Assessing the Safety and Effectiveness of Opioid Tapering in Large Health Systems

点击翻译按钮获取中文摘要

PROJECT ABSTRACT The United States is amidst a rapidly evolving opioid epidemic, with more than 200,000 deaths from opioid pain reliever overdoses since 1999. In 2016, the Centers for Disease Control and Prevention issued guidelines on the management of opioid therapy in chronic pain. These guidelines emphasized tapering opioids among individuals prescribed high-dose opioid therapy or with other overdose risk factors. While physicians are increasingly reducing or discontinuing (tapering) patients prescribed chronic opioid therapy, evidence to support the effectiveness and safety of tapering practices is lacking. While tapering opioids represents a reasonable approach to reduce the risk of overdose, there are also concerns that tapering opioids may lead to increased pain and withdrawal, prompting some individuals to seek illicit opioids such as heroin, paradoxically increasing their risk for overdose. The overall goal of this project is to conduct a longitudinal cohort study to determine the effectiveness and safety of opioid tapering in large health systems. We will leverage a data system (Pathways to Opioid Safety Datalink [POSD]) that contains linked data on more than 2.5 million people in rural and urban areas in Colorado and Wisconsin. Using the POSD data system, we will examine opioid tapering practices in a cohort of approximately 48,000 patients prescribed chronic opioid therapy from 2012 to 2020. We will employ novel analytic techniques to identify distinct tapering practices, such as gradual dose reduction, rapid dose reduction, dose discontinuation followed by re-initiation, and sustained dose discontinuation. We will then determine which patient and provider factors and behaviors are associated with opioid tapering. In the cohort, effectiveness and safety will be examined by comparing clinically important outcomes across tapering practices. Outcomes will include overdose, suicide, opioid use disorder, and health care utilization for pain and withdrawal. Lastly, we will develop and apply novel methods to minimize bias from loss to follow-up and misclassification of exposure and outcomes, including quantitative bias analysis, probabilistic bias analysis, and propensity scores. Our findings will have the potential to influence national clinical guidelines on opioid management. Our proposed research approach also represents a new paradigm for how to rigorously assess important opioid prescribing and substance use outcomes using complex health services data.