Associations of Age, Sex, and Diabetes Duration With Sensorineural Hearing Loss Severity in Patients With Type 2 Diabetes Mellitus: A Retrospective Study

BackgroundThis study examined the unadjusted associations of age, sex, and diabetes duration with sensorineural hearing loss (SNHL) severity among patients with type 2 diabetes mellitus (T2DM) and established SNHL. Materials and methodsThis observational cross-sectional study retrospectively reviewed medical records of patients who underwent pure-tone audiometry at an otolaryngology outpatient clinic between January 2024 and December 2025.All 112 eligible patients with T2DM and SNHL who had complete study data were included.Hearing loss severity was extracted from categories documented in medical records or audiometry reports, without recalculating hearing thresholds.Severity was grouped as mild to moderate or moderately severe to profound.Age was categorized as ≤46 or >46 years, and diabetes duration as <5 or ≥5 years.Associations were assessed separately using Pearson's chi-square test, with p < 0.05 considered statistically significant. ResultsOf the 112 patients, 92 (82.1%) were aged >46 years, 66 (58.9%) were male, and 69 (61.6%) had a diabetes duration of ≥5 years.Moderate SNHL was the most frequent category, affecting 44 patients (39.3%).Mild to moderate SNHL occurred in 78 patients (69.6%), while moderately severe to profound SNHL occurred in 34 (30.4%).Pearson's chi-square tests showed that the proportion with moderately severe to profound SNHL was higher among patients aged >46 years than those aged ≤46 years (35.9% versus 5.0%; χ²(1) = 7.405, p = 0.0065), among males than females (40.9% versus 15.2%; χ²(1) = 8.463, p = 0.0036), and among patients with diabetes duration ≥5 years than those with duration <5 years (42.0%versus 11.6%; χ²(1) = 11.581,p = 0.0007). ConclusionsAge >46 years, male sex, and diabetes duration ≥5 years were associated with greater SNHL severity in this sample of patients with T2DM and established SNHL.These associations were unadjusted for potential confounders and should be confirmed in studies incorporating multivariable analysis.