The proposed research collaboration between the Harvard School of Public Health and the Massachusetts Veterans Epidemiology and Information Center is directly responsive to research objectives #3 and #7 of this FOA: to evaluate the effectiveness of FDA actions (#3), and to inform policy debate on prescription drug overdose prevention using secondary data analyses (#7). General Objectives: 1) To determine whether the risk of unintentional overdose associated with initiating opioid therapy applies equally to all opioid classes, agents, dosages, schedules and durations of use or whether there are particular regimens with safety advantages that should be prescribed preferentially;2) To quantify the impact of recent FDA black box warnings on opioid prescribing practices. Importance: Physicians prescribe opioids in the near absence of sound empirical data about the risk of overdose associated with different opioid preparations. The proposed cohort study will generate clinically relevant data that could directly influence opioid prescribing decisions and guide future regulatory efforts.
Specific Aims : 1. To examine the relationship between onset of prescription opioid use among users who have been newly prescribed these agents and subsequent development of non-fatal overdose. The study cohort consists of opioid users only. We will assess the extent to which risk of non-fatal overdose varies by the following exposure characteristics: a. Specific agent (Aim 1a). b. Duration of use (i.e., time from initiating opioid use) (Aim 1b). c. Duration of action (i.e., short vs. long acting agents) (Aim 1c). d. Dose (in codeine equivalents) (Aim 1d). e. Controlled substance schedule (i.e., II vs. III vs. IV) (Aim 1e) In addition, we will explore whether and if so how risk estimates (hazards ratios) depend on potential effect modifiers including socio-demographic factors, co-morbidity, co-medication, indication for opioid use, and facility characteristics (Aim 1f). We will also assess the extent to which any observed associations identified in Aims 1a-1f appear to be mediated by the development of substance use disorders after initial opioid use (Aim 1g). In secondary analyses, we will repeat sub-aims a-g for all-cause mortality and calculate incident rates for all-cause mortality by agent, duration of use, action, dose, and controlled substance schedule. 2. To assess the impact of the FDA's recent regulatory actions on opioid utilization. Using time trend analyses, we will quantify changes in the rate of use, starting, switching and stopping specific opioid regimens following FDA black box warnings for particular opioid preparations. We will assess the effect of particular warnings on the agent in question and on prescribing overall. Study Design: retrospective cohort study. Setting/Participants: The proposed cohort study uses health care utilization data on over one million new opioid users who are veterans receiving care within the Veterans Healthcare Administration (VHA) system nationally over an 11-year period, 1999-2009. Study subjects are identified using pharmacy and diagnostic information. Intervention: None. Outcome measures: 1) Non-fatal unintentional overdose events (identified in emergency department or outpatient visits or inpatient hospitalizations using ICD-9-CM codes);2) all-cause mortality;3) opioid utilization (identified using individual-level pharmacy records).
Project Narrative The proposed research has the potential to reduce the incidence of death and injury from unintentional overdose, a central objective of this FOA and of the CDC's mission more generally. Findings will provide physicians with new information to guide their opioid prescribing practices and regulatory agencies with information about the impact of black box warnings on prescribing generally, by sub-specialty, and by other relevant prescriber characteristics. Because this project involves a partnership with the Veterans Healthcare Administration (VHA), findings could lead to particularly prompt adoption of safer opioid prescribing practices for veterans receiving care within the VHA system.
| Barber, Catherine; Gagnon, David; Fonda, Jennifer et al. (2017) Assessing the impact of prescribing directives on opioid prescribing practices among Veterans Health Administration providers. Pharmacoepidemiol Drug Saf 26:40-46 |
| Miller, Matthew; Barber, Catherine W; Leatherman, Sarah et al. (2015) Prescription opioid duration of action and the risk of unintentional overdose among patients receiving opioid therapy. JAMA Intern Med 175:608-15 |
