To assess the comparative effectiveness of educational interventions in neurological disease for healthcare workers and students.

Systematic review.

Medline, Embase and Cochrane through to 1 June 2025.

Studies evaluating neurological disease educational interventions with a comparator group (observational cohort/randomised controlled trial (RCT)) were included.

A Preferred Reporting Items for Systematic Reviews and Meta-Analyses-compliant systematic review was conducted (PROSPERO: CRD42023461838). Knowledge acquisition and educational methodologies were collected from each study. Study outcomes were classified using the Kirkpatrick and Kirkpatrick four-level model (learner reaction, knowledge acquisition, behavioural change, clinical outcome).¹ Risk of bias was assessed using the Newcastle-Ottawa scale for non-randomised studies and the Cochrane Risk of Bias tool for RCTs.^{2 3}

A total of 67 studies involving 4728 participants were included. Of these, 36 were RCTs, and 31 were observational studies. Virtual interventions were the most common (67.2%, n=45 studies), primarily targeting either medical students (46.3%, n=31 studies) or specialists (40.3%, n=27 studies). Overall, 70.1% (n=47) of studies demonstrated outcomes in favour of the intervention. However, few studies used K&K level 3/4 outcomes, with two studies evaluating behaviour change (level 3) and three assessing clinical outcomes (level 4 combined with other levels). No study exclusively assessed level 4 outcomes. Meta-analysis of 22 RCTs with calculable standardised mean differences (SMDs) (n=1748) showed a significant benefit of interventions (SMD 0.75, 95% CI 0.22 to 1.27, p=0.0056).

This review highlights a growing body of research particularly focusing on virtual techniques, specialist audiences and treatment-oriented content. Few studies assessed changes in practice or patient care. Non-specialists remain underrepresented. Future studies should prioritise assessing the clinical impact of educational interventions within non-specialist audiences.