An Evaluation of Preset Collection Size in Over-the-Counter Hearing Aids

Over-the-counter (OTC) hearing aids are available as self-fitting and preset-based devices. Preset-based devices may be more accessible to older adults with limited technological literacy, but it is unclear whether users prefer settings from larger preset collections that incorporate individual preference deviations or if simpler collections based only on prescriptive targets are sufficient. This study investigated whether potential OTC hearing aid users prefer gain-frequency configurations from larger preset collections, and whether these preferences impact speech perception in quiet and noise.

A within-subjects design was used to compare user preferences and speech perception outcomes across four hearing aid preset collections: small (4 presets), medium (16 presets), and large (32 presets) collections, as well as individualized NAL-NL2 targets. Presets were derived using NAL-NL2 real-ear insertion gain (REIG) targets for a 65 dB SPL input based on a large database of audiograms representing the OTC hearing aid market. The large (32) preset collection also incorporated empirically derived preference deviations away from prescription targets. A double-elimination tournament identified preferred settings, followed by speech testing. Forty-one older adults with mild-to-moderate sensorineural hearing loss matching OTC eligibility criteria participated. Participants listened to speech stimuli using presets delivered via a research hearing aid with eight-channel linear gain. Participants completed double-elimination style tournaments to determine their most preferred configuration in quiet and noise within each collection and overall. Preference rankings were analyzed using the Friedman and Conover tests. Speech reception thresholds were measured in quiet and noise. Linear mixed-effects models evaluated differences in speech reception thresholds, and regression analyses examined individual predictors of gain preferences and deviations from targets.

Participants significantly preferred presets from the 32-preset collection in quiet, with 49% ranking it highest and 44% ranking NAL-NL2 last. Preferences in noise showed similar trends but were more variable, and preference ranks were not statistically different. Speech reception thresholds differed slightly across collections in quiet but not in noise. Experienced hearing aid users preferred more gain and were less likely to prefer presets over NAL-NL2. When participants preferred a preset over NAL-NL2, the difference between their preferred preset and NAL-NL2 primarily reflected differences in overall gain (more or less than NAL-NL2 prescription gain), rather than differences in the spectral shape of the configuration.

Larger preset collections that incorporate user preference deviations away from prescription targets were more preferred, especially in quiet, without significant detriment to speech perception. Results suggest that users may prefer preset hearing aids with more preset options or at least the ability to adjust presets.