Depression as a Comorbidity to Diabetes: Implications for Management

Andrea A. Riley; Mindy L. McEntee; Linda Gerson; Cheryl R. Dennison


Journal for Nurse Practitioners. 2009;5(7):523-535. 

In This Article

Screening and Diagnosis

Although the link between depression and diabetes remains elusive, adverse effects can be minimized by adequate recognition and treatment. Because both conditions can produce similar somatic symptoms, concern has been raised that current diagnostic criteria for depression are not appropriate for the diabetic population.[80] While rates of sympathetic arousal, gastrointestinal complaints, and fatigue are higher in depressed individuals with co-occurring medical conditions, the differences are modest and did not significantly affect the diagnostic process.[81,82] Depressed diabetics experienced a similar course of depression and were just as likely to respond to treatment as depressed individuals without diabetes, even though they reported lower levels of somatic well-being and contentment.[83] Moreover, mean scores and symptom profiles on the Beck Depression Inventory (BDI) did not significantly differ between depressed individuals with diabetes and those without.[80] These findings suggest that diagnostic criteria and screening methods currently accepted for use in the general population are also suitable for diabetic patients.

Current research suggests only 25% to 50% of diabetic patients with depression are identified in primary care.[32,84,85] The American Diabetes Association (2008) advises regular screening throughout the course of diabetes management: at diagnosis, routine management visits, hospitalizations, when complications develop, or when problems with glucose control, quality of life, or adherence to self-care are identified.[86] Administration of a brief depression screening tool should be included with annual screening for diabetes-related parameters.

Selectively screening individuals based on the presence of known risk factors alone is not recommended, as presence of these factors is not as reliable as use of brief screening tools in detecting depression.[87] Providers should, however, be mindful of certain triggers in patients' medical history that may warrant additional screening, including unexplained psychosomatic symptoms, history of depression, comorbid psychological illnesses, chronic pain, substance abuse,[87] or reported diabetes symptoms disproportionately high with respect to the course of diabetes.[16] Patients with a personal history of depression or current anxiety disorder are significantly more likely to have depression and may warrant a diagnostic interview instead of initial screening.[87]

Depression Screening Tools

While guidelines agree routine screening for depression is necessary, they provide little guidance on selecting the best measure(s) to do so.[86,87] Several self-report questionnaires have been validated to assess depressive symptoms in primary care patients, although their use in diabetic populations has received considerably less study. Nevertheless, several screening tools have demonstrated sound psychometric properties detecting depression in samples of primary care patients with diabetes ( Table 1 ). In spite of differences in length, wording, and scoring criteria, each of these measures can be completed and scored within 5 to 10 minutes, allowing providers to follow up on the results of screening during the same visit.

Reliability and validity data for these instruments is also summarized in Table 1 . Cutoff scores optimizing the sensitivity and specificity of these measures were determined through receiving operator characteristics (ROC) analysis, while area under the ROC curve (area under the curve [AUC] index) assessed how well an instrument could discriminate between depressed and nondepressed individuals. Small variations in these statistics were reported across studies, though overall there was little difference between primary care (sensitivity 81–97%, specificity 63–99%, AUC 0.89–0.95) and diabetic patients (sensitivity 66–100%, specificity 52–92%, AUC 0.80–0.94). Because there is little evidence to suggest that any one of these measures is superior to the others, PCPs may wish to use the 2-item Patient Health Questionnaire (PHQ-2) as an initial screening tool for diabetic patients ( Table 2 ). For those who indicate they have been bothered in the last 2 weeks from either "little interest or pleasure in doing things" or "feeling down, depressed, or hopeless," providers can investigate further by administering the 9-item Patient Health Questionnaire-9 (PHQ-9) ( Table 3 ). This combination of tools, available in public domain, provides an effective method of screening for depression that also addresses time constraints frequently encountered by PCPs.