Question
I’ve just started using trainable hearing aids. I understand the potential benefits for new hearing aid users, but won’t experienced users simply train the settings so that the hearing aids sound like their old hearing aids? Some of the old fittings aren’t very good, so I’m not too sure I want this to happen.
Answer
Good question. We’re certainly familiar with the patient who says: “Can’t you make these new hearing aids sound like my old ones?” In fact, after doing a project with trainable hearing aids a few years ago, we had the very same question, and so we designed a follow-up study that we thought would give us some insights on this issue. In a nut shell, here is what we did, and what we found:
We recruited experienced hearing aid users and fit them with new hearing aids that had the trainable feature; the products we used (Siemens Pure mini-BTE RICs) had “compression training” and also allowed for some training of the frequency response. We fit the participants to the NAL-NL1, and verified this fitting using speech mapping for three inputs (55, 65 and 75 dB SPL). We also conducted probe-mic measures for the participant’s personal hearing aids for their “use setting,” which we then could use as a reference.
Using a remote control, the participants then made adjustments in the real world for two weeks. To encourage different listening experiences and possible instrument adjustments, they were given a diary to complete, which included a variety of suggested listening situations and tasks.
Following the two weeks of training, the participants returned to the clinic and we again did probe-mic speech mapping to record their trained settings. So we had three sets of speech mapping data for three different input levels: the real-ear output for the NAL-NL1 prescription, the real-ear output for the trained gain, and the real-ear findings for their personal hearing aids. Recall that these instruments had compression training (not just overall gain), so training could be (and was) different for different input levels. In general, here is what we found, based on average values:
- For the 75 dB input, NAL-NL1 target gain, trained gain, and the personal hearing aid gain were all within 1 or 2 dB of each other. This didn’t help answer our question.
- For the 65 dB input, the participants previous use gain was about 5 dB below the newly programmed NAL gain. Their trained gain was essentially the same as the NAL gain, significantly higher than what they previously had been using.
- The most interesting findings were for the 55 dB input signal. This is because the gain the participants had been using for this softer speech signal was ~10 dB below the NAL-NL1 prescribed settings, which was then programmed into the hearing aids for the field trial training. The resulting trained gain, however, did not drop to what the participants had been using for several years, but rather, was only about 2 dB below the NAL-NL1 fitting, or 7-8 dB above their previous use gain.
So to answer your question, it appears that on average, the participants did not train their hearing aids back to what they were accustomed to—these instruments had +/-16 dB of training possible, so they could have reduced gain for soft inputs (e.g., at least to some extent, gain for soft could have been reduced with only minimal change to gain for average and loud). Now, the skeptic might say, maybe they didn’t train at all, which is why the final output was very similar to the programmed NAL-NL1. A reasonable point, as we do know that the starting point can influence training—we reported on this back in 2008 in a JAAA article. We don’t believe this is true, however, at least not to a large extent, as data logging showed that all participants had at least 130 gain adjustments, and individual data revealed that 45% of the participants had gain training of 3 dB or more away from the NAL fitting.
The bottom line is that after the new fitting and the two-week training, these experienced hearing aid users, who all had been previously identified as “satisfied” with their instruments, were now using considerably more gain for soft inputs; 55% of them were using 10 dB or more gain for the 55 dB speech input. This would seem to be a good thing. For a more detailed description of this research, refer to our article on AudiologyOnline on this subject.
