Characterizing non-heroin opioid overdoses using electronic health records

Area of Science:

• Public Health
• Computational Medicine
• Epidemiology

Background:

• The opioid epidemic is a critical public health crisis.
• Current state-level interventions may not address individual patient needs.
• Individual-level analysis is crucial for effective opioid epidemic strategies.

Purpose of the Study:

• To characterize the opioid epidemic at the individual-level.
• To understand patient characteristics associated with opioid use disorder (OUD).
• To identify factors preceding first-time overdose events.

Main Methods:

• Utilized computational tools and electronic health record data from a large academic medical center.
• Employed a self-controlled analysis comparing pre- and post-overdose healthcare encounters.
• Contrasted patients with OUD to matched non-OUD controls.

Main Results:

• Confirmed a significant increase in hospital opioid overdoses from 2006-2015 (9% annual average).
• Identified preceding conditions of pain or malignancy before the first overdose.
• Suggests a potential link between pharmaceutical opioids for these conditions and subsequent overdose.

Conclusions:

• Informatics-based methods can enhance understanding of individuals with opioid dependency and overdose.
• This approach may inform future research and interventions.
• Potential to reduce morbidity and mortality associated with the opioid epidemic.