A novel algorithm to compute adherence from electronic adherence monitoring devices

项目摘要/摘要 服药依从性是国家癌症研究所当前的科学优先事项，也是临床的首要任务 审判。电子依从性监测设备(EAMD)是装有计算机芯片的药瓶或药盒 它记录了每个瓶子/盒子打开(或“启动”)的日期和时间。由于它们的准确性和能力 捕捉日常服药行为，EAMD越来越多地被引用(即，被FDA)作为 研究中每日抗癌药物依从性的首选衡量标准。不幸的是，研究人员目前 缺乏将原始EAMD激励准确高效地转换为遵从性数据所需的工具 目前完成这项任务的方法不标准化、容易出错且繁重。这样做的目的是 R21将开发一种新的算法，将EAMD驱动转换为粘合数据。最大限度地发挥潜力 吸收和易用性，我们将聘请一个多学科坚持科学焦点小组，以告知设计 算法和相关用户界面(目标1)。然后，我们将评估该算法在 对日常遵守数据进行分类(目标2)并获取使用质量反馈(目标3)，以便为最终改进提供信息 在大规模测试之前。

PROJECT SUMMARY/ABSTRACT Medication adherence is a current scientific priority of the National Cancer Institute and a top priority in clinical trials. Electronic adherence monitoring devices (EAMDs) are pill bottles or boxes that contain a computer chip that records the dates and times of each bottle/box opening (or “actuation”). Due to their accuracy and ability to capture day-to-day medication-taking behavior, EAMDs are increasingly cited (i.e., by the FDA) as the preferred measure of daily anticancer medication adherence in research. Unfortunately, researchers currently lack the tools they need to accurately and efficiently convert raw EAMD actuations into adherence data and current methods of completing this task are unstandardized, error-prone, and burdensome. The purpose of this R21 is to develop a novel algorithm to convert EAMD actuations into adherence data. To maximize potential uptake and usability, we will engage a multidisciplinary adherence science focus group to inform the design of the algorithm and associated user interface (Aim 1). We will then evaluate the accuracy of the algorithm in classifying daily adherence data (Aim 2) and obtain quality of use feedback (Aim 3) to inform final refinements prior to large-scale testing.