Data for"Association Between Preoperative Sleep Disorders and Chronic Postoperative Pain in Hysterectomy Patients: A Prospective Cohort Study"
Description
This dataset contains individual-level data from a prospective cohort study investigating the association between preoperative sleep disorders and chronic postsurgical pain. A total of 275 patients were stratified into two groups based on preoperative sleep quality scores: a no sleep disorders group （n=145）and a sleep disorders group（n=130）, with subsequent follow-up evaluations conducted to assess pain outcomes. This dataset originates from a prospective cohort study that enrolled 275 women aged 18 to 65 years, scheduled for elective hysterectomy at a single center, in accordance with predefined inclusion and exclusion criteria. Data1 includes preoperative information such as baseline characteristics (sex, age, body mass index, underlying diseases, and surgical history), pain intensity (Numeric Rating Scale, NRS) and sleep quality (Pittsburgh Sleep Quality Index, PSQI), as well as postoperative acute pain scores, chronic pain assessments, and analgesic usage. Data2 contains detailed PSQI component scores related to sleep quality. Data3 comprises perioperative hemodynamic measurements of the patients. All data have been fully de-identified: only study identification numbers and group labels are provided, with no personal identifiers such as names, dates of birth, or contact information included. All variables, units, and coding schemes are clearly annotated to facilitate data reuse and replication of analyses reported in the associated publication.
Files
Steps to reproduce
To reproduce the main analyses presented in the article, first download the following files（data1、data2、data3）. Each row in the raw data file corresponds to one study participant, and columns represent baseline characteristics, perioperative variables, and postoperative outcomes as detailed in the accompanying data dictionary. After importing the raw data into a statistical software package (e.g., R, SPSS, or Stata), apply the variable names, coding schemes, and value labels as specified in the data dictionary. Define the exposure groups as “No Sleep Disorder (PSQI ≤ 7)” and “Sleep Disorder (PSQI > 7)” based on preoperative Pittsburgh Sleep Quality Index scores. For baseline characteristics (presented as Table 1 in the article), summarize continuous variables using mean ± standard deviation or median with interquartile range depending on their distribution, and present categorical variables as frequencies and percentages by sleep disorder status. To evaluate the primary outcome—chronic postsurgical pain (CPSP) at 3 months—apply binary logistic regression to estimate odds ratios (OR) with 95% confidence intervals (CI), both unadjusted and adjusted for covariates such as age, BMI, acute postoperative pain, and ovarian preservation status. Propensity score methods (matching or inverse probability weighting) may be used for additional robustness, as detailed in the manuscript. Reproduction of ROC analysis for the predictive model can be performed using the variables indicated in the article, and discriminative performance should be reported as the area under the curve (AUC). All analytical steps—including handling of missing data, model specifications, and sensitivity analyses—follow the descriptions provided in the published methods section.
Institutions
- Nanjing University Medical School