Pandemic influenza in Papua New Guinea: a modelling study comparison with pandemic spread in a developed country

Milne, George J; Baskaran, Pravin; Halder, Nilimesh; Karl, Stephan; Kelso, Joel

While care was taken to obtain detailed data used to populate the Madang and Albany models, government data was found to be more limited in PNG. While a field survey was conducted in Madang to fill in missing data and make the models as directly comparable as possible, there is still scope for conducting further data collection. As with all modelling studies, the accuracy of results is limited by the quality of data used to construct the model. In the case of influenza epidemic modelling, outcomes are dependent on assumptions about the number, duration and physical closeness of contacts between people, and how these attributes of contact affect transmission probability. These data are very difficult to estimate, with the best data source available being [46] . In the absence of high-quality reliable data, we have made plausible assumptions regarding interpersonal contact, and, in the Albany case, these are reflective of the data in [46] .