Factors Associated With the Prevalence of Hypertension Among Residents in Fujian Province: Based on the Health Ecological Model

Abstract OBJECTIVE To explore the influencing factors of adult hypertension in Fujian Province based on the health ecological model (HEM) and to provide the scientific basis for formulating intervention strategies for hypertension. METHODS From August 2020 to April 2021, a multistage stratified random sampling method was used to select 9,725 permanent residents aged 18 and above in Fujian Province. Individual information was collected through questionnaires, physical examinations, and laboratory examinations, and the variables involved in the HEM were filtrated into the 5 dimensions: personal traits, behavioral characteristics, interpersonal relationships, living and working conditions, and policy environment. Single- and multifactor logistic regression analyses were used to explore the influencing factors of hypertension among residents in Fujian Province. A visual nomogram prediction model was constructed and its prediction ability was verified and evaluated through the area under the curve (AUC) of the receiver operating characteristic curve, calibration curve, and decision curve. RESULTS The prevalence of hypertension among residents aged 18 and above in Fujian Province was 34.2%, and the age-standardized prevalence rate was 35.9%. The results of multivariable logistic regression analysis showed that in the layer of personal traits, males, older age, overweight or obesity, central obesity, family history of hypertension, diabetes, dyslipidemia, and hyperuricemia were associated with a higher risk of hypertension; in the layer of behavioral characteristics, nonphysical activity, and drinking alcohol were associated with a higher risk of hypertension; in the layer of interpersonal relationship, current residence in a rural area was associated with a higher risk of hypertension; in the layer of living and working conditions, higher education, non-farmer, and higher self-assessed economic status were associated with a lower risk of hypertension (all P < 0.05). A nomogram was constructed based on the above factors. Internal validation showed that the AUC values of the training dataset and validation dataset were 0.835 (95% confidence interval [CI]: 0.826–0.844) and 0.843 (95% CI: 0.825–0.860), respectively. The calibration curve showed that the model had good calibration, and the decision curve verified that the model had clinical applicability. CONCLUSIONS The prevalence of hypertension among residents in Fujian Province is still at a high level. Most dimensions of HEM contain factors related to the prevalence of adult hypertension. The nomogram prediction model for hypertension constructed in this study can help medical staff identify high-risk groups at an early stage and formulate targeted comprehensive prevention and treatment strategies for hypertension from a multidimensional perspective that combines personal prevention and environmental prevention.