Why estimands are needed to define treatment effects in clinical trials.

BACKGROUND: The estimand for a clinical trial is a precise definition of the treatment effect to be estimated. Traditionally, estimates of treatment effects are based on either an ITT analysis or a per-protocol analysis. However, there are important clinical questions which are not addressed by either of these analyses. For example, consider a trial where patients take a rescue medication. The ITT analysis includes data after use of rescue, while the per-protocol analysis excludes these patients altogether. Neither of these analyses addresses the important question of what the treatment effect would have been if patients did not take rescue medication. MAIN TEXT: Trial estimands provide a broader perspective compared to the limitations of ITT and per-protocol analysis. Trial treatment effects depend on how events occurring after treatment initiation such as use of alternative medication or discontinuation of the intervention are included in the definition. These events can be accounted for in different ways, depending on the clinical question of interest. CONCLUSION: The estimand framework is an important step forward in improving the clarity and transparency of clinical trials. The centrality of estimands to clinical trials is currently not reflected in methods recommended by the Cochrane group or the CONSORT statement, the current standard for reporting clinical trials in medical journals. We encourage revisions to these guidelines.

Principles and recommendations for incorporating estimands into clinical study protocol templates.

Clinical study protocols are the foundation of good clinical studies. Prospective and multidisciplinary collaboration that pays attention to the design of all components of the study protocol can ensure that a clinical study will answer the research questions posed in a reliable manner that is meaningful for decision-makers and patients. The ICH E9(R1) addendum on estimands and sensitivity analysis in clinical trials provides a framework for clinical study planning to ensure alignment between study objectives, design, conduct, and analysis. The estimand or clinical question posed can be regarded as the backbone of the study and the clinical study protocol should reflect estimands accordingly. In practice, stakeholders are still learning how to embrace the estimand framework and how it impacts studies and study documents. In this paper, we anticipate that a protocol structure centred around estimands, or objectives rather than endpoints alone will prevail for all types of studies. To assist sponsors during this paradigm shift, this paper provides discussion and guidance for implementing the estimand framework in protocol templates.

Rationale and design of the phase 2b clinical trials to study the effects of the partial adenosine A1-receptor agonist neladenoson bialanate in patients with chronic heart failure with reduced (PANTHEON) and preserved (PANACHE) ejection fraction.

Despite major advances in the treatment of chronic heart failure (HF) with reduced ejection fraction (HFrEF), morbidity and mortality associated with the condition remain high, suggesting the need for additional treatment options, particularly haemodynamically neutral treatments that do not alter blood pressure, heart rate, or renal function. HF with preserved ejection fraction (HFpEF) is also associated with high morbidity and mortality and adequate treatment options are limited; thus there is a critical unmet need for the development of novel therapies for HFpEF. Chronic HFrEF and HFpEF are both systemic disorders that affect not only the heart but several other tissues and organs including skeletal muscle, leading to exercise intolerance and dyspnoea. Partial adenosine A1-receptor agonists represent a novel potential therapy for HF regardless of underlying ejection fraction given their minimal effect on heart rate and blood pressure, and preclinical data demonstrate several possible beneficial mechanisms, including improved mitochondrial function and sarcoplasmic reticulum Ca2+ -ATPase (SERCA2a) activity, enhanced energy substrate utilization, reverse ventricular remodelling, and anti-ischemic, cardioprotective properties. However, data on this class of drugs in humans are scarce, and the optimal dose of the partial adenosine A1 receptor, neladenoson bialanate, has not been defined. Here we describe the design and rationale of two randomized, double-blind, placebo-controlled, parallel-group, dose-finding phase 2b trials, PANTHEON (HFrEF) and PANACHE (HFpEF), that will advance our understanding of the potential benefit and optimal dose of neladenoson bialanate and provide critical information for the planning of future phase 3 trials.

Patterns of missing data in the use of the endometriosis symptom diary.

BACKGROUND: Endometriosis is a common, chronic condition in women of reproductive age that is characterized by the presence of functional endometriotic lesions outside the uterus. The Endometriosis Symptom Diary (ESD) is an electronic patient-reported outcome (ePRO) instrument that assesses women's experience of endometriosis symptoms, with pain scored using a 0-10 numeric rating scale. This study investigated patterns of data missing from the ESD in the VALEPRO study. METHODS: Post hoc analyses of missing data were conducted. RESULTS: Of 272 participants using the ESD, 26.5% had no missing diary entries, 46.7% had > 0-5% of entries missing, 13.2% had > 5-10% of entries missing and 13.6% had > 10% of entries missing over the entire study period. The duration of missing episodes (defined as ≥1 consecutive days with missing diary entries) was generally short; most (81.4%) were 1 day. The difference in mean worst pain scores between missing and complete episodes per participant was - 0.1, suggesting that missing episodes were not related to severity of pain. Entries were significantly more likely to be missing on Fridays (18.5%) and Saturdays (22.9%) compared with other days of the week (p < 0.0001). Participants in the USA had significantly more long missing episodes than those in Germany (proportions of missing episodes longer than 1 day, 22.6 and 10.5%, respectively; p < 0.0001). The proportions of women with ≥1 missing entry were 50.0, 70.2 and 79.8% for women with elementary education, secondary education, and a college or university education, respectively. The proportions of women with ≥1 missing entry were similar for those with and without children (72.2 and 74.3%, respectively). CONCLUSIONS: Most participants were highly compliant with entering data in the ESD and the amount of missing data was low. Entries were significantly more likely to be missing on Fridays and Saturdays compared with other days of the week, and participants in the USA had significantly more long missing episodes than participants in Germany. TRIAL REGISTRATION: Clinicaltrials.gov, NCT01643122 , registered 4 July 2012.

Major Adverse Limb Events and Mortality in Patients With Peripheral Artery Disease: The COMPASS Trial.

BACKGROUND: Patients with lower extremity peripheral artery disease (PAD) are at increased risk of major adverse cardiovascular events (MACE) and major adverse limb events (MALE). There is limited information on the prognosis of patients who experience MALE. OBJECTIVES: Among participants with lower extremity PAD, this study investigated: 1) if hospitalizations, MACE, amputations, and deaths are higher after the first episode of MALE compared with patients with PAD who do not experience MALE; and 2) the impact of treatment with low-dose rivaroxaban and aspirin compared with aspirin alone on the incidence of MALE, peripheral vascular interventions, and all peripheral vascular outcomes over a median follow-up of 21 months. METHODS: We analyzed outcomes in 6,391 patients with lower extremity PAD who were enrolled in the COMPASS (Cardiovascular Outcomes for People Using Anticoagulation Strategies) trial. COMPASS was a randomized, double-blind placebo-controlled study of low-dose rivaroxaban and aspirin combination or rivaroxaban alone compared with aspirin alone. MALE was defined as severe limb ischemia leading to an intervention or major vascular amputation. RESULTS: A total of 128 patients experienced an incident of MALE. After MALE, the 1-year cumulative risk of a subsequent hospitalization was 61.5%; for vascular amputations, it was 20.5%; for death, it was 8.3%; and for MACE, it was 3.7%. The MALE index event significantly increased the risk of experiencing subsequent hospitalizations (hazard ratio [HR]: 7.21; p < 0.0001), subsequent amputations (HR: 197.5; p < 0.0001), and death (HR: 3.23; p < 0.001). Compared with aspirin alone, the combination of rivaroxaban 2.5 mg twice daily and aspirin lowered the incidence of MALE by 43% (p = 0.01), total vascular amputations by 58% (p = 0.01), peripheral vascular interventions by 24% (p = 0.03), and all peripheral vascular outcomes by 24% (p = 0.02). CONCLUSIONS: Among individuals with lower extremity PAD, the development of MALE is associated with a poor prognosis, making prevention of this condition of utmost importance. The combination of rivaroxaban 2.5 mg twice daily and aspirin significantly lowered the incidence of MALE and the related complications, and this combination should be considered as an important therapy for patients with PAD. (Cardiovascular Outcomes for People Using Anticoagulation Strategies [COMPASS]; NCT01776424).

Rationale, Design and Baseline Characteristics of Participants in the Cardiovascular Outcomes for People Using Anticoagulation Strategies (COMPASS) Trial.

BACKGROUND: Long-term aspirin prevents vascular events but is only modestly effective. Rivaroxaban alone or in combination with aspirin might be more effective than aspirin alone for vascular prevention in patients with stable coronary artery disease (CAD) or peripheral artery disease (PAD). Rivaroxaban as well as aspirin increase upper gastrointestinal (GI) bleeding and this might be prevented by proton pump inhibitor therapy. METHODS: Cardiovascular Outcomes for People Using Anticoagulation Strategies (COMPASS) is a double-blind superiority trial comparing rivaroxaban 2.5 mg twice daily combined with aspirin 100 mg once daily or rivaroxaban 5 mg twice daily vs aspirin 100 mg once daily for prevention of myocardial infarction, stroke, or cardiovascular death in patients with stable CAD or PAD. Patients not taking a proton pump inhibitor were also randomized, using a partial factorial design, to pantoprazole 40 mg once daily or placebo. The trial was designed to have at least 90% power to detect a 20% reduction in each of the rivaroxaban treatment arms compared with aspirin and to detect a 50% reduction in upper GI complications with pantoprazole compared with placebo. RESULTS: Between February 2013 and May 2016, we recruited 27,395 participants from 602 centres in 33 countries; 17,598 participants were included in the pantoprazole vs placebo comparison. At baseline, the mean age was 68.2 years, 22.0% were female, 90.6% had CAD, and 27.3% had PAD. CONCLUSIONS: COMPASS will provide information on the efficacy and safety of rivaroxaban, alone or in combination with aspirin, in the long-term management of patients with stable CAD or PAD, and on the efficacy and safety of pantoprazole in preventing upper GI complications in patients receiving antithrombotic therapy.

Rationale, design and baseline characteristics of participants in the cardiovascular outcomes for people using anticoagulation strategies (COMPASS) trial

BACKGROUND: Long-term aspirin prevents vascular events but is only modestly effective. Rivaroxaban alone or in combination with aspirin might be more effective than aspirin alone for vascular prevention in patients with stable coronary artery disease (CAD) or peripheral artery disease (PAD). Rivaroxaban as well as aspirin increase upper gastrointestinal (GI) bleeding and this might be prevented by proton pump inhibitor therapy. METHODS: Cardiovascular Outcomes for People Using Anticoagulation Strategies (COMPASS) is a double-blind superiority trial comparing rivaroxaban 2.5 mg twice daily combined with aspirin 100 mg once daily or rivaroxaban 5 mg twice daily vs aspirin 100 mg once daily for prevention of myocardial infarction, stroke, or cardiovascular death in patients with stable CAD or PAD. Patients not taking a proton pump inhibitor were also randomized, using a partial factorial design, to pantoprazole 40 mg once daily or placebo. The trial was designed to have at least 90% power to detect a 20% reduction in each of the rivaroxaban treatment arms compared with aspirin and to detect a 50% reduction in upper GI complications with pantoprazole compared with placebo...

Prediction of worsening of skin fibrosis in patients with diffuse cutaneous systemic sclerosis using the EUSTAR database.

OBJECTIVES: To identify predictive parameters for the progression of skin fibrosis within 1 year in patients with diffuse cutaneous SSc (dcSSc). METHODS: An observational study using the EUSTAR database was performed. Inclusion criteria were dcSSc, American College of Rheumatology (ACR) criteria fulfilled, modified Rodnan skin score (MRSS) ≥7 at baseline visit, valid data for MRSS at 2nd visit, and available follow-up of 12±2 months. Worsening of skin fibrosis was defined as increase in MRSS >5 points and ≥25% from baseline to 2nd visit. In the univariate analysis, patients with progressive fibrosis were compared with non-progressors, and predictive markers with p<0.2 were included in the logistic regression analysis. The prediction models were then validated in a second cohort. RESULTS: A total of 637 dcSSc patients were eligible. Univariate analyses identified joint synovitis, short disease duration (≤15 months), short disease duration in females/patients without creatine kinase (CK) elevation, low baseline MRSS (≤22/51), and absence of oesophageal symptoms as potential predictors for progressive skin fibrosis. In the multivariate analysis, by employing combinations of the predictors, 17 models with varying prediction success were generated, allowing cohort enrichment from 9.7% progressive patients in the whole cohort to 44.4% in the optimised enrichment cohort. Using a second validation cohort of 188 dcSSc patients, short disease duration, low baseline MRSS and joint synovitis were confirmed as independent predictors of progressive skin fibrosis within 1 year resulting in a 4.5-fold increased prediction success rate. CONCLUSIONS: Our study provides novel, evidence-based criteria for the enrichment of dcSSc cohorts with patients who experience worsening of skin fibrosis which allows improved clinical trial design.

Rivaroxaban vs. vitamin K antagonists for cardioversion in atrial fibrillation.

AIMS: X-VeRT is the first prospective randomized trial of a novel oral anticoagulant in patients with atrial fibrillation undergoing elective cardioversion. METHODS AND RESULTS: We assigned 1504 patients to rivaroxaban (20 mg once daily, 15 mg if creatinine clearance was between 30 and 49 mL/min) or dose-adjusted vitamin K antagonists (VKAs) in a 2:1 ratio. Investigators selected either an early (target period of 1-5 days after randomization) or delayed (3-8 weeks) cardioversion strategy. The primary efficacy outcome was the composite of stroke, transient ischaemic attack, peripheral embolism, myocardial infarction, and cardiovascular death. The primary safety outcome was major bleeding. The primary efficacy outcome occurred in 5 (two strokes) of 978 patients (0.51%) in the rivaroxaban group and in 5 (two strokes) of 492 patients (1.02%) in the VKA group [risk ratio 0.50; 95% confidence interval (CI) 0.15-1.73]. In the rivaroxaban group, four patients experienced primary efficacy events following early cardioversion (0.71%) and one following delayed cardioversion (0.24%). In the VKA group, three patients had primary efficacy events following early cardioversion (1.08%) and two following delayed cardioversion (0.93%). Rivaroxaban was associated with a significantly shorter time to cardioversion compared with VKAs (P < 0.001). Major bleeding occurred in six patients (0.6%) in the rivaroxaban group and four patients (0.8%) in the VKA group (risk ratio 0.76; 95% CI 0.21-2.67). CONCLUSION: Oral rivaroxaban appears to be an effective and safe alternative to VKAs and may allow prompt cardioversion. NAME OF THE TRIAL REGISTRY: Clinicaltrials.gov; TRIAL REGISTRATION NUMBER: NCT01674647.

Rationale and design of the eXplore the efficacy and safety of once-daily oral riVaroxaban for the prEvention of caRdiovascular events in patients with nonvalvular aTrial fibrillation scheduled for cardioversion trial: A comparison of oral rivaroxaban once daily with dose-adjusted vitamin K antagonists in patients with nonvalvular atrial fibrillation undergoing elective cardioversion.

BACKGROUND: Anticoagulation before, during, and after cardioversion is effective in reducing stroke risk in patients with atrial fibrillation. OBJECTIVE: The objective of this study is to explore the efficacy and safety of rivaroxaban 20 mg once daily (15 mg if creatinine clearance is 30-49 mL/min) compared with dose-adjusted vitamin K antagonists (VKAs; international normalized ratio 2.0-3.0) in patients scheduled for elective cardioversion. METHODS: This is a prospective, randomized, open-label, parallel group comparison of approximately 1,500 patients from 17 countries with hemodynamically stable nonvalvular atrial fibrillation of >48 hours or unknown duration. Patients will be randomized 2:1 (rivaroxaban:VKA) using 2 cardioversion strategies: the first approach is early cardioversion with the precardioversion anticoagulation goal of 1 to 5 days using rivaroxaban or usual therapy (heparin + VKA). In these patients, transesophageal echocardiography will be encouraged to exclude atrial thrombi. The alternative approach is delayed cardioversion. Rivaroxaban or VKA will be administered for 21 to 56 days before cardioversion. All patients will receive study treatment for 6 weeks postcardioversion. The primary efficacy end point is a composite of all strokes, transient ischemic attacks, noncentral nervous system systemic emboli, myocardial infarctions, and cardiovascular deaths. Each primary end point component will be evaluated separately, and additional composites will be investigated. The principal safety end point is major bleeding. CLINICAL CONTEXT: This will be the first prospective study of a novel oral anticoagulant in the setting of cardioversion. It will provide important information regarding the use of rivaroxaban in the periods preceding and after cardioversion in a broad patient population.

Cost-effectiveness analysis of interferon beta-1b for the treatment of patients with a first clinical event suggestive of multiple sclerosis.

OBJECTIVES: To assess, from a Swedish societal perspective, the cost effectiveness of interferon ß-1b (IFNB-1b) after an initial clinical event suggestive of multiple sclerosis (MS) (ie, early treatment) compared with treatment after onset of clinically definite MS (CDMS) (ie, delayed treatment). METHODS: A Markov model was developed, using patient level data from the BENEFIT trial and published literature, to estimate health outcomes and costs associated with IFNB-1b for hypothetical cohorts of patients after an initial clinical event suggestive of MS. Health states were defined by Kurtzke Expanded Disability Status Scale (EDSS) scores. Model outcomes included quality-adjusted life years (QALYs), total costs (including both direct and indirect costs), and incremental cost-effectiveness ratios. Sensitivity analyses were performed on key model parameters to assess the robustness of model results. RESULTS: In the base case scenario, early IFNB-1b treatment was economically dominant (ie, less costly and more effective) versus delayed IFNB-1b treatment when QALYs were used as the effectiveness metric. Sensitivity analyses showed that the cost-effectiveness results were sensitive to model time horizon. Compared with the delayed treatment strategy, early treatment of MS was also associated with delayed EDSS progressions, prolonged time to CDMS diagnosis, and a reduction in frequency of relapse. CONCLUSION: Early treatment with IFNB-1b for a first clinical event suggestive of MS was found to improve patient outcomes while controlling costs.

Effects of interferon beta-1b on cognitive performance in patients with a first event suggestive of multiple sclerosis.

BACKGROUND: Cognitive dysfunction occurs at the earliest stages of multiple sclerosis (MS), including the stage of clinically isolated syndrome (CIS). METHODS: We evaluated the impact of interferon beta-1b (IFNß-1b) 250 µg on cognitive performance during the CIS stage in the BENEFITstudy. Cognition was assessed by Paced Auditory Serial Addition Test-3" (PASAT-3") scores. RESULTS: Improvement in PASAT-3" score from baseline to year two was greater for IFNß-1b treatment than placebo in patients not reaching clinically definite MS (CDMS) by year two. The treatment effect was maintained at year five and was statistically significant. CONCLUSIONS: To conclude, early IFNß-1b treatment had a sustained positive effect on PASAT-3" score over the 5-year BENEFIT study.

Magnetic resonance imaging predictors of conversion to multiple sclerosis in the BENEFIT study.

BACKGROUND: Several studies have confirmed the predictive value of baseline and follow-up magnetic resonance (MR) imaging variables for conversion to clinically definite multiple sclerosis (CDMS), depending on the population, follow-up duration, and treatment intervention. However, the timing of follow-up imaging and the effect of treatment intervention on the predictive value of baseline MR imaging variables require further elucidation. OBJECTIVES: To assess the prognostic value of baseline MR imaging variables for conversion to CDMS over 3 years and whether this was affected by treatment intervention and (2) to assess the increased risk for conversion posed by dissemination in time on follow-up MR imaging. DESIGN: Cohort study. SETTING: Multicenter randomized clinical trial. PATIENTS: Four hundred sixty-eight patients with a clinically isolated syndrome who had an initial clinical demyelinating event within the past 60 days who received early treatment (3 years of interferon beta-1b) or delayed treatment (placebo first, followed by > or =1 year of interferon beta-1b). Intervention Magnetic resonance imaging. Main Outcome Measure Time to CDMS. RESULTS: The overall conversion rate to CDMS was 42%. Barkhof criteria with the strongest prognostic value were the presence at baseline of at least 9 T2-weighted lesions (hazard ratio [HR], 1.64; 95% confidence interval [CI], 1.15-2.33; P = .006) and at least 3 periventricular lesions (1.66; 1.14-2.41; P = .009). No specific advantage was noted in using a fixed cutoff of at least 3 Barkhof criteria (HR, 1.31; 95% CI, 0.95-1.79; P = .10). The prognostic value of all MR imaging criteria was unaffected by treatment intervention (P > or = .20 for all). Dissemination in time resulted in increased risk for CDMS only in patients without dissemination in space at baseline and was most informative at the 9-month MR imaging (HR, 2.72; 95% CI, 1.26-5.87; P = .01). CONCLUSIONS: The modified Barkhof criteria showed moderate predictive value for conversion to CDMS, although all patients had received interferon beta-1b therapy for at least 1 year. The predictive value was unaffected by treatment intervention. Follow-up MR imaging was most informative after 9 months in patients without dissemination in space at baseline.

Long-term effect of early treatment with interferon beta-1b after a first clinical event suggestive of multiple sclerosis: 5-year active treatment extension of the phase 3 BENEFIT trial.

BACKGROUND: The Betaferon/Betaseron in newly emerging multiple sclerosis for initial treatment (BENEFIT) trial investigated the effect of treatment with interferon beta-1b after a clinically isolated syndrome. The 5-year active treatment extension compares the effects of early and delayed treatment with interferon beta-1b on time to clinically definite multiple sclerosis (CDMS) and other disease outcomes, including disability progression. METHODS: Patients with a first event suggestive of multiple sclerosis and a minimum of two clinically silent lesions in MRI were randomly assigned to receive interferon beta-1b 250 microg (n=292; early treatment) or placebo (n=176; delayed treatment) subcutaneously every other day for 2 years, or until diagnosis of CDMS. All patients were then eligible to enter a prospectively planned follow-up phase with open-label interferon beta-1b up to a maximum of 5 years after randomisation. Patients and study personnel remained unaware of initial treatment allocation throughout the study. Primary endpoints were time to CDMS, time to confirmed disability progression measured with the expanded disability status scale, and the functional assessment of multiple sclerosis trial outcomes index (FAMS-TOI) at 5 years. Analysis of the primary endpoints was by intention to treat. This trial is registered with ClinicalTrials.gov, number NCT00185211. FINDINGS: 235 (80%) patients from the early treatment and 123 (70%) from the delayed treatment group completed the 5-year study. Early treatment reduced the risk of CDMS by 37% (hazard ratio [HR] 0.63, 95% CI 0.48-0.83; p=0.003) compared with delayed treatment. The risk for confirmed disability progression was not significantly lower in the early treatment group (0.76, 0.52-1.11; p=0.177). At 5 years, median FAMS-TOI scores were 125 in both groups. No significant differences in other disability related outcomes were recorded. Frequency and severity of adverse events remained within the established safety and tolerability profile of interferon beta-1b. INTERPRETATION: Effects on the rate of conversion to CDMS and the favourable long-term safety and tolerability profile support early initiation of treatment with interferon beta-1b, although a delay in treatment by up to 2 years did not affect long-term disability outcomes. FUNDING: Bayer Schering Pharma.

Effect of early versus delayed interferon beta-1b treatment on disability after a first clinical event suggestive of multiple sclerosis: a 3-year follow-up analysis of the BENEFIT study.

BACKGROUND: Several controlled studies provide evidence that treatment with interferon beta in patients with a first event suggestive of multiple sclerosis (MS) delays conversion to clinically definite MS (CDMS). Our aim was to determine whether early initiation of treatment with interferon beta prevents development of confirmed disability in MS. METHODS: In the initial placebo-controlled phase of the double-blinded BENEFIT study, patients with a first event suggestive of MS and a minimum of two clinically silent lesions in MRI were randomised to receive either interferon beta-1b 250 microg (n=292) or placebo (n=176) subcutaneously every other day for 2 years, or until diagnosis of CDMS. Patients were then eligible to enter the follow-up phase with open-label interferon beta-1b. In the current prospectively planned analysis 3 years after randomisation, the effects of early interferon beta-1b treatment were compared with those of delayed treatment initiated after diagnosis of CDMS or after 2 years on the study. The primary outcomes of this ITT analysis were time to diagnosis of CDMS, time to confirmed expanded disability status scale (EDSS) progression, and score on a patient-reported functional assessment scale (FAMS-TOI). This trial is registered with ClinicalTrials.gov, number NCT00185211. FINDINGS: Of the 468 patients originally randomised, 418 (89%) entered the follow-up phase; 392 (84%) completed 3 years' post-randomisation follow-up. After 3 years, 99 (37%) patients in the early group developed CDMS compared with 85 (51%) patients in the delayed treatment group. Early treatment reduced the risk of CDMS by 41% (hazard ratio 0.59, 95% CI 0.44-0.80; p=0.0011; absolute risk reduction 14%) compared with delayed treatment. Over 3 years, 42 (16%) patients in the early group and 40 (24%) in the delayed group had confirmed EDSS progression; early treatment reduced the risk for progression of disability by 40% compared with delayed treatment (0.60, 0.39-0.92; p=0.022; absolute risk reduction 8%). The FAMS-TOI score was high and stable in both groups over the 3-year period (p=0.31). INTERPRETATION: Our data suggest that early initiation of treatment with interferon beta-1b prevents the development of confirmed disability, supporting its use after the first manifestation of relapsing-remitting MS.

Dimension reduction for physiological variables using graphical modeling.

In intensive care, physiological variables of the critically ill are measured and recorded in short time intervals. The proper extraction and interpretation of the essential information contained in this flood of data can hardly be done by experience alone. Typically, decision making in intensive care is based on only a few selected variables. Alternatively, for a dimension reduction statistical latent variable techniques like principal component analysis or factor analysis can be applied. However, the interpretation of latent components extracted by these methods may be difficult. A more refined analysis is needed to provide suitable bedside decision support. Graphical models based on partial correlations provide information on the relationships among physiological variables that is helpful for variable selection and for identifying interpretable latent components. In a comparative study we investigate how much of the variability of the observed multivariate physiological time series can be explained by variable selection, by standard principal component analysis and by extracting latent compo-nents from groups of variables identified in a graphical model.