What Is Automatic Audiometric Testing?

This is the second in a series of articles on factors affecting the quality of audiometric testing results. This post deals with automatic audiometric testing and why automatic testing is less reliable than manual audiometry by a skilled clinician.

How Do We Know Automatic Testing Was Used?

With over 10 years of experience in the industry, we have come across many instances where automatic testing procedures have been used. You can see it in the audiogram — usually a zig-zag or inconsistent pattern between frequencies or between ears, or responses consistently recorded at or below audiometric zero.

We also simply ask candidates. It is surprisingly common to hear: "the clinician walked out of the room and came back once my test was finished" or "I heard a long beep at the end." Both indicate that automatic audiometric testing was in place.

Automated pure tone audiometry is never taught or used in audiological practice — in other words, audiologists in clinics do not rely on automated results as a true indication of a person's hearing thresholds.

What the Australian Standard Says

The key clauses in AS 1269.4:2014 regarding automatic test procedures are primarily in Section 4.6:

• 4.6.1: Automatic audiometry may be used if the system meets the requirements specified in the standard.
• 4.6.2: The automatic audiometer must comply with AS/NZS 60645.1 and provide equivalent accuracy to manual audiometry.
• 4.6.3: The test environment must still comply with AS 1269.1.
• 4.6.4: The automatic system must include procedures to detect unreliable responses (e.g., false positives, inconsistent thresholds).

Clause 4.7 further requires that audiograms be reviewed for validity, that automatic systems flag questionable results, and that Clause 6.2 records must indicate whether the test was conducted manually or automatically.

Our Recommendation

Clauses 4.6.4, 4.7.1, and 4.7.2 together imply that there must be manual oversight over any audiometric test results obtained by automatic methods. We have not yet seen an automatic system that reliably detects unreliable responses or flags questionable results — so we cannot recommend it for occupational audiometric programs.

More importantly, it is impossible to meaningfully compare results from automated versus manual test procedures. We will go through the subtleties in a future post.

* The only notable exclusion is the US-based WAHTS automated audiometric screening system, which is currently excluded from audiometric testing practices in Australia but has attracted academic research interest.