Depression in Adults With Diabetes

Patients with diabetes are twice as likely to experience depression as those without diabetes (Anderson et al., 2001). The odds of depression are similar in type 1 and type 2 diabetes and are significantly higher in women than in men. Moreover, the psychiatric condition presents as major depression or elevated depression symptoms in 11% and 31%, respectively, of adults with diabetes (Anderson et al., 2001). The course of depression in diabetes tends to be severe, with recurrences being the norm and not the exception. Following successful treatment, fewer than 10% of patients remain depression-free over the ensuing five years, and afflicted patients tend to suffer about one episode of major depression annually (Lustman et al., 1997c).

Etiology

The exact cause of depression in patients with diabetes is unknown, but it is likely complex, involving interactions among psychological, physical and genetic factors. Diabetes is a challenging illness, and coping with the disease and its complications (such as loss of vision or sexual capacity) is difficult and may contribute to depression. While the clinical sentiment that depression is merely secondary to diabetes (i.e., "you'd be depressed, too, if you were this sick") is evident in some cases, this hypothesis insufficiently accounts for most cases of depression. The contribution of psychosocial factors to the initial onset of depression in diabetes is modest. However, in those with a prior history of depression, these factors significantly affect the risk of depression recurrence. Some cases of depression, particularly in type 1 diabetes, may be organic affective illnesses that reflect cerebrovascular, neurohormonal and neurotransmitter abnormalities caused by the medical illness (MacKenzie and Trulson, 1978; Popkin et al., 1988; Trulson and Himmel, 1985). There is also a greater risk of depression in patients with diabetes who have a family history of depression; this may be indicative of a genetic vulnerability (Lustman et al., 1987).

Last, it is possible that the coprevalence of these disorders results from depression and not from diabetes. A number of studies have shown increased insulin resistance in depressed patients without diabetes (Nathan et al., 1981; Wright et al., 1978). Recent prospective population studies that controlled for conventional risk factors (e.g., age, obesity, body mass index) found that depression was associated with a twofold increased risk of type 2 diabetes (Eaton et al., 1996; Kawakami et al., 1999).

Effects on Health

No matter what causes the co-occurrence of depression and diabetes, when they are both present, interactions at behavioral and physiologic levels are likely. Glucose dysregulation, depression, substance abuse, smoking, noncompliance, physical activity and weight gain all affect one another, and depression is related to all of the factors listed. There is evidence supporting all of these associations. In both cross-sectional and prospective studies, depression has been associated with poor glycemic control, the principle cause of diabetes complications (Lustman et al., 2000a). The average increase in glycosylated hemoglobin attributable to depression in these studies was approximately 1.0%, a clinically significant elevation (Lustman et al., 2000b). Depression is also associated with other factors related to poor glycemic control, such as physical inactivity, obesity and treatment noncompliance, which are all factors that could contribute to insulin resistance (Farmer et al., 1988; Frederick et al., 1988; Littlefield et al., 1992; Lustman et al., 1992).

Given depression's effects on glycemic control, it is not surprising that it is linked to the risk of diabetes complications. In a recent meta-analysis, depression was significantly associated with an increased risk for the occurrence of neuropathy, nephropathy, retinopathy and macrovascular disease in adults with diabetes (range of effect sizes (r)=0.17 to 0.32, p<0.05 for all) (de Groot et al., 2001). Prospective studies have found that depression accelerates the presentation of coronary heart disease in women with diabetes (Lloyd et al., 1997).

Depression influences perceptual processes and is thought to make people less tolerant of pain and physical symptoms. For example, neuropathic pain that would normally be only an occasional annoyance may feel constant and overwhelming during an episode of depression and can result in a visit to the doctor. Mood at the time of clinical evaluation is often a stronger predictor of symptoms than the biomedical markers that are presumed to cause these symptoms. In this manner, depression may influence the presentation of physical symptoms.

Recognition

According to the DSM-IV, the diagnosis of depressive disorder is assigned when five or more criterion symptoms occur, are severe and persist over a period of at least two weeks. Sad mood or a significant loss of interest or pleasure are the essential symptoms for the diagnosis, and one of them must be present along with at least four other criterion symptoms. Symptoms attributable to the medical condition cannot count toward the diagnosis. This rule discourages overdiagnosis of depression, which is important since hypo- and hyperglycemia may produce some of the symptoms of depression (e.g., fatigue, weight loss). Diabetes does not, however, directly produce the two sin qua non symptoms of depression (sadness and loss of interest or pleasure), thus reducing the likelihood of false-positive depression diagnoses.

Despite substantial evidence of its adverse effects on the presentation and course of diabetes, and new evidence that treatment of depression has therapeutic effects on both mood and diabetes (see treatment section), major depression is recognized and treated in only about one-third of cases (Lustman et al., 1997a). This situation has been blamed on a variety of factors, including the presumption by clinicians that depression is merely a secondary reaction to the medical illness, the view (often shared by patients and family members) that the diagnosis conveys additional burden, and the stigma attached to having a mental illness. Primary care physicians may also lack sufficient training and face financial disincentives to perform psychodiagnostic testing. The fact that the average physician-patient interaction in the United States is currently eight to 12 minutes long (Kathol R, personal communication, May, 1996) also contributes to poor depression recognition. Despite these rather formidable obstacles, simple and effective methods are available to readily identify those patients most likely to benefit from depression treatment. These include simple paper-and-pencil screening instruments, like the 21-item Beck Depression Inventory (BDI), which are self-administered and take under 10 minutes to complete. When used in the physician's office, these tests are accurate and sensitive for the detection of depression.

Treatment