Logistic函模型在预测慢病患病率中的应用.docVIP

Logistic函数模型在预测慢病患病率中的应用 严若华 王杨1 李卫1 摘要 目的。 方法 结果 结论实际， 关键词患病率预测 The Application of Logistic Model in Predicting the Prevalence of Chronic Non-communicable Diseases YAN Ruohua WANG Yang LI Wei State Key Laboratory of Cardiovascular Disease, Cardiovascular Institute and Fu Wai Hospital, CAMS and PUMC, Beijing, 100037, China Corresponding Author: Li Wei, Email: liwei@ Abstract Objective: To explore the application of Logistic model in predicting the prevalence of chronic non-communicable diseases. Methods: The prevalence of hypertension in China is used as an example. We take year, logarithm of per capita GDP and aging rate (the proportion of people over 65) respectively as independent variable, and prevalence rate as dependent variable, to build Logistic models. Compare these models’ fitting effects by the mean absolute error (MAE), the mean square error (MSE) and the determination coefficient (R2). Results: Take 40% as the upper limit of the models. If the independent variable is “year”, the predicted value of prevalence in 2010 is 20.3452%, and the prevalence rate will tend to stabilize in 2060 (MAE = 0.7345, MSE = 0.7038, R2 = 0.9633); if x refers to “log(per capita GDP)”, the prevalence will arrive to 23.8048% in 2010 (MAE = 0.8957, MSE = 0.9688, R2 = 0.9638); when it comes to “aging rate”, the prevalence at the same time is estimated to be 26.6285% (MAE = 1.0036, MSE = 1.6588, R2 = 0.9448). Conclusion: The Logistic models not only display a relatively flat trend of prevalence in the future, but also have clinical significance, thus providing a reliable prediction results. Keywords: Chronic Non-communicable Diseases; Hypertension; Prevalence; Prediction; Logistic model 随着我国经济的发展和人口的老龄化，慢性非传染性疾病日趋流行[1]，心脑血管疾病、恶性肿瘤作为慢病最主要的组成部分，在近几十年间持续快速增长，对我国人民造成了极大的健康威胁。在每年约1030万各种因素导致的死亡中，慢病所占比例超过80%，而慢病的疾病负担也达到所有病种的68.6%[2]。慢性非传染性疾病已经成为我国最重要的公共卫生问题之一[3]，因此，及时准确地了解人群的慢病患病情况，对流行病的预防和治疗工作意义重大。 然而全国性的慢病数据往往需要通过大规模的抽样调查采集，而这些调查又面临着范围广、周期长等困难，实施耗费大，难以频繁进行，因此我们并不能及时掌握如今的疾病流行状况。以高血压为