Table 4.
Multivariable adjusted percent differences in 25(OH)D concentration by exposure to demographic and seasonal factors.
| Parameter | Season | |||
|---|---|---|---|---|
| Summer (n = 318) 1 | Winter (n = 307) 1 | |||
| %ΔY (95%CI) or Intercept | p | %ΔY (95%CI) or Intercept | p | |
| Model 1 | ||||
| Intercept | 36 ng/mL | <0.001 | 9 ng/mL | <0.001 |
| Age, years | −1 (−1, 0) | <0.001 | −0 (−1, 0) | 0.07 |
| Female sex | −15 (−22, −9) | <0.001 | -6 (-13, 2) | 0.16 |
| Model 2 | ||||
| Intercept | 23 ng/mL | <0.001 | 6 ng/mL | <0.001 |
| Age, years | −1 (−1, −1) | <0.001 | −1 (−1, 0) | <0.01 |
| Female sex | −15 (−21, −10) | <0.001 | −5 (−11, 2) | 0.16 |
| Region | ||||
| Ulaanbaatar | ref | ref | ||
| Omnogobi | 64 (44, 88) | <0.001 | 69 (47, 94) | <0.001 |
| Bulgan | 45 (27, 67) | <0.001 | 69 (47, 94) | <0.001 |
| Khuvsgul | 36 (18, 55) | <0.001 | 37 (19, 57) | <0.001 |
| Tuv | 49 (30, 70) | <.001 | 55 (35, 78) | <0.001 |
| Sukhbaatar | 45 (26, 66) | <0.001 | 56 (35, 79) | <0.001 |
| Khovd | 6 (−7, 22) | 0.37 | 19 (3, 37) | 0.02 |
| Dornod | 31 (15, 50) | <0.001 | 22 (6, 40) | <0.01 |
| Outdoor occupation (vs. indoor) | 17 (10, 26) | <0.001 | 4 (−3, 12) | 0.24 |
| Model 3 | ||||
| Intercept | 19 ng/mL | <0.001 | 5 ng/mL | <0.001 |
| Age, years | −1 (−1, 0) | <0.001 | 0 (−1, 0) | <0.01 |
| Female sex | −17 (−22, −11) | <0.001 | −7 (−13, 0) | 0.06 |
| Region | ||||
| Ulaanbaatar | ref | ref | ||
| Omnogobi | 68 (45, 95) | <0.001 | 79 (53, 111) | <0.001 |
| Bulgan | 49 (27, 75) | <0.001 | 76 (53, 107) | <0.001 |
| Khuvsgul | 34 (17, 54) | <0.001 | 42 (25, 68) | <0.001 |
| Tuv | 54 (33, 78) | <.001 | 66 (43, 98) | <0.001 |
| Sukhbaatar | 39 (20, 62) | <.001 | 60 (40, 92) | <0.001 |
| Khovd | 9 (−7, 29) | 0.29 | 31 (10, 59) | <0.01 |
| Dornod | 35 (14, 60) | <0.001 | 33 (13, 64) | <0.01 |
| Outdoor occupation (vs. indoor) | 23 (11, 35) | <0.001 | 7 (−2, 19) | 0.11 |
| % body surface area exposed, 1 SD 2 | 6 (2, 10) | <0.01 | −0 (−4, 4) | 0.95 |
| Exposure duration score, per 1 SD 2 | −6 (−11, 0) | 0.04 | 1 (−3, 5) | 0.71 |
| Housing type | ||||
| Apartment | ref | ref | ||
| Yurt | 7 (−6, 21) | 0.32 | −9 (−21, 3) | 0.12 |
| House with central heating | 4 (−9, 18) | 0.61 | −10 (−22, 3) | 0.11 |
| House without central heating | 13 (−1, 29) | 0.06 | 1 (−13, 14) | 0.99 |
%ΔY for continuous parameters is interpreted as the % difference in seasonal 25(OH)D concentration associated with a one-unit increase in the parameter. %ΔY for levels of categorical variables are interpreted as the % change in seasonal 25(OH)D concentration associated with each level of the parameter. p-values are those associated with the difference in log(25(OH)D) for a one-unit change in a continuous parameter, a level of a categorical parameter relative to the reference category (“ref”), or the reference category itself. 1 In summer, two vitamin D outliers were excluded (n = 318), and in winter, one vitamin D outlier was excluded and 12 vitamin D measurements were missing (n = 307); 2 %ΔY for % body surface area exposed and exposure duration score are expressed in terms of a 1 SD change in the parameter.