Intratrial Exposure to Vitamin D and New-Onset Diabetes Among Adults With Prediabetes: A Secondary Analysis From the Vitamin D and Type 2 Diabetes (D2d) Study. Diabetes Care 2020;43:2916-2922.

We thank Dr. Davidson for his comment (1) on our article (2). We agree that weight loss is critically important in reducing risk of diabetes; weight loss can also result in increased 25-hydroxyvitamin D [25(OH)D] levels. We did not adjust for weight change because weight remained fairly stable during the trial. Nonetheless, in response to Dr. Davidson’s request, we have examined the impact of adjustment for change in weight (final − baseline weight) in the fully adjusted models shown in our article. In the group as a whole, the hazard ratios (HRs) changed from 0.57 (95% CI 0.41–0.79) to 0.55 (0.40–0.76) in the 25(OH)D category of 100–124 nmol/L and from 0.35 (0.24–0.50) to 0.34 (0.24–0.50) in the highest 25(OH)D category of ≥125 nmol/L. These minimal changes were expected because weight did not change much during the trial (mean change among the five 25(OH)D categories ranged from –0.5 kg to 0.7 kg). We also examined the impact of weight change in each group separately. In participants assigned to placebo, the additional adjustment for weight change altered the HR from 0.67 (0.40–1.12) to 0.69 (0.41–1.17) in the 100–124 nmol/L category and from 0.47 (0.15–1.52) to 0.58 (0.58–1.86) in the ≥125 nmol/L category. In participants assigned to vitamin D, the additional adjustment for weight change reduced the HRs from 0.48 (0.29–0.80) to 0.41 (0.24–0.69) in the 100–124 nmol/L category and from 0.29 (0.17–0.50) to 0.25 (0.15–0.43) in the ≥125 nmol/L category. These analyses confirm that weight loss did not account for the reduced incidence of diabetes observed in participants assigned to treatment with 4,000 IU per day of vitamin D who maintained high levels of 25(OH)D during the trial.