Background Despite increasing calls internationally for the inclusion of evidence-based decision-making (EBDM) processes in chronic disease prevention and control programming and policymaking, there is relatively sparse research assessing the current capacity of physicians and the factors influencing that capacity in China.Method A total of 892 physicians were collected from community healthcare centers (CHCs) in Shanghai, China. The experience-based chronic disease prevention (EBCDP) evaluation tool assessed physicians’ awareness, adoption, implementation and maintenance of EBCDP based on the RE-AIM framework. Linear regression analysis was used to assess associations between each EBCDP process and personal characteristics or organizational factors. Result Physicians from CHCs perceived their awareness (mean=4.90, SD=1.02) and maintenance (mean=4.71, SD=1.07) of EBCDP to be relatively low. Physicians with lower titles and monthly incomes >9,000 RMB per month tended to have relatively higher scores for the awareness, adoption, and implementation of EBCDP (P<0.05). Those who participated in one program were less likely to adopt (b=-0.284, P=0.007), implement (b=-0.292, P=0.004), and maintain (b=-0.225, P=0.025) EBCDP than those who participated in more programs. Physicians in general practice (Western medicine) had a lower level of awareness of EBCDP than those in other departments (P<0.0001). Those who were from the suburbs had lower scores regarding awareness (b=-0.150, P=0.047), implementation (b=-0.171, P=0.029), and maintenance (b=-0.237, P=0.002) compared with those from urban areas. Physicians in CHCs affiliated with universities had higher scores on all four EBCDP processes compared with those in CHCs not affiliated with a university. Conclusions This study provides evidence quantitatively illustrating the practice of EBCDP among physicians in CHCs with various personal and organizational characteristics. More solutions should be provided to increase their awareness of EBCDP to stimulate the use of EBCDP for chronic disease prevention and other public health priorities.

The prevention of hypertension in primary care requires an effective and suitable hypertension risk assessment model. The aim of this study was to develop and compare the performances of three machine learning algorithms in predicting the risk of hypertension for residents in primary care in Shanghai, China. A dataset of 40,261 subjects over the age of 35 years was extracted from Electronic Healthcare Records of 47 community health centres from 2017 to 2019 in the Pudong district of Shanghai. The XGBoost model outperformed the other two models and achieved an AUC of 0.765 in the testing set. Twenty features were selected to construct the model, including age, diabetes status, urinary protein, BMI, elderly health self-assessment, creatinine, systolic blood pressure of the upper right arm, waist circumference, smoking status, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, frequency of drinking, glucose, urea nitrogen, total cholesterol, diastolic blood pressure of the upper right arm, exercise frequency, time spent engaged in exercise, high salt consumption, and triglycerides. XGBoost outperformed random forest and logistic regression in predicting the risk of hypertension in primary care. Integration of such a risk assessment model into primary care may improve the prevention and management of hypertension in residents.