State-level predictors and outcomes associated with implementation of prescription drug monitoring program integration and mandatory use policies

Christian E Johnson

doi:10.25820/etd.006974

The opioid epidemic in the United States has resulted in significant morbidity, mortality, and economic costs affecting people of all ages, genders, races, and ethnicities. More than a million people who need treatment for opioid use disorder lack access to it despite the known benefits of medications for opioid use disorder on reducing opioid misuse, improving treatment retention, and preventing overdose death. Prescription drug monitoring programs (PDMPs) are tools states have implemented to address the opioid epidemic. These databases collect patients’ prescription opioid (and other controlled substances) history and make that information available to health care providers in an effort to reduce opioid prescribing and help identify people seeking prescription drugs for nonmedical use. This dissertation examines two methods intended to increase prescribers’ use of PDMPs: (1) mandatory use policies, which require prescribers to view PDMPs before prescribing additional opioids, and (2) PDMP integration, a technology that makes it easier for prescribers to view PDMP information in the context of a patient’s full health history. To better understand states decisions to implement these PDMP approaches and their impact on patient health, this dissertation aimed to: (1) identify state-level factors associated with implementation of PDMP integration and mandatory use policies, (2) examine the extent to which PDMP integration and mandatory use policies increase distribution of medications for opioid use disorder, and (3) examine the extent to which PDMP integration and mandatory use policies improve neonatal health outcomes associated with opioid use during pregnancy. For Aim 1, I performed an extended Cox regression to estimate the association of prior state rates of opioid dispensing, prescription opioid overdose deaths, and neonatal opioid withdrawal syndrome hospitalizations with implementation of PDMP integration and mandatory use policies from 2009-2020, controlling for demographic and economic factors, state government and political factors, and number of previously adopted opioid policies. In my main model, prior opioid dispensing (HR 2.31, 95% CI 1.17, 4.57), neonatal opioid withdrawal syndrome hospitalizations (HR 1.55, 95% CI 1.09, 2.19), and number of prior opioid policies (HR 2.13, 95% CI 1.13, 4.00) were associated with implementation of mandatory use policies. Prior prescription opioid overdose deaths (HR 1.21, 95% CI 1.08, 1.35) were also associated with implementation of mandatory use policies in a model that did not include opioid dispensing or neonatal opioid withdrawal syndrome hospitalizations. However, I did not observe any state-level factors associated with PDMP integration. For Aim 2, I used a difference-in-differences design to estimate the effect of PDMP integration and mandatory use policies on four outcomes (distribution of buprenorphine to opioid treatment programs, distribution of buprenorphine to pharmacies, distribution of methadone to opioid treatment programs, and the total combined distribution of methadone and buprenorphine), controlling for treatment supportive policies, policies aimed at reducing the supply of prescription opioids, and harm reduction policies for the years 2009-2021. Distribution of buprenorphine to pharmacies decreased 8% (95% CI -14%, -1%) following implementation of mandatory use policies. Conversely, distribution of methadone to opioid treatment programs increased 17% (95% CI 4%, 34%) and the total combined distribution of methadone and buprenorphine increased 6% (95% CI -0%, 14%) following the joint implementation of both approaches. For Aim 3, I used a difference-in-differences design to estimate the effect of PDMP integration and mandatory use policies on three outcomes (neonatal opioid withdrawal syndrome hospitalizations, length of stay, and cost per hospitalization), controlling for punitive policies associated with drug use during pregnancy, treatment supportive policies, and policies aimed at reducing the supply of prescription opioids for the years 2009-2020. Neither PDMP integration nor mandatory use policies had an effect on neonatal opioid withdrawal syndrome hospitalizations, length of stay, or cost per hospitalization (p<0.05). Overall, considering the results of all three dissertation aims, opioid dispensing and adverse health outcomes appeared to motivate state implementation of mandatory use policies but not PDMP integration. The joint implementation of PDMP integration and mandatory use policies resulted in small increases in the distribution of methadone and buprenorphine but had no discernible effect on neonatal opioid withdrawal syndrome hospitalizations. These findings suggest further work is needed to educate providers about the benefits of using buprenorphine to treat opioid use disorder, even among pregnant women. This dissertation focused on the effect of PDMPs which is primarily a tool aimed at decreasing harms associated with prescription opioid misuse. However, the epidemiology of opioid overdose mortality is shifting towards potent synthetic opioids obtained outside the healthcare system. Researchers and policymakers should continue to assess the utility of PDMPs and their potential unintended consequences in this new phase of the opioid epidemic. States should also consider additional approaches to prevent the misuse of drugs, treat those with substance use disorder, and minimize harm associated with drug use.

State-level predictors and outcomes associated with implementation of prescription drug monitoring program integration and mandatory use policies

Files and links (1)

Abstract

Details

Metrics