Background/Purpose: Patients with rheumatoid arthritis (RA) or spondyloarthritis (SpA) are more likely to report depression than the general population, and depression has been associated with poorer outcomes^1,2. A multidimensional health-assessment questionnaire (MDHAQ) has been found informative in many rheumatic diseases³, and includes two queries to screen for depression.  We compared these 2 MDHAQ depression items to 2 widely-used depression indices: the Patient Health Questionnaire (PHQ-9), and the Hospital Anxiety Depression Scale (HADS).

Methods: Consecutive patients attending two rheumatology clinical settings with a primary diagnosis of either RA or SpA completed a MDHAQ, which includes a 60 Yes/No symptom checklist (one symptom is depression), and a psychological index as the sum of 3 queries in the patient-friendly HAQ format (ranging from “without any difficulty” to “unable to do”) concerning dealing with depression, anxiety and sleep quality (each scored 0-3.3 for a total score=0-9.9). Patients also completed the PHQ-9, a depression screening tool (total=0-27, ≥10=depression), and HADS, a 14-item screening tool, 7 for depression (HADS-D) and 7 for anxiety (HADS-A) (total=0-21, ≥8=depression). Spearman correlations between the MDHAQ-psychological index and its 3 individual index items versus PHQ-9 and HADS were computed. Agreement and confidence intervals (CI) between MDHAQ-depression in the symptom checklist versus PHQ-9 ≥10 and HADS-D ≥8 as positive screens for depression was calculated.  Receiver-operator characteristic (ROC) curves were analysed for the MDHAQ psychological index, using PHQ-9≥10 or HADS-D≥8 as reference standards.

Results: Among 170 patients studied, 102 had RA and 68 had SpA, 82% and 47% females, mean age 58.8 and 55.6 years, respectively.  In RA, depression screening was positive in 26% of patients by MDHAQ-depression, 28% by PHQ-9 ≥10, and 33% by HADS-D ≥8, versus 35%, 43%, and 45%, respectively in SpA.  Agreement between MDHAQ-depression and PHQ-9≥10 was 83% (kappa= 0.60, 95% CI 0.47, 0.73), and with HADS-D≥8 was 79% (kappa= 0.52, 95% CI 0.39, 0.66). Agreement between PHQ-9≥10 and HADS-D≥8 was 82% (kappa= 0.60 95% CI 0.48, 0.72).  Correlations between the MDHAQ psychological index and individual items with PHQ-9 and HADS ranged from 0.62 to 0.82 (p< 0.001) (Table). The area under the ROC curve for the total MDHAQ psychological index was 0.91 (95% CI 0.87, 0.95) with PHQ-9≥10 and 0.86 (95% CI 0.79, 0.91) with HADS-D≥8 (Figure).

Conclusion: The prevalence of positive screening for depression was higher in SpA patients versus RA patients according to each screening tools. MDHAQ self-report of depression showed good agreement with PHQ-9 ≥10 and HADS ≥8 to screen for depression.  The three psychological queries on the MDHAQ can be combined into a psychological index to provide a useful clue to screen for depression in busy clinical settings.

References: 1.Matcham F. Rheumatology 2013,52:2136-48. 2.Zhao S. Arthritis Res Ther 2018,20:140. 3. Castrejon I. Bull Hosp Jt Dis 2017;75:93-100.

TABLE DEF MDHAQ ALL

Figure_ACR Abstract_ROC curves be

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/multidimensional-health-assessment-questionnaire-mdhaq-as-an-effective-screening-tool-to-identify-concomitant-depression-in-patients-with-rheumatoid-arthritis-and-spondyloarthritis-in-routine-care/