International Journal of Audiology 2015; Early Online: 1–7
Original Article
Prediction of the use of conventional hearing aids in Korean adults with unilateral hearing impairment Dong-Hee Lee* & Heil Noh† Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
*Department
of Otolaryngology-Head and Neck Surgery, Uijeongbu St. Mary’s Hospital, The Catholic University of Korea, Uijeongbu City, Gyeonggi-do, Korea, and †Department of Otolaryngology-Head and Neck Surgery, St. Vincent’s Hospital, The Catholic University of Korea, Suwon City, Gyeonggi-do, Korea
Abstract Objective: To determine the use of a hearing aid at six months post-fitting and to evaluate the predictors of its ongoing use in Korean adults with unilateral hearing impairment (HI). Design: Retrospective study at a secondary referral hospital over a 15-year period. Study sample: This study analysed 119 adults with unilateral HI who had been recommended for hearing amplification (55 men and 64 women, mean age, 58.0 ⫾ 11.7 years). Six months after the fitting, all of the participants were surveyed regarding subsequent decisions and actions about obtaining hearing aids. Results: General uptake rate for a hearing aid was 68.1% (58.0% of participants surveyed were successful users, and 10.1% were intermittent users). The most significant parameter associated with hearing-aid use was social and/or work activities (R2 ⫽ 0.457), and the significant predictors for successful hearing-aid use were social and/or work activities and method of signal processing (discriminatory power ⫽ 56.3%). Conclusions: Six months post-fitting, 68.1% of Korean adults with unilateral HI who had agreed to try a hearing aid continued to use it regularly. The predictors for hearing-aid use six months post-fitting included social and/or work activities and digital signal processing.
Key Words: Hearing aids; unilateral hearing loss; hearing disorders For individuals with a hearing impairment (HI), the reduced ability to detect, recognize and localize sounds in everyday life affects physical, cognitive, behavioral, and social factors, as well as quality of life. This is true for bilateral and unilateral HI. However, in the case of individuals with unilateral HI, the need to wear a hearing aid can be less apparent for some of the aforementioned factors because of normal hearing sensitivity in the better ear (Arlinger, 2003). Despite the benefits of hearing-aid use, only 20–25% of hearingimpaired individuals who would benefit from a hearing aid actually use one—also known as uptake (Chia et al, 2007). The Swedish Council on Health Technology Assessment in 2003 investigated the unmet need for hearing rehabilitation and found a slightly better estimate of uptake. Approximately 50% of adults who would benefit from hearing aids had been fitted (Arlinger, 2003). Attempting to explain the reasons why individuals may not uptake a hearing aid has been undertaken in several studies (e.g. Schumacher & Carruth, 1997; Bertoli et al, 2009; Cox et al, 2011; Laplante-Levesque et al, 2012). A more general explanation for such a decision has been provided by Stewart and DeMarco (Stewart & DeMarco, 2005). These authors reported on the economic theory of patient decision-making reaching the conclusions that (1) fully informed, rational patients will attempt
to maximize the net benefit of treatments, and (2) the net benefit is the difference between treatment benefits (such as improved hearing) and treatment burdens (such as money or discomfort). Treatment benefits and burdens differ among individuals, and the treatment point at which a net benefit is maximized varies among individuals. For many people, (3) the net benefit is maximized at a point that is well below the ideal treatment level recommended by physicians. For example, some individuals with bilateral HI will only purchase one hearing aid, although two are recommended, because of the limited financial burden, thus attempting to maximize his/her net benefit. Therefore, the ability to predict the uptake of the hearing aid has potential clinical benefits. Practical methods for predicting which individuals with unilateral HI will be satisfied with a hearing aid could result in more cost-effective auditory rehabilitation and higher patient satisfaction. To date, studies on unilateral HI have been generally confined to pediatric audiology and have involved bone-anchored hearing aids (Snik et al, 2008; Briggs et al, 2011; Roman et al, 2011), but this study addressed successful uptake of traditional amplification (i.e. BTE, ITC, CIC) in the adult population. Also, this is the first study to analyse this topic using a Korean population.
Correspondence: Dong-Hee Lee, Department of Otolaryngology-Head and Neck Surgery, Uijeongbu St. Mary’s Hospital, The Catholic University of Korea, 271 Cheonbo Street. Uijeongbu City, Gyeonggi-do, 480–717, Korea. E-mail: [email protected] (Received 1 October 2013; accepted 17 February 2015) ISSN 1499-2027 print/ISSN 1708-8186 online © 2015 British Society of Audiology, International Society of Audiology, and Nordic Audiological Society DOI: 10.3109/14992027.2015.1023902
2
D. Lee & H. Noh
Abbreviations
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
dB HL HI MCL PTA SDS
Decibel hearing level Hearing impairment Most comfortable level Pure-tone audiogram Speech discrimination score
The aim of this study was to determine how many unilaterally hearing-impaired Koreans used a hearing aid successfully. An additional aim was to identify any pre-fitting variables that could predict which unilaterally hearing-impaired Korean adults would use a hearing aid for better hearing. The Kaplan-Meier ‘time-to-event’ curve was analysed to illustrate the time course as well as the probability of having a hearing aid, as affected by any significant variable.
Materials and Methods Subjects and Procedures The study involved a detailed retrospective chart review of 130 Korean patients with unilateral HI who purchased a hearing aid at one secondary referral hospital over a 15-year period from 1999 to 2013. The inclusion criteria were as follows: (1) 19 years of age or older at the time of fitting; (2) unilateral stable HI that could benefit from hearing amplification; (3) self-reported good or excellent physical and mental health; (4) adequate cognitive competence to respond to questionnaires; and (5) willingness to pay for and to wear the hearing aid for a unilateral fitting (Korean National Health Insurance does not cover hearing aids). The exclusion criteria were as follows: (1) previous experience with a hearing aid; (2) observed or reported neurologic or psychiatric disorders; (3) retrocochlear lesions; (4) fluctuating HI; and (5) an average of four thresholds at 0.5-, 1-, 2and 4-kHz frequencies ⬎ 25 dB HL in the unaffected ear. In general, for prescriptive formula, the half-gain rule was applied to analogue hearing aids and NAL was applied to digital hearing aids. Eleven participants failed to respond to follow-up or phone calls, so the data for the remaining 119 participants (55 men and 64 women; mean age: 58.0 ⫾ 11.7 years; 49 right and 70 left ears) were analysed. Demographic data for the participants in this study are presented in Table 1. Social and/or work activities were notified in 65 participants (35 men and 30 women; mean age: 53.6 ⫾ 12.1 years) and not in 54 participants (20 men and 34 women; mean age: 63.0 ⫾ 9.1 years). Mean duration of HI was 13.3 ⫾ 15.6 years. Annoying tinnitus was notified in 64 participants (28 men and 36 women; mean age: 57.3 ⫾ 11.5 years) and not in 55 participants (27 men and 28 women; mean age: 59.0 ⫾ 12.0 years). The collected data included gender, age, social lifestyle and occupation, duration of HI and presence/absence of persistent annoying tinnitus. Data were collected primarily by chart review, a survey during the visit, and a contact interview. Social and/or work activities were defined as regular jobs or social activities and the latter was defined as the following five activities: (1) undergraduate/graduate students or regular attendance at educational or training courses; (2) regular volunteer or charity work; (3) regular caregiver for sick or disabled persons; (4) taking care of children or grandchildren all day; and (5) taking part in a community related organization. The degree and type of HI were classified using the average of four thresholds at 0.5, 1, 2, and 4 kHz according to the criteria set by the World Health Organization (WHO) (Mathers et al, 2003) and
Table 1. Demographic data for all included users (N ⫽ 119) (expressed as the mean ⫾ SD). Age (years) Gender Male Female PTA (dB HL)* of the affected ear(s) SDS (%) at MCL of affected ear(s) Types of HI Sensorineural Conductive Mixed Degree of HI Mild Moderate Severe Profound Configuration of HI Flat Rising/up-slope Sloping/down-slope Low frequency High frequency Hearing-aid type BTE BTE open fit ITC CIC Signal processing Digital, programmable Digital, non-programmable Analogue, non-programmable Analogue, programmable Number of channels One Two Three Four Eight Compression Linear compression Non-linear compression Output compression
58.0 ⫾ 11.7 (19–81) 46.2% 53.8%
(n ⫽ 55) (n ⫽ 64) 63.6 ⫾ 14.9 60.0 ⫾ 20.2
67.2% 12.6% 20.2%
(n ⫽ 80) (n ⫽ 15) (n ⫽ 24)
7.6% 31.9% 48.7% 11.8%
(n ⫽ 9) (n ⫽ 38) (n ⫽ 58) (n ⫽ 14)
67.2% 4.2% 21.0% 0.8% 6.7%
(n ⫽ 80) (n ⫽ 5) (n ⫽ 25) (n ⫽ 1) (n ⫽ 8)
5.0% 3.4% 19.3% 72.3%
(n ⫽ 6) (n ⫽ 4) (n ⫽ 23) (n ⫽ 86)
86.6% 1.7% 10.1% 1.7%
(n ⫽ 103) (n ⫽ 2) (n ⫽ 12) (n ⫽ 2)
18.5% 34.5% 0.8% 36.1% 10.1%
(n ⫽ 22) (n ⫽ 41) (n ⫽ 1) (n ⫽ 43) (n ⫽ 12)
0.8% 96.6% 2.5%
(n ⫽ 1) (n ⫽ 115) (n ⫽ 3)
PTA: pure-tone audiogram; SDS: speech discrimination score; MCL: most comfortable level. *The average of the four thresholds at 0.5-, 1-, 2-, and 4-kHz frequencies on a pure-tone audiogram was calculated.
Brannstrom and Wennerstrom (Brannstrom & Wennerstrom, 2010). The configuration of HI was classified as flat, sloping (down-slope or ski-slope), rising (up-slope or reverse-slope), low frequency or high frequency according to Stach’s classification (1998). The types and features of the hearing aids were chosen as appropriate for each participant considering the following: degree of HI and speech discrimination score (SDS) at the most comfortable level (MCL), tinnitus, education level, and job status. The age of identification of the HI and the duration of HI were also considered. The type of hearing aid was determined to be BTE (behindthe-ear), BTE open fit, ITC (in-the-canal), or CIC (completely-inthe-canal). The signal processing technique was categorized as digital, non-programmable digital, analogue, or programmable analogue.
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
Unilateral hearing aids in Koreans. Each hearing aid was categorized as one, two, three, four or eight depending upon the channels it used. Hearing aids also were divided into groups for those that used linear compression, non-linear compression or output compression. Six months after being fitted for the hearing aid, all of the participants or their parents were surveyed about subsequent decisions on whether or not to keep a hearing aid on. All of the participants responded to a three-item survey containing the following questions: (1) Have you used the hearing aid regularly for at least eight hours per day? (Yes or No). (2) If yes, how satisfied are you? (1 to 4). (3) If no, why did you stop using the hearing aid? The second question evaluated hearing-aid satisfaction using a four-point scale; ‘very satisfied’ and ‘rather satisfied’ were categorized as ‘satisfied,’ and ‘rather unsatisfied’ or ‘very unsatisfied’ were classified as ‘dissatisfied.’ The third question was free-response question. In this study, the outcomes for hearing-aid use were placed into three groups: successful users, intermittent users, and a failure group. Successful users in this study were defined as patients who wore the hearing aid satisfactorily for at least eight hours per day for six months or more post-fitting. Intermittent users were satisfied with the hearing aid, but wore it for less than eight hours per day. Failure was defined as voluntary cessation of hearing-aid use within the first six months post-fitting. Demographic data of three groups are presented in Table 2.
3
chi-squared test (gender, social and/or work activities, type of HI, configuration of HI, hearing-aid type, method for signal processing, number of channels, and amplification method). Differences in the numerical values among the three outcomes for hearing-aid use were examined using the Kruskal-Wallis test for multiple independent samples (age, duration of HI, PTA, and SDS). If a significant difference among these three was present in the Kruskal-Wallis test, the Mann Whitney U-test was used to analyse the significance of each intergroup. Furthermore, nominal regression analysis and discriminant analysis were used to identify associations between potential explanatory variables and hearing-aid use outcomes. Time-event analysis using the Kaplan-Meier method was used to analyse the probability of not having a hearing aid in relation to the time after the fitting, where the time was defined as years after the hearing-aid fitting and the event as the regular use of the hearing aid. This method is known to be one of the best options to be used to measure the fraction of subjects censored for an event after a defined point. On the Kaplan-Meier curve, the lengths of the horizontal lines along the X-axis of serial times represent the censored event (termination of a hearing aid in this study) for that interval and the vertical distances between horizontals are important because they illustrate the change in cumulative probability as the curve advances.
Results Uptake of a hearing aid 6 months post-fitting
Ethics The study design and experimental protocols were approved by the Institutional Review Board of St. Vincent’s Hospital, at the Catholic University of Korea.
In this study, the general uptake rate for a hearing aid was 68.1% in unilaterally hearing-impaired Korean adults. This study showed that 58.0% (n ⫽ 69) were in the successful user group; 10.1% (n ⫽ 12) made up the intermittent user group, and 31.9% (n ⫽ 38) were placed in the failure group because of a decision to not wear a hearing aid. The mean follow-up period was 4.9 ⫾ 3.8 years after the initial fitting (maximum, 13.9 years).
Statistical analysis Probability values of p ⬍ 0.05 were considered statistically significant. All of the statistical analyses were performed using IBM®SPSS® statistics (version 19, IBM Corporation, New York). The data were expressed as the mean ⫾ standard deviation. Statistical analysis compared the data among the successful users, intermittent users and failure groups. Differences in ordinary values among the three outcomes for hearing-aid use were tested using the
Predictors of user status six months post-fitting as either successful, intermittent, or failed The first question in this study was which factor for patients with unilateral HI could make successful use of a hearing aid possible. To find out any parameter which could explain the successful use of a hearing aid at six months post-fitting, nominal regression analysis was conducted for gender, social and/or work activities, tinnitus,
Table 2. Demographic data of successful users, intermittent users, and a failure group (expressed as the mean ⫾ SD)
Age (years) Gender (male : female) PTA (dB HL)* SDS (%) at MCL Types of HI (sensorineural : conductive : mixed) Degree of HI (mild : moderate : severe : profound) Configuration of HI (flat : rising : sloping : low frequency : high frequency) Hearing-aid type (BTE : BTE open fit : ITC : CIC) Signal processing (DP : DnP : AP : AnP) Number of channels (one : two : three : four : eight) Compression (linear : non-linear : output)
Successful (N ⫽ 69)
Intermittent (N ⫽ 12)
Failure (N ⫽ 38)
57.2 ⫾ 11.4 31:38 65.2 ⫾ 14.3 58.7 ⫾ 20.5 47:5:17 3:23:34:9 48:4:14:0:3 4:2:7:56 67:0:2:0 5:23:0:34:7 0.69:0
65.6 ⫾ 6.3 5:7 62.5 ⫾ 11.7 48.5 ⫾ 26.7 7:2:3 0:5:7:0 7:0:3:0:2 1:0:2:9 12:0:0:0 0:5:0:4:3 0:12:0
57.3 ⫾ 12.9 19:19 61.2 ⫾ 16.6 67.9 ⫾ 13.9 26:8:4 6:17:10:5 25:1:8:1:3 1:2:14:21 24:2:10:2 17:13:1:5:2 1:34:3
PTA: pure tone audiogram; SDS: speech discrimination score; MCL: most comfortable level; DP: digital programmable; DnP: digital non-programmable, AP: analogue programmable, AnP: analogue non-programmable. *The average of the four thresholds at 0.5-, 1-, 2-, and 4-kHz frequencies on a pure tone audiogram was calculated.
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
4
D. Lee & H. Noh
type of HI, configuration of HI, hearing-aid type, method of signal processing, and number of channels. The most significant parameter related with the outcome for hearing-aid use was participation in social and/or work activities (p ⫽ 0.020, R2 ⫽ 0.457). To predict the successful use of a hearing aid, a discriminant analysis (stepwise method) was conducted for age, gender, social and/or work activities, tinnitus, type of HI, configuration of HI, hearing-aid type, method of signal processing, and number of channels. The significant predictors were social and/or work activities and method of signal processing; the discriminate function equation was D ⫽ ⫺ 1.079 ⫹ (⫺ 1.131 * social and/or work activities) ⫹ (1.337 * method of signal processing) (discriminatory power ⫽ 56.3%). That is, individuals with more social/work related activities are more likely to be successful hearing-aid users and using digital-programmable signal processing will increase the likelihood of successful use. Table 3 showed the difference between real data and data predicted using the equation of discriminate function. Time-event analysis using the Kaplan-Meier method determined that the first drop of hearing-aid use was within the first year, and a second downward trend occurred approximately 4–5 years after the fitting, regardless of whether the patient participated in social and/or work activities (Figure 1, A). This drop pattern of rate of successful use of a hearing aid was also found in both groups regardless of social and/or work activities (Figure 1, B).
Further differentiation of successful hearing-aid users (Figure 2) The next step was to find out any difference between the successful user and the failure groups. There was a significant difference in outcomes for hearing-aid use among the hearing-aid types (p ⫽ 0.043). BTE and CIC were used more frequently in the successful user group, and ITC was prescribed more often in the failure group (four successful users vs. one failure users for BTE; 56 successful users vs. 21 failure users for CIC; and seven successful users vs. 14 failure users for ITC). This study showed that the degree of HI was significantly different according to hearing aid type (p ⬍ 0.0001). Figure 2 demonstrated that the average PTA of BTE was significantly higher than those of BTE open fit and CIC (p ⬍ 0.0001 and p ⫽ 0.004, respectively). It also showed that the average PTA of BTE open fit was significantly lower than those of ITC and CIC (p ⫽ 0.009 and p ⫽ 0.041, respectively). Average PTA of BTE, BTE open fit, ITC, and CIC were 83.5 ⫾ 13.0 dB HL, 41.9 ⫾ 18.0 dB HL, 67.6 ⫾ 13.2 dB HL, and 62.2 ⫾ 13.6 dBHL, respectively. No significant differences in speech discrimination scores were found among the hearing-aid types (p ⫽ 0.775). Mean SDS of BTE, BTE open fit, ITC and CIC were 52.4 ⫾ 28.1%, 55.0 ⫾ 14.1%, 67.7 ⫾ 15.7% and 59.1 ⫾ 20.5%, respectively. Table 3. Comparison of the outcomes of hearing-aid use using discriminant analysis. The equation of discriminate function was D ⫽ ⫺ 1.079 ⫹ (⫺ 1.131 * social and/or work activities) ⫹ (1.337 * method of signal processing); its discriminatory power was 56.3%. Original results
Predicted results Successful user Intermittent user Failure Total
Successful user
Intermittent user
Failure
Total
46 21 2 69
5 7 0 12
8 16 14 38
59 44 16 119
Figure 1. Kaplan-Meier curve. The x-axis represents the time (years) after the hearing-aid fitting, and the y-axis indicates the probability of having a hearing aid at that time. (A) Overall rate of successful use of a hearing aid. (B) Rate of successful use of a hearing aid according to social and/or work activities. The solid line represents the presence of social and/or work activities, and the dotted line indicates their absence.
There was a significant difference in outcomes for hearing-aid use among the methods of signal processing (p ⬍ 0.0001). Digital processing was used more frequently in the successful user group, and analogue processing was found more frequently in the failure group (67 successful users vs. 26 failure users for digital signal processing; and two successful users vs. 12 failure users for analogue signal processing). There was also a significant difference in outcomes for hearingaid use according to the number of channels (p ⬍ 0.0001). Four- or eight-channel devices were used more frequently in the successful user group, and one- or two-channel device was prescribed more frequently in the failure group (five successful users vs. 17 failure
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
Unilateral hearing aids in Koreans.
Figure 2. Boxplots of PTA (average of four thresholds at 0.5-, 1-, 2- and 4-kHz frequencies on a pure-tone audiogram) according to the type of hearing aid. The length of the box means the interquartile range (25th∼75th percentile), a line drawn across the box means the median, and two whiskers represents the maximum and minimum, respectively.
users for one-channel; 23 successful users vs. 13 failure users for two-channel; 34 successful users vs. five failure users for four-channel; and seven successful users vs. two failure users for eight-channel). There was a significant difference in the outcomes for hearing-aid use between groups with members that participated in social and/or work activities, including a job, and those who did not (p ⫽ 0.002). Successful users were more likely to participate in social and/or work activities (22 successful users vs. 25 failure users for persons without social and/or work activities; but 47 successful users vs. 13 failure users for persons with social and/or work activities). This study also showed that the CIC was used in 78.5% of participants with social and/or work activities, and the CIC and ITC were used in 64.8% and 31.5% of participants without social and/or work activities, respectively. There was no significant difference in the average PTA among outcomes for hearing-aid use (p ⫽ 0.403), but a significant difference in SDS was found among outcomes for hearing aid use (p ⫽ 0.033). The SDS score of intermittent users was significantly lower than that of the failure group (58.6 ⫾ 22.0% in successful users, 43.1 ⫾ 29.8% in intermittent users, and 64.8 ⫾ 19.8% in the failure group; p ⫽ 0.043). No significant differences in gender, age, duration of HI, degree of HI, type of HI, configuration of HI, or tinnitus were found among outcomes for hearing aid use (p ⫽ 0.833, p ⫽ 0.154, p ⫽ 0.451, p ⫽ 0.610, p ⫽ 0.158, p ⫽ 0.646, and p ⫽ 0.106, respectively). No significant differences in gender, degree of HI, tinnitus, type of HI or configuration of HI were found between groups with and without social and/or work activities (p ⫽ 0.096, p ⫽ 0.586, p ⫽ 0.356, p ⫽ 0.771, and p ⫽ 0.682, respectively). In this study, 63.6% of males and 46.9% of females responded that they regularly participated in social and/or work activities.
Discussion This study showed that 68.1% of unilaterally hearing-impaired Korean adults consistently wore a hearing aid and were satisfied
5
with its function. This suggested that a considerable proportion (31.9%) of patients with unilateral hearing impairment failed to feel a hearing aid was beneficial. Considering the advantages of binaural hearing or better hearing by unilateral fitting, these findings were unexpected. This study revealed some variables that were significantly different among successful users, intermittent users and failure groups: hearing-aid type, method of signal processing, number of channels, and social and/or work activities. There were no significant differences in gender, age, duration of HI, degree of HI, type of HI, configuration of HI, and tinnitus. This study also found that social and/or work activities and method of signal processing were significant predictors of a successful user. In their study of hearing-aid preferences in bilaterally symmetrical HI, Cox et al (2011) found that the preference for binaural hearing (using two hearing aids) rather than monaural hearing (using one hearing aid) was predictable from a combination of greater perceived daily problems and binaural advantage, as well as reduced binaural imbalance. They also found that audiometric hearing loss and auditory lifestyle were not predictive of aid preferences (Cox et al, 2011). Recent research has provided clinicians with valuable information regarding the predictors of intervention uptake and intervention outcomes. In a study of 153 uni- or bilaterally hearing-impaired participants 50 years of age and older with two intervention options (hearing aid(s) or communication programs), Laplante-Levesque et al (2012) found that applications for subsidized hearing services, higher socioeconomic status, lesser communication self-efficacy, and greater contemplation stage of change increased the uptake of hearing aids. They also determined that higher socioeconomic status, greater initial self-reported hearing disability, lower pre-contemplation stage of change, greater action stage of change, lower chance locus of control, greater hearing disability perceived by others, and self-increased intervention outcomes were predictors of successful intervention outcomes. These findings suggest that socio-behavioral factors, not audiological factors, influenced the uptake and outcome of hearing aids and emphasized the active role of clinicians who manage and counsel hearing-impaired persons. Cox et al (2011) have also reported the reasons for participants’ preferences for either one or two hearing aids. Because of improved speech comprehension in quiet situations as well as in noisy ones, the naturalness/comfort of their own voice, and the inconvenience in telephone use, patients preferred one hearing aid over two. However, patients who preferred two hearing aids reported feeling balanced, experiencing clarity of sound and feeling comfort/capable/ safe (Cox et al, 2011). Together with these reports, our results show that clinicians must evaluate the patients’ primary need, their stage of change, their lifestyle and their audiological environment, and counsel them thoroughly and in depth. Successful decisions regarding hearing aids and their uptake requires clinicians to understand their patients comprehensively and also to consider their economic status. For example, for unilaterally hearing-impaired persons, resolving the discomfort of daily life may not be the main target of a hearing aid. In this case, clinicians should weigh the advantages of binaural hearing and the disadvantages, as well as the expense associated with a hearing aid. Readers should not overlook the reality that binaural hearing can produce better real-world outcomes than monaural hearing. People who report greater hearing problems in daily life, who experience greater binaural loudness summation, and whose ears are more equivalent in dichotic listening are more likely to prefer to use two hearing aids (Cox et al, 2011). Clinicians must do their best to
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
6
D. Lee & H. Noh
make sure that patients have accurate and clear information in the decision-making process. Although a patient s own decision must be respected, there are some caveats associated with the choice of monaural hearing rather than binaural hearing. Persons who prefer long-term monaural hearing are potentially vulnerable to the deprivation effect in their hearing-impaired ear. In this condition, the ability to recognize words declines in the hearing-impaired ear over time, even though audiogram thresholds remain symmetrical in both ears (Silman et al, 1984; Neuman, 1996; Hurley, 1999). This study indicated that ITC was used more frequently in the failure group. Considering that the hearing threshold of ITC and CIC users was similar in this study, this result suggests that patients with moderate or severe HI may be more suitable for CIC. It may be because people with more social and/or work activities preferred the appearance of the smaller aid, or some other unknown explanation. This study also showed that analogue signal processing and fewer channels were prescribed more frequently in the failure group. Generally, analogue signal processing and fewer channels are not suitable for not-flat HI. Even in the 80 cases of flat HI in this study, analogue signal processing was used in 32.0% (n ⫽ 8), and 1 or 2 channels were used in 80.0% (n ⫽ 20) of the failure group (n ⫽ 25). These results were in contrast to the findings of the successful/intermittent user groups (n ⫽ 55), in which analogue signal processing was used in 3.6% (n ⫽ 2) of the cases and 1 or 2 channels were used in 43.6% (n ⫽ 24) of the cases. These findings suggest that even for flat HI, analogue signal processing and fewer channels should not be prescribed based solely on the patient’s economic grounds. In the past, a wrong recommendation of hearing aids had a negative impact on many patients and their families in cases where the choice of a hearing aid was decided by matching the price with the patients’ financial condition to determine what they could afford. A patient’s economic status often creates a dilemma regarding the cost versus performance of hearing aids. Therefore, an application for subsidized hearing services has been identified as an uptake predictor of hearing aids (Laplante-Levesque et al, 2012). In this study, Kaplan-Meier survival analysis revealed that the first decrease in hearing-aid use occurred within one year after the initial fitting, which was similar to previous reports (Schumacher & Carruth, 1997). However, a second drop was found to occur approximately four to five years after the fitting, which is coincident with the time when patients are due to update their hearing aid(s). We surmise that many persons with HI contemplate whether to purchase a new hearing aid when it is time to replace their first hearing aid with a new one. The major limitation of this study was the possibility of confounding factors because it was a retrospective study. This study was not randomized or blind. This retrospective study might not exclude recall bias. A second limitation of this study included the lack of more hearing-specific assessments on the performance of a hearing aid, such as international outcome inventory for hearing aids (IOI-HA), abbreviated profile of hearing aid benefit (APHAB), or the device oriented subjective outcome (DOSO) scale. This study evaluated participant’s satisfaction about his or her hearing aid using just a three-item survey with four-point scale. A final limitation was the long span of the enrollment period in this study because hearing-aid technology is always improving rapidly. In this study, the bias might result from diverse types of the hearing aid because the technology of the hearing aid has been moving too fast during a 15-year period.
This study showed that the SDS score of intermittent users was significantly lower than that of the failure group (p ⫽ 0.033). However, considering that the distribution of SDS was too dispersed in intermittent and successful users and our recommendation for hearing aids was based on a conventional formula in all patients, these results should be interpreted cautiously.
Conclusion On the basis of eight hours or longer of hearing-aid use and selfreported satisfaction, general uptake rate for a hearing aid was 68.1% in unilaterally hearing-impaired Korean adults. Successful users comprised 58.0% of the study population, and intermittent users made up 10.1%. Pre-fitting variables found to be predictors of hearing-aid uptake were social and/or work activities and method of signal processing. This study revealed that persons with social and/or work activities had better odds of being a successful hearing-aid user. Variables that were significantly different according to hearing-aid outcomes were the hearing-aid type, the method of signal processing, the number of channels, and social and/or work activities.
Acknowledgements The authors thank Mu Gil Heo (Audiologist, Jeil Hearing Aid Center, Seoul, Korea), who helped with data collection. The authors contributed equally to this project and should be considered co-first authors Declaration of interest: The authors have no conflicts of interest to report. The authors alone are responsible for the content and writing of this paper.
References Arlinger S. 2003. Negative consequences of uncorrected hearing loss: A review. Int J Audiol, 42, Suppl 2, 2S17–20. Arlinger S.B., Lagerbring C., Leijon A., Rosenhall U. & Scherstén T. 2003. Hearing aids for adults: Benefits and costs. The SBU Board of Directors and the SBU Scientific Advisory Committee. Bertoli S., Staehelin K., Zemp E., Schindler C., Bodmer D. et al. 2009. Survey on hearing aid use and satisfaction in Switzerland and their determinants. Int J Audiol, 48, 183–95. Brannstrom K.J. & Wennerstrom I. 2010. Hearing aid fitting outcome: Clinical application and psychometric properties of a Swedish translation of the international outcome inventory for hearing aids (IOI-HA). J Am Acad Audiol, 21, 512–21. Briggs L., Davidson L. & Lieu J.E. 2011. Outcomes of conventional amplification for pediatric unilateral hearing loss. Ann Otol Rhinol Laryngol, 120, 448–54. Chia E.M., Wang J.J., Rochtchina E., Cumming R.R., Newall P. et al. 2007. Hearing impairment and health-related quality of life: The Blue Mountains Hearing Study. Ear Hear, 28, 187–95. Cox R.M., Schwartz K.S., Noe C.M. & Alexander G.C. 2011. Preference for one or two hearing AIDS among adult patients. Ear Hear, 32, 181–97. Hurley R.M. 1999. Onset of auditory deprivation. J Am Acad Audiol, 10, 529–34. Laplante-Levesque A., Hickson L. & Worrall L. 2012. What makes adults with hearing impairment take up hearing aids or communication programs and achieve successful outcomes? Ear Hear, 33, 79–93. Mathers C., Smith A. & Concha M. 2003. Global burden of hearing loss in the year 2000. GBD 2000 Working Paper. Geneva: World Health Organization.
Unilateral hearing aids in Koreans.
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
Neuman A.C. 1996. Late-onset auditory deprivation: A review of past research and an assessment of future research needs. Ear Hear, 17, 3S–13S. Roman S., Nicollas R. & Triglia J.M. 2011. Practice guidelines for boneanchored hearing aids in children. Eur Ann Otorhinolaryngol Head Neck Dis, 128, 253–8. Schumacher D.U. & Carruth J.A. 1997. Long-term use of hearing aids in patients with presbyacusis. Clin Otolaryngol Allied Sci, 22, 430–3.
7
Silman S., Gelfand S.A. & Silverman C.A. 1984. Late-onset auditory deprivation: Effects of monaural versus binaural hearing aids. J Acoust Soc Am, 76, 1357–62. Snik A., Leijendeckers J., Hol M., Mylanus E. & Cremers C. 2008. The boneanchored hearing aid for children: Recent developments. Int J Audiol, 47, 554–9. Stewart D.O. & DeMarco J.P. 2005. An economic theory of patient decisionmaking. J Bioeth Inq, 2, 153–164.
Original Article
Prediction of the use of conventional hearing aids in Korean adults with unilateral hearing impairment Dong-Hee Lee* & Heil Noh† Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
*Department
of Otolaryngology-Head and Neck Surgery, Uijeongbu St. Mary’s Hospital, The Catholic University of Korea, Uijeongbu City, Gyeonggi-do, Korea, and †Department of Otolaryngology-Head and Neck Surgery, St. Vincent’s Hospital, The Catholic University of Korea, Suwon City, Gyeonggi-do, Korea
Abstract Objective: To determine the use of a hearing aid at six months post-fitting and to evaluate the predictors of its ongoing use in Korean adults with unilateral hearing impairment (HI). Design: Retrospective study at a secondary referral hospital over a 15-year period. Study sample: This study analysed 119 adults with unilateral HI who had been recommended for hearing amplification (55 men and 64 women, mean age, 58.0 ⫾ 11.7 years). Six months after the fitting, all of the participants were surveyed regarding subsequent decisions and actions about obtaining hearing aids. Results: General uptake rate for a hearing aid was 68.1% (58.0% of participants surveyed were successful users, and 10.1% were intermittent users). The most significant parameter associated with hearing-aid use was social and/or work activities (R2 ⫽ 0.457), and the significant predictors for successful hearing-aid use were social and/or work activities and method of signal processing (discriminatory power ⫽ 56.3%). Conclusions: Six months post-fitting, 68.1% of Korean adults with unilateral HI who had agreed to try a hearing aid continued to use it regularly. The predictors for hearing-aid use six months post-fitting included social and/or work activities and digital signal processing.
Key Words: Hearing aids; unilateral hearing loss; hearing disorders For individuals with a hearing impairment (HI), the reduced ability to detect, recognize and localize sounds in everyday life affects physical, cognitive, behavioral, and social factors, as well as quality of life. This is true for bilateral and unilateral HI. However, in the case of individuals with unilateral HI, the need to wear a hearing aid can be less apparent for some of the aforementioned factors because of normal hearing sensitivity in the better ear (Arlinger, 2003). Despite the benefits of hearing-aid use, only 20–25% of hearingimpaired individuals who would benefit from a hearing aid actually use one—also known as uptake (Chia et al, 2007). The Swedish Council on Health Technology Assessment in 2003 investigated the unmet need for hearing rehabilitation and found a slightly better estimate of uptake. Approximately 50% of adults who would benefit from hearing aids had been fitted (Arlinger, 2003). Attempting to explain the reasons why individuals may not uptake a hearing aid has been undertaken in several studies (e.g. Schumacher & Carruth, 1997; Bertoli et al, 2009; Cox et al, 2011; Laplante-Levesque et al, 2012). A more general explanation for such a decision has been provided by Stewart and DeMarco (Stewart & DeMarco, 2005). These authors reported on the economic theory of patient decision-making reaching the conclusions that (1) fully informed, rational patients will attempt
to maximize the net benefit of treatments, and (2) the net benefit is the difference between treatment benefits (such as improved hearing) and treatment burdens (such as money or discomfort). Treatment benefits and burdens differ among individuals, and the treatment point at which a net benefit is maximized varies among individuals. For many people, (3) the net benefit is maximized at a point that is well below the ideal treatment level recommended by physicians. For example, some individuals with bilateral HI will only purchase one hearing aid, although two are recommended, because of the limited financial burden, thus attempting to maximize his/her net benefit. Therefore, the ability to predict the uptake of the hearing aid has potential clinical benefits. Practical methods for predicting which individuals with unilateral HI will be satisfied with a hearing aid could result in more cost-effective auditory rehabilitation and higher patient satisfaction. To date, studies on unilateral HI have been generally confined to pediatric audiology and have involved bone-anchored hearing aids (Snik et al, 2008; Briggs et al, 2011; Roman et al, 2011), but this study addressed successful uptake of traditional amplification (i.e. BTE, ITC, CIC) in the adult population. Also, this is the first study to analyse this topic using a Korean population.
Correspondence: Dong-Hee Lee, Department of Otolaryngology-Head and Neck Surgery, Uijeongbu St. Mary’s Hospital, The Catholic University of Korea, 271 Cheonbo Street. Uijeongbu City, Gyeonggi-do, 480–717, Korea. E-mail: [email protected] (Received 1 October 2013; accepted 17 February 2015) ISSN 1499-2027 print/ISSN 1708-8186 online © 2015 British Society of Audiology, International Society of Audiology, and Nordic Audiological Society DOI: 10.3109/14992027.2015.1023902
2
D. Lee & H. Noh
Abbreviations
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
dB HL HI MCL PTA SDS
Decibel hearing level Hearing impairment Most comfortable level Pure-tone audiogram Speech discrimination score
The aim of this study was to determine how many unilaterally hearing-impaired Koreans used a hearing aid successfully. An additional aim was to identify any pre-fitting variables that could predict which unilaterally hearing-impaired Korean adults would use a hearing aid for better hearing. The Kaplan-Meier ‘time-to-event’ curve was analysed to illustrate the time course as well as the probability of having a hearing aid, as affected by any significant variable.
Materials and Methods Subjects and Procedures The study involved a detailed retrospective chart review of 130 Korean patients with unilateral HI who purchased a hearing aid at one secondary referral hospital over a 15-year period from 1999 to 2013. The inclusion criteria were as follows: (1) 19 years of age or older at the time of fitting; (2) unilateral stable HI that could benefit from hearing amplification; (3) self-reported good or excellent physical and mental health; (4) adequate cognitive competence to respond to questionnaires; and (5) willingness to pay for and to wear the hearing aid for a unilateral fitting (Korean National Health Insurance does not cover hearing aids). The exclusion criteria were as follows: (1) previous experience with a hearing aid; (2) observed or reported neurologic or psychiatric disorders; (3) retrocochlear lesions; (4) fluctuating HI; and (5) an average of four thresholds at 0.5-, 1-, 2and 4-kHz frequencies ⬎ 25 dB HL in the unaffected ear. In general, for prescriptive formula, the half-gain rule was applied to analogue hearing aids and NAL was applied to digital hearing aids. Eleven participants failed to respond to follow-up or phone calls, so the data for the remaining 119 participants (55 men and 64 women; mean age: 58.0 ⫾ 11.7 years; 49 right and 70 left ears) were analysed. Demographic data for the participants in this study are presented in Table 1. Social and/or work activities were notified in 65 participants (35 men and 30 women; mean age: 53.6 ⫾ 12.1 years) and not in 54 participants (20 men and 34 women; mean age: 63.0 ⫾ 9.1 years). Mean duration of HI was 13.3 ⫾ 15.6 years. Annoying tinnitus was notified in 64 participants (28 men and 36 women; mean age: 57.3 ⫾ 11.5 years) and not in 55 participants (27 men and 28 women; mean age: 59.0 ⫾ 12.0 years). The collected data included gender, age, social lifestyle and occupation, duration of HI and presence/absence of persistent annoying tinnitus. Data were collected primarily by chart review, a survey during the visit, and a contact interview. Social and/or work activities were defined as regular jobs or social activities and the latter was defined as the following five activities: (1) undergraduate/graduate students or regular attendance at educational or training courses; (2) regular volunteer or charity work; (3) regular caregiver for sick or disabled persons; (4) taking care of children or grandchildren all day; and (5) taking part in a community related organization. The degree and type of HI were classified using the average of four thresholds at 0.5, 1, 2, and 4 kHz according to the criteria set by the World Health Organization (WHO) (Mathers et al, 2003) and
Table 1. Demographic data for all included users (N ⫽ 119) (expressed as the mean ⫾ SD). Age (years) Gender Male Female PTA (dB HL)* of the affected ear(s) SDS (%) at MCL of affected ear(s) Types of HI Sensorineural Conductive Mixed Degree of HI Mild Moderate Severe Profound Configuration of HI Flat Rising/up-slope Sloping/down-slope Low frequency High frequency Hearing-aid type BTE BTE open fit ITC CIC Signal processing Digital, programmable Digital, non-programmable Analogue, non-programmable Analogue, programmable Number of channels One Two Three Four Eight Compression Linear compression Non-linear compression Output compression
58.0 ⫾ 11.7 (19–81) 46.2% 53.8%
(n ⫽ 55) (n ⫽ 64) 63.6 ⫾ 14.9 60.0 ⫾ 20.2
67.2% 12.6% 20.2%
(n ⫽ 80) (n ⫽ 15) (n ⫽ 24)
7.6% 31.9% 48.7% 11.8%
(n ⫽ 9) (n ⫽ 38) (n ⫽ 58) (n ⫽ 14)
67.2% 4.2% 21.0% 0.8% 6.7%
(n ⫽ 80) (n ⫽ 5) (n ⫽ 25) (n ⫽ 1) (n ⫽ 8)
5.0% 3.4% 19.3% 72.3%
(n ⫽ 6) (n ⫽ 4) (n ⫽ 23) (n ⫽ 86)
86.6% 1.7% 10.1% 1.7%
(n ⫽ 103) (n ⫽ 2) (n ⫽ 12) (n ⫽ 2)
18.5% 34.5% 0.8% 36.1% 10.1%
(n ⫽ 22) (n ⫽ 41) (n ⫽ 1) (n ⫽ 43) (n ⫽ 12)
0.8% 96.6% 2.5%
(n ⫽ 1) (n ⫽ 115) (n ⫽ 3)
PTA: pure-tone audiogram; SDS: speech discrimination score; MCL: most comfortable level. *The average of the four thresholds at 0.5-, 1-, 2-, and 4-kHz frequencies on a pure-tone audiogram was calculated.
Brannstrom and Wennerstrom (Brannstrom & Wennerstrom, 2010). The configuration of HI was classified as flat, sloping (down-slope or ski-slope), rising (up-slope or reverse-slope), low frequency or high frequency according to Stach’s classification (1998). The types and features of the hearing aids were chosen as appropriate for each participant considering the following: degree of HI and speech discrimination score (SDS) at the most comfortable level (MCL), tinnitus, education level, and job status. The age of identification of the HI and the duration of HI were also considered. The type of hearing aid was determined to be BTE (behindthe-ear), BTE open fit, ITC (in-the-canal), or CIC (completely-inthe-canal). The signal processing technique was categorized as digital, non-programmable digital, analogue, or programmable analogue.
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
Unilateral hearing aids in Koreans. Each hearing aid was categorized as one, two, three, four or eight depending upon the channels it used. Hearing aids also were divided into groups for those that used linear compression, non-linear compression or output compression. Six months after being fitted for the hearing aid, all of the participants or their parents were surveyed about subsequent decisions on whether or not to keep a hearing aid on. All of the participants responded to a three-item survey containing the following questions: (1) Have you used the hearing aid regularly for at least eight hours per day? (Yes or No). (2) If yes, how satisfied are you? (1 to 4). (3) If no, why did you stop using the hearing aid? The second question evaluated hearing-aid satisfaction using a four-point scale; ‘very satisfied’ and ‘rather satisfied’ were categorized as ‘satisfied,’ and ‘rather unsatisfied’ or ‘very unsatisfied’ were classified as ‘dissatisfied.’ The third question was free-response question. In this study, the outcomes for hearing-aid use were placed into three groups: successful users, intermittent users, and a failure group. Successful users in this study were defined as patients who wore the hearing aid satisfactorily for at least eight hours per day for six months or more post-fitting. Intermittent users were satisfied with the hearing aid, but wore it for less than eight hours per day. Failure was defined as voluntary cessation of hearing-aid use within the first six months post-fitting. Demographic data of three groups are presented in Table 2.
3
chi-squared test (gender, social and/or work activities, type of HI, configuration of HI, hearing-aid type, method for signal processing, number of channels, and amplification method). Differences in the numerical values among the three outcomes for hearing-aid use were examined using the Kruskal-Wallis test for multiple independent samples (age, duration of HI, PTA, and SDS). If a significant difference among these three was present in the Kruskal-Wallis test, the Mann Whitney U-test was used to analyse the significance of each intergroup. Furthermore, nominal regression analysis and discriminant analysis were used to identify associations between potential explanatory variables and hearing-aid use outcomes. Time-event analysis using the Kaplan-Meier method was used to analyse the probability of not having a hearing aid in relation to the time after the fitting, where the time was defined as years after the hearing-aid fitting and the event as the regular use of the hearing aid. This method is known to be one of the best options to be used to measure the fraction of subjects censored for an event after a defined point. On the Kaplan-Meier curve, the lengths of the horizontal lines along the X-axis of serial times represent the censored event (termination of a hearing aid in this study) for that interval and the vertical distances between horizontals are important because they illustrate the change in cumulative probability as the curve advances.
Results Uptake of a hearing aid 6 months post-fitting
Ethics The study design and experimental protocols were approved by the Institutional Review Board of St. Vincent’s Hospital, at the Catholic University of Korea.
In this study, the general uptake rate for a hearing aid was 68.1% in unilaterally hearing-impaired Korean adults. This study showed that 58.0% (n ⫽ 69) were in the successful user group; 10.1% (n ⫽ 12) made up the intermittent user group, and 31.9% (n ⫽ 38) were placed in the failure group because of a decision to not wear a hearing aid. The mean follow-up period was 4.9 ⫾ 3.8 years after the initial fitting (maximum, 13.9 years).
Statistical analysis Probability values of p ⬍ 0.05 were considered statistically significant. All of the statistical analyses were performed using IBM®SPSS® statistics (version 19, IBM Corporation, New York). The data were expressed as the mean ⫾ standard deviation. Statistical analysis compared the data among the successful users, intermittent users and failure groups. Differences in ordinary values among the three outcomes for hearing-aid use were tested using the
Predictors of user status six months post-fitting as either successful, intermittent, or failed The first question in this study was which factor for patients with unilateral HI could make successful use of a hearing aid possible. To find out any parameter which could explain the successful use of a hearing aid at six months post-fitting, nominal regression analysis was conducted for gender, social and/or work activities, tinnitus,
Table 2. Demographic data of successful users, intermittent users, and a failure group (expressed as the mean ⫾ SD)
Age (years) Gender (male : female) PTA (dB HL)* SDS (%) at MCL Types of HI (sensorineural : conductive : mixed) Degree of HI (mild : moderate : severe : profound) Configuration of HI (flat : rising : sloping : low frequency : high frequency) Hearing-aid type (BTE : BTE open fit : ITC : CIC) Signal processing (DP : DnP : AP : AnP) Number of channels (one : two : three : four : eight) Compression (linear : non-linear : output)
Successful (N ⫽ 69)
Intermittent (N ⫽ 12)
Failure (N ⫽ 38)
57.2 ⫾ 11.4 31:38 65.2 ⫾ 14.3 58.7 ⫾ 20.5 47:5:17 3:23:34:9 48:4:14:0:3 4:2:7:56 67:0:2:0 5:23:0:34:7 0.69:0
65.6 ⫾ 6.3 5:7 62.5 ⫾ 11.7 48.5 ⫾ 26.7 7:2:3 0:5:7:0 7:0:3:0:2 1:0:2:9 12:0:0:0 0:5:0:4:3 0:12:0
57.3 ⫾ 12.9 19:19 61.2 ⫾ 16.6 67.9 ⫾ 13.9 26:8:4 6:17:10:5 25:1:8:1:3 1:2:14:21 24:2:10:2 17:13:1:5:2 1:34:3
PTA: pure tone audiogram; SDS: speech discrimination score; MCL: most comfortable level; DP: digital programmable; DnP: digital non-programmable, AP: analogue programmable, AnP: analogue non-programmable. *The average of the four thresholds at 0.5-, 1-, 2-, and 4-kHz frequencies on a pure tone audiogram was calculated.
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
4
D. Lee & H. Noh
type of HI, configuration of HI, hearing-aid type, method of signal processing, and number of channels. The most significant parameter related with the outcome for hearing-aid use was participation in social and/or work activities (p ⫽ 0.020, R2 ⫽ 0.457). To predict the successful use of a hearing aid, a discriminant analysis (stepwise method) was conducted for age, gender, social and/or work activities, tinnitus, type of HI, configuration of HI, hearing-aid type, method of signal processing, and number of channels. The significant predictors were social and/or work activities and method of signal processing; the discriminate function equation was D ⫽ ⫺ 1.079 ⫹ (⫺ 1.131 * social and/or work activities) ⫹ (1.337 * method of signal processing) (discriminatory power ⫽ 56.3%). That is, individuals with more social/work related activities are more likely to be successful hearing-aid users and using digital-programmable signal processing will increase the likelihood of successful use. Table 3 showed the difference between real data and data predicted using the equation of discriminate function. Time-event analysis using the Kaplan-Meier method determined that the first drop of hearing-aid use was within the first year, and a second downward trend occurred approximately 4–5 years after the fitting, regardless of whether the patient participated in social and/or work activities (Figure 1, A). This drop pattern of rate of successful use of a hearing aid was also found in both groups regardless of social and/or work activities (Figure 1, B).
Further differentiation of successful hearing-aid users (Figure 2) The next step was to find out any difference between the successful user and the failure groups. There was a significant difference in outcomes for hearing-aid use among the hearing-aid types (p ⫽ 0.043). BTE and CIC were used more frequently in the successful user group, and ITC was prescribed more often in the failure group (four successful users vs. one failure users for BTE; 56 successful users vs. 21 failure users for CIC; and seven successful users vs. 14 failure users for ITC). This study showed that the degree of HI was significantly different according to hearing aid type (p ⬍ 0.0001). Figure 2 demonstrated that the average PTA of BTE was significantly higher than those of BTE open fit and CIC (p ⬍ 0.0001 and p ⫽ 0.004, respectively). It also showed that the average PTA of BTE open fit was significantly lower than those of ITC and CIC (p ⫽ 0.009 and p ⫽ 0.041, respectively). Average PTA of BTE, BTE open fit, ITC, and CIC were 83.5 ⫾ 13.0 dB HL, 41.9 ⫾ 18.0 dB HL, 67.6 ⫾ 13.2 dB HL, and 62.2 ⫾ 13.6 dBHL, respectively. No significant differences in speech discrimination scores were found among the hearing-aid types (p ⫽ 0.775). Mean SDS of BTE, BTE open fit, ITC and CIC were 52.4 ⫾ 28.1%, 55.0 ⫾ 14.1%, 67.7 ⫾ 15.7% and 59.1 ⫾ 20.5%, respectively. Table 3. Comparison of the outcomes of hearing-aid use using discriminant analysis. The equation of discriminate function was D ⫽ ⫺ 1.079 ⫹ (⫺ 1.131 * social and/or work activities) ⫹ (1.337 * method of signal processing); its discriminatory power was 56.3%. Original results
Predicted results Successful user Intermittent user Failure Total
Successful user
Intermittent user
Failure
Total
46 21 2 69
5 7 0 12
8 16 14 38
59 44 16 119
Figure 1. Kaplan-Meier curve. The x-axis represents the time (years) after the hearing-aid fitting, and the y-axis indicates the probability of having a hearing aid at that time. (A) Overall rate of successful use of a hearing aid. (B) Rate of successful use of a hearing aid according to social and/or work activities. The solid line represents the presence of social and/or work activities, and the dotted line indicates their absence.
There was a significant difference in outcomes for hearing-aid use among the methods of signal processing (p ⬍ 0.0001). Digital processing was used more frequently in the successful user group, and analogue processing was found more frequently in the failure group (67 successful users vs. 26 failure users for digital signal processing; and two successful users vs. 12 failure users for analogue signal processing). There was also a significant difference in outcomes for hearingaid use according to the number of channels (p ⬍ 0.0001). Four- or eight-channel devices were used more frequently in the successful user group, and one- or two-channel device was prescribed more frequently in the failure group (five successful users vs. 17 failure
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
Unilateral hearing aids in Koreans.
Figure 2. Boxplots of PTA (average of four thresholds at 0.5-, 1-, 2- and 4-kHz frequencies on a pure-tone audiogram) according to the type of hearing aid. The length of the box means the interquartile range (25th∼75th percentile), a line drawn across the box means the median, and two whiskers represents the maximum and minimum, respectively.
users for one-channel; 23 successful users vs. 13 failure users for two-channel; 34 successful users vs. five failure users for four-channel; and seven successful users vs. two failure users for eight-channel). There was a significant difference in the outcomes for hearing-aid use between groups with members that participated in social and/or work activities, including a job, and those who did not (p ⫽ 0.002). Successful users were more likely to participate in social and/or work activities (22 successful users vs. 25 failure users for persons without social and/or work activities; but 47 successful users vs. 13 failure users for persons with social and/or work activities). This study also showed that the CIC was used in 78.5% of participants with social and/or work activities, and the CIC and ITC were used in 64.8% and 31.5% of participants without social and/or work activities, respectively. There was no significant difference in the average PTA among outcomes for hearing-aid use (p ⫽ 0.403), but a significant difference in SDS was found among outcomes for hearing aid use (p ⫽ 0.033). The SDS score of intermittent users was significantly lower than that of the failure group (58.6 ⫾ 22.0% in successful users, 43.1 ⫾ 29.8% in intermittent users, and 64.8 ⫾ 19.8% in the failure group; p ⫽ 0.043). No significant differences in gender, age, duration of HI, degree of HI, type of HI, configuration of HI, or tinnitus were found among outcomes for hearing aid use (p ⫽ 0.833, p ⫽ 0.154, p ⫽ 0.451, p ⫽ 0.610, p ⫽ 0.158, p ⫽ 0.646, and p ⫽ 0.106, respectively). No significant differences in gender, degree of HI, tinnitus, type of HI or configuration of HI were found between groups with and without social and/or work activities (p ⫽ 0.096, p ⫽ 0.586, p ⫽ 0.356, p ⫽ 0.771, and p ⫽ 0.682, respectively). In this study, 63.6% of males and 46.9% of females responded that they regularly participated in social and/or work activities.
Discussion This study showed that 68.1% of unilaterally hearing-impaired Korean adults consistently wore a hearing aid and were satisfied
5
with its function. This suggested that a considerable proportion (31.9%) of patients with unilateral hearing impairment failed to feel a hearing aid was beneficial. Considering the advantages of binaural hearing or better hearing by unilateral fitting, these findings were unexpected. This study revealed some variables that were significantly different among successful users, intermittent users and failure groups: hearing-aid type, method of signal processing, number of channels, and social and/or work activities. There were no significant differences in gender, age, duration of HI, degree of HI, type of HI, configuration of HI, and tinnitus. This study also found that social and/or work activities and method of signal processing were significant predictors of a successful user. In their study of hearing-aid preferences in bilaterally symmetrical HI, Cox et al (2011) found that the preference for binaural hearing (using two hearing aids) rather than monaural hearing (using one hearing aid) was predictable from a combination of greater perceived daily problems and binaural advantage, as well as reduced binaural imbalance. They also found that audiometric hearing loss and auditory lifestyle were not predictive of aid preferences (Cox et al, 2011). Recent research has provided clinicians with valuable information regarding the predictors of intervention uptake and intervention outcomes. In a study of 153 uni- or bilaterally hearing-impaired participants 50 years of age and older with two intervention options (hearing aid(s) or communication programs), Laplante-Levesque et al (2012) found that applications for subsidized hearing services, higher socioeconomic status, lesser communication self-efficacy, and greater contemplation stage of change increased the uptake of hearing aids. They also determined that higher socioeconomic status, greater initial self-reported hearing disability, lower pre-contemplation stage of change, greater action stage of change, lower chance locus of control, greater hearing disability perceived by others, and self-increased intervention outcomes were predictors of successful intervention outcomes. These findings suggest that socio-behavioral factors, not audiological factors, influenced the uptake and outcome of hearing aids and emphasized the active role of clinicians who manage and counsel hearing-impaired persons. Cox et al (2011) have also reported the reasons for participants’ preferences for either one or two hearing aids. Because of improved speech comprehension in quiet situations as well as in noisy ones, the naturalness/comfort of their own voice, and the inconvenience in telephone use, patients preferred one hearing aid over two. However, patients who preferred two hearing aids reported feeling balanced, experiencing clarity of sound and feeling comfort/capable/ safe (Cox et al, 2011). Together with these reports, our results show that clinicians must evaluate the patients’ primary need, their stage of change, their lifestyle and their audiological environment, and counsel them thoroughly and in depth. Successful decisions regarding hearing aids and their uptake requires clinicians to understand their patients comprehensively and also to consider their economic status. For example, for unilaterally hearing-impaired persons, resolving the discomfort of daily life may not be the main target of a hearing aid. In this case, clinicians should weigh the advantages of binaural hearing and the disadvantages, as well as the expense associated with a hearing aid. Readers should not overlook the reality that binaural hearing can produce better real-world outcomes than monaural hearing. People who report greater hearing problems in daily life, who experience greater binaural loudness summation, and whose ears are more equivalent in dichotic listening are more likely to prefer to use two hearing aids (Cox et al, 2011). Clinicians must do their best to
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
6
D. Lee & H. Noh
make sure that patients have accurate and clear information in the decision-making process. Although a patient s own decision must be respected, there are some caveats associated with the choice of monaural hearing rather than binaural hearing. Persons who prefer long-term monaural hearing are potentially vulnerable to the deprivation effect in their hearing-impaired ear. In this condition, the ability to recognize words declines in the hearing-impaired ear over time, even though audiogram thresholds remain symmetrical in both ears (Silman et al, 1984; Neuman, 1996; Hurley, 1999). This study indicated that ITC was used more frequently in the failure group. Considering that the hearing threshold of ITC and CIC users was similar in this study, this result suggests that patients with moderate or severe HI may be more suitable for CIC. It may be because people with more social and/or work activities preferred the appearance of the smaller aid, or some other unknown explanation. This study also showed that analogue signal processing and fewer channels were prescribed more frequently in the failure group. Generally, analogue signal processing and fewer channels are not suitable for not-flat HI. Even in the 80 cases of flat HI in this study, analogue signal processing was used in 32.0% (n ⫽ 8), and 1 or 2 channels were used in 80.0% (n ⫽ 20) of the failure group (n ⫽ 25). These results were in contrast to the findings of the successful/intermittent user groups (n ⫽ 55), in which analogue signal processing was used in 3.6% (n ⫽ 2) of the cases and 1 or 2 channels were used in 43.6% (n ⫽ 24) of the cases. These findings suggest that even for flat HI, analogue signal processing and fewer channels should not be prescribed based solely on the patient’s economic grounds. In the past, a wrong recommendation of hearing aids had a negative impact on many patients and their families in cases where the choice of a hearing aid was decided by matching the price with the patients’ financial condition to determine what they could afford. A patient’s economic status often creates a dilemma regarding the cost versus performance of hearing aids. Therefore, an application for subsidized hearing services has been identified as an uptake predictor of hearing aids (Laplante-Levesque et al, 2012). In this study, Kaplan-Meier survival analysis revealed that the first decrease in hearing-aid use occurred within one year after the initial fitting, which was similar to previous reports (Schumacher & Carruth, 1997). However, a second drop was found to occur approximately four to five years after the fitting, which is coincident with the time when patients are due to update their hearing aid(s). We surmise that many persons with HI contemplate whether to purchase a new hearing aid when it is time to replace their first hearing aid with a new one. The major limitation of this study was the possibility of confounding factors because it was a retrospective study. This study was not randomized or blind. This retrospective study might not exclude recall bias. A second limitation of this study included the lack of more hearing-specific assessments on the performance of a hearing aid, such as international outcome inventory for hearing aids (IOI-HA), abbreviated profile of hearing aid benefit (APHAB), or the device oriented subjective outcome (DOSO) scale. This study evaluated participant’s satisfaction about his or her hearing aid using just a three-item survey with four-point scale. A final limitation was the long span of the enrollment period in this study because hearing-aid technology is always improving rapidly. In this study, the bias might result from diverse types of the hearing aid because the technology of the hearing aid has been moving too fast during a 15-year period.
This study showed that the SDS score of intermittent users was significantly lower than that of the failure group (p ⫽ 0.033). However, considering that the distribution of SDS was too dispersed in intermittent and successful users and our recommendation for hearing aids was based on a conventional formula in all patients, these results should be interpreted cautiously.
Conclusion On the basis of eight hours or longer of hearing-aid use and selfreported satisfaction, general uptake rate for a hearing aid was 68.1% in unilaterally hearing-impaired Korean adults. Successful users comprised 58.0% of the study population, and intermittent users made up 10.1%. Pre-fitting variables found to be predictors of hearing-aid uptake were social and/or work activities and method of signal processing. This study revealed that persons with social and/or work activities had better odds of being a successful hearing-aid user. Variables that were significantly different according to hearing-aid outcomes were the hearing-aid type, the method of signal processing, the number of channels, and social and/or work activities.
Acknowledgements The authors thank Mu Gil Heo (Audiologist, Jeil Hearing Aid Center, Seoul, Korea), who helped with data collection. The authors contributed equally to this project and should be considered co-first authors Declaration of interest: The authors have no conflicts of interest to report. The authors alone are responsible for the content and writing of this paper.
References Arlinger S. 2003. Negative consequences of uncorrected hearing loss: A review. Int J Audiol, 42, Suppl 2, 2S17–20. Arlinger S.B., Lagerbring C., Leijon A., Rosenhall U. & Scherstén T. 2003. Hearing aids for adults: Benefits and costs. The SBU Board of Directors and the SBU Scientific Advisory Committee. Bertoli S., Staehelin K., Zemp E., Schindler C., Bodmer D. et al. 2009. Survey on hearing aid use and satisfaction in Switzerland and their determinants. Int J Audiol, 48, 183–95. Brannstrom K.J. & Wennerstrom I. 2010. Hearing aid fitting outcome: Clinical application and psychometric properties of a Swedish translation of the international outcome inventory for hearing aids (IOI-HA). J Am Acad Audiol, 21, 512–21. Briggs L., Davidson L. & Lieu J.E. 2011. Outcomes of conventional amplification for pediatric unilateral hearing loss. Ann Otol Rhinol Laryngol, 120, 448–54. Chia E.M., Wang J.J., Rochtchina E., Cumming R.R., Newall P. et al. 2007. Hearing impairment and health-related quality of life: The Blue Mountains Hearing Study. Ear Hear, 28, 187–95. Cox R.M., Schwartz K.S., Noe C.M. & Alexander G.C. 2011. Preference for one or two hearing AIDS among adult patients. Ear Hear, 32, 181–97. Hurley R.M. 1999. Onset of auditory deprivation. J Am Acad Audiol, 10, 529–34. Laplante-Levesque A., Hickson L. & Worrall L. 2012. What makes adults with hearing impairment take up hearing aids or communication programs and achieve successful outcomes? Ear Hear, 33, 79–93. Mathers C., Smith A. & Concha M. 2003. Global burden of hearing loss in the year 2000. GBD 2000 Working Paper. Geneva: World Health Organization.
Unilateral hearing aids in Koreans.
Int J Audiol Downloaded from informahealthcare.com by Kainan University on 04/25/15 For personal use only.
Neuman A.C. 1996. Late-onset auditory deprivation: A review of past research and an assessment of future research needs. Ear Hear, 17, 3S–13S. Roman S., Nicollas R. & Triglia J.M. 2011. Practice guidelines for boneanchored hearing aids in children. Eur Ann Otorhinolaryngol Head Neck Dis, 128, 253–8. Schumacher D.U. & Carruth J.A. 1997. Long-term use of hearing aids in patients with presbyacusis. Clin Otolaryngol Allied Sci, 22, 430–3.
7
Silman S., Gelfand S.A. & Silverman C.A. 1984. Late-onset auditory deprivation: Effects of monaural versus binaural hearing aids. J Acoust Soc Am, 76, 1357–62. Snik A., Leijendeckers J., Hol M., Mylanus E. & Cremers C. 2008. The boneanchored hearing aid for children: Recent developments. Int J Audiol, 47, 554–9. Stewart D.O. & DeMarco J.P. 2005. An economic theory of patient decisionmaking. J Bioeth Inq, 2, 153–164.
