Tracking persistent postoperative opioid use: a proof-of-concept study demonstrating a use case for natural language processing
Study context
- Population
- Patients who underwent major spine surgery, July 2015-August 2021
- Participants
- 965 patients
- Comparison
- Natural language processing (NLP) engine vs Clinician manual review
- Outcome
- Persistent postoperative opioid use (continued opioid use ≥3 months after surgery), ascertained by clinician manual review
- Follow-up
- ≥3 months after surgery
Methodological detailsproof-of-concept, retrospective EHR review
- Proof-of-concept
- Retrospective EHR review
Performance
As reported in the abstract
Identification of persistent postoperative opioid use
| Test or model | Opioid use status correctly determined | Persistent opioid use correctly identified | No persistent opioid use correctly identified |
|---|---|---|---|
| NLP engine | 92.9% | 95.6% | 86.1% |
NLP engine
Opioid use status correctly determined92.9%
Persistent opioid use correctly identified95.6%
No persistent opioid use correctly identified86.1%
Further findings
Persistent opioid use following surgery (prevalence)
- Patients determined to have persistent opioid use following surgery
- 705 (73.1%)
Original publication
Tracking persistent postoperative opioid use: a proof-of-concept study demonstrating a use case for natural language processing. Reg Anesth Pain Med. Published 6 July 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →