255 1.00
 Q2 195 to 268 1,732 432 0.75 (0.58 to 0.97) 893 0.75 (0.62 to 0.90)
 Q3 269 to 363 1,741 297 0.68 (0.47 to 0.99) 777 0.69 (0.55 to 0.87)
 Q4 ≥364 1,735 220 0.43 (0.26 to 0.70) 695 0.61 (0.45 to 0.83)
P for trend 0.002 0.002
Supplemental magnesium intake (mg/day) 
 Non-user 3,922 1,440 1.00 2,548 1.00
 1 to 49 669 109 0.53 (0.34 to 0.81) 334 1.14 (0.91 to 1.43)
 50 to 99 506 43 0.35 (0.17 to 0.71) 205 0.80 (0.53 to 1.22)
 ≥100 1,865 83 0.30 (0.17 to 0.52) 533 0.68 (0.52 to 0.88)
P for trend <0.001 0.008

aSurvey logistic regression models used to estimate odds ratios and 95% confidence intervals. Models adjusted for age, sex, body mass index (BMI), educational attainment, race, household income, smoking status, alcohol use, physical activity, month of blood collection, dietary intakes of total energy and total calcium intake. Models were also adjusted for total intake of vitamin D or magnesium mutually.

Deng et al.

Deng et al. BMC Medicine 2013 11:187   doi:10.1186/1741-7015-11-187

Open Data