Glycemic control and major depression in patients with type 1 and type 2 diabetes mellitus

Abstract

The current study evaluated the association of glycemic control and major depression in 33 type 1 and 39 type 2 diabetes mellitus patients. Type 1 patients with a lifetime history of major depression showed significantly worse glycemic control than patients without a history of psychiatric illness (t=2.09; df=31, p<0.05). Type 2 diabetes patients with a lifetime history of major depression did not have significantly worse control than those with no history of psychiatric illness. Findings from this study indicate different relationships between lifetime major depression and glycemic control for patients with type 1 and type 2 diabetes. Treatment implications for glycemic control in type 1 and type 2 diabetes patients are discussed.

Introduction

Depressive disorders have been found to occur at increased prevalence rates among patients with type 1 (formerly known as juvenile-onset) and type 2 (formerly known as adult-onset) diabetes mellitus. Lifetime prevalence rates of major depression among type 1 and type 2 diabetes patients range between 14.4% and 32.5% 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11. Despite variation in the degree of prevalence in this population, these studies suggest that a sizable proportion of patients with diabetes show evidence of clinically meaningful depressive disorders.

A few studies have explored the relationship between depression and key medical variables in diabetes. One such critical indicator is glycemic control. Prolonged poor glycemic control has been demonstrated to result in more rapid onset and progression of retinopathy, neuropathy, and nephropathy 12, 13, 14, 15, 16, which have been linked with decreases in quality of life 12, 17, 18 and increased mortality [19]. Studies documenting the impact of depression on glycemic control have varied in the methods of assessment, from self-reported questionnaire data 20, 21, 22 to standardized psychiatric interview protocols 2, 4, 23. Although self-report questionnaires provide symptom endorsement, they do not allow clinical evaluation of the degree to which symptoms are organized into episodes, impair functioning, or represent somatic manifestations of diabetes-related complications unrelated to mood syndromes.

Three principal studies have utilized psychiatric interviews for the diagnosis of major depression 2, 4, 23. Lustman and colleagues [2] found, in a mixed group of type 1 and type 2 patients, those with a diagnosis of major depression within 12 months of diagnostic interview had significantly worse glycemic control (HbA1) than patients with no history of any psychiatric illness. Psychiatric illness was evaluated using the Diagnostic Interview Schedule (DIS) by clinical interviewers who were blind to glycemic status.

Robinson and colleagues [4] studied a combined sample of type 1 and type 2 adults to examine the prevalence of depression and presence of complications in four groups of diabetes patients compared with controls: normals; depression cases; subthreshold or borderline depression cases; and past history or recovered cases. A modified version of the Present State Exam (PSE) was used to assess psychiatric illness. The investigators did not find significant differences in current glycemic control, as indexed by HbA1, among normal controls and patients who were currently depressed cases, borderline depression, or patients who had a past history of depression. Significant differences in mean postprandial blood glucose values were found. Patients with borderline depression reported the highest blood glucose values. Patients with a past history of depression reported the lowest mean blood glucose values.

Finally, in an investigation of the relationship between lifetime history of psychiatric illness and the severity of early retinopathy, Cohen et al. [23] compared HbA1 values among three type 1 samples: those with a history of any psychiatric illness; those with a lifetime history of affective illness; and those with a lifetime history of major depression. Cohen’s group found significantly worse glycemic control in the affective illness and any psychiatric history samples when compared to a psychiatric-illness-free reference sample. The small subgroup of patients with a lifetime history of major depression showed elevated glycemic levels compared to the reference sample, but the difference did not meet statistical significance.

Variable findings from these studies may be attributable to differing samples. Lustman et al. [2] and Robinson et al. [4] used combined type 1 and type 2 samples to conduct their analyses, whereas Cohen et al. [23] selected a type 1 sample with limited disease duration (8 to 15 years) in order to examine the effect of psychiatric illness history and the development of diabetic retinopathy. The combination of type 1 and type 2 patients into a single sample poses challenges to understanding a relationship of glycemic control to depression. Type 1 and type 2 diabetes are distinct diseases with differing etiologies, ages of onset, risk factors, and treatment regimens. Among patients with type 2 diabetes, patients vary considerably in their prescribed treatment regimens; that is, diet only, oral agents, and/or insulin injections. Given these differences, the relationship between glycemic control and depression may differ significantly between diabetes types and across diabetes treatment regimens. Neither Robinson [4] nor Lustman [2] conducted separate analyses by diabetes type. Thus, it is not yet known whether the relationship of depression to glycemic control differs in type 1 and type 2 diabetic patients.

The timeframe for the diagnosis of major depression episodes also differed across these studies. Lustman and colleagues reported significant findings for patients with a recent history of major depression, but did not report results for patients with a lifetime history of major depression. In the studies that examined lifetime history of major depression, Cohen et al. [23] reported a nonsignificant trend toward worsened glycemic control, whereas Robinson et al. [4] reported worsened postprandial blood glucose levels for patients with a past history of depressive illness. The correspondence of time and glycemic control is a potentially important indicator of an underlying mechanism that might be responsible for a relationship between psychiatric history and glycemic control. An enduring effect of past episodes of major depression on glycemic functioning would have important consequences for the development of long-term diabetes complications such as retinopathy, neuropathy, and nephropathy. Further exploration of a possible long-term effects is necessary to clarify this point.

This study was proposed to examine the relationship of glycemic control to both current and lifetime major depression in a cohort of adult patients with type 1 and type 2 diabetes. Specifically, we hypothesized that: (1) type 1 patients with current major depression would show worse glycemic control than type 1 patients with no history of psychiatric illness; (2) type 2 patients with current major depression would show worse glycemic control than type 2 patients with no history of psychiatric illness; (3) type 1 patients with a lifetime history of major depression would show worse glycemic control than type 1 patients with no history of psychiatric illness; and (4) type 2 patients with a lifetime history of major depression would show worse glycemic control than type 2 patients with no history of psychiatric illness.