A Multisite Randomized Controlled Trial Comparing the Effectiveness of Two Self-Fit Methods to the Best-Practices Method of Hearing Aid Fitting.

Abstract

Trial design: This is a randomized multisite noninferiority comparative-effectiveness clinical trial with three parallel branches comparing a best-practices audiologist-fit method to two experimental self-fit (person-fit) methods. Outcomes were measured at 6 weeks and 6 months post-fit.

Method: Participants: Five hundred eighty-four participants met the following inclusion criteria and enrolled: (a) age between 50 and 79 years; (b) never used or tried hearing aids previously; (c) can speak, read, and understand English well; (d) willing to purchase the study hearing aids for $650/pair; (e) no diagnosis of a memory or cognitive impairment; (f) 25-item Hearing Handicap Inventory for the Elderly (HHIE) score > 4; (g) Montreal Cognitive Assessment score ≥ 23; and (h) not excluded due to specific audiometric criteria. The audiometric exclusion criteria, based on air-conduction pure-tone thresholds, were as follows: (a) thresholds at all frequencies (250-8000 Hz) < 20 dB HL, for both ears (no hearing loss); (b) pure-tone average for 500, 1000, 2000, and 4000 Hz (PTA4) in the better ear > 50 dB HL (greater than moderate hearing loss); and (c) interaural difference > 20 dB at three or more frequencies or ≥ 40 dB at 500 or 1000 Hz. Interventions: Three groups received the same hearing aids fitted either by an audiologist using best practices (Group AB) or by themselves using one of two efficacious self-fit methods (Groups CD and EF). Objectives: The effectiveness of each of the self-fit methods, CD and EF, was hypothesized to be noninferior to the professional-fit method, AB. Outcomes: The primary outcome measure was the global score from the Profile of Hearing Aid Benefit (PHAB), and the secondary outcome measure was the benefit score (unaided - aided) for the 25-item HHIE. Randomization: Participants were stratified into one of three hearing loss categories based on better-ear PTA4: normal (≤ 20 dB HL), mild (20.1-35 dB HL), or moderate (35.1-50 dB HL). The random assignment of sequential enrollees within each hearing loss category to the three treatment groups made use of a site-specific pregenerated randomization list produced from a random-numbers table. Blinding: It was not possible to blind the participants as to the fitting method used, but the research personnel assessing the outcomes were blinded to the treatment group.

Results: Numbers randomized: The 584 enrollees were randomized to one of the three treatment groups: AB (n = 190), CD (n = 193), and EF (n = 201).

Trial status: The trial began on November 1, 2020, and 6-month outcomes were obtained from the last participant on March 29, 2024. Numbers analyzed: At the 6-week measurement interval, outcome measures were completed for 182 AB, 172 CD, and 178 EF participants, representing 91.1% of the 584 individuals who enrolled. At the 6-month interval, completed outcomes were available for 166 AB, 148 CD, and 151 participants, representing 79.6% of the 584 original enrollees. Outcome: Bootstrapped (N = 1,000) means and 95% confidence intervals (CIs) at 6 weeks were 16.4 (95% CI [14.3, 18.7]), 16.8 (95% CI [14.9, 18.6]), and 15.2 (95% CI [13.5, 17.1]) for the AB, CD, and EF treatment groups on the primary outcome measure (PHAB global). At 6 months, primary outcome scores were 15.5 (95% CI [13.4, 17.8]), 16.3 (95% CI [14.4, 18.2]), and 15.3 (95% CI [13.5, 17.3]) for the AB, CD, and EF treatment groups, respectively. At both measurement intervals, there were no significant effects of treatment group (p > .05) for the primary or secondary outcome measures. Harms: No significant adverse events or side effects were observed.

Conclusions: At both the 6-week and 6-month postfit intervals, each of the two self-fit methods (CD, EF) yielded outcomes (primary and secondary) that were noninferior to the professional-fit best-practices (AB) fitting method. In addition, on average, clinically meaningful benefit was provided regardless of the fitting method.

Supplemental material: https://doi.org/10.23641/asha.28585145.