Project Summary/Abstract
The opioid addiction crisis in the US is increasingly affecting older adults. Hospitalization rates for opioid
overdoses grew five-fold from 1993 to 2012 among older Medicare beneficiaries. High rates of prescription
opioid use among older adults are particularly concerning given that adverse outcomes related to prescription
opioid use - including falls, fractures, and all cause-mortality - are known to be greater in older adults.
An important, common, and unstudied scenario for new opioid prescription is at the time of discharge from
a medical hospitalization (e.g., non-surgical conditions such as pneumonia, heart failure, sepsis, delirium, etc.).
Elderly patients who are hospitalized with medical conditions are at elevated risk of adverse outcomes related
to opioid use not only due to clinical fragility but also fragmented transitions in care, which may make
monitoring of opioids’ side effects by outpatient providers difficult. Yet, little data exist on the extent to which
patients hospitalized for medical conditions are prescribed an opioid at time of discharge, the medical
conditions for which prescribing is greatest, and the effects of these opioid prescriptions on adverse patient
outcomes and long-term opioid dependence.
It is also unknown how individual physicians and hospitals contribute to inappropriate prescription opioid
use among patients hospitalized with general medical conditions, as well as the physician- and hospital-level
factors that correlate with opioid prescribing. Variation in physician prescribing is thought to be an important
driver of the opioid epidemic, in part because existing guidelines are underutilized, allowing for non-uniform
prescribing to emerge. Differences in hospital prescribing cultures may also drive variation in prescription
opioid use across hospitals.
The proposed grant will provide national estimates of the rate of opioid prescribing following general
medical hospitalizations in Medicare; identify conditions for which prescribing and opioid-related adverse
events are most common; quantify patient, physician, and hospital risk-factors that are associated with opioid
prescribing; and use novel quasi-experimental methods to determine the effects of provider (both physician
hospital) opioid prescribing patterns on opioid-related adverse events and long-term opioid use.
The proposed research will be the first to comprehensively study how a single hospitalization can
precipitate opioid-related adverse outcomes and long-term opioid dependence. The study’s findings will help
policymakers and clinicians define the risks of transition to long-term opioid use and opioid-related adverse
events in the post-hospitalization setting and will help inform targeted opioid policy to help elderly patients
avoid opioid-related adverse consequences and dependency. The project will build a data infrastructure that
allows for up-to-date data access, ensuring that the project’s findings are relevant to the current clinical
practice and opioid policy.