Longitudinal Changes in Hearing Aid Satisfaction and Usage in the Elderly Over a Period of One or Two Years After Hearing Aid Delivery

Transcription

1 Longitudinal Changes in Hearing Aid Satisfaction and Usage in the Elderly Over a Period of One or Two Years After Hearing Aid Delivery Larry E. Humes, Dana L. Wilson, Nancy N. Barlow, Carolyn B. Garner, and Nathan Amos Objective: The objective of this study was to measure hearing aid satisfaction and usage for extended periods of time, up to 2 yr after hearing aid delivery, to determine whether longitudinal changes occur in the elderly for these outcome measures. Design: A longitudinal study of hearing aid satisfaction and usage was performed in a group comprised of 134 elderly hearing aid wearers with these outcome measures obtained at 1, 6, and 12 mo postfit. A total of 49 of the original 134 elderly returned 2 yr after hearing aid delivery to complete the satisfaction and usage measures again. Multiple self-report measures of hearing aid satisfaction and hearing aid usage were obtained at each follow-up session. Results: Most between-interval comparisons of the various measures of satisfaction and usage failed to reveal significant changes over time. However, some significant changes were observed in both satisfaction and usage. When such changes occurred, they were always in a direction that suggested a decrease in satisfaction or usage over time. Correlations across various postfit interval pairs were found to be positive, statistically significant, and moderate to strong in strength for all measures of satisfaction and usage. Conclusions: Although there were slight declines in hearing aid usage and satisfaction over time, measures of hearing aid satisfaction obtained at 1 mo postfit and of hearing aid usage obtained at 6 mo postfit are generally stable for up to 2 yr after hearing aid delivery. (Ear & Hearing 2002;23; ) Hearing aid outcome measures represent a topic of increasing interest to a variety of parties or stakeholders (Humes, 1999, 2001; Weinstein, 1997). Humes and colleagues (Humes,1999, 2001; Humes, Garner, Wilson, & Barlow, 2002a) have demonstrated that there are several independent dimensions of hearing aid outcome, including subjective Department of Speech and Hearing Sciences, Indiana University, Bloomington, Indiana. DOI: /01.AUD B benefit, aided speech-recognition performance, objective benefit, satisfaction and usage, among others. Over the past decade, there has been keen interest in changes in hearing aid outcome measures after delivery of the hearing aid, especially regarding subjective benefit, objective benefit, and aided speech-recognition performance. In this regard, Gatehouse (1992) and Cox and Alexander (1992) were among the first to document improvements in objective and subjective benefit over the first few months of hearing aid use with Gatehouse coining the term acclimatization for such effects. Turner, Humes, Bentler, and Cox (1996) performed a comprehensive review of the existing literature on acclimatization at the time and found that the size of the effects, when they existed, were small. Studies conducted since this review, moreover, including a large-scale study from our laboratory (Humes, Wilson, Barlow, & Garner, 2002b), have generally been consistent with the conclusions of this review (e.g., Saunders & Cienkowski, 1997; Surr, Cord, & Walden, 1998). The focus in the literature on acclimatization has clearly been on measures of aided performance and benefit; two key dimensions of hearing aid outcome. The time course for changes, if any, in other dimensions of outcome, however, has received little attention. In particular, little is known about how hearing aid usage or satisfaction change during the first couple of years after hearing aid delivery. Bentler, Niebuhr, Getta, and Anderson (1993) followed 65 participants having a mean age of 63.8 yr for a period of 1 yr postfit and reported that satisfaction remained stable throughout this period. The measure of satisfaction, however, was a single rating regarding overall satisfaction with the hearing aid(s). Nonetheless, the individual satisfaction ratings obtained 6 and 12 mo after hearing aid delivery were positively and moderately correlated (r 0.73), suggesting reasonable stability of the individual data as well. A variety of hearing aid styles and circuit types were evaluated by Bentler et al. (1993), moreover, and stability of satisfaction ratings over time was observed regardless of technology. Humes, 0196/0202/02/ /0 Ear & Hearing Copyright 2002 by Lippincott Williams & Wilkins Printed in the U.S.A. 428

2 EAR &HEARING, VOL. 23 NO Halling, and Coughlin (1996) found that hearing aid satisfaction did not change during the first year after hearing aid delivery. The study by Humes et al. (1996), however, made use of a brief 7-item survey to measure satisfaction and followed only 20 hearing aid wearers. Nonetheless, the individual satisfaction ratings were reasonably consistent over time with between-interval correlations that were positive, moderate to strong in magnitude, and statistically significant. The use of a single and brief (1- or 7-item) measure of satisfaction by Bentler et al. (1993) and Humes et al. (1996) may have lacked sufficient sensitivity to observe changes in satisfaction over time. In addition, the time course of changes in hearing aid usage has received even less attention. Humes et al. (1996) measured hearing aid usage through self report and with a data-logging device within the hearing aid for the first year of hearing aid use. No significant changes in hearing aid usage were observed with either method and significant betweeninterval correlations were observed among the 20 elderly participants in that study. Given the foregoing, the purpose of the present study was to examine measures of hearing aid usage and satisfaction in a large number of elderly adults and over periods of 1 to 2 yr postdelivery. In addition to obtaining data over a longer period of time than heretofore reported and for a larger number of participants, multiple measures of hearing aid satisfaction and usage were obtained at each measurement interval. Further, the same set of clinicians making use of identical test protocols performed the measures throughout the 2-yr study period and all participants were fit with identical hearing instruments. Thus, sources of variance associated with different test protocols, examiners or hearing instruments were minimized. METHOD Participants The participants in this study were recruited for a large-scale, longitudinal study on hearing aid outcome measures via newspaper ads, flyers posted in the community, printed announcements in church/ synagogue bulletins and word of mouth. All participants enrolled in the study met the following selection criteria: 1) age between 60 and 89 yr; 2) hearing loss that was flat or gently sloping (from 250 to 4000 Hz, no inter-octave change in hearing thresholds of more than 20 db); 3) hearing loss that was of sensorineural origin (normal tympanometry and airbone gaps no greater than 10 db at three or more frequencies); 4) hearing loss that was bilaterally symmetrical (interaural difference within 30 db at TABLE 1. Comparison of some descriptive characteristics for the 1-yr and 2-yr datasets. Dataset Variable 1-Year 2-Year Sample size (N) HFPTA-right (db HL) HFPTA-left (db HL) HHIE (total score) Age (yr) WAIS-R, IQ % Males % New HA Users all octave and half-octave intervals from 250 to 4000 Hz); 5) pure-tone thresholds within the following ranges at frequencies of 250, 500, 1000, 1500, 2000, 3000, 4000, and 6000 Hz, respectively: 5 to 85, 5 to 85, 10 to 90, 20 to 95, 25 to 95, 30 to 120, 30 to 120, and 30 to 120 db HL (ANSI, 1989); 6) no known medical or surgically treatable ear-related condition; 7) no known fluctuating or rapidly progressing hearing loss; 8) no cognitive, medical, or languagebased conditions that may have limited the participant s ability to complete the procedures used in the longitudinal study of outcome measures; 9) no use of medications that could affect hearing or cognition; and 10) completion of a signed medical clearance form, or waiver of such by the participant, and a signed informed consent form. Approximately 400 individuals had been identified as potential participants and 205 subsequently enrolled in the large-scale, longitudinal study. The present study reports on a group of 134 of these individuals who represent all the participants who completed a variety of measures at various times throughout the first year of enrollment in the longitudinal study. In particular, these 134 participants completed a series of hearing aid outcome measurements at 1 mo, 6 mo, and 1 yr after delivery of the hearing aids and their data constitute the 1-yr dataset. Of these 134 participants, 49 returned for a 2-yr follow-up and their data constitute the 2-yr dataset. Table 1 summarizes several key features for the participants comprising each dataset, including high-frequency (1000, 2000, and 4000 Hz) puretone average (HFPTA) for each ear, perceived hearing handicap without amplification (Hearing Handicap Inventory for the Elderly [HHIE], Ventry & Weinstein, 1982), age, general cognitive function measured with the Wechsler Adult Intelligence Scale-Revised (WAIS-IQ) (Wechsler, 1981), gender (% males) and prior hearing aid experience (% new users). In general, the participants comprising the 1-yr and 2-yr datasets are virtually indistinguishable from one another.

3 430 EAR &HEARING /OCTOBER 2002 Procedures After audiologic testing, participants returned for an extensive battery of cognitive, psychologic and auditory tests administered before hearing aid delivery and over a series of five 90- to 120-minute sessions. This prefit testing was conducted for other purposes and will not be considered further here. Once this prefit testing was completed, the participants returned for the initial fit and delivery of their hearing aids. Based on the previously obtained audiologic information and using the Hearing Aid Selection Program (HASP, Version 2.07) fitting software, produced and distributed by the National Acoustics Laboratories (NAL), NAL-R targets, including corrections to targets for severe or profound hearing loss (Byrne, Parkinson, & Newall, 1990), were generated for each ear and the corresponding circuit was selected and ordered. The HASP software returns a variety of targets, including real-ear insertion gain and full-on coupler gain. The latter was adjusted downward by the 10 db reserve gain incorporated into the HASP software to create recommended coupler-gain values. All hearing aids made use of linear circuits with output-limiting compression and Class D amplifiers. All instruments were full-shell, in-the-ear (ITE) devices and included a telecoil switch on one instrument (determined by the wearer s preference). Active tone (low-cut only) and output-limiting controls, adjustable select-a-vent venting, and wax guards were included on all devices. The volume-control wheels were marked by the manufacturer with a small white dot at the perimeter to provide a visual reference for its position and adjustment. Using real-ear insertion gain targets for the NAL-R prescription formula incorporated in the HASP fitting software, the clinician adjusted the settings of the controls and vent to achieve the closest match possible to target gain for a 60 db SPL swept pure-tone signal using either Frye 6500 or Audioscan real-ear measurement equipment. Matching criteria were 10 db from 250 to 2000 Hz and 15 db at 3000 and 4000 Hz. If a match could not be obtained using these fairly broad matching criteria, the participant was not permitted to continue in the study. In fact, no participants were eliminated from the study for this reason and the quality of the matches to target gain were considerably better than the broad exclusion criteria cited above, as will be seen below. A previous report from this project (Humes, Barlow, Garner, & Wilson, 2000) on 55 participants followed for a full year demonstrated that: 1) users consistently preferred a gain setting that was typically 6 to 9 db below that prescribed; 2) the experimenters were able to return the volume control for all aided measurements to a consistent position that matched more closely the prescribed target gain for subsequent aided testing; and 3) problems, such as dead batteries, cerumen blockage of receiver port, etc., were relatively uncommon. To examine the stability of hearing aid gain in the present study, coupler gain for each ear was examined for both the clinician-fit and as-worn gain measurements. Means and standard errors for the gain responses obtained at each postfit interval for each dataset were provided recently by Humes et al. (2002b). In all cases, the clinician-fit gain during the initial fitting is in excellent agreement with that measured at the final session either 1 or 2 yr later. As-worn gain was found to be lower than the clinician-fit gain in both datasets, especially in the high frequencies. In summary, the data from Humes et al. (2002b) indicated that the clinician-fit gain was very stable throughout the study and the as-worn gain, perhaps more representative of the gain during the everyday usage conditions to which the subjective measures of satisfaction and usage refer, was stable throughout as well. After the delivery of the hearing aids and their initial fitting by the clinician, a hearing aid orientation was conducted in which the following general topics were reviewed with each participant: 1) matters pertaining to the hearing aid purchase (user manual, warranty, 30-day trial period); 2) location and function of hearing aid components (microphone, volume control, battery door, telecoil switch, wax guard, etc.); 3) hearing aid battery (size, type, insertion and removal, etc.); 4) demonstration and practice in hearing aid insertion and removal; 5) demonstration and practice in care of hearing aids; and 6) counseling regarding benefits and limitations of amplification, communication strategies to optimize benefit, and instructions about the completion of a daily hearing aid usage diary. Participants were instructed to use their hearing aids at least 4 hr per day and to begin use in easier listening conditions (quiet, one-on-one conversation, etc.) when possible. Each participant returned 2 wk later for a follow-up session. At the beginning of this session, gain measurements were again made in the coupler to evaluate the instruments and the aids were removed, inspected, and subsequently adjusted as needed to restore their function to that recorded in the initial session. The hearing aid usage diary was collected, photocopied, and returned to the participant. The participant was also instructed to increase their minimum daily hearing aid usage to at least 6 hr. Approximately 2 wk later, the participant returned for the 1-mo follow-up visit. The hearing aids were again examined, evaluated in the testbox and adjusted as needed to return their function to the

4 EAR &HEARING, VOL. 23 NO Figure 1. Mean hearing aid satisfaction ratings for the HASS scales (top) and the GHABP (bottom) for the 1-yr (circles) and 2-yr (triangles) datasets as a function of postfit interval. Vertical bars represent plus or minus one standard error about the mean values.

5 432 EAR &HEARING /OCTOBER 2002 target levels from the initial fitting session. The hearing aid usage diary was collected, photocopied, and returned to the participant. The participant was also instructed to increase their minimum daily hearing aid usage to at least 8 hr. Next, each participant completed several surveys, including multiple measures of hearing aid satisfaction and usage. Measures of Hearing Aid Satisfaction and Usage The measures of hearing aid satisfaction used in this study were a slightly modified version of the MarkeTrak-IV satisfaction survey (Kochkin, 1997), referred to hereafter as the Hearing Aid Satisfaction Survey (HASS) (Humes et al., 2002a), and the satisfaction items extracted from the prescribed portion of the Glasgow Hearing Aid Benefit Profile (GHABP) (Gatehouse, 1999), consisting of four prototypical listening situations. The HASS yields two subscale satisfaction ratings, one for dispenser-related satisfaction items and one for device-related satisfaction items. Measures of hearing aid usage included the hours logged in each participant s hearing aid diary, the usage portion of the GHABP, and the usage portion of the Hearing Aid Disability and Benefit Inventory (HDABI) (Gatehouse, 1999), a precursor to the GHABP that included an assessment of the hearing aid in 14 common listening situations. For the GHABP, a 5-point rating scale of satisfaction is used with a rating of 1 being the lowest amount of satisfaction or usage and a rating of 5 being the highest amount. In this study, a shortened version of the GHABP was used in which only the four prespecified prototypical listening situations were employed. That is, four additional patientspecific listening situations were not constructed. For the HDABI, 14 listening situations are specified and range from listening to music to talking on the telephone to having a conversation with someone at a large noisy gathering. For each of these hearing activities, three questions are asked and one of these pertains to hearing aid usage (does the individual wear his or her hearing aids in the specified listening situation?). The five possible answers range from always (1) to never (5). Thus, lower scores on the HDABI indicate greater hearing aid use than higher scores. The other two questions for each listening situation in the HDABI concerned the helpfulness of the hearing aid and the frequency with which the listening situation was encountered by the hearing aid wearer. Responses to these two items are not considered here because the focus of this report is on measures of hearing aid satisfaction and usage. The satisfaction and usage measures described above were obtained at the 1-mo, 6-mo, and 1-yr follow-up sessions for the 1-yr dataset and again at the 2-yr follow-up session for the 2-yr dataset. The daily hearing aid diaries, however, were used only during the first year after hearing aid delivery. RESULTS AND DISCUSSION Figure 2. Mean hearing aid usage ratings for the HDABI scale (top) and the GHABP (bottom) for the 1-yr (circles) and 2-yr (triangles) datasets as a function of postfit interval. Vertical bars represent plus or minus one standard error about the mean values. Note that higher scores on the HDABI reflect lower hearing aid usage whereas the opposite is true for the GHABP. Figure 1 shows the means and standard errors for the three measures of hearing aid satisfaction included in this study and for each of the datasets. The top panel presents data for both subscales of the HASS whereas the bottom panel depicts the results for the satisfaction item of the abbreviated GHABP instrument. Before examining the results in each

6 EAR &HEARING, VOL. 23 NO TABLE 2. Summary of General Linear Model (GLM) analyses for the effect of time on each measure of satisfaction for the 1-yr and 2-yr datasets. Dataset Dependent Variable F (df) N Significant Paired t-tests** 1-yr HASS-global 12.2 (2,236)* yr 1mo HASS-dispenser 4.9 (2,246)* 124 None GHABP-satisfaction 15.2 (2,254)* yr 1mo;6mo 1mo 2-yr HASS-global 7.5 (3,117)* 40 1 yr 6mo;2yr 6mo HASS-dispenser 7.5 (3,120)* 41 2 yr 1mo;2yr 6mo GHABP-satisfaction 8.2 (3,132)* 45 2 yr 1mo *p ** p 0.01 with Bonferroni-adjustment for multiple comparisons. panel in detail, several general statements can be made about the data in Figure 1. First, regarding ratings from the HASS for the device-related items (white symbols), participants in both datasets were generally satisfied with their hearing aids as reflected by average scores of approximately 4 on the HASS. Second, regarding the dispenser-related items of the HASS, participants in both datasets were generally very satisfied with this aspect of the hearing aid delivery process as reflected in average scores between 4.5 and 5.0 across time. Third, average satisfaction ratings on the GHABP of about 3.0 to 3.5 for both datasets across time indicate that the participants were between reasonably satisfied (a rating of 3 ) and very satisfied (a rating of 4 ). Finally, in addition to yielding similar ratings of satisfaction, the HASS and GHABP measures reveal consistent patterns across datasets and time. A repeated-measures version of the General Linear Model (GLM) was used to analyze the data in each panel. Separate repeated-measures GLM analyses were conducted for each dataset and for each satisfaction measure. Significant main effects of postfit interval were followed by Bonferroni-adjusted paired-sample t-tests. The significance level for all GLM and post hoc analyses was p Table 2 summarizes the results of the GLM analyses of the satisfaction measures for the 1-yr and 2-yr datasets. Significant main effects of time were observed for all three satisfaction measures. For the 1-yr dataset, there are three possible post hoc comparisons for each satisfaction measure: 1) 1-mo versus 6-mo; 2) 1-mo versus 1-yr; and 3) 6-mo versus 1-yr. For the device-related items of the HASS, only the 1-mo versus 1-yr difference was found to be significant with the satisfaction rating at 1 yr being lower than that at 1 mo. There were no significant effects observed in the post hoc analyses for the dispenser-related items of the HASS. Finally, for the GHABP satisfaction item, both the 1-yr and 6-mo ratings were significantly lower than the 1-mo ratings. In summary, three of the nine post hoc comparisons of satisfaction ratings for the 1-yr dataset indicated that ratings decreased over time, especially from the 1-mo to the 1-yr measurement intervals. Two-thirds of the possible paired comparisons, however, revealed no significant changes in satisfaction over time. Regarding the 2-yr dataset, there are six possible post hoc paired comparisons, the same three as for the 1-yr dataset plus comparison of the 2-yr measurements to those of each of the other three time intervals (1-mo, 6-mo, and 1-yr). Of the six possible paired comparisons for the device-related portion of the HASS, two were found to be significant. Both the 1-yr and 2-yr satisfaction ratings were significantly lower than those obtained at 6 mo for the 2-yr dataset. Regarding the dispenser-related items of the HASS, the 2-yr ratings were found to be significantly lower than those at both the 1-mo and 6-mo postfit intervals. Finally, for the satisfaction portion of the abbreviated GHABP, only the 2-yr rating was found to be significantly lower than that obtained after 1 mo of hearing aid use. In summary, of the 18 possible paired comparisons (six postfit interval pairs for each of three satisfaction measures), five were found to be significantly different. Of these five, four involved comparisons of the satisfaction ratings obtained at the 2-yr postfit interval to those at either the 1-mo or 6-mo postfit interval and all five reflected significant decreases in hearing aid satisfaction over time. Thirteen of the 18 paired comparisons (72.2%) for the 2-yr dataset, however, revealed no significant changes in hearing aid satisfaction over time. Figure 2 provides plots of hearing aid usage as a function of time for both datasets and for two of the usage measures, the HDABI (top panel) and the GHABP (bottom panel). Visual inspection of these group data reveals an apparent increase in HDABI usage ratings and an apparent decrease in GHABP usage ratings over time. The reader should recall, however, that the GHABP scores are directly proportional to usage whereas those for the HDABI are inversely proportional to usage. For example, whereas ratings of 3 and 4 on the GHABP correspond to proportional usage ratings of about 1/2 of the time and about 3/4 of the time, respec-

7 434 EAR &HEARING /OCTOBER 2002 Figure 3. Histograms of score differences for three measures of hearing aid satisfaction (left panels) and hearing aid usage (right panels) for the 1-yr dataset. Except for the HDABI-use measure, score differences were computed by subtracting the 1-mo rating from the 1-yr rating such that negative score differences represent a decrease in usage or satisfaction from the 1-mo to the 1-yr postfit intervals. To maintain this association of negative score differences with decreases in usage or satisfaction over time for the HDABI, the 1-yr ratings were subtracted from the 1-mo ratings. Bin widths were 0.5 for the HASS, GHABP and HDABI measures and 1.0 for the Daily Use Diary. For example, the bin labeled 0 for the HASS-global histogram in the top left panel represents the number of score differences that were greater than 0.25 and less than 0.24(99999). tively, ratings of 1, 2, and 3 on the HDABI correspond to usage ratings of always, often, and sometimes, respectively. In general, averaging across both datasets and all postfit intervals, it appears that the participants in this study used their hearing aids about 3/4 of the time (GHABP) or often (HDABI). Because the average hours of daily hearing aid usage computed from the daily diaries kept by the participants were available only for the first year of hearing aid use, these data are not presented in Figure 2. Rather, means and standard errors for the average daily amount of hearing aid usage (hr/day) computed from the hearing aid diaries are presented in Table 3 for the participants comprising each dataset and for the three measurement intervals available (1-mo, 6-mo, and 1-yr). Diaries were completed throughout the first year by about 70 to 80% of the participants in each dataset with the number of participants providing completed diaries in each group indicated in Table 3. In general, visual inspection of the data in Table 3 is consistent with the patterns from the other measures of hearing aid usage presented previously in Figure 2. Specifically, both datasets show a slight trend toward decreased hearing aid usage during the first year. In a manner identical to that described previously for the analysis of hearing aid satisfaction ratings, repeated-measures GLM analyses were performed for TABLE 3. Means and standard errors (in parentheses), in hours/ day, for hearing aid usage estimates calculated for each dataset from the daily diaries maintained by the participants. Postfit Interval Group N 1mo 6mo 1yr 1-yr (0.30) 8.72 (0.40) 8.65 (0.41) 2-yr (0.53) 9.68 (0.65) 9.63 (0.63) N represents the number of participants with valid data for all three postfit intervals from which the means and standard errors were calculated.

8 EAR &HEARING, VOL. 23 NO TABLE 4. Summary of General Linear Model (GLM) analyses for the effect of time on each measure of usage for the 1-yr and 2-yr datasets. Dataset Dependent Variable F (df) N Significant Paired t-tests** 1-yr HDABI-use 6.4 (2246)* yr 1mo Daily Use Diary 10.8 (2208)* yr 1mo;6mo 1mo GHABP-use 13.0 (2254)* yr 1mo;6mo 1mo 2-yr HDABI-use 5.7 (3120)* 41 2 yr 6mo GHABP-use NS 45 None *p ** p 0.01 with Bonferroni-adjustment for multiple comparisons. each of the hearing aid usage measures to examine the effects of postfit measurement interval, with significant GLM effects followed by post hoc Bonferroniadjusted paired-sample t-tests. Table 4 presents a summary of the GLM analyses of hearing aid usage for each of the datasets in this study. For the 1-yr dataset, significant effects of postfit interval were found for all three measures of hearing aid usage (HDABI, GHABP, and use diary). Of the nine postfit intervals pairs, three for each dataset, five were found to be significant. For the use diary and the usage portion of the GHABP, the usage measures at 6 mo and 1 yr were both found to be significantly lower than that at 1-mo postfit. For the HDABI measure of hearing aid usage, only the usage at 1 yr was found to be significantly less than that at 1 mo. For the 2-yr dataset, usage data throughout the 2-yr period after hearing aid delivery were available only for the GHABP and HDABI usage measures. As indicated in Table 4, the GLM analysis of the GHABP measure of hearing aid usage revealed no significant effects of postfit interval on usage. The GLM analysis of the HDABI measure of usage, however, revealed a significant effect of postfit interval. Post hoc testing revealed that the HDABI usage rating reflected less usage (higher rating) at the 2-yr postfit interval when compared with that measured at the 6-mo postfit interval. In summary, of the 12 possible paired comparisons across postfit intervals for the 2-yr group, six for the HDABI measure of usage and six for the usage portion of the GHABP, only one was found to be significant. In summary, few significant changes in hearing aid usage were observed in the mean data as a function of postfit interval. Considering both datasets and both the GHABP and HDABI usage measures, a total of 21 potential across-interval differences are represented and, of these, only six were found to reflect a significant difference in usage over time. Most of the significant changes in hearing aid usage occurred during the first year, as reflected by the five significant differences between postfit intervals, out of the nine possible, for the 1-yr dataset. In all cases, however, when significant differences were observed, they were consistent with a decrease in usage over time. Thus far, the focus has been placed on group data from the participants in this study. Individual data are of importance too, however, as it is possible that significant portions of each dataset might change their satisfaction ratings over time, but in opposite directions, with some indicating increasing satisfaction or usage over time and others moving in the opposite direction. Two approaches were pursued in examining the individual data. First, score differences from the 1-mo to either the 1-yr or 2-yr measurement intervals were calculated for each participant and histograms of these scores differences were generated. Second, Pearson-r correlation coefficients were calculated for all possible postfit interval pairs to examine the consistency of ordering among participants over time. That is, if an individual had one of the highest satisfaction ratings 1 mo after hearing aid delivery, was that still the case 6 or 12 mo after delivery? Figure 3 presents the histograms of score differences for the participants comprising the 1-yr dataset for measures of hearing aid satisfaction (left panels) and usage (right panels). For all measures except the HDABI usage measure, the data for the 1-mo postfit interval were subtracted from the data gathered at the 1-yr postfit interval. This was reversed for the HDABI measure because, unlike all the other usage and satisfaction measures in this figure, higher ratings on this measure reflect lower usage. As a result, decreased satisfaction or usage over the first year after hearing aid delivery is reflected as negative score differences in each histogram whereas increases in satisfaction or usage over time are represented as positive difference scores. For all of the histograms, except that for the daily hearing aid diaries (lower right), the bin width used along the abscissa is in 0.5 scale units. For example, the bin for score differences of 0 included score differences from 0.25 to 0.249, the bin for score differences of 0.5 included differences from 0.25 to 0.749, and so on. For daily usage derived from the hearing aid diaries, a bin width of 1 hr/day was used

9 436 EAR &HEARING /OCTOBER 2002 such that score differences of 0 included all those differences from 0.50 to 0.49 hr/day. Clearly, no bimodal distributions of score differences are evident in Figure 3 for these six measures. A score difference of 0 is the modal value for all six distributions in this figure with most of the histograms skewed slightly toward negative values, consistent with the trends in the group data toward slight reduction in usage and satisfaction from the 1-mo to the 1-yr postfit intervals. Table 5 presents the Pearson-r correlation coefficients for the 1-yr dataset for each postfit-interval pair and each outcome measure. The total number of participants (N) with valid data from which the correlations were calculated is also provided in parentheses following each correlation in the table. All the correlations are positive, moderate (r 0.50) to strong (r 0.95) in strength, and statistically significant (p 0.01). TABLE 5. Between-interval Pearson-r correlation coefficients for the 1-yr group and for each measure of hearing aid satisfaction and usage. Dependent Variable 1movs. 6mo Postfit Interval Pairs 1movs. 1yr 6movs. 1yr HASS-global 0.71 (119) 0.65 (128) 0.82 (122) HASS-dispenser 0.52 (124) 0.50 (131) 0.60 (127) GHABP-satisfaction 0.70 (128) 0.59 (134) 0.69 (128) HDABI-use 0.69 (125) 0.70 (132) 0.87 (125) Daily Use Diary 0.85 (111) 0.78 (107) 0.95 (106) GHABP-use 0.74 (128) 0.59 (134) 0.79 (128) The number of participants with valid data for each between-interval comparison is shown in parentheses for each correlation. All correlations in this table are statistically significant (p 0.01). Thus, there was reasonable consistency in the satisfaction and usage ratings from individual participants in the 1-yr dataset. Fig. 4. Histograms of score differences for three measures of hearing aid satisfaction (left panels) and hearing aid usage (right panels) for the twoyear dataset. The score differences were calculated as described in Figure 3. Bin widths were 0.5 for all measures in this figure.

10 EAR &HEARING, VOL. 23 NO TABLE 6. Between-interval Pearson-r correlation coefficients for the 2-yr group and for each measure of hearing aid satisfaction and usage. Postfit Interval Pairs Dependent Variable 1 mo vs. 6 mo 1 mo vs. 1 yr 1 mo vs. 2 yr 6 mo vs. 1 yr 6 mo vs. 2 yr 1 yr vs. 2 yr HASS-global 0.72 (40) 0.63 (44) 0.60 (44) 0.81 (41) 0.73 (42) 0.87 (47) HASS-dispenser 0.49 (41) 0.53 (45) 0.61 (45) 0.71 (44) 0.40 (44) 0.71 (48) GHABP-satisfac 0.56 (45) 0.57 (48) 0.58 (48) 0.60 (45) 0.60 (45) 0.52 (48) HDABI-use 0.62 (43) 0.73 (46) 0.54 (46) 0.88 (43) 0.69 (43) 0.75 (46) GHABP-use 0.79 (45) 0.56 (48) 0.57 (48) 0.60 (45) 0.65 (45) 0.63 (48) Number of participants with valid data for each between-interval comparison is shown in parentheses for each correlation. All correlations in this table are statistically significant (p 0.01). Figure 4 displays the histograms of score differences for the satisfaction and usage measures from the 2-yr dataset. The layout of this figure is identical to that of Figure 3 except that usage measures derived from the daily diaries were not available for the 2-yr postfit interval. The histograms in Figure 4 confirm that the individual data are generally consistent with the group findings reported previously (Tables 2 and 4). That is, the mode for all five histograms of score differences in Figure 4 is 0 indicating that many participants failed to exhibit much change in satisfaction or usage over a 2-yr period. In several cases, however, the distributions of score differences are again skewed slightly toward negative values, which is consistent with the trend toward reduced satisfaction or usage from the 1-mo to 2-yr postfit intervals in the group data. Table 6 presents the Pearson-r correlation coefficients for each postfit-interval pair for all five outcome measures from the 2-yr dataset. Again, all correlations are positive, moderate (0.40) to strong (0.88) in magnitude, and statistically significant (p 0.01). Thus, there is reasonable consistency across participants in their satisfaction and usage ratings through the first 2 yr after hearing aid delivery. In summary, the group and individual data indicate that the hearing aid satisfaction and usage ratings obtained in this study were fairly stable throughout the first 2 yr after delivery of the hearing aids. This is consistent with the prior findings of Bentler et al. (1993) and Humes et al. (1996) obtained with abbreviated measures of satisfaction and usage and from a smaller number of participants. In several cases, however, significant effects of time on either satisfaction or usage ratings were observed. Although this occurred only a low percentage of the time, it is noteworthy that the direction of all such changes, when they did occur, was toward less satisfaction or less usage over time. In addition, the percentage of such significant changes in the group data was greatest for the participants in the 1-yr dataset and then diminished for the 2-yr dataset. This undoubtedly reflects the greater sensitivity to such differences in the data for the larger 1-yr dataset. Larger sample sizes mean smaller standard errors, which, in turn, increases the likelihood that small differences between means will be statistically significant. Thus, in principle, hearing aid satisfaction and hearing aid usage were observed to decrease statistically over time, but the size of the changes was small so as to be of little practical significance to individual hearing aid wearers. The latter statement is borne out by the histograms of score differences, which show that the vast majority of individual ratings changed by less than a scale unit over the time period being evaluated. All told, it appears that measures of hearing aid satisfaction at 1-mo postfit will provide a reasonable picture of hearing aid satisfaction throughout the first 2 yr. For hearing aid usage, the usage estimates provided at 1-mo postfit tend to be higher than those measured at other postfit intervals, but stabilize by 6-mo postfit. The histograms of score differences over time and the moderate to strong correlations in satisfaction and usage ratings across time suggest that the ranking of hearing aid wearers is fairly stable over time. That is, a hearing aid wearer with a relatively high satisfaction rating at 1-mo postfit is likely to maintain that high rating through the next few years and to remain more satisfied than an individual with an initially lower satisfaction rating. The same is generally true for hearing aid usage, except that there is a greater tendency for a large number of wearers to decrease their usage over time, especially during the first 6 mo after delivery. ACKNOWLEDGMENT: This research was supported, in part, by research grant R01- AG08293 from the National Institute on Aging. Larry E. Humes, Ph.D., Department of Speech and Hearing Sciences, Indiana University, Bloomington, IN E- mail: humes@indiana.edu. Received September 7, 2001; accepted May 3, 2002

11 438 EAR &HEARING /OCTOBER 2002 REFERENCES ANSI (1989). Specifications for audiometers, ANSI S New York: ANSI. Bentler, R. A., Niebuhr, D. P., Getta, J. P., & Anderson, C. V. (1993). Longitudinal study of hearing aid effectiveness. II: Subjective measures. Journal of Speech and Hearing Research, 38, Byrne, D., Parkinson, A., & Newall, P. (1990). Hearing aid gain and frequency response requirements for the severely/profoundly hearing impaired. Ear and Hearing, 11, Cox, R., & Alexander, G. (1992). Maturation of hearing aid benefit: Objective and subjective measurements. Ear and Hearing, 13, Gatehouse, S. (1992). The timecourse and magnitude of perceptual acclimatization to frequency responses: Evidence from monaural fitting of hearing aids. Journal of the Acoustical Society of America, 92, Gatehouse, S. (1999). Glasgow Hearing Aid Benefit Profile: Derivation and validation of a client-centered outcome measure for hearing aid services. Journal of the American Academy of Audiology, 10, Humes, L. E. (1999). Dimensions of hearing aid outcome. Journal of the American Academy of Audiology, 10, Humes, L. E. (2001). Issues in evaluating the effectiveness of hearing aids in the elderly: What to measure and when. Seminars in Hearing, 22, Humes, L. E., Barlow, N. N., Garner, C. B., & Wilson, D. L. (2000). Prescribed clinician-fit versus as-worn coupler gain in a group of elderly hearing-aid wearers. Journal of Speech-Language-Hearing Research, 43, Humes, L. E., Garner, C. B., Wilson, D. L., & Barlow, N. N. (2002a). Hearing-aid outcome measures following one month of hearing aid use by the elderly. Journal of Speech-Language- Hearing Research, 44, Humes, L. E., Halling, D., & Coughlin, M. (1996). Reliability and stability of various hearing aid outcome measures in a group of elderly hearing aid wearers. Journal of Speech and Hearing Research, 39, Humes, L. E., Wilson, D. L., Barlow, N. N., & Garner, C. B. (2002b). Changes in hearing-aid benefit following 1 or 2 years of hearing-aid use by older adults. Journal of Speech-Language-Hearing Research, 45, Kochkin, S. (1997). Subjective measures of satisfaction and benefit: establishing norms. Seminars in Hearing, 18, Saunders, G. H., & Cienkowski, K. M. (1997). Acclimatization to hearing aids. Ear and Hearing, 18, Surr, R. K., Cord, M. T., & Walden, B. E. (1998). Long-term versus short-term hearing aid benefit. Journal of the American Academy of Audiology, 9, Turner, C. W., Humes, L. E., Bentler, R. A., & Cox, R. M. (1996). A review of past research on changes in hearing aid benefit over time. Ear and Hearing, 17, 14S 25S. Ventry, I., & Weinstein, B. (1982). The hearing handicap inventory for the elderly: A new tool. Ear and Hearing, 3, Wechsler, D. (1981). The Wechsler Adult Intelligence Scale-Revised. New York: The Psychological Corporation. Weinstein, B. (1997). Outcome measures in the hearing aid fitting/selection process. Trends in Amplification, 2,