Association between occupational noise exposure and diabetes: A systematic review and meta-analysis

In addition to the well-known risk factors of diabetes, evidence is accumulating on the negative role of environmental and occupational factors such as noise exposure. We conducted a systematic review and meta-analysis on the association between long-term occupational noise exposure and diabetes.

We systematically searched evidence in PubMed, Scopus, and Web of Science (until August 2022) according to the PRISMA protocol. Risk of bias was assessed using the Newcastle-Ottawa scale. Random-effects meta-analysis was applied separately for risk ratio (odds ratio, relative risk) and hazard ratio. We evaluated the heterogeneity and publication bias. We applied meta-regressions to identify sources of heterogeneity. The overall body of evidence was assessed using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) framework.

Of 533 retrieved articles, twelve studies (11 on non-gestational, and one on gestational diabetes) on total 106,045 population (23,996 diabetic cases) met our inclusion criteria; of which eight studies were cross-sectional, three were cohorts, and one was case-control. Only 40% of papers (five out of 12) had fair, good or very good quality, and most of the papers had poor or very poor quality in terms of risk of bias. We observed a non-significant increased risk of diabetes in association with occupational noise exposure (combined risk estimates: 1.16, 95% confidence interval [CI]: 0.97: 1.34; I² = 57.7%). Doing separate meta-analyses on cohort and rest of studies, we found similar findings (cohort studies (n = 3): combined risk estimate: 1.17; 95% CI: 0.84: 1.50; I² = 79%; cross-sectional studies (n = 8): combined risk estimate: 1.26; 95% CI: 0.93: 1.58; I² = 50.4%). We found no indication of publication bias.

The overall evidence on the association between occupational noise exposure and diabetes is heterogeneous, limited, and mostly with low quality. More robust studies in terms of population selection, exposure and outcome assessment, and adjustment for confounders are necessary.