基于体位操训练的慢性便秘患者列线图预测模型的构建

Establishment of a Nomogram prediction model for chronic constipation patients based on body position exercise training

  • 摘要:
    目的 构建基于体位操训练的慢性便秘患者列线图预测模型。
    方法 将2021年3月—2023年3月收治的327例慢性便秘患者根据体位操合格情况分为合格组(n=279)和不合格组(n=48)。收集患者一般临床资料, 采用单因素和多因素Logistic回归分析影响慢性便秘患者体位操不合格的危险因素; 采用R软件构建预测慢性便秘患者体位操不合格的列线图模型,并应用受试者工作特征(ROC)曲线及校准曲线验证列线图模型的区分度和一致性。
    结果 2组患者性别、文化程度、吸烟史、酗酒史、高血压史、腹部手术史、结肠镜检查史、服药方法是否正确比率相比,差异均无统计学意义(P>0.05); 不合格组患者年龄、慢性便秘病程、糖尿病史比率均高于合格组患者,差异有统计学意义(P < 0.05); 年龄、慢性便秘病程、糖尿病史是影响慢性便秘患者体位操不合格的独立危险因素(P < 0.05)。构建的列线图预测模型具有较优的区分度(曲线下面积为0.923, 95%CI为0.888~0.949, 灵敏度、特异度分别为87.50%、81.00%)和一致性(H-L拟合优度检验χ2=2.246, P=0.973)。高风险阈值概率在0.06~0.80时应用该列线图模型预测慢性便秘患者体位操不合格的临床价值较高。
    结论 年龄、慢性便秘病程、糖尿病史是影响慢性便秘患者体位操不合格的独立危险因素,根据多因素结果构建的预测慢性便秘患者体位操不合格的列线图模型有较高的区分度和一致性。

     

    Abstract:
    Objective To construct a Nomogram prediction model based on body position exercise training for patients with chronic constipation.
    Methods A total of 327 patients with chronic constipation from March 2021 to March 2023 were divided into qualified group (n=279) and unqualified group (n=48) according to their performance in body position exercises. General clinical materials of the patients were collected, and univariate and multivariate Logistic regression analyses were performed to identify risk factors for unqualified body position exercises in patients with chronic constipation; the R software was used to construct a Nomogram model for predicting unqualified body position exercises in patients with chronic constipation, and the receiver operating characteristic (ROC) curve and calibration curve were applied to verify the discrimination and consistency of the Nomogram model.
    Results There were no significant differences in gender, education level, history of smoking, history of alcohol abuse, history of hypertension, history of abdominal surgery, history of colonoscopy, and correct medication administration ratio between the two groups (P>0.05); the unqualified group had higher age, longer duration of chronic constipation, and a higher proportion of diabetes history compared to the qualified group (P < 0.05); age, duration of chronic constipation, and diabetes history were independent risk factors for unqualified body position exercises in patients with chronic constipation (P < 0.05). The constructed Nomogram prediction model demonstrated excellent discrimination (area under the curve was 0.923, 95%CI, 0.888 to 0.949, with sensitivity and specificity of 87.50% and 81.00% respectively) and consistency (H-L goodness-of-fit test χ2=2.246, P=0.973). The clinical value of this Nomogram model in predicting unqualified body position exercises for patients with chronic constipation was high when the high-risk threshold probability ranged from 0.06 to 0.80.
    Conclusion Age, duration of chronic constipation, and diabetes history are independent risk factors for unqualified body position exercises in patients with chronic constipation, and the Nomogram model constructed based on these multi-factor results has high discrimination and consistency in predicting unqualified body position exercises for patients with chronic constipation.

     

/

返回文章
返回