We thank Drs Mezuk and Grant for their thoughtful comments. Dr Mezuk raised an important point that stress, rather than depression, may be the underlying causal factor for diabetes. Stressful life events are well-known triggers for depression, and these events may also be linked to an increased risk of diabetes.1 Thus, it is plausible that stress explains the as sociation between depression and diabetes risk. To test this hypothesis and examine whether the association between depression and diabetes risk is attenuated by stress, we could include specific stressful life events as covariates in our model. We were unable to test this hypothesis with our current data because detailed data on stressful life events were not collected. However, even if the association between depression and diabetes is attenuated by adjusting for stress, this finding would not invalidate the relationship between depression and diabetes because depression may be a mediating factor between stress and diabetes. Consistent with previous studies,2,3 we found that treatment with antidepressants was actually associated with an increased risk of diabetes. We must be cautious in our interpretation of these results, however, because treatment with antidepressants may be a marker of the severity of depression and the medications may have direct deleterious effects on glucose metabolism. Depression may result from long-term stress associated with the management and treatment of diabetes and its debilitating complications. Thus, psychosocial stress may contribute to depression among patients with diabetes.
Pan A, Hu FB. Low Serum 25-Hydroxyvitamin D Levels and the Bidirectional Association Between Depression and Type 2 Diabetes Mellitus in Women—Reply. Arch Intern Med. 2011;171(11):1041-1042. doi:10.1001/archinternmed.2011.223