Deciphering risk factors for severe postherpetic neuralgia in patients with herpes zoster: an interpretable machine learning approach
Authors' conclusion
Leveraging machine learning analysis, this study identifies an optimal model for predicting severe PHN and highlights key associated risk factors.
Study context
- Population
- Patients diagnosed with herpes zoster (HZ), January 2010 to June 2020of 23,326 reviewed
- Participants
- 8,878 patients
- Exposure
- Five machine learning algorithms for predicting severe postherpetic neuralgia (PHN)
- Outcome
- Progression to severe postherpetic neuralgia (PHN; pain persisting ≥90 days post-HZ)
Methodological detailselectronic medical records database, five machine learning algorithms compared
- Electronic medical records database
- Five machine learning algorithms compared
Main finding
As reported in the abstract
Prediction of severe PHN (best model)
| Group | F1 score | Accuracy | AUC |
|---|---|---|---|
| XGBoost | 0.351 | 0.900 | 0.787 |
XGBoost
F1 score0.351
Accuracy0.900
AUC0.787
Observational data: associations, not effects.
Further findings
Incidence of severe PHN
- Progressed to severe PHN
- 801 (9.0%)
Risk factors for severe PHN (XGBoost)
- Eight major risk factors: older age, female sex, history of shingles and cancer, use of immunosuppressants and antidepressants, intensive initial pain, neutrophil-to-lymphocyte ratio
PHN in model-defined high-risk vs low-risk patients
- PHN seven times more likely in the high-risk group
- P < 0.001
Original publication
Deciphering risk factors for severe postherpetic neuralgia in patients with herpes zoster: an interpretable machine learning approach. Reg Anesth Pain Med. Published 8 January 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →