Treatment of Depression in Diabetes: an Update

Frank Petrak; Stephan Herpertz


Curr Opin Psychiatry. 2009;22(2):211-217. 

In This Article

Treatment of Depression in Diabetes: Current Knowledge and Future Directions

For an overview of current knowledge, a literature search for all published RCTs that evaluated treatment of depression in diabetes was conducted; the search yielded 11 RCTs ( Table 1 ).[20,21,22,23•,24,25,26,27,28,29•,30••] The RCTs were grouped according to the interventions that were used: pharmacological, psychological or mixed (pharmacological and psychological).

Pharmacological Treatment Trials for Depression in Diabetes

Up to now, four pharmacological trials for depression therapy in diabetes have been published. In the first RCT, nortriptyline was tested in comparison with placebo in patients with depression and poorly controlled diabetes. There was a significant improvement in the intervention group regarding depression, but deterioration in glycemic control was observed in the intervention group in subsequent analyses.[20]

The next trial evaluated the effectiveness of fluoxetine compared with placebo in patients with diabetes and depression. Again, there was a significant improvement in depression after 2 months of treatment in the intervention group. There was a trend toward better glycemic control for fluoxetine; however, this trend did not reach statistical significance, a result that may be due to insufficient statistical power.[21]

A more recent RCT that included patients with type 1 or type 2 diabetes evaluated the effect of sertraline on prevention of relapse of depression.[22] The first phase of the trial was a noncontrolled, open-label treatment with sertraline in which 43% of the patients achieved remission of depression. Those treatment responders were subsequently included in a RCT comparing sertraline with placebo for relapse prevention. Patients were followed for a maximum of 12 months or until depression recurred. Depression recurred in one-third of the patients, but there was a clear and significant advantage of sertraline over placebo (median time to recurrence was 57 days in the placebo group compared with 226 days in patients treated with sertraline). Regarding glycemic control, there was a significant improvement in the whole sample in the nonrandomized phase 1 of the study when every participant was treated with antidepressants. In the second phase, in which those who had responded to antidepressant treatment were randomized to long-term sertraline or placebo, there was no significant difference in the change in glycemic control in the two groups. The clearest result from the RCT part of the study was that sertraline is an effective agent for the prevention of relapse of depression in diabetes.

In the most recent pharmacological RCT, paroxetine was compared with placebo in elderly patients with minor depression.[23•] The results yielded no statistically significant differences between the groups on the primary psychological and medical outcome variables. In conclusion, there is no evidence to support the use of paroxetine for older patients with minor depression and diabetes.

Psychological Treatment Trials for Depression in Diabetes

The only trial of cognitive behavioral therapy (CBT) included 52 patients with type 2 diabetes and major depression.[24] For 10 weeks, patients were randomized to diabetes education and CBT for depression or to diabetes education only. The follow-up included a 6-month interval during which 70% of the patients in the CBT group achieved remission compared with just 33% in the education-only group. Regarding medical outcomes, there was a clear advantage of CBT compared with the control group 6 months after treatment was delivered but not during the 10-week intervention phase; HbA1c deteriorated in the control group (+0.9%), whereas it improved in the CBT group (-0.7%). This result was statistically significant (P = 0.04) and relevant from a clinical point of view; moreover, it was the first time that a psychotherapeutic approach was demonstrated to be effective for the treatment of depression in diabetes. Considering the various limitations of this trial (small sample, monocentric trial with only one therapist, and no replication of the results), more data are needed to generalize these results.

The second RCT regarding psychological treatment was a trial to evaluate group counseling in China; 59 patients were randomized to group counseling (with a focus on social support) or treatment as usual for 3 months.[25] Significant improvements were reported in the group counseling condition for depression and glycemic control; however, the methodological limitations were comparable to those of the aforementioned trial.

The most recent trial was a pilot study to evaluate supportive psychotherapy in patients with diabetic foot syndrome who also had depressive symptoms.[26] Patients were randomized to either supportive psychotherapy or standard medical treatment for a period of 6 weeks on average. Results, which were reported for posttreatment evaluation but not for follow-up data, demonstrated a moderate improvement in depressive symptoms. Given the trial's short duration, it is not surprising that no difference was observed for glycemic control or other medical outcome variables.

Mixed (Psychological and Medical) Treatments for Depression in Diabetes

The effectiveness of algorithm-based, flexible interventions using a combination of psychological and pharmacological treatments compared with standard care was evaluated in four RCTs. The psychological modules of these treatments included problem solving training[27,28] and counseling[29•] or interpersonal therapy.[30••] In addition, in all four trials, antidepressants were given according to the patients' preferences or following a predefined treatment algorithm.

A significant improvement in depression was observed for the combination of antidepressant medication with problem solving training[27,28] or counseling[29•] compared with standard care; however, regarding metabolic control, no significant differences between the intervention and control groups were observed.

In a recent algorithm-based care trial that included 123 patients with depression and self-reported diabetes (among 584 patients without diabetes), interpersonal therapy and citalopram (in combination or alone) were compared with care as usual.[30••] The results of a secondary analysis demonstrated that this intervention led to a significant decrease in mortality after 5 years. But because depression was not assessed or specified, no conclusion can be drawn regarding depression treatment.

Finally, in 2008, a RCT was conducted in which patients with diabetes and depression were randomized to a so-called multifaceted psychiatric intervention or to usual care.[29•] The intervention group was given the options of counseling, a case conference or referral to a psychiatrist. Antidepressant medication was an option in all treatment conditions. The results were significantly better for the intervention group regarding depression, but no positive effect on medical outcome was observed.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: