The objective of our research is to evaluate and compare the performance of ChatGPT, Google Bard, and medical students in performing START triage during mass casualty situations. We conducted a cross-sectional analysis to compare ChatGPT, Google Bard, and medical students in mass casualty incident (MCI) triage using the Simple Triage And Rapid Treatment (START) method. A validated questionnaire with 15 diverse MCI scenarios was used to assess triage accuracy and content analysis in four categories: "Walking wounded," "Respiration," "Perfusion," and "Mental Status." Statistical analysis compared the results. Google Bard demonstrated a notably higher accuracy of 60%, while ChatGPT achieved an accuracy of 26.67% (p = 0.002). Comparatively, medical students performed at an accuracy rate of 64.3% in a previous study. However, there was no significant difference observed between Google Bard and medical students (p = 0.211). Qualitative content analysis of 'walking-wounded', 'respiration', 'perfusion', and 'mental status' indicated that Google Bard outperformed ChatGPT. Google Bard was found to be superior to ChatGPT in correctly performing mass casualty incident triage. Google Bard achieved an accuracy of 60%, while chatGPT only achieved an accuracy of 26.67%. This difference was statistically significant (p = 0.002).

Background Economic losses from extreme weather events (EWEs) are rising and are increasingly reported in settings with limited historical exposure to such hazards. Proactive disaster risk management is widely promoted to reduce impacts and improve value for investment, yet the economic evidence is dispersed across hazards, strategy types, and evaluation methods. As a result, decision-makers often struggle for a clear basis for comparing options across contexts. In addition, economic results are frequently reported without a structured synthesis of the barriers and facilitators that influence adoption and implementation. This review aims to synthesise and compare global evidence on proactive disaster risk management strategies for extreme weather events, alongside the barriers and facilitators associated with their adoption. Methods This protocol is registered in PROSPERO (CRD420251023343) and will be reported in accordance with PRISMA guidance. Peer-reviewed primary studies published from 2015 onwards will be eligible if they evaluate proactive strategies addressing extreme weather events and report at least one economic outcome. Eligible hazards must be extreme weather events. Interventions will be categorised as technology-based, infrastructure-based, community-based, nature-based, risk-financing, or hybrid strategies. Searches will be conducted in six electronic databases, with screening and data extraction managed in Covidence. Two reviewers will complete screening, extraction, and quality appraisal, with disagreements resolved through discussion or adjudication by a third reviewer. Data items will include study setting (Global North or South and World Bank income group), hazard type, intervention characteristics and scale, economic methods, and economic results. Economic outcomes will be collected. The study quality will be assessed using the Joanna Briggs Institute tools. Given anticipated heterogeneity, findings will be narratively synthesised alongside subgroup comparisons and a framework-based thematic synthesis of adoption barriers and facilitators.

Growing extreme weather risks increase the need for economically justified investment in climate adaptation and disaster risk reduction, yet evidence remains fragmented across strategies, hazards, and settings. This systematic review synthesised economic effectiveness and stated-preference evidence for proactive management of extreme weather risks. Six electronic search platforms covering eight database sources were searched for English-language peer-reviewed studies from January 2015 to May 2025. Heterogeneity required narrative synthesis rather than meta-analysis. Fifty-three unique studies were included: 32 contributed economic effectiveness evidence, 23 contributed stated-preference evidence, and two contributed to both evidence strands. Among the 32 economic effectiveness studies, 15 reported favourable conclusions, 15 mixed or conditional conclusions, one an unfavourable conclusion, and one an unclear conclusion. All 23 stated-preference studies reported positive willingness to pay, willingness to purchase or related stated demand, indicating expressed demand or perceived value rather than demonstrated economic effectiveness. Economic conclusions frequently depended on site, design, scale, time horizon, discounting, valuation boundaries, and modelling assumptions. Economic effectiveness evidence was absent for community-based and risk-financing strategies and concentrated on flood risk, Asia, and North America. Economic and financial factors were the most common barriers, whereas social and cultural factors were the most common facilitators. Investment decisions should consider evidence type, robustness, assumptions, and context.