Dataset Information

Development and prospective validation of postoperative pain prediction from preoperative EHR data using attention-based set embeddings.

ABSTRACT: Preoperative knowledge of expected postoperative pain can help guide perioperative pain management and focus interventions on patients with the greatest risk of acute pain. However, current methods for predicting postoperative pain require patient and clinician input or laborious manual chart review and often do not achieve sufficient performance. We use routinely collected electronic health record data from a multicenter dataset of 234,274 adult non-cardiac surgical patients to develop a machine learning method which predicts maximum pain scores on the day of surgery and four subsequent days and validate this method in a prospective cohort. Our method, POPS, is fully automated and relies only on data available prior to surgery, allowing application in all patients scheduled for or considering surgery. Here we report that POPS achieves state-of-the-art performance and outperforms clinician predictions on all postoperative days when predicting maximum pain on the 0-10 NRS in prospective validation, though with degraded calibration. POPS is interpretable, identifying comorbidities that significantly contribute to postoperative pain based on patient-specific context, which can assist clinicians in mitigating cases of acute pain.

SUBMITTER: Liu R

PROVIDER: S-EPMC10654400 | biostudies-literature | 2023 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Development and prospective validation of postoperative pain prediction from preoperative EHR data using attention-based set embeddings.

Liu Ran R Gutiérrez Rodrigo R Mather Rory V RV Stone Tom A D TAD Santa Cruz Mercado Laura A LA Bharadwaj Kishore K Johnson Jasmine J Das Proloy P Balanza Gustavo G Uwanaka Ekenedilichukwu E Sydloski Justin J Chen Andrew A Hagood Mackenzie M Bittner Edward A EA Purdon Patrick L PL

NPJ digital medicine 20231116 1

Preoperative knowledge of expected postoperative pain can help guide perioperative pain management and focus interventions on patients with the greatest risk of acute pain. However, current methods for predicting postoperative pain require patient and clinician input or laborious manual chart review and often do not achieve sufficient performance. We use routinely collected electronic health record data from a multicenter dataset of 234,274 adult non-cardiac surgical patients to develop a machin ...[more]

PMID: 37973817

Similar Datasets

Project description:BackgroundPostoperative pain management is key for patient satisfaction. Pressure pain threshold (PPT) has been studied in some surgical cohorts but has not been studied in relationship to acute postoperative pain in short-stay patients undergoing anorectal surgery. We hypothesized that preoperative finger PPT measurements can identify respective patients with higher postoperative pain. Aiming to understand the relationship with subjective postoperative pain perception, we tested the hypotheses that preoperative PPT is associated with postoperative Visual Analog Scale (VAS) pain scores and correlates with postoperative analgesic consumption in short-stay patients undergoing anorectal surgery.MethodsWe prospectively assessed preoperative PPT in a cohort undergoing anorectal surgery, known as a moderately to severely painful procedure. Linear mixed-effects models were used to assess the relationship with postoperative VAS pain scores at 1 and 3 days as well as 4 weeks postoperatively. Logistic regression was used to study the relationship with additional postoperative analgesic consumption.ResultsWe studied 128 patients and found that preoperative PPT is significantly associated with postoperative pain (P value for interaction = .025). Logistic regression modeling additionally revealed an association between the preoperative PPT and the need for additional postoperative analgesics, with odds of requiring additional analgesia decreasing by about 10% for each 1-point increase in PPT (odds ratio [OR] = 0.90; 95% confidence interval [CI], 00.81-0.98; P = .012).ConclusionsPreoperative finger PPT is associated with postoperative pain and might help identify patients who are at risk of developing more severe postoperative pain on anorectal surgery. Especially in ambulatory and short-stay settings, this approach can help to address patients' high variability in pain sensitivity to facilitate appropriate postoperative analgesia, timely discharge, and prevent readmission.

Dataset Information

Development and prospective validation of postoperative pain prediction from preoperative EHR data using attention-based set embeddings.

Publications

Development and prospective validation of postoperative pain prediction from preoperative EHR data using attention-based set embeddings.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets