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

It is apparent from Table 1 that there is a greater proportion of cases occurring within households in both Madang models compared to that of the Albany model (see Table 1 ). This arises from the average Madang household size being larger than in Albany. The number of cases occurring in Madang due to household transmission is ~8,000 compared to ~3,500 in Albany; note that Madang has a total population of ~35,000 compared to ~30,000 in Albany. School closure scenario The school closure intervention in the Madang model is found to be approximately five times less effective compared to that of the Albany model. Simulation results suggest a 9.4% reduction (46.6% to 42.2%) in the illness attack rate can be achieved in Madang due to school closure whereas a 48.6% reduction (31.7% to 16 .3%) can be achieved in Albany (see Table 1 ). For the no-hub Madang-nnh model, school closure is ~3.5 times less effective compared to the Albany model. Table 2 (middle panel) indicates an illness attack rate of ~15,000 in Madang compared to ~4,800 in Albany. Much higher case numbers occur in households, workplaces and the wider community (including neighbourhood hubs) than in Albany, highlighting the poorer efficacy of this intervention strategy in a PNG setting. The Madang-nnh model also has higher household and workplace transmission than in Albany. Note that while there is a significantly higher proportion of under 17 year old children in Madang (40%) compared to Albany (28%), our field survey indicates that 50% of school age children in Madang do not attend school, hence school closure is less effective in disrupting the transmission chain (household 1 / school / household 2) compared to Australia. This feature also is present in the rigorous social distancing intervention.