Recognizing and Managing Depression in Primary Care A Standardized Patient Study

Standardized patients also recorded the amount of time spent with the physicians and the charges for the visit. Three weeks after the final study visit, physicians were informed that the standardized patient visits had occurred. They were asked to either describe or name the patient. If the patient was detected as being a standardized patient, physicians were asked when during the encounter the detection occurred. Detection occurred in 22.8% of cases. The majority of detections occurred at the end of the second visit or in retrospect after the second visit had occurred, which likely did not influence the performance of the physicians to a significant degree; therefore, all detected visits were included in the analysis. If physicians did not recommend follow-up, the standardized patients initiated a follow-up visit by calling the practice and noting that their symptoms, though improved somewhat, were still present. All physicians forwarded standardized patient medical records for abstraction and took part in a telephone debriefing during which we determined what alerted them to this diagnosis or why they did not recognize depression.

For classifying physicians as recognizing depression, the following indicators were used: discussion of depression with the standardized patient, diagnosis of depression in the medical record, prescription for an antidepressant, or a recommendation for ongoing counseling with a social worker, psychiatrist, or psychologist.

Analysis

Descriptive statistics were used to characterize study participants and encounters. Comparisons between scenarios A and B for pursuit of presenting complaints, DSM-III-R criteria, other diagnostic criteria, and depression management were evaluated using the chi-squared or Fisher’s exact test when indicated. Time and charges were analyzed using t tests and Pearson’s product moment correlation coefficient. Analysis of data on scheduling follow-up between the first and second visits across cases was performed using 1-way analysis of variance (mixed model). All tests comparing performance across the scenarios were 2-tailed. An a level of 0.05 was considered statistically significant, except where more than one test was used on the same variable (eg, sleep and other criteria comparisons analyzed by scenario and to identify predictors of a diagnosis of depression in a pooled analysis). In these cases, the a levels were set at 0.025. Inter-observer reliability was determined using percent agreement.

Results

One hundred forty-nine internists and family physicians meeting eligibility requirements were visited by an unannounced standardized patient. Sixty-nine physicians participated in northern New England, 50 in Seattle, and 30 in Tuscaloosa, Alabama. Physicians from northern New England were overrepresented because of specialty and sex-specific recruitment difficulties at other sites. The mean age of physician participants was 42.8 years (standard deviation [SD] = 7.7); 67.5% were male. Fifty-one percent were family physicians; the other 49% were internists. Ninety-two percent were board certified. The mean number of years at the current practice location was 9.4 (SD = 7.0), and the estimated mean number of adult patients seen per week was 90.5 (SD = 36.1). Participants reported that 80% of their time was devoted to primary care. A total of 288 visits were made by our standardized patients: 135 in northern New England, 93 in Seattle, and 60 in Tuscaloosa. Second visits for 10 of the participants could not be made within the study period.

Recognition of Depression

One hundred percent of physicians visited by scenario A patients diagnosed depression, 91% at the first visit. Depression was diagnosed without a specific prompt in 49% of scenario B patients (38 of 77). Twenty-six percent of the diagnoses occurred during the first visit and 23% during the second visit. An additional 30% of providers (23 of 77) diagnosed depression in scenario B after the prompt, “I’ve had a really tough year,” was delivered. In 3% (2 patients) of scenario B encounters, the standardized patients were unable to deliver the prompt, and in 18% (14 of 77) depression was not diagnosed according to our a priori criteria, even though the prompt was delivered.

Table 2 shows the use of DSM-III-R criteria by physicians according to patient presentation. Among patients given a diagnosis of a depressive disorder, including all 72 who presented with scenario A and the 61 of 77 who presented with scenario B, physicians made similar use of criteria. No differences were noted in recognizing depression by specialty (family physician or internal medicine), and no statistical differences were noted in the characteristics of physicians assigned to scenario A or scenario B.

When we clustered depression criteria as outlined in DSM-III-R 26 into physical/neurovegetative symptoms (sleep disturbance, weight changes, attention/concentration problems, fatigue, psychomotor disturbance) or psychiatric/psychosocial symptoms (depressed mood, anhedonia, guilt feelings, suicidal thoughts) and assessed inquiry about them for association with a diagnosis of depression, both sets of criteria were statistically significant at P <.001. The mean number of physical/neurovegetative symptoms asked about in diagnosing depression was 3.4; the mean number of psychiatric/psychosocial symptoms was 1.8. If fewer than 2 symptoms were asked about during the encounter, depression was not diagnosed.