Changes to which Drugs and Drug Combinations are Prescribed in Response to Opioid Supply-side Policies

点击翻译按钮获取中文摘要

ABSTRACT Claiming the lives of nearly 400,000 Americans between 1999 and 2017, drug overdoses caused by the misuse of and addiction to opioids – including prescription pain relievers, heroin, and synthetic opioids such as fentanyl – have become a major public health crisis in the U.S. In response to the crisis, state and federal governments have enacted many policies, mostly targeting the supply-side to restrict the issuance of prescriptions. However, physicians face persistent patient demand for pain relief. Consequently, as supply-side policies restrict opioid prescriptions, physicians’ pharmacologic options to address patient demand are limited, including: 1) supplement opioids with off-label, concomitant prescriptions, and 2) prescribe nonopioid pharmaceutical substitutes. In this exploratory R21 we will examine how the policies aimed at addressing the opioid crisis have changed physicians’ prescribing patterns. Preliminary evidence indicates that prescriptions of nonopioid alternatives have been increasing, as have concomitant prescriptions of opioids with benzodiazepines or gabapentinoids. However, there is, according to our knowledge, no empirical evidence linking these trends to the opioid supply-side policies. Accordingly, our study will break new ground in our understanding of how physicians are attempting to meet patient demand for pain relief while facing increasing opioid prescription restrictions. The supply-side policy that we will focus on is Prescription Drug Monitoring Programs (PDMPs). We will use the RAND Opioid Policy and Tools Information Center (OPTIC) database which documents policies related to opioid prescribing, as well as the Truven Health MarketScan (2005-2020) database of commercial and Medicare prescription drug claims, and the Centers for Disease Control’s Multiple Cause of Death (MCOD) data. With this combination of data, we will first examine the effects of PDMPs on the concomitant prescription of opioids with benzodiazepines or gabapentinoids. Then, we will explore PDMPs’ effects on the prescription of nonopioid substitute pharmaceuticals. Finally, we will examine the population health effects, specifically the effects of PDMPs on fatal overdoses – all fatal overdoses, those involving opioids (both licit and illicit), and those involving combinations of opioids with benzodiazepines or gabapentinoids. For each aim, we will leverage state- year variation in the timing of PDMP implementation and use the controlled interrupted time series method. Additionally, we will perform descriptive analyses (e.g. trends), robustness checks (e.g. alternative outcome specifications such as dosage and duration), and heterogeneity analyses (e.g. different effects of PDMP types). While PDMPs have been studied with regards to their effects on opioid prescriptions, the existing literature does not consider the role of PDMPs in the prescription of benzodiazepines, gabapentinoids, and nonopioid pharmaceutical substitutes despite recent FDA warnings and evidence of the increasingly frequent involvement of these substances in overdose fatalities. Thus, this exploratory R21 will begin an important and novel research into physician responses to opioid supply-side policies.