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

The interventions used in the present study are based on social distancing. Surveys have shown that a higher proportion of low-income countries have social distancing interventions planned in the case of a pandemic, compared to high income European 17 countries due to their feasibility and cost-effectiveness. [3, 4, 6, 8, 47, 48] Pharmaceutical-based interventions were not considered in this study due to cost and delivery issues found with low-income countries. [4, 8] Consideration may need to be given to some form of antiviral agent use, possibly for case treatment, given the poor performance of purely social distancing interventions in developing countries as suggested by this study. The advantage of antiviral use is that it may be targeted at reducing within-household transmission, a location of transmission otherwise not impacted by social distancing interventions. The present study shows that larger numbers of transmissions occur within households due to the larger household sizes in countries such as PNG. Furthermore, low-income countries generally have population health profiles not seen in developed countries, including the prevalence of diseases such as malaria, tuberculosis and HIV. [49] Antiviral treatment of influenza cases co-infected with such diseases may be an appropriate intervention strategy given the poorer health outcomes which co-morbidity may impose. [12] The provision and use of antiviral agents will have challenges not faced by industrialised countries, due to the cost of providing an antiviral stockpile and subsequent distribution of antivirals in countries which have health systems already stretched to capacity. [15] The importance of suitable pandemic influenza preparedness plans for low-income countries is recognized by the WHO [50, 51]. Disease transmission models will play a role in guiding policy makers in determining the effectiveness of possible pandemic countermeasures. History tells us that low-income countries are likely to be more affected, yet almost all models deal with mitigation strategies for developed nations with good health infrastructure and low prevalence of endemic diseases. These countries have poorer access to antiviral drug and vaccine-based interventions and higher extant disease burdens, compared to developed countries. For the reasons, development of appropriate intervention strategies focused on countries such as PNG needs to occur. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 (top table) , Madang (centre table) and Madang-nnh (bottom table) models are presented. One of the key differences between the Albany and the Madang models is that ~50% of school age children do not attend school in Madang compared to Albany. There are ~8900 school age children in Madang among which ~4300 do not attend school. Another key difference is ~5700 individuals (including unemployed adults and the ~50% of children not attending school) in Madang are assumed to mix in neighbourhood hubs. If no neighbourhood hubs are included, then the individuals who are in neighbourhood hubs in the Madang model in daytime would mix only in their households, as in the Madangnnh model. The baseline assumption in the main results is that individuals in neighbourhood hubs mix with 10 other individuals. If this value is assumed to be 20 or 30 in