Decision making under uncertainty, therapeutic inertia, and physicians' risk preferences in the management of multiple sclerosis (DIScUTIR MS).

BACKGROUND: The management of multiple sclerosis (MS) is rapidly changing by the introduction of new and more effective disease-modifying agents. The importance of risk stratification was confirmed by results on disease progression predicted by different risk score systems. Despite these advances, we know very little about medical decisions under uncertainty in the management of MS. The goal of this study is to i) identify whether overconfidence, tolerance to risk/uncertainty, herding influence medical decisions, and ii) to evaluate the frequency of therapeutic inertia (defined as lack of treatment initiation or intensification in patients not at goals of care) and its predisposing factors in the management of MS. METHODS/DESIGN: This is a prospective study comprising a combination of case-vignettes and surveys and experiments from Neuroeconomics/behavioral economics to identify cognitive distortions associated with medical decisions and therapeutic inertia. Participants include MS fellows and MS experts from across Spain. Each participant will receive an individual link using Qualtrics platform(©) that includes 20 case-vignettes, 3 surveys, and 4 behavioral experiments. The total time for completing the study is approximately 30-35 min. Case vignettes were selected to be representative of common clinical encounters in MS practice. Surveys and experiments include standardized test to measure overconfidence, aversion to risk and ambiguity, herding (following colleague's suggestions even when not supported by the evidence), physicians' reactions to uncertainty, and questions from the Socio-Economic Panel Study (SOEP) related to risk preferences in different domains. By applying three different MS score criteria (modified Rio, EMA, Prosperini's scheme) we take into account physicians' differences in escalating therapy when evaluating medical decisions across case-vignettes. CONCLUSIONS: The present study applies an innovative approach by combining tools to assess medical decisions with experiments from Neuroeconomics that applies to common scenarios in MS care. Our results will help advance the field by providing a better understanding on the influence of cognitive factors (e.g., overconfidence, aversion to risk and uncertainty, herding) on medical decisions and therapeutic inertia in the management of MS which could lead to better outcomes.

Atrial fibrillation in ischemic stroke: predicting response to thrombolysis and clinical outcomes.

BACKGROUND AND PURPOSE: Atrial fibrillation (AF) increases the risk of stroke and is associated with poor stroke outcomes. Limited tools are available to evaluate clinical outcomes and response to thrombolysis in stroke patients with AF. METHODS: We applied the iScore (http://www.sorcan.ca/iscore), a validated risk score, to consecutive patients with an acute ischemic stroke admitted to stroke centers in the Registry of the Canadian Stroke Network. The main outcome considered was a favorable outcome (defined as a modified Rankin scale 0-2) at discharge after thrombolysis. Secondary outcomes included intracerebral hemorrhage, death at 30 days, and at 1 year stratified by terciles of the iScore. RESULTS: Among 12 686 patients with an acute ischemic stroke, 2185 (17.2%) had AF. Overall, AF patients had higher risk of death at 30 days (22.3% versus 10.2%; P<0.0001), 1 year (37.1% versus 19.5%; P<0.0001) and death or disability at discharge (69.7% versus 54.7%; P<0.0001) compared with non-AF patients. After adjustment, thrombolysis was associated with a favorable outcome for patients without AF (relative risk, 1.18; 95% CI, 1.10-1.27), but no benefit was observed for patients with AF (relative risk, 0.91; 95% CI, 0.71-1.17). There was a modestly increased risk of intracranial hemorrhage (any type) (16.5% versus 11.6%; relative risk, 1.42; 95% CI, 1.05-1.91) after thrombolysis among AF compared with non-AF patients. In the logistic regression analysis, there was an interaction between tPA and iScore for a favorable outcome (P-value interaction <0.001). The interaction also was significant (P<0.0012) among patients without AF, but did not reach significance (P=0.17) in patients with AF. CONCLUSIONS: Stroke patients with AF have higher mortality, greater risk of intracerebral hemorrhage, and a similar response trend to thrombolysis compared with non-AF patients.

Real-World Treatment Sequencing, Toxicities, Health Utilities, and Survival Outcomes in Patients with Advanced ALK-Rearranged Non-Small-Cell Lung Cancer.

OBJECTIVES: This real-world analysis describes treatment patterns, sequencing and clinical effectiveness, toxicities, and health utility outcomes in advanced-stage, incurable ALK-positive NSCLC patients across five different ALK-TKIs. MATERIALS AND METHODS: Clinicodemographic, treatment, and toxicity data were collected retrospectively in patients with advanced-stage ALK-positive NSCLC at Princess Margaret Cancer Centre. Patient-reported symptoms, toxicities, and health utilities were collected prospectively. RESULTS: Of 148 ALK-positive NSCLC patients seen July 2009-May 2021, median age was 58.9 years; 84 (57%) were female; 112 (76%) never-smokers; 54 (47%) Asian and 40 (35%) white; 139 (94%) received at least one ALK-TKI: crizotinib (n = 74; 54%) and alectinib (n = 61; 44%) were administered mainly as first-line ALK-TKI, ceritinib, brigatinib and lorlatinib were administered primarily after previous ALK-TKI failure. Median overall survival (OS) was 54.0 months; 31 (21%) patients died within two years of advanced-stage diagnosis. Treatment modifications were observed in 35 (47%) patients with crizotinib, 19 (61%) with ceritinib, 41 (39%) with alectinib, 9 (41%) with brigatinib and 8 (30%) with lorlatinib. Prevalence of dose modifications and self-reported toxicities were higher with early versus later generation ALK-TKIs (P<.05). The presence of early treatment modification was not negatively associated with progression-free survival (PFS) and OS analyses. CONCLUSION: Serial ALK-TKI sequencing approaches are viable therapeutic options that can extend quality of life and quantity-of-life, though a fifth of patients died within two years. No best single sequencing approach could be determined. Clinically relevant toxicities occurred across all ALK-TKIs. Treatment modifications due to toxicity may not necessarily compromise outcomes, allowing multiple approaches to deal with ALK-TKI toxicities.

Predicting clinical outcomes and response to thrombolysis in acute stroke patients with diabetes.

OBJECTIVE: Few tools are available to evaluate clinical outcomes and response to thrombolysis (tPA) in stroke patients with diabetes. We explored how the iScore (www.sorcan.ca/iscore), a validated risk score, predicts clinical outcomes in stroke patients with and without diabetes. RESEARCH DESIGN AND METHODS: We applied the iScore to stroke patients presenting to stroke centers participating in the Registry of the Canadian Stroke Network. Main outcomes included favorable outcome, defined as a modified Rankin scale (mRS) 0-2 at discharge, and intracerebral hemorrhage (ICH) after tPA. RESULTS: Among 12,686 patients with an acute ischemic stroke, 3,228 (25.5%) had diabetes. Among patients receiving tPA (n = 1,689), those with diabetes had a lower rate of a favorable outcome compared with their counterparts (24.3 vs. 31.1%; RR 0.90 [95% CI 0.82-0.98]). The risk of ICH was not significantly different in patients with or without diabetes (for any type 12.6 vs. 12.5%, RR 1.01 [0.72-1.40]; for symptomatic ICH 7.5 vs. 6.8%, RR 1.11 [0.70-1.72]). The regression analysis revealed a decline in the probability of a favorable outcome after tPA with increments in the iScore (P value for iScore × tPA interaction <0.001). There was no difference in the response to tPA predicted by the iScore between stroke patients with and without diabetes (P value = 0.07). CONCLUSIONS: Stroke patients with diabetes have poorer outcomes compared with patients without diabetes, which is not explained by ICH. The iScore similarly predicts response to tPA between stroke patients with and without diabetes.