Normative Data for the Attitudes towards Loss of Hearing Questionnaire

Transcription

1 J Am Acad Audiol 16: (2005) Normative Data for the Attitudes towards Loss of Hearing Questionnaire Gabrielle H. Saunders* Kathleen M. Cienkowski Anna Forsline* Stephen Fausti* Abstract Investigations have shown that patient attitudes toward hearing loss and hearing aids impact self-reported handicap and disability, hearing aid benefit, and hearing aid use. The Attitudes towards Loss of Hearing Questionnaire (ALHQ) was developed by Saunders and Cienkowski (1996) to examine some of the psychosocial factors underlying the use and acquisition of hearing aids. Here we report data from a new version of questionnaire (ALHQ v2.1), which examines attitudes towards hearing loss and hearing aids on five scales: Denial of Hearing Loss, Negative Associations, Negative Coping Strategies, Manual Dexterity and Vision, and Hearing-Related Esteem. Reliability values, internal consistency values, and cut points for typical and atypical scores are provided, along with comparison of the scores of women, men, current hearing aid users, non hearing aid users, and paying versus nonpaying individuals. The ALHQ takes about ten minutes to complete and identifies for the clinician some of the issues that might jeopardize successful hearing aid outcome. Key Words: Attitudes, counseling, hearing aids, questionnaire Abbreviations: ALHQ = Attitudes towards Loss of Hearing Questionnaire; PTA = pure-tone average (mean of thresholds at 0.5, 1.0, 2.0 khz) Sumario Las investigaciones han mostrado que las actitudes de los pacientes hacia los trastornos auditivos y los auxiliares auditivos tienen un impacto sobre el autoreporte del impedimento y la discapacidad, sobre el beneficio y el uso del auxiliar auditivo. Saunders y Cienkowski desarrollaron el Cuestionario de Actitudes hacia la Pérdida de la Audición (ALHQ) (1996) para examinar algunos de los factores psico-sociales relacionados con el uso y adquisición de auxiliares auditivos. Este examina aspectos asociados con el impacto psico-social de la hipoacusia, con el estigma del auxiliar auditivo, la destreza manual, y la aceptación de la pérdida auditiva. Aquí se reportan datos sobre la confiabilidad del cuestionario y se aportan datos normativos para hombre y mujeres, usuarios actuales de auxiliares auditivos, pacientes que no utilizan audífono, e individuos que pagan y los que no pagan los servicios auditivos. El instrumento toma alrededor de diez minutos para completarlo e identifica para el clínico algunos de los problemas que pueden amenazar un resultado exitoso en el uso del auxiliar auditivo. Palabras Clave: Actitud, consejería, auxiliares auditivos, cuestionario Abreviaturas: ALHQ = Cuestionario de Actitudes hacia la Pérdida de la Audición; PTA = promedio tonal puro (media de las frecuencias de 0.5, 1.0 y 2.0 khz) *National Center for Rehabilitative Auditory Research, Portland VA Medical Center, Portland, Oregon; Department of Communication Sciences, University of Connecticut, Storrs, Connecticut Gabrielle H. Saunders, Ph.D., National Center for Rehabilitative Auditory Research, Portland VA Medical Center, 3710 SW US Veterans Hospital Road, Portland, OR 97207; Phone: , ext ; Fax: ; Gabrielle.saunders@med.va,.gov This work was supported by the Department of Veterans Affairs, Veterans Health Administration, Rehabilitation Research and Development Service Grant #C2645R, and the National Center for Rehabilitative Auditory Research. 637

2 Journal of the American Academy of Audiology/Volume 16, Number 9, 2005 In recent years, clinicians and researchers alike have become aware that psychosocial attitudes towards hearing loss and hearing aids affect hearing aid outcome. Studies have shown that hearing aid use is lower among individuals reporting general negativity towards amplification (Hickson et al, 1986; Wilson and Stephens, 2002), who perceive their hearing impairment has little impact upon them (e.g., van den Brink et al, 1996; Brooks and Hallam, 1998) and who consider hearing aids to be stigmatizing (van den Brink et al, 1996). Unfortunately, it is not always easy to discern how patients perceive their need and desire for, or their expectations and opinion of, amplification. While it would seem logical that there would be a positive relationship between degree of hearing loss and perceived hearing difficulties and/or perceived need for amplification, this is often not the case. For instance, Cox et al (2003) recently reported that audiometric data accounted for less than half of the variance in self-report data obtained by the APHAB, while Hawes and Niswander (1985) found low correlations between audiometric variables and the Hearing Performance Inventory. Similarly, it would seem logical to expect that individuals with more severe hearing impairment would perceive greater benefit from hearing aids than individuals with less impairment. While there is a trend in this direction, the majority of studies do not find a significant relationship between the two (e.g., Bentler et al, 1993; Gatehouse, 1994; Jerram and Purdy, 2001; Saunders and Jutai, 2004; and see Wong et al, 2003, for a review). Nonauditory factors such as pre-use expectations (Kricos et al, 1991; Cox and Alexander, 2000), motivation and desire to acquire amplification (Hickson et al, 1986; Brooks and Hallam, 1998), stigma associated with hearing aids (Stock et al, 1997), and personality (Gatehouse, 1994) have all been shown to relate to reported hearing aid satisfaction, use, and/or benefit. For this reason, questionnaires that attempt to evaluate some psychosocial attitudes quickly and efficiently have been developed; these include the Hearing Attitudes in Rehabilitation Questionnaire (HARQ) by Hallam and Brooks (1996), the Hearing Aid Selection Profile (HASP) by Jacobson et al (2001), and the Attitudes towards Loss of Hearing Questionnaire (ALHQ) by Saunders and Cienkowski (1996). This paper reports on data collected using the ALHQ. The ALHQ was developed with two purposes in mind: first, as a tool to elucidate some of the underlying psychosocial issues that lead to the refusal to acquire or to use amplification, and, second, as a counseling tool prior to fitting a hearing aid. At this time, however, only the first application is available. Ultimately, counseling modules associated with each ALHQ subscale will be developed. The clinician will use these modules to counsel the patient when his or her ALHQ score on a particular subscale or subscales indicates an attitude that is potentially detrimental to hearing aid outcome. Counseling modules will be particularly helpful because audiologists receive little training in counseling patients on emotional issues related to hearing impairment (e.g., Culpepper et al, 1994; Crandell, 1997; English et al, 2000) and because studies have shown that counseling individuals regarding attitudinal issues at the time of hearing aid fitting can increase hearing aid use and/or decrease perceived handicap (Surr et al, 1978; Brooks, 1979, 1989; Abrams et al, 1992; Kapteyn et al, 1997). The ALHQ was first published in 1996 by Saunders and Cienkowski. That version (v1.0) consisted of 24 items in five subscales that evaluated the social and emotional impact of hearing loss, lack of acceptance of and adjustment to hearing loss, perceived absence of support from significant others, hearing aid stigma, and awareness of hearing loss. Importantly, ALHQ scores were not highly correlated with age or degree of hearing loss; that is, the attitudes measured were not simply another way to measure severity of impairment. In that paper, it was reported that the reliability of the subscales was adequate for the development of the assessment tool but that two of the five subscales (hearing aid stigma and awareness of hearing loss) had reliability values lower than would be acceptable for general clinical use. Since that time, further changes have been made to the ALHQ. First, questions were added to capture information about additional constructs, and interim principal components analyses were conducted. This resulted in the R-ALHQ, which was a 25- item questionnaire (Cienkowski and Saunders, 2000). Second, the questionnaire response scale was improved from an 638

3 Attitudes towards Loss of Hearing Questionnaire/Saunders et al agree/disagree or yes/no format into a fivepoint scale resulting in the ALHQ v2.1. This revision widens the response scale and increases the variance in scores, thus allowing for more sensitive measurements (Nunnally, 1967; Cox, 1980). In this study, the ALHQ v2.1 was used to collect data from a large and diverse sample of subjects in order to establish norms for different population groups. The population subgroups investigated were non hearing aid users and current hearing aid users, men and women, and individuals who pay for auditory rehabilitation services and individuals who do not. These particular population subgroups were investigated because there is reason to believe that attitudes among them will differ. For instance, Saunders and Jutai (2004) found that experienced hearing aid users reported hearing aids to be less detrimental to their perceived personal image than new users. Garstecki and Erler (1999) found that women considered effective communication more important than did men, and that women showed greater awareness and less denial of their hearing difficulties than men. Finally, Kochkin (2003) showed that cost was highly related to consumer satisfaction with hearing aids. In this publication, then, we present confirmatory psychometric analyses of ALHQ v2.1, which includes extraction and reliability of the subscales along with normative values for the different population subgroups. These confirmatory analyses were felt necessary in light of the changes made to the response scale and the widened sample population. Subjects METHODS Data were collected from 325 subjects with sensorineural hearing loss, as defined by the presence of one or more thresholds greater than 25 db HL at 0.25 khz though 4 khz. Subjects were aged between 45 and 86 years (mean: 67.8 yr., SD: 8.8 yr.). Seventy-one were female, 254 were male; 168 wore binaural hearing aids, 157 were non hearing aid users. One hundred seventy-seven subjects were tested at the VA National Center for Rehabilitative Auditory Research (NCRAR); 104 were tested by the research team at Decibel Instruments Inc., a hearing aid company; and 44 were tested at the Audiology Clinic of the University of Connecticut. This sample size fulfills the requirements of principal components analysis used to extract the ALHQ subscales, which requires that there be at least five to ten respondents per questionnaire item (Tinsley and Tinsley, 1987). Of the 168 hearing aid users, 95 were known to have purchased their hearing aids, and 54 were known to have received their hearing aids free of charge through the Veterans Administration or from the Lions Club. A subset of 64 subjects, 37 from the NCRAR site and 27 from the University of Connecticut site, completed the questionnaire twice in order to confirm test-retest reliability. Thirty-three of these individuals were non hearing aid users; 31 were hearing aid users. Test Measures 1. Pure-tone audiometry, otoscopy, and tympanomety. Air-conduction thresholds were measured at octave frequencies between 250 Hz and 8 khz, along with interoctave frequencies of 1.5, 3.0, and 6.0 khz. Otoscopy and tympanometry were conducted to check for cerumen and conductive pathology, respectively. Any subjects with conductive pathology were excluded from the study. For later analyses, a three-frequency pure-tone average was computed (mean of thresholds at 0.5, 1.0, and 2.0). 2. The Attitudes towards Loss of Hearing Questionnaire (ALHQ v2.1). The ALHQ v2.1 was completed by all subjects. Each questionnaire item consists of a single statement, such as, I try to avoid small talk because of my hearing difficulties. Subjects state the extent to which they agree or disagree with the statement on a five-point scale ranging from a = Strongly disagree to e = Strongly agree. Two forms of the ALHQ are available; one for nonusers of hearing aids and one for current users of hearing aids. The forms differ in the wording of six questions. For example, the nonusers form, I am pretty sure that I don t need hearing aids, versus the current users form, I really don t think that I need my hearing aids. 639

4 Journal of the American Academy of Audiology/Volume 16, Number 9, 2005 Procedures In a single test session, subjects underwent audiometric evaluation and then completed the ALHQ. Subjects completed the form appropriate to their history of hearing aid use, using paper and pencil format. One subject with visual impairment required the experimenter to read questions and response options to him. A subset of 64 subjects returned to the laboratory between 2 and 14 days after the first test session to complete the ALHQ a second time to enable confirmation of testretest reliability. RESULTS Initial analyses were carried out for all data combined in order to have representation from individuals with a range of degrees of hearing loss, a range of backgrounds, and from both genders. Audiometric Thresholds Figure 1 shows a plot of pure-tone thresholds at each frequency with error bars showing +/- one standard error for the non hearing aid users and current hearing aid users separately. It is seen that individuals with a range of degrees of hearing impairment took part. Repeated measures analysis of Figure 1. Mean pure-tone thresholds in db HL with +/-1 standard error bars separated by hearing aid user status. Non hearing aid users are depicted by open circles; current hearing aid users are depicted by an asterisk. variance (ANOVA) shows that the thresholds of the current users are significantly poorer than those of the nonusers (F = 117.4, p < 0.001), and post hoc tests show that thresholds differ significantly at each frequency. ALHQ Subscale Extraction Questionnaire responses were scored as follows: one point was given for an a response, 2 for a b response, 3 for a c response, 4 for a d response, and 5 points for an e response. Principal components analysis (PCA) with varimax rotation was then used to extract psychometrically valid scales from the ALHQ data. PCA uses the inter-item correlations to group the questionnaire items into factors/scales. The items in each factor have stronger relationships to each other than they do to the questionnaire items in other factors. A fivefactor solution emerged that converged in eight iterations and explained 56.7% of the total variance. Factor 1 explained 15.3% of the variance, Factor 2 explained 13.1%, Factor 3 explained 11.9%, Factor 4 explained 9.8%, and Factor 5 explained 6.6%. Any questionnaire item with a factor loading (coefficient) of less than 0.5 was omitted from further analysis and from the questionnaire. The final solution included 22 questions. One item (question 1) appeared in two of the extracted factors. The way in which these factors were interpreted is provided below. Reliability analyses using Cronbach s alpha (Cronbach, 1951) were then used to determine the internal consistency of each factor/scale. Reliability analysis evaluates the intercorrelations between items in each scale and thus examines the extent to which the items in the scale measure the same underlying construct. For these analyses, any items with a negative factor loading were reverse scored such that a response of a was given 5 points, b 4 points, c 3 points, d 2 points, and e 1 point. Any item that decreased the Cronbach s alpha value of a scale was excluded from that scale; any item that increased the alpha-value of more than one scale was included in both. The combination of PCA and reliability analysis resulted in five groupings of questionnaire items (scales) that were interpreted as measuring: 640

5 Attitudes towards Loss of Hearing Questionnaire/Saunders et al 1. Denial of Hearing Loss. This scale consists of six items addressing issues of acceptance and acknowledgement of hearing loss. Three relate to acknowledgement of hearing difficulties, and three concern acceptance of the need for hearing aids. 2. Negative Associations. The four items in this scale query whether the patient considers hearing aids to have negative associations with aging or considers them to be embarrassing. 3. Negative Coping Strategies. There are eight items in this subscale. Three of the questions directly address the impact of hearing loss on social interactions; four address emotional reactions to the hearing loss; and one addresses the patient s interpretation of how others perceive the hearing loss. 4. Manual Dexterity and Vision. The three items in this scale address manual dexterity and vision with a view to assessing the individual s ability to manipulate a hearing aid. 5. Hearing-Related Esteem. The two items in this scale concern the impact hearing loss has had upon the individual s selfconfidence. The final questionnaire (ALHQ 3.0) can be found in Appendix 1. Detailed instructions for scoring the ALHQ are described in Appendix 2. Table 1 provides statistical information about each item and scale. Column 1 shows the item number in the questionnaire. Column 2 shows the factor loading of each item. The factor loading is the correlation between the item in question and the scale it is in. The higher the correlation, the more closely the item is associated with the scale. Column 3 shows the alpha value of each scale for both groups combined, for nonusers and for hearing aid users separately. Scoring the ALHQ Scores for each scale are computed by adding up the points obtained on each question and dividing by the total number of questions in the scale. For items with a positive factor loading, a response of a is given 1 point, a response of b is given 2 points, and so on. Items with a negative factor loading are reverse scored. Thus, for these items, a response of a is given 5 points, a response of b is given 4 points, and so forth. For ease of interpretation, a high score on each scale is associated with a negative attitude; Table 1. Results of Principal Components Analysis and Reliability Analysis Item number Factor loading Alpha-value Subscale All subjects Non hearing Current hearing (n = 325) aid users (n = 95) aid users (n = 168) Denial of Hearing Loss Negative Associations Negative Coping Strategies Manual Dexterity and Vision Hearing-Related Esteem

6 Journal of the American Academy of Audiology/Volume 16, Number 9, 2005 Figure 2. Histograms showing the distribution of scores on each ALHQ scale. Number of cases are shown on the y-axis, ALHQ score on the x-axis. that is, denial rather than acceptance of hearing loss, negative associations with hearing aids, poor coping strategies, poor manual dexterity and/or visual acuity, and low hearing-related esteem. Thus, low ALHQ scores are preferable to high ALHQ scores in terms of probable hearing aid outcome. Figure 2 shows the distribution of scores on each of the five ALHQ scales. As can be seen, scores obtained on each scale ranged from the minimum possible (1 point) to the maximum possible (5 points). Scores on the Denial of Hearing Loss, Negative Coping Strategies, and Hearing-Related Esteem scales are normally distributed. Scores on the Negative Associations and Manual Dexterity and Vision scales are positively skewed. This indicates that the mean score on the scale is higher than the median score on the scale; that is, on average subjects tend 642

7 Attitudes towards Loss of Hearing Questionnaire/Saunders et al Table 2. Correlations between Scores on Each ALHQ Subscale Scale Denial of Negative Negative Manual Dexterity Hearing Loss Associations Coping Strategies and Vision Negative Associations Negative Coping Strategies Manual Dexterity and Vision Hearing-Related Esteem to have positive rather than negative attitudes on these scales. Normally distributed scores are generally desirable for statistical analyses, but for the current purposes, in which the distribution of scores will be used to define typical and atypical scores for later counseling, skewed distributions will not affect the recommendations that will emerge from the ALHQ. Relationships between ALHQ Subscales Pearson correlations were computed in order to determine the extent to which the ALHQ scales are independent of each other, that is, measure different constructs. The lower the r-value, the less related the scales. The results are shown in Table 2. Correlations are low between all scales, with the exception of the relationship between the Negative Coping Strategies scale and the Denial of Hearing Loss scale (r = , p < 0.001). Thus, although there is some overlap in the constructs measured by the ALHQ scales, the scales are generally independent. Test-Retest Reliability Data from the 64 subjects who completed the ALHQ a second time were analyzed to examine test-retest reliability. Pearson correlations were carried out that yielded the following r-values for each scale: Denial of Hearing Loss: r = 0.85, Negative Associations: r = 0.84, Negative Coping Strategies: r = 0.88, Manual Dexterity and Vision: r = 0.79, Hearing-Related Esteem: r = Statisticians consider test-retest values greater than 0.8 to be excellent, and values between 0.6 and 0.8 to be good. The Hearing-Related Esteem scale has the lowest test-retest reliability, presumably because it consists of only two items; thus, variability in a response will have a greater impact on the testretest variance than a scale with more items. Normative Data for the ALHQ As discussed in the introduction, the ALHQ is intended for use as a prefitting counseling tool to determine whether a particular hearing aid candidate or current hearing aid user might benefit from counseling regarding their attitudes towards hearing loss and hearing aids, and if so, on which aspect or aspects. The ALHQ is scored such that a high score on any scale is indicative of a less favorable attitude. It is postulated that such attitudes could lead to potential problems with hearing aid acceptance and that counseling about that/those issues will improve the likelihood of a successful hearing aid outcome. In order to differentiate those individuals who may benefit from counseling from those who may not, it is necessary to define typical and atypical scores for each scale. It was decided that individuals with scores in the upper 20th percentile of any scale would potentially benefit from counseling. Such scores are thus defined as atypical. All other scores are defined as typical. This definition is purely based upon the consideration that provision of counseling to 20% of patients, that is, one in five, is clinically practical. Line 1 of each row in Table 3 shows the cut points for typical and atypical scores, along with the means and standard deviations for each subscale for all data combined. It is postulated that any individual with an atypical score (that is, a score above the cut point on a particular scale) would benefit from counseling. There is ongoing work to validate this hypothesis. Relationship between ALHQ Scores and Demographic Variables The data in Table 3 are for all subject populations combined. Prior to using these norms, it is necessary to determine whether ALHQ scores are indeed independent of age, 643

8 Journal of the American Academy of Audiology/Volume 16, Number 9, 2005 Table 3. Normative Data Values for the ALHQ for All Data Combined ALHQ scale Mean Standard Cut point deviation (80th percentile) Denial of Hearing Loss All data* Negative Associations All data Negative Coping Strategies All data Manual Dexterity and Vision All data Hearing-Related Esteem All data *See Table 5 for norms and cut points by hearing aid user status and degree of hearing loss. gender, hearing acuity, and hearing aid experience. In order to investigate this, two sets of analyses were conducted. First, stepwise multiple linear regression was used to determine the extent to which age, gender, hearing acuity, and hearing aid user status explain variance in each ALHQ scale score. The results of these analyses are shown in Table 4. The first column shows the independent variable entering the regression equation. The second column shows the adjusted r-squared change. This number, multiplied by 100, is the percentage of the total variance in scores that is explained by that variable. The third column shows the significance of the variable in the final model, and the fourth column shows the β-value. This is the weighting of the variable in the final equation. A positive β-value shows a positive correlation between the dependent and independent variables; a negative value shows an inverse correlation. For the analyses, hearing aid users were coded as 1, nonusers as 0; men were coded as 1, women as 2. Only those variables in the final regression equation that are significant at p < 0.05 are shown in this table. The analyses show that high Denial of Hearing Loss scores are related to better hearing, being older, and being a non hearing Table 4. Results of Stepwise Linear Regression for ALHQ Subscales Variable Adjusted r 2 change p-value in final equation β-value Denial of Hearing Loss PTA Age User status % total variance explained 27.4% Negative Associations Age User status Gender % total variance explained 7.2% Negative Coping Strategies PTA Age Gender % total variance explained 16.4% Manual Dexterity and Vision No variables entered na na na % total variance explained 0.0% Hearing-Related Esteem PTA % total variance explained 1.1% 644

9 Attitudes towards Loss of Hearing Questionnaire/Saunders et al aid user. High Negative Associations scores are associated with being younger, a non hearing aid user, and male. The use of Negative Coping Strategies is related to poorer hearing, being younger, and being female. Poorer Hearing-Related Esteem is associated with poorer hearing. The Manual Dexterity and Vision scale is not related to any of the independent variables. While age, gender, and hearing aid experience together explained only up to 7.2% of the total variance, degree of hearing loss explained over 22% of the variance in Denial of Hearing Loss scores and 10% of the variance in Negative Coping Strategies scores, such that poorer hearing is associated with lower Denial of Hearing Loss scores and use of poorer coping strategies. The strong relationship between hearing loss and Denial of Hearing Loss exists because questions on the Denial of Hearing Loss scale directly address issues regarding the perceived need for hearing assistance. Individuals with better hearing have less need for hearing assistance and thus score lower on this scale. A second series of analyses were then carried out to further investigate the need for additional ALHQ scale norms. Univariate ANOVAs were conducted for each scale separately, using gender, hearing loss, and hearing aid user status as the fixed factors. The pure-tone average (PTA) was used to define three categories of hearing loss: mild (PTA <40 db HL), moderate (PTA db HL), and moderately severe (56 70 db HL) hearing loss. The results of each ANOVA are presented below. 1. Denial of Hearing Loss. ANOVA results showed significant main effects of hearing loss (F = 7.8, p < 0.001) and hearing aid user status (F = 5.5, p < 0.02) but a nonsignificant main effect of gender (F = 2.3, p = 0.14). The only significant interaction was that between hearing loss and hearing aid user status (F = 3.0, p = 0.05). Separate group norms were therefore generated for each hearing loss by hearing aid user status combination, but not for men and women. These means, standard deviations, and cut points for typical/atypical scores are shown in Table 5. As in Table 3, within each group, atypical scores were defined as scores in the upper 20th percentile. When these norms are applied, the Denial of Hearing Loss score above which counseling is recommended is higher for individuals with better hearing than it is for individuals with poorer hearing, and is higher for non hearing aid users than for hearing aid users. In other words, non hearing aid users with good hearing are least likely to receive counseling, while hearing aid users with poorer hearing are most likely to receive counseling for issues associated with Denial of Hearing Loss. Note that there were only four individuals with a PTA >55 that did not wear hearing aids; thus, the mean, standard deviation, and cut point is not provided for this subgroup of subjects. 2. Negative Associations. A univariate ANOVA to determine whether degree of hearing loss, gender, and hearing aid user status significantly impact Negative Associations scores was next conducted. Results showed nonsignificant main effects of hearing loss (F = 0.2, p = 0.84), gender (F = 2.6, p = 0.11), and hearing aid user status (F = 0.3, p = 0.57) and nonsignificant two- and three-way interactions. Thus, additional norms were not generated for the Negative Associations scale. 3. Negative Coping Strategies. The univariate ANOVA to examine the relationship between Negative Coping Strategies scores and degree of hearing loss, gender, and hearing aid user status Table 5. Normative Data Scores for the Denial of Hearing Loss Scale of the ALHQ by Degree of Hearing Loss and Hearing Aid User Status User status Non hearing aid users Hearing aid users Hearing loss No. cases Mean SD Cut point No. cases Mean SD Cut point PTA 40 db HL PTA >40 db HL and 55 db PTA >55 db HL Too few cases

10 Journal of the American Academy of Audiology/Volume 16, Number 9, 2005 showed, as expected from the multiple regression results, a significant main effect of hearing loss (F = 9.1, p < 0.001). In addition, the main effect of gender was also significant (F = 3.8, P = 0.05), but hearing aid user status was not (F = 0.4, p = 0.50). All two- and three-way interactions were also nonsignificant. The relationship between Negative Coping Strategies and hearing loss is such that individuals with worse hearing use poorer coping strategies than do individuals with better hearing, and men use poorer coping strategies than women. However, additional norms based upon degree of hearing loss and gender were not computed because we believe that counseling individuals who use poor coping strategies, whatever their degree of hearing loss or gender, will be beneficial to those individuals. 4. Manual Dexterity and Vision. The univariate ANOVA to examine the relationship between Manual Dexterity and Vision scores and degree of hearing loss, gender, and hearing aid user status revealed nonsignificant main effects of hearing loss (F = 0.4, p = 0.68), gender (F = 1.5, P = 0.22), and hearing aid user status (F = 0.1, p = 0.99), and nonsignificant interactions between all variables; thus, additional norms are not needed for this scale. 5. Hearing-Related Esteem. The univariate ANOVA to examine the relationship between Hearing-Related Esteem scores and degree of hearing loss, gender, and hearing aid user status showed a significant main effect of hearing loss (F = 5.9, P = 0.003) but nonsignificant main effects of gender (F = 1.3, P = 0.25) and hearing aid user status (F = 0.7, p = 0.39), and nonsignificant interactions between all three variables. Tukey-B post hoc analyses show individuals with a PTA 40 db HL to have higher hearingrelated esteem than individuals with PTAs >40 db and 55 db HL. No other across-group comparisons were significant. Cut points for the three hearing loss groups were computed. Due to the distribution of scores, all three were the same (4.0); thus, no additional norms are provided in Table 5. A final set of analyses were conducted to compare the ALHQ scores of private-pay and nonpaying individuals. The mean age of the individuals who received their hearing aids from the VA or other nonpaying sources was 68.5 years (SD = 9.9); the mean age of the individuals who purchased their hearing aids was 70.0 years (SD = 6.8). One-way ANOVA showed this difference to be nonsignificant (F = 1.2, p = 0.27). The mean PTA of individuals who received their hearing aids from the VA or other nonpaying sources was 43.7 db HL (SD = 15.5); the mean PTA of the individuals who purchased their hearing aids was 49.1 db HL (SD = 14.9). One-way ANOVA showed this difference to be significant (F = 4.5, p = 0.036). The ALHQ scores of individuals in these two groups were then compared with ANOVA. The ANOVA showed that the ALHQ scores of individuals who had received their hearing aids free of charge did not differ from individuals who had purchased their hearing aids privately (F = 1.4, p = 0.23). GENERAL DISCUSSION As described in the introduction, the purpose of this study was to confirm the psychometric properties of the ALHQ on a large sample of subjects and to provide norms for data that includes a diverse selection of individuals. Data were collected from 325 subjects, and principal components analysis was run. It resulted in a 22-item questionnaire with five scales: Denial of Hearing Loss, Negative Associations, Negative Coping Strategies, Manual Dexterity and Vision, and Hearing-Related Esteem. These v3.0 subscales differ from those published by Saunders and Cienkowski in 1996 (ALHQ v1.0) because many additional questionnaire items have been added since the initial development of the ALHQ. However, these subscales are very similar to those of R-ALHQ, published by Cienkowski and Saunders (2000) albeit with different subscale names. More specifically, four of the five questions in R-ALHQ Denial of Hearing Loss scale remain in the Denial of Hearing Loss scale here; all items in the R-ALHQ Stigma scale remain in the Negative Associations scale here; and all three items in the R-ALHQ Dexterity scale are in the Dexterity and Vision scale here. Finally, items in the R-ALHQ Social Impact of Hearing Loss scale are now divided between the Negative Coping Strategies scale and the Hearing-Related Esteem scale here. The 646

11 Attitudes towards Loss of Hearing Questionnaire/Saunders et al internal consistency values of the scales are good and have improved over those published by Cienkowski and Saunders (2000). Four of the five scales now have Cronbach s α-values greater than 0.80, and thus also have the necessary sensitivity for clinical application as supported by Nunnally s criterion (1967). The test-retest reliability of responses is excellent, and the low interscale correlations show that each scale measures a different construct. The rationale for including each of the five scales in the final ALHQ is described below. The authors acknowledge that these rationales are based upon previous work, rather than empirically proven findings. 1. Denial of Hearing Loss. A high score on this scale indicates that the individual does not consider his/her hearing loss to be a problem and that he/she does not feel the need for hearing aids. Studies have shown that individuals with lower reported handicap are more likely to abandon hearing aids than individuals who report more handicap, even when hearing loss, age, and gender are accounted for (Humes et al, 2003). Similarly, it has been shown that individuals who minimize their hearing loss use their hearing aids less than individuals who are acknowledging of their hearing loss (Brooks and Hallam, 1998). Thus, a high score on this scale is likely to lead to poor uptake, poor use, or abandonment of hearing aids. 2. Negative Associations. A high score on this scale indicates that the individual associates hearing aids with aging and embarrassment. This has been shown to be a common sentiment among the young and elderly alike (e.g., Kricos et al, 1991; Kochkin, 1993; Erler and Garstecki, 2002) and is a problem because studies have shown that negative attitudes towards hearing aids result in less hearing aid use and lower satisfaction (e.g., Brooks, 1989; Garstecki and Erler, 1998; Wilson and Stephens, 2002). 3. Negative Coping Strategies. The items on this scale evaluate the extent to which individuals use undesirable behavioral techniques, such as withdrawing or pretending to hear, to cope with their hearing loss. Although cause and effect are difficult to differentiate from one another, poor coping is associated with negative psychological consequences, such as depression and loneliness (see Arlinger, 2003, for review). Aural rehabilitation programs in which participants are taught improved communications skills have shown positive outcomes, such as decreased handicap following hearing aid fitting when compared to a control group (Abrams et al, 1992; Beynon et al, 1997) and increased hearing aid use and satisfaction (Eriksson-Mangold et al, 1990). 4. Manual Dexterity and Vision. A high score on the Manual Dexterity and Vision scale is reflective of someone who has poor fine motor skills and/or poor visual acuity resulting in limited ability to manipulate small objects. These questions were included to establish whether an individual is likely to have difficulty managing a hearing aid (inserting the aid, changing the battery, checking for cerumen, etc.). Manual dexterity has been shown to correlate with hearing aid outcome, use, and satisfaction (Kumar et al, 2000; Wilson and Stephens, 2002; Humes, et al, 2003) and was the only factor that differentiated older and younger individuals in their reasons for dissatisfaction with hearing aids (Meister and von Wedel, 2003). In another study it was concluded that ease of use of a hearing aid was a major factor in hearing aid preference among a group of elderly first-time users (Baumfield and Dillon, 2001). Although it is ultimately up to the clinician to ensure that individuals select an appropriate hearing aid style for their manual dexterity and/or visual limitations, individuals tend to select the smaller less visible models and are resistant to larger styles of aid that are easier to see and handle. However, we postulate that an individual who acknowledges having poor manual dexterity and/or vision might be more accepting of a clinician s advice than an individual who does not. 5. Hearing-Related Esteem. A high Hearing-Related Esteem score indicates the individual has lost confidence in his/her ability to hear. Research in this area is sparse; however, as discussed by 647

12 Journal of the American Academy of Audiology/Volume 16, Number 9, 2005 Kricos (2000), self-esteem is associated with self-confidence and the belief that one can succeed at a particular task. The prospect of learning the required skills for use and maintenance of hearing aids requires the individual to have the confidence to take the first step towards such. Furthermore, although again it is hard to distinguish cause from effect, Harless and McConnell (1982) showed that individuals who were successful hearing aid users had a higher selfconcept than non hearing aid users. Thus, there is evidence to believe that awareness of low hearing-related esteem, prior to hearing aid fitting and intervention with counseling, might improve hearing aid outcome. In addition to the evidence cited above suggesting that high scores on the ALHQ might result in hearing aid abandonment or low usage, preliminary data from this laboratory also supports this notion. Fifty-four of the experienced hearing aid users in this study also completed the Satisfaction with Amplification in Daily Living (SADL, Cox and Alexander, 1999) questionnaire and rated the benefit provided by their hearing aids on a scale of 0 to 100. Pearson correlations showed a negative correlation between overall hearing aid satisfaction as measured by the SADL and scores on the Negative Associations scale, Negative Coping Strategies scale, and Manual Dexterity and Vision scale of the ALHQ. Furthermore, Negative Associations scores were negatively correlated with reported benefit and hours of hearing aid use per day. That is, negative attitudes as measured by the ALHQ were correlated with lower reported hearing aid use, satisfaction, and benefit. The relationship between degree of hearing loss and Denial of Hearing Loss can be interpreted as providing evidence that the Denial of Hearing Loss scale is a valid measure of denial, in that our data show that individuals with little hearing loss have higher Denial of Hearing Loss scores than individuals with greater hearing loss. Further evidence for the validity of this scale comes from the finding that hearing aid users with a PTA 40 db HL have lower Denial of Hearing Loss scores than non hearing aid users with the same degree of hearing loss. That is, those individuals who do not perceive the need for hearing assistance have not sought assistance. Brooks and Hallam (1998) report a similar finding; they found that minimization of hearing loss and a lack of desire for amplification, as measured by the HARQ, were negatively associated with hearing aid use. Men and women had significantly different scores on the Negative Coping Strategies scale, with men using poorer strategies than women. Many of these poor strategies involve withdrawing or pretending to hear. Hetu et al (1993) reported similar data. This may be because, as noted by Garstecki and Erler (1998), women place more importance upon social communication than do men, and thus women would be less likely than men to feign hearing if it were detrimental to communication. There were no differences in the ALHQ scores of private-pay versus nonpaying hearing aid users; however, it must be acknowledged that all of the nonpaying sample were veterans and that only 2% of the nonpaying population were female, while 26% of the paying sample were female. Newman et al (1993) reported no difference in the aided Hearing Handicap Inventory for the Elderly scores (Ventry and Weinstein, 1982) of paying versus nonpaying patients. On the other hand, Kochkin (2000) in his MarkeTrak V survey found that the fifth most common reason provided by individuals who had abandoned their hearing aids was cost related, and in MarkeTrak VI that hearing aid cost was negatively correlated with satisfaction (Kochkin, 2003). Further work must be done to clarify these findings. FUTURE WORK AND CLINICAL APPLICATION Unlike the majority of questionnaires available to audiologists, the ALHQ is neither a measure of hearing disability or handicap, nor is it an outcome measure; it is a tool for elucidating some of the psychosocial attitudes towards hearing loss and hearing aids that might result in hearing aid abandonment or low usage. As discussed above, it appears to be effective for such, since negative attitudes on the ALHQ were associated with less reported hearing aid satisfaction and less daily hearing aid use. There are two other questionnaires available that have been developed with similar 648

13 Attitudes towards Loss of Hearing Questionnaire/Saunders et al purposes in mind: the HARQ and the HASP. The HARQ was developed in the United Kingdom and was designed to assess the attitudes of middle-aged to elderly persons towards their own hearing impairment and the prospect of being fitted with a hearing aid. The HASP, on the other hand, was designed as a tool for assisting the clinician with the selection of an appropriate hearing aid for an individual. It deals with some of the issues addressed by the ALHQ, such as manual dexterity and motivation. However, both of these questionnaires appear to have been developed as items in themselves. We, on the other hand, are in the process of developing counseling modules to accompany the AHLQ scales so that once clinicians have identified potentially detrimental attitudes with the ALHQ, they can tailor counseling easily to the needs of a particular individual. The data presented above showing correlations between SADL scores and the ALHQ scales, and the fact that the Denial of Hearing Loss scale correlates negatively with degree of hearing loss, provide preliminary evidence that the ALHQ is a valid measure of attitudes towards hearing loss and hearing aids. Studies to provide this evidence are underway. SUMMARY AND CONCLUSIONS The ALHQ is a 22-item questionnaire with five scales that evaluates the attitudes towards hearing loss and hearing aids of both non hearing aid users and current hearing aid users. It takes about ten minutes to complete and identifies for the clinician some of the issues that might jeopardize successful hearing aid outcome. Counseling modules associated with each scale are currently under development. Acknowledgment. We thank Rebecca Cox and Kristin Vasil for data collection. REFERENCES Abrams HB, Hnath-Chislom T, Guerreiro SM, Ritterman SI. (1992) The effects of intervention strategy on self-perception of hearing handicap. Ear Hear 13: Arlinger S. (2003) Negative consequences of uncorrected hearing loss -a review. Int J Audiol 42(Suppl. 2):2S Baumfield A, Dillon H. (2001) Factors affecting the use and perceived benefit of ITE and BTE hearing aids. Br J Audiol 35: Bentler RA, Neibuhr DP, Getta JP, Anderson CV. (1993) Longitudinal study of hearing aid effectiveness. II. Subjective measures. J Speech Hear Res 36: Beynon GJ, Thornton FL, Poole C. (1997) A randomized, controlled trial of the efficacy of a communication course for first time hearing aid users. Br J Audiol 31: Brooks DN. (1979) Counseling and its effect on hearing aid use. Scand Audiol 8: Brooks DN. (1989) The effect of attitude on benefit obtained from hearing aids. Br J Audiol 23:3 11. Brooks DN, Hallam RS. (1998) Attitudes to hearing difficulty and hearing aids and the outcome of audiological rehabilitation. Br J Audiol 32: Cienkowski KM, Saunders GH. (2000) Clientattitude questionnaire provides data that can help shape counseling strategies. Hear J 53: Cox EP. (1980) The optimal number of response alternatives for a scale: a review. J Mark Res 17: Cox RM, Alexander GC. (1999) Measuring Satisfaction with Amplification in Daily Life: the SADL Scale. Ear Hear 20: Cox RM, Alexander GC. (2000) Expectations about hearing aids and their relationship to fitting outcome. J Am Acad Audiol 11: Cox RM, Alexander GC, Gray GA. (2003) Audiometric correlates of the unaided APHAB. J Am Acad Audiol 14: Crandell CC. (1997) An update on counseling instruction in audiology training programs. J Acad Rehabil Audiol 30:1 10. Cronbach LJ. (1951) Coefficient alpha and the internal structure of tests. Psychometrika 16: Culpepper B, Mendel LL, McCarthy PA. (1994) Counseling experience and training offered by ESBaccredited programs. ASHA 36: English K, Rojeski T, Branham K. (2000) Acquiring counseling skills in mid-career: outcomes of a distance learning education course for practicing audiologists. J Am Acad Audiol 11: Eriksson-Mangold M, Ringdahl A, Bjorklund A-K, Wahlin B. (1990) The active-fitting (AF) programme of hearing aids: a psychological perspective. Br J Audiol 24: Erler SF, Garstecki DC. (2002) Hearing loss- and hearing aid-related stigma: perceptions of women with age-normal hearing. Am J Audiol 11: Garstecki DC, Erler SF. (1998) Hearing loss, control and demographic factors influencing hearing aid use among older adults. J Speech Lang Hear Res 41: Garstecki DC, Erler SF. (1999) Older adult performance on the Communication Profile for the Hearing Impaired: gender difference. J Speech Lang Hear Res 42: Gatehouse S. (1994) Components and determinants of hearing aid benefit. Ear Hear 15:

14 Journal of the American Academy of Audiology/Volume 16, Number 9, 2005 Hallam RS, Brooks DN. (1996) Development of the Hearing Attitudes in Rehabilitation Questionnaire (HARQ). Br J Audiol 30: Harless EL, McConnell F. (1982) Effects of hearing aid use on self concept in older persons. J Speech Hear Disord 47: Hawes NA, Niswander PS. (1985) Comparison of the revised Hearing Performance Inventory with audiometric measures. Ear Hear 6: Hetu, R., Jones, L. and Getty, L. (1993). The impact of acquired hearing impairment on intimate relationships: implications for rehabilitation. Audiology 32: Hickson L, Hamilton L, Orange SP. (1986) Factors associated with hearing aid use. Aust J Audiol 8: Humes LE, Wilson DL, Humes AC. (2003) Examination of differences between successful and unsuccessful elderly hearing aid candidates matched for age, hearing loss and gender. Int J Audiol 42: Jacobson GP, Newman CW, Fabry DA, Sandridge SA. (2001) Development of the three-clinic Hearing Aid Selection Profile (HASP). J Am Acad Audiol 12: Jerram JC, Purdy S. (2001) Technology, expectations, and adjustment to hearing loss: predictors of hearing aid outcome. J Am Acad Audiol 12: Kapteyn TS, Wijkel D, Hackenitz E. (1997) The effects of involvement of the general practitioner and guidance of the hearing impaired on hearing-aid use. Br J Audiol 31: Saunders GH, Cienkowski KM. (1996) Refinement and psychometric evaluation of the Attitudes Towards Loss of Hearing Questionnaire. Ear Hear 17: Saunders GH, Jutai JW. (2004) Hearing specific and generic measures of the psychosocial impact of hearing aids. J Am Acad Audiol 15(3): Stock A, Fichtl E, Heller O. (1997) Comparing determinants of hearing instrument satisfaction in Germany and in the United States. High Perform Hear Solut 2: Surr RK, Schuchman GI, Montgomery AA. (1978) Factors influencing the use of hearing aids. Arch Otolaryngol 104: Tinsley HEA, Tinsley DJ. (1987) Uses of factor analysis in counseling psychology research. J Couns Psychol 34. van den Brink RHS, Wit HP, Kempden GIJM, van Heuvelen MJG. (1996) Attitude and help-seeking for hearing impairment. Br J Audiol 30: Ventry IM, Weinstein BE. (1982) The Hearing Handicap Inventory for the Elderly: a new tool. Ear Hear 3: Wilson C, Stephens D. (2002) Reasons for referral and attitudes toward hearing aids: do they affect outcome? Clin Otolaryngol 28: Wong LLN, Hickson L, McPherson B. (2003) Hearing aid satisfaction: what does research from the past 20 years say? Trends Amplif 7: Kochkin S. (1993) MarkeTrak III: why 20 million Americans don t use hearing aids for their hearing loss. Hear J 46: Kochkin S. (2000) MarkeTrak V: Why my hearing aids are in the drawer : the consumers perspective. Hear J 53: Kochkin S. (2003) On the issue of value: hearing aid benefit, price, satisfaction and brand repurchase rates. Hear Rev 10: Kricos PB. (2000) The influence of nonaudiological variables on audiological rehabilitation outcomes. Ear Hear 21:7S 14S. Kricos P, Lessner SA, Sandridge SA. (1991) Expectations of older adults regarding the use of hearing aids. J Am Acad Audiol 2: Kumar M, Hickey S, Shaw S. (2000) Manual dexterity and successful hearing aid use. J Laryngol Otol 114: Meister H, von Wedel H. (2003) Demands on hearing aid features special signal processing for elderly users? Int J Audiol 42:S58 S62. Newman CW, Hug GA, Wharton JA, Jacobson GP. (1993) The influence of hearing aid cost on perceived benefit in older adults. Ear Hear 14: Nunnally JC. (1967) Psychometric Theory. New York: McGraw Hill. 650