Influence of Multimorbidity on New Treatment Initiation and Achieving Target Disease Activity Thresholds in Active Rheumatoid Arthritis: A Cohort Study Using the Rheumatology Informatics System for Effectiveness Registry

Changes in Long-Term Glucocorticoid Use Among Older Adults After New Diagnosis of Late-Onset Rheumatoid Arthritis

BACKGROUND

We evaluated changes in long-term glucocorticoid (GC) use and factors associated with persistent GC use in older adults with late-onset rheumatoid arthritis (LORA).

METHODS

Using 20% Medicare data from 2008 to 2017, we identified adults ≥66 years with a new diagnosis of LORA, disease-modifying antirheumatic drug (DMARD) use or at least two rheumatologist visits, and at least 12 months of follow-up data. Older adults were categorized as DMARD-exposed or DMARD-unexposed based on treatment during the 12 months after LORA diagnosis (index date). For each quarter after the index date, long-term GC use was defined as having oral GC prescriptions for at least 30 days with a dose >5 mg/day prednisone equivalent. We compared long-term GC use between quarter (Q)1 and Q4 and performed stratified mixed-effects logistic regression for factors associated with persistent GC use, defined as long-term GC use in Q2 to Q4.

RESULTS

The cohort included 15,425 individuals with two-thirds (62.5%) being DMARD-exposed. Between Q1 and Q4, the proportion of older adults on long-term GC declined from 44.1 to 24.9% (∆19.2%) among the DMARD-exposed and from 25.8 to 17.9% (∆7.9%) among the DMARD-unexposed. One year after the index date, 13.5% of the DMARD-exposed and 9.8% of the DMARD-unexposed were persistent GC users. In stratified mixed-effects logistic models, persistent GC use was associated with low-income subsidy status among the DMARD-exposed and with greater comorbidity burden among DMARD-unexposed.

CONCLUSION

Long-term GC use declined more among DMARD-exposed than DMARD-unexposed patients. One in seven DMARD-exposed and one in ten DMARD-unexposed have persistent GC use which is associated with financial barriers and multimorbidity that may limit the use of steroid-sparing DMARDs.

RheuMetric Physician 0 to 10 Estimates of Inflammation, Damage, and Patient Distress at Initial Versus Follow-Up Visits in Contemporary Rheumatology Care

OBJECTIVE

We aimed to analyze the RheuMetric physician 0 to 10 visual numeric subscale (VNS) estimates of inflammatory activity (DOCINF), organ damage (DOCDAM), and patient distress (DOCDIS) at initial and follow-up routine rheumatology visits for possible incremental information to clarify physician estimate of global assessment (DOCGL).

METHODS

A retrospective cross-sectional study compared mean DOCGL, DOCINF, DOCDAM, and DOCDIS and the percentage contributed by inflammation, damage, and distress to DOCGL (total = 100%) at initial and follow-up visits in 563 unselected routine care patients, classified into four diagnosis categories: inflammatory (rheumatoid arthritis, systemic lupus erythematosus [SLE], spondylarthritis, vasculitis, and gout), primary osteoarthritis (OA), primary fibromyalgia (FM), and "other" diagnoses. Differences between initial and follow-up visits were estimated using t-tests.

RESULTS

In all patients, mean DOCGL was 4.0/10, DOCINF 1.6/10, DOCDAM 2.9/10, and DOCDIS 2.4/10, indicating higher estimates for damage and distress than for inflammation, including in all inflammatory diagnoses other than SLE. Highest mean estimates were 2.2 for DOCINF in inflammatory diagnoses, 4.9 for DOCDAM in primary OA, 6.3 for DOCDIS in primary FM. However, DOCDAM was 2.8 (0.6 uniyts higher than DOCINF) in inflammatory diagnoses. RheuMetric estimates of inflammation were significantly higher at initial than at follow-up visits, and estimates of damage were significantly lower at initial than at follow-up visits in all patients and in those with inflammatory diagnoses. DOCGL did not differ significantly at initial versus follow-up visits.

CONCLUSION

DOCINF, DOCDAM, and DOCDIS add feasibly recorded, clinically relevant incremental information to DOCGL. Despite excellent contemporary control of inflammation, joint damage and patient distress remain important clinical problems in contemporary routine rheumatology care, documented by quantitative RheuMetric estimates.

Multimorbidity Patterns and Rheumatoid Arthritis Disease Outcomes: Findings From a Multicenter, Prospective Cohort

OBJECTIVE

To determine whether unique multimorbidity patterns are associated with long-term rheumatoid arthritis (RA) disease severity.

METHODS

We conducted a cohort study within the Veterans Affairs Rheumatoid Arthritis registry. We applied previously derived multimorbidity patterns based on the presence of diagnostic codes for relevant conditions prior to enrollment using linked administrative data. Disease activity and functional status were assessed longitudinally up to 5 years after enrollment. The association of multimorbidity patterns with disease activity and functional status were assessed using generalized estimating equations models adjusting for relevant confounders.

RESULTS

We studied 2,956 participants, of which 88.2% were male, 76.9% reported white race, and 79.3% had a smoking history. Mental health and substance abuse (β 0.12 [95% confidence interval {CI} 0.00, 0.23]), cardiovascular (β 0.25 [95% CI 0.12, 0.38]), and chronic pain (β 0.21 [95% CI 0.11, 0.31]) multimorbidity were associated with higher Disease Activity Score in 28 joints (DAS28) scores. Mental health and substance abuse (β 0.09 [0.03, 0.15]), cardiovascular (β 0.11 [95% CI 0.04, 0.17]), and chronic pain multimorbidity (β 0.15 [95% CI 0.10, 0.20]) were also associated with higher Multidimensional Health Assessment Questionnaire (MDHAQ) scores. The metabolic pattern of multimorbidity was not associated with DAS28 or MDHAQ. The number of multimorbidity patterns present was highly associated with DAS28 and MDHAQ (P trend < 0.001), and patients with all four multimorbidity patterns had the highest DAS28 (β 0.59 [95% CI 0.36, 0.83]) and MDHAQ (β 0.27 [95% CI 0.16, 0.39]) scores.

CONCLUSION

Mental health and substance abuse, chronic pain, and cardiovascular multimorbidity patterns are associated with increased RA disease activity and poorer functional status. Identifying and addressing these multimorbidity patterns may facilitate achieving RA treatment targets.

Real-World Effectiveness of Pegloticase Associated With Use of Concomitant Immunomodulatory Therapy

OBJECTIVE

The objective of this study was to ascertain pegloticase persistence and adverse events associated with concomitant immunomodulatory drug treatment in patients with gout.

METHODS

We conducted a retrospective analysis of patients with gout using the American College of Rheumatology's Rheumatology Informatics System for Effectiveness registry from January 2016 through June 2020. The first pegloticase infusion defined the index date. Based on concomitant immunomodulatory drug treatment, we identified three exposure groups: (1) immunomodulatory drug initiators (patients initiating an immunomodulatory prescription ±60 days from the index date), (2) prevalent immunomodulatory drug recipients (patients receiving their first immunomodulatory drug prescription >60 days before the index date with at least one prescription within ±60 days of the index date), and (3) immunomodulatory nonrecipients (patients receiving pegloticase without concomitant immunomodulatory drugs). We calculated the proportion of patients who achieved serum urate levels ≤6 mg/dL and who had laboratory abnormalities (white blood cell count <3.4 x 109/L, platelet count <135,000, hematocrit level <30%, alanine aminotransferase or aspartate aminotransferase level ≥1.5 times the upper limit normal value) within 180 days after the index date. Cox regression analyzed time to pegloticase discontinuation, controlling for potential confounders.

RESULTS

We identified 700 pegloticase recipients (91 immunomodulatory drug initiators, 33 prevalent immunomodulatory drug recipients, and 576 nonrecipients), with a median follow-up of 14 months. Immunomodulatory drug recipients were less likely to discontinue pegloticase. The adjusted hazard ratios of pegloticase discontinuation associated with concomitant immunomodulatory drug initiation and prevalent treatment were 0.52 (95% confidence interval [CI] 0.37-0.75) and 0.69 (95% CI 0.42-1.16), respectively. Laboratory abnormalities were uncommon (<5%) and were not higher in concomitant immunomodulatory drug treatment.

CONCLUSION

Consistent with clinical trials, results from this large observational registry suggest that concomitant immunomodulatory drug treatment improves pegloticase persistence.

PROMIS-29 in rheumatoid arthritis patients who screen positive or negative for fibromyalgia on MDHAQ FAST4 (fibromyalgia assessment screening tool) or 2011 fibromyalgia criteria

BACKGROUND

PROMIS-29 T-scores query health-related quality of life (HRQL) in 7 domains, physical function, pain, fatigue, anxiety, depression, sleep quality, and social participation, to establish population norms. An MDHAQ (multidimensional health assessment questionnaire) scores these 7 domains and includes medical information such as a FAST4 (fibromyalgia assessment screening tool) index. We analyzed PROMIS-29 T-scores in rheumatoid arthritis (RA) patients vs population norms and for positive vs negative fibromyalgia (FM) screens and compared PROMIS-29 T-scores to MDHAQ scores to assess HRQL.

METHODS

A cross-sectional study was performed at one routine visit of 213 RA patients, who completed MDHAQ, PROMIS-29, and reference 2011 FM Criteria. PROMIS-29 T-scores were compared in RA vs population norms and in FM+ vs FM- RA patients, based on MDHAQ/FAST4 and reference criteria. Possible associations between PROMIS-29 T-scores and corresponding MDHAQ scores were analyzed using Spearman correlations and multiple regressions.

RESULTS

Median PROMIS-29 T-scores indicated clinically and statistically significantly poorer status in 26-29% FM+ vs FM- RA patients, with larger differences than in RA patients vs population norms for 6/7 domains. MDHAQ scores were correlated significantly with each of 7 corresponding PROMIS-29 domains (|rho|≥0.62, p<0.001). Linear regressions explained 55-73% of PROMIS-29 T-score variation by MDHAQ scores and 56%-70% of MDHAQ score variation by PROMIS-29 T-scores.

CONCLUSIONS

Scores for 7 PROMIS-29 domains and MDHAQ were highly correlated. The MDHAQ is effective to assess HRQL and offers incremental medical information, including FAST4 screening. The results indicate the importance of assessing comorbidities such as fibromyalgia screening in interpreting PROMIS-29 T-scores.

Classifying Multimorbidity Using Drug Concepts via the Rx-Risk Comorbidity Index: Methods and Comparative Cross-Sectional Study

OBJECTIVE

The study objective was to update a method to identify comorbid conditions using only medication information in circumstances in which diagnosis codes may be undercaptured, such as in single-specialty electronic health records (EHRs), and to compare the distribution of comorbidities across Rx-Risk versus other traditional comorbidity indices.

METHODS

Using First Databank, RxNorm, and its web-based clients, RxNav and RxClass, we mapped Drug Concept Unique Identifiers (RxCUIs), National Drug Codes (NDCs), and Anatomical Therapeutic Chemical (ATC) codes to Rx-Risk, a medication-focused comorbidity index. In established rheumatoid arthritis (RA) and osteoarthritis (OA) cohorts within the Rheumatology Informatics System for Effectiveness registry, we then compared Rx-Risk with other comorbidity indices, including the Charlson Comorbidity Index, Rheumatic Disease Comorbidity Index (RDCI), and Elixhauser.

RESULTS

We identified 965 unique ingredient RxCUIs representing the 46 Rx-Risk comorbidity categories. After excluding dosage form and ingredient related RxCUIs, 80,911 unique associated RxCUIs were mapped to the index. Additionally, 187,024 unique NDCs and 354 ATC codes were obtained and mapped to the index categories. When compared to traditional comorbidity indices in the RA cohort, the median score for Rx-Risk (median 6.00 [25th percentile 2, 75th percentile 9]) was much greater than for Charlson (median 0 [25th percentile 0, 75th percentile 0]), RDCI (median 0 [25th percentile 0, 75th percentile 0]), and Elixhauser (median 1 [25th percentile 1, 75th percentile 1]). Analyses of the OA cohort yielded similar results. For patients with a Charlson score of 0 (85% of total), both the RDCI and Elixhauser were close to 1, but the Rx-Risk score ranged from 0 to 16 or more.

CONCLUSION

The misclassification and under-ascertainment of comorbidities in single-specialty EHRs can largely be overcome by using a medication-focused comorbidity index.

Assessing clusters of comorbidities in rheumatoid arthritis: a machine learning approach

BACKGROUND

Comorbid conditions are very common in rheumatoid arthritis (RA) and several prior studies have clustered them using machine learning (ML). We applied various ML algorithms to compare the clusters of comorbidities derived and to assess the value of the clusters for predicting future clinical outcomes.

METHODS

A large US-based RA registry, CorEvitas, was used to identify patients for the analysis. We assessed the presence of 24 comorbidities, and ML was used to derive clusters of patients with given comorbidities. K-mode, K-mean, regression-based, and hierarchical clustering were used. To assess the value of these clusters, we compared clusters across different ML algorithms in clinical outcome models predicting clinical disease activity index (CDAI) and health assessment questionnaire (HAQ-DI). We used data from the first 3 years of the 6-year study period to derive clusters and assess time-averaged values for CDAI and HAQ-DI during the latter 3 years. Model fit was assessed via adjusted R2 and root mean square error for a series of models that included clusters from ML clustering and each of the 24 comorbidities separately.

RESULTS

11,883 patients with RA were included who had longitudinal data over 6 years. At baseline, patients were on average 59 (SD 12) years of age, 77% were women, CDAI was 11.3 (SD 11.9, moderate disease activity), HAQ-DI was 0.32 (SD 0.42), and disease duration was 10.8 (SD 9.9) years. During the 6 years of follow-up, the percentage of patients with various comorbidities increased. Using five clusters produced by each of the ML algorithms, multivariable regression models with time-averaged CDAI as an outcome found that the ML-derived comorbidity clusters produced similarly strong models as models with each of the 24 separate comorbidities entered individually. The same patterns were observed for HAQ-DI.

CONCLUSIONS

Clustering comorbidities using ML algorithms is not computationally complex but often results in clusters that are difficult to interpret from a clinical standpoint. While ML clustering is useful for modeling multi-omics, using clusters to predict clinical outcomes produces models with a similar fit as those with individual comorbidities.

The Multimorbidity Web in rheumatoid arthritis

Multimorbidity, the presence of multiple chronic conditions, is highly prevalent in people with RA. An essential characteristic of multimorbidity is the interrelatedness of the different conditions that may develop in a multimorbid person. Recent studies have begun to identify and describe the Multimorbidity Web by elucidating unique multimorbidity patterns in people with RA. The primary multimorbidity patterns in this web are cardiopulmonary, cardiometabolic, and mental health and chronic pain multimorbidity. Once caught in the Multimorbidity Web, the consequences can be devastating, with reduced quality of life, physical function, survival, and treatment responses observed in multimorbid RA persons. The development of effective management and preventive approaches for multimorbidity in people with RA is in its infancy. Determining how best to assess, intervene, and prevent multimorbidity in RA is crucial to optimize long-term outcomes in people with RA.

Disease-Modifying Antirheumatic Drug Use and Its Effect on Long-term Opioid Use in Patients With Rheumatoid Arthritis

BACKGROUND/OBJECTIVES

The prevalence of chronic pain is high in patients with rheumatoid arthritis (RA), increasing the risk for opioid use. The objective of this study was to assess disease-modifying antirheumatic drug (DMARD) use and its effect on long-term opioid use in patients with RA.

METHODS

This cohort study included Medicare beneficiaries with diagnosis of RA who received at least 30-day consecutive prescription of opioids in 2017 (n = 23,608). The patients were grouped into non-DMARD and DMARD users, who were further subdivided into regimens set forth by the American College of Rheumatology. The outcome measured was long-term opioid use in 2018 defined as at least 90-day consecutive prescription of opioids. Dose and duration of opioid use were also assessed. A multivariable model identifying factors associated with non-DMARD use was also performed.

RESULTS

Compared with non-DMARD users, the odds of long-term opioid use were significantly lower among DMARD users (odds ratio, 0.89; 95% confidence interval, 0.83-0.95). All regimens except non-tumor necrosis factor biologic + methotrexate were associated with lower odds of long-term opioid use relative to non-DMARD users. The mean total morphine milligram equivalent, morphine milligram equivalent per day, and total days of opioid use were lower among DMARD users compared with non-DMARD users. Older age, male sex, Black race, psychiatric and medical comorbidities, and not being seen by a rheumatologist were significantly associated with non-DMARD use.

CONCLUSION

Disease-modifying antirheumatic drug use was associated with lower odds of long-term opioid use among RA patients with baseline opioid prescription. Factors associated with non-DMARD use represent a window of opportunity for intervention to improve pain-related quality of life in patients living with RA.

Multimorbidity status and annual healthcare expenditures of rheumatoid arthritis patients: a Dutch hospital-centered versus population-based comparison

The prevalence of multimorbidity among rheumatoid arthritis (RA) patients is increasing and associated with worse outcomes. Therefore, management of multimorbid patients requires a multidisciplinary approach. However, healthcare systems consist of mono-disciplinary subsystems, which limits collaboration across subsystems. To study the importance of a multidisciplinary, integrated approach, associations between expenditures and multimorbidity are assessed in real-life data. Retrospective data on RA patients from a Dutch single-hospital are analyzed and compared to the Dutch RA population data. The Elixhauser index is used to measure the multimorbidity prevalence. Regression analyses were conducted to derive the relationship between multimorbidity, healthcare costs and self-reported quality of life (e.g. EQ-5D). When analyzing the impact of multimorbidity within RA patients in context of a single-hospital context, multimorbidity is only partially captured: 13% prevalence versus 24% of the Dutch population. Multimorbidity is associated with higher care expenditures. Depending on the type of multimorbidity, expenditures are €43-€5821 higher in a single-hospital and from €2259-€9648 in population data. Finally, medication use associated with chronic diseases and self-reported aspects of well-being are associated with similar increases in healthcare expenditures as multimorbidity based on hospital care. Within RA, a single-hospital approach underestimates the association between multimorbidity and healthcare expenditures as 43% of healthcare utilization and expenditures are missed. To overcome a single-provider perspective in healthcare and efficiently coordinate multimorbid patients, besides providing holistic care, professionals also need to use data providing comprehensive pictures of patients.