Using systems science to optimize the impact of point-of-care viral load testing for pediatric HIV management

摘要 HIV感染儿童的病毒载量（VL）抑制比成人差，发病率高， mortality.定期进行儿科VL监测对于指导儿科治疗和临床护理至关重要。VL测试是 通常建议治疗开始后第一年每6个月一次，此后每年一次，但 标准治疗（SOC）VL系统的周转时间很慢（中位数为21天）。床旁（POC）技术 已经彻底改变了HIV诊断和治疗的开始，POC VL检测是HIV的下一个前沿领域 治疗监测。目前正在评估儿童POC VL检测的临床有效性， 正在肯尼亚进行的随机试验（R34 PI：Patel，Abuogi）;然而，临床研究之间存在较大差距 POC技术的有效性和实际应用。使用系统工程优化POC VL测试 这种方法将有助于实现POC VL监测投资的全面影响。在目前的提案中，我们采取 一个系统工程的方法，并借用不同领域的方法，如制造业，心理学， 和软件开发，以实现以下目标：目标1：以确定有限的最佳位置 中心辐射式与平台共享模式中的POC VL机器，以平衡预算影响并最大限度地减少 周转时间我们将创建一个排队模型--用于工业工程中对等待时间进行建模--以 确定肯尼亚POC机器的最佳位置。我们将模拟减少周转时间， 与在选定的“中心”设施（具有POC机器的站点）中放置POC机器相关的等待时间 以及“辐条”（将样本发送到中心的站点）设施与在设施组之间共享的平台。在不同 在预算水平上，我们将确定POC机器的最佳数量和位置。目标2：确定 政策制定者对目标1中模型工具可用性的看法。我们将目标1的模型转换为 用户友好的、基于Excel的模型，供决策者用于决策。我们将进行可用性 访谈（涵盖可学习性，效率，记忆力，错误恢复和满意度）约20 卫生管理人员和政策制定者对该工具的了解。这种新颖的应用不同的方法借鉴了 工业工程，软件工程，心理学和质量改进提出了一个创新的 提高儿童POC VL检测可扩展性的方法。

ABSTRACT HIV-infected children have poorer viral load (VL) suppression than adults and experience high morbidity and mortality. Regular pediatric VL monitoring is critical to direct pediatric treatment and clinical care. VL testing is generally recommended every 6 months for the first year after treatment initiation and annually thereafter, but turnaround time of standard of care (SOC) VL systems is slow (median 21 days). Point of care (POC) technology has revolutionized HIV diagnosis and treatment initiation, and POC VL testing is the next frontier for HIV treatment monitoring. Clinical effectiveness of POC VL testing for children is currently being assessed in an ongoing randomized trial in Kenya (R34 PIs: Patel, Abuogi); however, large gaps exist between clinical effectiveness and real-world use for POC technology. Optimizing POC VL testing using a systems engineering approach will help realize the full impact of investments in POC VL monitoring. In the current proposal, we take a systems engineering approach and borrow methods from diverse fields, such as manufacturing, psychology, and software development to achieve the following aims: Aim 1: To determine the optimal placement of limited POC VL machines within a hub-and-spoke vs. platform sharing models, to balance budget impact and minimize turnaround time. We will create a queuing model – used in industrial engineering to model waiting times – to identify optimal placement of POC machines in Kenya. We will model the reduction in turnaround time and waiting time associated with placement of POC machines in select “hub” facilities (sites with a POC machine) and “spoke” (sites that send samples to a hub) facilities vs. platform sharing amongst sets of facilities. At different budget levels, we will identify the optimal number and placement of POC machines. Aim 2: To determine policymakers’ opinions about usability of the model tool from Aim 1. We will convert the model from Aim 1 into a user-friendly, Excel-based model for policymakers to use for decision-making. We will conduct usability interviews (covering learnability, efficiency, memorability, error recovery, and satisfaction) with approximately 20 health administrators and policymakers about the tool. This novel application of diverse methods borrowed from industrial engineering, software engineering, psychology, and quality improvement presents an innovative approach to increase scalability of POC VL testing for children.