Adult patients with diabetes may be twice as likely to experience moderate-to-severe hearing loss compared with those without diabetes, according to a systematic review and meta-analysis.
Investigators estimated the prevalence of moderate-to-severe hearing loss among adult patients with diabetes or prediabetes and compared hearing loss odds with controls. The review was prospectively registered in the International Prospective Register of Systematic Reviews and conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The investigators searched PubMed, Scopus, Web of Science, SPORTDiscus, and CINAHL for observational studies published from January 2000 to August 2025 as well as searched gray literature and reference lists. Eligible studies included adult patients with diabetes or prediabetes who underwent objective audiometric testing. Two independent reviewers performed study selection, data extraction, and quality assessment using the Newcastle-Ottawa Scale. The investigators used random-effects meta-analyses to estimate pooled prevalence and hearing loss odds, with prespecified subgroup analyses based on study-level mean age, diabetes duration, and World Bank income classification.
Among 3,490 identified records, 29 studies met the inclusion criteria, and 27 contributed to the quantitative meta-analysis. The studies included patients with type 2 diabetes, while one included patients with prediabetes. Twenty-three studies involving 5,221 patients with diabetes contributed to the prevalence analysis, which found a pooled prevalence of moderate-to-severe hearing loss of 24%. Sensitivity analyses showed that no individual study materially changed the pooled estimate despite substantial between-study heterogeneity.
Eleven studies compared 1,881 patients with diabetes with 15,670 controls without diabetes. Those with diabetes had more than twice the odds of moderate-to-severe hearing loss compared with controls. Sensitivity analyses produced similar findings, although funnel plot asymmetry and Egger's test suggested small-study effects and possible publication bias.
Subgroup analyses found a stronger association in studies with a mean participant age younger than 60 years, in which adults with diabetes had approximately three times the odds of clinically significant hearing loss. The association was not statistically significant in studies with a mean participant age of 60 years or older. Pooled odds were numerically higher in studies conducted in low- and middle-income countries compared with in upper-middle- or high-income countries.
Patients with diabetes for less than 10 years had increased odds of moderate-to-severe hearing loss compared with adults without diabetes. Only one study evaluated patients with diabetes for at least 10 years, precluding pooled analysis of longer disease duration. Evidence regarding prediabetes remained limited because only one study included that population and did not demonstrate a statistically significant association with hearing loss.
The investigators noted several limitations, including substantial heterogeneity among the studies, the predominance of cross-sectional study designs that precluded causal inference, variation in audiometric protocols and adjustment for confounders, publication bias, evidenced by funnel plot asymmetry and Egger's test, and limited data regarding prediabetes.
"Early detection through systematic screening could facilitate timely intervention, offering a crucial opportunity to prevent progression and significantly improve communicative function and overall quality of life for individuals living with diabetes," wrote lead study author Mehwish Nisar, of the Centre for Hearing Research at The University of Queensland School of Health and Rehabilitation Sciences in Australia, and colleagues.
The study authors reported no funding and no conflicts of interest.
