Table 2.
Statistical analysis of Study 2 scores of wellbeing (Winter 2002–2003).
| Intent-to-treat analysis | Per-protocol analysis | ||||||||
| Dose of Vitamin D | N | Age | N | ||||||
| mcg/day (IU/day) | Total in group, (% female) | yr (SD) | 25(OH)D nmol/L (SD) | December 2002 Score (out of 16) |
February 2003 Score (out of 16) |
Total in group, (% female) |
December 2002 Score (out of 16) |
February 2003 Score (out of 16) |
|
| CONTINUERS FROM STUDY 1 (on Vit D since previous year) | |||||||||
| 15 (600) | 22 (77%) | 54 (14) | 69 (26) | 7.2 (4.5) | 4.4 (3.4) | 15 (73%) | 6.9 (4.8) | 4.4 (3.4) b | |
| 100 (4000) | 24 (84%) | 56 (9) | 126 (45) a | 4.4 (4.4) a | 4.0 (3.7) | 16 (88%) | 4.6 (4.6) | 4.0 (3.7) | |
| NEW PATIENTS FOR STUDY 2 | |||||||||
| 15 (600) | 33 (68%) | 48 (13) | 39 (9) | 8.0 (5.2) | 5.4 (4.3) | 25 (64%) | 8.7 (5.5) | 5.4 (4.3) b | |
| 100 (4000) | 33 (85%) | 50 (14) | 39 (9) | 8.4 (5.5) | 3.9 (3.6) c | 26 (89%) | 8.1 (5.6) | 3.9 (3.6) bc | |
a Different from 15 mcg (600 IU)/day group (the value above the mean marked by this footnote) by t-test p < 0.04; lower (better) than in the 600 IU/day group by Mann-Whitney p = 0.039.
b Paired t-test, December score vs February Score (the value to the left of the mean marked by this footnote) p < 0.012; also significant by the non-parametric equivalent to paired t-test, the Wilcoxan test, p < 0.012.
c For New patients, low vs high dose group, 2-tail unpaired t test p = 0.188; Mann-Whitney p = 0.183; i.e. not significantly different.