糖尿病足发病风险预测模型的系统评价

A systematic review of risk prediction models for diabetic foot

目的系统地分析、评价糖尿病足发病风险预测模型,以期为护理实践提供参考依据。方法检索PubMed、Embase、Web of Science、Cochrane图书馆、中国生物医学文献数据库、中国知网和万方数据库中有关糖尿病足发病风险预测模型的研究,检索时限为建库至2020年3月1日。由2名研究者独立筛选文献和提取数据,并应用预测模型研究的偏倚风险评估工具分析纳入文献的偏倚风险和适用性。结果共纳入7项糖尿病足发病风险预测模型研究,包括6项开发研究和1项验证研究。7个模型的受试者工作特征曲线下面积为0.65~0.88。所有纳入模型中包含最多的预测因子是单丝试验不敏感。7项研究的适用性较好,但存在一定的偏倚,主要是因为未采取或未报告盲法、应变量事件数不足、忽略缺失数据、缺乏模型性能评估以及模型过度拟合。结论糖尿病足发病风险预测模型的研究还处于发展阶段,未来应开发性能优良、偏倚风险低的预测模型,并进行内部验证或外部验证。

Objective We systematically analyzed and compared the relevant studies on risk prediction models for diabetic foot,so as to provide references for nursing practice.Methods We searched PubMed,Embase,Web of Science,Cochrane Library,CBM,CNKI,Wanfang Data from the inception to March 1,2020,in order to collect studies on risk prediction models for diabetic foot.2 researchers independently screened the literature,extracted information,and assessed the risk of bias and applicability concerns of the included literature by the prediction model risk of bias assessment tool.Results The search identified 7 relevant studies,including 6 development studies and 1 verification study.The area under the curve of 7 models was 0.65~0.88.The presence of monofilament insensitivity was the most commonly reported predictive factor of diabetic foot among all included models.All included studies have low concerns regarding applicability,but with a certain bias,which mostly due to the absence or non-reported of the blind method,the insufficient number of events per variable,omitting samples with missing data,the lack of model performance assessment and overfitting.Conclusion The research on risk prediction models for diabetic foot is still in the development stage.In the future,prediction models with good performance and low risk of bias should be developed and verified internally or externally.