Optimal Decentralization of Early Infant Diagnosis of HIV in Resource-Limited Settings

Early infant diagnosis (EID) programs in many resource-limited settings are aimed at diagnosing infants born to HIV positive mothers. Due to the complexity of the diagnostic technology, EID programs are often highly centralized with few laboratories testing blood samples from a large network of health facilities. This leads to long diagnostic delays and consequent failure of patients to collect results in a timely manner. Several point-of-care (POC) devices that provide rapid diagnosis within the health facilities are being developed to mitigate these drawbacks of centralized EID networks. We study the decision of which facilities should receive the POC device (the placement plan) using the EID program in Mozambique as a case-study. We argue that the choice of an appropriate plan is critical to maximizing the public health impact of POC devices in the presence of tight budget constraints. To formalize this argument, we develop a detailed simulation model to evaluate the impact of a placement plan. It comprises two parts: an operational model that quantifies the impact of a POC placement plan on the diagnostic delay and a behavioral part that quantifies the impact of diagnostic delay on the likelihood of result collection by infants’ caregivers. We also develop an approximate version of these operational and patient behavior dynamics and embed them in an optimization model to generate candidate POC placement plans. We find that the optimization based plan can result in up to 30% more patients collecting their results compared to rules of thumb that have practical appeal. Finally, we show that the effectiveness of POC devices is much higher than other operational improvements to the EID network such as increased laboratory capacity, reduced transportation delay, and more regularized transport