Is There a Better “Hearing Number”?

Some thoughts on World Hearing Day

For World Hearing Day 2025, the WHO declared “We invite individuals of all ages to empower themselves to ensure healthy ears and hearing for themselves and others” including checking one’s hearing regularly. The ability to do so has become vastly more accessible thanks to broad availability of smartphone-based hearing screeners. This is especially true after Dr. Frank Lin and team at the Johns Hopkins Bloomberg School of Public Health, along with Mimi Hearing Technologies, created a public awareness campaign and a pure tone screening app called the Hearing Number, followed by Apple, working in consultation with Dr. Lin, implementing a similar screener using the popular AirPods Pro 2 earbuds and an iPhone.

Since the release of the Apple hearing test, validity of pure tone screeners and the Hearing Number have been more hotly debated than ever before. According to Dr. Lin, the Hearing Number “gives everyone a way to easily understand and think about their hearing over their lifetimes beginning as a teenager.” In this context the correlation between the Hearing Number and actual ability to understand speech in different settings can be looser, since tracking changes over time is the goal. It is worth asking if the same can be said when the Hearing Number or similar screeners are used to provide an immediate diagnosis.

Keeping it simple

A goal of consumer screeners is to provide results in the simplest possible format to reach the broadest possible audience. This is nothing new. But how simple is too simple? Pure tone audiograms have long been boiled down to a single number, the Pure Tone Average (PTA), typically using the frequencies 0.5, 1, 2 and 4 kHz, often called the PTA4. This raises two questions:

• Is the Hearing Number / PTA4 an accurate reflection of hearing difficulty in different situations?
• Do various interpretations of the PTA4 properly communicate the results?

Both are neatly encapsulated in my own audiogram from the Apple hearing test shown above.

What’s in a name?

The Hearing Number website offers rather conservative, If somewhat blurred, interpretations of PTAs. Similar explanations appear in the app.

It takes some thought to distinguish between my Hearing Number result and the one beyond it, though both fit with my actual experience. I would actually categorize myself as being in the 50-64 group. We'll get to that.

While the Hearing Number app and website offer functional descriptions of the various hearing loss levels, the PTA4 generated by other tests is often classified in the simplest possible terms in order to deliver a bite-sized summary. Consider the table above v. my Apple diagnosis. I either have "mild loss" in my right ear or "difficulty hearing conversational speech." Huh?

Studies have shown that there is a respectable correlation between PTA4 and Speech Reception Threshold (SRT), implying that pure tone averages can be useful as a consumer screening tool provided appropriate guidance is given. I would suggest that in the quest for simplicity, two very important data are often being left off consumer-level testing apps:

• Interpretation of the PTA4 beyond one-word descriptions like "mild" or "moderate."
• A statement that if one's perceived hearing ability is worse than stated, a hearing care professional should be consulted. This is needed to validate that one might experience more difficult hearing in noise that the PTA would indicate, or have other auditory processing issues.

Taking a golf lesson

Interestingly, I have read studies that point to the 4 kHz threshold as being the least correlated with SRT. One of these is Prediction of pure tone thresholds using the speech reception threshold and age in elderly individuals with hearing loss by R. Dindamrongkul et al, published in BMC Research Notes.

This leaves me wondering if the common hearing loss categories tend to understate the difficulty understanding speech with steeply sloping audiograms like mine, especially in noise. Is it possible that somehow weighting the PTA4 in consumer screeners would offer a better interpretation without resorting to inordinately complex calculations? It is both difficult to tell from the published data and beyond my own ability to determine. It is still a question worth asking.

The sport of golf confronted a similar issue. Golf has long had a handicap system to equalize scores between golfers of different abilities. However, a straight handicap system based on scores ignores the fact that the spread between two golfers depends on the difficulty of the course. On an easier course the better golfer will, on average, win by fewer strokes than they would on a more difficult course. Therefore the "slope" system was created. Weaker players receive more handicap points on difficult courses than they would on easier ones, making competitions more fun for everyone regardless of how difficult the course is.

Would the same thinking be valid in considering steeply sloping audiograms v. flatter ones? I suggest that with consumer-grade screeners specifically in mind, an analysis needs to be made to determine if a better calculation exists than straight PTA4 for giving guidance.

A call to action

The WHO frames World Hearing Day in terms of actionable items. In the same spirit I offer this call to action to make the Hearing Number and similar consumer hearing tests more relevant without sacrificing simplicity:

• Refrain from using simplistic terms like "mild" hearing loss. State the PTA along with more comprehensive but still brief interpretations and recommendations as the Hearing Number app and website do. This can be done directly or through a phrase with link such as "what does this mean?"
• Include a statement like "If you believe your hearing ability is worse than indicated here, a professional hearing evaluation is highly recommended."
• Investigate if a calculation other than a straight average of pure tones, for example PTA4 * slope rating, can more closely align the result with typical actual experience.

Postscript 5 March

Several people wrote either in the comments or privately that there are far better ways to test hearing than the pure tone test. No disagreement there. I chose to focus on it for two reasons:

• The combined weight of 苹果 and Johns Hopkins Cochlear Center for Hearing and Public Health , whose Hearing Number was codified by the Consumer Technology Association as the ANSI/CTA-2118 standard, guarantees that we will see it for years to come.
• The likelihood that companies making OTC hearing aids and music personalized earbuds will use pure tone testing for "one stop" screening and configuration.

Thus my call to make pure tone testing and results reporting as good as possible.