A direct comparison of presenting characteristics of depressed outpatients from primary vs. specialty care settings: Preliminary findings from the STAR*D clinical trial

Purpose: No study has directly compared the clinical features of depression for patients entering clinical trials using identical enrollment criteria at primary care (PC) and specialty care (SC) settings. The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study (http://www.star-d.org) provides a unique opportunity to provide this comparison for patients with a major depressive disorder (MDD) requiring treatment. Subjects and Methods: We report baseline data for the first 1500 patients enrolled in this trial involving 41 clinic sites (18 PC, 23 SC). Broadly inclusive eligibility criteria required that patients have a DSM-IV diagnosis of nonpsychotic MDD, have not failed an adequate medication trial during their current episode and score ≥14 on the 17-item Hamilton Rating Scale for Depression (HAM-D₁₇). Primary outcomes included the 30-item Inventory of Depressive Symptomatology-Clinician-Rated (IDS-C₃₀) and the HAM-D₁₇. Results: Specialty care and PC patients had equivalent degrees of depressive severity (IDS-C₃₀=35.8; HAM-D₁₇=20.4). Specialty care patients were almost twice as likely to report a prior suicide attempt than PC patients (21% vs. 12%, P<.0001) and slightly less likely to endorse suicidal ideation in the past week (45.0% vs. 50.8%, P=.006). The only other distinguishing core symptoms were a slightly lower likelihood of PC patients endorsing depressed mood (95.2% vs. 97.7%, P=.032) or anhedonia (66.3% vs. 70.7%, P=.042, IDS-C₃₀) and a lower likelihood of PC patients endorsing weight loss (IDS-C₃₀). HAM-D₁₇ results were identical. Conclusion: Depressive severity was not different, and symptomatic presentations did not differ substantially. Major depressive disorder is more similar than different among patients at SC and PC settings. Thus, similar clinical and research methods for screening, detecting and measuring treatment outcomes can be applied in both settings.