Five RCTs (four unblinded randomised and one cross-over; overall quality fair) and 20 observational studies (all cross-sectional; overall quality good) were included in the review.
Screening (one RCT and 20 observational studies)
One fair quality RCT (n=2,305 participants) found that screening using the AudioScope and/or Hearing Handicap Inventory for the Elderly-Screening (HHIE-S) for hearing loss resulted in a statistically significantly greater use of hearing aids at one year (AudioScope 6.3%, HHIE-S 4.1%, combined 7.4%) compared with no screening (3.3%) (p=0.03 for between-group differences). Post-hoc stratified analysis showed that hearing aid use was greater among patients with perceived hearing loss at baseline compared with those without perceived hearing loss. There was no statistically significant difference in the proportion of patients experiencing a minimum clinically important difference in hearing-related function, measured using the Inner Effectiveness of Aural Rehabilitations scale.
Twenty studies (n=7,946 participants; range 30 to 3,471) assessed the diagnostic accuracy of tests; seven were rated as good quality and 13 as fair. Six good quality studies directly compared different screening tests. One good quality study found that the watch-tick and finger-rub tests had significantly stronger positive likelihood ratios (watch-tick test positive LR 70, 95% CI 4.4 to 1,120; finger-rub test positive LR 10, 95% CI 2.6 to 43) and similar negative likelihood ratios (watch-tick negative LR 0.57, 95% CI 0.49 to 0.66; finger-rub LR 0.75, 95% CI 0.68 to 0.84) compared with the whispered voice test (positive LR 2.3, 95% CI 1.3 to 3.8; negative LR 0.73, 95% CI 0.61 to 0.87) or a single screening question (positive LR 2.5, 95% CI 1.0 to 5.9, negative LR 0.82, 95% CI 0.68 to 0.99). There was a consistent trade-off between lower sensitivity and higher specificity for the HHIE-S compared with a single screening question (three studies), reflecting stronger positive likelihood ratios and weaker negative likelihood ratios.
Studies comparing screening tests with a control also provided evidence that screening tests are useful for identifying patients at higher risk for hearing loss (see other Publications of Related Interest for full 2011 report results).
Diagnostic odds ratios and other results were reported in the review.
Hearing loss treatment (four RCTs; n=571 participants; one good, two fair, and one poor quality)
One good quality RCT found that immediate hearing aids compared with waiting list control groups improved hearing-specific quality of life and communication difficulties in veterans up to 12 months, but no difference was reported for general quality of life. Two fair quality RCTs found no difference between treatment and control groups in function or quality of life. A fourth RCT did not report results clearly.
No evidence was identified on the adverse effects of hearing loss screening or treatment of hearing loss.