Summary of findings for the main comparison. Hearing aids versus no hearing aids for mild to moderate hearing loss in adults.

Hearing aids versus no hearing aids for mild to moderate hearing loss in adults
Patient or population: adults with mild to moderate hearing loss Setting: audiology services and clinics Intervention: hearing aids Comparison: no hearing aids (waiting list) or placebo hearing aids
Outcomes No. of participants (studies)	Anticipated absolute effects (95% CI)			Quality	What happens*
	Without hearing aids	With hearing aids	Difference
Hearing‐specific HRQoL assessed with: HHIE (range 0 to 100) Follow‐up: range 6 to 16 weeks No. of participants: 722 (3 RCTs)	The mean hearing‐specific HRQoL score was 39	—	Mean 26 lower (42 to 11 lower)	⊕⊕⊕⊝ MODERATE 1,2,3,4	Lower score indicates better hearing‐specific HRQoL. The mean difference corresponds to a large effect size (SMD ‐1.38, 95% CI ‐2.02 to ‐0.75) favouring hearing aids.
Health‐related QoL assessed with: WHO‐DAS II (range 0 to 100) and the SELF (range 54 to 216) Follow‐up: range 2 months to 16 weeks No. of participants: 568 (2 RCTs)	—	—	SMD 0.38 lower (0.55 lower to 0.21 lower)	⊕⊕⊕⊝ MODERATE 1,4	Lower score indicates better HRQoL. The SMD corresponds to a small effect size favouring hearing aids, which is equivalent to a 6‐point decrease (9‐ to 3‐point decrease) on the 0 to 100 scale of the WHO‐DAS II5.
Listening ability assessed with: PHAP (range 0 to 1) and APHAB (range 0 to 100) Follow‐up: 6 weeks to 2 months No. of participants: 534 (2 RCTs)	—	—	SMD 1.88 lower (3.24 lower to 0.52 lower)	⊕⊕⊕⊝ MODERATE 1,2,3,4	Lower score indicates improved listening ability. The SMD corresponds to a large effect size favouring hearing aids, which is equivalent to a 29‐point decrease (50‐ to 8‐point decrease) on the 0 to 100 scale of the APHAB6.
Adverse effect ‐ pain No. of participants: 48 (1 RCT)	Adverse effects related to pain were measured in 1 study: none were reported.			⊕⊝⊝⊝ VERY LOW7	There was too little information to estimate the risk of pain.
Adverse effect ‐ noise‐induced hearing loss No. of participants: 48 (1 RCT)	Adverse effects related to noise‐induced hearing loss were measured in 1 study: none were reported.			⊕⊝⊝⊝ VERY LOW7	There was too little information to estimate the risk of noise‐induced hearing loss.
*The equivalent change in the intervention group (and its 95% confidence interval) is based on the standard deviation in the comparison group from a representative study (see footnotes for each outcome) and the relative effect of the intervention (and its 95% CI). APHAB: Abbreviated Profile of Hearing Aid Benefit; CI: confidence interval; HHIE: Hearing Handicap Inventory for the Elderly; HRQoL: health‐related quality of life; MD: mean difference; PHAP: Profile of Hearing Aid Performance; RCT: randomised controlled trial; SELF: Self Evaluation of Life Function; SMD: standardised mean difference; WHO‐DAS II: WHO Disability Assessment Schedule II
GRADE Working Group grades of evidence High quality: We are very confident that the true effect lies close to that of the estimate of the effect Moderate quality: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low quality: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low quality: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

¹Quality of evidence downgraded by one level because all studies have either a rating of unclear and/or high risk bias in at least one of these domains: selection bias, performance and/or detection bias.
 ²We considered downgrading for inconsistency due to observed statistical heterogeneity but we did not apply this. The data consistently showed large beneficial effects of using hearing aids for mild to moderate hearing loss despite the apparent differences in study designs and populations. Our confidence in the size of the effect is not affected.
 ³We considered downgrading due to indirectness as some data were obtained after a short follow‐up period (six weeks) but we did not apply this. Large beneficial effects were observed regardless of duration of follow‐up.
 ⁴We considered downgrading due to indirectness as some analyses included data from male military veterans but we did not apply this. Effect sizes were consistent within each outcome despite differences in study samples and designs (small beneficial effect for HRQoL; large beneficial effect for hearing‐specific HRQoL and listening ability).
 ⁵Equivalent change calculated assuming a standard deviation of 15.99 in WHO‐DAS II scores in the no hearing aid group.
 ⁶Equivalent change calculated assuming a standard deviation of 15.30 in APHAB scores in the no hearing aid group.
 ⁷Very serious imprecision as the sample size was very small. There was serious indirectness because only people with mild to moderate Alzheimer's disease were included in the study.