A DYNAMIC DECISION ANALYSIS PROCESS FOR EVALUATING PANDEMIC INFLUENZA INTERVENTION STRATEGIES
Author(s)
Fasth T1, Talantsev A2, Brouwers L1, Larsson A2
1Public Health Agency of Sweden, Solna, Sweden, 2Stockholm University, Kista, Sweden
OBJECTIVES: Public health authorities (PHA) experience challenges during outbreaks of pandemic influenza. There are (1) a high public attention and pressure for fast and effective actions, (2) uncertainty regarding development and potential impact of the pandemic, (3) complexity of the decisions on selecting interventions – this involves considering multitude of criteria and interests of stakeholders. Therefore we suggest a dynamic decision analysis process, which supports comprehensiveness, transparency, and consistency in decisions regarding interventions. METHODS: The process design was informed by scenario analysis, multi-criteria decision analysis methodology, and adapted to the Swedish context. The process was then validated in a pilot decision conference (DC) conducted at the PHA of Sweden. The process consists of elicitation and evaluation. Elicitation comprises developing scenarios, intervention strategies, and a multi-criteria model (MCM) against which the strategies are evaluated. In the evaluation process the strategies are scored, criteria weighted, and the results analyzed. RESULTS: In the pilot DC we considered a mild pandemic scenario, similar to the influenza pandemic of 2009. Three intervention strategies were evaluated, based on four interventions; vaccination (VAC), antiviral treatment (AVT), antiviral prophylaxis (AVP), school closure (SC). Strategy (1) included AVT to everyone with influenza-like illness (ILI), VAC to risk-groups, (2) AVT and AVP to risk-groups, SC, and VAC to everyone, and (3) AVT to everyone with ILI, AVP to risk-groups, and VAC to everyone. Guided by the process the seven participants outlined the MCM through a post-it session, evaluated the strategies against the criteria, and expressed criteria weights. Finally the results were aggregated and visualized. The analysis showed that strategy (1) was the most preferred. CONCLUSIONS: The process supports a dynamic, comprehensive, and transparent decision-making. An integrated decision-support toolset will be developed in forthcoming research along with contextualising and validating the process with Sweden related stakeholders.
Conference/Value in Health Info
2017-11, ISPOR Europe 2017, Glasgow, Scotland
Value in Health, Vol. 20, No. 9 (October 2017)
Code
PIN120
Topic
Health Technology Assessment
Topic Subcategory
Decision & Deliberative Processes
Disease
Infectious Disease (non-vaccine)