Listening-Related Fatigue in Adolescents With Prelingual Deafness and Device Experience.

Cross-sectional analysis of 271 participants aged 16–19 from the LOCHI longitudinal cohort, all with prelingual hearing loss and fitted with a hearing device by age 3. Subgroups: 136 hearing-aid users, 116 cochlear-implant users, 4 bone-anchored-device users, 15 unaided. Fatigue was measured with the Vanderbilt Fatigue Scale (10-item self-report, 12-item parent-proxy with mental and physical subscales); item-response-theory scale scores were used for analyses. Additional measures: communication difficulty (SAC/SOAC, 6 items, range 6–30), expressive language (CC-SR/CCA, z-scores), anxiety and depression (RCADS-25, t-scores), speech discrimination (Language Independent Test, percent correct), and non-verbal IQ (TONI-4). Two sets of multiple linear regression models were fitted: a pooled model (hearing-aid + implant users) for shared predictors, and device-specific models for service-related and functional factors. Multiple imputation (20 datasets) handled 0–16% missing data. Inferential statistics were limited to the hearing-aid and implant groups because the bone-anchored (n = 4) and unaided (n = 15) groups were too small.

At the group level, mean fatigue scores (parent mental, parent physical, and self-report) fell within 1 SD of the normative mean for all device groups. However, the proportion exceeding the clinical threshold (mean + 1 SD) was higher than the roughly 16% expected under a normal distribution, particularly among implant users. Using item-level 'extreme' responses (3 or more items rated 'often' or 'almost always'), the proportion was even higher, with implant users showing about 9 percentage points more than hearing-aid users in both self- and parent-report. In the pooled regression, being female was associated with higher fatigue across all subscales; additional disability raised parent-reported but not self-reported fatigue; lower socioeconomic status raised self-reported but not parent-reported fatigue. Hearing level (better-ear 4-frequency PTA) was not a significant predictor. In device-specific models, perceived listening difficulty was the strongest predictor in both groups and both reporter types (p < 0.001 in all four models). Anxiety/depression was also significant in all four models. In implant users, more frequent school accommodations were associated with higher parent-reported fatigue (mental p = 0.039, physical p = 0.004). In hearing-aid users, worse self-reported expressive language was associated with higher self-reported fatigue (p = 0.005). Device type, device use, age at fitting, and speech discrimination were not significant in any model. Parent and self-report fatigue were significantly correlated (all p < 0.001) but with substantial individual variability; of 110 flagged as high-fatigue by either informant, 68 were flagged by both, 42 by self-report only, and 42 by parent-report only.

This paper is directly about the population and outcomes an SLP manages. Three practical takeaways: (1) Do not infer that a well-fitted device and good speech-discrimination score mean the adolescent is not experiencing listening fatigue—those audiological variables were not significant predictors. Ask about fatigue directly, ideally with a listening-specific tool like the Vanderbilt Fatigue Scale rather than a generic tiredness question. (2) Get both the adolescent's and the parent's report. In this sample, 42 of 110 adolescents flagged as high-fatigue were identified by only one informant, so relying on a single source will miss roughly a third of them. (3) When an adolescent reports listening fatigue, the likely drivers are perceived communication difficulty and anxiety or low mood, not the degree of hearing loss or the type of device. Exploring those functional and emotional factors is where the clinical leverage is, not re-checking the audiogram.

Cross-sectional, single time point: fatigue was measured once, so it is unclear whether it is persistent, episodic, or tied to specific listening situations. No age-matched hearing-typical control group. The VFS normative data were derived from a broader 6–17 age range (median 14 for self-report, 11 for parent-report), making it hard to separate hearing-loss-related fatigue from normal adolescent fatigue. Bone-anchored (n = 4) and unaided (n = 15) groups are too small for any statistical inference; their results are descriptive only. Device use was a yes/no self-report; objective datalogging was available for only 111 of 136 hearing-aid users and was not used in the main analyses. Anxiety/depression, communication difficulty, and expressive language were measured by questionnaire, not by diagnostic or performance-based assessment, so the associations reflect reported symptoms rather than confirmed conditions. No measures of compensatory strategies (self-advocacy, pacing, environmental modifications), so their potential influence on fatigue is unknown. The association between more school accommodations and higher parent-reported fatigue in implant users is cross-sectional and cannot establish whether accommodations cause fatigue or mark greater underlying need.

Funded by the US National Institutes of Health (2R01DC008080-11A1) and the Australian Government Department of Health, Disability and Ageing. The authors declared no conflicts of interest.

Do not read the group mean falling within the normal range as evidence that these adolescents are not experiencing listening fatigue. The mean can mask a right-tailed distribution, and a subset exceeding the expected 16% scored above the clinical threshold. Conversely, do not read the association between more school accommodations and higher parent-reported fatigue in implant users as 'accommodations cause fatigue'; the cross-sectional design cannot establish direction, and the authors note it is more likely a marker of greater underlying need.