Credit: peterschreiber.media/Getty Images
A randomized clinical trial across 77 Chinese hospitals has shown that the use of an AI-based clinical decision support tool significantly improved long-term outcomes and quality of care for stroke patients. Published today in The BMJ, results show a 27% reduction in new vascular events after the initial stroke after 12 months.
“The clinical decision support system (CDSS) is an innovative and promising approach for improving stroke outcomes and healthcare services,” write the authors of the study, led by Zixiao Li, MD, PhD, associate professor at Beijing Tiantan Hospital. “However, the majority of AI applications for stroke healthcare have not been rigorously evaluated through randomized controlled trials. As a result, the use of CDSS in treating cerebrovascular disease is currently limited.”
Li and colleagues developed a stroke CDSS system that integrates AI-assisted imaging analysis, classification of stroke causes, and evidence-based treatment recommendations to improve the quality of care and clinical outcomes in routine clinical practice.
The GOLDEN BRIDGE II clinical trial involved more than 21,000 patients with acute ischemic stroke across 77 hospitals in China, which were randomized to give patients either the standard medical care or an intervention with the support of the stroke CDSS. All patients were admitted into one of the hospitals within seven days of symptom onset, and physicians assigned to the intervention group were trained on using the AI support tool.
The primary outcome of the trial was the proportion of patients who suffered a new cardiovascular event after three, six, and twelve months following the initial stroke. These included another ischemic stroke, hemorrhagic stroke, myocardial infarction, and vascular death.
Results showed a 26% reduction in new cardiovascular events after three months (2.9% of patients in the intervention group compared to 3.9% in the control group) and a 27% reduction after 12 months (4% of patients in the intervention group versus 5.5% in the control group). No significant differences were found between both groups at any time point when it came to bleeding, disability, and all-cause mortality.
In addition, the stroke CDSS significantly improved composite stroke performance measures as well as several specific stroke indicators including dual antiplatelet treatment, anticoagulation for atrial fibrillation, dysphagia screening, and deep venous thrombosis prophylaxis. Overall, the AI-based support tool was found to be easy to use and integrate into hospital information systems.
Although hospital region and grade, patient age, medication history, and lifestyle were taken into account, the clinical trial randomized hospitals rather than individual patients, meaning some differences in care across institutions might impact the findings. Future studies will look in more detail at the cost-effectiveness of employing the stroke CDSS to evaluate any potential financial advantages of implementing the system across hospitals.
The study authors concluded: “The stroke CDSS offers a promising approach to providing high quality care for patients with acute ischemic stroke admitted to hospital, particularly for resource constrained regions with a heavy burden of cerebrovascular diseases like China.”