Overview
This prospective observational study will externally validate a previously developed and locked prediction model for identifying adult patients at risk of adverse postoperative pain trajectories after elective orthopedic surgery. Approximately 400 patients will be consecutively recruited at the First Affiliated Hospital of Chongqing Medical University.
Postoperative pain will be assessed during the first 48 hours after surgery using the 0-10 Numeric Rating Scale. Pain scores will be organized into eight consecutive 6-hour time windows to determine each participant's postoperative pain trajectory according to prespecified rules. Clinical information available at the predefined prediction time point will be entered into the locked prediction model to calculate each participant's predicted probability of an adverse pain trajectory.
The study will evaluate the model's discrimination, calibration, overall prediction performance, and potential clinical utility. The model will not be modified or retrained using data from this validation cohort. Individual prediction results will not be provided to patients or treating clinicians and will not be used to guide treatment during the study.
Description
This is a single-center, prospective, observational external validation study of an early prediction model for adverse postoperative pain trajectories in adult patients undergoing elective orthopedic surgery at the First Affiliated Hospital of Chongqing Medical University.
Eligible participants will receive routine perioperative care. The study will not assign any experimental treatment or alter the participants' surgical, anesthetic, analgesic, nursing, or rehabilitation management. Study personnel will collect prespecified clinical variables from patient interviews, electronic medical records, anesthesia information systems, nursing records, and standardized postoperative pain assessments.
The end of surgery will be defined as postoperative hour 0. Postoperative pain intensity will be assessed using the 0-10 Numeric Rating Scale during the first 48 hours after surgery. This period will be divided into eight consecutive 6-hour windows: 0-6, 6-12, 12-18, 18-24, 24-30, 30-36, 36-42, and 42-48 hours. When more than one valid pain score is available within a time window, the median value will be used for the primary analysis.
Before enrollment of the first participant, the prediction model, prediction time point, predictor definitions, variable coding, data-processing procedures, missing-data handling rules, model structure, model parameters and hyperparameters, and outcome classification rules will be locked. Data from the external validation cohort will not be used to reselect predictors, retune hyperparameters, redefine cutoffs, or otherwise modify the original model.
After completion of the 48-hour observation period and data verification, each participant's probability of developing an adverse postoperative pain trajectory will be calculated using the locked model. The observed postoperative pain trajectory will be determined independently according to the prespecified trajectory classification and mapping rules.
External model performance will be evaluated using the area under the receiver operating characteristic curve, Brier score, calibration curve, calibration intercept, and calibration slope. Additional analyses will include sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and decision curve analysis at prespecified risk thresholds. Exploratory analyses may describe model performance across major orthopedic surgical categories.
Model predictions will be calculated offline and will not be communicated to participants, surgeons, anesthesiologists, nurses, or other treating clinicians during the study. Prediction results will therefore not influence clinical management.
Eligibility
Inclusion Criteria:
- Aged 18 to 90 years, regardless of sex.
- Scheduled to undergo an eligible elective orthopedic surgical procedure at the First Affiliated Hospital of Chongqing Medical University.
- Expected to remain hospitalized for at least 48 hours after surgery.
- Conscious before surgery, with sufficient understanding and communication ability to understand and complete the 0-10 Numeric Rating Scale for pain.
- Able and willing to provide written informed consent before any study-specific data collection.
Exclusion Criteria:
- Younger than 18 years or older than 90 years.
- Emergency surgery, unplanned surgery, or a surgical procedure outside the prespecified orthopedic surgical scope of the study.
- Unable to reliably understand the study or complete the Numeric Rating Scale because of impaired consciousness, severe cognitive impairment, psychiatric disorder, persistent deep sedation, mechanical ventilation, severe hearing impairment, language or communication difficulties, or similar conditions.
- Lacks full capacity to provide valid informed consent.
- Concurrent participation in another clinical study that may affect pain assessment, analgesic treatment, or study outcomes.
- Expected discharge or transfer within 48 hours after surgery, or inability to complete the main study observation because of clinical circumstances.
- Any other condition that, in the investigator's judgment, makes the patient unsuitable for participation in the study.
