International Journal of Audiology 2015; 54: 384–390
Original Article
College students’ personal listening device usage and knowledge Kathleen Hutchinson Marron*, Kendrah Marchiondo*, Sarah Stephenson*, Sarah Wagner*, Ian Cramer†, Theresa Wharton*, Michael Hughes‡, Brittany Sproat* & Helaine Alessio† *Department of Speech Pathology and Audiology, Miami University, Oxford, Ohio, USA, †Department of Kinesiology and Health, Miami University, Oxford, Ohio, USA, ‡Department of Statistics, Miami University, Oxford, Ohio, USA
Abstract Objective: To determine the usage and knowledge of safe limits on personal listening devices (PLD) among college students. Design: First, information on health history was collected. Second, microphone in real ear techniques determined eardrum to free-field correction factors. Third, hearing levels were evaluated and information gathered about knowledge of safe listening behaviors. Study sample: 180 college students participated in a one-hour session using their PLDs and earphones set to their personal preference. Results: Virtually all participants reported knowledge of hearing loss risk due to PLD use and accurately recognized their own PLD listening levels (p ⫽ .01) as either within or exceeding safe sound limits. Forty-four subjects listened at greater than 80-dBA free-field equivalent levels. Only 7% of these participants were aware of these hazardous levels and 15% of participants’ exposure surpassed free-field equivalent levels normalized to eight hours. Conclusions: Despite reported knowledge of hearing loss risk due to PLD use in virtually all college students, 1 in 4 were found to listen to their PLDs at free-field equivalent levels greater than 80-dBA, with 94% unaware of their potential risk. Further research is needed to provide accurate PLD listening information and evaluate the possibility of long term PLD intensities that surpass recommended safety levels on hearing loss in adults over time.
Key Words: Noise-induced hearing loss; music; personal listening device; health knowledge; hearing protection; hearing perceptions
Noise pollution remains the most common environmental hazard accounting for hearing loss (Chepesiuk, 2005). The harmful effects of exposure to high intensity music during leisure activities have been documented in many nations (Biassoni et al, 2005; Cone et al, 2010; Job et al, 2000; Kumar et al, 2009; SCENIHR, 2008) with major sources of music-related exposure settings including live concerts, bars or nightclubs, and use of personal listening devices (PLDs). Mainstream media have recently drawn attention to increased reports in prevalence of noise-induced hearing loss (NIHL) in adolescents and young adults that may be caused by use of PLDs (Blue, 2008; Kirschner, 2013). A large variation exists in the listening habits among young adults who use PLDs and who attend nightclubs or participate in music performances, both between and within studies (Keppler et al, 2010). For this reason, it is difficult to pinpoint a single recreational source as the main cause of hearing loss. Yet, current PLDs can store thousands of songs, connect to the internet, and play for many hours. Coupled with the potential for high output levels, PLDs can exceed
SCENHIR (2008) guidelines leading to permanent hearing loss (Fligor & Cox, 2004). Meyer-Bisch (1996) recommends giving hearing conservation information about the risk of hearing loss to young people who use PLDs as well as limiting output levels of amplified music in places such as discos and live concerts. Despite published educational programs in hearing loss prevention, adolescents continue to exhibit poor listening behaviors (Bohlin et al, 2011) and only 8% of individuals between the ages 13 and 35 classify hearing loss as a ‘very big problem’ (Chung et al, 2005). Even when given information about NIHL, PLD users are not inclined to consistently modify potentially harmful behaviors (Daniel, 2007). Presently, the only reported concerns emerging from studies regarding PLD safety were distracted driving effects (Hoover & Krishnamurti, 2010). Listening to a PLD at too high levels for just one afternoon may cause a temporary threshold shift and tinnitus, which PLD users do not associate with immediate hearing damage (Daniel, 2007; Mercier & Hohmann, 2002; Punch et al, 2011).
Correspondence: Kathleen Hutchinson Marron, Department of Speech Pathology and Audiology, Miami University, 301 S Patterson Ave., Oxford, OH 45056, USA. E-mail: [email protected] (Received 19 April 2013; accepted 4 November 2014 ) ISSN 1499-2027 print/ISSN 1708-8186 online © 2014 British Society of Audiology, International Society of Audiology, and Nordic Audiological Society DOI: 10.3109/14992027.2014.986691
PLD use and Knowledge
Abbreviations ANOVA AUC KEMAR MIRE NIHL PLD SCENIHR
Analysis of variance Area under the curve endpoints Knowles electronics manikin for acoustical research Microphone in real ear technique Noise-induced hearing loss Personal listening device Scientific committee on emerging and newly identified health risks
Although more costly than survey research, measurement of individual transducers may be more reliable than self-report due to the potential for decreased subject awareness and candor. Comparing self-report of listening levels to physical measurements in a controlled setting may determine the extent of self-monitoring integrity among PLD users. If self-report is found to be accurate and awareness is present, credibility for surveys reporting PLD listening habits may be validated (Epstein et al, 2010; Danhauer et al, 2009; Hoover & Krishnamurti, 2010). If self-report is inaccurate, the results will aid educators in developing PLD technology to increase selfmonitoring behaviors (Daniel, 2007). The purpose of the present investigation was to determine if: (1) PLD users accurately report and self-monitor listening levels, and (2) participants are aware of a PLD’s potential to compromise hearing. Possible music genre and gender-related differences in listening behavior were also investigated in this study.
Design and Methods College students were recruited to participate in a study to examine PLD music listening behaviors using a convenience sampling method. PLD output for three music selections were measured with a probe-microphone placed in the ear with the participant’s earphones. The University Committee on the Use of Human Subjects in Research approved this study protocol.
Participants A total of 129 females and 51 males (total, 180) were recruited as participants (mean age ⫽ 19.8 years; range ⫽ 17–25 years). Subjects were recruited from the area surrounding Miami University, Oxford, Ohio, using scripted e-mails sent to listservs, advertisement fliers, and classroom announcements. All participants were screened for middle-ear disease using otoscopy and tympanometry. Demographic information from intake surveys indicated that 89% of the participants were white and 11% were black, Asian, or another ethnic group. Students represented 38 different majors studying in four college divisions at Miami University. Ten percent of participants were majors in the Department of Speech Pathology and Audiology, while the majority of participants reported studies in Business, Engineering, Fine Arts, and Education divisions.
Instrumentation Participants were instructed to bring their own PLD and headphones to testing. A standard set of earbuds and a PLD were provided to
385
18 subjects who neglected to bring one. Upon arrival, subjects were given a health history questionnaire to assess history of or risk for hearing loss due to reasons other than noise exposure (Supplementary Appendix A available online at http://informahealthcare. com/doi/abs/10.3109/14992027.2014.986691). Thirty-three percent of the subjects reported a history of otologic disease, stemming mostly from ear infections as children and swimmer’s ear. Hearing thresholds were measured with a diagnostic, clinical audiometer (Madsen, GSI 33) using the ascending method (ANSI/ASA, 2004). Pulsed tones were presented through earphones (Telephonics, TDH-50P) mounted in supra-aural cushions (MX-51/AR) while participants sat in a double-walled sound booth (Industrial Acoustics Company). Burk and Wiley (2004) found that pulsed tones had several advantages, including increased awareness of the signal. Annual calibration of the equipment was performed according to the American National Standards Institute (ANSI) guidelines (ANSI, 2004). A listening check was performed daily on the equipment. Following pure-tone testing, participants received a second listening habits survey (Supplementary Appendix B available online at http://informahealthcare.com/doi/abs/10.3109/14992027.2014.986691) to investigate self-reported music listening patterns (i.e. hours per day / days per week of use, years of use, preferred music genres) and most frequently used device types (Danhauer et al, 2009). Participants were placed in four equal groups for usage comparison purposes: rare users (⬍ 2 hours per week), minimum users (⬎ 2 and ⬍ 4.75 hours per week), moderate users (ⱖ 4.75 and ⬍ 7.5 hours per week), and extreme users (⬎ 7.5 hours per week). A probe tube from the probe microphone was marked 30 mm from the tip for males and 28 mm from the tip for females. The probe microphone was carefully slid into the ear canal until the marker ring approached the intertragal notch. Then the front cord was moved forward to hold the probe tube in place. An otoscope was used to verify that the tube was 2 to 5 mm from the eardrum following manufacturer’s specifications. Probe tube placement was done prior to listening level adjustment to simplify and expedite the recording process. All subjects were read identical instructions for each part of the study including directions for setting the volume level for their PLD. Researchers placed the participant’s headphones over the probe microphone, instructed the participant to select the first of three favorite songs, and requested the participant to adjust the volume to the most frequently used intensity level. After the selected song played for 2 minutes, a 15-second long-term average speech spectrum (LTASS) run provided 1/3 octave-band levels from 250 to 6000 Hz. In practice, a 10-second average provides a stable spectrum curve (Cox & Moore, 1988). The input envelope and average are calculated over the music passage to provide stable and repeatable data (Cole, 2009). Measurements were taken for each of three songs selected by the participant to obtain a range of listening levels and music genres. Data were displayed on graphs and tables, saved to a secure USB drive, and reviewed at a later time. Values recorded were rounded to the nearest whole decibel (Etymonic Design Incorporated, 2009). Participants concluded testing with a final survey investigating knowledge of potential risks for hearing loss related to PLD use (Supplementary Appendix B available online at http:// informahealthcare.com/doi/abs/10.3109/14992027.2014.986691). Items which queried knowledge of listening levels were addressed after all testing had been completed to reduce subject bias. Descriptive statistics were performed on the data set using SAS Software version 9.2 for Windows (SAS Institute Inc., 2011). The data were first summarized and examined for outliers and consis-
386
K. H. Marron et al.
tency. First, pure-tone thresholds in dB HL were converted to SPL (ANSI, 2004). Several preparatory steps were then taken to ready the data for analysis. Individual pure-tone threshold levels in dB SPL were then averaged from both ears and transformed to area under the curve (AUC) endpoints. Pynchon et al (1998) set a precedent using a broad response model with the AUC metric to provide an alternative method of quantifying a spectrum response in electrophysiologic analyses. The AUC analysis allows comparison of thresholds across the full frequency range for both ears for which a single response, a snapshot of sorts, can be analysed. Usage of AUC as a technique of deriving an endpoint for analysis is widely used in repeated measure designs on a wide range of response variables. The area under the receiver operator characteristic (ROC) curve is a well-established measure for determining the efficacy of tests in correctly classifying diseased and non-diseased individuals (Altman, 2006). Analyses of AUC provides an advantage in this study by creating a spectrum area for analyses that includes both ears as well as the range of frequency responses between 500–8000 Hz in a single statistical endpoint for each subject. Standards exist that describe how to reliably measure the output of earphones by adjusting correction factors from measurements taken with microphone-in-real-ear technology (Berger et al, 2009). Microphone in real ear technique (MIRE) Table 1 of the ISO was used to determine eardrum to free-field correction factors (transfer function of the outer ear [TFOE] measurements) that are needed to compare to damage-risk criteria that are commonly recommended or enforced. Part II of the ISO also includes a manikin technique for measurement of sound from insert phones. However, a manikin was not available for this study. After this transfer to sound field was done, A-weighting was applied at individual frequencies by applying the A-weighting adjustment values (Berger, 1994). From these data, participants’ song selection outputs were transformed to area under the curve (AUC) endpoints and aggregated to use as an independent variable in a standard ANOVA. From the health survey, responses from the medical history, years in school, number of loud noise exposure events, and usage time of the PLD also served as independent variables. Puretone threshold AUC measures were used as the dependent variable. A Kruskal-Wallis test (SAS, 2011) was performed to evaluate the relationship between subjective report of harmful listening behavior and free-field equivalent AUC values. Briefly, the Kruskal-Wallis test is a version of an ANOVA of independent measures that can be performed on ordinal data (Noether, 1971). Overall free-field equivalent values were compared to all the same variables as the AUC endpoints in a second analysis. A value of p ⬍.05 was set as the level of statistical significance for all tests reported.
dB HL for octave frequencies from 250 to 8000 Hz, and normal middle-ear function. Average threshold levels were 4.48 dB HL (SD 3.43) from 250 to 8000 Hz. No hearing sensitivity requirements were required for participant participation. Investigators were interested in possible associations between hearing levels and PLD use.
Listening habits
Mean age of participants was 19.7 years with 51 male and 129 female students. Although not specified for recruitment purposes, all but one of the 180 participants presented with pure-tone thresholds ⱕ 25
Sixteen subjects were excluded from this analysis for not fully responding to necessary survey questions. Over 77% of students reported using PLDs for 10 to 12 years. Subjects reported an average of 6.93 hours of use per week (SD ⫽ 9.51 hours/week); 53 subjects (30.1%) reported infrequent noise exposure (⬍ 2 hours/ week) and 43 subjects (24.4%) reported frequent exposure to noise (⬎ 7.5 hours/week). Approximately half of the participants (55%) reported exposure to loud noise events at bars or concerts, while athletic events (7%) and car speakers (7%) showed a much lower incidence. Only 8% of the subjects reported using hearing protection on an ‘as needed’ basis for work and recreational sports. Self-report data from questions in (Supplementary Appendix B available online at http://informahealthcare.com/doi/abs/10.3109/ 14992027.2014.986691) were used to investigate the frequency of listening at ‘harmful’ levels and classified into three categories: ‘yes,’ ‘no,’ or ‘at times’ (Table 1). Ten of the 171 participants who responded to the question chose ‘yes,’ 55 chose ‘no,’ and 106 chose ‘at times.’ Table 2 further describes participant report of length of listening time deemed to be dangerous to hearing levels. Most individuals thought that 1–2 hours of continuous listening would be ‘harmful to hearing’. Table 3 shows that most participants listen with small portable listening devices by iPod®. Indeed, Danhauer et al (2009) suggested that the iPod’s extended memory parameters increase risk for hearing loss. Additionally, iPod stock earbuds do not block out background noise which may result in volume level increases for many listeners. Participants reported the use of earbuds (166) more often than over-the-ear types of headphones (21). Only 15 students reported the use of noise-cancelling earphones. Nine subjects reported more than one transducer type as ‘most often used,’ accounting for the larger total report than the total number of subjects. For music genre, ‘Pop’ was the category most frequently reported by participants followed by ‘Rock’ and ‘Hip-Hop/Rhythm & Blues.’ Figure 1 illustrates the varied music types by these students. The measured mean free-field equivalent listening level was 73-dBA (Range ⫽ 46.1–103.3 dBA; sd ⫽ 11.2 dB) averaged over all music selections. Incorporating the reported listening duration data, results indicate that on average, 96% of participants in this study approached but did not exceed the SCENIHR noise exposure recommendation (REL) for both daily and weekly exposures. The SCENIHR recommend an equivalent sound-exposure level of less than 80-dBA free-field equivalent levels. Seven participants in the current study listened to potentially harmful levels when listening duration was taken into account.
Table 1. Perceived participants.
Table 2. Length of listening time perceived to be harmful to hearing.
Results Participant information
Self-report Yes At times No
listening
# Participants 10 106 55
at
‘harmful’ levels
of
171
Mean free-field equivalents in dBA
SD (dB)
78.9 74.3 69.9
12.8 10.6 11.5
Self-report
# Participants
30 minutes 1 hour 2 hours 3 hours
20 61 64 35
PLD use and Knowledge Table 3. Listening device most often used. Self-report iTouch® iPod Nano® iPod Classic® iPhone® iPod Shuffle® Zune iTouch, iPod nano, LG optimus iTouch, iPhone iPod Nano, iPhone iTouch, Blackberry LG optimus Other Cell phone Sansa Nokia N8 HTC evo Droid phone Computer Laptop Portable Radio
# Participants 60 49 19 25 4 3 1 3 1 2 1 3 4 1 1 1 1 1 1 1
387
analysis was not associated with the participants’ pure-tone hearing levels (F (2, 172) ⫽ 1.54, p ⫽ 0.12). No gender differences were found between listening levels (p ⫽ .78). Furthermore, reliability coefficients for test-retest reliability were high (r ⫽ .78, p ⫽ ⬍.001) for the 21 subjects who returned for a second probe-microphone measurement series. To quantify the significance of listening habits on pure-tone hearing levels, a three-way analysis of variance (ANOVA) was performed using hearing levels as the dependent variable with years of earbud use, reported daily usage levels, and free-field equivalent levels as factors. The AUC pure-tone hearing levels were the same for the participants who reported a long history of earbud use versus those who did not (F (3, 160) ⫽ ⫺ 0.60, p ⫽ 0.55). No association was found between the number of exposures ‘to loud noise’ incidents and puretone hearing levels (F (2, 172) ⫽ ⫺ 1.15, p ⫽ 0.25). Adjusting for other partial contributing factors, statistical differences were found in the pure-tone AUC levels based on report of weekly usage times. That is, longer use times have the potential to explain worse puretone hearing levels (F (2, 172) ⫽ 2.22, p ⫽ 0.03). Figure 2 shows the plots of discrete pure-tone threshold responses from 2000 to 8000 Hz for two groups of subjects by two listening levels, demonstrating the mean and range of threshold levels and the overlap between groups. The graph on the right shows responses for students who listened to music above 85 dBA in at least one song.
Multivariate analysis results A multivariate analysis compared participants’ mean free-field equivalent continuous A-weighted sound pressure levels to subjects’ reported typical listening levels to evaluate information consistency between real-ear sound measurements and subjective data. Comparison of participants’ free-field equivalent listening levels to self-reported occurrence of harmful levels indicated participants are likely to be aware of listening (or of not listening) at harmful levels; F( 2, 168) ⫽ 4.45, p ⫽ .01. The Kruskal Wallis test also indicated consistent perception of selected listening levels (p ⫽ .03) with a difference in free-field equivalent continuous A-weighted sound pressure levels between the participants who reported listening at harmful levels and those that stated they did not. Table 1 shows the consistency of the measured listening levels to perceived ‘harmful’ listening levels. The PLD free-field equivalent levels in the AUC
Figure 1. Varied types of music reported by student participants. Supplementary Appendix B asked participants: What genres of music do you generally listen to a majority of the time under headphones? ‘Pop’ was the category most frequently reported by participants followed by ‘Rock’ and ‘Hip-Hop / Rhythm & Blues.’.
Discussion The American Medical Association has published warnings against excessive noise exposure since 1956 and now recognizes the importance of promoting healthy listening behavior with the popularity of in-ear headphones (Glorig et al, 1956; McCaffree, 2008). College students, in particular, were found to listen to music with PLDs in excess of safe listening levels and durations that posed a risk for NIHL (Levey et al, 2011). The results of this study suggest that the listeners who reported longer usage time exhibited statistically significant decrements in hearing. Of the 180 subjects, 44 participants showed free-field equivalent levels greater than 80-dBA, yet only seven of these were listening at durations exceeding the RELs outlined by SCENIHR (2008). Average free-field equivalent exposure levels for all participants across a typical day were 73-dBA. In the current study, most subjects selected their preferred music genre with their own music devices and earphones. Epstein et al (2010) measured free-field equivalent sound levels worn by young adult PLD users in a large urban area. They found that none of the 64 listeners exceeded allowable occupational doses. In a quiet environment, 81% of our participants selected relatively safe intensities below 80-dBA free-field equivalent levels. Results of the current study indicated that subjects were able to accurately report their own music listening intensity levels. Thirty-one percent of all participants were certain that they were not using their PLDs at harmful levels and following probe microphone measurements, 87% of these were correct in their assumptions. Ten participants (6%) believed they were using their device at harmful levels; however, only three were correct in the belief. While earphone transducer type can influence output (Fligor & Cox, 2004), results of the current study indicated that participants can accurately report a generalized concept of the intensity at which they listen to a device. Participants who said ‘yes’ to their belief that their listening levels were high were more likely to be incorrect, participants who said ‘no’ or ‘at times’ to high listening levels were generally consistent. However, the chi-square frequency count of free-field
388
K. H. Marron et al. < 7.5 hrs/wk 2000 4000 8000
< 85 dB A
> 7.5 hrs/wk 2000 4000 8000
> 85 dB A
Pure Tone Threshold (dB SPL)
35
30
25
20
15
10 Frequency (hz) Usage hrs/wk
2000 4000 8000 < 7.5 hrs/wk
2000 4000 8000 > 7.5 hrs/wk
Figure 2. Pure-tone thresholds in dB SPL with means and standard error lines. Each panel represents usage in hours/week and free-field equivalent levels greater than and less than 85 dBA.
equivalent levels was based on the measured levels alone, not taking daily listening lengths into account. The potential for hazardous output levels in modern PLDs and the capacity to listen for longer periods of time creates increased risk potential. Preferred listening levels might also be influenced by the genre of music. Consistent with the students in the current study, Worthington et al (2004) also found that ‘Pop’ and ‘Rock’ were the most popular genre of music among young adults. Fligor and Cox (2004) measured maximum SPLs for different types of music in comparison to white noise by placing the headphones of a personal cassette player on a Knowles electronics manikin for acoustical research (KEMAR). Highest third octave band SPLs were found during the Rock and Pop music samples. In the current study, Pop/Rock and Alternative music produced the highest 1/3-octave band levels in the 250 to 2000 Hz range. Measured intensity levels may have been affected by the differences in transducer outputs, although it can be argued that participants would perceive the difference across headphones and adjust the levels accordingly. And ultimately, it is the listener who sets the volume control setting with a selected PLD. Research investigating the role of gender on PLD habits has produced mixed results. Hoover and Krishnamurti (2010) surveyed 428 frequent PLD users in college and found no significant difference between male and female listening habits. Kumar et al (2009) also found no differences in their measurements of male versus female
Table 4. Participants that would limit daily use of device after knowledge of listening at harmful levels. Self-report Yes No
# Participants 155 25
PLD levels. Conversely, Torre’s (2008) survey analysis showed that men are more likely to listen to PLDs at higher volume levels and for longer time periods. Measuring the average A-weighted eight-hour equivalent continuous noise exposure level, Williams (2005) also found a statistically significant tendency towards higher listening levels among men compared to females. Worthington et al (2004) also measured free-field equivalent recordings for 23 young adults and found that the male subjects listened at 6.1 dBA higher than their female counterparts in a quiet setting. However, when accounting for other listener behaviors, the main effect of gender was not statistically different. In this particular sample of students, there was also no gender difference. Hodgetts et al (2007) found that while students did turn up the PLD volume in a background noise, the increase varied by headphone and noise type. Students in the current study reported that different environments influence their volume levels. The range of environments reported included, but were not limited to: when walking to class, when riding a bus, when studying, when exercising, and when doing yard work or working. Recordings of free-field equivalent levels in all of these settings would provide realistic measures of participant volume levels. Laboratory-based values are in essence only estimates of field measures under controlled conditions. The current study attempted to provide an applicable baseline environment and procedure for practical measurement of PLD use levels. The validity of the estimates was assessed and substantiated by careful participant instructions, strict experimental procedures, and reliability testing on the 21 participants who returned for retesting. Further study should include specific types of background noise that can be applicable to a variety of listening situations. Concerns associated with the recent generation of PLDs are justified when considering the long length of time one can listen to music coupled with improved sound quality. Our results suggest that listeners who use their PLD for more than 7.5 hours in a week
PLD use and Knowledge exhibit statistically significant worse hearing levels. Meyer-Bisch (1996) also documented increased hearing thresholds in 54 subjects using their PLDs for longer than seven hours per week, compared to better thresholds in 195 subjects using PLDs only two to seven hours per week. Similarly, Buffe et al, (1986) concluded that listeners who consistently exceeded seven hours per week of moderate intensity listening with PLDs were at risk of hearing loss, needing preventative intervention. In the current study 4% of all participants were found to listen at levels greater than free-field equivalent levels of 80-dBA for greater than 7.5 hours/week. It is important to note that NIOSH (1998) exposure recommendations may not be sufficient as all of these limits are based on a 40-hour work week and 8-hours/day exposure times. The SCENHIR (2008) recommendations are more appropriate as they based on a daily continuous level of noise. During the exit interview, 84% of students stated that they would shorten PLD use duration or reduce their listening level if given information by audiologists or doctors, indicating a willingness to change their behavior if deemed to be unsafe (Table 4). Chung et al (2005) showed that 66% of over 9000 young adults could be motivated to try ear protection if they were aware of the potential for hearing loss. Although results revealed a high level of support from individuals for the concept of receiving and utilizing information on hearing loss prevention, results also showed that only a small percentage of students use ear protection strategies regularly (Bohlin et al, 2011). Explicit information detailing when the listener exceeds safety standards for volume and duration, as well as the influence of earbud type, has the potential to guide PLD users in safer listening behaviors. In a study of hearing conservation measures, Rubinstein et al (2013) recruited participants from both university faculty and students to assess attitudes and behaviors regarding potential risks of music-related risk for hearing loss. The authors found that over 50% of their college student sample were interested in measuring the level coming from their PLDs. Technology interventions might address risk for NIHL by monitoring (or minimizing) the integration of sound level with usage times and provision of educational feedback.
Conclusion The goal of the present study was to evaluate if PLD users accurately report and self-monitor listening levels and to assess if participants are aware of a PLD’s potential for hearing damage. Results showed that the average PLD user was able to discriminate a listening level that is safe versus unsafe and was aware of the harmful effects of high intensity levels of music. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References Altman D. 2006. Practical Statistics for Medical Research (2nd ed.). New York: Chapman & Hall/CRC. American National Standards Institute (ANSI). 2004. Specification for audiometers (ANSIS3.6-2004). New York: ANSI. American National Standards Institute (ANSI)/Acoustical Society of America (ASA) S3.21-2004. (R 2009). American National Standard methods for manual pure-tone threshold audiometry (Revision of ANSI S3.21-1978). Berger E. 1994. Hearing protectors: Specifications, fitting, use, and performance. In: Lipscomb (ed.). Hearing Conversation in Industry, Schools, and the Military. San Diego: Singular Publishing Group.
389
Berger E., Megerson S. & Stergar M. 2009. August 11. Personal music players: Are we measuring the sound levels correctly? The ASHA Leader. Biassoni E.C., Serra M.R., Richter U., Joekes S., Yacci M.R. et al. 2005. Recreational noise exposure and its effects on the hearing of adolescents. Part II: Development of hearing disorders. Int J Audiol, 44, 74–85. Blue L. 2008, July 28. How bad are iPods for your hearing? Time. Retrieved from http://content.time.com [accessed 20 October 2013]. Bohlin M.C., Sorbring E., Widen S.E. & Erlandsson S.I. 2011. Risks and music: Patterns among young women and men in Sweden. Noise Health, 13(53), 310–319. Buffe P., Cudennec Y.F., Ben Azzouz M., Bassoumi T. & Ferron J.J. 1986. Survey on the harmfulness of listening to music with headphones. Ann Otolaryngol Chir Cervicofac, 103(5), 351–355. Burk M.H. & Wiley T.L. 2004. Continuous versus pulsed tones in audiometry. Am J Audiol, 13, 54–61. Chepesiu R. 2005. Decibel hell: The effects of living in a noisy world. Environ Health Perspect, 113(1), A34–A41. Chung J.H., Des Roches C.M., Meunier J. & Eavey R.D. 2005. Evaluation of noise-induced hearing loss in young people, using a web-based survey technique. Pediatrics, 115(4), 861–867. Cole B. 2009. Verifit and RM500SL test signals and analysis. Audionote 2, REV 3. [online]. Available from http://www.audioscan.com/news/Audionotes/audionote2.pdf [accessed 5 February 2014]. Cone B.K., Wake M., Tobin S., Poulakis Z., Rickards F.W. 2010. Slight-mild sensorineural hearing loss in children: Audiometric, clinical, and risk factor profiles. Ear Hear, 31(2), 202–212. Cox R.M. & Moore J.N. 1988. Composite speech spectrum for hearing and gain prescriptions. J Speech Hear Res, 31, 102–107. Danhauer J.L., Johnson C.E., Byrd A., DeGood L., Meuel C. et al. 2009. Survey of college students on iPod use and hearing health. J Am Acad Audiol, 20, 5–27. Daniel E. 2007. Noise and hearing loss: A review. J Sch Health, 77(5), 225–231. Earshen J.J. 2003. Sound measurement: Instrumentation and noise description. In: Berger, Royster, Royster, Driscoll, Layne (eds.). The Noise Manual. Fairfax, USA: American Industrial Hygiene Association. Epstein M., Marozeau J. & Cleveland S. 2010. Listening habits of iPod users. J Speech Lang Hear Res, 53, 1472–1477. Etymonic Design Incorporated. 2009. Audioscan Verifit user’s guide version 3.4, 1-138. Canada. Fligor B.J. & Cox L.C. 2004. Output levels of commercially available compact disc players and the potential risk to hearing. Ear Hear, 25, 513–527. Glorig A., Wheeler D.E. & House H.P. 1956. Your ear and noise. JAMA Otolaryngol Head Neck Surg, 63(4), 449–453. Hodgetts W.E., Riecker J.M. & Szarko R.A. 2007. The effects of listening environment and earphone style on preferred listening levels of normalhearing adults using an MP3 player. Ear Hear, 28, 290–297. Hoover A. & Krishnamurti S. 2010. Survey of college students’ MP3 listening: Habits, safety issues, attitudes, and education. Am J Audiol, 19, 73–83. International Organization for Standardization (ISO). 2013. Acoustics Determination of sound immission from sound sources placed close to the ear - Part 1: Technique using a microphone in a real ear (MIRE technique). ISO 11904-1. Switzerland. ipod. (n.d.). Dictionary.com Unabridged [online], available from: http://dictionary.reference.com/browse/ipod [accessed 20 September 2014]. Job A., Raynal M., Tricoire A., Signoret J. & Rondet P. 2000. Hearing status of French youth aged from 18 to 24 years in 1997: A cross-sectional epidemiological study in the selection centres of the army in Vincennes and Lyon. Revue d’Epidémiologie et de Santé Publique, 48 (3), 227–237. Kirschner C. 2013, October 16. Are headphones bad for hearing. The Huffington Post. [online], available from http://www.huffingtonpost.com [accessed 20 October 2013]. Kumar A., Mathew M., Alexander S.A. & Kiran C. 2009. Output sound pressure levels of personal music systems and their effect on hearing. Noise Health, 11(13), 2–40. Levey S., Levey T. & Fligor B.J. 2011. Noise exposure estimates of urban MP3 player users. J Speech Lang Hear Res, 54, 263–277.
390
K. H. Marron et al.
Lutman M.E., Davis A.C. & Ferguson M.A. 2008. Epidemiological Evidence for the Effectiveness of the Noise at Work Regulations. Norwich, UK: Health and Safety Executive. McCaffree M.A. 2008. Portable listening devices and noise-induced hearing loss. Action of the AMA House of Delegates [online], available from: www.amaassn.org/resources/doc/csaph/csaph6a08.pdf [accessed 20 October 2013]. Mercier V. & Hohmann B. 2002. Is electronically amplified music too loud: What do young people think? Noise Health, 4(1), 47–55. Meyer-Bisch C. 1996. Epidemiological evaluation of hearing damage related to strongly amplified music (personal cassette players, discotheques, rock concerts) high-definition audiometric survey on 1364 subjects. Audiology, 35, 121–142. National Institute for Occupational Safety and Health (NIOSH). 1998. Criteria for recommended noise exposure: Revised criteria [online], available from: http://www.cdc.gov/niosh/docs/98–126/ [accessed 20 October 2013]. Noether G.E. 1971. Introduction to Statistics A Nonparametric Approach. Boston: Houghton Mifflin Company. Punch J.L., Elfenbein J.L. & James R.R. 2011. Targeting hearing health messages for users of personal listening devices. Am J Audiol, 20(1), 69–82.
Supplementary material available online Supplementary Appendix A and B.
Pynchon K., Tucker D., Ruth R., Barrett K. & Herr D. 1998. Area-under-thecurve measure of the auditory middle latency response (AMLR) from birth to early adulthood. Am J Audiol, 7, 45–49. Rubinstein A., Cherry R. & Bader E. 2013. College faculty views about hearing protection use and hearing conservation training. Hear Rev, 20 (2), 24–34. SAS Institute Inc. 2011. System Requirements for SAS 9.2 Foundation for Microsoft Windows. Cary, USA: SAS Institute Inc. Scientific committee on emerging and newly identified health risks (SCENIHR). 2008. Potential health risks of exposure to noise from personal music players and mobile phones including a music playing function, [online] available from: ⬍http://ec.europa.eu/health/ph_risk/ risk_en.htm ⬎ [accessed 20 October 2013]. Torre P. 2008. Young adults’ use and output level settings of personal music systems. Ear Hear, 29, 791–799. Williams W. 2005. Noise exposure levels from personal stereo use. Int J Audiol, 44, 231–236. Worthington D.A., Siegel J.H., Wilber L.A., Faber B.M., Dunckley K.T. et al. 2009. Comparing two methods to measure preferred listening levels of personal listening devices. J Acoust Soc Am, 125(6), 3733–374.
Copyright of International Journal of Audiology is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Original Article
College students’ personal listening device usage and knowledge Kathleen Hutchinson Marron*, Kendrah Marchiondo*, Sarah Stephenson*, Sarah Wagner*, Ian Cramer†, Theresa Wharton*, Michael Hughes‡, Brittany Sproat* & Helaine Alessio† *Department of Speech Pathology and Audiology, Miami University, Oxford, Ohio, USA, †Department of Kinesiology and Health, Miami University, Oxford, Ohio, USA, ‡Department of Statistics, Miami University, Oxford, Ohio, USA
Abstract Objective: To determine the usage and knowledge of safe limits on personal listening devices (PLD) among college students. Design: First, information on health history was collected. Second, microphone in real ear techniques determined eardrum to free-field correction factors. Third, hearing levels were evaluated and information gathered about knowledge of safe listening behaviors. Study sample: 180 college students participated in a one-hour session using their PLDs and earphones set to their personal preference. Results: Virtually all participants reported knowledge of hearing loss risk due to PLD use and accurately recognized their own PLD listening levels (p ⫽ .01) as either within or exceeding safe sound limits. Forty-four subjects listened at greater than 80-dBA free-field equivalent levels. Only 7% of these participants were aware of these hazardous levels and 15% of participants’ exposure surpassed free-field equivalent levels normalized to eight hours. Conclusions: Despite reported knowledge of hearing loss risk due to PLD use in virtually all college students, 1 in 4 were found to listen to their PLDs at free-field equivalent levels greater than 80-dBA, with 94% unaware of their potential risk. Further research is needed to provide accurate PLD listening information and evaluate the possibility of long term PLD intensities that surpass recommended safety levels on hearing loss in adults over time.
Key Words: Noise-induced hearing loss; music; personal listening device; health knowledge; hearing protection; hearing perceptions
Noise pollution remains the most common environmental hazard accounting for hearing loss (Chepesiuk, 2005). The harmful effects of exposure to high intensity music during leisure activities have been documented in many nations (Biassoni et al, 2005; Cone et al, 2010; Job et al, 2000; Kumar et al, 2009; SCENIHR, 2008) with major sources of music-related exposure settings including live concerts, bars or nightclubs, and use of personal listening devices (PLDs). Mainstream media have recently drawn attention to increased reports in prevalence of noise-induced hearing loss (NIHL) in adolescents and young adults that may be caused by use of PLDs (Blue, 2008; Kirschner, 2013). A large variation exists in the listening habits among young adults who use PLDs and who attend nightclubs or participate in music performances, both between and within studies (Keppler et al, 2010). For this reason, it is difficult to pinpoint a single recreational source as the main cause of hearing loss. Yet, current PLDs can store thousands of songs, connect to the internet, and play for many hours. Coupled with the potential for high output levels, PLDs can exceed
SCENHIR (2008) guidelines leading to permanent hearing loss (Fligor & Cox, 2004). Meyer-Bisch (1996) recommends giving hearing conservation information about the risk of hearing loss to young people who use PLDs as well as limiting output levels of amplified music in places such as discos and live concerts. Despite published educational programs in hearing loss prevention, adolescents continue to exhibit poor listening behaviors (Bohlin et al, 2011) and only 8% of individuals between the ages 13 and 35 classify hearing loss as a ‘very big problem’ (Chung et al, 2005). Even when given information about NIHL, PLD users are not inclined to consistently modify potentially harmful behaviors (Daniel, 2007). Presently, the only reported concerns emerging from studies regarding PLD safety were distracted driving effects (Hoover & Krishnamurti, 2010). Listening to a PLD at too high levels for just one afternoon may cause a temporary threshold shift and tinnitus, which PLD users do not associate with immediate hearing damage (Daniel, 2007; Mercier & Hohmann, 2002; Punch et al, 2011).
Correspondence: Kathleen Hutchinson Marron, Department of Speech Pathology and Audiology, Miami University, 301 S Patterson Ave., Oxford, OH 45056, USA. E-mail: [email protected] (Received 19 April 2013; accepted 4 November 2014 ) ISSN 1499-2027 print/ISSN 1708-8186 online © 2014 British Society of Audiology, International Society of Audiology, and Nordic Audiological Society DOI: 10.3109/14992027.2014.986691
PLD use and Knowledge
Abbreviations ANOVA AUC KEMAR MIRE NIHL PLD SCENIHR
Analysis of variance Area under the curve endpoints Knowles electronics manikin for acoustical research Microphone in real ear technique Noise-induced hearing loss Personal listening device Scientific committee on emerging and newly identified health risks
Although more costly than survey research, measurement of individual transducers may be more reliable than self-report due to the potential for decreased subject awareness and candor. Comparing self-report of listening levels to physical measurements in a controlled setting may determine the extent of self-monitoring integrity among PLD users. If self-report is found to be accurate and awareness is present, credibility for surveys reporting PLD listening habits may be validated (Epstein et al, 2010; Danhauer et al, 2009; Hoover & Krishnamurti, 2010). If self-report is inaccurate, the results will aid educators in developing PLD technology to increase selfmonitoring behaviors (Daniel, 2007). The purpose of the present investigation was to determine if: (1) PLD users accurately report and self-monitor listening levels, and (2) participants are aware of a PLD’s potential to compromise hearing. Possible music genre and gender-related differences in listening behavior were also investigated in this study.
Design and Methods College students were recruited to participate in a study to examine PLD music listening behaviors using a convenience sampling method. PLD output for three music selections were measured with a probe-microphone placed in the ear with the participant’s earphones. The University Committee on the Use of Human Subjects in Research approved this study protocol.
Participants A total of 129 females and 51 males (total, 180) were recruited as participants (mean age ⫽ 19.8 years; range ⫽ 17–25 years). Subjects were recruited from the area surrounding Miami University, Oxford, Ohio, using scripted e-mails sent to listservs, advertisement fliers, and classroom announcements. All participants were screened for middle-ear disease using otoscopy and tympanometry. Demographic information from intake surveys indicated that 89% of the participants were white and 11% were black, Asian, or another ethnic group. Students represented 38 different majors studying in four college divisions at Miami University. Ten percent of participants were majors in the Department of Speech Pathology and Audiology, while the majority of participants reported studies in Business, Engineering, Fine Arts, and Education divisions.
Instrumentation Participants were instructed to bring their own PLD and headphones to testing. A standard set of earbuds and a PLD were provided to
385
18 subjects who neglected to bring one. Upon arrival, subjects were given a health history questionnaire to assess history of or risk for hearing loss due to reasons other than noise exposure (Supplementary Appendix A available online at http://informahealthcare. com/doi/abs/10.3109/14992027.2014.986691). Thirty-three percent of the subjects reported a history of otologic disease, stemming mostly from ear infections as children and swimmer’s ear. Hearing thresholds were measured with a diagnostic, clinical audiometer (Madsen, GSI 33) using the ascending method (ANSI/ASA, 2004). Pulsed tones were presented through earphones (Telephonics, TDH-50P) mounted in supra-aural cushions (MX-51/AR) while participants sat in a double-walled sound booth (Industrial Acoustics Company). Burk and Wiley (2004) found that pulsed tones had several advantages, including increased awareness of the signal. Annual calibration of the equipment was performed according to the American National Standards Institute (ANSI) guidelines (ANSI, 2004). A listening check was performed daily on the equipment. Following pure-tone testing, participants received a second listening habits survey (Supplementary Appendix B available online at http://informahealthcare.com/doi/abs/10.3109/14992027.2014.986691) to investigate self-reported music listening patterns (i.e. hours per day / days per week of use, years of use, preferred music genres) and most frequently used device types (Danhauer et al, 2009). Participants were placed in four equal groups for usage comparison purposes: rare users (⬍ 2 hours per week), minimum users (⬎ 2 and ⬍ 4.75 hours per week), moderate users (ⱖ 4.75 and ⬍ 7.5 hours per week), and extreme users (⬎ 7.5 hours per week). A probe tube from the probe microphone was marked 30 mm from the tip for males and 28 mm from the tip for females. The probe microphone was carefully slid into the ear canal until the marker ring approached the intertragal notch. Then the front cord was moved forward to hold the probe tube in place. An otoscope was used to verify that the tube was 2 to 5 mm from the eardrum following manufacturer’s specifications. Probe tube placement was done prior to listening level adjustment to simplify and expedite the recording process. All subjects were read identical instructions for each part of the study including directions for setting the volume level for their PLD. Researchers placed the participant’s headphones over the probe microphone, instructed the participant to select the first of three favorite songs, and requested the participant to adjust the volume to the most frequently used intensity level. After the selected song played for 2 minutes, a 15-second long-term average speech spectrum (LTASS) run provided 1/3 octave-band levels from 250 to 6000 Hz. In practice, a 10-second average provides a stable spectrum curve (Cox & Moore, 1988). The input envelope and average are calculated over the music passage to provide stable and repeatable data (Cole, 2009). Measurements were taken for each of three songs selected by the participant to obtain a range of listening levels and music genres. Data were displayed on graphs and tables, saved to a secure USB drive, and reviewed at a later time. Values recorded were rounded to the nearest whole decibel (Etymonic Design Incorporated, 2009). Participants concluded testing with a final survey investigating knowledge of potential risks for hearing loss related to PLD use (Supplementary Appendix B available online at http:// informahealthcare.com/doi/abs/10.3109/14992027.2014.986691). Items which queried knowledge of listening levels were addressed after all testing had been completed to reduce subject bias. Descriptive statistics were performed on the data set using SAS Software version 9.2 for Windows (SAS Institute Inc., 2011). The data were first summarized and examined for outliers and consis-
386
K. H. Marron et al.
tency. First, pure-tone thresholds in dB HL were converted to SPL (ANSI, 2004). Several preparatory steps were then taken to ready the data for analysis. Individual pure-tone threshold levels in dB SPL were then averaged from both ears and transformed to area under the curve (AUC) endpoints. Pynchon et al (1998) set a precedent using a broad response model with the AUC metric to provide an alternative method of quantifying a spectrum response in electrophysiologic analyses. The AUC analysis allows comparison of thresholds across the full frequency range for both ears for which a single response, a snapshot of sorts, can be analysed. Usage of AUC as a technique of deriving an endpoint for analysis is widely used in repeated measure designs on a wide range of response variables. The area under the receiver operator characteristic (ROC) curve is a well-established measure for determining the efficacy of tests in correctly classifying diseased and non-diseased individuals (Altman, 2006). Analyses of AUC provides an advantage in this study by creating a spectrum area for analyses that includes both ears as well as the range of frequency responses between 500–8000 Hz in a single statistical endpoint for each subject. Standards exist that describe how to reliably measure the output of earphones by adjusting correction factors from measurements taken with microphone-in-real-ear technology (Berger et al, 2009). Microphone in real ear technique (MIRE) Table 1 of the ISO was used to determine eardrum to free-field correction factors (transfer function of the outer ear [TFOE] measurements) that are needed to compare to damage-risk criteria that are commonly recommended or enforced. Part II of the ISO also includes a manikin technique for measurement of sound from insert phones. However, a manikin was not available for this study. After this transfer to sound field was done, A-weighting was applied at individual frequencies by applying the A-weighting adjustment values (Berger, 1994). From these data, participants’ song selection outputs were transformed to area under the curve (AUC) endpoints and aggregated to use as an independent variable in a standard ANOVA. From the health survey, responses from the medical history, years in school, number of loud noise exposure events, and usage time of the PLD also served as independent variables. Puretone threshold AUC measures were used as the dependent variable. A Kruskal-Wallis test (SAS, 2011) was performed to evaluate the relationship between subjective report of harmful listening behavior and free-field equivalent AUC values. Briefly, the Kruskal-Wallis test is a version of an ANOVA of independent measures that can be performed on ordinal data (Noether, 1971). Overall free-field equivalent values were compared to all the same variables as the AUC endpoints in a second analysis. A value of p ⬍.05 was set as the level of statistical significance for all tests reported.
dB HL for octave frequencies from 250 to 8000 Hz, and normal middle-ear function. Average threshold levels were 4.48 dB HL (SD 3.43) from 250 to 8000 Hz. No hearing sensitivity requirements were required for participant participation. Investigators were interested in possible associations between hearing levels and PLD use.
Listening habits
Mean age of participants was 19.7 years with 51 male and 129 female students. Although not specified for recruitment purposes, all but one of the 180 participants presented with pure-tone thresholds ⱕ 25
Sixteen subjects were excluded from this analysis for not fully responding to necessary survey questions. Over 77% of students reported using PLDs for 10 to 12 years. Subjects reported an average of 6.93 hours of use per week (SD ⫽ 9.51 hours/week); 53 subjects (30.1%) reported infrequent noise exposure (⬍ 2 hours/ week) and 43 subjects (24.4%) reported frequent exposure to noise (⬎ 7.5 hours/week). Approximately half of the participants (55%) reported exposure to loud noise events at bars or concerts, while athletic events (7%) and car speakers (7%) showed a much lower incidence. Only 8% of the subjects reported using hearing protection on an ‘as needed’ basis for work and recreational sports. Self-report data from questions in (Supplementary Appendix B available online at http://informahealthcare.com/doi/abs/10.3109/ 14992027.2014.986691) were used to investigate the frequency of listening at ‘harmful’ levels and classified into three categories: ‘yes,’ ‘no,’ or ‘at times’ (Table 1). Ten of the 171 participants who responded to the question chose ‘yes,’ 55 chose ‘no,’ and 106 chose ‘at times.’ Table 2 further describes participant report of length of listening time deemed to be dangerous to hearing levels. Most individuals thought that 1–2 hours of continuous listening would be ‘harmful to hearing’. Table 3 shows that most participants listen with small portable listening devices by iPod®. Indeed, Danhauer et al (2009) suggested that the iPod’s extended memory parameters increase risk for hearing loss. Additionally, iPod stock earbuds do not block out background noise which may result in volume level increases for many listeners. Participants reported the use of earbuds (166) more often than over-the-ear types of headphones (21). Only 15 students reported the use of noise-cancelling earphones. Nine subjects reported more than one transducer type as ‘most often used,’ accounting for the larger total report than the total number of subjects. For music genre, ‘Pop’ was the category most frequently reported by participants followed by ‘Rock’ and ‘Hip-Hop/Rhythm & Blues.’ Figure 1 illustrates the varied music types by these students. The measured mean free-field equivalent listening level was 73-dBA (Range ⫽ 46.1–103.3 dBA; sd ⫽ 11.2 dB) averaged over all music selections. Incorporating the reported listening duration data, results indicate that on average, 96% of participants in this study approached but did not exceed the SCENIHR noise exposure recommendation (REL) for both daily and weekly exposures. The SCENIHR recommend an equivalent sound-exposure level of less than 80-dBA free-field equivalent levels. Seven participants in the current study listened to potentially harmful levels when listening duration was taken into account.
Table 1. Perceived participants.
Table 2. Length of listening time perceived to be harmful to hearing.
Results Participant information
Self-report Yes At times No
listening
# Participants 10 106 55
at
‘harmful’ levels
of
171
Mean free-field equivalents in dBA
SD (dB)
78.9 74.3 69.9
12.8 10.6 11.5
Self-report
# Participants
30 minutes 1 hour 2 hours 3 hours
20 61 64 35
PLD use and Knowledge Table 3. Listening device most often used. Self-report iTouch® iPod Nano® iPod Classic® iPhone® iPod Shuffle® Zune iTouch, iPod nano, LG optimus iTouch, iPhone iPod Nano, iPhone iTouch, Blackberry LG optimus Other Cell phone Sansa Nokia N8 HTC evo Droid phone Computer Laptop Portable Radio
# Participants 60 49 19 25 4 3 1 3 1 2 1 3 4 1 1 1 1 1 1 1
387
analysis was not associated with the participants’ pure-tone hearing levels (F (2, 172) ⫽ 1.54, p ⫽ 0.12). No gender differences were found between listening levels (p ⫽ .78). Furthermore, reliability coefficients for test-retest reliability were high (r ⫽ .78, p ⫽ ⬍.001) for the 21 subjects who returned for a second probe-microphone measurement series. To quantify the significance of listening habits on pure-tone hearing levels, a three-way analysis of variance (ANOVA) was performed using hearing levels as the dependent variable with years of earbud use, reported daily usage levels, and free-field equivalent levels as factors. The AUC pure-tone hearing levels were the same for the participants who reported a long history of earbud use versus those who did not (F (3, 160) ⫽ ⫺ 0.60, p ⫽ 0.55). No association was found between the number of exposures ‘to loud noise’ incidents and puretone hearing levels (F (2, 172) ⫽ ⫺ 1.15, p ⫽ 0.25). Adjusting for other partial contributing factors, statistical differences were found in the pure-tone AUC levels based on report of weekly usage times. That is, longer use times have the potential to explain worse puretone hearing levels (F (2, 172) ⫽ 2.22, p ⫽ 0.03). Figure 2 shows the plots of discrete pure-tone threshold responses from 2000 to 8000 Hz for two groups of subjects by two listening levels, demonstrating the mean and range of threshold levels and the overlap between groups. The graph on the right shows responses for students who listened to music above 85 dBA in at least one song.
Multivariate analysis results A multivariate analysis compared participants’ mean free-field equivalent continuous A-weighted sound pressure levels to subjects’ reported typical listening levels to evaluate information consistency between real-ear sound measurements and subjective data. Comparison of participants’ free-field equivalent listening levels to self-reported occurrence of harmful levels indicated participants are likely to be aware of listening (or of not listening) at harmful levels; F( 2, 168) ⫽ 4.45, p ⫽ .01. The Kruskal Wallis test also indicated consistent perception of selected listening levels (p ⫽ .03) with a difference in free-field equivalent continuous A-weighted sound pressure levels between the participants who reported listening at harmful levels and those that stated they did not. Table 1 shows the consistency of the measured listening levels to perceived ‘harmful’ listening levels. The PLD free-field equivalent levels in the AUC
Figure 1. Varied types of music reported by student participants. Supplementary Appendix B asked participants: What genres of music do you generally listen to a majority of the time under headphones? ‘Pop’ was the category most frequently reported by participants followed by ‘Rock’ and ‘Hip-Hop / Rhythm & Blues.’.
Discussion The American Medical Association has published warnings against excessive noise exposure since 1956 and now recognizes the importance of promoting healthy listening behavior with the popularity of in-ear headphones (Glorig et al, 1956; McCaffree, 2008). College students, in particular, were found to listen to music with PLDs in excess of safe listening levels and durations that posed a risk for NIHL (Levey et al, 2011). The results of this study suggest that the listeners who reported longer usage time exhibited statistically significant decrements in hearing. Of the 180 subjects, 44 participants showed free-field equivalent levels greater than 80-dBA, yet only seven of these were listening at durations exceeding the RELs outlined by SCENIHR (2008). Average free-field equivalent exposure levels for all participants across a typical day were 73-dBA. In the current study, most subjects selected their preferred music genre with their own music devices and earphones. Epstein et al (2010) measured free-field equivalent sound levels worn by young adult PLD users in a large urban area. They found that none of the 64 listeners exceeded allowable occupational doses. In a quiet environment, 81% of our participants selected relatively safe intensities below 80-dBA free-field equivalent levels. Results of the current study indicated that subjects were able to accurately report their own music listening intensity levels. Thirty-one percent of all participants were certain that they were not using their PLDs at harmful levels and following probe microphone measurements, 87% of these were correct in their assumptions. Ten participants (6%) believed they were using their device at harmful levels; however, only three were correct in the belief. While earphone transducer type can influence output (Fligor & Cox, 2004), results of the current study indicated that participants can accurately report a generalized concept of the intensity at which they listen to a device. Participants who said ‘yes’ to their belief that their listening levels were high were more likely to be incorrect, participants who said ‘no’ or ‘at times’ to high listening levels were generally consistent. However, the chi-square frequency count of free-field
388
K. H. Marron et al. < 7.5 hrs/wk 2000 4000 8000
< 85 dB A
> 7.5 hrs/wk 2000 4000 8000
> 85 dB A
Pure Tone Threshold (dB SPL)
35
30
25
20
15
10 Frequency (hz) Usage hrs/wk
2000 4000 8000 < 7.5 hrs/wk
2000 4000 8000 > 7.5 hrs/wk
Figure 2. Pure-tone thresholds in dB SPL with means and standard error lines. Each panel represents usage in hours/week and free-field equivalent levels greater than and less than 85 dBA.
equivalent levels was based on the measured levels alone, not taking daily listening lengths into account. The potential for hazardous output levels in modern PLDs and the capacity to listen for longer periods of time creates increased risk potential. Preferred listening levels might also be influenced by the genre of music. Consistent with the students in the current study, Worthington et al (2004) also found that ‘Pop’ and ‘Rock’ were the most popular genre of music among young adults. Fligor and Cox (2004) measured maximum SPLs for different types of music in comparison to white noise by placing the headphones of a personal cassette player on a Knowles electronics manikin for acoustical research (KEMAR). Highest third octave band SPLs were found during the Rock and Pop music samples. In the current study, Pop/Rock and Alternative music produced the highest 1/3-octave band levels in the 250 to 2000 Hz range. Measured intensity levels may have been affected by the differences in transducer outputs, although it can be argued that participants would perceive the difference across headphones and adjust the levels accordingly. And ultimately, it is the listener who sets the volume control setting with a selected PLD. Research investigating the role of gender on PLD habits has produced mixed results. Hoover and Krishnamurti (2010) surveyed 428 frequent PLD users in college and found no significant difference between male and female listening habits. Kumar et al (2009) also found no differences in their measurements of male versus female
Table 4. Participants that would limit daily use of device after knowledge of listening at harmful levels. Self-report Yes No
# Participants 155 25
PLD levels. Conversely, Torre’s (2008) survey analysis showed that men are more likely to listen to PLDs at higher volume levels and for longer time periods. Measuring the average A-weighted eight-hour equivalent continuous noise exposure level, Williams (2005) also found a statistically significant tendency towards higher listening levels among men compared to females. Worthington et al (2004) also measured free-field equivalent recordings for 23 young adults and found that the male subjects listened at 6.1 dBA higher than their female counterparts in a quiet setting. However, when accounting for other listener behaviors, the main effect of gender was not statistically different. In this particular sample of students, there was also no gender difference. Hodgetts et al (2007) found that while students did turn up the PLD volume in a background noise, the increase varied by headphone and noise type. Students in the current study reported that different environments influence their volume levels. The range of environments reported included, but were not limited to: when walking to class, when riding a bus, when studying, when exercising, and when doing yard work or working. Recordings of free-field equivalent levels in all of these settings would provide realistic measures of participant volume levels. Laboratory-based values are in essence only estimates of field measures under controlled conditions. The current study attempted to provide an applicable baseline environment and procedure for practical measurement of PLD use levels. The validity of the estimates was assessed and substantiated by careful participant instructions, strict experimental procedures, and reliability testing on the 21 participants who returned for retesting. Further study should include specific types of background noise that can be applicable to a variety of listening situations. Concerns associated with the recent generation of PLDs are justified when considering the long length of time one can listen to music coupled with improved sound quality. Our results suggest that listeners who use their PLD for more than 7.5 hours in a week
PLD use and Knowledge exhibit statistically significant worse hearing levels. Meyer-Bisch (1996) also documented increased hearing thresholds in 54 subjects using their PLDs for longer than seven hours per week, compared to better thresholds in 195 subjects using PLDs only two to seven hours per week. Similarly, Buffe et al, (1986) concluded that listeners who consistently exceeded seven hours per week of moderate intensity listening with PLDs were at risk of hearing loss, needing preventative intervention. In the current study 4% of all participants were found to listen at levels greater than free-field equivalent levels of 80-dBA for greater than 7.5 hours/week. It is important to note that NIOSH (1998) exposure recommendations may not be sufficient as all of these limits are based on a 40-hour work week and 8-hours/day exposure times. The SCENHIR (2008) recommendations are more appropriate as they based on a daily continuous level of noise. During the exit interview, 84% of students stated that they would shorten PLD use duration or reduce their listening level if given information by audiologists or doctors, indicating a willingness to change their behavior if deemed to be unsafe (Table 4). Chung et al (2005) showed that 66% of over 9000 young adults could be motivated to try ear protection if they were aware of the potential for hearing loss. Although results revealed a high level of support from individuals for the concept of receiving and utilizing information on hearing loss prevention, results also showed that only a small percentage of students use ear protection strategies regularly (Bohlin et al, 2011). Explicit information detailing when the listener exceeds safety standards for volume and duration, as well as the influence of earbud type, has the potential to guide PLD users in safer listening behaviors. In a study of hearing conservation measures, Rubinstein et al (2013) recruited participants from both university faculty and students to assess attitudes and behaviors regarding potential risks of music-related risk for hearing loss. The authors found that over 50% of their college student sample were interested in measuring the level coming from their PLDs. Technology interventions might address risk for NIHL by monitoring (or minimizing) the integration of sound level with usage times and provision of educational feedback.
Conclusion The goal of the present study was to evaluate if PLD users accurately report and self-monitor listening levels and to assess if participants are aware of a PLD’s potential for hearing damage. Results showed that the average PLD user was able to discriminate a listening level that is safe versus unsafe and was aware of the harmful effects of high intensity levels of music. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References Altman D. 2006. Practical Statistics for Medical Research (2nd ed.). New York: Chapman & Hall/CRC. American National Standards Institute (ANSI). 2004. Specification for audiometers (ANSIS3.6-2004). New York: ANSI. American National Standards Institute (ANSI)/Acoustical Society of America (ASA) S3.21-2004. (R 2009). American National Standard methods for manual pure-tone threshold audiometry (Revision of ANSI S3.21-1978). Berger E. 1994. Hearing protectors: Specifications, fitting, use, and performance. In: Lipscomb (ed.). Hearing Conversation in Industry, Schools, and the Military. San Diego: Singular Publishing Group.
389
Berger E., Megerson S. & Stergar M. 2009. August 11. Personal music players: Are we measuring the sound levels correctly? The ASHA Leader. Biassoni E.C., Serra M.R., Richter U., Joekes S., Yacci M.R. et al. 2005. Recreational noise exposure and its effects on the hearing of adolescents. Part II: Development of hearing disorders. Int J Audiol, 44, 74–85. Blue L. 2008, July 28. How bad are iPods for your hearing? Time. Retrieved from http://content.time.com [accessed 20 October 2013]. Bohlin M.C., Sorbring E., Widen S.E. & Erlandsson S.I. 2011. Risks and music: Patterns among young women and men in Sweden. Noise Health, 13(53), 310–319. Buffe P., Cudennec Y.F., Ben Azzouz M., Bassoumi T. & Ferron J.J. 1986. Survey on the harmfulness of listening to music with headphones. Ann Otolaryngol Chir Cervicofac, 103(5), 351–355. Burk M.H. & Wiley T.L. 2004. Continuous versus pulsed tones in audiometry. Am J Audiol, 13, 54–61. Chepesiu R. 2005. Decibel hell: The effects of living in a noisy world. Environ Health Perspect, 113(1), A34–A41. Chung J.H., Des Roches C.M., Meunier J. & Eavey R.D. 2005. Evaluation of noise-induced hearing loss in young people, using a web-based survey technique. Pediatrics, 115(4), 861–867. Cole B. 2009. Verifit and RM500SL test signals and analysis. Audionote 2, REV 3. [online]. Available from http://www.audioscan.com/news/Audionotes/audionote2.pdf [accessed 5 February 2014]. Cone B.K., Wake M., Tobin S., Poulakis Z., Rickards F.W. 2010. Slight-mild sensorineural hearing loss in children: Audiometric, clinical, and risk factor profiles. Ear Hear, 31(2), 202–212. Cox R.M. & Moore J.N. 1988. Composite speech spectrum for hearing and gain prescriptions. J Speech Hear Res, 31, 102–107. Danhauer J.L., Johnson C.E., Byrd A., DeGood L., Meuel C. et al. 2009. Survey of college students on iPod use and hearing health. J Am Acad Audiol, 20, 5–27. Daniel E. 2007. Noise and hearing loss: A review. J Sch Health, 77(5), 225–231. Earshen J.J. 2003. Sound measurement: Instrumentation and noise description. In: Berger, Royster, Royster, Driscoll, Layne (eds.). The Noise Manual. Fairfax, USA: American Industrial Hygiene Association. Epstein M., Marozeau J. & Cleveland S. 2010. Listening habits of iPod users. J Speech Lang Hear Res, 53, 1472–1477. Etymonic Design Incorporated. 2009. Audioscan Verifit user’s guide version 3.4, 1-138. Canada. Fligor B.J. & Cox L.C. 2004. Output levels of commercially available compact disc players and the potential risk to hearing. Ear Hear, 25, 513–527. Glorig A., Wheeler D.E. & House H.P. 1956. Your ear and noise. JAMA Otolaryngol Head Neck Surg, 63(4), 449–453. Hodgetts W.E., Riecker J.M. & Szarko R.A. 2007. The effects of listening environment and earphone style on preferred listening levels of normalhearing adults using an MP3 player. Ear Hear, 28, 290–297. Hoover A. & Krishnamurti S. 2010. Survey of college students’ MP3 listening: Habits, safety issues, attitudes, and education. Am J Audiol, 19, 73–83. International Organization for Standardization (ISO). 2013. Acoustics Determination of sound immission from sound sources placed close to the ear - Part 1: Technique using a microphone in a real ear (MIRE technique). ISO 11904-1. Switzerland. ipod. (n.d.). Dictionary.com Unabridged [online], available from: http://dictionary.reference.com/browse/ipod [accessed 20 September 2014]. Job A., Raynal M., Tricoire A., Signoret J. & Rondet P. 2000. Hearing status of French youth aged from 18 to 24 years in 1997: A cross-sectional epidemiological study in the selection centres of the army in Vincennes and Lyon. Revue d’Epidémiologie et de Santé Publique, 48 (3), 227–237. Kirschner C. 2013, October 16. Are headphones bad for hearing. The Huffington Post. [online], available from http://www.huffingtonpost.com [accessed 20 October 2013]. Kumar A., Mathew M., Alexander S.A. & Kiran C. 2009. Output sound pressure levels of personal music systems and their effect on hearing. Noise Health, 11(13), 2–40. Levey S., Levey T. & Fligor B.J. 2011. Noise exposure estimates of urban MP3 player users. J Speech Lang Hear Res, 54, 263–277.
390
K. H. Marron et al.
Lutman M.E., Davis A.C. & Ferguson M.A. 2008. Epidemiological Evidence for the Effectiveness of the Noise at Work Regulations. Norwich, UK: Health and Safety Executive. McCaffree M.A. 2008. Portable listening devices and noise-induced hearing loss. Action of the AMA House of Delegates [online], available from: www.amaassn.org/resources/doc/csaph/csaph6a08.pdf [accessed 20 October 2013]. Mercier V. & Hohmann B. 2002. Is electronically amplified music too loud: What do young people think? Noise Health, 4(1), 47–55. Meyer-Bisch C. 1996. Epidemiological evaluation of hearing damage related to strongly amplified music (personal cassette players, discotheques, rock concerts) high-definition audiometric survey on 1364 subjects. Audiology, 35, 121–142. National Institute for Occupational Safety and Health (NIOSH). 1998. Criteria for recommended noise exposure: Revised criteria [online], available from: http://www.cdc.gov/niosh/docs/98–126/ [accessed 20 October 2013]. Noether G.E. 1971. Introduction to Statistics A Nonparametric Approach. Boston: Houghton Mifflin Company. Punch J.L., Elfenbein J.L. & James R.R. 2011. Targeting hearing health messages for users of personal listening devices. Am J Audiol, 20(1), 69–82.
Supplementary material available online Supplementary Appendix A and B.
Pynchon K., Tucker D., Ruth R., Barrett K. & Herr D. 1998. Area-under-thecurve measure of the auditory middle latency response (AMLR) from birth to early adulthood. Am J Audiol, 7, 45–49. Rubinstein A., Cherry R. & Bader E. 2013. College faculty views about hearing protection use and hearing conservation training. Hear Rev, 20 (2), 24–34. SAS Institute Inc. 2011. System Requirements for SAS 9.2 Foundation for Microsoft Windows. Cary, USA: SAS Institute Inc. Scientific committee on emerging and newly identified health risks (SCENIHR). 2008. Potential health risks of exposure to noise from personal music players and mobile phones including a music playing function, [online] available from: ⬍http://ec.europa.eu/health/ph_risk/ risk_en.htm ⬎ [accessed 20 October 2013]. Torre P. 2008. Young adults’ use and output level settings of personal music systems. Ear Hear, 29, 791–799. Williams W. 2005. Noise exposure levels from personal stereo use. Int J Audiol, 44, 231–236. Worthington D.A., Siegel J.H., Wilber L.A., Faber B.M., Dunckley K.T. et al. 2009. Comparing two methods to measure preferred listening levels of personal listening devices. J Acoust Soc Am, 125(6), 3733–374.
Copyright of International Journal of Audiology is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
