Interpreting audiograms

This is a huuuuuge subject area and one page on this can only hope to skim the surface of the subject, but hopefully there are enough pointers and further links here to be helpful. The idea is to give enough information so a non-specialist who may see an audiogram has enough basic knowledge to understand what it is saying.

Audiometric tests at work generate not only a result category, (the bit most employers are interested in and which the HSE tend to look at the most), but also a table or graph of the actual hearing performance at various frequencies. They often look something like this: 

 Example audiogram - the graph of an audiometry result.

Example audiogram - the graph of an audiometry result.

Even simply saying 'this is what the graph looks like' can't be taken too literally, as this is how the graph looks on one specific type of audiometer, in this case the GM Instruments ASRA audiometer. On the one above you will see two dotted slightly-diagonal lines, one black and one red. This is specific to this audiometer so ignore them for now as they don't appear on any graph from any other manufacturer.

Presentation of the graph

You may get one graph for each ear such as the one above, or sometimes both lines on one single graph, with the 'O' line always meaning the right ear and the 'X' line always meaning the left.

Sometimes where you have two separate graphs, the right ear is on the right and the left is on the left, as you may imagine makes sense. But that would be too easy as you also get some such as the one above where the left is on the right and right on the left. That's a bit of a hangover from the days of proper old-fashioned audiometers with big dials and buttons the technician could push with relish. In those tests the technician sat facing the person being tested so their right ear was indeed on their left and their left ear was on the right - see, there is some logic in it somewhere.

Similarity between the two lines

There is always a slight difference between the two ears and almost never are the results for each ear exactly the same, so don't worry about small differences in performance between the two graphs. The one above is my own result and as you can see, the right ear is just a fraction lower than the left. It's been like that for years and is perfectly fine. Although a cause cannot be confirmed from a test like this, I put it down to the three years I spent 25 years ago frequently doing front-stage security for bands. To do that I was always stood with a large speaker stack very close to my right ear, and although the noise will impact both ears almost equally, the sheer severity of the exposure did cause some issues with the right ear over time. I lost almost all the hearing in that ear for a good week or so after one particularly loud weekend. 

How the audiogram is laid out

The left hand side of the graph is the low frequencies and the right hand side is the higher ones. Kind of bass on the left and treble on the right. Almost, although it doesn't quite expand quite low enough to be considered proper 'bass'.

You get a 'O' or a 'X' at each point measured by the audiometer. 

There are two types of test and in one you will always get the 'O' or the 'X' exactly on the horizontal line as it tests hearing in 5dB steps, while with the other it tests in 1dB steps so the marks can be anywhere. Both types of test are equally valid for workplace screening audiometry.

Performance at the frequencies

When someone is young then their hearing performance tends to be fairly even across the frequency range, like this made-up and suspiciously identical graph:

Audiogram Normal hearing.png

As people start knocking on a bit, hearing starts to naturally deteriorate and it happens first at the higher frequencies, so a middle aged person will naturally have lower results, and more variation between the ears. A normal result for an older person therefore often means the higher frequencies are slightly worse performing than the lower ones.

 This is a result for an older person and although there are some losses, it is still fine for their age. 

This is a result for an older person and although there are some losses, it is still fine for their age. 

Be aware that this ageing effect means that someone in their 50s should have a lower result than someone in their 20s. This in itself is perfectly fine and nothing to worry about and is a normal part of getting on in life - the closer you get to receiving the Saga insurance adverts through the post, the worse your hearing should be. It's only when losses start to be excessive for their age that we need to worry and that's covered in the category of result and the advice of the technician.

The 'Noise Notch'

In workplace audiometric screening, accepted wisdom is that a dip at the 4kHz, or 6kHz, frequencies is indicative of noise exposure - a graph like the plough constellation. The HSE say in L108 that a Category 3 result, the calculation for which focuses on the frequencies in this notch, "suggests significant Noise Induced Hearing Loss". To see what one looks like, have a gander above at that first audiogram of mine at the very top - a classic noise notch.

This is very out-dated thinking and is these days a long way behind more recent research. The International Journal of Audiology, among other sources, have published studies which confirmed that this pattern of result where there is a notch in the higher frequencies, is just the way some people's hearing is. Almost 40% of people will have that pattern of hearing irrespective of any noise exposure. 

For workplace audiometric screening, the presence of the notch can only be used as one tool among many when looking at someone's noise exposure and cannot be used to confirm damage by noise on it's own. 

By the way, just to confuse things even more, pretty much all excessive noise exposure causes exactly the same type of notch in the result meaning it is IMPOSSIBLE to tell what the source of the noise is from an audiometric test result, as all noise exposure will cause the same result. It could indeed be workplace noise, but it could also be loud pubs or clubs, loud music in headphones, a love of gardening or DIY and a lot of time spent using loud equipment, or even no high noise exposure at all now but a nicely disreputable youth 20 years ago where the damage will still be showing. 

What I am aiming at here is that when looking at a graph of an audiometry result, a notch can be an indication of an excessive exposure, but you cannot conclusively rule it as such.

So, the notch is a guide that there may be some noise exposures going on, but it is not diagnostic of it.