Occupational Noise and Hypertension Risk: A Systematic Review and Meta-Analysis

Abstract

A number of epidemiological studies report an association between occupational noise exposure and arterial hypertension. Existing systematic reviews report conflicting results, so we conducted an updated systematic review with meta-analysis. We registered the review protocol with PROSPERO (registration no.: CRD 42019147923) and searched for observational epidemiological studies in literature databases (Medline, Embase, Scopus, Web of Science). Two independent reviewers screened the titles/abstracts and full texts of the studies. Two reviewers also did the quality assessment and data extraction. Studies without adequate information on recruitment, response, or without a comparison group that was exposed to occupational noise under 80 dB(A) were excluded. The literature search yielded 4583 studies, and 58 studies were found through hand searching. Twenty-four studies were included in the review. The meta-analysis found a pooled effect size (ES) for hypertension (systolic/diastolic blood pressure ≥140/90 mmHg) due to noise exposures ≥80 dB(A) of 1.81 (95% CI 1.51-2.18). There is no substantial risk difference between men and women, but data concerning this question are limited. We found a positive dose-response-relationship: ES = 1.21 (95% CI 0.78-1.87) ≤ 80 dB(A), ES = 1.77 (95% CI 1.36-2.29) >80-≤85 dB(A), and ES = 3.50 (95% CI 1.56-7.86) >85-≤90 dB(A). We found high quality of evidence that occupational noise exposure increases the risk of hypertension.

Keywords: arterial hypertension; blood pressure; dose response relationship; meta-analysis; noise; occupation; systematic review; work.

Figure 1

PRISMA Flowchart.

Figure 2

Forest plot for noise exposure >80 dB(A) versus ≤80 dB(A) grouped according to hypertension definitions. Studies marked with * indicate that we calculated the effect size (ES) from the reported prevalences (generally this ES was unadjusted except for job-complexity: Melamed 2001; age: Ha & Kim 1991, Parameswarappa & Narayana 2015, Giordano 2001). Studies marked with ** indicate that the odds ratio was corrected to represent the prevalence ratio. † indicates that a physician diagnosis of hypertension was included in hypertension definition, and ‡ indicates that anti-hypertensive use was included in the hypertension definition.

Figure 3

Forest plot for noise exposure in the range of >80 to ≤85 dB(A) and >85 to ≤90 dB(A) using the 140/90 hypertension definition. Studies marked with * indicate that we calculated the effect size (ES) from the reported prevalence. Studies marked with ** indicate that the odds ratio was corrected to represent the prevalence ratio. † indicates that a physician diagnosis of hypertension was included in hypertension definition, and ‡ indicates that anti-hypertensive use was included in the hypertension definition.

Figure 4

Forest plot of risk per 10 dB(A) × years increase in cumulative noise exposure (CNE) among studies where the average noise and duration of employment was reported for both the exposure and comparison groups, and where hypertension was defined as blood pressure exceeding 140/90 mmHg. The effect estimates of studies marked with * indicate that effect size (ES) were calculated from the reported prevalence, and studies marked with ** indicate that the odds ratio was corrected to represent the prevalence ratio.

Figure 5

Noise exposure levels L_Aeq,Te in dB(A) and durations of exposure in years resulting in a doubling of risk.

Figure 6

Forest plot of study results stratified by risk of bias. Studies marked with * indicate that we calculated the effect size (ES) from the reported prevalence. Studies marked with ** indicate that the odds ratio was corrected to represent the prevalence ratio. † indicates that a physician diagnosis of hypertension was included in hypertension definition, and ‡ indicates that anti-hypertensive use was included in the hypertension definition.

Figure 7

Forest plot depicting study results stratified by sex. Studies marked with * indicate that we calculated the effect size (ES) from the reported prevalence. Studies marked with ** indicate that the odds ratio was corrected to represent the prevalence ratio. † indicates that a physician diagnosis of hypertension was included in hypertension definition, and ‡ indicates that anti-hypertensive use was included in the hypertension definition.

Figure 8

Funnel plot of effect estimates included in the main analysis (Figure 2).