Lancet OncolJClub
Assessment of comprehensiveness of national cancer control plans using a multiagent artificial intelligence model: a global analysis.
The Lancet Oncology · · Journal Article
Kumar, Reddy + more
Abstract ↗AI summary
The abstract is read at the publisher; the summary is JClub's.
Median national cancer control plan comprehensiveness across 50 countries is 2·78 (IQR 2·25–3·09), with Europe and lower-middle-income countries highest and low-income lowest.
- Why it matters: Expert-led reviews of national cancer control plans are resource intensive and limit cross-country comparability, leaving uncertainty about how comprehensiveness varies globally.
- What they did: Using PRISM, a validated multiagent AI system, the team scored 50 most recent NCCPs with a 12-element, 76-sub-element framework, validated against masked dual-expert ratings for six NCCPs.
- The result: Comprehensiveness is highest for implementation (4·17; 3·54–4·38) and governance and organisation (3·67; 3·35–3·98) but lowest for financing (1·88; 1·00–2·69) and outcomes (1·83; 1·33–2·33), with financing showing the largest gaps.
The findingWhy it mattersWhat they didThe result