Objective Implementation of machine learning (ML) may be limited by patients’ right to “meaningful information about the logic involved” when ML influences healthcare decisions. Given the complexity of healthcare decisions, it is likely that ML outputs will need to be understood and trusted by physicians, and then explained to patients. We therefore investigated the association between physician understanding of ML outputs, their ability to explain these to patients, and their willingness to trust the ML outputs, using various ML explainability methods. Materials and Methods We designed a survey for physicians with a diagnostic dilemma that could be resolved by an ML risk calculator. Physicians were asked to rate their understanding, explainability, and trust in response to 3 different ML outputs. One ML output had no explanation of its logic (the control) and 2 ML outputs used different model-agnostic explainability methods. The relationships among understanding, explainability, and trust were assessed using Cochran-Mantel-Haenszel tests of association. Results The survey was sent to 1315 physicians, and 170 (13%) provided completed surveys. There were significant associations between physician understanding and explainability (P < .001), between physician understanding and trust (P < .001), and between explainability and trust (P < .001). ML outputs that used model-agnostic explainability methods were preferred by 88% of physicians when compared with the control condition; however, no particular ML explainability method had a greater influence on intended physician behavior. Conclusions Physician understanding, explainability, and trust in ML risk calculators are related. Physicians preferred ML outputs accompanied by model-agnostic explanations but the explainability method did not alter intended physician behavior.

No abstract

Background: In ischemic stroke patients, studies have suggested that clinical outcomes following endovascular thrombectomy are worse after general anesthesia (GA) compared with conscious sedation (CS). Most data are from observational trials, which are prone to measure and unmeasure confounding. We performed a systematic review and meta-analysis of thrombectomy trials where patients were randomized to GA or CS, and compared efficacy and safety outcomes. Methods: The Medline, Embase, and Cochrane databases were searched for randomized controlled trials comparing GA to CS in endovascular thrombectomy. Efficacy outcomes included successful recanalization (Thrombolysis in Cerebral Infarction score of 2b to 3), and good functional outcome, defined as a modified Rankin Scale score of 0 to 2 at 3 months. Safety outcomes included intracerebral hemorrhage and 3-month mortality. Results: Four studies were identified and included in the random effects meta-analysis. Patients treated with GA achieved a higher proportion of successful recanalization (odds ratio [OR]: 2.14, 95% confidence interval [CI]: 1.26-3.62; P=0.005) and good functional outcome (OR: 1.71, 95% CI: 1.13-2.59; P=0.01). For every 7.9 patients receiving GA, one more achieved good functional outcome compared with those receiving CS. There were no significant differences in intracerebral hemorrhage (OR: 0.61, 95% CI: 0.20-1.85; P=0.38) or 3-month mortality (OR: 0.62, 95% CI: 0.33-1.17; P=0.14) between GA and CS patients. Conclusions: In centers with high quality, specialized neuroanesthesia care, GA treated thrombectomy patients had superior recanalization rates and better functional outcome at 3 months than patients receiving CS.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.