The evidence for the efficacy of tinnitus intervention has been plagued by poorly designed trials without appropriate randomization and controls. Henry et al. (2016) performed a multi-site randomized control trial (RCT) to compare two methods of tinnitus intervention to two control conditions. The study was completed at four Veterans Affairs Hospital sites. The two intervention methods included tinnitus masking (TM) and tinnitus retraining therapy (TRT). The two controls included a tinnitus education (TED) group and wait-list control (WLC). The Tinnitus Handicap Inventory (THI) served as the outcome measure.
One-hundred and forty-eight subjects were randomized to the four conditions. The TM, TRT, and TED group all received ear level devices (hearing aid, sound generator, or combination device) based on candidacy and counseling structured to match TRT with respect to length. TM counseling focused on general concepts of masking and the TED on generic concepts on how we hear, causes of tinnitus, and strategies to minimize tinnitus. The TED group was considered an attention-control group to control for the Hawthorne or attention effect. TRT utilized a structured TRT approach (Jastreboff and Hazell 2004). The analysis found the effectiveness (as measured by change in THI) of the four conditions, the four sites, and time using repeated measure analysis of variance.
The subjects in the four groups did not differ significantly in regards to baseline demographics of clinical characteristics (e.g. THI, tinnitus loudness, age, previous use of amplification, etc.). At six months the results demonstrated that TM, TRT, and TED were more effective than WLC in reducing tinnitus severity. However, neither the TM or TRT was more effective than TED. Similar findings were reported at 18 months. Treatment effectiveness did not differ by treatment site.
Clinical significance for THI is considered to be a 20-point change. Similar levels of clinically significant improvement were seen for TM, TRT, and TED. However, greater report of worsening of tinnitus was seen in the TED (25 percent) and WLC (33 percent) groups compared to TM (9.5 percent) and TRT (5.8 percent).
The only baseline predictors that significantly affected THI at 18 months were baseline THI score and previous use of hearing aids. Lower initial baseline THI score (0 to 38 score) and subjects reporting the use of hearing aids were less likely to show significant improvement.
The primary author (Henry) had previously demonstrated a difference between TM and TRT (TRT showing more improved THI at 18 months) in a previous study. The rationale for lack of statistical difference in the current study was attributed to the use of multiple-sites and clinicians versus one site and one clinician where differences were reported. There was also significant dropout from the current study (29.6 percent). Those who dropped out may not have been motivated to participate.
In summary, the results suggest that TM, TRT, and basic TED are beneficial about wait-listed controls. The lack of difference between TM, TRT, and TED may have been influenced by variance in clinician experience across the four sites. In other words, clinicians with greater expertise and years of experience with various intervention methods may be more effective and influence outcomes.
Henry JA, Stewart BJ, Griest S, Kaeiln C, Zaugg TL, Carlson K (2016). Multisite randomized controlled trial to compare two methods of tinnitus intervention to two control conditions. Ear & Hearing, published ahead of print.