Arthritis Care & Research最新文献

Objective: Pain in spondyloarthritis reflects not only inflammatory activity but also social context. We quantified the association between a composite Social Factors Index (SFI) and patient-reported pain severity on the Visual Analog Scale (VAS).

Methods: We analyzed 2,042 patients from REGISPONSER (N = 1,514) and RESPONDIA (N = 528). Missing data were handled with multiple imputation (m = 20). The SFI combined education, employment, housing, income source, and Graffar class using penalized regression (sparse-group lasso). Mutual information (MI) quantified the SFI's relative importance. Propensity scores (logistic regression) supported 1:1 caliper matching (0.2 SD on the logit), comparing the highest (Q4) and lowest (Q1) SFI quartiles.

Results: Before matching, Q1 and Q4 included on average 918 and 829 patients per imputation, differing mainly in race (White 80.8% vs 86.8%; standardized mean difference [SMD] = -0.16) and country (Spain 66.2% vs 78.2%; SMD = -0.27). After matching, a mean of 745 pairs (range 731-761) were retained with excellent covariate balance (all |SMD| < 0.10). In the matched sample, patients in Q4 reported higher pain (average treatment effect on the treated +0.38 VAS points; 95% confidence interval [CI] 0.05-0.71; P = 0.023). Postmatching regression yielded an identical estimate (β = +0.37; 95% CI 0.04-0.70; P = 0.030). MI indicated the SFI carried more information about VAS than any single social variable. Stratified analyses suggested heterogeneity: Spain +0.31 (95% CI 0.08-0.71) and Latin America +0.62 (95% CI -0.22 to 1.45).

Conclusion: Social disadvantage, summarized by the SFI, is independently associated with greater pain severity in spondyloarthritis after adjustment. These findings support incorporating social determinants into clinical assessment and pain management.

Objective: To quantify the degree of financial distress and identify its determinants in adults with rheumatoid arthritis (RA) given the frequent prolonged use of expensive disease-modifying therapies.

Methods: We identified adults enrolled in the FORWARD databank with either RA or noninflammatory musculoskeletal disease (NIMSKD) completing the Functional Assessment of Chronic Illness Therapy-Comprehensive Score for Financial Toxicity (FACIT-COST) questionnaire. In this cross-sectional study, FACIT-COST was analyzed as a continuous (higher score indicates less financial distress) and binary variable (presence of financial distress with threshold <26). Least Absolute Shrinkage and Selection Operator (LASSO) was applied to linear and logistic regression to select covariates for inclusion in multivariable models.

Results: Participants with RA (n = 2,277) had lower FACIT-COST scores, indicating greater financial distress, than those with NIMSKD (n = 1,340) (mean of 30.2 ± SD 9.4 vs mean of 34.0 ± SD 8.4; unadjusted P < 0.001). Assessed as a binary outcome, financial distress was more frequent in participants with RA than participants with NIMSKD (29% vs 15%; unadjusted P < 0.001). Differences in financial distress by diagnosis persisted following multivariable adjustment. Among those with RA, determinants identified in multivariable models included depression (adjusted odds ratio 1.12; 95% confidence interval 1.09-1.16) and disease severity.

Conclusion: Financial distress is prevalent in adults with RA and appears to be greatest in those with comorbidities, specifically depression, identifying a potential area for intervention. Notably, expensive biologic or targeted synthetic disease-modifying antirheumatic drugs were not associated with FACIT-COST scores.

Objective: Ultrasound (US) has been proposed as a potential tool for assessing skin fibrosis in systemic sclerosis (SSc). However, a large-scale comparison of US-based assessment with histologic markers of skin fibrosis has not been reported. We evaluated US-based skin assessments for their face validity (differentiation between involved SSc and healthy control [HC] skin), construct validity (comparison to modified Rodnan skin score [mRSS]), and criterion validity (comparison to histologic and gene expression fibrosis markers).

Methods: Twenty HCs and 52 patients with SSc underwent clinical and US assessment followed by a forearm skin biopsy. Predefined areas were assessed on the finger, hand, and forearm bilaterally. mRSS, US, histologic and molecular (reverse transcription quantitative polymerase chain reaction) evaluations were performed by blinded, independent assessors. Dermal thickness, echogenicity, and elastography were assessed using a high-frequency GE LOGIQ P9 US machine.

Results: Except in the hand area, the US variables could not differentiate between HC and clinically affected SSc skin (face validity). There was only a weak to moderate correlation between US-based measurements and mRSS in the hand and finger areas (construct validity). US-based thickness showed moderate correlation with histologic thickness (ρ = 0.43; P = 0.002) but no statistically significant correlations with other histologic or gene expression markers of fibrosis in patients with SSc (criterion validity).

Conclusion: High-frequency US assessment of SSc skin does not show consistent and strong face, construct, or criterion validity. The role of this assessment tool remains limited, underscoring the need for further development of a quantitative and accurate tool for assessing SSc skin fibrosis.

Objective: This systematic review aimed to assess the diagnostic accuracy of algorithms used to identify rheumatoid arthritis (RA) and juvenile idiopathic arthritis (JIA) in electronic health records (EHRs).

Methods: We searched MEDLINE, Embase, and CENTRAL databases and included studies that validated case definitions against a reference standard such as rheumatologist-confirmed diagnosis or ACR/EULAR classification criteria. Title/abstract screening, full-text review, data extraction and quality assessment were all completed in duplicate. Results were synthesised narratively and using a bivariate random-effects meta-analysis of sensitivity and specificity.

Results: A total of 35 studies were included. Algorithms varied widely in complexity, ranging from single ICD codes to combinations including disease-modifying antirheumatic drugs (DMARDs), hospitalisation records, and specialist diagnosis. Algorithms combining ICD codes with DMARD prescriptions (pooled sensitivity= 0.79 95% CI 0.61-0.90, specificity= 0.96 95% CI 0.72-1.00, PPV= 0.78 95% CI 0.63-0.88) or requiring an ICD code assigned by a rheumatologist (pooled sensitivity= 0.91 95% CI 0.70-0.98, specificity= 0.94 95% CI 0.49-1.00, PPV= 0.70 95% CI 0.64-0.75) showed the highest accuracy, with balanced sensitivity, specificity, and positive predictive value (PPV). Less restrictive algorithms demonstrated high sensitivity but lower PPV. Substantial heterogeneity was observed across studies, likely due to differences in algorithm structure, data sources, and validation methods. Despite this variability, we used conceptually coherent categories to allow for meaningful synthesis, prioritising clinical interpretability.

Conclusions: These findings support the use of more specific algorithms when diagnostic certainty is essential and highlight the need for further validation of high-performing algorithms across diverse healthcare systems.

Objective: Antinuclear antibodies (ANAs) are present at high titers in 2% of the general population but their clinical significance in individuals without an autoimmune (AI) disease is not known. We tested the hypothesis that the presence of a high ANA titer in non- AI conditions is associated with disease.

Methods: We conducted a retrospective case-control study in the Vanderbilt University Medical Center's de-identified electronic medical record system. Individuals without AI disease who had an ANA test were classified into 3 groups: high titer (HT, ANA≥1:640), low titer (LT, ANA≤1:80), and negative ANA (NG). The prevalence of diagnoses recorded within 90 days of the ANA test were compared among groups in a phenome-wide association study (PheWAS) adjusting for age at ANA testing, sex, median body mass index (BMI), and reported race. A P-value <5x10^-5 was considered significant.

Results: A total of 28,781 individuals qualified for the study: 3.1% in the HT, 12.3% in the LT, and 84.6% in the NG groups. BMI was similar between groups (P-value=0.345), but individuals in the HT group were older (P=3.9x10^-73). A high ANA titer increased risk of 46 and 67 clinical diagnoses when comparing the HT group with the LT and the NG groups, respectively. The most significant associations in both comparisons included liver disorders/complications and risk factors for liver disease.

Conclusions: A high ANA titer in the absence of an AI disease was associated with increased risk of liver disorders and related risk factors and complications.

Objective: Opioid use remains common despite effective therapies for axial spondyloarthritis (axSpA). We assessed whether early tumor necrosis factor inhibitor (TNFi) use is associated with reduced risk of chronic opioid use.

Methods: Using the Merative MarketScan Commercial Database containing health insurance billing claims data, we conducted a time-stratified, propensity score (PS)-matched cohort study of adults aged 18 to 65 years with axSpA. Early TNFi exposure was defined using pharmacy and medical claims as any incident TNFi use within 6 months of axSpA diagnosis. We used PSs to match early TNFi users 1:1 to those not starting a TNFi within 6 months in 1-year cohort accrual blocks. We compared the risk of chronic opioid use (≥90 days' prescription) among early TNFi users versus comparators using Cox proportional hazard models, overall and stratified by prior opioid use (within 12 months).

Results: We included 8,508 individuals with axSpA after PS matching (4,254 early TNFi initiators and 4,254 comparators) with a mean age of 42 years; 50% were female. Chronic opioid use occurred in 20.9% (22.3% early TNFi initiators and 19.6% comparators). Early TNFi initiators had 17% higher risk of chronic opioid use versus matched comparators (95% confidence interval [CI] 1.06-1.28), with higher risk of chronic opioid use for early TNFi initiators in the opioid-naive but not in the opioid-experienced stratum (hazard ratio [HR] 1.96; 95% CI 1.41-2.74 vs HR 1.06; 95% CI 0.96-1.17).

Conclusion: Early TNFi initiation was not associated with reduced risk of chronic opioid use versus later or no TNFi initiation. Confounding by indication in administrative claims data limit result interpretation.