Detection and Treatment of Depression in a VA Primary Care Clinic

An estimated 6 to 8 percent of patients seen in primary care settings suffer from depression (1). Among untreated psychiatric disorders, major depression has been found to have the greatest negative impact on primary care patients' physical, social, and emotional functioning (2). Depression among medical patients is associated with a poorer prognosis for other medical conditions (3) and increased outpatient and emergency room visits (4).

Despite demonstrations of efficacious treatments for depressed primary care patients (5), depression is often unrecognized or misdiagnosed in primary care settings, and treatment of depression in these settings is often inadequate (6).

To facilitate the diagnosis and treatment of depression in outpatient settings, in 1993 the Agency for Health Care Policy and Research (AHCPR) published guidelines for the detection, diagnosis, and treatment of depression in primary care (7). In 1997 the Department of Veterans Affairs' Veterans Health Administration released a clinical guideline for major depressive disorder (8). Both the AHCPR and the VA guidelines target depressive disorders in primary care, but the VA guideline is more current and offers greater depth and breadth. The VA guideline addresses depressive disorders in the specialty mental health sector, as well as comorbid conditions such as substance use disorders, which are frequently encountered in the health care sector at large, and posttraumatic stress disorder, often encountered among patients served by the Veterans Health Administration.

In the study reported here, a pilot implementation of the VA guideline for major depressive disorder was conducted in one primary care clinic within a VA medical center. The study examined how this guideline affected detection and treatment of depression for primary care outpatients.

Methods

The procedure followed the clinical pathway for primary care clinics outlined in the VA clinical guideline for major depressive disorder. During January and February 1997, a total of 574 male veteran outpatients were screened for depression at one of the Salem VA Medical Center's primary care clinics using a three-item screening measure developed by Rost and associates (9). The initial screening constituted the first-stage assessment of depression. Patients screened positive for depression if they had suffered one or more days of depression during the past week and if they indicated either that they had experienced depression for two weeks or more during the past year or that they had experienced depression for two or more years of their life and much of the past year. In previous studies of medical outpatients, this scale demonstrated adequate sensitivity (80 percent) and specificity (95 percent), positive predictive value (33 percent), and negative predictive value (99 percent) (9).

Results

Of the 574 patients screened, 501 patients, or 87 percent, screened negative for depression, and 73 patients, or 13 percent, screened positive. Patients who screened positive then completed the Depression Outcomes Module (DOM), Part B (10), as a second-stage assessment of depression. The 11 items in this inventory address the presence and severity of symptoms of a major depressive episode as defined by DSM-IV. These items were summed to yield a total score, ranging from 11 (asymptomatic) to 42 (many severe symptoms).

Patients who met DSM-IV criteria for major depressive disorder and patients who did not meet the full DSM-IV criteria but who were still considered by their physician to be clinically depressed were then assessed for their willingness to receive treatment. Of the 73 patients with positive screens, 22, or 30 percent, were not treated at the VA hospital for various reasons; eight patients refused treatment, four were already receiving treatment, and ten evidenced subclinical levels of depression. Fourteen patients, or 19 percent, were referred within the hospital for treatment by psychologists (eight patients) or psychiatrists (six patients). However, 37 patients, or 51 percent, were willing to receive treatment only through the primary care team. Of these 37 depressed patients treated within primary care, seven, or 23 percent, met DSM-IV criteria for major depression, and 30, or 77 percent, did not.

Analysis of treatment outcome used data from readministration of the DOM Part B during follow-up visits with a sample of 16 of the 37 patients treated in the primary care setting. Improvement in severity of depressive symptoms was measured by changes in DOM Part B scores from baseline to follow-up.

All 16 patients in the follow-up sample were male veterans who ranged in age from 41 to 80 years, with a mean of 63.1 years. Ten patients had disabilities related to their military service. Follow-up visits occurred an average of 63 days after initiation of treatment, with a range from 16 to 199 days.

Sertraline was prescribed for all 16 patients for whom follow-up data were collected, at either 100 mg daily, for 11 patients, or 50 mg daily, for five patients.

Data were analyzed using one-tailed repeated-measures t tests for dependent samples. The mean±SD score on the DOM Part B for the 16 patients decreased from 24.13±4.77 at baseline to 22.13±4.15 at follow-up, a 15 percent change in scores and a significant difference (t=1.85, df=15, p<.05).

Of the 11 items in the DOM Part B, only those reflecting mood and anhedonia showed a significant difference in scores from baseline to follow-up. Scores for item 1, "How often in the past four weeks have you felt depressed, blue, or in low spirits for most of the day?" decreased significantly from a mean±SD of 2.88±.81 to 2.13±.62, representing a 40 percent change in score (t=5.2, df=15, p<.001). Similarly, for item 2, "How often in the past four weeks did you have days in which you experienced little or no pleasure in most of your activities?" scores decreased significantly from a mean±SD of 2.75±.77 at baseline to 2.31±.79 at follow-up, reflecting a 25 percent change in score (t=1.96, df=15, p<.05). Other items in the scale did not show significant differences from baseline to follow-up.

Discussion and conclusions

Findings from this pilot study appear to offer preliminary support for the utility of implementing the VA clinical guideline for major depressive disorder in VA primary care settings. Of the 13 percent of patients who initially screened positive for depression, 70 percent were previously unidentified depressed patients who subsequently received treatment. Besides identifying major depression, the screen was useful in detecting patients without major depression who still had significant symptoms that warranted treatment.

More than half of the depressed veterans who were identified in the study were willing to receive treatment through the VA medical center's primary care clinic, and 19 percent were willing to be seen in psychology or psychiatry services. Patients treated in the primary care clinic who were seen for follow-up assessment showed but modest statistically significant improvement in overall severity of depression, with mood and anhedonia most markedly affected.

The quasiexperimental, uncontrolled study design and small subject sample treated within only one clinic limit interpretation of the outcome data. Sample size was limited due to the labor-intensive undertaking of manually tracking and flagging patients for follow-up assessments. It is possible that selection factors might create differences in the severity of depression at follow-up between the subjects who were examined and those who were not. Due to the limited sample size, it is unknown if varying levels of severity of depression would differentially affect treatment outcomes.

Future research would gain a more accurate picture of treatment response by including control groups, tracking patients seen outside of the primary care setting, and monitoring the type of depression experienced by patients and the treatment they receive.

Dr. Foster is a clinical psychologist with the Bedford-Somerset Mental Health and Mental Retardation Program in Somerset, Pennsylvania. Dr. Ragsdale, Dr. Dunne, Dr. Jones, Mr. Lentz, and Dr. Gilmore are with the Veterans Affairs Medical Center in Salem, Virginia. Dr. Ihnen is with the department of psychology at the Illinois Institute of Technology in Chicago. Address correspondence to Dr. Jones, Psychiatric Services, Veterans Affairs Medical Center, 1970 Roanoke Boulevard, Salem, Virginia 24153 (e-mail, [email protected]).