Although a causal relationship exists between military occupational noise exposure and hearing loss, researchers have struggled to identify and/or characterize specific operational noise exposures that produce measurable changes in hearing function shortly following an exposure. Growing evidence suggests that current standards for noise-exposure limits are not good predictors of true hearing damage. In this study, the aim was to capture the dose-response relationship during military rifle training exercises for noise exposure and hearing threshold. To capture exposure, a wearable system capable of measuring impulse noise simultaneously on-body and in-ear, behind hearing protection was used. To characterize hearing threshold changes, portable audiometry was employed within 2 h before and after exposure. The median 8-h time-weighted, protected, free-field equivalent in-ear exposure was 87.5 dBA at one site and 80.7 dBA at a second site. A significant dose-response correlation between in-ear noise exposure and postexposure hearing threshold changes across our population ( R = 0.40 , p = 0.0281) was observed. The results demonstrate an approach for establishing damage risk criteria (DRC) for in-ear, protected measurements based on hearing threshold changes. While an in-ear DRC does not currently exist, it may be critical for predicting the risk of injury for noise environments where protection is mandatory and fit status can vary.

Noise-induced hearing loss (NIHL) continues to be among the most common permanent injuries suffered by military personnel, resulting in reduced operational effectiveness and impaired quality of life (Tepe , 2017). Although a causal relationship exists between military noise exposure and hearing loss (HL), Department of Defense (DoD) researchers have struggled to identify and/or characterize specific operational noise exposures that produce measurable changes in hearing function over a short (within a 24 h) time-frame (Brungart , 2019). Military and civilian standards and hearing conservation programs often focus on permanent threshold shifts (PTS) when trying to protect military personnel from NIHL. However, new research shows that having previously experienced temporary threshold shifts (TTSs) predicts subjective hearing difficulties in the present (Brungart , 2019). Despite this, TTS is not well understood (Hecht , 2019), indicating the need for an improved understanding of the causes of TTS to prevent it.

Design criteria for military noise exposure in the U.S. are advised by MIL-STD-1474E (U.S. Dept. of Defense, 2015). No consensus exists, however, on which damage risk metrics should be used to predict auditory risk (Murphy and Kardous, 2012; Price , 2017; Zagadou , 2016), leading to multiple methods being supported by the standard: auditory risk units (ARUs) and the time-weighted average of the A-weighted sound pressure energy ( L Aeq 8 hr dBA). ARUs are computed using the Auditory Hazard Assessment Algorithm for Humans (AHAAH) model (Price and Kalb, 1991). The AHAAH model treats the ear as a circuit to estimate the energy that reaches the cochlea, taking into account reflexes that activate when an exposure is anticipated by the ear. For occupational exposures that occur more than one time per week, the limit shall not exceed 200 ARUs per 24-h sliding window period. L IAeq 100 ms is an “equal energy” model that characterizes total energy over an impulse calculated for 100 ms. Damage risk criteria (DRC) and occupational exposure limits are most often defined based on the A-weighted energy, including NIOSH and OSHA (NIOSH, 1998). We note that these standards often specify free-field or on-body measurements. For further review, see Nakashima and Farinaccio (2015).

The goal of auditory DRC is typically to protect the individual or population from PTS and does not explicitly address TTSs. Historical work has suggested that a limit of 25 dB change in hearing thresholds at any frequency would recover and no permanent damage to the cochlear hair cells would result (Melnick, 1991). However, the relationship between TTS and permanent damage remains difficult to predict (Ryan , 2016). A growing body of evidence suggests that current standards for noise-exposure limits are not good predictors of damage to the auditory system. For example, a TTS that fully recovers with no permanent threshold shift may still result in permanent damage to the cochlear synapses (Kujawa and Liberman, 2009). Vulnerability varies by species (for review, see Trevino and Lobarinas, 2022) and is not directly dependent on the amount of TTS (Fernandez , 2020; Fernandez , 2015). It is unknown what that threshold might be for humans and if 25 dB (or 500 ARU for the AHAAH model; Price and Kalb, 2018) is the correct limit. This synaptic damage can also occur as a result of either continuous noise or blast-induced exposure (Hickman , 2018; Tepe , 2017). Other challenges to current DRC include the equal energy hypothesis, which assumes that only the total amount of sound pressure energy matters and not the timing or impulsivity of the exposure. In a recent study by Qiu (2020), accumulated evidence from a study of thousands of factory workers suggests that current noise metrics underpredict the risk of hearing loss due to complex noise exposures. These studies emphasize the need to establish objective measures of damage or injury based on exposure and corresponding auditory function data.

Further complicating the prediction of risk in many military and industrial environments is accounting for the attenuation provided by hearing protection devices (HPDs). The fit of the hearing protection can vary over time, and the degree of compliance might be the largest factor for prediction of permanent injury risk (Federman and Duhon, 2016; Voix and Hager, 2009). Today, it is now possible to measure the personal attenuation rating (PAR) in the field through microphone in real ear (MIRE) or behaviorally, through the real-ear attenuation test, or possibly even visually (Smalt , 2021) pre- and postexposure. This measurement of the PAR in dB can be accounted for in the risk prediction through a derating procedure (Berger, 1996) or an acoustic model of the hearing protector attenuation (Kalb, 2010). Alternatively, in-ear dosimetry, as performed in this experiment, has the benefit of continuous measurements of the PAR, which can result in the ability to report the in-ear dose (Davis , 2019). While there are some challenges to in-ear dosimetry, including wearer-disturbances and knocking artifacts, an in-ear exposure standard or guidelines might improve hearing conservation efforts and reduce noise-induced hearing injury (NIHI; Bonnet , 2015).

In this study, we propose in-ear dosimetry, paired with prospective portable field audiometry pre- and postexposure, as a means to develop and evaluate DRC. Integration of auditory function tests along with enhanced noise-dosimetry data in tactical and operational environments are necessary to inform the refinement and individualization of noise-exposure standards to better protect service personnel (Hecht , 2019; Kulinski , 2022). We hypothesize that protected in-ear exposure measurements are more likely to correlate with injury than free-field measurements because the protected measurements account for the variable levels of hearing protection that are achieved in noisy occupational environments. We also suggest this methodology as an alternative to more controlled human laboratory studies in which the amount of exposure must be limited due to ethical concerns (Theis , 2012). Here, we seek to demonstrate a methodology to gather dose and response data in the field during service members' normal duties. Critically, this study took efforts to minimize disturbance to regular work duties of military personnel while still gathering an adequate quantity of data. We first describe the design of the study for characterizing the dose-response between in-ear exposure and postexposure hearing threshold change. We then review our results for this relationship and provide some guidelines for future studies.

Field measurements were collected at two military rifle ranges: Camp Pendleton Weapons and Field Training Battalion Edson Range (“Camp Pendleton”) and Marine Corps Air Station Miramar Hathcock Range (“Air Station Miramar”). A total of 27 range safety instructors volunteered as participants in the study. The range safety instructor role is to observe students who are shooting munitions. The instructors do not shoot any munitions themselves. The participants wore noise exposure monitors during their normal duties for one or two days and completed one or more hearing tests before and after the measured noise exposure. To facilitate field noise-exposure measurements, we used a prototype wearable system capable of measuring impulse noise simultaneously on-body (up to 173 dBP) and in-ear (up to 154 dBP) suitable for military occupational environments (Davis , 2019; Smalt , 2017). On-body measurements capture the noise environment for an individual during military training exercises that involve free-movement. In-ear measurements allow for exposure assessment in environments where hearing protection is routinely worn and can allow for the incorporation of hearing protection fit status. To characterize temporary changes in hearing, we performed testing with a portable automated audiometry system, the Creare Hearing Assessment (CHA) system (Creare LLC, Hanover, NH), as close as possible before and after training exercises. All of the audiograms were automated, but a trained technician was present during testing.

Measurements of noise exposure were obtained in a manner similar to those of our previous study (Davis , 2019). Figure 1 shows the in-ear and on-body devices used to measure noise exposure for all of the participants. The TASCAM DR-100MKIII audio recorder (TEAC, Montebello, CA) recorded the signals from the in-ear and on-body devices. It was selected due to its low cost, high dynamic range (110 dB), high sampling rate (96 kHz), and phantom power, which is capable of driving two external high-sound pressure level (SPL) microphones for several hours. Audio inputs to the recorder were provided by (1) an in-ear microphone (discreetTM 4062, DPA Microphones, Alleroed, Denmark) set in a custom-designed housing that couples to a military-grade foam hearing protector (ComplyTM Canal Tip, St. Paul, MN); and (2) an on-body GRAS 47DX 1/8″ pressure microphone (GRAS Sound and Vibration A/S, Holte Denmark) clipped to the shoulder toward the subject's back as shown in Fig. 1. This location for the on-body microphone was chosen to prevent the dosimeter's cords from interfering with the participants' motions. In general, we tried to place the microphone as close as possible to the top of the shoulder such that it was facing up. Because the 47DX is a pressure microphone, it means that incoming pressure waveforms at 0 deg elevation (from any azimuth) angle would not need any correction. The in-ear microphone was attached to a foam hearing protector that served as the participant's hearing protection in the ear in which the in-ear microphone was worn. The participant's regular in-ear hearing protection was worn in the other ear. Both microphones were selected for wideband-widths and high-SPL measurements with an in-ear upper limit of 154 dB and an on-body upper limit of 173 dB.

FIG. 1.

(Color online) On-body and in-ear dosimeters each connected to the TASCAM DR-100MKIII audio recorder. The in-ear microphone's upper limit is 154 dB, and the on-body microphone's upper limit is 173 dB.

FIG. 1.

(Color online) On-body and in-ear dosimeters each connected to the TASCAM DR-100MKIII audio recorder. The in-ear microphone's upper limit is 154 dB, and the on-body microphone's upper limit is 173 dB.

Close modal

To characterize the noise-exposure levels associated with each participant, we calculate three standard metrics for the in-ear and on-body exposures: (1) an 8-h equivalent of the noise energy, termed L Aeq 8 hr; (2) ARU from the AHAAH model; and (3) the peak SPL. Because noise-exposure measurements were collected inside the ear canal (“in-ear,” behind hearing protection) and on the body (“on-body,” outside of the hearing protection), this results in six distinct definitions of exposure to analzye against observed threshold shifts.

Acoustic measurements from the in-ear dosimeter were processed offline to produce a free-field equivalent exposure. This in-ear to free-field transformation was performed for two reasons: first, exposure standards are defined around free-field values, thus, by calculating a free-field equivalent, the results can be compared against existing standards. Second, using a free-field equivalent allows calculation of the PAR value by allowing direct comparison between the free-field equivalent of the in-ear exposure level and the on-body exposure level.

The transfer function that makes this conversion was derived from the average ratio of frequency responses between free-field 150 dB peak SPL pressure waveforms and the corresponding unoccluded eardrum pressure waveforms as measured from an acoustic test fixture. For a full description of this process, see Davis (2019). All in-ear exposure values reported in this paper are free-field equivalents.

Artifacts can occur in the acoustic measurements and are caused by rapid acceleration of one of the microphones, such as from readjusting a sensor or other equipment. These artifacts can appear as high sound levels despite not contributing to the individual's auditory exposure. Since artifacts are likely to inflate the noise exposure totals, we have taken steps to identify and remove them from the data. This was achieved by conducting consistency checks between the in-ear and on-body data as described previously (Davis , 2019).

Audiograms were administrated using the automated CHA system. The system consists of a hand-held unit containing custom circuitry and firmware to produce calibrated tones and execute audimetric tests. The CHA was connected to a Sennheiser HDA-200 headset (Wedemark, Germany) for the Hughson-Westlake (H-W) and Bekesy tests performed in this study. It was calibrated using a Bruel and Kjaer type 4153 flat plate coupler (Naerum, Denmark) according to ANSI S3.6 (2004). Data were saved to comma separated text files and exported to matlab for analysis. The CHA system has been used in many previous studies (Buckey , 2015; Rieke , 2017; Zhan , 2018). We estimate that the test-retest reliability of our system at the frequencies tested (4 kHz and 6 kHz) is at most 4.4 dB for H-W audiometry measurements and at most 3.3 dB for Bekesy audiometry measurements. We make this estimate because the measurement setup and population are similar to recent studies (Kulinski 2022; Meinke 2017). Although the current study uses the handheld CHA and the referenced studies the Wireless Automatic Hearting Test System (WAHTS; Creare, Hanover, NH), both device setups use the same algorithm. All of the audiograms in this study were automated.

This study was approved by the Navy Medicine Readiness and Training Command San Diego (NMRTC SD) Institutional Review Board, and participants provided written informed consent to take part in the study. Participant information can be found in Table I. Study participants belonged to two cohorts. The seven Pendleton participants were stationed at Camp Pendleton and were range safety instructors for recruits firing M4 rifles. Some of the participants had already suffered what appears to be PTS at 4 and 6 kHz prior to the start of the current study as evidenced by starting thresholds up to 60.5 dB HL. Their participation in the study spanned three days: On day 1, they received a “pre-exposure” H-W audiogram. Day 1 was chosen such that the participants had not experienced vocational exposures on the previous four days. Days 2 and 3 were exposure days during which the participants performed their normal duties while wearing the the in-ear and on-body measurement microphones. Median in-ear exposure ( L Aeq 8 hr) of the Pendleton participants on the exposure days was 87.5 dBA. At the conclusion of the normal work duties on the second exposure day (day 3), the participants were given a “postexposure” H-W audiogram. Post- versus pre-exposure threshold shifts were calculated by taking the difference between the postexposure and pre-exposure thresholds for each participant at 4 and 6 kHz and reporting the maximum threshold shifts at those two frequencies. We chose 4 and 6 kHz because there is evidence from other recent studies in DoD personnel that this might be the range that is most likely to see short term threshold changes (Kulinski , 2022). Since audiograms were taken for the left and right ears, threshold shifts for each ear were treated as independent data-points in the dose-response analysis. No follow-up data were available for the Pendleton participants to determine whether the threshold shifts observed for these participants were recovered after a quiet rest-period. Because the study participants were range safety instructors, they were not shooting rifles, only observing students shooting. This means that there was no head-shadow effect from their own weapon, making it likely that with similar hearing protection insertion, both ears received similar levels of exposure. In this paper, “dose” is a generic term referring to the amount of acoustic exposure as measured and calculated in a variety of methods. “Response” generically refers to the change or potential change in hearing thresholds as can be measured and calculated in multiple ways.

TABLE I.

The study cohorts, Camp Pendleton and Air Station Miramar, in which audiograms for Pendleton participants were collected for the left (L) and right (R) ears. Audiograms for Miramar participants were collected in the right ear only. Ranges are represented as median [min,max].

Cohort Pendleton participants Miramar participants
Location  Camp Pendleton  Air Station Miramar 
Number of participants  7 (L and R ears)  20 (R ear) 
Ages during study (approximate)  26 [22,32]  21 [20,29] 
Genders  7 male, 0 female  19 male, 1 female 
Job roles  Range safety instructor  Range safety instructor 
Pre-exposure thresholds at 4 kHz (dB HL)  10.5 [−4.5,60.5]  11.7 [1.3,25.7] 
Pre-exposure thresholds at 6 kHz (dB HL)  8 [3,43]  20.8 [8,37] 
Daily in-ear exposure ( LAeq 8 h r FFeq, dBA)  87.5 [78.4,102.3]  80.7 [72.2,91.3] 
Cohort Pendleton participants Miramar participants
Location  Camp Pendleton  Air Station Miramar 
Number of participants  7 (L and R ears)  20 (R ear) 
Ages during study (approximate)  26 [22,32]  21 [20,29] 
Genders  7 male, 0 female  19 male, 1 female 
Job roles  Range safety instructor  Range safety instructor 
Pre-exposure thresholds at 4 kHz (dB HL)  10.5 [−4.5,60.5]  11.7 [1.3,25.7] 
Pre-exposure thresholds at 6 kHz (dB HL)  8 [3,43]  20.8 [8,37] 
Daily in-ear exposure ( LAeq 8 h r FFeq, dBA)  87.5 [78.4,102.3]  80.7 [72.2,91.3] 

The 20 Miramar participants were stationed at Air Station Miramar and were range safety instructors for Marines completing annual weapons qualification for pistols or rifles. The Miramar participants' participation in the study spanned 2 days. On the morning of day 1, participants received a pre-exposure Bekesy audiogram using the CHA. Day 1 was chosen such that the participants had not experienced vocational exposures on the previous 1–4 days, depending on the range schedule. Following the pre-exposure audiogram, participants performed their normal duties on the shooting range. Median in-ear exposure ( L Aeq 8 hr) of the Miramar participants was 80.7 dBA. This exposure level is lower than that for the Pendleton participants. One reason for this is that the Miramar range is smaller than the range at Pendleton with fewer Marines firing and fewer total rounds fired. The ammunition type was also different between the two ranges. When the participants had finished their range duties for the day, they received a postexposure Bekesy audiogram using the CHA system. The threshold shift was calculated in the same way as it was calculated for the Pendleton participants. The next morning, day 2 of the study, the participants received a “follow-up exposure” Bekesy audiogram using the CHA. This audiogram was used to determine whether threshold shifts were recovered. If recovered, the shift is verified as a TTS. If not, it is possible that the subject would recover over the following weeks or the shift could become permanent. As audiograms were taken on only the right ear for the Miramar participants, each Miramar data-point represents a single participant. The choice to test only the right ear was due to time constraints imposed by the need not to interfere with normal work duties. The in-ear dosimeter was in the right ear for all of the subjects, meaning that the in-ear exposure measurement was in the same ear as the ear that was subjected to audiograms. In this study, we chose to combine the Pendleton and Miramar cohorts in the dose-response analysis to strengthen the results with a larger sample size than would be possible with just one cohort. In addition, we chose to present all valid data that were acquired, including participants where only in-ear or on-body recordings were successfully captured with the prototype noise measurement equipment.

Figure 2 shows a Global Positioning System (GPS) time-lapse map of exposures for two of the Pendleton participants over a 6-h period. Each dot represents an impulse measured on the in-ear device. Levels are represented as free-field equivalents of the in-ear measurements in dB. Dots on the hotter end of the color spectrum (red to black) represent louder measurements than those at the cooler end (green to yellow). At Camp Pendleton, recruits need to qualify at 200, 300, and 500 yards from the target, which corresponds to the clusters of data-points of the instructors at those locations. Additional data-points are visible from when the participants were walking between the qualification distances. The two participants in Fig. 2 had very different in-ear exposures. Since they were exposed to similar stimuli, the contrast is likely due to differences in the effectiveness of their hearing protection (which are captured in our in-ear free-field equivalent exposure metric, LAeq 8 h r FFeq). Participant 3 [Fig. 2(top)] had a peak SPL of 136 dB and an L Aeq 8 hr of 83 dBA, which are both below the DoD limit. Participant 7 [Fig. 2(bottom)] had a peak SPL of 160 dB and an L Aeq 8 hr of 99 dBA, both of which are above the DoD limit.

FIG. 2.

(Color online) A time-lapse map of in-ear peak-impulse levels for two Pendleton participants. Each dot represents an impulse measured on the in-ear device. Participant 3 (top) had a peak SPL of 136 dB and an L Aeq 8 hr of 83 dBA. Participant 7 (bottom) had a peak SPL of 160 dB and an L Aeq 8 hr of 99 dBA.

FIG. 2.

(Color online) A time-lapse map of in-ear peak-impulse levels for two Pendleton participants. Each dot represents an impulse measured on the in-ear device. Participant 3 (top) had a peak SPL of 136 dB and an L Aeq 8 hr of 83 dBA. Participant 7 (bottom) had a peak SPL of 160 dB and an L Aeq 8 hr of 99 dBA.

Close modal

Figure 3 shows exposure versus time for a representative participant from the Miramar cohort on the exposure day. The top two panels of Fig. 3 show the on-body and in-ear levels recorded for this participant over the course of the day. The in-ear levels are represented as the free-field equivalent values rather than the raw values measured from the in-ear microphone. On-body with HPD model values are also included as are artifacts that were removed from the dataset. The difference between on-body and in-ear levels for each detected impulse represents an instantaneous PAR, which is plotted on the bottom panel of Fig. 3. Unlike a normal PAR value, which is measured as a spot-check, the PAR values shown here are calculated for each impulse. This demonstrates a new way to evaluate PAR for impulsive noise that more accurately reflects the real-world values. While this participant consistently had PAR values between 20 and 40 dB, there were moments when PAR was near zero, including the last half hour of the exposure period when the individual deliberately removed the microphone from their ear, effectively removing any hearing protection.

FIG. 3.

(Color online) The exposure metrics for a representative Miramar participant over the course of a day of occupational exposure. The top panel shows exposure represented by the AHAAH metric. The middle panel shows the same exposure represented as LAeq evaluated over 100 ms intervals. The bottom panel shows a real-time PAR value calculated as the difference between the on-body and in-ear free-field equivalent exposures.

FIG. 3.

(Color online) The exposure metrics for a representative Miramar participant over the course of a day of occupational exposure. The top panel shows exposure represented by the AHAAH metric. The middle panel shows the same exposure represented as LAeq evaluated over 100 ms intervals. The bottom panel shows a real-time PAR value calculated as the difference between the on-body and in-ear free-field equivalent exposures.

Close modal

A histogram of total in-ear and on-body exposure for all participants is shown in Fig. 4. The mean on-body L Aeq 8 hr is 100 dBA with a standard deviation of 4.1 dBA. The mean in-ear exposure was 84 dBA with a standard deviation of 7.0 dBA. The in-ear exposure is weighted toward lower exposure levels, indicating successful use of hearing protection for most participants.

FIG. 4.

(Color online) In-ear and on-body exposures for all of the participants as measured by the in-ear and on-body microphones. The in-ear microphones also serves as hearing protection, resulting in the in-ear dose being lower compared to the on-body dose.

FIG. 4.

(Color online) In-ear and on-body exposures for all of the participants as measured by the in-ear and on-body microphones. The in-ear microphones also serves as hearing protection, resulting in the in-ear dose being lower compared to the on-body dose.

Close modal

Figure 5 shows the dose-response results for all of the participants (i.e., ears) for whom we could calculate a valid post- versus pre-exposure threshold shift and who had valid in-ear and/or on-body exposure data. The y axes in Fig. 5 represent the response (post- versus pre-exposure threshold shifts). The x axes represent the dose (noise exposure expressed as L Aeq 8 hr, AHAAH ARU or peak SPL exposure as measured either in-ear or on-body). Black data-points represent unique ears where the in-ear dosimeter was used (all Miramar data-points and five of the Pendleton data-points). The red data-points on the in-ear dose plots represent the Pendleton participant ears that did not contain the in-ear dosimeter but where we have a valid measurement of hearing pre-/postexposure. For these data-points, the in-ear dosimeter measurement from their other ear was used for the x axis position. We opted to use all of the ears with valid audiograms in our analysis, largely due to the limited quantity of data. We calculated the Spearman correlation for the data in the six subplots to assess if a monotonic relationship existed between the dose and response as defined for each subplot. The results are shown in Table II, where a significant Spearman monotonic correlation of R = 0.40 , p = 0.0281 was observed.

FIG. 5.

(Color online) Dose-response correlations for various damage risk metrics, measured in-ear (left column) and on-body (right column). Each data-point represents a single ear from one of the participants. The overall highest correlation observed between dose and response was using the in-ear LAeq 8 FFeq noise metric. The red data-points are the ears without direct measurement of in-ear exposure (the in-ear dosimeter was in the other ear).

FIG. 5.

(Color online) Dose-response correlations for various damage risk metrics, measured in-ear (left column) and on-body (right column). Each data-point represents a single ear from one of the participants. The overall highest correlation observed between dose and response was using the in-ear LAeq 8 FFeq noise metric. The red data-points are the ears without direct measurement of in-ear exposure (the in-ear dosimeter was in the other ear).

Close modal
TABLE II.

The dose-response correlations for various damage risk metrics shown as R coefficients and p-values. A (*) indicates significance p <0.05. The overall highest correlation between dose and response was achieved using in-ear noise exposure data and the L Aeq 8 hr noise metric.

Exposure metric RSpearman p-value
L Aeq 8 hr in-ear  0.40  0.0281* 
L Aeq 8 hr on-body  0.09  0.6536 
AHAAH unwarned in-ear (log)  0.31  0.0918 
AHAAH unwarned on-body (log)  0.33  0.0952 
Peak level in-ear  0.19  0.3136 
Peak level on-body  −0.24  0.2265 
Exposure metric RSpearman p-value
L Aeq 8 hr in-ear  0.40  0.0281* 
L Aeq 8 hr on-body  0.09  0.6536 
AHAAH unwarned in-ear (log)  0.31  0.0918 
AHAAH unwarned on-body (log)  0.33  0.0952 
Peak level in-ear  0.19  0.3136 
Peak level on-body  −0.24  0.2265 

To assess whether changes to the audiometric thresholds are truly temporary or resulted in permanent threshold loss, follow-up testing is needed to evaluate the recovery. In our data collections, follow-up audiograms were only conducted for the Miramar population (the Pendleton population only had pre- and postexposure audiograms). Follow-up audiograms for the Miramar population were collected the morning after the postexposure audiogram. Figure 6 shows the changes in hearing threshold between post- and pre-exposure Fig. 6(left) and follow-up and pre-exposure Fig. 6(right). While a slight (nonsignificant) linear dose-response relationship may be present between dose and response for the post- versus pre-exposure threshold shifts at these relatively low exposure levels [Fig. 6(left), R 2 = 0.07, p =0.14), no such relationship is apparent between the follow-up versus pre-exposure threshold shift [Fig. 6(right), R 2 = 0.05, p =0.80]. These results indicate that the (often-slight) threshold shifts observed for the Miramar participants were temporary and did not persist as permanent shifts.

FIG. 6.

(Color online) Miramar participants' dose-responses where response (y axis) is the threshold shift calculated between post- and pre-exposure (left) and follow-up and pre-exposure (right). The threshold shifts are calculated from the maximum of the shifts at 4 and 6 kHz.

FIG. 6.

(Color online) Miramar participants' dose-responses where response (y axis) is the threshold shift calculated between post- and pre-exposure (left) and follow-up and pre-exposure (right). The threshold shifts are calculated from the maximum of the shifts at 4 and 6 kHz.

Close modal

A significant monotonic dose-response correlation was observed in our range safety instructor population when dose was defined as L Aeq 8 hr as measured in-ear and the response was defined as the maximum threshold shift between post- and pre-exposure at 4 and 6 kHz (Fig. 5). The temporary nature of the shift was confirmed for one of the two study cohorts through follow-up measurements the next day, which suggest recovery. Based on the available data, we cannot say if the relationship is linear througout the full range of exposure levels. Instead, it appears more likely that above a certain exposure threshold, the risk of a large threshold shift dramatically increases. In this dataset, there was only one participant who experienced a threshold shift of more than 10 dB. This significant response occurred for one of two ears of the participant whose measured in-ear exposure was greater than 100 dBA L Aeq 8 hr. Based on further investigation of the H-W adaptive tracking data for that participant, our conclusion is that this participant may be an outlier in terms of threshold change and exposure, but we do not suspect it is due to invalid or noisy data. This result, where the postexposure hearing threshold change can be significant but rare, is consistent with a recent study that also observed outliers with significant TTS (Kulinski , 2022).

We also observe from the data that the in-ear dose measurements provide more differentiation between the subjects' exposures compared to the on-body dose measurements, which is similar to our previous work (Davis , 2019). To take L Aeq 8 hr as an example, the range of observed exposures was 16.5  for on-body measurements but 30 dBA for in-ear measurements. This greater individual variability (perhaps due to hearing protection fit) likely contributes to stronger correlation between dose and response for the in-ear doses compared to the on-body doses. This strengthened correlation holds even with the inclusion of the non-instrumented ears, indicating that future studies that measure in-ear dose can draw stronger conclusions for the threshold shifts in both ears than studies that measure dose on-body. In the case of the subject with the 40 dB threshold shift, their non-instrumented ear had a negligible threshold shift. It is likely that the non-instrumented ear did not experience the same exposure as the instrumented ear (possibly the in-ear dosimeter was improperly inserted while their regular hearing protection in the non-instrumented ear was inserted correctly). However, even with this ear included in the dataset, the monotonic relationship between dose and response is still significant.

In general, there are several advantages of using dosimeter-based data to develop DRC. Zhao (2010) recently considered a kurtosis adjustment to the L Aeq 8 hr damage risk metric using an on-body dosimeter. Their study focused on long-term permanent changes to hearing and involved hundreds of participants. While historical studies of impulse noise have used more controlled environments (Kryter , 1966; Ward, 1968), the finding that permanent damage to the cochlear synapses may persist even if “temporary” threshold shifts recover (Kujawa and Liberman, 2009) makes future such studies unlikely. Therefore, we suggest that dosimeter data can be used in situations where exposure will occur regardless, such as during military operations. These results are a first step in observing this relationship in a field, and suggest a path-forward for validating and developing new DRC.

Our approach demonstrates a viable methodology for evaluating DRC for in-ear protected measurements based on postexposure hearing threshold changes. While an in-ear DRC does not currently exist, it may be critical for predicting the risk for noise environments where protection is mandatory and fit status can largely affect risk. It also takes into account variability of hearing protection fit across individuals, so risk can be characterized more accurately. Important applications of these integrated tests include improved prediction of NIHI, informing the effective design of HPDs, and identifying activities and environments associated with high risk of NIHI.

In our study, we confirm previous finding that the overall peak level of the exposure is less correlated with threshold shifts than energy-based DRC (AHAAH, A-weighted energy). This finding is consistent with historical animal (Murphy and Kardous, 2012) and human DRC research. As more of this type of data is taken, more fine-grained model comparisons can be made as well as supporting data driven approaches (Zhao , 2019). Our results are in contrast to Rabinowitz (2013), which did not find any dose-response relationship, but that study only tracked noise exposures less than 85 dBA (mean LA eq = 76 dBA ).

One caveat to our finding is relatively small changes in audiometric threshold. Our study found that a significant postexposure hearing threshold change, i.e., 15 dB at any frequency is rare. It is still unknown if 85 dBA is a sufficient time-weighted average daily (8-h) exposure for in-ear measurements for a study like this. We suspect it is not. Negative changes in the TTS (potential improvement) are not excluded from our analysis, and the minimum significant threshold shift and variability in audiometric measurements can complicate the interpretation. Additional data are needed to overcome this challenge and allow for the evaluation across the rare events where large threshold shifts occur.

A focus of the current study was to start establishing the methods to conduct a TTS-focused dose-response study in the field. This included fielding custom devices to monitor in-ear and on-body doses. Future studies can build on this technology development and therein, we would recommend increasing the number of participants and ensuring that all of the participants had a pre-exposure, postexposure, and follow-up audiogram. One challenge of such a field study is to conduct it without excessively interfering with normal work duties. Since TTSs are heavily time dependent (Hecht , 2019), we would recommend obtaining more frequent audiograms or other measures of hearing throughout the exposure day to better analyze the time-dependence of TTSs. In addition, minimizing and tracking the time between the exposure and follow-up tests would likely improve sensitivity to detecting smaller TTSs from the relatively lower exposure levels in this study. Given potential damage to the cochlear synapses or other parts of the auditory pathway, we would also recommend complementing the audiograms with additional objective measures, such as otoacoustic emissions (OAEs), speech-in-noise tests, and binaural hearing (Brungart , 2022; Le Prell and Brungart, 2016). In addition, it would be beneficial to collect future measurements from a community that expects greater noise exposures than those found in this study, as well as a community that is exposed to continuous noise or a combination of continuous and impulse noise.

This material is based on work supported by the Department of the Navy under Air Force Contract No. FA8702-15-D-0001. The views expressed in this paper are those of the authors and do not necessarily reflect the official policy or position of the U.S. Department of the Navy, U.S. Department of Defense, nor the U.S. Government. The authors acknowledge support for this work from the Office of Naval Research under FY21 Military Interdepartmental Purchase Request No. N0001420MP00152. The study protocol was approved by the Navy Medicine Readiness and Training Command San Diego (NMRTC SD) Institutional Review Board in compliance with all applicable Federal regulations governing the protection of human subjects. Research data derived from approved NMRTC SD Institutional Review Board protocol Nos. NMCSD.2013.0090 and NMCSD.2016.0042. Commander S.W. was a military service member. This work was prepared as part of his official duties. Title 17, United States Code (U.S.C.) Sec. 105 provides that copyright protection under this title is not available for any work of the U.S. Government. Title 17, U.S.C., Sec. 101 defines a U.S. Government work as a work prepared by a military service member or employee of the U.S. Government as part of that person's official duties.

1.
American National Standards Institute (
2004
). ANSI-S3.6-2004, Specification for Audiometers (
American National Standards Institute
,
New York
).
2.
Berger
,
E.
(
1996
). “
Scientific basis of noise-induced hearing loss
,” in
International Review of Field Studies of Hearing Protector Attenuation
(
Thieme
,
New York
), pp.
361
377
.
3.
Bonnet
,
F.
,
Voix
,
J.
, and
Nélisse
,
H.
(
2015
). “
The opportunities and challenges of in-ear noise dosimetry
,”
Can. Acoust.
43
(
3
),
80
81
.
4.
Brungart
,
D. S.
,
Barrett
,
M. E.
,
Schurman
,
J.
,
Sheffield
,
B.
,
Ramos
,
L.
,
Martorana
,
R.
, and
Galloza
,
H.
(
2019
). “
Relationship between subjective reports of temporary threshold shift and the prevalence of hearing problems in military personnel
,”
Trends Hear.
23
,
2331216519872601
.
5.
Brungart
,
D. S.
,
Sherlock
,
L. P.
,
Kuchinsky
,
S. E.
,
Perry
,
T. T.
,
Bieber
,
R. E.
,
Grant
,
K. W.
, and
Bernstein
,
J. G.
(
2022
). “
Assessment methods for determining small changes in hearing performance over time
,”
J. Acoust. Soc. Am.
151
(
6
),
3866
3885
.
6.
Buckey
,
J. C.
,
Fellows
,
A. M.
,
Clavier
,
O. H.
,
Allen
,
L. V.
,
Brooks
,
C. A.
,
Norris
,
J. A.
,
Gui
,
J.
, and
Meinke
,
D. K.
(
2015
). “
DPOAE level mapping for detecting noise-induced cochlear damage from short-duration music exposures
,”
Noise Health
17
(
78
),
263
.
7.
Davis
,
S. K.
,
Calamia
,
P. T.
,
Murphy
,
W. J.
, and
Smalt
,
C. J.
(
2019
). “
In-ear and on-body measurements of impulse-noise exposure
,”
Int. J. Audiol.
58
(
sup1
),
S49
S57
.
8.
Federman
,
J.
, and
Duhon
,
C.
(
2016
). “
The viability of hearing protection device fit-testing at navy and marine corps accession points
,”
Noise Health
18
(
85
),
303
311
.
9.
Fernandez
,
K. A.
,
Guo
,
D.
,
Micucci
,
S.
,
De Gruttola
,
V.
,
Liberman
,
M. C.
, and
Kujawa
,
S. G.
(
2020
). “
Noise-induced cochlear synaptopathy with and without sensory cell loss
,”
Neuroscience
427
,
43
57
.
10.
Fernandez
,
K. A.
,
Jeffers
,
P. W.
,
Lall
,
K.
,
Liberman
,
M. C.
, and
Kujawa
,
S. G.
(
2015
). “
Aging after noise exposure: Acceleration of cochlear synaptopathy in ‘recovered’ ears
,”
J. Neurosci.
35
(
19
),
7509
7520
.
11.
Hecht
,
Q. A.
,
Hammill
,
T. L.
,
Calamia
,
P. T.
,
Smalt
,
C. J.
, and
Brungart
,
D. S.
(
2019
). “
Characterization of acute hearing changes in united states military populations
,”
J. Acoust. Soc. Am.
146
(
5
),
3839
3848
.
12.
Hickman
,
T.
,
Smalt
,
C.
,
Bobrow
,
J.
,
Quatieri
,
T.
, and
Liberman
,
M.
(
2018
). “
Blast-induced cochlear synaptopathy in chinchillas
,”
Sci. Rep.
8
(
1
),
10740
.
13.
Kalb
,
J.
(
2010
). “
A hearing protector model for predicting impulsive noise hazard
,”
J. Acoust. Soc. Am.
127
(
3
),
1879
.
14.
Kryter
,
K. D.
,
Ward
,
W. D.
,
Miller
,
J. D.
, and
Eldredge
,
D. H.
(
1966
). “
Hazardous exposure to intermittent and steady-state noise
,”
J. Acoust. Soc. Am.
39
(
3
),
451
464
.
15.
Kujawa
,
S. G.
, and
Liberman
,
M. C.
(
2009
). “
Adding insult to injury: Cochlear nerve degeneration after ‘temporary’ noise-induced hearing loss
,”
J. Neurosci.
29
(
45
),
14077
14085
.
16.
Kulinski
,
D.
,
Dirks
,
C.
,
Carr
,
W.
,
Sheffield
,
B.
,
Kamimori
,
G.
, and
Brungart
,
D. S.
(
2022
). “
Field assessment of acute auditory responses to environmental exposures in close quarters tactics training
,”
Int. J. Audiol.
(published online).
17.
Le Prell
,
C. G.
, and
Brungart
,
D. S.
(
2016
). “
Speech-in-noise tests and supra-threshold auditory evoked potentials as metrics for noise damage and clinical trial outcome measures
,”
Otol. Neurotol.
37
(
8
),
e295
e302
.
18.
Meinke
,
D. K.
,
Norris
,
J. A.
,
Flynn
,
B. P.
, and
Clavier
,
O. H.
(
2017
). “
Going wireless and booth-less for hearing testing in industry
,”
Int. J. Audiol.
56
(
sup1
),
41
51
.
19.
Melnick
,
W.
(
1991
). “
Human temporary threshold shift (TTS) and damage risk
,”
J. Acoust. Soc. Am.
90
(
1
),
147
154
.
20.
Murphy
,
W. J.
, and
Kardous
,
C. A.
(
2012
). “
A case for using a-weighted equivalent energy as a damage risk criterion
,”
EPHB Report No. 350-11a
,
National Institute for Occupational Safety and Health
, Washington,
DC
.
21.
Nakashima
,
A.
, and
Farinaccio
,
R.
(
2015
). “
Review of weapon noise measurement and damage risk criteria: Considerations for auditory protection and performance
,”
Mil. Med.
180
(
4
),
402
408
.
22.
NIOSH
(
1998
). “
Criteria for a recommended standard: Occupational noise exposure
,”
Technical Report NIOSH Pub. No. 98-126
.
23.
Price
,
G. R.
, and
Kalb
,
J. T.
(
1991
). “
Insights into hazard from intense impulses from a mathematical model of the ear
,”
J. Acoust. Soc. Am.
90
(
1
),
219
227
.
24.
Price
,
G. R.
, and
Kalb
,
J. T.
(
2018
). “
The philosophy, theoretical bases, and implementation of the ahaah model for evaluation of hazard from exposure to intense sounds
,”
U.S. Army Research Laboratory, Adelphi, MD
, Technical Report No. ARL-TR-8333.
25.
Price
,
G. R.
,
Kalb
,
J. T.
, and
Jokel
,
C. R.
(
2017
). “
Critical examination of the article impulse noise injury prediction based on the cochlear energy by Zagadou, Chan, Ho, and Shelly
,”
U.S. Army Research Laboratory, Adelphi, MD
, Technical Report No. ARL-TR-7958.
26.
Qiu
,
W.
,
Murphy
,
W.
, and
Suter
,
A.
(
2020
). “
Kurtosis: A new tool for noise analysis
,”
Acoust. Today
16
,
39
47
.
27.
Rabinowitz
,
P. M.
,
Galusha
,
D.
,
Dixon-Ernst
,
C.
,
Clougherty
,
J. E.
, and
Neitzel
,
R. L.
(
2013
). “
The dose-response relationship between in-ear occupational noise exposure and hearing loss
,”
Occup. Environ. Med.
70
(
10
),
716
721
.
28.
Rieke
,
C. C.
,
Clavier
,
O. H.
,
Allen
,
L. V.
,
Anderson
,
A. P.
,
Brooks
,
C. A.
,
Fellows
,
A. M.
,
Brungart
,
D. S.
, and
Buckey
,
J. C.
(
2017
). “
Fixed-level frequency threshold testing for ototoxicity monitoring
,”
Ear Hear.
38
(
6
),
e369
e375
.
29.
Ryan
,
A. F.
,
Kujawa
,
S. G.
,
Hammill
,
T.
,
Le Prell
,
C.
, and
Kil
,
J.
(
2016
). “
Temporary and permanent noise-induced threshold shifts: A review of basic and clinical observations
,”
Otol. Neurotol.
37
(
8
),
e271
e275
.
30.
Smalt
,
C. J.
,
Ciccarelli
,
G. A.
,
Rodriguez
,
A. R.
, and
Murphy
,
W. J.
(
2021
). “
A deep neural-network classifier for photograph-based estimation of hearing protection attenuation and fit
,”
J. Acoust. Soc. Am.
150
(
2
),
1067
1075
.
31.
Smalt
,
C. J.
,
Lacirignola
,
J.
,
Davis
,
S. K.
,
Calamia
,
P. T.
, and
Collins
,
P. P.
(
2017
). “
Noise dosimetry for tactical environments
,”
Hear. Res.
349
,
42
54
.
32.
Tepe
,
V.
,
Smalt
,
C.
,
Nelson
,
J.
,
Quatieri
,
T.
, and
Pitts
,
K.
(
2017
). “
Hidden hearing injury: The emerging science and military relevance of cochlear synaptopathy
,”
Mil. Med.
182
(
9-10
),
e1785
e1795
.
33.
Theis
,
M. A.
,
Gallagher
,
H. L.
,
McKinley
,
R. L.
, and
Bjorn
,
V. S.
(
2012
). “
Hearing protection with integrated in-ear dosimetry: A noise dose study
,” in
Noise Control and Acoustics Division Conference
(
American Society of Mechanical Engineers
,
New York
), Vol.
45325
, pp.
237
241
.
34.
Trevino
,
M.
, and
Lobarinas
,
E.
(
2022
). “
Current topics in hearing research: Deafferentation and threshold independent hearing loss
,”
Hear. Res.
419
,
108408
.
35.
U.S. Dept. of Defense
(
2015
). “
MIL-STD-1474E: DOD Design Criteria Standard; Noise Limits
” (
Washington, DC
).
36.
Voix
,
J.
, and
Hager
,
L. D.
(
2009
). “
Individual fit testing of hearing protection devices
,”
Int. J. Occupational Safety Ergonom.
15
(
2
),
211
219
.
37.
Ward
,
W. D.
(
1968
). “
Proposed damage-risk criterion for impulse noise (gunfire)
,”
National Academy of Sciences – National Research Council
, Washington, DC, Technical Report No. AD 673–223.
38.
Zagadou
,
B.
,
Chan
,
P.
,
Ho
,
K.
, and
Shelley
,
D.
(
2016
). “
Impulse noise injury prediction based on the cochlear energy
,”
Hear. Res.
342
,
23
38
.
39.
Zhan
,
Y.
,
Fellows
,
A. M.
,
Qi
,
T.
,
Clavier
,
O. H.
,
Soli
,
S. D.
,
Shi
,
X.
,
Gui
,
J.
,
Shi
,
Y.
, and
Buckey
,
J. C.
(
2018
). “
Speech in noise perception as a marker of cognitive impairment in HIV infection
,”
Ear Hear.
39
(
3
),
548
554
.
40.
Zhao
,
Y.
,
Li
,
J.
,
Zhang
,
M.
,
Lu
,
Y.
,
Xie
,
H.
,
Tian
,
Y.
, and
Qiu
,
W.
(
2019
). “
Machine learning models for the hearing impairment prediction in workers exposed to complex industrial noise: A pilot study
,”
Ear Hear.
40
(
3
),
690
699
.
41.
Zhao
,
Y-m.
,
Qiu
,
W.
,
Zeng
,
L.
,
Chen
,
S.-s.
,
Cheng
,
X.-r.
,
Davis
,
R. I.
, and
Hamernik
,
R. P.
(
2010
). “
Application of the kurtosis statistic to the evaluation of the risk of hearing loss in workers exposed to high-level complex noise
,”
Ear Hear.
31
(
4
),
527
532
.