Does a coexisting anxiety disorder predict persistence of depressive illness in primary care patients with major depression?

Abstract

We assessed whether a coexisting anxiety disorder predicts risk for persistent depression in primary care patients with major depression at baseline. Patients with major depression were identified in a 12-month prospective cohort study at a University-based family practice clinic. Presence of an anxiety disorder and other potential prognostic factors were measured at baseline. Persistent depressive illness (major depression, minor depression, or dysthymia) was determined at 12 months. Of 85 patients with major depression at baseline, 43 had coexisting anxiety disorder (38 with social phobia). The risk for persistent depression at 12 months was 44% greater [Risk Ratio (RR) = 1.44, 95% confidence interval (CI) 1.02–2.04] in those with coexisting anxiety. This risk persisted in stratified analysis controlling for other prognostic factors. Patients with coexisting anxiety had greater mean depressive severity [repeated measures analysis of variance (ANOVA), p < 0.04] and total disability days (54.9 vs 19.8, p < 0.02) over the 12-month study. Patients with social phobia had similar increased risk for persistent depression (RR = 1.40, 95% CI 0.98–2.00). A coexisting anxiety disorder indicates risk for persistent depression in primary care patients with major depression. Social phobia may be important to recognize in these patients. Identifying anxiety disorders can help primary care clinicians target patients needing more aggressive treatment for depression.

Introduction

Depressive disorders in primary care are common and carry substantial morbidity. Major depression, the most severe form of the depressive disorders, has a prevalence of 6%–8%, making it more common than hypertension in this setting [1]. It carries substantial morbidity, matching or exceeding that associated with other common chronic medical conditions [2].

Other depressive disorders less severe than major depression are also fairly common in primary care settings. Dysthymia, a chronic low grade depression lasting at least 2 years, is found in 2.1%–3.7% of primary care patients [3]. Minor depression, defined as two to four depressive symptoms over a 2-week period, is found in 3.6%–9.1% of this population 4, 5. Although frequently seen in the primary care setting and noted as a specific type of subthreshold depressive condition in the Primary Care version of the DSM-IV [6], minor depression is not currently a discrete DSM-IV diagnosis and is included under the category “Depressive disorder not otherwise specified” [7]. Both dysthymia and minor depression, though less severe than major depression, are associated with morbidity nearly as great as that with major depression and much greater than in patients without depressive symptoms 2, 8.

Not surprisingly, primary care clinicians play a pivotal role in managing depressive disorders. They provide nearly half of the outpatient care for patients with depression [9] and record approximately the same number of yearly patient visits for an antidepressant prescription as do psychiatrists [10].

Despite meeting criteria for depressive disorders, patients with depression seen in primary care settings have a milder course than do those seen in specialty mental health settings. Depressive illness may be less severe for the former group, with fewer psychiatric symptoms, a lower likelihood of having received prior treatment for depression, and a lower risk of lifetime psychiatric hospitalization 11, 12. These patients also have a better short-term prognosis, with a higher rate of response to treatment [13] and a greater chance of recovery at 1 year follow-up [12], and this improved prognosis appears independent of whether adequate antidepressant medication is provided [14].

Given that depression is so common and that many patients improve relatively quickly, a key issue for primary care physicians is to determine which patients are at risk for persistent depressive illness. Unfortunately, in primary care settings, there is little knowledge of the factors that influence the likelihood of persistent depression. Identified predictors include depressive severity and high neuroticism scores 3, 15, the presence of dysthymia [16], the presence of comorbid medical illness 3, 17, less education [18], less physical activity, unemployment, and low levels of social support [19], and poor social functioning 17, 20. Of these, only depressive illness and comorbid physical illness appear as potential areas for direct intervention by primary care physicians.

A potential predictor of persistent depression in the primary care setting which has received relatively limited examination is the presence of a coexisting anxiety disorder. Previous primary care research, while suggesting that current coexisting anxiety disorders should not be considered as predictors of persistence in a multivariate analysis, involved only panic disorder and generalized anxiety disorder (GAD), and follow-up was limited to 4 months [15]. A coexisting anxiety disorder remains a good candidate as a predictor of persistence for a number of reasons. First, anxiety disorders have been identified as a predictor of persistent depressive illness in mental health settings, where they have been associated with subsequent depressive morbidity 21, 22, decreased treatment responsiveness 22, 23, 24, 25, and greater social dysfunction [26]. Second, anxiety disorders commonly coexist with major depression in primary care settings, with reports of comorbidity ranging from 28% to 66% 15, 27, 28, 29, 30, 31. GAD is the most common, coexisting in 54%–62% of patients with major depression. Panic disorder co-occurs in 10%–44% and is felt to be the most clinically important. Social phobia and agoraphobia, however, have been less well studied. Most reports on major depression comorbid with an anxiety disorder have either grouped all anxiety disorders together 28, 29, grouped all phobic disorders together [31], screened for (rather than diagnosed) the anxiety disorder [31], or ignored anxiety disorders other than panic disorder and GAD 15, 30.

Third, a lifetime history of either a comorbid panic disorder or GAD does appear to be a risk factor for a worse course of illness in primary care patients with major depression. Patients with a lifetime history of panic disorder have greater depressive severity [32], greater impairment in physical and psychosocial functioning [32], and a poorer response to treatment than those with major depression alone 32, 33. Primary care patients with major depression with a lifetime history of either panic disorder or GAD tend to terminate treatment prematurely more often than those without a comorbid anxiety disorder 32, 33.

Finally, determining whether a coexisting anxiety disorder predicts persistence of depressive symptoms in primary care patients is clinically important. Unlike many other possible predictors of depression, anxiety disorders are treatable. Knowledge of this comorbidity can alert the clinician to a possible need to modify the treatment intervention and to more aggressively manage the patient’s depression.

The following then is our research question: For primary care patients with major depression, does a current coexisting anxiety disorder predict persistent depressive illness at 12-month follow-up?