Patients with depression are more likely to develop type 2 diabetes, and patients diagnosed with diabetes are more likely to develop depressive symptoms than the general population.

A longitudinal cohort study, from the US, used data from a population-based sample of 6,800 men and women aged 45 to 84. Participants enrolled in 2000-2002 and were followed up until 2004-2005.

A CES-D questionnaire was completed at baseline and at the final follow-up visit. A score of 16 or higher (mild to moderate depression or dysthymia) and/or the use of antidepressant medication was used to define elevated depressive symptoms. Fasting blood glucose was measured at each follow-up visit.

Participants with elevated depressive symptoms at baseline were significantly more likely to develop diabetes (relative hazard 1.42, 95% CI 1.02-1.95). However, after adjustment for lifestyle factors the relative hazard was slightly reduced to 1.34, with loss of statistical significance, suggesting that these may be partly responsible.

Compared with participants with normal fasting glucose levels at baseline, patients with untreated type 2 diabetes were less likely, and patients with recognised and treated diabetes significantly more likely, to develop elevated depressive symptoms (relative odds of 0.75 and 1.54).

Although the CES-D has a high sensitivity, it has a low specificity and positive predictive value for major depression.¹ Screening for depression is particularly difficult in patients with diabetes, where some somatic symptoms, such as fatigue, sleep disturbance and altered appetite, can be caused by either condition.² Therefore it would have been better to have followed up positive screening scores with a diagnostic interview.

Adults with diabetes are twice as likely to have depression compared with the general population.³ This study suggests a two-way relationship between depression and diabetes.

The increased risk of diabetes in patients with depression appears to be mediated in part by lifestyle factors such as lack of exercise and diet.

Neuroendocrine effects are also likely to be important. Untreated diabetes appears to reduce the risk of depression, and it has been suggested that insulin resistance may lead to higher serotonin levels.⁴

Diabetes is only psychologically harmful once it has been recognised. This may result from the psychological impact of diagnosis and disease management, or the burden of dealing with complications.

It is possible that the apparent association is because a sedentary lifestyle increases the risks both of depression and diabetes. However, the relative odds of developing elevated depressive symptoms were not significantly affected by adjustment for lifestyle factors.

This study suggests that we should not only promote exercise as an effective treatment for patients with mild to moderate depression but also as a way of reducing their risk of developing diabetes.