Release immediately or sequentially? Strategies for allocating scarce therapeutic resources during disease outbreaks
研究了疫情爆发时，从中央库存向地方分配稀缺医疗资源（如抗病毒药物、呼吸机）的策略，通过数学模型比较立即释放与分批释放的效果，发现最优策略取决于资源效益随时间的变化和不同人群的受益差异。
During infectious disease outbreaks, public health authorities face the challenge of fairly distributing limited medical therapeutics, such as antiviral drugs and respirators, to save lives. Such resources are not always widely available at the start of the outbreak when the progression of the outbreak is uncertain. Then, resources may be allocated from a centralized inventory — such as a federal stockpile — to regional authorities for further distribution on a periodic basis. Using mathematical optimization models for spatiotemporal resource allocations, we address a fundamental question of whether such stockpiles should be fully released at the outset or gradually distributed throughout the threat. Immediately releasing the therapeutics risks its consumption by non-symptomatic populations who are concerned about the disease; this is particularly so in the absence of a diagnostic test that distinguishes such individuals from those that are genuinely sick. We find that if the benefit a resource provides to an individual, whether genuinely sick or not, decreases as the outbreak progresses, then an immediate release is optimal. However, if some population groups are expected to benefit more than others, then sequential releases that conserve resources and match peak demand are advisable. Sequential release policies can pose challenging ethical and political dilemmas for policymakers under public pressure to provide immediate relief. To this end, using simulated scenarios sampled from six historic pandemics including the COVID-19 pandemic, we provide computational experiments for the state of Texas, with over 25 million people, that empirically demonstrate a significant number of lives can be saved by following optimal release policies.