Zhang Siyuan, Chen Jerry Yongqiang, Pang Hee Nee, Lo Ngai Nung, Yeo Seng Jin, Liow Ming Han Lincoln
Yong Loo Lin School of Medicine, National University of Singapore, 1E Kent Ridge Road, NUHS Tower Block, Level 11, Singapore, 119228, Singapore.
Department of Orthopaedic Surgery, Singapore General Hospital, 20 College Road, Academia, Level 4, Singapore, 169856, Singapore.
Arthroplasty. 2021 Sep 2;3(1):33. doi: 10.1186/s42836-021-00087-3.
Patient satisfaction is a unique and important measure of success after total hip arthroplasty (THA). Our study aimed to evaluate the use of machine learning (ML) algorithms to predict patient satisfaction after THA.
Prospectively collected data of 1508 primary THAs performed between 2006 and 2018 were extracted from our joint replacement registry and split into training (80%) and test (20%) sets. Supervised ML algorithms (Random Forest, Extreme Gradient Boosting, Support Vector Machines, Logistic LASSO) were developed with the training set, using patient demographics, comorbidities and preoperative patient reported outcome measures (PROMs) (Short Form-36 [SF-36], physical component summary [PCS] and mental component summary [MCS], Western Ontario and McMaster's Universities Osteoarthritis Index [WOMAC] and Oxford Hip Score [OHS]) to predict patient satisfaction at 2 years postoperatively. Predictive performance was evaluated using the independent test set.
Preoperative models demonstrated fair discriminative ability in predicting patient satisfaction, with the LASSO model achieving a maximum AUC of 0.76. Permutation importance revealed that the most important predictors of dissatisfaction were (1) patient's age, (2) preoperative WOMAC, (3) number of comorbidities, (4) preoperative MCS, (5) previous lumbar spine surgery, and (6) low BMI (< 18.5).
Machine learning algorithms demonstrated fair discriminative ability in predicting patient satisfaction after THA. We have identified modifiable and non-modifiable predictors of postoperative satisfaction which could enhance preoperative counselling and improve health optimization prior to THA.
患者满意度是全髋关节置换术（THA）后成功与否的一项独特且重要的衡量指标。我们的研究旨在评估使用机器学习（ML）算法预测THA术后患者满意度的情况。
从我们的关节置换登记处提取2006年至2018年间进行的1508例初次THA的前瞻性收集数据，并将其分为训练集（80%）和测试集（20%）。使用训练集开发监督式ML算法（随机森林、极端梯度提升、支持向量机、逻辑套索回归），利用患者人口统计学数据、合并症和术前患者报告结局指标（PROMs）（简短健康调查问卷36项版[SF-36]、身体成分汇总[PCS]和精神成分汇总[MCS]、西安大略和麦克马斯特大学骨关节炎指数[WOMAC]以及牛津髋关节评分[OHS]）来预测术后2年的患者满意度。使用独立测试集评估预测性能。
术前模型在预测患者满意度方面表现出中等判别能力，套索回归模型的最大曲线下面积（AUC）为0.76。排列重要性分析显示，导致不满意的最重要预测因素为：（1）患者年龄，（2）术前WOMAC，（3）合并症数量，（4）术前MCS，（5）既往腰椎手术史，以及（6）低体重指数（<18.5）。
机器学习算法在预测THA术后患者满意度方面表现出中等判别能力。我们已经确定了术后满意度的可改变和不可改变的预测因素，这可以加强术前咨询并改善THA术前的健康优化。
