Episode 23: Good Tinnitus Science, Bad Tinnitus Science

[Tinnitus Talk]

This episode is close to our hearts. We read the latest tinnitus studies and listen to the presentations at research conferences. Of specific interest are of course the studies that test the effectiveness of new treatments – could this treatment finally be what we've been waiting for?!? All too often, we're disappointed. Not just with the results but with how studies are conducted. Results are often misleading and provide false promises to patients.

During this episode, we dive deep into concepts like research design, patient selection, outcome measures, statistical analysis, and everything else required for high-quality studies. We focus on studies that assess tinnitus interventions. How do we ensure that such studies generate valuable information for patients?

We discuss these topics with Inge Stegeman, an epidemiologist from the University of Utrecht, and Jorge Simões, Assistant Professor in data science and mental health at the University of Twente.

​

Listen on:

• Tinnitus Talk Podcast (incl. timestamps and transcript)
  
• SoundCloud
  
• Apple Podcasts
  
• Spotify

We welcome any discussion, but please keep the following in mind when commenting:

• Off-topic comments, i.e. not directly responding to the content of the podcast, will be removed. So please do listen before commenting!
• While we welcome constructive criticism on ideas or policies, we do not tolerate direct attacks on individuals.

This episode was produced by @Markku and @Hazel, with the transcript created by @Hazel as well. We also want to thank Inge and Jorge for their time and wise words! Do you want to take part in creating the podcast? Let us know.

We thank our Patreon supporters for making this podcast possible.

Don't forget to:

• Send Ideas and Suggestions for Future Podcast Episodes

Sharing Is Caring
Share the podcast on your social media and with your tinnitus friends!

 

[Nick47]

There was some interesting general discussion on research quality and data collection and analysis. I agree with the sentiments expressed on UNITI, and they frankly need all funding pulled with immediate effect. What they proposed would fall short of a school leavers project/GCSE level. It's completely regressive nonsense.

We know the elephant in the room regarding collaboration and data sharing is that it is driven by money (individual researchers and businesses) and patents.

With regards to placebo controlling a trial. In tinnitus, a large placebo response is often seen. Due to this, a crossover design is often better, where the participant is their own placebo.

My biggest concern is seeing misleading scams in academia (Lenire & Neosensory) promoted by reputable charities, companies, and media outlets.

 

[2noist]

My biggest concern is seeing misleading scams in academia (Lenire & Neosensory) promoted by reputable charities, companies, and media outlets.

This is a BIG problem. Tinnitus UK and the American Tinnitus Association should NOT take money from Lenire and have them sponsor their events and website. They do mention in small print that they don't endorse Lenire but that doesn't help: it still gives credibility to Lenire in most people's minds.

Shame on these organizations. Instead of taking money from them, they should be discrediting them and highlighting the problems in their clinical trials.

 

[Nick47]

Shame on these organizations. Instead of taking money from them, they should be discrediting them and highlighting the problems in their clinical trials.

I agree completely. Remember Science Daily and their headline 'breakthrough tinnitus treatment' of some silly fucking app? And Hearing Health Foundation and Healthy Hearing promoting Neosensory and Lenire? It is criminal. These organizations have a job to critique the studies. MSM has a job to do 'investigational journalism,' not just promote horseshit to the community.

 

[UKBloke]

A big thank you to @Markku and @Hazel for their tireless effort in putting together these excellent discussions. Also, a big thank you to Inge and Jorge for agreeing to participate, especially in such an open and frank way.

My first takeaway from this episode is that it illustrates (again) the very real problem with TFI. The fact this flimsy, outmoded method remains the metric by which the success of most tinnitus trials is measured really is the elephant in the room. It's the bifurcation point between good tinnitus science and bad.

Inge spoke quite a bit about collaboration. I think a really useful one would be for clinical research teams to partner with organizations like the Bionics Institute to try and incorporate their objective tinnitus measurement system into ongoing trials. Even if these developmental systems were used as secondary metrics, partnering researchers would enjoy a quid-pro-quo, and I'm sure a lot of patients would also start to take these trials a lot more seriously.

Regarding other parts of the discussion, it was interesting to hear Jorge mention the need for people/institutions to foster a different outlook on tinnitus and also have a platform for that with Open Science feeding in. In many ways, Tinnitus Talk is already ahead of the game, and this podcast episode was a good illustration, yet my main concern with Open Science lies with the fact that we would still operate within a centralized paradigm.

My two issues with the centralized paradigm are funding (interests) and the fact that centralization begins in the schools and winds its way through the universities and evidently into "Big Pharma." In principle, that shouldn't matter; however, this model is vulnerable to a kind of institutionalization resulting in conservatism (not necessarily a bad thing) but also situations that (because of the resistance to change) bad science negatively affects good science - the problem with TFI being a fine example of that. This is where I think something like Open Source Research can help.

Open Source Research differs from Open Science in that we, the public, can also present our own theses to mainstream researchers. Of course, there can be serious issues with that - researchers drowning in a deluge of papers, for example, but this is where a filtering mechanism (a new area of Tinnitus Talk perhaps?) can be employed; "trusted partnerships," so to speak.

The key aspect of Open-Source Research is that patients are not beholden to the restrictions a centralized model might place upon them. We also don't have professional reputations to uphold, so in many ways, this allows us a little more creative freedom at times.

With all that said, the most interesting takeaway from this discussion is that it seems we all agree that the scope of tinnitus research needs to change. Of course, change as a term by itself is rather nebulous, so I suppose the question is, can we pinpoint where the change we desire might occur?

I've long believed that from an engineering perspective, it's ludicrous to try and solve a problem that can't be properly defined. Therefore the kind of change I would like to see happen is for researchers to go back to basics and understand the underlying mechanism of tinnitus, or like Jorge says, we must try and understand what tinnitus is and how it manifests in the brain. This, to me, sounds like common sense, and I'm pretty certain that common sense in that respect will eventually translate into good science.