Assignment optimization of pandemic influenza antiviral drugs in Urban pharmacies

Abstract

© 2018, Springer-Verlag GmbH Germany, part of Springer Nature. Antiviral drugs have benefited public health officers to elucidate outbreak risks by controlling influenza pandemics efficacious, especially effective in the early stage of epidemic outbreaks. To limit explosive strain on hospitals, commercial pharmacies have joined, as antiviral drug-dispensing partners, in governments’ pandemic response plans. Existing researches focus on site selection by optimizing the single objective of access to the target population. However, there are substantial inevitable but essential social factors (such as social unbalance, spatial unbalance and resource unbalance) needed to consider to benefit the society best. In this paper, we propose a network-perspective optimization model across multiple social scales (e.g, access, social unbalance, spatial unbalance and resource unbalance) to assign antiviral drugs to the urban dispensing pharmacies. In the network-based frame, we transfer these considerations to the constraints of group, edge, and node. The constrained optimization model is studied and solved using methods of willingness-to-travel model, L12 norm and network lasso, corresponding to each considerations. Taking Shanghai in a cohort of 11 million individuals as an example, we have shown the flexibility of the proposed multi-objective model, comparing with the traditional methods. For example, we found that there are 29 pharmacies needed with covering 81% districts by tradition single-objective method. In the contrast, only 12 pharmacies are needed with similar access ability but can still cover 75% districts. Or more pharmacies are assigned with covering 87% districts. This research can supply an initial exploration of pharmacy-based distribution of antiviral drugs for the studying construction of strategic national stockpile in some countries.