Study on risk factors of hypertension among noise-exposed workers and prediction model

Objective To explore the influencing factors of hypertension among occupationally noise-exposed workers, and to establish a risk model of predicting hypertension for individual in this specific population.

Methods A total of 4 951 male occupationally noise-exposed workers in two automobile manufacturing enterprises were studied in 2017. The prevalence of hypertension was analyzed based on occupational health examination and survey data. The influencing factors of hypertension were analyzed by lasso-logistic regression, and the predictive nomogram model was evaluated and validated by ROC curve, decision curve and bootstrap internal verification.

Results The prevalence of hypertension among 4 951 subjects averagely aged(29.9 ±4.5) years old was 6.65%. The cumulative noise exposure(CNE) level played an important role and lasso-logistic regression analysis showed that the risk of hypertension among workers with higher CNE of 90.1-95.0 dB(A)·a, 95.1-100.0 dB (A)·a and more than 100.0 dB(A)·a was 4.666, 11.810, 9.785(P < 0.01), compared with CNE of lower than 90.1 dB(A)·a, respectively. The workers with elevated hearing threshold had higher risk (1.348) compared with workers with normal hearing threshold(P < 0.01). The risk of hypertension among workers with age of 31-35 years and more than 35 years was 3.669, 7.353 (P < 0.01) compared with age of lower than 26 years, respectively (P < 0.01). The risk of hypertension among workers with hyperhemoglobin was 1.498 times higher compared with the normal group (P < 0.01).The analysis showed that other risks of hypertension of these noise-exposed workers were overweight or obesity, fatty liver, and hyperglycemia and the risks varied from 1.448 to 5.839(P < 0.05). The AUC of ROC curve of the hypertensive risk predictive nomogram model was 0.705 and bootstrap internal verification showed a C index of 0.696. Decision curve analysis showed that the intervention could be meaningful and useful while hypertension risk threshold probability was higher than 0.03 in the predictive nomogram.

Conclusion The occupationally noise-exposed workers had a higher risk of hypertension and their age, BMI and other individual also affected their blood pressure. The hypertensive risk predictive nomogram model based on lasso-logistic regression was operable.