Psychometric properties of HSCL-5 and HSCL-10 to detect depression symptoms in Primary Health Care in Spain

Maria Rodriguez Barragan, María Isabel Fernández San Martín, Ana Clavería Fontán, Eva Peguero Rodríguez, Patrice Nabbe, Jean Yves Le Reste

Keywords: Primary Health Care; Family Practice; General Practitioners; Depression; Depressive Disorder; Questionnaires; Psychometrics; Validation Studies.

Background:

As a collaborative project of the Family Practice Depression and Multimorbidity group of European General Practice Research Network, the Hopkins Symptom Checklist-25 (HSCL-25) was identified as reproducible, reliable, and easy to use. It was translated and adapted to 13 European languages. There are different short versions of this scale that can be useful in the Primary Care (PC) settings.

Research questions:

What are the psychometric properties of the Spanish version of HSCL-10 and HSCL-5?

Method:

Multicentre, cross-sectional study. HSCL-25 was administered to outpatients aged from 45 to 75 in six Spanish PC Centres. All patients also conducted the structured Composite International Diagnostic Interview (CIDI). HSCL-10 and HSCL-5 were assessed and compared to HSCL-25 regarding total score correlation, internal consistency, and criterion validity against gold standard CIDI.

Results:

767 patients out of 790 complimented HSCL-25 (97.1% response rate). Mean age was 58.4 years (± 8.2), 54.4% were women. 736 participated in the CIDI interview. Total score mean was 1.57 (SD 0.45) for HSCL-25, 1.55 (SD 0.50) for HSCL-10, and 1.72 (SD 0.59) for HSCL-5. Pearson’s correlation was 0.955 (CI 95%, 0.95–0.96%) between HSCL-25 and HSCL-10, and 0.91 (CI 95%, 0.90–0.92%) between HSCL-25 and HSCL-5. Cronbach’s alpha was 0.84 for HSCL-10 and 0.77 for HSCL-5. Sensitivity was 79.7% (CI 95%, 67.2-89.0%) for HSCL-10, and 78.0% (CI 95%, 65.3-87.8%) for HSCL-5, whereas specificity was 83% (CI 95%, 80.0-85.8%) for HSCL-10, and 72.8% (CI 95%, 69.3-76.1%) for HSCL-5. Area under the curve (AUC) against CIDI was 0.88 (CI 95%, 0.84–0.92%) for HSCL-10, and 0.85 (CI 95%, 0.81–0.89%) for HSCL-5.

Conclusions:

Spanish versions of the HSCL-10 and the HSCL-5 show acceptable psychometric properties to diagnose depression in PC and could be used for daily practice and epidemiological research.

Points for discussion:

Are these results similar in other language versions of HSCL-10 and HSCL-5?

Are short versions useful tools in Primary Care?

#109