The ears are the organs of both hearing and balance. Together, they facilitate communication with others, warn us of impending danger, and help bipeds maintain their equilibrium. Hearing loss often occurs as individuals age, are exposed to loud sounds, or succumb to certain diseases. It can be caused by a problem in any of the three ear sections.

Otolaryngologists and audiologists use audiometers and other instruments, such as tympanometers, to determine if there is a hearing loss, quantify the loss, aid in diagnosis, and measure improvement over the course of treatment. Audiometers generate sounds at specific pitches (frequencies) and sound pressures (levels) to measure individual threshold hearing levels, the quietest sound an individual can hear 50% of the time. The results are shown on an audiogram in graphic or tabular forms, or both (see Figure 1). Next, either the audiologist or the audiometer itself compares the individuals' measured levels to a standard range of normal values of the population at large, a previously obtained baseline audiogram (the first audiogram of an individual, taken as a basis for comparison to determine later hearing loss), or both. A clinical determination of follow-up care is made if there appears to be a significant hearing loss.

Figure 1.

A typical example of an audiogram that shows the current readings in both the chart and graphically, the baseline readings in the chart, and the difference between the current audiogram and the baseline audiogram.

Figure 1.

A typical example of an audiogram that shows the current readings in both the chart and graphically, the baseline readings in the chart, and the difference between the current audiogram and the baseline audiogram.

Close modal

Audiometers can be divided into two general categories: screening and clinical. Screening audiometers perform rapid, semi-automated determinations of threshold hearing levels. They are used to quickly determine whether or not there is a measurable hearing loss. Some are capable of testing multiple individuals simultaneously and are used in schools for mass screening programs and in industrial operations to periodically test the workforce. Some screening audiometers adjust the results for age.

While screening audiometers can do an excellent job of assessing the degree of hearing loss, they do not determine the reason for hearing loss. That is the where the clinical (or diagnostic) audiometer enters the picture. Clinical audiometers, along with noise exposure history, are used to definitively determine the cause of a previously identified hearing loss. Although they are more operator-intensive to use, clinical audiometers provide more flexibility for the examiner in developing a diagnosis. Clinical audiometers also allow the use of special testing protocols on infants and young children.

All legacy audiometers are comprised of four sections—the oscillator, amplifier, attenuator, and headphones. The oscillator produces the audio frequencies generated during the examination. The amplifier increases the sound level of the output of the oscillator. The attenuator controls output of the amplifier and provides a precise sound level. The headphones provide the precise sound level to each of the test subject's ears. Microprocessor and computer-controlled audiometers perform the same functions, but the sections are less discernible since the process is software controlled. During the hearing test, a tone is repeatedly presented at an increased sound level (volume) until the person acknowledges hearing the sound. The lowest sound level at which the test subject hears the tone is called the threshold. The oscillator is changed to the next frequency being tested and this process is repeated with the new tone.

Screening (also called Bekesy) audiometers follow this standard pattern to perform their assessment of the test subject. Generally, they present the test subject with an amplified pure (single frequency sinusoidal wave) tone in one ear through a set of headphones. The headphone's diaphragm vibrates the air around it, and these vibrations are felt by the eardrum. The volume of the tone increases until the subject hears the tone and affirmatively responds, usually by depressing a hand switch. This tone can either be continuous or pulsed, with a pulsed tone being preferred for testing children and individuals with tinnitus. Sometimes white or pink noise is provided in the headphone of the non-tested ear to mask sound conducted to the non-tested ear through the skull. The audiometer records the sound level of the tone, then goes to the next tone and the sequence is repeated. After readings for one ear are completed, the audiometer tests the other ear following the same pattern. Standard screening frequencies are 500, 1,000, 2,000, 3,000, 4,000, 6,000, and 8,000 hertz (Hz). Sound intensity begins at 0 decibels (dB) (see sidebar) and increases in 5 dB increments. Negative values are not tested since they would represent above-average sensitivity.

To perform these tests automatically, screening audiometers rely on microprocessor control of the testing steps and sequence. Standalone instruments have this circuitry built into the device, while screening audiometers, capable of simultaneously testing multiple individuals, rely on a standalone personal computer (PC) and custom software to perform the tests. The basic design of multi-subject audiometers centers on a single PC that controls tone generator modules (one per test subject) connected to the PC and provides headphone outputs and hand switch input. These instruments are often used in large industrial hygiene/hearing conservation programs at manufacturing plants.

Once a screening audiometer identifies a hearing loss, or if the individual is otherwise referred, the audiologist employs a clinical audiometer to assist in providing additional insight into the individual's hearing problems.

Most functions of a clinical audiometer are manually performed based upon the audiologist's exam protocol and results of related tests. Clinical audiometers can perform a number of specialized tests. They allow the examiner to send pure tones to either the headphones, a bone-conduction vibrator that is placed on the forehead or the mastoid process, or a pair of high-quality loudspeakers located in the testing environment. Additionally, clinical audiometers generally provide a wider range of frequencies to test the upper and lower limits of the test subject's hearing, and even allow the examiner to use spoken words introduced through a microphone or playback device (phonograph, tape recorder, CD, etc.) to perform specialized speech tests.

Most clinical audiometry is performed through air-conduction headphones, which tests the outer, middle, and inner ear, and the nerves connecting the ear to the brain. Bone-conduction tests only assess the function of the inner ear and nerves. Comparing the measured hearing losses from both tests provides valuable diagnostic information to the audiologist. Two tests in particular, the speech-reception test (SRT) and the speech-discrimination test (SDT), measure the subject's ability to distinguish one syllable words at a high sound intensity, a problem for patients with severe inner ear diseases. Where headphones are impractical, for example, in young children and in cases of severe physical injury in adults, free-field testing is performed using loudspeakers, with the test subject placed between them. Free-field testing employs warble tones (modulated at a specific rate), narrow-band noise (several equal intensity frequencies mixed together), or spoken material, depending on the test protocol being followed by the audiologist.

The Origin and Evolution of the Device

The earliest device comparable to the audiometer as we know it today was invented by David Hughes, the inventor of the microphone, and first introduced by Dr. B.W. Richardson to a meeting of the Royal Society in London in 1879. This device produced a controllable sound when connected to a telephone. The examining physician could vary the intensity of the sound by moving an induction coil over a bar that was graduated from 200 (full volume) to 0 (completely silent). This provided the first objective measurement of a patient's hearing ability. In the centuries prior to this invention, physicians made subjective determinations of a patient's hearing ability by making subtle noises, snapping their fingers, and clapping their hands. Even though it was the first practical device to objectively assess hearing ability, it offered little to the medical community since much was unknown about the causes of hearing loss. At that time, the only hearing aid was a large metal horn as often seen in old movies.

The 20th century history of audiometers is even fuzzier. Different sources credit Dr. Harvey Fletcher in 1922, Ralph Allison in 1935, and Leland Watson in 1937 of developing the modern-day instrument and coining the term “audiometer.” What is sure is that Watson founded Maico Diagnostics in 1937. Maico produced the first audiometer with a zero reference level, and it was the first audiometer accepted by the American Medical Association (AMA). Along the pathway to today's computer-controlled device was Dr. Georg von Békésy's 1947 concept and design of an adaptive automatic audiometer using clicking noises. Later, Dr. Wayne Rudmose developed the RA-101, an audiometer that allowed the test subject to find his own hearing threshold and used specially designed earphones called “Otocups” which were individually calibrated for each frequency. In 1955, he demonstrated the first fully automated screening audiometer, what was to become the first in the ARJ-series, using a card (about the size of a computer punch card) lying on a table above which moved a stylus whose movement was controlled by the test subject. At the beginning of the test, the table moved at a constant speed (driven by a long rotating screw) and the stylus moved perpendicularly to the table. At the same time, the first sound was produced, getting louder as the stylus moved across the table. When the test subject heard the tone, he would press a button. The button press both reversed the direction of the stylus and caused the audiometer to change frequency. The test continued until its conclusion at the highest frequency being tested, at which time a medical technician removed the card, replaced it with a fresh one, reset the table position, and performed the test on another subject. This was the forerunner of the modern screening audiometer, capable of producing an accurate audiogram without the constant involvement of the audio technician.

Incremental improvements continued in both screening and diagnostic audiometers as they migrated from the mixed electromechanical-electronic design of the ARJ-series to totally electronic. Today, screening audiometers are either microprocessor- or PC-controlled, but the basic elements (oscillator, amplifier, attenuator, and transducer) are still present. In Dr. Rudmose's original test protocol, which is still followed today, the sound level increases at a fixed rate until the test subject hears the sound and presses a button.

In parallel with the increase in medical knowledge about the mechanics of hearing, the functions of the outer, middle, and inner ear, the acoustic nerves, and the brain's auditory processing centers, diagnostic audiometers have become more useful in diagnosing hearing losses and deafness. Although modern audiometers no longer utilize motor-driven attenuators, the basic principle of test subject-controlled testing is still widely used in screening audiometers.

Due to the level of importance placed on hearing conservation by the U.S. Occupational Safety and Health Administration (OSHA), audiometers should be individually managed, by serial number and/or unique maintenance management control number, by the supporting maintenance organization. The maintenance management database must document both scheduled and remedial maintenance services. Calibration must be scheduled at least annually and performed any time a repair affects either the frequency produced or the output level.

Audiometers, because of their importance in hearing conservation programs, are one of the few medical devices that have the interest of both federal and state regulators. Additionally, 29 CFR 1910.95 (OSHA's Occupational noise exposure regulation) incorporates American National Standards Institute (ANSI) S3.6-1969 by reference in subparagraph 1910.95(h)(2).

The OSHA regulation not only specifies the definition and core requirements of a workplace hearing conservation program, it also contains specifications for audiometers and the test environment. The areas important to biomedical equipment technicians (BMET) include requirements for record keeping, annual audiometer calibrations, the calibration procedure followed, and equipment used in the calibration, including its calibration history. These are contained in Appendix E of 29 CFR 1910.95.


ECRI Healthcare Product Comparison System for Audiometers & Audio Booths

ANSI Standard S3.6-2004, American National Standard Specification for Audiometers

OSHA Standard 29 CFR 1910.95, Occupational Safety and Health Standards, Occupational Health and Environmental Control, Occupational Noise Exposure.

Risk mitigation relies heavily on the audiologist operating the audiometer properly and performing timely operator maintenance. Operator-level maintenance responsibilities include the replacement of headphone cushions when they are no longer soft and pliable or if they contain visible cracks. Additionally, the examiner should ensure the correct headphones are connected to the audiometer, the left and right earphone plugs are connected to the appropriate jacks and are not reversed, and that the headphones are properly placed on the test subjects' ears.

From a financial perspective, risk management also includes the financial liability and potential impact of violating OSHA standards and mandated procedures. Not only is the employer subject to direct fines from federal and state agencies for violations of OSHA standards, the violations extend the time frame for potential hearing loss claims and lawsuits decades into the future.

Modern day audiometers are remarkably stable and reliable and utilize proven technology with no known identified systemic weaknesses. The BMET must, however, ensure the instrument is recalibrated after any maintenance affecting the frequencies produced or the sound level of the outputs. If a headphone is replaced, calibration is also required since the audiometer and headphones combine to form a test system.

A common problem for the examiner is the test subject reversing the headphones so that each earphone is on the wrong ear. The convention is blue earphone on the left ear and red earphone on the right ear. Remember, match ‘R’ (red) and ‘R’ (right). The problem for the BMET occurs in calibration. If audiometers are calibrated in-house, care must be taken to adjust the left side when taking frequency and output readings from the blue headphone and vice versa. One fine point on audiometer calibration: The audiometer and its headphones constitute a system and headphones cannot simply be interchanged between audiometers without recalibration. For this reason, the headphone often carries the same serial number as the audiometer with which it was calibrated.

Common shop tools and test equipment can be used to service most audiometers. The circuitry of the tone generator portion is relatively straightforward with or without microprocessor or computer control. Good service literature is required if in-house maintenance will be the standard of practice.

Calibration must be traceable to the National Institute of Standards and Technology (NIST) to comply with OSHA requirements. This is especially important for screening audiometers since one of their primary purposes is to uncover occupationally related hearing losses that can lead to workers' compensation claims. Although the calibration process itself is not particularly complex, many organizations prefer to contract calibration to other organizations or the audiometer manufacturer's local representative or dealer. Special purpose, dedicated test equipment (sound level meter, acoustic coupler, calibrated weights) is required for audiometers and it is somewhat expensive. Unless the biomedical maintenance shop supports a number of audiometers, outsourcing the calibration is often the most cost-effective solution.

Current audiometer designs are considered mature technology. Microprocessor controls have automated much of the test procedure as well as the necessary calculations, providing virtually instantaneous results in both tabular and graphic forms. Personal computers controlling eight or more exam stations allow speedy individual testing of large groups as well as the capability to store the results on both the internal hard drive and on compact disks. One current developmental effort being pursued is tele-audiology—Internet-based audiology via computer—to support patients who are hospitalized or for other reasons cannot travel to an audiologist. Another developmental effort is the application of active noise reduction to attenuate background noise at test locations lacking an audio booth, such as in mobile testing applications in elder care locations and business settings.

What's a Decibel?

The decibel (dB) is a unit used in medicine, audio electronics, communications, and other mediums to measure many things and its precise meaning changes with the medium. Unlike the volt, ampere, ohm, etc., a decibel does not represent a finite unit of measurement, but rather a ratio between a reference and a measured value. If the reference and measured value are equal, it is expressed as “zero decibels” or 0 dB. Values below and above the reference are expressed as minus or plus values such as −3 dB or +5 dB.

The reference value varies with the industry. For example, in some commercial audio applications, the reference is one volt root mean square (RMS) and the level is written as “dBV.” In other audio applications involving telephone lines, the reference is one milliwatt (1 mw) through a 600 ohm (Ω) load and is often expressed as “dBm”, alluding to the one milliwatt reference. In communications, some designs cause an antenna to project more power in one direction than another. This is called “gain” which is the ratio of power radiated in one direction verses another and is expressed as “dBi” for that antenna. In audiology and audiometers in particular, the reference (or 0 dB) is the average threshold response of a group of 18-to-25-year-olds with no history of otologic disease. In terms of pressure on the eardrum, this corresponds to approximately 20 micropascals or 2.9 × 10−9 pounds per square inch (psi), a miniscule amount of pressure.

The human ear responds to a surprisingly wide range of sound pressure levels. The loudest sound that does not cause permanent hearing loss is about 20,000 times louder than the quietest sound undamaged ears can hear. Because decibels use a logarithmic scale, they can conveniently represent this large range of sound pressure level ratios. For example, if the sound pressure level is twice as loud as the reference, the difference is 3 dB; if it is 10 times louder, the difference is 10 dB; and if it is a million times louder, it is 60 dB. Oddly enough, our hearing seems to work logarithmically (not linearly) since a sound that we perceive to be twice as loud has, curiously enough, a 10 dB higher sound level.

As complex and confusing as decibels may be, the three key points to remember are:

  • The decibel represents a ratio between a measured value and the reference.

  • The reference varies with the medium.

  • Decibels are measured on a logarithmic scale thus allowing large ratios to be expressed as reasonably sized numbers.

Author notes

Robert Dondelinger, CBET-E, MS, is the senior medical logistician at the U.S. Military Entrance Processing Command in North Chicago, IL. Email: