Clinical Procedure Guidance.

Pure Tone Audiometry (PTA) Expectations

Feedback from students indicates there is varying PTA practice in placements and some confusion
over PTA technique.
In order for placement teams to be aware of what students are being taught in PTA practicals please
see the statement below from the PTA Practical Lead:

Starting point for masking

The masking channels of Audiometers are calibrated in effective masking (EM). The starting point of
masking is the EM level that corresponds to the threshold in the non-test ear. For example; If the
threshold of the non-test ear is 30 dB HL then the masking start point is 30 dB EM (30 dB on the
dial). An alternative exists in the protocol for situations when the audiometer masking channel has
not been calibrated to BS EN ISO 389-4 effective masking. When not calibrated to EM you find the
threshold of the masker which is M. M+10 dB is then used as the starting point equivalent to EM. We
are aware that this M and M+10 is sometimes incorrectly used. M+10 is ONLY needed when the
audiometer masking channel has not been calibrated using EM. The University of Manchester is not
aware of any current audiometry equipment, and calibration services, where the masking channel is
calibrated to something other than EM. Therefore, there is no need to ever find M and then use
M+10. We have heard of instances where EM, M and M+10 are confused, and this has resulted in
over-masking. To clarify to our knowledge ALL audiometers in the UK are now calibrated to EM.
Always start masking at the EM level that corresponds to the pure tone threshold of the non-test ear
(also see below). Do not confuse EM, M and M+10.
BSA recommended procedures require that you start masking at the EM level that corresponds to
the pure tone threshold in the non-test ear. You need a three point masking plateaux to at least
EM+30 dB. We are aware that some audiologists prefer to start masking at EM+10 and then proceed
until at least EM+30, as it would appear that the first EM test is redundant. However, we do not
agree that the first EM level is redundant. The first EM point is there for a reason and we
recommend always starting at EM as stated in the BSA protocol. Starting at EM familiarises the test
subject to simultaneously hearing the masker and tone at a masking level at which the tone should
still be audible (unless there is some central masking). The protocol requires that the not masked
threshold is first re-established at each frequency before introducing masker at EM. This is essential
to remind the test subject of what the tone sounds like. We also stress the “wait a few seconds” for
erroneous responses when first presenting the masker. Rushing ahead without being sure that the
response has come from the tone and not the masker is not in keeping with the protocol
requirements.
The old BSA procedures used to state that you can use an abbreviated threshold method for finding
the threshold of the tone in the presence of maskers. However, the new protocols do NOT state this.
The current procedures clearly require that you find masked thresholds using the same steps
required for not-masked thresholds. The protocol states that you should first use the same steps for
finding threshold without masking, including the initial familiarisation by starting at clearly audible
level. For subsequent thresholds after the masker has been increased by 10 dB the protocol states
repeat the “Re-measure hearing threshold” step. We interpret that to mean fully find threshold each
time the masker is increased. We know of cases where audiologists still apply the old protocol and
use an abbreviated threshold measurement for masked thresholds. For example; We know that
some people use the following method. Present masker. Present tone once. If no response increase
tone in 5 dB steps until a response. Take first response as threshold and carry on to the next masker
level. This is NOT establishing threshold. Threshold is clearly defined as a response at two out of two
three or four presentations. We require that threshold is properly established at each masking level,
as stated in the BSA procedures, using the two out of two three of four procedures. To clarify again;
One presentation before increasing the tone level in the presence of a masker alone is NOT
recommended, this is not threshold. Properly establishing threshold prevents over-masking and
prevents missing masking plateaus which might otherwise have been found.
The protocols now say that you need to use the “familiarisation” step starting at clearly audible level
each time you find threshold in the presence of masker. Once again to be clear, it is recommended
that you re-establish threshold for each new masker level, which includes the two out of two, three,
or four, down 10 up 5 procedures for the tone, and the familiarisation step. Abbreviated methods
could be used if needed under the general part of the protocol that permits deviations when
required. However, all deviations from the standard protocols must be noted on the audiogram and
the reasons stated.

Masking Plateaux

8.8 Procedure for masking (Step 6)

6. When three successive levels of masking yield the same tonal threshold, or one
threshold is no more than 5 dB different from the other two, this is the ‘plateau’ (see
Figures 1 and 2, and Section 8.10). The mode (i.e. the threshold which occurs 2/3) of
the three hearing threshold levels at plateau is taken as the correct hearing threshold
of the test ear and no further masking is required. Withdraw the masking noise and plot
the hearing threshold level on the audiogram.

We are aware of some confusion over which threshold to take from the masking function. The
extract, shown above, from the BSA Recommended Procedures for Audiometry 2018 clearly defines
threshold.
A masking plateau occurs when there are three thresholds within 5 dB of each other. The mode
(most frequently occurring) of the three thresholds is taken as the masked threshold.
Figure 1 depicts the simplest masking function where all three points on the plateau are identical;
clearly the threshold is 65 dB. Figure 2 depicts a common situation where the final two points of the
plateau are 5 dB higher than the first. In this case the threshold is the mode of the three points
which is 70 dB. Figure 3 demonstrates a second common situation where the final point on the
masking function is 5 dB higher than the first two points. Once again, in this case the threshold is the
mode of the three points which is 65 dB.
Figure 1 Masking Function One. Plateau has three identified thresholds:65,65,65 .Threshold is
65dB.

Figure 2 Masking Function Two. Plateau has three identified thresholds: 65, 70, 70.
Threshold is 70dB.
Figure 3 Masking Function Three. Plateau has three identified thresholds: 65, 65, 70.
Threshold is 65dB

Measuring BC thresholds
It has been identified that many departments do not plug the ear when testing not masked bone
conduction at 3 and 4 KHz. The new protocols now do not recommend testing BC thresholds at 3 or
4 kHz. So the point about plugging test ear at those frequencies has been removed.
NB: you must plot thresholds on the side which BC vibrator was placed. However, most computer
based Audiometry modules do not allow both a not-masked and masked BC threshold to be plot on
the same ear. This means that it is impossible to correctly plot not-masked BC on the worst ear when
masked BC has been performed. In such instances the workaround is to copy the not-masked BC to
the non-test ear prior to commencing masking, and then clearly note on the audiogram
comments/description line which ear the not-masked BC was performed on. This is not ideal but in
some cases there is no other solution. Moreover, the not-masked BC might be needed on the non-
test ear (the other ear from where it was measured) for the fitting and verification software to
correctly calculate targets ( Figure 6, below, is an example where this could be the case).

When to mask BC in both ears or only one ear

There is sometimes some confusion about when you need to mask BC in the better ear for AC. We
have heard of instances of the phrase “the not masked BC must have come from somewhere” being
misused as a reason to not mask BC on the better ear for AC, following masking the BC on the worst
ear for AC. Section 8.6.2 Rule 2 of the 2018 BSA Recommended Procedures for PTA clearly defines
when to mask BC on one, or both ears, without ambiguity. The section reads:
 8.6.2 Rule 2

8.6.2 Rule 2
Masking is needed at any frequency where the not-masked b-c threshold is better
than the air-conduction threshold of either ear by 10 dB or more. The worse ear (by
air-conduction) would then be the test ear and the better ear would be the non-test
ear to be masked.
Notes on Rule 2:

Although this rule may frequently indicate the need for masking, there will be
occasions where this is not warranted, depending on the purpose of the
investigation. For example it may not benefit patient management to mask more
than two b-c frequencies on one ear, or to mask small air-bone gaps.

If the b-c threshold with masking is more than 10 dB worse than the not masked b-c
threshold, then the not-masked b-c threshold can be attributed to the other (non-
test) ear. However if the b-c threshold with masking is not more than 10 dB worse
than the not masked b-c threshold then it is possible that the not-masked b-c result
was from the test ear, and it may be necessary to test the b-c thresholds of the other
ear with masking.

The BLUE text explains when masking is needed. The need for masking is rarely confused for the
worst ear. The RED text clarifies when masking is needed in the better ear for AC.

Figure 4 Situation where masked BC on the better ear for AC not needed

Figure 4 depicts a situation where the left AC is the worst ear. Not-masked and then masked BC have
been performed on the left ear. The masked BC has moved by only 5 dB, but in this instance there is
no need to perform masked BC on the opposite (RIGHT) ear. The not-masked BC is only 5 dB more
acute than the AC on the right. There is no possibility of an air-bone gap on the better (RIGHT) ear. In
this case the true BC thresholds of the left and right ear are identical (within 5 dB).
Figure 5 Situation where masked BC on better ear for AC is needed. Audiogram is incomplete, right
masked BC needed

Figure 5 depicts an incomplete audiogram where masked BC on the right is required to determine
the status of the right ear (better ear for AC) which could be a conductive, mixed, or sensorineural
loss. In this situation the masked BC performed on the left ear has moved by only 5 dB compared to
the not-masked BC. This means that the not-masked BC could still be attributed to the left ear (the
worst ear for AC), and the true (masked) BC threshold for the right ear is uncertain. In this case the
statement “the not-masked BC must have come from somewhere” applies equally to either the left
or right ear. The masked BC thresholds of the left ear firmly reveal the status of the left cochlear
alone. The not-masked BC could have come from the left ear or right ear. The right BC must now also
be masked to determine the type of loss in the right ear. This is the situation described by the final
note in 8.6.2 of the BSA procedures shown in RED text above. Note that, in this example, performing
masked BC on the right might be problematic due to cross-masking.
Figure 6 Situation where masked BC on the better ear for AC is not needed. Masked BC moved by
more than 10 dB.

Figure 6 depicts a similar situation to Figure 5. However, in this case the masked BC on the left ear
(worst ear for AC) has moved by more than 10 dB. This means that the not-masked BC measured on
the left ear “must have come from somewhere” and is attributed to the right ear. Note that the BSA
procedures require that you plot the not-masked BC on the ear it was measured and not the ear that
it is attributed to. The audiogram in Figure 6 is complete. It is up to the person reading the
audiogram to know that the not-masked BC must, in this case, be attributed to the right ear. There is
no need in this instance to mask the right BC. In actual fact attempting to mask the right BC, in this
case, is not only not required, it is also unwise as it could lead to erroneous results due to the strong
possibility of cross-masking.

General Comments
It is also important to stress that the BSA procedures do allow some flexibility from the prescribed
protocol. It might, in certain circumstances, be acceptable to use abbreviated protocols. The
protocol specifically allows this when needed. The protocol also specifically states that you may not
always need to mask all frequencies, nor mask to find small air-bone gaps depending on the reason
for the audiogram once you have enough data for the purposes required. However, you MUST
record on the audiogram any deviation from the normal procedures and the reason why this was
done. Using abbreviated methods might give different thresholds which might not be comparable to
normative data and previous tests for that subject.