Estimating Optimal Weights for Compound Scores: A Multidimensional IRT Approach.

A method is proposed for constructing indices as linear functions of variables such that the reliability of the compound score is maximized. Reliability is defined in the framework of latent variable modeling [i.e., item response theory (IRT)] and optimal weights of the components of the index are found by maximizing the posterior variance relative to the total latent variable variance. Three methods for estimating the weights are proposed. The first is a likelihood-based approach, that is, marginal maximum likelihood (MML). The other two are Bayesian approaches based on Markov chain Monte Carlo (MCMC) computational methods. One is based on an augmented Gibbs sampler specifically targeted at IRT, and the other is based on a general purpose Gibbs sampler such as implemented in OpenBugs and Jags. Simulation studies are presented to demonstrate the procedure and to compare the three methods. Results are very similar, so practitioners may be suggested the use of the easily accessible latter method. A real-data set pertaining to the 28-joint Disease Activity Score is used to show how the methods can be applied in a complex measurement situation with multiple time points and mixed data formats.

Predictors of satisfactory improvements in pain for patients with early rheumatoid arthritis in a treat-to-target study.

OBJECTIVE: The aim of this study was to identify baseline predictors of achieving patient-perceived satisfactory improvement (PPSI) in pain after 6 months of treat to target in patients with early RA. METHODS: Baseline and 6 month data were used from patients included in the Dutch Rheumatoid Arthritis Monitoring remission induction cohort study. Simple and multivariable logistic regression analyses were used to identify significant predictors of achieving an absolute improvement of 30 mm or a relative improvement of 50% on a visual analogue scale for pain. RESULTS: At 6 months, 125 of 209 patients (59.8%) achieved an absolute PPSI and 130 patients (62.2%) achieved a relative PPSI in pain. Controlling for baseline pain, having symmetrical arthritis was the strongest independent predictor of achieving an absolute [odds ratio (OR) 3.17, P = 0.03] or relative (OR 3.44, P = 0.01) PPSI. Additionally, anti-CCP positivity (OR 2.04, P = 0.04) and having ≤12 tender joints (OR 0.29, P = 0.01) were predictive of achieving a relative PPSI. The total explained variance of baseline predictors was 30% for absolute and 18% for relative improvements, respectively. CONCLUSION: Symmetrical joint involvement, anti-CCP positivity and fewer tender joints at baseline are prognostic signs for achieving satisfactory improvement in pain after 6 months of treat to target in patients with early RA.

A short tutorial on item response theory in rheumatology.

OBJECTIVES: The aim is to familiarize physicians and researchers with the most important concepts of item response theory (IRT) and with its usefulness for improving test administration and data collection in health care. Special attention is given to the versatility of its use within the rheumatic field. METHODS: This short tutorial describes the most important basic principles of item response theory, including the underlying assumptions, the model parameters, and the different models that can be applied. Practical applications are discussed to demonstrate the potential utility of IRT within clinical practice. RESULTS: IRT has proven to be useful for the development and evaluation of both clinical measures as well as patient reported outcomes used for measuring health status in observational studies and clinical trials. Promising features of IRT for the future of test administration are the assessment of local reliability and differential item functioning, the cross-cultural validation or equation of instruments, the development of large item banks, and the administration of computerised adaptive tests. These modern techniques have the ability to maximise measurement precision while simultaneously minimise response burden. CONCLUSIONS: IRT provides a theoretical basis for developing alternatives to the existing tools for assessing health outcome measures in rheumatology.

How age and sex affect the erythrocyte sedimentation rate and C-reactive protein in early rheumatoid arthritis.

BACKGROUND: The erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) are two commonly used measures of inflammation in rheumatoid arthritis (RA). As current RA treatment guidelines strongly emphasize early and aggressive treatment aiming at fast remission, optimal measurement of inflammation becomes increasingly important. Dependencies with age, sex, and body mass index have been shown for both inflammatory markers, yet it remains unclear which inflammatory marker is affected least by these effects in patients with early RA. METHODS: Baseline data from 589 patients from the DREAM registry were used for analyses. Associations between the inflammatory markers and age, sex, and BMI were evaluated first using univariate linear regression analyses. Next, it was tested whether these associations were independent of a patient's current disease activity as well as of each other using multiple linear regression analyses with backward elimination. The strengths of the associations were compared using standardized beta (ß) coefficients. The multivariate analyses were repeated after 1 year. RESULTS: At baseline, both the ESR and CRP were univariately associated with age, sex, and BMI, although the association with BMI disappeared in multivariate analyses. ESR and CRP levels significantly increased with age (ß-ESR=0.017, p<0.001 and ß-CRP=0.009, p=0.006), independent of the number of tender and swollen joints, general health, and sex. For each decade of aging, ESR and CRP levels became 1.19 and 1.09 times higher, respectively. Furthermore, women demonstrated average ESR levels that were 1.22 times higher than that of men (ß=0.198, p=0.007), whereas men had 1.20 times higher CRP levels (ß=-0.182, p=0.048). Effects were strongest on the ESR. BMI became significantly associated with both inflammatory markers after 1 year, showing higher levels with increasing weight. Age continued to be significantly associated, whereas sex remained only associated with the ESR level. CONCLUSIONS: Age and sex are independently associated with the levels of both acute phase reactants in early RA, emphasizing the need to take these external factors into account when interpreting disease activity measures. BMI appears to become more relevant at later stages of the disease.

Performance of the Dutch SF-36 version 2 as a measure of health-related quality of life in patients with rheumatoid arthritis.

BACKGROUND: The aim of this study was to examine the measurement properties of the Dutch SF-36 version 2 (SF-36v2) health survey in patients with rheumatoid arthritis (RA). METHODS: Scaling assumptions, internal reliability, and internal construct validity were examined using available data from 1884 RA patients included in the Dutch Rheumatoid Arthritis Monitoring (DREAM) registry. External construct validity and responsiveness to change were examined using baseline and 6-month follow-up data from a subset of 387 early RA patients participating in the DREAM remission induction cohort. RESULTS: The individual items of the SF-36v2 adequately met scaling assumptions, although four items correlated too highly with items from different scales. Internal consistency was high for all eight scales and the physical and mental health components underlying the scales were replicated, supporting the use of the standard scoring algorithms. The SF-36v2 scales demonstrated minimal floor effects and ceiling effects were noteworthy only for the role-physical, social functioning, and role-emotional scales. Correlations with other core measures were as expected and the SF-36v2 showed excellent known-groups validity in distinguishing between patients with low or moderate-high disease activity. All scales related to physical health showed moderate to large responsiveness to change in patients that achieved low disease activity at six months. CONCLUSION: The SF-36v2 appears to be a psychometrically sound tool for the assessment of health-related quality of life of Dutch patients with RA.

Translation and validation of the Dutch version of the Effective Consumer Scale (EC-17).

PURPOSE: The Effective Consumer Scale (EC-17) measures the skills of musculoskeletal patients in managing their own healthcare. The objectives of this study were to translate the EC-17 into Dutch and to further evaluate its psychometric properties. METHODS: The EC-17 was translated and cognitively pretested following cross-cultural adaptation guidelines. Two hundred and thirty-eight outpatients (52 % response rate) with osteoarthritis or fibromyalgia completed the EC-17 along with other validated measures. Three weeks later, 101 patients completed the EC-17 again. RESULTS: Confirmatory factor analysis supported the unidimensional structure of the scale. The items adequately fit the Rasch model and only one item demonstrated differential item functioning. Person reliability was high (0.92), but item difficulty levels tended to cluster around the middle of the scale, and measurement precision was highest for moderate and lower levels of skills. The scale demonstrated adequate test-retest reliability (ICC = 0.71), and correlations with other measures were largely as expected. CONCLUSION: The results supported the validity and reliability of the Dutch version of the EC-17, but suggest that the scale is best targeted at patients with relatively low levels of skills. Future studies should further examine its sensitivity to change in a clinical trial specifically aimed at improving effective consumer skills.

Validity of summing painful joint sites to assess joint-pain comorbidity in hip or knee osteoarthritis.

BACKGROUND: Previous studies in patients with hip and knee osteoarthritis (OA) have advocated the relevance of assessing the number of painful joint sites, other than the primary affected joint, in both research and clinical practice. However, it is unclear whether joint-pain comorbidities can simply be summed up. METHODS: A total of 401 patients with hip or knee OA completed questionnaires on demographic variables and joint-pain comorbidities. Rasch analysis was performed to evaluate whether a sum score of joint-pain comorbidities can be calculated. RESULTS: Self-reported joint-pain comorbidities showed a good fit to the Rasch model and were not biased by gender, age, disease duration, BMI, or patient group. As a group, joint-pain comorbidities covered a reasonable range of severity levels, although the sum score had rather low reliability levels suggesting it cannot discriminate well among patients. CONCLUSIONS: Joint-pain comorbidities, in other than the primary affected joints, can be summed into a joint pain comorbidity score. Nevertheless, its use is discouraged for individual decision making purposes since its lacks discriminative power in patients with minimal or extreme joint pain.

Modern psychometrics applied in rheumatology--a systematic review.

BACKGROUND: Although item response theory (IRT) appears to be increasingly used within health care research in general, a comprehensive overview of the frequency and characteristics of IRT analyses within the rheumatic field is lacking. An overview of the use and application of IRT in rheumatology to date may give insight into future research directions and highlight new possibilities for the improvement of outcome assessment in rheumatic conditions. Therefore, this study systematically reviewed the application of IRT to patient-reported and clinical outcome measures in rheumatology. METHODS: Literature searches in PubMed, Scopus and Web of Science resulted in 99 original English-language articles which used some form of IRT-based analysis of patient-reported or clinical outcome data in patients with a rheumatic condition. Both general study information and IRT-specific information were assessed. RESULTS: Most studies used Rasch modeling for developing or evaluating new or existing patient-reported outcomes in rheumatoid arthritis or osteoarthritis patients. Outcomes of principle interest were physical functioning and quality of life. Since the last decade, IRT has also been applied to clinical measures more frequently. IRT was mostly used for evaluating model fit, unidimensionality and differential item functioning, the distribution of items and persons along the underlying scale, and reliability. Less frequently used IRT applications were the evaluation of local independence, the threshold ordering of items, and the measurement precision along the scale. CONCLUSION: IRT applications have markedly increased within rheumatology over the past decades. To date, IRT has primarily been applied to patient-reported outcomes, however, applications to clinical measures are gaining interest. Useful IRT applications not yet widely used within rheumatology include the cross-calibration of instrument scores and the development of computerized adaptive tests which may reduce the measurement burden for both the patient and the clinician. Also, the measurement precision of outcome measures along the scale was only evaluated occasionally. Performed IRT analyses should be adequately explained, justified, and reported. A global consensus about uniform guidelines should be reached concerning the minimum number of assumptions which should be met and best ways of testing these assumptions, in order to stimulate the quality appraisal of performed IRT analyses.

Technology to Support Integrated Antimicrobial Stewardship Programs: A User Centered and Stakeholder Driven Development Approach.

The rise of antimicrobial resistance (AMR) is a severe global health problem. Tackling this problem requires the prudent prescribing of antimicrobials. This is promoted through Antimicrobial Stewardship Programs (ASPs). In this position paper we describe i) how a socio-technical multidisciplinary approach (based on the CeHRes Roadmap) can be applied in the development and implementation of Antimicrobial Stewardship technologies and ii) how this approach can be of value to support Antimicrobial Stewardship in practice. The CeHRes Roadmap entails five different phases to explore and test how an eHealth technology can be tailored to the target group and successfully implemented in practice: i) contextual inquiry, ii) value specification, iii) design, iv) operationalization, v) evaluation. In this position paper we describe the lessons learned from research and practice to guide future developments of technology based ASP interventions. Since AMR is a huge wicked problem on a global level, it requires innovative methods and models to empower general public and professionals to be proactive rather than reactive in a digitalized world. We highlight how to combat the dangerous rise of antimicrobial resistance in the future.

Exploring Fatigue Trajectories in Early Symptomatic Knee and Hip Osteoarthritis: 6-year Results from the CHECK Study.

OBJECTIVE: To examine whether different groups of fatigue trajectories can be identified among patients with early symptomatic osteoarthritis (OA) of the knee or hip, to describe the level of fatigue severity within each of these fatigue groups, and to investigate the involvement of age, sex, use of medication, comorbidity, and OA severity in relation to group membership. METHODS: Six years of followup data on fatigue (Medical Outcomes Study Short Form-36 Vitality scale) came from the Cohort Hip and Cohort Knee (CHECK) cohort. Growth mixture modeling was applied to identify distinct fatigue trajectories as well as to take into account the effects of the patient characteristics. RESULTS: Three fatigue trajectories were identified: low fatigue, low-to-high fatigue, and high fatigue. Latter trajectories showed considerable overlap from years 2 to 6, but differed in some patient characteristics in comparison with each other and in comparison with the low fatigue group. Comorbidity, medication use, and sex were significantly associated with the identified trajectories. Women, individuals with a comorbid disease, and those who used medication were more likely to follow a high fatigue trajectory. CONCLUSION: These findings suggest heterogeneous development of fatigue in the early OA population associated with varying patient characteristics. Further, this study shows that a considerable number of patients with OA already experience elevated levels of fatigue at an early stage of OA. While these findings need to be replicated, the identification of these trajectories with differing patient characteristics may warrant tailored psychosocial interventions for patients with elevated levels of fatigue.

Tackling wicked problems in infection prevention and control: a guideline for co-creation with stakeholders.

BACKGROUND: Infection prevention and control can be seen as a wicked public health problem as there is no consensus regarding problem definition and solution, multiple stakeholders with different needs and values are involved, and there is no clear end-point of the problem-solving process. Co-creation with stakeholders has been proposed as a suitable strategy to tackle wicked problems, yet little information and no clear step-by-step guide exist on how to do this. The objectives of this study were to develop a guideline to assist developers in tackling wicked problems using co-creation with stakeholders, and to apply this guideline to practice with an example case in the field of infection prevention and control. METHODS: A mixed-method approach consisting of the integration of both quantitative and qualitative research was used. Relevant stakeholders from the veterinary, human health, and public health sectors were identified using a literature scan, expert recommendations, and snowball sampling. The stakeholder salience approach was used to select key stakeholders based on 3 attributes: power, legitimacy, and urgency. Key values of stakeholders (N = 20) were derived by qualitative semi-structured interviews and quantitatively weighted and prioritized using an online survey. RESULTS: Our method showed that stakeholder identification and analysis are prerequisites for understanding the complex stakeholder network that characterizes wicked problems. A total of 73 stakeholders were identified of which 36 were selected as potential key stakeholders, and only one was seen as a definite stakeholder. In addition, deriving key stakeholder values is a necessity to gain insights into different problem definitions, solutions and needs stakeholders have regarding the wicked problem. Based on the methods used, we developed a step-by-step guideline for co-creation with stakeholders when tackling wicked problems. CONCLUSIONS: The mixed-methods guideline presented here provides a systematic, transparent method to identify, analyze, and co-create with stakeholders, and to recognize and prioritize their values, problem definitions, and solutions in the context of wicked problems. This guideline consists of a general framework and although it was applied in an eHealth context, may be relevant outside of eHealth as well.

Distinct trajectories of disease activity over the first year in early rheumatoid arthritis patients following a treat-to-target strategy.

Objective: Although treat-to-target (T2T) strategies are effective in early RA patients, important individual variations exist in the course towards remission. Growth mixture modeling (GMM) provides more insight into this heterogeneity by identifying subgroups of patients with similar response patterns. This study aimed to identify distinct trajectories of disease activity in early RA patients following a T2T strategy, during their first year. Methods: Data on various clinical and patient-reported measures were collected from the DREAM remission induction cohort. GMM was applied to examine the impact of T2T on subgroups characterized by different types of growth trajectories, as measured with the Disease Activity Score for 28 joints. Results: Three distinct trajectories of disease activity were found. The normative trajectory contained most patients (82.6%), showing a quickly decreasing disease activity, stabilizing at remission after 9 months. This group performed best on clinical and patient-reported measures over time and were more likely to be men. A smaller group (14.1%) also approached remission, but demonstrated a slower response to treatment. Finally, a minority (3.3%) showed no improvement after 1 year, despite an initial quick decrease in disease activity during the first months of treatment. Conclusion: Disease activity in early RA patients during the first year of a T2T strategy does not follow a linear pattern, nor is a single developmental trajectory applicable to all patients. Future studies should attempt to identify more specific risk factors for poor outcome to enable early identification of patients in need of alternative therapeutic approaches.

Further optimization of the reliability of the 28-joint disease activity score in patients with early rheumatoid arthritis.

BACKGROUND: The 28-joint Disease Activity Score (DAS28) combines scores on a 28-tender and swollen joint count (TJC28 and SJC28), a patient-reported measure for general health (GH), and an inflammatory marker (either the erythrocyte sedimentation rate [ESR] or the C-reactive protein [CRP]) into a composite measure of disease activity in rheumatoid arthritis (RA). This study examined the reliability of the DAS28 in patients with early RA using principles from generalizability theory and evaluated whether it could be increased by adjusting individual DAS28 component weights. METHODS: Patients were drawn from the DREAM registry and classified into a "fast response" group (N = 466) and "slow response" group (N = 80), depending on their pace of reaching remission. Composite reliabilities of the DAS28-ESR and DAS28-CRP were determined with the individual components' reliability, weights, variances, error variances, correlations and covariances. Weight optimization was performed by minimizing the error variance of the index. RESULTS: Composite reliabilities of 0.85 and 0.86 were found for the DAS28-ESR and DAS28-CRP, respectively, and were approximately equal across patients groups. Component reliabilities, however, varied widely both within and between sub-groups, ranging from 0.614 for GH ("slow response" group) to 0.912 for ESR ("fast response" group). Weight optimization increased composite reliability even further. In the total and "fast response" groups, this was achieved mostly by decreasing the weight of the TJC28 and GH. In the "slow response" group, though, the weights of the TJC28 and SJC28 were increased, while those of the inflammatory markers and GH were substantially decreased. CONCLUSIONS: The DAS28-ESR and the DAS28-CRP are reliable instruments for assessing disease activity in early RA and reliability can be increased even further by adjusting component weights. Given the low reliability and weightings of the general health component across subgroups it is recommended to explore alternative patient-reported outcome measures for inclusion in the DAS28.

Extending the floor and the ceiling for assessment of physical function.

OBJECTIVE: To improve the assessment of physical function by enhancing precision of physical function assessment as it pertains to subjects at extreme ends of the health continuum (i.e., subjects with extremely poor function ["floor"] or extremely good health ["ceiling"]). METHODS: Under the Patient-Reported Outcomes Measurement Information System (PROMIS) (a National Institutes of Health initiative), we developed new items to assess floor and ceiling physical function in order to supplement the existing item bank. Using item response theory and standard PROMIS methodology, we developed 31 floor items and 31 ceiling items and administered the items during a 12-month prospective, observational study of 737 subjects whose health status was at either extreme. Effect size was calculated and change over time was compared across anchor instruments and across items. Using the observed changes in scores, we back-calculated sample size requirements for the new and comparison measures. RESULTS: We studied 444 subjects who had been diagnosed as having a chronic illness and/or were of old age and 293 generally fit subjects (including athletes in training). Item response theory analyses confirmed that the new floor and ceiling items outperformed reference items (P < 0.001). The estimated post hoc sample size requirements were reduced by a factor of 2-4 for the floor population and a factor of 2 for the ceiling population. CONCLUSION: Extending the range of items by which physical function is measured can substantially improve measurement quality, reduce sample size requirements, and improve research efficiency. The paradigm shift from assessing disability to assessing physical function focuses assessment on the entire spectrum of physical function, signals improvement in the conceptual base of outcome assessment, and may be transformative as medical goals more closely approach societal goals for health.

Interchangeability of 28-joint disease activity scores using the erythrocyte sedimentation rate or the C-reactive protein as inflammatory marker.

This paper aims to examine the interchangeability of the disease activity score in 28 joints (DAS28)-erythrocyte sedimentation rate (ESR) and DAS28-CRP scores in a diverse sample of rheumatoid arthritis (RA) patients and to evaluate generalizability over gender, age, and disease duration. A sample of 682 patients was drawn from the DREAM registry. Agreement between the two DAS28 scores was analyzed using the intraclass correlation coefficient (ICC), Bland Altman plots, and a matrix of classification agreement over DAS28 disease activity categories. Despite a strong linear correlation between the DAS28 scores and a high ICC value of 0.931, a considerable lack of individual agreement could be observed, with Bland-Altman 95% limits of agreement ranging between -0.85 and +1.25 points. On average, DAS28-CRP scores were 0.20 points lower than DAS28-ESR scores, and data stratification on age and gender showed that this systematic bias was most severe in older women (0.39 points). The overall classification agreement across DAS28 categories was 76.69%, with the agreement being lowest (35.37%) in the low disease activity group. Patients were more easily classified as being in remission when using the DAS28-CRP measure. DAS28-ESR and DAS28-CRP scores are not interchangeable within individuals. The DAS28-CRP tends to yield lower values of disease activity than the DAS28-ESR, resulting in substantial classification differences.

Contribution of assessing forefoot joints in early rheumatoid arthritis patients: insights from item response theory.

OBJECTIVE: To evaluate the contribution of assessing forefoot joints to the measurement range and measurement precision of joint counts in early rheumatoid arthritis (RA) using item response theory. METHODS: Baseline measures of tender and swollen joint counts were analyzed in 459 early RA patients from the Dutch Rheumatoid Arthritis Monitoring remission induction cohort. The contribution of forefoot joints was studied by evaluating their effect on the measurement range and measurement precision of measures based on 28-joint counts. In addition, the alignment between the patient and joint distributions was investigated to determine whether the forefoot joints were informative for measuring joint tenderness or swelling of an early RA patient. RESULTS: In total, 233 patients (50.76%) experienced tenderness and 200 patients (43.57%) experienced swelling in ≥1 forefoot joint. Forefoot joints were more informative for measuring joint tenderness than joint swelling, but did not significantly improve the measurement range and measurement precision of the 28-joint counts. Furthermore, including forefoot joints did not remove the existing discrepancy between the joint and patient distributions in both joint counts. CONCLUSION: Forefoot joints were frequently affected on an individual level, but did not significantly improve the measurement range or precision of 28-joint counts in patients with early RA. From a measurement perspective, reduced joint counts are appropriate for use on a population level. The contribution of assessing forefoot joints on an individual level requires further investigation. Additionally, the results should be cross-validated in patients with longer disease durations to determine whether the pattern of joint involvement is similar in later stages of RA.

Validating the 28-tender joint count using item response theory.

OBJECTIVE: To examine the construct validity of the 28-tender joint count (TJC-28) using item response theory (IRT)-based methods. METHODS: A total of 457 patients with early stage rheumatoid arthritis (RA) were included. Internal construct validity of the TJC-28 was evaluated by determining whether the TJC-28 fit a 2-measure logistic IRT model. As well, we tested whether the discrimination and difficulty parameters of the joints properly reflected the known left-right symmetry of joint involvement. External validity was evaluated by correlations with other established measures of disease activity, including pain, disability, general health, erythrocyte sedimentation rate (ESR), and the 28-swollen joint count. RESULTS: The TJC-28 showed a good fit with the 2-parameter logistic model, with no relevant differential item functioning across sex, age, and time and with excellent reliability. The 28 joints covered a reasonable range of disease activity, even though they were mainly targeted at patients with moderate or high disease activity levels. The joint parameters reflected the left-right symmetry of joint involvement for all pairs of joints except one. All disease activity measures, except ESR, were significantly correlated with the TJC-28. Most correlations were of the expected magnitude. CONCLUSION: The TJC-28 showed good internal and acceptable external construct validity for patients with early-stage RA. The IRT analyses did point to some potential limitations of the instrument, a major problem being its limited measurement range. Future research should examine whether instrument modifications might lead to a more robust assessment of disease activity in patients with RA.