Measurement Variability:  The human factor

As Audiologists, there will be times when you will encounter measurement variability and errors despite having properly functioning and calibrated equipment.  How can this occur you may ask?  In the medico-legal assessment of patients, patients who present with hearing loss are sometimes eligible for compensation of various forms.  With this added incentive, there will be some patients who will try and exaggerate their hearing loss in hopes of gaining greater compensation.  Pure tone audiometry is of course a subjective hearing assessment method, and as with all subjective means of assessment, human factors can significantly affect results and performance. 

According to Cooper et al 1999, the actual percentage of exaggerators for medico-legal claimants reported varies between studies, ranging from 8 – 30 percent. 

To obtain accurate test results from those individuals where exaggerating can be expected, a thorough medico-legal assessment needs to include both objective and subjective methods of testing.  Subjective testing is common and familiar daily assesment method. A less practiced clinical method is the Cortical electric response audiometry. CERA is a good objective test which is not affected by patient exaggeration.  As suggested by King et al 1992, PTA thresholds should only be accepted as actual thresholds and used in medico-legal calculations when they are better than or within 10 dB of CERA thresholds. 

Exaggeration Strategies:  Consistency in measurement variability

Exaggeration strategies based on loudness memory and response delay allows a person to produce seemingly consistent and accurate thresholds.  This involves that patient remembering a level of loudness and setting it to be what he excepts as the level of his threshold.  Anything above this level he will respond and any level below he will not, even though he may hear at this level.  This discrimination ability can be quite consistent and produce responses at a fix level above thresholds across a frequency range. 

Most Audiologists including myself have been trained using the Hughson-Westlake method (the BSA method for those colleagues in the UK) of beginning at 30 dB and decreasing 10 dB for responses and increasing 5 for non-responses.  An ascending method whereby inaudible tones are presented in an ascending manner have been proven to be on average 20 dB better than those using the other method (Harris, 1958; Kerr et al., 1975; Cherry and Ventry, 1976). 

A combined method (PTA_mod) whereby tone presentation begins with non-audible tones and increased by 10 dB steps can be used.  Once a response is obtained, then the 10 down 5 up process is resumed.  I have practiced this method during my clinical experience and have found it to be the default method of choice when dealing with compensation patients. 

A study by Cooper et al 1999 examined using a combination objective and subjective test methods.  The PTA_mod and PTA methods were compared against each other, with the CERA method acting as a control.  The PTA_mod method yielded results 20 dB (on average) lower than that of the conventional method. 

In summary, what Audiologist need to be aware of is that measurement error and variability does not necessarily come about as a result of miscalibrated equipment.  A good Audiologist therefore must be able to adjust his style of testing to when he expects his patient to be exaggerating his actually hearing loss.

Note:  This site is best viewed using IE 4.X or higher under 800x600 resolution