Background/Purpose: Many studies have shown that rheumatologic conditions are associated with a higher risk of depression. The two most common methods to detect depression in these studies used DSM/ICD criteria; alternatively, self-report screening questionnaires, such as the Patient Health Questionnaire (PHQ). However, using one or the other method to detect depression might lead to decreased sensitivity to detect depression, and the statistical analysis can be fraught with difficulty since numerous confounding variables affect depression. In this study, we aim to implement both methods to detect depression in patients with Systemic sclerosis (SSc), Dermatomyositis (DM), and Fibromyalgia(FM) using propensity-score matching in a national cohort database.

Methods: ICD 9/10 codes were used to identify SSc, DM, and FM, of which a propensity score matching (PSM) analysis was used to evaluate the effect of disease on the risk of developing depression in a cohort of 240,353 Americans enrolled in the “All of Us database.” ICD 9/10 codes were used, and it can be inferred that the ACR classification criteria for the diseases were met. Covariates for matching included age, gender, race, ethnicity, income level, education level, marital status, cigarette exposure, alcohol exposure, drug exposure, and comorbidities. The control groups were matched with a ratio of 1 to each patient group based on propensity scores. The PSM reduces any confounding factors between the case and control cohorts. A logistic regression analysis determined the association between each disorder and depression. Global Mental Health Scores (GMHS) were compared between control and disease groups using a t-test or Mann-Whitney U test. GMHS is a patient-reported perspective of their mental health, with a score range of 0-20, with the maximum score representing optimum health status.

Results: Out of 240,353 individuals, 638 patients with SSc, 358 patients with DM, and 21,263 FM patients were identified. Control groups for each disease were statistically not different (p ≥0.05) from the case groups for age, gender, race, ethnicity, income level, education level, marital status, cigarette exposure, alcohol exposure, drug exposure, and Charlson comorbidity scores. Only FM was associated with increased risk of depression (OR 2.32 [95% CI: 2.23-2.41], p< 0.001), which was in line with the lower average GMHS of FM patients (13.1) compared to control (13.8) (p< 0.001). SSc (OR 0.94, 95% CI: 0.75-1.17, p=0.571) and DM (OR 1.07, 95% CI: 0.80-1.44, p=0.649) were not associated with an increased risk of depression. The average GMHS of SSc (13.6) and DM (13.4) patients were statistically not different from their respective control groups.

Conclusion: When used concurrently, diagnostic codes for depression and self-reported screening questionnaires were congruent in assessing the association of depression with common rheumatologic disorders. In addition, the use of propensity score matching helped eliminate confounding variables.

« Back to ACR Convergence 2023

ACR Meeting Abstracts - https://acrabstracts.org/abstract/assessing-risk-of-depression-in-common-rheumatologic-disorders-using-diagnostic-codes-survey-scores-and-propensity-score-matching-methodology/