First Validation of the Full PROMIS Pain Interference and Pain Behavior Item Banks in Patients With Rheumatoid Arthritis

Utilizing a learning health system to capture real-world patient data: Application of the reliable change index to evaluate and improve the outcome of a pain rehabilitation program

BACKGROUND AND OBJECTIVES

The learning healthcare system (LHS) has been developed to integrate patients' clinical data into clinical decisions and improve treatment outcomes. Having little guidance on this integration process, we aim to explain (a) an applicable analytic tool for clinicians to evaluate the clinical outcomes at a group and an individual level and (b) our quality improvement (QI) project, analyzing the outcomes of a new outpatient pain rehabilitation program ("Back-in-Action": BIA) and applying the analysis results to modify our clinical practice.

METHODS

Through our LHS (CHOIR; https://choir.stanford.edu), we administered the Pain Catastrophizing Scale (PCS), Chronic Pain Acceptance Questionnaire (CPAQ), and Patient-Reported Outcomes Measures (PROMIS)® before and after BIA. After searching for appropriate analytic tools, we decided to use the Reliable Change Index (RCI) to determine if an observed change in the direction of better (improvement) or worse (deterioration) would be beyond or within the measurement error (no change).

RESULTS

Our RCI calculations revealed that at least a 9-point decrease in the PCS scores and 10-point increase in the CPAQ scores would indicate reliable improvement. RCIs for the PROMIS measures ranged from 5 to 8 T-score points (i.e., 0.5-0.8 SD). When evaluating change scores of the PCS, CPAQ, and PROMIS measures, we found that 94% of patients showed improvement in at least one domain after BIA and 6% showed no reliable improvement.

CONCLUSIONS

Our QI project revealed RCI as a useful tool to evaluate treatment outcomes at a group and an individual level, and RCI could be incorporated into the LHS to generate a progress report automatically for clinicians. We further explained how clinicians could use RCI results to modify a clinical practice, to improve the outcomes of a pain program, and to develop individualized care plans. Lastly, we suggested future research areas to improve the LHS application in pain practice.

Protocol for the development of a tool to map systemic sclerosis pain sources, patterns, and management experiences: a Scleroderma Patient-centered Intervention Network patient-researcher partnership

INTRODUCTION

Systemic sclerosis (SSc) is a rare, complex autoimmune rheumatic disease with multiple factors that contribute to pain. People with SSc emphasize the effect pain has on their quality of life, but no studies have systematically examined the frequency and relative importance of different SSc pain sources, patterns of pain from different sources, and pain management experiences. Our objectives are to (1) develop a tool, jointly with researchers, health care providers, and patients, to map sources of pain in SSc, determine patterns of pain from different sources, and understand pain management experiences; and (2) administer the final tool version to participants in the large multinational Scleroderma Patient-centered Intervention Network (SPIN) Cohort.

METHODS

First, we will use validated pain assessment tools as templates to develop an initial version of our pain assessment tool, and we will obtain input from patient advisors to adapt it for SSc. The tool will include questions on pain sources, pain patterns, pain intensity, pain management techniques, and barriers to pain management in SSc. Second, we will conduct nominal group technique sessions with people living with SSc and health care providers who care for people with SSc to further refine the tool. Third, we will conduct individual usability testing sessions with SPIN Cohort participants. Once the tool has been finalized, we will administer it to individuals in the multinational SPIN Cohort, which currently includes over 1,300 active participants from 54 sites in 7 countries. We will perform unsupervised clustering using the KAy-Means for MIxed LArge data (KAMILA) method to identify participant subgroups with similar profiles of pain sources (present or absent) and to evaluate predictors of subgroup membership. We will use latent profile analysis to identify subgroups of participants with similar profiles based on pain intensity scores for each pain source and evaluate predictors.

DISCUSSION

Once completed, our pain assessment tool will allow our team and other researchers to map sources of pain in SSc and to understand pain management experiences of people living with SSc. This knowledge will provide avenues for studies on the pathophysiology of pain in SSc and studies of interventions to improve pain management.

Capturing Patient Value in an Economic Evaluation

OBJECTIVE

Economic evaluations predominantly use generic outcomes, such as the Euro Quality of Life-5 Dimension (EQ-5D), to assess health status. However, because of the generic nature, they are less suitable to capture the quality of life of patients with specific conditions. Given the transition to patient-centered (remote) care delivery, this study aims to evaluate the possibility of using disease-specific measures in a cost-effectiveness analysis.

METHODS

A real-life cohort from Maasstad Hospital (2020-2021) in the Netherlands, with 772 patients with rheumatoid arthritis (RA), was used to assess the cost-effectiveness of electronic consultations (e-consultations) compared with face-to-face consultations. The Incremental Cost-Effectiveness Ratio (ICER), based on the generic EQ-5D, was compared with ICER's based on RA-specific measures: the Rheumatoid Arthritis Impact of Disease (RAID) and Health Assessment Questionnaire-Disability Index (HAQ-DI). To compare the cost-effectiveness of these different measures, HAQ-DI and RAID were expressed in quality-adjusted life-years (QALYs) via estimated conversion equations.

RESULTS

Disease-specific patient-reported outcome measures (PROMs) offer a promising alternative for traditional measures in economic evaluations, capturing patient-relevant domains more comprehensively. Because PROMs are increasingly applied in clinical practice, the next step entails modeling of an RA patient-wide conversion equation to implement PROMs in economic evaluations.

CONCLUSION

The conventional ICER (eg, EQ-5D) indicates that e-consultations are cost-effective with cost savings of -€161,000 per QALY gained for a prevalent RA cohort treated in a secondary trainee hospital. RA-specific measures show similar results, with ICERs of -€163,000 per HAQ-DI (QALY) and -€223,000 per RAID (QALY) gained. RA-specific measures capture patient-relevant domains and offer the opportunity to improve the assessment and treatment of the disease impact.

Country-specific reference values for PROMIS® pain, physical function and participation measures compared to US reference values

INTRODUCTION

Patient-Reported Outcomes Measurement Information System (PROMIS^®) is commonly used across medical conditions. To facilitate interpretation of scores across countries, we calculated Dutch reference values for PROMIS Physical Function (PROMIS-PF), Pain Interference (PROMIS-PI), Pain Behavior (PROMIS-PB), Ability to Participate in Social Roles and Activities (PROMIS-APSRA), and Satisfaction with Social Roles and Activities (PROMIS-SSRA), as compared to US reference values.

PATIENTS AND METHODS

A panel completed full PROMIS-PF (n=1310), PROMIS-PI and PROMIS-PB (n=1052), and PROMIS-APSRA and PROMIS-SSRA (n=1002) item banks and reported their level of health per domain (no, mild, moderate, severe limitations). T-scores were calculated by sample and subgroups (age, gender, self-reported level of domain). Distribution-based and anchor-based thresholds for mild, moderate, and severe scores were determined.

RESULTS

Mean T-scores were close to the US mean of 50 for PROMIS-PF (49.8) and PROMIS-APSRA (50.6), lower for PROMIS-SSRA (47.5) and higher for PROMIS-PI (54.9) and PROMIS-PB (52.0). Distribution-based thresholds for mild, moderate, and severe scores were comparable to US recommended cut-off values (except for PROMIS-PI) but participants reported limitations 'earlier' than suggested thresholds.

CONCLUSION

Dutch reference values were close to US reference values for some PROMIS domains but not all. We recommend country-specific reference values to facilitate worldwide PROMIS use.KEY MESSAGESPROMIS offers universally applicable IRT-based efficient and patient-friendly measures to assess commonly relevant patient-reported outcomes across medical conditions.To support the use of PROMIS in daily clinical practice and research across the world, country-specific general population reference values should be obtained.More research is necessary to obtain reliable and valid cut-off values for what constitutes mild, moderate and severe scores from the patients' perspective.

Measuring PROMIS pain interference in German patients with chronic conditions: calibration, validation, and cross-cultural use of item parameters

PURPOSE

To calibrate the item parameters of the German PROMIS® Pain interference (PROMIS PI) items using an item-response theory (IRT) model and investigate psychometric properties of the item bank.

METHODS

Forty items of the PROMIS PI item bank were collected in a convenience sample of 660 patients, which were recruited during inpatient rheumatological treatment or outpatient psychosomatic medicine visits in Germany. Unidimensionality, monotonicity, and local independence were tested as required for IRT analyses. Unidimensionality was examined using confirmatory factor analyses (CFA) and exploratory factor analysis (EFA). Unidimensional and bifactor graded-response IRT models were fitted to the data. Bifactor indices were used to investigate whether multidimensionality would lead to biased scores. To evaluate convergent and discriminant validity, the item bank was correlated with legacy pain instruments. Potential differential item functioning (DIF) was examined for gender, age, and subsample. To investigate whether U.S. item parameters may be used to derive T-scores in German patients, T-scores based on previously published U.S. and newly estimated German item parameters were compared with each other after adjusting for sample specific differences.

RESULTS

All items were sufficiently unidimensional, locally independent, and monotonic. Whereas the fit of the unidimensional IRT model was not acceptable, a bifactor IRT model demonstrated acceptable fit. Explained common variance and Omega hierarchical suggested that using the unidimensional model would not lead to biased scores. One item demonstrated DIF between subsamples. High correlations with legacy pain instruments supported construct validity of the item bank. T-scores based on U.S. and German item parameters were similar suggesting that U.S. parameters could be used in German samples.

CONCLUSION

The German PROMIS PI item bank proved to be a clinically valid and precise instrument for assessing pain interference in patients with chronic conditions.

The use of PROMIS measures in clinical studies in patients with inflammatory arthritis: a systematic review

PURPOSE

Although the use of Patient-Reported Outcomes Measurement Information System (PROMIS) measures is widely advocated, little is known on their use in patients with inflammatory arthritis. We systematically describe the use and outcomes of PROMIS measures in clinical studies involving people with rheumatoid arthritis (RA) or axial spondyloarthritis (axSpA).

METHODS

A systematic review was conducted according to the PRISMA guidelines. Through a systematic search of nine electronic databases, clinical studies including patients with RA or axSpA and reporting the use of PROMIS measure were selected. Study characteristics, details of PROMIS measures and their outcomes, if available, were extracted.

RESULTS

In total, 29 studies described in 40 articles met the inclusion criteria, of which 25 studies included RA patients, three studies included axSpA patients and one study included both RA and axSpA patients. The use of two general PROMIS measures (PROMIS Global Health, PROMIS-29) and 13 different domain-specific PROMIS measures was reported, of which the PROMIS Pain Interference (n = 17), Physical Function (n = 14), Fatigue (n = 13), and Depression (n = 12) measures were most frequently used. Twenty-one studies reported their results in terms of T-scores. Most T-scores were worse than the general population mean, indicating impairments of health status. Eight studies did not report actual data but rather measurement properties of the PROMIS measures.

CONCLUSION

There was considerable variety regarding the different PROMIS measures used, with the PROMIS Pain interference, Physical function, Fatigue, and Depression measures being the most frequently used. In order to facilitate the comparisons across studies, more standardization of the selection of PROMIS measures is needed.

Responsiveness and Minimal Important Change of the PROMIS Pain Interference Item Bank in Patients Presented in Musculoskeletal Practice

We evaluated the responsiveness of the Patient Reported Outcome Information System Pain Interference item bank in patients with musculoskeletal pain by testing predefined hypotheses about the relationship between the change scores on the item bank, change scores on legacy instruments and Global Ratings of Change (GRoC), and we estimated Minimal Important Change (MIC). Patients answered the full Dutch-Flemish V1.1 item bank. From the responses we derived scores for the standard 8-item short form (SF8a) and a CAT-score was simulated. Correlations between the change scores on the item bank, GRoC and legacy instruments were calculated, together with Effect Sizes, Standardized Response Means, and Area Under the Curve. GRoC were used as an anchor for estimating the MIC with (adjusted) predictive modeling. Of 1,677 patients answering baseline questionnaires 960 completed follow-up questionnaires at 3 months. The item bank correlated moderately high with the GRoC (Spearman's rho 0.63) and with the legacy instruments (Pearson's R ranging from .45 to .68). It showed a high ES (.97) and Standardized Response Means (.71), and could distinguish well between improved and not improved patients based on the GRoC (Area Under the Curve .77). Comparable results were found for the derived SF8a and CAT-scores. The MIC was estimated to be 3.2 (CI 2.6-3.7) T-score points. PERSPECTIVE: Our study supports the responsiveness of the PROMIS-PI item bank in patients with musculoskeletal complaints. Almost all predefined hypotheses were met (94%). The PROMIS-PI item bank correlated well with several legacy instruments which supports generic use of the item bank. MIC for PROMIS-PI was estimated to be 3.2 T-score points.

Response to Treatment with Intravenous Golimumab or Infliximab in Rheumatoid Arthritis Patients: PROMIS Results from the Real-World Observational Phase 4 AWARE Study

INTRODUCTION

To assess changes in the Patient-Reported Outcomes Measurement Information System (PROMIS^®) outcomes related to social, mental, and physical well-being after approximately 1 year of intravenous (IV) golimumab or infliximab treatment in patients with rheumatoid arthritis (RA) using real-world evidence from AWARE.

METHODS

AWARE was a prospective, noninterventional, multicenter, observational, U.S.-based phase 4 study of 1270 RA patients who initiated treatment with IV golimumab or infliximab. PROMIS-29 and PROMIS short form (SF) Fatigue 7a and Pain Interference 6b questionnaires were administered at baseline and infusions 2, 5, and 8 (approximately weeks  4, 28, and 52 for IV golimumab and weeks  2, 22, and 46 for infliximab). Mean changes from baseline in all PROMIS-29 domains and respective SFs and response rates for achieving ≥ 3, ≥ 5, or ≥ 10-point improvements were determined.

RESULTS

Among all patients, baseline mean ± SD PROMIS T-scores were consistent between treatment groups and indicated worse physical function (38.2 ± 6.8 IV golimumab, 38.0 ± 6.9 infliximab), more pain interference (63.0 ± 7.6 IV golimumab, 63.9 ± 7.8 infliximab), and greater fatigue (58.4 ± 9.9 IV golimumab, 59.4 ± 10.0 infliximab) in these patients vs the general U.S. population (T-score = 50). Through the 8th infusion of either treatment, IV golimumab- and infliximab-treated patients achieved meaningful improvements (≥ 3-point improvement in T-scores) in all PROMIS-29 domains and respective SFs, and the proportions of patients with ≥ 3, ≥ 5, or ≥ 10-point improvements in T-scores increased from infusion 2 through infusion 8.

CONCLUSIONS

RA patients treated with IV golimumab or infliximab achieved comparable improvements across social, mental, and physical well-being PROMIS measures. Additionally, PROMIS detected meaningful clinical changes in patient-reported outcomes in both treatment groups.

CLINICALTRIALS

GOV REGISTRATION NUMBER

NCT02728934.

Machine Learning Applied to Patient-Reported Outcomes to Classify Physician-Derived Measures of Rheumatoid Arthritis Disease Activity

OBJECTIVE

Patient-reported outcome (PRO) data have assumed increasing importance in the care of patients with rheumatoid arthritis (RA), yet physician-derived disease activity measures, such as Clinical Disease Activity Index (CDAI), remain the most accepted metrics to assess disease activity. The possibility that newer longitudinal PRO data might be used as a proxy for the CDAI has not been evaluated.

METHODS

Using data from a large pragmatic trial, we evaluated patients with RA initiating golimumab intravenous or infliximab. The classification target was low disease activity (LDA) (CDAI ≤10) at the first visit between months 3 and 12. Data were randomly partitioned into training (80%) and test (20%) data sets. Multiple machine learning (ML) methods (eg, random forests, gradient boosting, support vector machines) were used to classify CDAI disease activity category, conduct feature selection, and assess feature importance. Model performance evaluated cross-validated error, comparing different ML approaches using both training and test data.

RESULTS

A total of 494 patients were analyzed, and 36.4% achieved LDA. The most important classification features included several Patient-Reported Outcomes Measurement Information System measures (social participation, pain interference, pain intensity, and physical function), patient global, and baseline CDAI. Among all ML methods, random forests performed best. Overall model accuracy and positive predictive values for all ML methods were approximately 80%.

CONCLUSION

ML methods coupled with longitudinal PRO data appear useful and can achieve reasonable accuracy in classifying LDA among patients starting a new biologic. This approach has promise for real-world evidence generation in the common circumstance when physician-derived disease activity data are not available yet PRO measures are.

A workplace wellness program at an academic health center influences employee health, satisfaction, productivity and the rate of workplace injury

PURPOSE

Workplace wellness (WP) programs are an employer strategy to improve employee health and satisfaction. Their impact on productivity and benefit expense remains unclear. This study examines the association of a voluntary WP at an academic health center with both employee metrics and employer costs.

METHODS

We retrospectively reviewed prospectively collected data from January 2016 to April 2018 for employees who voluntarily underwent screening for a WP at an academic medical center. We used their demographic, social, work, and clinical data to address the central research question. The primary outcomes included wellness measures from the Patient-Reported Outcomes Measurement Information System (PROMIS), secondary wellness outcomes such as body mass index, job-related outcomes such as job satisfaction, and workers' compensation metrics such as the claim amount. The key independent variables were whether an observation was from before or after the WP. For workers' compensation metrics, additional key independent variables were intervention/control group, and an interaction between the before/after and intervention/control variables. We conducted univariate and bivariate/unadjusted analyses, and estimated multivariable linear, logistic, and gamma regression models that also controlled for confounders.

RESULTS

The study included 370 employees. Participation in the program was associated with significant improvements in the PROMIS pain interference, fatigue and sleep quality domains. Hip circumference diminished, and functional movement outcomes were improved. Job satisfaction improved by 4.4 percentage points (95% Confidence Interval [CI]: 0.3-8.5, p = 0.04) and self-reported productivity by 14.5 percentage points (95% Confidence Interval [CI]: 9.5-19.5, p < 0.001). The likelihood of a new compensation claim during the 12-month follow-up period fell by 10.1% (95% Confidence Interval [CI]: - 15.5 to - 4.7, p < 0.001). However, the value of a new claim was unchanged.

CONCLUSION

Employees who completed a WP at an academic medical center demonstrated improvements in several recognized patient-reported outcome measures, in job satisfaction and self-reported productivity, a decrease in hip circumference, an improvement in functional motion and a decreased rate of compensable injury.

Pain levels and associated factors in the Scleroderma Patient-centered Intervention Network (SPIN) cohort: a multicentre cross-sectional study

BACKGROUND

Pain is an important and detrimental feature of systemic sclerosis but is often overlooked or deprioritised in research and clinical care. Raynaud's phenomenon, arthritis, and cutaneous ulcers are among the commonly reported disease manifestations of systemic sclerosis that could be associated with pain. We aimed to assess levels of pain intensity and interference and to evaluate disease factors associated with pain intensity and interference.

METHODS

In this multicentre cross-sectional study, participants from the Scleroderma Patient-centered Intervention Network cohort who completed pain intensity and interference measures (Patient Reported Outcomes Information System-29 profile, version 2·0) as part of baseline assessments were included. Patients were recruited from 46 centres in Australia, Canada, France, Mexico, Spain, the UK, and the USA between April 15, 2014, and Jan 7, 2020. Eligible patients included those aged 18 years or older who met the criteria for systemic sclerosis devised by the American College of Rheumatology and the European League Against Rheumatism. Associations of pain intensity and pain interference with systemic sclerosis-related variables and overlap syndromes, controlling for sociodemographic variables, were assessed with multiple linear regression. Continuous independent variables were standardised.

FINDINGS

Among 2157 participants with systemic sclerosis (268 [12%] males and 1889 [88%] females), 1870 (87%) reported mild, moderate, or severe pain (defined as ≥1 on a 0 to 10 scale), and 815 (38%) reported moderate or severe pain (defined as ≥5). Moreover, 757 (35%) participants reported moderate or severe pain interference. Greater pain intensity was independently associated with female sex (0·58 points [95% CI 0·26-0·90]), non-White race or ethnicity (0·50 points [0·21-0·79]), fewer years in formal education (0·30 points per SD [0·19-0·41]), country (reference: USA; Canada: 0·29 points [0·01-0·57] and UK: 0·58 points [0·21-0·95]), greater body-mass index (0·35 points per SD [0·24-0·45]); joint contractures (0·67 points [0·39-0·94]), digital ulcers (0·33 points [0·10-0·55]), gastrointestinal involvement (0·66 points [0·33-0·98]), skin involvement (measured using modified Rodnan skin score; 0·22 points per SD [0·10-0·35]), rheumatoid arthritis (0·96 points [0·50-1·43]), and Sjögren's syndrome (0·42 points [0·01-0·83]). Pain interference results were similar.

INTERPRETATION

Pain is common among people with systemic sclerosis. Controlling for sociodemographic variables, greater pain was associated with multiple systemic sclerosis-related manifestations, including joint contractures, digital ulcers, gastrointestinal involvement, skin involvement, and the presence of overlap syndromes. Health-care providers should work with patients to address pain, including identifying and addressing systemic sclerosis manifestations associated with their pain, and supporting behavioural approaches to minimise impact on function and quality of life.

FUNDING

Canadian Institutes of Health Research, Arthritis Society, The Lady Davis Institute for Medical Research of the Jewish General Hospital, Jewish General Hospital Foundation, McGill University, Scleroderma Society of Ontario, Scleroderma Canada, Sclérodermie Québec, Scleroderma Manitoba, Scleroderma Atlantic, Scleroderma Association of BC, Scleroderma SASK, Scleroderma Australia, Scleroderma New South Wales, Scleroderma Victoria, and Scleroderma Queensland.

Why It Hurts: The Mechanisms of Pain in Rheumatoid Arthritis

Pain is a near-universal feature of rheumatoid arthritis, but peripheral joint inflammation may not suffice to explain the etiology of pain in all patients with rheumatoid arthritis. Inflammation in rheumatoid arthritis releases several algogens that may generate pain. Also, central nervous system processes may play a crucial role in the regulation and perpetuation of pain. Several methods for assessing pain in rheumatoid arthritis exist, and recently the role of assessing therapeutics in treating specific etiologies of pain has gained interest.

Responsiveness to change over time and test-retest reliability of the PROMIS and Neuro-QoL mental health measures in persons with Huntington disease (HD)

BACKGROUND

The majority of persons with Huntington disease (HD) experience mental health symptoms. Patient-reported outcome (PRO) measures are capable of capturing unobservable behaviors and feelings relating to mental health. The current study aimed to test the reliability and responsiveness to self-reported and clinician-rated change over time of Neuro-QoL and PROMIS mental health PROs over the course of a 24-month period.

METHODS

At baseline, 12-months, and 24-months, 362 participants with premanifest or manifest HD completed the Neuro-QoL Depression computer adaptive test (CAT), PROMIS Depression short form (SF), Neuro-QoL Anxiety CAT, PROMIS Anxiety SF, PROMIS Anger CAT and SF, Neuro-QoL Emotional/Behavioral Dyscontrol CAT and SF, Neuro-QoL Positive Affect and Well-Being CAT and SF, and Neuro-QoL Stigma CAT and SF. Participants completed several clinician-administered measures at each time point, as well as several global ratings of change at 12- and 24-months. Reliability (test-retest reliability and measurement error) and responsiveness (using standardized response means and general linear models) were assessed.

RESULTS

Test-retest reliability and measurement error were excellent for all PROs (all ICC ≥ .90 for test-retest reliability and all SEM percentages ≤ 6.82%). In addition, 12- and 24-month responsiveness were generally supported for the Neuro-QoL and PROMIS mental health PROs; findings relative to clinician-rated anchors of change (e.g., SRMs for the group with declines ranged from .38 to .91 for 24-month change and .09 to .45, with the majority above .25 for 12-month change) were generally more robust than those relative to self-reported anchors of change (e.g., SRMs for the group with declines ranged from .02 to .75, with the majority above .39 for 24-month change and .09 to .45, with the majority above .16 for 12-month change).

CONCLUSIONS

The Neuro-QoL and PROMIS mental health PROs demonstrated strong psychometric reliability, as well as responsiveness to self-reported and clinician-rated change over time in people with HD.