| Purpose of Study: This descriptive correlational study was designed to identify factors that can lead to unique pain management regimens based on individual attributes. The variables for this study included preoperative pain intensity, age, sex, body mass index, smoking status, and race/ethnicity. All variables were examined over three postoperative days.Methods: Data was collected from the electronic medical records of 1123 patients who were 18 to 89 years of age and had undergone elective, unilateral, total knee arthroplasty. These records were identified by the International Classification of Diseases (ICD-9) Medical Procedure Code 81.54. Due to the impact the following comorbidities have on pain intensity and pain management, records with documented comorbidities of addictive disorder, selected chronic pain syndromes, and depression, also identified by (ICD-9) coding, were excluded from the study. Only patient records with consistent use of the numeric rating scale for pain intensity were included in the sample. The study sample consisted primarily of Caucasians (95%) and females (63%). The age range was 26 to 89 years of age with a mean age of 65 years. The body mass index ranged from 16.50 to 79.40. Using BMI categories that are consistent with those used by the National Heart, Lung, and Blood Institute, the World Health Organization, and the Centers for Disease Control, the sample was mainly overweight (25%) and obese (66%). More than half (51.02%) of the sample had never smoked cigarettes followed closely by another 31.43% who had quit smoking more than a year prior to surgery. All research questions were analyzed using generalized estimating equations (GEE). Multiple measurements on the same subject at different points in time resulted in correlated data. GEE is an extension of generalized linear models and is an appropriate method for analyzing correlated outcome data. A criterion measure known as a QIC (quasilikelihood information criterion) was used to evaluate goodness of fit. Three predictive models for pain intensity, opioid consumption, and comfort goal attainment were generated by this study The best predictive models are linear combinations of the regression coefficient estimates for each variable included in the model.Results: The most consistent predictors found in all three best predictive models were race, smoking, and age. In this study black patients, younger patients, and smokers tended to have significantly different pain intensity, opioid consumption, and comfort goal attainment than other groups. The predictor models can be used to calculate pain intensity, opioid consumption, and the predicted difference between desired comfort goal and actual pain intensity postoperatively.Conclusions: The predictive models generated in this study will assist health care providers in assessing the risk of postoperative pain, increased opioid consumption, and failure to attain desired comfort goals in the TKA population. Early identification of these risks will allow for aggressive intervention prior to side effects or adverse outcomes. |
