社区康复脊髓损伤患者就业影响因素分析及预测模型构建

关键词：脊髓损伤  就业  影响因素  预测模型

Keywords:spinal cord injury  employment  influencing factors  prediction model

摘要：

      摘要 目的：探讨社区康复脊髓损伤（spinal cord injury， SCI）患者就业的影响因素并构建就业预测模型，为政策制定者、医疗专业人员和社会服务机构提供科学依据，以改善社区康复SCI患者的就业状况和生活质量。 方法：通过多阶段分层整群随机抽样，在江苏省和四川省的28所医院招募了809例符合条件的社区康复SCI患者。采用最小绝对值压缩选择模型（least absolute shrinkage and selection operator， LASSO）筛选可能的影响因素，并用Logistic回归计算OR值及95%CI。基于建立的回归模型，构建列线图预测社区康复SCI患者就业概率并对其预测性能进行内部验证。 结果：年龄（OR=0.96，95%CI：0.94—0.97）、精神健康评分（OR=1.27，95%CI：1.10—1.47）、自我感知健康得分（OR=1.45，95%CI：1.14—1.87）以及环境障碍数量（OR=0.74，95%CI：0.68—0.80）显著影响SCI患者就业。通过Logistic回归构建的预测模型预测性能良好，受试者工作特征曲线下面积为0.864，灵敏度为72.39%，特异度为88.53%。绘制的列线图可以对社区康复SCI患者进行简便的就业概率评估。 结论：社区康复SCI患者就业率较低，年龄、受教育年限、家庭收入、精神健康评分、自我感知健康得分、环境障碍数量等影响因素可对社区康复SCI患者就业概率进行有效预测。

Abstract:

      Abstract Objective： To explore the factors influencing employment among spinal cord injury （SCI） patients in the community and construct an employment prediction model to provide scientific evidence for policymakers， healthcare professionals， and social service agencies， aiming to improve the employment status and quality of life of SCI patients. Method： Using a multi-stage stratified cluster random sampling method，we recruited 809 eligible SCI patients in community rehabilitation from 18 hospitals in Jiangsu Province and 10 hospitals in Sichuan Province. We used the Least Absolute Shrinkage and Selection Operator（LASSO） model to screen factors and constructed a logistic regression model to analyze the factors influencing employment rates among individuals with SCI patients in community rehabilitation. We then performed internal validation. Result： Age （OR=0.96，95% CI：0.94—0.97）， mental health score（OR=1.27，95% CI：1.10—1.47），self-perceived health score （OR=1.45，95% CI：1.14—1.87），and the number of environmental barriers（OR=0.74，95% CI：0.68—0.80） significantly influenced employment among SCI patients in community rehabilitation. The predictive performance of the logistic regression model was good， with an area under the receiver operating characteristic curve（AUC） of 0.864， a sensitivity of 72.39%， and a specificity of 88.53%.The nomogram drawn can facilitate the employment probability assessment for SCI patients in the community. Conclusion： The employment rate of SCI patients in community rehabilitation is relatively low. Influencing factors such as age， years of education， family income， mental health scores， self-perceived health scores， and the number of environmental barriers can effectively predict the employment probability of SCI patients in community rehabilitation.