A composite risk score to identify older adults at high risk of hearing loss in a community screening program

Background Age-related hearing loss (HL) is highly prevalent among older adults, yet it often goes undetected and untreated. Routine screening in community settings is not widespread, and hearing aid uptake remains very low. We aimed to construct a composite risk score to identify individuals at high risk of HL for targeted audiometric screening. Methods We conducted a cross-sectional study using data from a community-based health screening program in Shenzhen, China. Participants underwent pure-tone audiometry at 500-8000 Hz to determine hearing thresholds. Moderate or greater HL was defined as a pure-tone average (PTA) >= 35dB in the better ear. Stepwise multivariable regression was used to identify predictors of HL, which were then used to develop a cumulative Hearing Risk Score (HRS). Results A total of 2,490 adults (mean age, 67.5 years, SD 5.8 years) were included; 32.5% (810 participants) had moderate or greater hearing loss. Of 22 risk factors included in the stepwise regression model, seven were identified: self-reported hearing difficulty, age 65 years or older, male sex, social isolation, cardiovascular disease, and metabolic disease. These were incorporated into the HRS, with total scores ranging from 1 to 23. The HRS was strongly associated with moderate or greater hearing loss, with adjusted odds ratios increasing from 4.50 (95% confidence interval (CI), 1.57-12.88) for a score of 1 to 39.11 (13.50-113.33) for a score of 6 or more (P for trend < 0.001). Similar dose-response patterns were observed at all frequencies tested (0.5 to 8 kHz). Conclusions The HRS showed a clear dose-response relationship with HL and may serve as a practical tool to target older adults for confirmatory audiologic evaluation.