A novel framework for evaluating the impact of individual decision-making on public health outcomes and its potential application to study antiviral treatment collection during an influenza pandemic

摘要

The importance of accounting for social and behavioural processes when studying public health emergencies has been well-recognised. For infectious disease outbreaks in particular, several methods of incorporating individual behaviour have been put forward, but very few are based on established psychological frameworks. In this paper, we develop a decision framework based on the COM-B model of behaviour change to investigate the impact of individual decision-making on public health outcomes. We demonstrate the application of our decision framework in a proof-of-concept case study based on the 2009 A(H1N1) influenza pandemic in the UK. The National Pandemic Flu Service (NPFS) was set up in England during the pandemic as a means to provide antiviral (AV) treatment to clinically ill patients with influenza-like illness, via telephone calls or internet screening, thereby averting the need to see a doctor. The evaluated patients based on a clinical algorithm and authorised AV drugs for collection via community collection points. We applied our behavioural framework to evaluate the influence of human behaviour on AV collection rates, and subsequently to identify interventions that could help improve AV collection rates. Our model was validated against empirically collected pandemic data from 2009 in the UK. We also performed a sensitivity analysis to identify potentially effective interventions by varying model parameters. Using our behavioural framework in a proof-of-concept case study, we found that interventions geared towards increasing people’s ‘Capability’ and ‘Opportunity’ are likely to result in increased AV collection, potentially resulting in fewer influenza-related hospitalisations and deaths. We note that important behavioural data from public health emergencies are largely scarce. Insights obtained from models such as ours can, not only be very useful in designing healthcare interventions, but also inform future data collection.