ixxmu
commented
1 year ago 一文告诉你，到底如何看DCA，发文时到底怎么描述！以及错误文献与完美文献赏析 by 精鼎统计
精鼎79期《SPSS+R语言临床预测模型实战》（2023年03月11-12日)
案例1.
如下这篇文章，根据融入点描述，融入点大约在0.55左右，因此，作者描述模型预测概率<0.55时有更高的净获益；
但是细看离开点，并不是在0的位置，所以不是作者所说的0-0.55，精细一点应该大约0.05-0.55左右。可能因为0.05过小忽视了吧！
The decision curve shows that if the threshold probability of an individual is< 55%, using this model to predict pneumothorax post-CT-TNB adds more beneft than either the treat-all tactics or the treat-none tactics.
案例2
这篇文章采用的时离开点描述，离开点大概0.04，然而融入点已经是最大值1了，因此，作者说在>0.04时，模型有更好的净获益。
Figure 5 Decision curves of the different scoring systems for predicting SAP. Thenet benefit was calculated by adding the true-positives and subtracting the false_x0002_positives. For a threshold probability >4%, application of the SAP nomogram would add net benefit compared to either the treat-all strategy or the treat-none strategy.In addition, the SAP nomogram always showed a greater net benefit than theA2DS2, ISAN, and PANTHERIS scores for predicting SAP with a threshold probability >4%.
案例3
这篇文章完全按照离开点和融入点进行的描述，离开点大约0.1，融入点大概0.45，因此作者说在0.1-0.45范围内模型表现更好的净获益。
The decision curve analysis (DCA) for the prediction nomogram and that for the model with single predictor is presented in Fig. 3. Te fnal DCA showed that if the threshold probability of patients or clinicians is between 10 and 45%, screening strategies based on our nomogram’ colorectal cancer risk estimates resulted in superior net beneft than screen-none or screen-all strategies. Within this range, the predictive efect of nomogram is better than that of a single predictor, respectively
如下松哥再带大家看2篇错误离谱文章的表述，以及1篇完美表达与展示文章的表述。
https://mp.weixin.qq.com/s/JltaSSGSKP1GVAIweECsYQ