2016 American College of Rheumatology/European League Against Rheumatism Criteria for Minimal, Moderate, and Major Clinical Response in Juvenile Dermatomyositis: An International Myositis Assessment and Clinical Studies Group/Paediatric Rheumatology International Trials Organisation Collaborative Initiative.

To develop response criteria for juvenile dermatomyositis (DM). We analysed the performance of 312 definitions that used core set measures from either the International Myositis Assessment and Clinical Studies Group (IMACS) or the Paediatric Rheumatology International Trials Organisation (PRINTO) and were derived from natural history data and a conjoint analysis survey. They were further validated using data from the PRINTO trial of prednisone alone compared to prednisone with methotrexate or cyclosporine and the Rituximab in Myositis (RIM) trial. At a consensus conference, experts considered 14 top candidate criteria based on their performance characteristics and clinical face validity, using nominal group technique. Consensus was reached for a conjoint analysis-based continuous model with a total improvement score of 0-100, using absolute per cent change in core set measures of minimal (≥30), moderate (≥45), and major (≥70) improvement. The same criteria were chosen for adult DM/polymyositis, with differing thresholds for improvement. The sensitivity and specificity were 89% and 91-98% for minimal improvement, 92-94% and 94-99% for moderate improvement, and 91-98% and 85-86% for major improvement, respectively, in juvenile DM patient cohorts using the IMACS and PRINTO core set measures. These criteria were validated in the PRINTO trial for differentiating between treatment arms for minimal and moderate improvement (p=0.009-0.057) and in the RIM trial for significantly differentiating the physician's rating for improvement (p<0.006). The response criteria for juvenile DM consisted of a conjoint analysis-based model using a continuous improvement score based on absolute per cent change in core set measures, with thresholds for minimal, moderate, and major improvement.

2016 American College of Rheumatology/European League Against Rheumatism criteria for minimal, moderate, and major clinical response in adult dermatomyositis and polymyositis: An International Myositis Assessment and Clinical Studies Group/Paediatric Rheumatology International Trials Organisation Collaborative Initiative.

To develop response criteria for adult dermatomyositis (DM) and polymyositis (PM). Expert surveys, logistic regression, and conjoint analysis were used to develop 287 definitions using core set measures. Myositis experts rated greater improvement among multiple pairwise scenarios in conjoint analysis surveys, where different levels of improvement in 2 core set measures were presented. The PAPRIKA (Potentially All Pairwise Rankings of All Possible Alternatives) method determined the relative weights of core set measures and conjoint analysis definitions. The performance characteristics of the definitions were evaluated on patient profiles using expert consensus (gold standard) and were validated using data from a clinical trial. The nominal group technique was used to reach consensus. Consensus was reached for a conjoint analysis-based continuous model using absolute per cent change in core set measures (physician, patient, and extramuscular global activity, muscle strength, Health Assessment Questionnaire, and muscle enzyme levels). A total improvement score (range 0-100), determined by summing scores for each core set measure, was based on improvement in and relative weight of each core set measure. Thresholds for minimal, moderate, and major improvement were ≥20, ≥40, and ≥60 points in the total improvement score. The same criteria were chosen for juvenile DM, with different improvement thresholds. Sensitivity and specificity in DM/PM patient cohorts were 85% and 92%, 90% and 96%, and 92% and 98% for minimal, moderate, and major improvement, respectively. Definitions were validated in the clinical trial analysis for differentiating the physician rating of improvement (p<0.001). The response criteria for adult DM/PM consisted of the conjoint analysis model based on absolute per cent change in 6 core set measures, with thresholds for minimal, moderate, and major improvement.

2017 European League Against Rheumatism/American College of Rheumatology classification criteria for adult and juvenile idiopathic inflammatory myopathies and their major subgroups.

OBJECTIVE: To develop and validate new classification criteria for adult and juvenile idiopathic inflammatory myopathies (IIM) and their major subgroups. METHODS: Candidate variables were assembled from published criteria and expert opinion using consensus methodology. Data were collected from 47 rheumatology, dermatology, neurology and paediatric clinics worldwide. Several statistical methods were used to derive the classification criteria. RESULTS: Based on data from 976 IIM patients (74% adults; 26% children) and 624 non-IIM patients with mimicking conditions (82% adults; 18% children), new criteria were derived. Each item is assigned a weighted score. The total score corresponds to a probability of having IIM. Subclassification is performed using a classification tree. A probability cut-off of 55%, corresponding to a score of 5.5 (6.7 with muscle biopsy) 'probable IIM', had best sensitivity/specificity (87%/82% without biopsies, 93%/88% with biopsies) and is recommended as a minimum to classify a patient as having IIM. A probability of ≥90%, corresponding to a score of ≥7.5 (≥8.7 with muscle biopsy), corresponds to 'definite IIM'. A probability of <50%, corresponding to a score of <5.3 (<6.5 with muscle biopsy), rules out IIM, leaving a probability of ≥50 to <55% as 'possible IIM'. CONCLUSIONS: The European League Against Rheumatism/American College of Rheumatology (EULAR/ACR) classification criteria for IIM have been endorsed by international rheumatology, dermatology, neurology and paediatric groups. They employ easily accessible and operationally defined elements, and have been partially validated. They allow classification of 'definite', 'probable' and 'possible' IIM, in addition to the major subgroups of IIM, including juvenile IIM. They generally perform better than existing criteria.

2016 ACR-EULAR adult dermatomyositis and polymyositis and juvenile dermatomyositis response criteria-methodological aspects.

Objective: The objective was to describe the methodology used to develop new response criteria for adult DM/PM and JDM. Methods: Patient profiles from prospective natural history data and clinical trials were rated by myositis specialists to develop consensus gold-standard ratings of minimal, moderate and major improvement. Experts completed a survey regarding clinically meaningful improvement in the core set measures (CSM) and a conjoint-analysis survey (using 1000Minds software) to derive relative weights of CSM and candidate definitions. Six types of candidate definitions for response criteria were derived using survey results, logistic regression, conjoint analysis, application of conjoint-analysis weights to CSM and published definitions. Sensitivity, specificity and area under the curve were defined for candidate criteria using consensus patient profile data, and selected definitions were validated using clinical trial data. Results: Myositis specialists defined the degree of clinically meaningful improvement in CSM for minimal, moderate and major improvement. The conjoint-analysis survey established the relative weights of CSM, with muscle strength and Physician Global Activity as most important. Many candidate definitions showed excellent sensitivity, specificity and area under the curve in the consensus profiles. Trial validation showed that a number of candidate criteria differentiated between treatment groups. Top candidate criteria definitions were presented at the consensus conference. Conclusion: Consensus methodology, with definitions tested on patient profiles and validated using clinical trials, led to 18 definitions for adult PM/DM and 14 for JDM as excellent candidates for consideration in the final consensus on new response criteria for myositis.

Scientific factors and current issues in biosimilar studies.

Biological drugs are much more complicated than chemically synthesized, small-molecule drugs; for instance, their size is much larger, their structure is more complicated, they can be sensitive to environmental conditions such as temperature or pressure, and they may expose patients to immunogen reactions. Consequently, the assessment of biosimilarity calls for greater circumspection than the evaluation of bioequivalence. The present communication discusses scientific factors and some current issues related to biosimilarity and the interchangeability of drug products. The scientific factors include questions involving endpoint selection, the one-size-fits-all criterion, and the need for a more flexible approach, e.g., evaluation of the degree of similarity (i.e., responding to the question of "how similar is similar?"; a review of study designs that are useful for the assessment of biosimilarity and drug interchangeability; and tests for the comparability of critical quality attributes at various stages of the manufacturing process). Current issues include the choice of reference standards and the relevant study designs; criteria for biosimilarity, as well as for interchangeability and for comparability; the determination of the noninferiority margin; and the concepts of the stepwise approach to biosimilarity studies and of their assessment by the totality of the evidence. The calculation of sample sizes is discussed for crossover (including some higher-order schemes) and parallel designs.

Comments on the FDA draft guidance on biosimilar products.

The Food and Drug Administration issued on February 9, 2012, drafts of three new guidance documents about the demonstration of biosimilarity. One of these deals with scientific considerations. It suggests, among others, that demonstration of biosimilarity be developed by a stepwise (step-by-step) approach and that it be assessed by considering the totality of the evidence. This communication provides comments on some scientific factors and issues that still remain unanswered or unsolved. They include the question 'how similar is considered to be highly similar?' considerations of criteria for and the degree of biosimilarity; alternatives of study design and sample size requirements; statistical methods for achieving the totality of the evidence needed for biosimilarity; and methods needed for the assessment of drug interchangeability. It is anticipated that the comments will assist the revision of the guidance documents.

Scientific considerations for assessing biosimilar products.

The problem for assessing biosimilarity and drug interchangeability of follow-on biologics (biosimilar products) is studied. Unlike the generic products, the development of biosimilar products is much more complicated because of fundamental differences in functional structures and manufacturing processes. As a result, the criteria and standard methods for the design and analysis of bioequivalence assessment of generic drug products may not be directly applicable to assessing biosimilarity of biosimilar products. In this article, we provide some scientific considerations for criteria, design, and analysis regarding the assessment of biosimilarity and drug interchangeability of biosimilar products. In addition, we discuss scientific and practical issues raised at the 2010 FDA public hearing and the 2011 FDA public meeting on biosimilar products.

Developing international consensus on measures of improvement for patients with myositis.

We discuss methods of developing consensus in measuring improvement in myositis. We consider selecting candidate variables, reliability and validity, percentage improvement/worsening rules, rules based on CART and logistic regression. We discuss criteria for determining an acceptable rule that include both numerical measures and physician acceptance.

2016 American College of Rheumatology/European League Against Rheumatism Criteria for Minimal, Moderate, and Major Clinical Response in Adult Dermatomyositis and Polymyositis: An International Myositis Assessment and Clinical Studies Group/Paediatric Rheumatology International Trials Organisation Collaborative Initiative.

OBJECTIVE: To develop response criteria for adult dermatomyositis (DM) and polymyositis (PM). METHODS: Expert surveys, logistic regression, and conjoint analysis were used to develop 287 definitions using core set measures. Myositis experts rated greater improvement among multiple pairwise scenarios in conjoint analysis surveys, where different levels of improvement in 2 core set measures were presented. The PAPRIKA (Potentially All Pairwise Rankings of All Possible Alternatives) method determined the relative weights of core set measures and conjoint analysis definitions. The performance characteristics of the definitions were evaluated on patient profiles using expert consensus (gold standard) and were validated using data from a clinical trial. The nominal group technique was used to reach consensus. RESULTS: Consensus was reached for a conjoint analysis-based continuous model using absolute percent change in core set measures (physician, patient, and extramuscular global activity, muscle strength, Health Assessment Questionnaire, and muscle enzyme levels). A total improvement score (range 0-100), determined by summing scores for each core set measure, was based on improvement in and relative weight of each core set measure. Thresholds for minimal, moderate, and major improvement were ≥20, ≥40, and ≥60 points in the total improvement score. The same criteria were chosen for juvenile DM, with different improvement thresholds. Sensitivity and specificity in DM/PM patient cohorts were 85% and 92%, 90% and 96%, and 92% and 98% for minimal, moderate, and major improvement, respectively. Definitions were validated in the clinical trial analysis for differentiating the physician rating of improvement (P < 0.001). CONCLUSION: The response criteria for adult DM/PM consisted of the conjoint analysis model based on absolute percent change in 6 core set measures, with thresholds for minimal, moderate, and major improvement.

2017 European League Against Rheumatism/American College of Rheumatology Classification Criteria for Adult and Juvenile Idiopathic Inflammatory Myopathies and Their Major Subgroups.

OBJECTIVE: To develop and validate new classification criteria for adult and juvenile idiopathic inflammatory myopathies (IIM) and their major subgroups. METHODS: Candidate variables were assembled from published criteria and expert opinion using consensus methodology. Data were collected from 47 rheumatology, dermatology, neurology, and pediatric clinics worldwide. Several statistical methods were utilized to derive the classification criteria. RESULTS: Based on data from 976 IIM patients (74% adults; 26% children) and 624 non-IIM patients with mimicking conditions (82% adults; 18% children), new criteria were derived. Each item is assigned a weighted score. The total score corresponds to a probability of having IIM. Subclassification is performed using a classification tree. A probability cutoff of 55%, corresponding to a score of 5.5 (6.7 with muscle biopsy) "probable IIM," had best sensitivity/specificity (87%/82% without biopsies, 93%/88% with biopsies) and is recommended as a minimum to classify a patient as having IIM. A probability of ≥90%, corresponding to a score of ≥7.5 (≥8.7 with muscle biopsy), corresponds to "definite IIM." A probability of <50%, corresponding to a score of <5.3 (<6.5 with muscle biopsy), rules out IIM, leaving a probability of ≥50-<55% as "possible IIM." CONCLUSION: The European League Against Rheumatism/American College of Rheumatology (EULAR/ACR) classification criteria for IIM have been endorsed by international rheumatology, dermatology, neurology, and pediatric groups. They employ easily accessible and operationally defined elements, and have been partially validated. They allow classification of "definite," "probable," and "possible" IIM, in addition to the major subgroups of IIM, including juvenile IIM. They generally perform better than existing criteria.

EULAR/ACR classification criteria for adult and juvenile idiopathic inflammatory myopathies and their major subgroups: a methodology report.

OBJECTIVE: To describe the methodology used to develop new classification criteria for adult and juvenile idiopathic inflammatory myopathies (IIMs) and their major subgroups. METHODS: An international, multidisciplinary group of myositis experts produced a set of 93 potentially relevant variables to be tested for inclusion in the criteria. Rheumatology, dermatology, neurology and paediatric clinics worldwide collected data on 976 IIM cases (74% adults, 26% children) and 624 non-IIM comparator cases with mimicking conditions (82% adults, 18% children). The participating clinicians classified each case as IIM or non-IIM. Generally, the classification of any given patient was based on few variables, leaving remaining variables unmeasured. We investigated the strength of the association between all variables and between these and the disease status as determined by the physician. We considered three approaches: (1) a probability-score approach, (2) a sum-of-items approach criteria and (3) a classification-tree approach. RESULTS: The approaches yielded several candidate models that were scrutinised with respect to statistical performance and clinical relevance. The probability-score approach showed superior statistical performance and clinical practicability and was therefore preferred over the others. We developed a classification tree for subclassification of patients with IIM. A calculator for electronic devices, such as computers and smartphones, facilitates the use of the European League Against Rheumatism/American College of Rheumatology (EULAR/ACR) classification criteria. CONCLUSIONS: The new EULAR/ACR classification criteria provide a patient's probability of having IIM for use in clinical and research settings. The probability is based on a score obtained by summing the weights associated with a set of criteria items.

2016 American College of Rheumatology/European League Against Rheumatism Criteria for Minimal, Moderate, and Major Clinical Response in Juvenile Dermatomyositis: An International Myositis Assessment and Clinical Studies Group/Paediatric Rheumatology International Trials Organisation Collaborative Initiative.

OBJECTIVE: To develop response criteria for juvenile dermatomyositis (DM). METHODS: We analyzed the performance of 312 definitions that used core set measures from either the International Myositis Assessment and Clinical Studies Group (IMACS) or the Paediatric Rheumatology International Trials Organisation (PRINTO) and were derived from natural history data and a conjoint analysis survey. They were further validated using data from the PRINTO trial of prednisone alone compared to prednisone with methotrexate or cyclosporine and the Rituximab in Myositis (RIM) trial. At a consensus conference, experts considered 14 top candidate criteria based on their performance characteristics and clinical face validity, using nominal group technique. RESULTS: Consensus was reached for a conjoint analysis-based continuous model with a total improvement score of 0-100, using absolute percent change in core set measures of minimal (≥30), moderate (≥45), and major (≥70) improvement. The same criteria were chosen for adult DM/polymyositis, with differing thresholds for improvement. The sensitivity and specificity were 89% and 91-98% for minimal improvement, 92-94% and 94-99% for moderate improvement, and 91-98% and 85-86% for major improvement, respectively, in juvenile DM patient cohorts using the IMACS and PRINTO core set measures. These criteria were validated in the PRINTO trial for differentiating between treatment arms for minimal and moderate improvement (P = 0.009-0.057) and in the RIM trial for significantly differentiating the physician's rating for improvement (P < 0.006). CONCLUSION: The response criteria for juvenile DM consisted of a conjoint analysis-based model using a continuous improvement score based on absolute percent change in core set measures, with thresholds for minimal, moderate, and major improvement.

Relative contributions of interface pressure, shear stress, and temperature on ischemic-induced, skin-reactive hyperemia in healthy volunteers: a repeated measures laboratory study.

Although the primary risk factors for pressure ulcer development - pressure, shear, skin temperature, moisture, and friction - have been identified for decades, the relative contribution of each to this risk remains unclear. To confirm the results of and expand upon earlier research into the relative contributions of interface pressures, shear stress, and skin temperature among 4 healthy volunteers, a study involving 6 additional healthy 40- to 75-year-old volunteers was conducted and results of the 2 studies were pooled. All 3 variables (interface pressures, shear stress, and skin temperature) were systematically and randomly varied. In the prone position, volunteers each underwent 18 test conditions representing different combinations of temperature (28˚ C, 32˚ C, 36˚ C), pressure (8.0 and 13.3 kPa), and shear (0, 6.7, and 14.0 kPa) using a computer-controlled indenter applied to the sacrum for 20 minutes exerting weights of 100 g and 200 g to induce 0.98 N and 1.96 N of shear force, respectively. Each condition was tested twice, resulting in a total of 360 trials. Magnitude of postload reactive hyperemia as an index of ischemia was measured by laser Doppler flowmetry. Fixed effects regression models were used to predict 3 different indices of reactive hyperemic magnitude. Friedman tests were performed to compare the reactive hyperemia among 3 different skin temperatures or shear stresses under the same amount of localized pressure. In all regression models, pressure and temperature were highly significant predictors of the extent of reactive hyperemia (P <0.0001 and P <0.0001, respectively); the contributions of shear stress were not statistically significant (P = 0.149). With higher temperature, reactive hyperemia increased significantly, especially at greater localized pressure and shear stress, and the difference was more profound between 32˚ C and 36˚ C than between 28˚ C and 32˚ C. These results confirm that, in laboratory settings, temperature is an important factor in tissue ischemia. Additional studies examining the relative importance of pressure, shear, and temperature and potential effects of lowering temperature on tissue ischemia in healthy volunteers and patients at risk for pressure ulcer development are warranted. Because deformation at weight-bearing areas often results in blood flow occlusion, actively lowering the temperature may reduce the severity of ischemia and lower pressure ulcer risk. In this study, shear did not appear to contribute to ischemia in the dermal tissues when assessed using laser Doppler; further work is needed to examine its effect on deeper layers, particularly with regard to nonischemic mechanisms.

Fatalities caused by TRALI.

This article includes a retrospective review of fatalities caused by transfusion-related acute lung injury (TRALI) over a 5-year period (from 1997 to 2002) that were reported to the Center for Biologics Evaluation and Research involving 58 recipient deaths and the corresponding 63 blood component donors. Descriptive statistics are presented. Recipient characteristics include age, sex, and admitting diagnosis. Cardiovascular disease, pulmonary disorders, and cancer were the most frequent diagnoses of transfusion recipients. Reported deaths did not appear to be associated with age, sex, reason for transfusion, or transfusion component. Implicated blood component(s) and clinical symptoms at the time of reaction were recorded. Fresh frozen plasma was implicated in one half of the cases, whereas red blood cells played a role in approximately one third. The clinical characteristics described most often in TRALI reports included shortness of breath, frothy sputum, pulmonary infiltrates, and hypoxia. Donor variables included age, sex, parity, and laboratory tests for antibodies to HLA and/or antigranulocyte antibodies. Laboratory tests showed HLA antibodies and/or antigranulocyte antibodies were positive in the majority of donors tested. More data are needed to better describe the role of antibodies in these reactions. Greater awareness is crucial for the practitioner to be alert for signs and symptoms of TRALI and to be aware of the necessary steps in treatment.

Analysis of data with excess zeros.

This paper is an introduction to a set of papers on two-part models that were presented at the 2000 Joint Statistical Meetings of American Statistical Association, International Biometric Society (ENAR), International Biometric Society (WNAR), Institute of Mathematical Statistics, and Statistical Society of Canada. The second part of the paper summarizes results of a study of the size and power of two-part models. It proposes that two-part models are a useful alternative to the t-test and the Wilcoxon test.

Early illness features associated with mortality in the juvenile idiopathic inflammatory myopathies.

OBJECTIVE: Because juvenile idiopathic inflammatory myopathies (IIMs) are potentially life-threatening systemic autoimmune diseases, we examined risk factors for juvenile IIM mortality. METHODS: Mortality status was available for 405 patients (329 with juvenile dermatomyositis [DM], 30 with juvenile polymyositis [PM], and 46 with juvenile connective tissue disease-associated myositis [CTM]) enrolled in nationwide protocols. Standardized mortality ratios (SMRs) were calculated using US population statistics. Cox regression analysis was used to assess univariable associations with mortality, and random survival forest (RSF) classification and Cox regression analysis were used for multivariable associations. RESULTS: Of 17 deaths (4.2% overall mortality), 8 (2.4%) were in juvenile DM patients. Death was related to the pulmonary system (primarily interstitial lung disease [ILD]) in 7 patients, gastrointestinal system in 3, and multisystem in 3, and of unknown etiology in 4 patients. The SMR for juvenile IIMs overall was 14.4 (95% confidence interval [95% CI] 12.2-16.5) and was 8.3 (95% CI 6.4-10.3) for juvenile DM. The top mortality risk factors in the univariable analysis included clinical subgroup (juvenile CTM, juvenile PM), antisynthetase autoantibodies, older age at diagnosis, ILD, and Raynaud's phenomenon at diagnosis. In multivariable analyses, clinical subgroup, illness severity at onset, age at diagnosis, weight loss, and delay to diagnosis were the most important predictors from RSF; clinical subgroup and illness severity at onset were confirmed by multivariable Cox regression analysis. CONCLUSION: Overall mortality was higher in juvenile IIM patients, and several early illness features were identified as risk factors. Clinical subgroup, antisynthetase autoantibodies, older age at diagnosis, and ILD are also recognized as mortality risk factors in adult myositis.

Clinical and laboratory features distinguishing juvenile polymyositis and muscular dystrophy.

OBJECTIVE: To differentiate juvenile polymyositis (PM) and muscular dystrophy, both of which may present with chronic muscle weakness and inflammation. METHODS: We studied 39 patients with probable or definite juvenile PM and 9 patients with muscular dystrophies who were initially misdiagnosed as having juvenile PM. Differences in demographic, clinical, and laboratory results; outcomes; and treatment responses were evaluated by Fisher's exact and rank sum tests. Random forests classification analysis and logistic regression were performed to examine significant differences in multivariable models. RESULTS: Clinical features and serum muscle enzyme levels were similar between juvenile PM and dystrophy patients, except 89% of dystrophy patients had muscle atrophy compared with 46% of juvenile PM patients. Dystrophy patients had a longer delay to diagnosis (median 12 versus 4 months) and were less frequently hospitalized than juvenile PM patients (22% versus 74%). No dystrophy patients, but 54% of juvenile PM patients, had a myositis autoantibody. Dystrophy patients more frequently had myopathic features on muscle biopsy, including diffuse variation of myofiber size, fiber hypertrophy, and myofiber fibrosis (44-100% versus 8-53%). Juvenile PM patients more frequently had complex repetitive discharges on electromyography and a complete response to treatment with prednisone or other immunosuppressive agents than dystrophy patients (44% versus 0%). Random forests analysis revealed that the most important features in distinguishing juvenile PM from dystrophies were myositis autoantibodies, clinical muscle atrophy, and myofiber size variation on biopsy. Logistic regression confirmed muscle atrophy, myofiber fibrosis, and hospitalization as significant predictors. CONCLUSION: Muscular dystrophy can present similarly to juvenile PM. Selected clinical and laboratory features are helpful in combination in distinguishing these conditions.

Variable selection when missing values are present: a case study.

We consider variable selection when missing values are present in the predictor variables. We compare using complete cases with multiple imputation using backward selection (backwards stepping) and least angle regression. These are studied using a data set from a rheumatological disease (myositis). We find that the coefficients are slightly different and the estimated standard errors are smaller in the complete cases (not a surprise). This seems to be due to the fact that because the estimated residual variance is small the complete cases are more homogeneous than the full data cases.

On Determining the Effects of Therapy on Disease Damage in Non- randomized Studies with Multiple Treatments: A study of Juvenile Myositis.

This article discusses predicting damage in patients with juvenile myositis after treatment with various medications. These data were taken from medical records and not randomized. Investigators advocate using propensity scores for analysis of such non-randomized studies in order to reduce the effect of selection of treatment. Thus far, the studies have typically been comparing drug administration versus no drug after including a propensity score to compensate for potential bias in selecting patients for use of the agent. In this study, we use propensity scoring for multiple treatments given singly or in combination. We study two methods. We use a multiple logistic regression model with continuous propensity scores and a model that develops strata based on the dichotomous treatment assignment (received drug or not). We find the multiple logistic regression models predict damage better than the dichotomous model. In many cases, the propensity score also accounts for the effect of the treatment.