This study investigated the accuracy, reliability, and characteristics of three brands of personal noise dosimeters (N = 7 units) in both pink noise (PN) environments and natural environments (NEs) through the acquisition of decibel readings, Leq readings and noise doses. Acquisition periods included repeated PN conditions, choir room rehearsals and participant (N = 3) Leq and noise dosages procured during a day in the life of a music student. Among primary results: (a) All dosimeters exhibited very strong positive correlations for PN measurements across all instruments; (b) all dosimeters were within the recommended American National Standard Institute (ANSI) SI.25-1991 standard of ±2 dB (A) of a reference measurement; and (c) all dosimeters were within the recommended ANSI SI.25-1991 standard of ±2 dB (A) when compared with each other. Results were discussed in terms of using personal noise dosimeters within hearing conservation and research contexts and recommendations for future research. Personal noise dosimeters were studied within the contexts of PN environments and NEs (choral classroom and the day in the life of collegiate music students). This quantitative study was a non-experimental correlation design. Three brands of personal noise dosimeters (Cirrus doseBadge, Quest Edge Eg5 and Etymotic ER200D) were tested in two environments, a PN setting and a natural setting. There were two conditions within each environment. In the PN environment condition one, each dosimeter was tested individually in comparison with two reference measuring devices (Ivie and Easera) while PN was generated by a Whites Instrument PN Tube. In condition two, the PN procedures were replicated for longer periods while all dosimeters measured the sound levels simultaneously. In the NE condition one, all dosimeters were placed side by side on a music stand and recorded sound levels of choir rehearsals over a 7-h rehearsal period. In NE, condition two noise levels were measured during a day in the life of college music students. Three participants each wore two types of dosimeters for an 8-h period during a normal school day. Descriptive statistical analyzes including means, standard deviation and Pearson product-moment correlation. The primary finding is that the dosimeters in this study recorded results within ±2 dB of either a reference measurement or within dosimeters in all four conditions examined. All dosimeters studied measured steady noise source accurately and consistently, with strong positive correlations across all instruments. Measurements acquired during choral rehearsals indicated a maximum of 1.5 dB difference across dosimeters. The Etymotic research personal noise dosimeters (ER200D) could provide individuals and schools of music with a relatively inexpensive tool to monitor sound doses. Findings from this study suggest that the three brands of dosimeters tested will provide reliable Leq levels and hearing dosages in both PN and natural settings.
Keywords: Hearing dose, noise dose, noise dosimeter, noise induced hearing loss
|How to cite this article:|
Cook-Cunningham SL. Personal noise dosimeters: Accuracy and reliability in varied settings. Noise Health 2014;16:143-8
| Introduction|| |
Noise induced hearing loss (NIHL) should be of concern to all musicians, including university student musicians. As concern over NIHL has risen, an increasing number of schools of music have established hearing conservation programs. ,,,, In the year 2011, the National Association of Schools of Music and the Performing Arts Medicine Association recommended schools take scientific measurements of sound levels using either sound-level meters or noise dosimeters.  While a growing body of research addresses potentially dangerous sound levels in music settings, little research to date has examined the consistency of the instruments responsible for collecting that data.
The American National Standard Institute (ANSI) outlined specifications for personal noise dosimeters in 1978 with the goal of minimizing variabilities between different dosimeter models.  These standards were revised in 1991 and reaffirmed in 2007 to provide instrument characteristics and tolerances.  Dosimeters used in industrial hearing protection compliance settings are required to conform to ANSI standards.
Sound level meters (SLM) and noise dosimeters are given class designations equivalent to their levels of precision. A type 0 SLM or dosimeter is specific to laboratory use while a type 1 designation is for precise field measurements and has accuracy of ±1 dB. A type 2 instrument has ±2 dB accuracy and is considered sufficient for general-purpose noise surveys. Dosimeters that comply with ANSI SI.25-1991 must either list the class designation or a means by which to obtain that information. 
ANSI SI.25-1991 was developed for industrial settings with the primary focus of maintaining healthy hearing environments for employees. SLM and noise dosimeters were also designed for industrial settings where noise is generally constant both in duration and at sound pressure levels. There has been no measurement tool designed specifically for determining sound levels in music settings, where there may be inherent fluctuation of sound pressure levels and dynamic changes. Personal noise dosimeters have typically been the instrument of choice in measuring sound pressure levels in music situations due to their mobile nature and ability to assess an individual's sound level exposure.
Music researchers have collected sound exposure data using a variety of brands of noise dosimeters including Brüel & Kjær, Larson Davis, Quest, Cassella, Cirrus doseBadge, Quest Edge and Etymotic. ,,,,,,,,,,,,,,,, Noise dosimeters that conform to ANSI S1.25-1991 and are designated a type 2 noise dosimeter range in cost from $1500 to $4000. This high cost could deter music schools from purchasing a noise dosimeter and thus taking sound level measurements. Etymotic research developed a cost-effective (about $250.00/unit) personal noise dosimeter (ER200D), which measures continuous sound pressure level readings (Leq), computes noise dose and is user friendly. According to the manufacturers, ER200D dosimeters were not designed to be type 2 devices but do conform to some of the type 2 criteria, particularly linearity, microphone response and frequency response.
The purpose of this study was to investigate the accuracy, reliability and characteristics of three brands (Cirrus doseBadge, Quest Edge Eg5 and Etymotic ER200D) of personal noise dosimeters of varying characteristics and costs that are currently being used in music research.
The following research questions guided this investigation:
- What do data acquired through the three brands of noise dosimeters (Cirrus doseBadge (n = 2), Quest Edge Eg5 (n = 1) and Etymotic ER200D (n = 4) indicate about dosimeter reliability when measuring pink noise (PN) in a PN environment in comparison to reference measurements and the ANSI Standard S1.25-1991 of ±2 dB accuracy?
- What do data acquired through the three brands of noise dosimeters indicate about dosimeter accuracy in natural environments (NEs) (choral rehearsal, a day in the life of a student), when compared with each other and the ANSI Standard S1.25-1991 of ±2 dB accuracy?
| Methods|| |
Three brands of dosimeters were tested:
- Cirrus Research CR 110A doseBadges (n = 2),
- Etymotic ER200D personal noise dosimeters (n = 4) and
- Quest Edge Eg5 (n = 1) for a total of seven dosimeters.
All three brands have been used previously in music research.
The Cirrus CR:110A doseBadges (CR dBadge) used in this study, were a type 2 instrument that weighed 8 oz. and attached to the wearer's shoulder with a set of clips. Shaped like a small cone, the units were designed to be unobtrusive. The microphone was 13 mm and was positioned in the top of the doseBadge. The doseBadge was controlled by an RC:110A handheld Reader unit through an infra-red link on the dosimeter. The reader unit contained an integral calibrator designed specifically to fit the top of the doseBadge, allowing the operator to calibrate the unit before and after each use.
The Etymotic (ER200D) dosimeters used for this study were designed to provide the wearer with a user-friendly estimate of noise dose. Each unit measured approximately 15 cm long by 2 cm wide and attached at the wearer's collar with a clip, similar to that of a writing pen. There was a stop/start button on the Etymotic with a noise dose display in percentages on the front of the unit. It contained an omnidirectional microphone (flat from 100 Hz to 15 kHz) located on the top of the unit. It was factory calibrated with accuracy of ±2.5 dB (A) with the ability to verify the calibration through the software program. The dosimeter captured dose values every 220 ms averaged over a 3.75-min interval and saved in non-volatile memory (16 times/h).
The Quest Edge Eg5 unit used in this study, attached to the shoulder with the manufacturer's clips and weighed approximately 7 oz. It had an on/off button and a start/stop button to activate the meter when running measurements. The unit included a manufacturer's calibrator and specified the dosimeter should be calibrated before and after every noise collection. The Eg5 had a display panel and study measurements could be viewed by scrolling through the menu. It conformed to ANSI S1.25 personal noise dosimeter standards.
The personal noise dosimeters (N = 7) were tested in two environments; a PN setting and a natural setting. There were two test conditions within each environment. All of the dosimeters were set to NIOSH standards with a criterion level of 85 dBA, an exchange rate of 3 dBA and a criterion time of 8 h.
The seven dosimeters measured PN in an acoustics classroom. Dosimeter results were compared with measurements from two reference instruments, an Ivie IE-45 audio analysis system and an Easera computer program, designed for electronic and acoustic system evaluation and response analysis.
The Ivie IE-45 was a hand-held calibrated instrument that measured sound pressure levels. It had an accuracy of ±0.5 dB and was connected to a Viliv ultra model PC, which displayed the measured decibel. A Larson Davis 1136 microphone was connected to the Ivie audio analysis device using an Ie-6P pre-amplifier. The microphone was calibrated to 94 dB(A) using a Cirrus Research IEC 942 microphone calibrator.
The Easera computer program provided the second reference measurement tool. It was loaded on a Dell 1557 notebook computer through an Easera Gateway audio interface device. The sound was captured with an Earthworks high fidelity, omni-directional M-30 #8168 microphone and was connected to the Easera interface device. The microphone was calibrated to 94 dBA using a Cirrus Research IEC 942 microphone calibrator.
A White Instruments PN Tube generated PN at a constant decibel level. The PN was fed into a Mackie 1402-VLZ mixing board that powered a QSC Power Amplifier, model CX 254. The amplifier fed two Electro-Voice zX1 loudspeakers positioned in the front corners of the room, approximately 4 m from the microphones.
Each personal noise dosimeter was tested individually in comparison with the two reference measuring devices (Ivie and Easera). The noise dosimeter being assessed was affixed to a Manhasset 48 Black Standard Music Stand using the dosimeter manufacturer's provided attachment tool. The stand was positioned directly between the two microphone stands holding the Earthworks and Larson Davis microphones with approximately 4 cm between each device.
PN was initiated at approximately 80 dB(A) with the two reference measuring units in agreement by ±0.2 dB. Each dosimeter recorded the decibel levels for a 2-min interval.
The PN procedures were replicated for a longer time periods and in a more reverberant space, a Midwestern University choir room. During this condition, the dosimeters (N = 7) were attached to the top of a Manhasset 48 Black Standard Music Stand using manufacturer's standard clips. The first dosimeter was attached approximately 10 cm from the left side of the stand with each successive instrument arranged side by side in the following order: ER200-D #30044, ER200-D #30141, CR 110-A doseBadge #766, Quest Edge Eg5, ER200-D #30078, ER200-D #30140 and CR 110-A doseBadge #778.
Directly behind the dosimeters was a Brüel & Kjær Type 2209 Precision Sound Level Meter with a Grayson Stadler 1136 type one microphone. The microphone was calibrated to 94 dB (A) using a Cirrus IEC942 Calibrator and extended over the stand at the same level as the dosimeters. The SLM provided a comparison reference measurement for the condition.
A Whites Instrument PN Tube generated the PN. The tube was connected to a SHURE FP11 Pre-Amplifier, powered by an ART Phantom II 48 v phantom power supply with the sound being directed through a Fostex 6301 B powered analog speaker.
PN was played for 30 min at three separate decibel levels, 80 dB SPL, 86 dB SPL and 92 dB SPL. The noise was produced 3 times at each decibel level for a total of nine measurements. During this condition, all dosimeters measured the sound levels simultaneously. The dosimeters were started and stopped within 30 s of each other. Sound levels were monitored by the SLM throughout each 30-minute treatment.
The dosimeters obtained sound levels in two NEs, during six consecutive choir rehearsals and throughout a day in the life of a music student.
NE: Condition one
The seven dosimeters were situated on a stationary music stand in front of the same Midwestern University choir room that was used in PN condition two. The dosimeters recorded sound levels of choir rehearsals over a 7-h rehearsal period. The live music provided a real-life hearing situation with fluctuating sound pressure levels inherent to choral music.
Six different choirs rehearsed throughout the day. Five of the choirs practiced for 1 h time periods and one ensemble rehearsed for a 2-h time frame. The choirs were seated in chairs in a semi-circle with the piano and conductor in the center of the formation. The dosimeters were attached to a music stand positioned in front of the room approximately 2.5 m behind the conductor. The noise dosimeters were arranged in the same left to right formation as in PN, condition two (ER200D #30044, ER200D #30141, CR 110-A doseBadge #766, Quest Edge Eg5, ER200D #30078, ER200D #30140 and CR 110-A doseBadge #778). The dosimeters were all turned on within 30 s of each other and recorded noise levels continuously throughout the 7-h rehearsal period. At the conclusion of the final choir practice, the dosimeters were turned off within 30 s of each other.
NE: Condition two
NE condition two measured noise levels during a day in the life of collegiate music students. Three participants, pseudonyms Sarah, Bob and Thomas, each wore two types of dosimeters for an 8-h period during their school-day regular rehearsals, classes and activities. The dosimeters measured continuous sound pressure levels during the 8-h time frame. Sarah was 20 years of age and Bob and Thomas were both 22 years old. The dosimeters were positioned side by side in the middle of the participant's right shoulder and were started and stopped by the researcher.
| Results|| |
Results are presented according to the research questions posed for this investigation.
Research question one: Dosimeter data from a PN environment
[Table 1] and [Table 2] present the results of PN tests.
|Table 1: Dosimeter data and variance from reference measurements for PN, condition one|
Click here to view
|Table 2: Means, standard deviations and variance from reference measurement for PN, condition two|
Click here to view
Results from individual testing of each dosimeter in comparison with two reference measurements indicated a less than ±2 dB difference between tested dosimeters (N = 7) and reference instruments. All dosimeters registered measurements below those recorded by the reference instruments. The ER 30141 recorded the greatest difference in comparison with the Ivie of −1.4 dB (A) and the ER 30141 also measured the largest difference with the Easera at −1.2 dB (A). The Quest Edge Eg5 recorded the least difference with both the Ivie and Easer with measurements of −0.5 and −0.3 respectively.
The data indicated that during three repeated testing procedures at three dB levels, all dosimeters were within a ±2 dB range when compared to the reference measurement. The greatest differences between the reference measurements and dosimeters were recorded by the dBadge 766 at 80 dBA (+1.8 dBA), the ER 30078 at 86 dBA (−1.8 dBA) and the ER 30140 at 92 dB SPL (+1.7 dBA).
[Table 3] presents correlation data between all noise dosimeters during PN, condition two.
Pearson product-moment correlation results between all dosimeters indicated very strong positive correlations for PN measurements across all instruments. The highest correlations (r = 0.997) were between the Etymotic dosimeters (ER 30044, ER 30141 and ER 30078) and the lowest correlation (r = 0.970) was between the CR 778 and Eg5; all positive correlations were extremely high.
Research question two: Dosimeter data from a NE
[Figure 1] and [Figure 2] present the results of recorded Leq and dose percentages acquired during 7 h of choir rehearsals. The Etymotic dosimeters registered both the highest and lowest Leq readings during this condition. There was a difference of 1.5 dB between the ER 30140 and ER 30078 which were adjacent to each other. The noise dose percentages varied in accordance with the recorded Leq levels with a 1.3% difference between the highest and lowest recorded doses.
[Figure 3] and [Figure 4] present dosimeter data from Sarah, Bob and Thomas acquired as they wore two dosimeters simultaneously during an average university day. The difference between the recorded Leq from the paired dosimeters were all under the ANSI S1.35-1991 standard of ±2 dB difference. The Quest Edge and Etymotic brands measured within 0.5 dBA of each other, the dBadge and the Etymotic were within 0.8 dBA and the smallest difference was between the dBadge and the Quest Edge at 0.1 dBA.
|Figure 3: Participants' Leq from a day in the life of a university music students during rehearsals/class|
Click here to view
| Discussion|| |
Few studies to date compare the performance of a variety of personal noise dosimeters in multiple contexts. The primary finding of this study is that the dosimeters in this study recorded results within ±2 dB of either a reference measurement or within dosimeters in all four conditions examined.
Results from the PN trials indicate that all dosimeters studied measured the steady noise source accurately and consistently, with strong positive correlations across all instruments. Measurements acquired during choral rehearsals indicated a maximum of 1.5 dB difference across dosimeters. Of particular interest were the readings from the music students who wore the dosimeters for a day.
Thomas experienced a Leq of 92 dB(A) and his dose (579.8-590%) exceeded the NIOSH 8-h daily standard by more than 5 times. Thomas' activities that day included a 3-h solo saxophone rehearsal session in a practice room. This result appears consistent with findings from the Phillips & Mace study  in which mean sound pressure levels in university practice rooms were 87-95 dB(A). Future studies might investigate varying acoustic properties of practice rooms and what impact if any the acoustics play on noise doses.
Hearing protection among musicians is another area that merits further research. According to Thomas, he has hearing protection in the form of earplugs, but uses them only when he is in a band rehearsal. He does not use the earplugs when practicing because he "does not like the way [his] instrument sounds in [his] head." Future investigations might consider the effect of an acclimation period on musicians' perceptions of ear protection devices.
For those schools of music wishing to purchase personal noise dosimeters or participate in a hearing protection campaign, reliability, ease of use and cost might factor into the decision making process. The Etymotic ER200D's are designed to be a user-friendly and cost-effective device (approximately $250/unit) that could easily be initiated and terminated by the participant. The criterion level, exchange rate and threshold level are programed through the accompanying software and cannot be changed on the instrument itself, thus eliminating potential contamination of the results. The software for the Etymotic ERD200s is self-explanatory and accessible. The dosimeter was somewhat limited however in that it computed only two measurements, overall Leq and final hearing dose.
The Cirrus dosimeters provide dual channels that can be programmed to different standards and allow for two sets of measurements to be taken simultaneously, i.e., can take measurements according to NIOSH and OSHA standards at the same time. The Cirrus doseBadges are initiated with either the hand-held reader or by a remote control key fob, reducing the possibility of the wearer tampering with the device. The software, though more intricate than the ER200D, is accessible and provides more measurements than the Etymotic. The initial cost of the Cirrus doseBadge system is approximately $3700, which might preclude some schools from purchasing this dosimeter.
The Quest Edge dosimeter provides three different programmable channels, each set to a different standard. The Quest Edge dosimeter has a control pad on the unit itself; however, it can be locked to safeguard against potential user interference. The software for the Quest Edge and Cirrus were similar in complexity and data production. The Quest Edge costs between $2200 and $2400, which might be a deterrent to some music schools.
In sum, findings from this study suggest that the three brands of dosimeters tested will provide reliable Leq levels and hearing dosages in both PN and natural settings. Researchers and schools of music who may be interested in using these units, of course, will need to make decisions according to the specific factors that inform use of personal noise dosimeters within their particular contexts.
| References|| |
|1.||Chesky K. Preventing music-induced hearing loss. Mus Educ J 2008;94:36-41. |
|2.||Chesky K. Schools of music and conservatories and hearing loss prevention. Int J Audiol 2011;50 Suppl 1:S32-7. |
|3.||Hodges DA. Brains and music, whales and apes, hearing and learning…and more. Update App Res Mus Educ 2009;27:62-75. |
|4.||Miller VL, Stewart, M, Lehman M. Noise exposure levels for student musicians. Med Prob Perform Art 2007;22:160-65. |
|5.||Phillips SL, Henrich VC, Mace ST. Prevalence of noise-induced hearing loss in student musicians. Int J Audiol 2010;49:309-16. |
|6.||National Association of Schools of Music and Performing Arts Medicine Association. Basic information on hearing health: Information and recommendations for faculty and staff in schools of music 2011. Available from: http://www.nasm.arts-accredit.org/index.jsp?page=NASM-PAMA_Hearing_Health (last accessed on 10 Feb, 2014.) |
|7.||ANSI S1.25-1978. Specifications for Personal Noise Dosimeters. New York: American National Standards Institue; 1978. |
|8.||ANSI S1.25-1991. Specifications for Personal Noise Dosimeters. New York: American National Standards Institue; 2007. |
|9.||Occupational Safety and Health Administration. Occupational noise exposure; Hearing conservation amendment: Final rule. (Fed. Reg. 48:9738-9785). Washington, DC: US Department of Health and Human Service Publication; 1983. |
|10.||Fearn R. Hearing loss in musicians. J Sound Vib 1993;163:372-8. |
|11.||McBride D, Gill F, Proops D, Harrington M, Gardiner K, Attwell C. Noise and the classical musician. BMJ 1992;305:1561-3. |
|12.||Obeling L, Poulsen T. Hearing ability in Danish symphony orchestra musicians. Noise Health 1999;1:43-9. |
|13.||Schmidt JH, Pedersen ER, Juhl PM, Christensen-Dalsgaard J, Andersen TD, Poulsen T, et al. Sound exposure of symphony orchestra musicians. Ann Occup Hyg 2011;55:893-905. |
|14.||Royster JD, Royster LH, Killion MC. Sound exposures and hearing thresholds of symphony orchestra musicians. J Acoust Soc Am 1991;89:2793-803. |
|15.||Laitinen HM, Toppila EM, Olkinuora PS, Kuisma K. Sound exposure among the Finnish National Opera personnel. Appl Occup Environ Hyg 2003;18:177-82. |
|16.||Owens DT. Sound pressure levels experienced by the high school band director. Med Probl Perform Art 2004;19:109-15. |
|17.||Toppila E, Koskinen H, Pyykkö I. Hearing loss among classical-orchestra musicians. Noise Health 2011;13:45-50. |
|18.||Henoch M, Chesky K. Sound exposure levels experienced by a college jazz band ensemble - Comparison with OSHA risk criteria. Med Probl Perform Art 2011;15:17-22. |
|19.||Behar A, MacDonald E, Lee J, Cui J, Kunov H, Wong W. Noise exposure of music teachers. J Occup Environ Hyg 2004;1:243-7. |
|20.||Qian CL, Behar A, Wong W. Noise exposure of musicians of a ballet orchestra. Noise Health 2011;13:59-63. |
|21.||O′Brien I, Wilson W, Bradley A. Nature of orchestral noise. J Acoust Soc Am 2008;124:926-39. |
|22.||Mace ST. A descriptive analysis of university music performance teachers′ sound level exposures during a typical day of teaching, performing and rehearsing. In: 9 th International Conference on Music Perception and Cognition August, 2006; Alma Mater Studorium University of Bologna. |
|23.||Phillips SL, Mace S. Sound level measurements in music practice rooms. Mus Perf Res 2008;2:36-47. |
|24.||Walter JS. Sound exposure levels experienced by university wind band members. Med Probl Perform Art 2009;24:63-70. |
|25.||Deiters K, Flamme G, Roth E, Nordgren A. Daily sound exposures among college-level musicians. Paper Presented at the National Hearing Conservation Association National Meeting, Orlando, FL; February 2010. |
|26.||Cook-Cunningham SL, Grady ML, Nelson HR. Hearing doses and perceptions of hearing and singing effort among university choral singers in varied rehearsal and performance settings. Int J Res Chor Sing 2012;4:19-35. |
Dr. Sheri Lynn Cook-Cunningham
University of Central Arkansas, 201 Donaghey Ave, Conway, AR. 72035
Source of Support: None, Conflict of Interest: None
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3]