| Article Access Statistics|
| Viewed||11542 |
| Printed||468 |
| Emailed||8 |
| PDF Downloaded||45 |
| Comments ||[Add] |
| Cited by others ||1 |
|Year : 2011
: 13 | Issue : 53 | Page
|Using the Extended Parallel Process Model to create and evaluate the effectiveness of brochures to reduce the risk for noise-induced hearing loss in college students
Michael R Kotowski1, Sandi W Smith2, Patti M Johnstone3, Erin Pritt3
1 School of Communication Studies, University of Tennessee, Knoxville, 293 Communications Building, Knoxville, TN 37996
2 Department of Communication, Michigan State University, 573A Communication Arts and Sciences Building, East Lansing, MI 48824
3 Department of Audiology and Speech Pathology, University of Tennessee, Health Science Center, 444 S. Stadium Hall, Knoxville, TN 37996
Click here for correspondence address
|Date of Web Publication||14-Jul-2011|
Brochures containing messages developed according to the Extended Parallel Process Model were deployed to increase intentions to use hearing protection for college students. These brochures were presented to one-half of a college student sample, after which a questionnaire was administered to assess perceptions of threat, efficacy, and behavioral intentions. The other half of the sample completed the questionnaire and then received brochures. Results indicated that people receiving the brochure before the questionnaire reported greater perceptions of hearing loss threat and efficacy to use ear plugs when in loud environments, however, intentions to use ear plugs were unchanged. Distribution of the brochure also resulted in greater perceptions of hearing loss threat and efficacy to use over-the-ear headphones when using devices such as MP3 players. In this case, however, intentions to use over-the-ear headphones increased. Results are discussed in terms of future research and practical applications.
Keywords: Extended Parallel Process Model, noise-induced hearing loss, health communication
|How to cite this article:|
Kotowski MR, Smith SW, Johnstone PM, Pritt E. Using the Extended Parallel Process Model to create and evaluate the effectiveness of brochures to reduce the risk for noise-induced hearing loss in college students. Noise Health 2011;13:261-71
|How to cite this URL:|
Kotowski MR, Smith SW, Johnstone PM, Pritt E. Using the Extended Parallel Process Model to create and evaluate the effectiveness of brochures to reduce the risk for noise-induced hearing loss in college students. Noise Health [serial online] 2011 [cited 2021 Feb 26];13:261-71. Available from: https://www.noiseandhealth.org/text.asp?2011/13/53/261/82958
| Introduction|| |
Using the Extended Parallel Process Model (EPPM) to create and evaluate the effectiveness of brochures to reduce the risk for noise-induced hearing loss (NIHL) in college students who are exposed to excessive exposure of loud noise is a well-recognized hazard. According to the National Institute on Deafness and Other Communication Disorders,  nearly 15% of American Adults (26 million people) between the ages of 20 and 69 years may have suffered permanent hearing damage as a result of hazardous noise exposure. NIHL may be caused by one-time or prolonged exposure to hazardous noise.  The intensity and duration of noise determines its level of harm. Depending on the intensity of the sound and the duration of the exposure, permanent sensorineural hearing loss and tinnitus (ringing in the ears) can ensue.  The intensity of sound is measured in decibels (dB) ranging from 0-180 dB(A). Reference points often cited include: Whisper (30 dB[A]); normal conversation (60 dB[A]); lawnmower (90 dB[A]); pneumatic drill (100 dB[A]); auto horn (115 dB[A]); jet engine (140 dB[A]); and rocket pad during launch (180 dB[A]). Consistent exposure to noise above 85 dB(A) causes an excess risk of hearing impairment  and every 3 dB increase in intensity halves the allowable exposure time. NIHL is 100% preventable and the most effective ways to avert NIHL are by avoiding or reducing hazardous noise.
Although, NIHL is well recognized as an occupational hazard, less recognized perhaps, are non-occupational leisure activities or environments that can put people of all ages at risk for hazardous noise exposure. ,,, In fact, college students are exposed regularly to noisy situations that exceed NIOSH recommended sound intensity limits.  For example: Music concerts, 92-110 dB(A);  nightclubs, 96-110 dB(A);  sporting events, 100-140 dB(A); , and use of portable listening devices, 90 -120 dB(A). , In addition, various studies have shown that young adult survey respondents (age 19-27 years) report frequent symptoms of NIHL including temporary hearing threshold shift (61%), tinnitus (43%), and/or ear pain (28%) after participating in leisure activities associated with hazardous noise levels. ,
Wearing hearing protection, turning down the volume, and use of headphones with personal listening devices have all been found to help reduce the risk of NIHL, but young adults typically do not choose to heed warnings to protect their hearing. , Some possible reasons for this include discomfort, design, lack of knowledge about NIHL, and peer pressure. , It has been proposed, however, that young adults would be more likely to take precautions if they understood that they were personally vulnerable to permanent NIHL.  It is imperative, therefore, to determine if prevention programs targeting young people can effectively convey the risk of NIHL and change their behavior to help reduce risk of NIHL. 
The purpose of this study is to determine the effectiveness of a specific message model, the Extended Parallel Process Model (EPPM)  in informing college students of the risks of NIHL associated with MP3 player usage and how use of headphones instead of earbud transducers can lower the risk. The focus of this research endeavor on headphone use with MP3 players (instead of lowering the volume) was used for the following reasons. First, students entering college in the USA are very likely to have been exposed to NIHL prevention messages that focus on reducing volume. An example of two recent national campaigns targeting young people to lower the volume while using portable listening devices include: "Turn It to the Left," a campaign promoted by the American Academy of Audiology;  and "Listen to Your Buds," a campaign promoted by the American Speech Language Hearing Association.  Second, incoming college students are unlikely to have been exposed to information detailing the difference in level between headphones and earbuds (~3 - 4 dB SPL less for headphones),  which, according to NIOSH standards, translates into a doubling of exposure time without increasing risk. In other words, by switching to headphones a student could listen to music for twice as long as when using earbuds. Finally, by selecting a message that teens are unlikely to have been exposed to while in primary school, the effect of a specific communication model or theory on health risk knowledge and intentions could be more accurately measured in this study.
Using an approach based in communication theory to achieve the goal of effectively informing young people of the threat they face for NIHL and then outlining effective ways to avert that threat seems particularly useful. This study used the EPPM  to design and evaluate the creation of brochures targeting college students that highlight threats associated with exposure to extreme noise and the efficacy with which hearing protection can be used. The current study represents the first step towards the goal of reducing risk of NIHL in undergraduate college students (typically 18 to 21 years of age) and simultaneously tests the ability of the EPPM to guide effective message development in this population. This goal is particularly important given that frequent or prolonged exposure to hazardous noise levels leads to permanent NIHL (and often constant tinnitus) the consequence of which can include decreased enjoyment and quality of life, depression, lower academic performance, and reduced productivity. ,, Furthermore, NIHL has implications for the communication abilities of those suffering from the ailment. Specifically, the impact of NIHL on communication includes: Difficulty hearing speech from a distance, in noisy and reverberant environments, and when it is softly spoken. , These symptoms are particularly problematic to college students who are expected to listen and learn in complex acoustic environments.
| The Extended Parallel Process Model|| |
The EPPM is a model of persuasive communication that posits a mechanism responsible for the effectiveness of a particular class of persuasive messages referred to as fear [Figure 1].  Fear appeals should present a relevant threat and then show a relatively simple response that will avert the threat, when enacted. To the extent that this combined presentation of threat and action instigates protective motivation, recipients reduce the threat in the way advocated by the appeal. , The EPPM has been successfully applied in a variety of public health contexts such as: Health promotion;  breast self exams;  alcohol warnings;  firearms;  drinking water quality;  cardiovascular disease;  tobacco use;  burn prevention;  and road safety. 
The first step in applying the EPPM is to present a threat of a hazard (a message) to the target population, in this case college students. According to the EPPM, perceptions of threat require that the young person receiving the message believe that the health hazard is severe and that they are susceptible to injury or harm. When applying the EPPM to leisure NIHL, the college student, for example, must perceive that hearing loss is a real threat to their well-being and that they are susceptible to hearing loss. If a threat is considered to be trivial or irrelevant, the model predicts that the student will not be motivated to process and accept the message and no action will be taken to prevent injury. 
Once a threat is established, the next step is to induce perceived efficacy, i.e. the perception that the college studnet can effectively avert the hazard by performing a simple task. According to the EPPM, efficacy is composed of two elements: Perceived self-efficacy and response efficacy. Perceived self-efficacy refers to a student's perception of his/her ability to perform the recommended response to avoid harm.  For college students, self-efficacy involves the perception that hearing protection devices (HPDs) are easy to use while engaging in leisure activities or headphone-style transducers are easy to use while listening to an MP3 player. Response efficacy refers to the perception that the recommended action to avert the threat (or injury) really will protect a college student from harm. Response efficacy refers to the perception that HPDs will adequately protect against future hearing loss due to hazardous leisure noise exposure and will protect against the negative personal and financial consequences that hearing loss can bring.
According to the EPPM, an effective prevention message is created with a balance between threat and efficacy. If both perceived threat and efficacy are high (i.e., real danger is recognized but perceived to be easily and effectively avoided), the EPPM predicts that college students will engage in a danger-control process. The danger-control process is desired because it motivates a young person to employ protection behaviors, which will prevent the negative outcomes associated with health hazard. In the application of leisure NIHL, if a danger-control process is engaged the college student would use hearing protection because he/she perceives it to be an effective way to avoid a real hazard with undesirable, negative consequences.
When the perceived threat far outweighs the perceived efficacy, the EPPM predicts that people will engage in a fear-control process. In the fear-control process, college students engage in denial, defensive avoidance, reactance, and ultimately reject the message in order to escape their fear Witte.  In relation to leisure NIHL, if the risk of NIHL is exaggerated or the steps to avoid injury too onerous, a fear-control process could be engaged and the young person would deny the severity of the noise hazard, minimize the consequences, and fail to use hearing protection while engaged in noisy leisure activities.
Therefore, because the brochures developed for this study were designed to include the EPPM components of severity, susceptibility, response efficacy, and self-efficacy messages the following hypotheses are offered:
Hypothesis 1: College students receiving messages describing the severity of NIHL and their susceptibility to NIHL will rate the threat of NIHL higher than those who do not receive the messages.
Hypothesis 2: College students receiving messages describing the efficacy of the recommended response and their self-efficacy of enacting the response will rate the efficacy of using hearing protection higher than those who do not receive the messages.
Hypothesis 3: College students who receive messages describing the threat of NIHL and the efficacy of the recommended response will have higher intent to use hearing protection than will those who do not receive the messages.
| Methods|| |
Undergraduates at a large southeastern university were convenience sampled at several different campus events, at several different points in time over the course of a semester until a sample of 200 subjects was obtained. No more than 40 subjects were grouped at any one event, and subjects were only allowed to participate in the study once. This sample was reduced to 176 after removing subjects with a self-reported history of hearing loss. This plan resulted in a sample that comprised 45% males and 55% females, where subjects were on average, Mean (M)=21.02 (Standard Deviation [SD]=3.62) years old and in their M=2.88 (SD=1.60) year in school.
Design and procedures
This study employed a randomized two group post-test only design where after selection for inclusion in the study, subjects were assigned randomly to either a control condition or treatment condition. In the control condition, subjects responded to a questionnaire designed to measure the EPPM variables of severity, susceptibility, response efficacy, self-efficacy, and behavioral intentions, as well as several demographic variables. After completing the questionnaire, these subjects received the brochure so as not to withhold any benefits of the brochure from the control condition. Subjects assigned to the treatment condition were presented with the same brochure first. After being given a few minutes to review the brochure, subjects responded to the same questionnaire that was administered to subjects in the control condition. In both conditions, subjects were instructed to ask the researcher if they had questions or needed clarification while completing the measure. None of the subjects needed any clarification.
Brochure and measurement
brochure was designed specifically for a target population of college students focussing on two different types of behavior. Specifically, the two qualitatively different activities of portable audio device use and student leisure activity engagement were considered. While the former is an individual behavior, the latter is a collective behavior. Consequently, people's motivations to engage in behaviors to protect hearing may be very different across the two behaviors resulting in the possibility of observing a differential pattern of results. Thus, each page of the brochure contained messages developed to make salient the threat college students face for NIHL and increase the perceived efficacy of behaviors that can avert the threat. To increase perceptions of threat, the first through third pages of the brochure presented students with messages hypothesized to increase perceptions of NIHL severity and susceptibility (e.g., "Hearing loss is permanent and irreversible!" and "College students are at high risk for hearing loss!"). These first three pages also summarize NIOSH  and Occupational Safety and Health Administration  standards for noise level exposure in addition to images of college student leisure activities (e.g., football games and nightclubs) and portable listening devices (e.g., CD and MP3 players) along with corresponding decibel levels.
The rear two pages of the brochure presented additional severity messages such as a listing of the side effects of NIHL (e.g., "Difficulty hearing conversations in groups." and "The need to use hearing aids."). These latter pages of the brochure, however, were dedicated primarily to messages designed to increase perceptions of response efficacy and self-efficacy by emphasizing the ease of preventing NIHL (e.g., "Hearing loss is one of the easiest hazards to prevent against.") and recommending behaviors that can avert the threat of NIHL (e.g., "You can begin to save your hearing loss today even if you have experienced some hearing loss by using formable foam plugs when in loud environments or using over-the-ear headphones when listening to your MP3 player"). The messages contained in the brochure were collaboratively designed by an EPPM expert and an audiologist with NIHL expertise. The NIHL data contained in the brochure was provided by an audiologist with expertise in the area. A copy of the brochure can be found in [Figure 2].
|Figure 2: Two-sided tri-fold brochure for college students. The panels are in order of pages 5, 6, 1, 2, 3, and 4 in order to accommodate a tri-fold order|
Click here to view
questionnaire was a combination of several demographic items and items designed to measure the EPPM variables of threat (i.e., severity and susceptibility), efficacy (i.e., response efficacy, and self-efficacy), and behavioral intentions adapted from the risk diagnosis behavior scale.  The demographic variables consisted of the subject's age, sex, year in school, whether they have a history of hearing loss, and how frequently they are exposed to environmental risk factors for NIHL. The EPPM variables were measured using five-point Likert-type scales ranging from "disagree strongly" to "agree strongly" where higher scores corresponded to greater item endorsement. Using this scaling format, the EPPM portion of the questionnaire was designed to measure the severity of NIHL, the susceptibility to NIHL, the response efficacy and self-efficacy of using formable ear plugs when in loud environments, the response efficacy and self-efficacy of using over-the-ear headphones when using MP3 players, and behavioral intentions to enact the recommended behaviors (i.e., using formable ear plugs when in loud environments and using over-the-ear headphones when using MP3 players). The Flesch-Kincaid grade level of the measure was 7.5. A copy of the measure can be found in Appendix A.-[Additional file 1]
| Results|| |
Measurement models - Validity and reliability
specified by the EPPM, efficacy was hypothesized to be a second-order unidimensional construct composed of the constructs of response efficacy (RE) and self-efficacy (SE). Threat was also hypothesized to be second-order unidimensional and composed of the constructs of susceptibility (SUS) and severity (SEV). RE, SE, SUS, and SEV were each measured by a set of five items. Only responses to the RE and SE items for wearing formable earplugs in loud environments were considered for these analyses. Analyses of responses to the RE and SE items for wearing over-the-ear headphones when listening to MP3 players were conducted separately.
After examining the response distributions of each item for normality, the 20 item inter-item correlation matrix was examined for items lacking internal consistency and parallelism.  These initial analyses resulted in eliminating numerous items, either because they lacked either internal consistency, parallelism, or both. The results of these analyses yielded four potentially content valid indicators of RE, four of SE, four of SEV, and five of SUS. The AMOS 16.0 Confirmatory Factor Analysis (CFA) algorithm using a maximum likelihood parameter estimation method was employed to test the hypothesized measurement model. The analysis revealed that the model was consistent with the data. Examination of the obtained parameter estimates presented in [Figure 3] reveal that they are all ample. Furthermore, examination of the modification indices suggested there were no localized areas of substantial model misfit. Global model fit was also acceptable: χ2 (114)=197.58, Goodness of Fit Index (GFI)=0.89, Root Mean Residual (RMR)=0.05. All of this is convincing evidence of this measure's validity.
Given the CFA results and considering the face validity of the items, two indices were created. An efficacy index was created by averaging (arithmetic mean) across responses to the four RE items and four SE items, whereas a threat index was created by averaging across the five SUS items and four SEV items. The distribution of the efficacy index was approximately normal: α=0.83, M=3.98, SD=0.61. The distribution of the threat index was also approximately normal: α=0.83, M=3.97, SD=0.58.
responses by subjects who self-reported as owning an MP3 player (N=137) and only responses to the RE and SE items for wearing over-the-ear headphones when listening to MP3 players were considered in these analyses. As was the case in the loud environment analysis, efficacy was hypothesized to be a second-order unidimensional construct composed of the constructs of RE and SE and threat was hypothesized to be second-order unidimensional and composed of the constructs of SUS and SEV for the MP3 player analysis. RE, SE, SUS, and SEV were each measured by a set of five items.
Using the same procedures of initial analysis as applied to the loud environment data, the same four potentially content valid indicators of RE, SE, and SEV and five potentially content valid indicators of SUS emerged in the MP3 data. CFA also using the same procedures as applied earlier was performed to test the hypothesized measurement model with the MP3 data. The analysis revealed that the model was consistent with the data. The obtained parameter estimates presented in [Figure 4] are all ample, and the modification indices suggested there were no localized areas of substantial model misfit. Again, global model fit was acceptable: χ2 (114)=225.21, GFI=0.85, RMR=0.06. Again, this evidence is indicative of the measure's validity.
Based on the evidence from the CFA, efficacy and threat indices were created for the MP3 data. The efficacy index was created by averaging across responses to the four RE items and four SE items, whereas a threat index was created by averaging across the five SUS items and four SEV items. The efficacy index was distributed approximately normal: α=0.92, M=3.42, SD=0.85. The distribution of the threat index was also approximately normal: α=0.86, M=3.91, SD=0.62.
effects of the EPPM brochure on threat, efficacy, and intent to wear ear plugs in loud environments are presented in [Table 1]. Hypothesis one predicted that college students who received severity and susceptibility of hearing loss messages would rate the threat of hearing loss higher than students who did not receive the messages. The results of a multiple regression analysis showed that when threat was regressed onto experimental condition, current use of hearing protection, and weekly hours spent in loud environments, as well as all two- and three-way interaction terms, only the experimental condition emerged as a substantial and statistically significant predictor, β=0.26 (t (168)=3.40, P<0.05). Therefore, the analyses revealed that students receiving the severity and susceptibility messages in the EPPM brochure perceived a greater threat (M=4.00, SD=0.54) than those in the control group (M=3.72, SD=0.60). These results indicate that the loud environment data were indeed consistent with hypothesis one.
Hypothesis two predicted that college students who received messages with response efficacy and self-efficacy components would rate efficacy of using ear plugs in loud environments higher than those who did not receive the messages. The results of a multiple regression analysis showed that when efficacy was regressed onto experimental condition, current use of hearing protection, and weekly hours spent in loud environments, as well as all two and three-way interaction terms, only the experimental condition emerged as a substantial and statistically significant predictor, β=0.21 (t (168)=2.66, P<0.05). Therefore, students receiving the self-efficacy and response efficacy messages in the EPPM brochure perceived greater efficacy (M=4.09, SD=0.59) than those who did not have exposure to the brochure in the control group (M=3.86, SD=0.62). These analyses were consistent with hypothesis two.
Finally, hypothesis three predicted that students who received messages with threat and efficacy components would have more intent to use hearing protection than those who did not receive the messages. Intent was measured by a single item focusing on the intent to use ear plugs when in loud environments in the future. The distribution of the item was slightly positively skewed: M=1.85, SD=0.97. Comparing levels of intent for EPPM brochure receiving students (M=2.02, SD=1.05) against control students (M=1.65, SD=0.82) revealed a slight difference between groups in the hypothesized direction. A multiple regression analysis showed, however, that when regressing intentions onto experimental condition, current use of hearing protection, and weekly hours spent in loud environments, as well as all two-way and three-way interaction terms, current use of hearing protection emerged as the substantial statistically significant predictor, β=0.50 (t (168)=6.25, P<0.05). These analyses indicated that the data were not consistent with hypothesis 3.
MP3 Players. [Table 1] also presents the effects of the EPPM brochure on threat, efficacy, and intent to wear over-the-ear headphones when using an MP3 player for subjects who self-reported as owning an MP3 player (N=137). Hypothesis 1 predicted that college students who received severity and susceptibility of hearing loss messages would rate the threat of hearing loss more highly than those who did not receive the messages. The results of a multiple regression analysis indicated that the MP3 player data were consistent with hypothesis one. The analysis showed that when threat was regressed onto experimental condition, current use of over-the-ear headphones, and daily hours spent listening to MP3 players, as well as all two-way and three-way interaction terms, only the experimental condition emerged as a substantial and statistically significant predictor, β=0.25 (t(129)=2.86, P<0.05). Students exposed to the severity and susceptibility messages in the EPPM brochure perceived a greater threat (M=4.03, SD=0.56) than those in the control group (M=3.75, SD=0.65).
Hypothesis 2 predicted that college students who received messages with response efficacy and self-efficacy components would rate efficacy of using over-the-ear headphones when listening to their MP3 players higher than students who did not receive the messages. The results of a multiple regression analysis showed that when perceived efficacy was regressed onto experimental condition, current use of over-the-ear headphones, daily hours spent listening to MP3 players, as well as all two-way and three-way interaction terms, the experimental condition emerged as a substantial and statistically significant predictor, β=0.44 (t(129)=5.46, P<0.05). Thus, students receiving the efficacy messages in the EPPM brochure perceived greater efficacy (M=3.73, SD=0.87) than those who did not have exposure to the brochure in the control group (M=3.02, SD=0.63). These analyses were consistent with hypothesis 2.
Finally, hypothesis three predicted that students receiving messages with threat and efficacy components would have greater intent to use over-the-ear headphones than would those who did not receive the messages. One item, focusing on intentions to use over-the-ear headphones while listening to MP3 players in the future, was used to measure intent. The distribution of the item was slightly positively skewed: M=2.58, SD=1.40. Comparing levels of intent to use over-the-ear headphones when listening to an MP3 player in the future for students receiving the EPPM brochure (M=2.70, SD=1.29) against control students (M=2.41, SD=1.52) revealed a slight difference between groups in the hypothesized direction. Regressing intentions onto experimental condition, current use of over-the-ear headphones, and daily hours spent listening to MP3 players, as well as all two-way and three-way interaction terms in a multiple regression analysis and controlling for current use of over-the-ear headphones (β=0.80, t (129)=15.58, P<0.05), revealed that intentions were predicted by the exposure to the EPPM brochure (β=0.17, t (129)=3.45, P<0.05). According to these analyses, the data were consistent with hypothesis 3.
| Discussion|| |
The analyses show that the data are consistent with the hypotheses predicted by the EPPM for the MP3 player data but not for the loud environment data. Exposure to the brochure created perceptions of increased threat of NIHL resulting from loud environments as well as MP3 players. Brochure exposure also resulted in increased perceptions of the efficacy of using ear plugs as well as the efficacy of using over-the-ear headphones to avert the threat of NIHL. Exposure to the brochure only increased behavioral intentions to use over-the-ear headphones, however, and not behavioral intentions to use ear plugs. Put differently, the brochure produced statistically significant changes in only the threat and efficacy variables for the loud environment data whereas the changes produced in all three variables for the MP3 data were ample and statistically significant.
Although it could be argued that the lack of effect on behavioral intentions to use ear plugs when in loud environments is due to a failure of the EPPM, an alternative argument could be made. It is possible that wearing ear plugs is a response that no matter how effective the intervention used, will not be changed easily, if at all. Close examination of [Table 1] shows that behavioral intentions in the control condition were almost at the bottom of the scale (M=1.65, SD=0.82). Perhaps when intentions to engage in the recommended response are initially so low, even if the brochure has an impact on threat (β=0.26) and efficacy (β=0.21), the intention baseline was below a threshold. In other words, the impact of threat and efficacy on intentions when the baseline is below the threshold is smaller than the impact of threat and efficacy on intentions when the baseline is above the threshold. Such threshold effects, or more precisely effect nonlinearity, result from the presence of a moderator variable. , Therefore, although wearing ear plugs when in loud environments was seen as efficacious, the level of a moderator variable such as a characteristic of the recommended response (i.e., comfort, acceptability, accessibility, etc.) may have attenuated the effect on intentions. Similar effects have been observed in research focused on encouraging college students to wear bicycle helmets while riding their bikes on campus.  This study found that bicycle helmets were not seen as cool or acceptable causing persuasive interventions to fail.
Thus, it is possible that the recommended response of wearing earplugs in loud environments is perceived as particularly uncomfortable, unacceptable, inaccessible, or given the social nature of the behavior just not cool, even though the behavior was simultaneously seen as efficacious. As a result, although the brochure brought about increases in the perceptions of threat and created a substantial increase in the efficacy of the recommended response, an unobserved moderator variable may have attenuated the effect on intentions causing people to engage in fear-control processes that resulted in people failing to engage in the response recommended to avert the threat. 
On the other hand, in addition to being seen as efficacious, the recommended response of wearing over-the-ear headphones when listening to an MP3 player could have been perceived as more comfortable, acceptable, accessible, or cool. As a result of the higher level of one of these moderator variables the effect on intentions was not attenuated and people engaged in danger-control processes and intentions to wear over-the-ear headphones when using MP3 players in the future were increased. 
| Limitations|| |
Future research would be well served including a measure of the potential moderator variable(s) that could be responsible for this study's results. Based on the findings of this study, the likely moderator variable is related generally to the desirability of performing the recommended response. This study did not include such a measure and it resulted in the need to make inferences of desirability based on the efficacy, threat, and intention perceptions that were measured. For example, although perceived as efficacious, the recommended response of wearing ear plugs in loud environments could have been seen as undesirable, perhaps because it would be perceived to get in the way of hearing sounds in the environment that need to be heard. Having a way to measure issues of response desirability would allow the researcher to diagnose its status as a moderator variable, determine how much impact it had on EPPM processes, and in turn control for or reduce that impact.
Future examinations of the EPPM would also be well served at improving the quality of instrumentation used to measure the EPPM components. Specifically, there are two main areas of improvement for future studies. The measures employed in this study were improved versions of the the risk diagnosis behavior scale.  The measures in this study refined the wording of the original items to improve their face validity and clarity. The measures in this study also included additional items to measure each of the variables. Although, fewer items on a questionnaire can be advantageous pragmatically, too few a items can result in problems assessing the measure's validity and reliability. Therefore, paying a small price in practicality allows for the disposal of items displaying misfit during testing of the measurement model, but also an increase in reliability estimates as long as the items are of equal quality to the existing items. 
The strategy of revised and additional items in this study increased the average instrument reliability from previous research using earlier versions of the measures employed.  Even so, the amounts of measurement error in the measures of threat, efficacy, and the unknown amount in intent (e.g., intent was measured with a single item) in this study were somewhat high for applied settings where high-stakes policy decisions might be made. Therefore, it is recommended that the response scales for each item be increased from five-point to seven-point response scales, which will serve to relieve the equivalent of a restriction in range problem attenuating the reliability coefficient.  Further, at least one additional intent item needs to be added so a reliability estimate can be obtained.
| Practical applications|| |
This study contributes to the substantial body of knowledge resulting from the application of the EPPM. Topics addressed through application of the EPPM include hearing loss;  vaccine risk;  farm hazards;  general health promotion;  risk behaviors;  breast self-exams;  alcohol warnings;  firearms;  drinking water;  terror management;  eating disorders;  rape;  cardiovascular disease;  genetic testing and tobacco use;  burn prevention;  road safety;  and jail encounter programs.  The results of this research add hearing loss prevention in college students to this list of populations who have benefitted from application of the EPPM to health hazards that they face.
This study along with the others shows clearly that the EPPM can be deployed to develop effective persuasive messages in the health context. When the message development phase of a communication effort begins, at a minimum, the messages will have to include threat and efficacy components. Moreover, the threat component must convince the intended message recipients of their susceptibility to the condition by making arguments that the condition can and does happen to people just like them. In addition, the threat component must include messages that convey the severity of the condition. In other words, the message must convince recipients that they are at risk of grave potential harm. Leaving the message recipients without an efficacious means to deal with the threat, however, may lead to counterproductive responses intended to control fear.  Therefore, the message must also recommend responses that can avert the threat (response efficacy) and that the recipients are capable of performing such responses (self-efficacy).
The present research demonstrated, however, that for some possible recommended responses, baseline intentions will be so low that it will be unlikely for any message to increase them to a level high enough to motivate the intended message recipient to act. Therefore, for these situations brochures can be helpful to inform of the threat and efficacy of a response but will not be sufficient to improve intentions to engage in the response. In these cases, practitioners will need to search for alternative responses that have similar levels of efficacy but higher levels of baseline intent. For situations where only one recommended response is possible, further research must uncover the potential moderators that may be attenuating the EPPM's effect on intentions and design messages aimed at reducing their impact on the target's intentions to engage in the recommended response.
| References|| |
|1.||National Institute on Deafness and Other Communication Disorders. (2010). Noise induced hearing loss. Available from: http://www.nidcd.nih.gov/health/hearing/noise.asp [Last Accessed on 2010, May 20]. |
|2.||Bohne BA, Harding GW, Lee SC. Death pathways in noise-damaged outer hair cells. Hear Res 2007;223:61-70. |
|3.||Daniel E. Noise and hearing loss: A review. J Sch Health 2007;77:225-31. |
|4.||National Institute for Occupational Safety and Health. (1998). Criteria for a recommended standard: Occupational noise exposure. Available from: http://www.cdc.gov/niosh/98-126pd.html [Last accessed on 2006, June 1]. |
|5.||Morata T. Young people: Their noise and music exposure and risk of hearing loss. Int J Audiol 2007;46:411-2. |
|6.||Mostafapour SP, Lahargoue K, Gates GA. Noise-induced hearing loss in young adults: The role of personal listening devices and other sources of leisure noise. Laryngoscope 1998:108:1832-9. |
|7.||Nistar AS, Kieszak SM, Holmes AE, Estaban E, Rubin C, Brody DJ. Estimated prevalence of noise-induced hearing threshold shifts among children ages 6 to 19 years of age: The third national health and nutrition survey, 1988-1994. Am Acad Pediatr 2001;108:40-3. |
|8.||Rabinowitz PM, Slade MD, Galusha D, Dixon-Ernst C, Cullen MR. Trends in the prevalence of hearing loss among young adults entering an industrial workforce 1985 to 2004. Ear Hear 2006;27:369-75. |
|9.||Peters RJ. The role of hearing protectors in leisure noise. Noise Health 2003:5:47-55. |
|10.||Bogoch II, House RA, Kudla L. Perceptions about hearing protection and noise-induced hearing loss of attendees of rock concerts. Can J Public Health 2005;96:69-72. |
|11.||Bray A, Szymañski M, Mills R. Noise induced hearing loss in dance music disc jockeys and an examination of sound levels in nightclubs. J Laryngol Otol 2004;118:123-8. |
|12.||Hodgetts WE, Liu R. Can hockey playoffs harm your hearing? CMAJ 2006;175:1541-2. |
|13.||Van Campen LE, Morata T, Kardous CA, Gwin K, Wallingford KM, Dallaire J, et al. Ototoxic occupational exposures for a stock car racing team: I, noise survey. J Occup Environ Hyg 2007;2:383-90. |
|14.||Fligor BJ, Cox LC. Output levels of commercially available portable compact disc players and the potential risk to hearing. Ear Hear 2004;25:513-27. |
|15.||Hodgetts WE, Rieger JM, Szarko RA. The effects of listening environment and earphone style on preferred listening levels of normal hearing adults using an MP3 player. Ear Hear 2007;28:290-7. |
|16.||Chung JH, Des Roches CM, Meunier J, Eavey RD. Evaluation of noise-induced hearing loss in young people using a web-based survey technique. Pediatrics 2005;115:861-7. |
|17.||Holmes AE, Widén SE, Carver CL, White LL. Perceived hearing status and attitudes toward noise in young adults. Am J Audiol 2007;16:S182-9. |
|18.||Vogel I, Brug J, van der Ploeg CP, Raat H. Young people's exposure to loud music: A summary of the literature. Am J Prev Med 2007;33:124-33. |
|19.||Witte K. Putting the fear back into fear appeals: The extended parallel process model. Commun Monogr 1992;59:329-49. |
|20.||Baron Witte American Academy of Audiology 2008. New public education campaign encourages kids to turn it to the left. Available from: http://www.turnittotheleft.com/news/pm05.htm [Last accessed on 2010 May 18]. |
|21.||American Speech Language Hearing Association 2008. Making kids safe in sound. Available from: http://www.listentoyourbuds.org/[Last accessed on 2010 May 18]. |
|22.||Griest SE, Folmer RL, Hal Martin W. Effectiveness of 'dangerous decibels,' a school-based hearing loss prevention program. Am J Audiol 2007;16:S165-81. |
|23.||Kujawa SG, Liberman MC. Acceleration of age-related hearing loss by early noise exposure: Evidence of misspent youth. J Neurosci 2006;26:2115-23. |
|24.||Ricketts TA, Hornsby BW. Distance and reverberation effects on directional benefit. Ear Hear 2003;24:472-82. |
|25.||Payton KL, Uchanski RM, Braida LD. Intelligibility of conversational speech and clear speech in noise and reverberation for listeners with normal and impaired hearing. J Acoust Soc Am 1993;95:1851-92. |
|26.||Witte K. Fear control and danger control: A test of the extended parallel process model (EPPM). Commun Monogr 1994;61:113-34. |
|27.||Witte K. Fear as motivator, fear as inhibitor: Using the extended parallel process model to explain fear appeal successes and failures. In: Andersen PA, Guerrero LK, editors. Handbook of communication and emotion: Research, theory, applications, and contexts. San Diego, CA: Academic Press; 1998. |
|28.||Dutta-Bergman MJ. Theory and practice in health communication campaigns: A critical interrogation. Health Commun 2005;18:103-22. |
|29.||Kaplowitz SA, Campo S, Chiu WT. Cancer patients' desires for communication of prognosis information. Health Commu 2002;14:221-41. |
|30.||Campo S, Poulos G, Sipple JW. Prevalence and profiling: Hazing among college students and points of intervention. Am J Health Behav 2005;29:137-49. |
|31.||Meyer G, Roberto AJ, Atkin CK. A radio-based approach to promoting gun safety: Process and outcome evaluation implications and insights. Health Commun 2003;15:299-318. |
|32.||Johnson BB. Do reports on drinking water quality affect customers' concerns? Experiments in report content. Risk Anal 2003;23:985. |
|33.||McKay DL, Berkowitz JM, Blumberg JB, Goldberg JP. Communicating cardiovascular disease risk due to elevated homocysteine levels: Using the EPPM to develop print materials. Health Educ Behav 2004;31:355-71. |
|34.||Sanderson SC, Wardle J. Will genetic testing for complex diseases increase motivation to quit smoking? Anticipated reactions in a survey of smokers. Health Educ Behav 2005;32:640-53. |
|35.||Shani E, Ayalon A, Hammad IA, Sikron F. What picture is worth a thousand words? A comparative evaluation of a burn prevention programme by type of medium in Israel. Health Promot Int 2003;18:361-71. |
|36.||Simons-Morton BG, Hartos JL, Leaf WA, Preusser DF. Increasing parent limits on novice young drivers: Cognitive mediation of the effect of persuasive messages. J Adolesc Res 2006;21:83-105. |
|37.||Occupational Safety and Health Administration. (2006). Occupational safety and health standards: Occupational noise exposure. Available from: http://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=STANDARDS&p_id =9735 [Last Accessed on 2011, March 16] |
|38.||Witte K, Meyer G, Martell D. Effective health risk messages: A step-by-step guide. Thousand Oaks, CA: Sage; 2001. |
|39.||Hunter JE, Gerbing DW. Unidimensional measurement, second order factor analysis, and causal models. Res Organ Behav 1982;4:267-320. |
|40.||Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986;51:1173-82. |
|41.||Cohen J, Cohen P, West SG, Aiken LS. Applied multiple regression/correlation analysis for the behavioral sciences 3 rd ed. Mahwah, NJ: Lawrence Erlbaum Associates; 2003. |
|42.||Mitchell MM, Banas J. Why won't people wear bicycle helmets?: An examination of counter-arguments. Paper presented at the annual meeting of the National Communication Association. New Orleans, LA: National Communication Association; 2002 |
|43.||Nunnally JC, Bernstein IH. Psychometric theory. 3 rd ed. New York, NY: McGraw Hill; 1994. |
|44.||Smith SW, Rosenman K, Kotowski MR, Glazer E, McFeters C, Keesecker N, et al. Using the EPPM to create and evaluate the effectiveness of brochures to increase the use of hearing loss protection in farmers and landscape workers. J Appl Commun Res 2008;36:200-18. |
|45.||Murray-Johnson L, Witte K, Patel D, Orrego V, Zuckerman C, Maxfield AM, et al. Using the Extended Parallel Process Model to prevent noise-induced hearing loss among coal miners in Appalachia. Health Educ Behav 2004;31:741-55. |
|46.||Ball LK, Evans G, Bostrom A. Risky business: Challenges in vaccine risk communication. Pediatrics 1998;101:453-58. |
|47.||Cole HP. Cognitive-behavioral approaches to farm community safety education: A conceptual analysis. J Agric Saf Health 2002;8:145-59. |
|48.||Rimal RN, Real K. Perceived risk and efficacy beliefs as motivators of change: Use of the risk perception attitude (RPA) framework to understand health behaviors. Hum Commun Res 2003;29:370-99. |
|49.||Shehryar O, Hunt DM. A terror management perspective on the persuasiveness of fear appeals. J Consum Psychol 2005;15:275-87. |
|50.||Smalec JL, Klingle RS. Bulimia interventions via interpersonal influence: The role of threat and efficacy in persuading bulimics to seek help. J Behav Med 2000;23:37-57. |
|51.||Morrison K. Motivating women and men to take protective action against rape: Examining direct and indirect persuasive fear appeals. Health Commun 2005;18:237-56. |
|52.||Windell JO, Allen Jr JS. An application of fear appeal messages to enhance the benefits of a jail encounter program for youthful offenders. Youth Violence Juv Justice 2005;3:388-94. |
Michael R Kotowski
School of Communication Studies, University of Tennessee - Knoxville, 293 Communications Building, Knoxville, TN 37996
Source of Support: None, Conflict of Interest: None
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
|This article has been cited by|
||A Threat- and Efficacy-Based Framework to Understand Confidence in Vaccines among the Public Health Workforce
| ||Daniel Barnett,Nicole Errett,Lainie Rutkow |
| ||Vaccines. 2013; 1(2): 77 |
|[Pubmed] | [DOI]|