Article Text
PDF
Abstract
Introduction Postherpetic neuralgia (PHN) is a common complication of herpes zoster (HZ). This study aimed to use a large real-world electronic medical records database to determine the optimal machine learning model for predicting the progression to severe PHN and to identify the associated risk factors.
Methods We analyzed the electronic medical records of 23,326 patients diagnosed with HZ from January 2010 to June 2020. PHN was defined as pain persisting for ≥90 days post-HZ, based on diagnostic and prescription codes. Five machine learning algorithms were compared with select the optimal predictive model and a subsequent risk factor analysis was conducted.
Results Of the 23,326 patients reviewed, 8,878 met the eligibility criteria for the HZ cohort. Among these, 801 patients (9.0%) progressed to severe PHN. Among the various machine learning approaches, XGBoost—an approach that combines multiple decision trees to improve predictive accuracy—performed the best in predicting outcomes (F1 score, 0.351; accuracy, 0.900; area under the receiver operating characteristic curve, 0.787). Using this model, we revealed eight major risk factors: older age, female sex, history of shingles and cancer, use of immunosuppressants and antidepressants, intensive initial pain, and the neutrophil-to-lymphocyte ratio. When patients were categorized into low-risk and high-risk groups based on the predictive model, PHN was seven times more likely to occur in the high-risk group (p<0.001).
Conclusions Leveraging machine learning analysis, this study identifies an optimal model for predicting severe PHN and highlights key associated risk factors. This model will enable the establishment of more proactive treatments for high-risk patients, potentially mitigating the progression to severe PHN.
- Neuralgia
- CHRONIC PAIN
- Quality Improvement
- Pain Measurement
Data availability statement
Data are available upon reasonable request. Data may be obtained from a third party and are not publicly available. Data may be obtained from a third party and are not publicly available.
Statistics from Altmetric.com
Data availability statement
Data are available upon reasonable request. Data may be obtained from a third party and are not publicly available. Data may be obtained from a third party and are not publicly available.
Footnotes
SJP and JH contributed equally.
Contributors SJP and JH contributed equally to this work as co-first authors. They were responsible for conceptualization, data collection, analysis, and drafting the initial manuscript. JH performed data analysis, statistical modeling, and contributed to the interpretation of the results. JBC and SKM provided clinical expertise, supervised the study, and assisted in the manuscript revision. JP and SC oversaw the study design, methodology development, and coordinated the overall research efforts. SC is the guarantor of this study and is responsible for its overall content.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.