Designed to identify people with hearing loss and to be used as hearing aid outcome tools, the Self-Assessment of Communication (SAC) and its companion questionnaire, the Significant Other Assessment of Communication (SOAC), are structured on recommendations of the World Health Organization (WHO). The current study assesses important statistical properties of new 2007 computerized versions of SAC and SOAC and provides comparisons with other commonly used self-report tools.

Michael Hodes, AuD, is a recent AuD graduate of Idaho State University; Ronald L. Schow, PhD, is a professor of audiology; and Jeff Brockett, EdD, is associate professor of audiology in the Department of Communication Sciences & Disorders and Education of the Deaf at Idaho State University.

Hearing questionnaires are often used by those who test and treat hearing loss. They were first reported as being used by audiologists in 1964.1 Recently, they have been used for a variety of clinical purposes including use as an outcome measure.2-4 In 2007, Janota5 reported on clinical practice patterns by ASHA-certified audiologists and found that 35% were using “self questionnaires” to validate treatment outcomes. In a 2000 ASHA survey there was 53% use of self report when outcomes plus other uses are included.6

Frequently used self-assessment questionnaires include the Abbreviated Profile of Hearing Aid Benefit (APHAB),7 the Client Oriented Scale of Improvement (COSI),8 the Hearing Handicap Inventory for the Elderly (HHIE),9 and the Hearing Handicap Inventory for Adults (HHIA).10 This report contains statistical information to support the use of the Self-Assessment of Communication (SAC) and a companion questionnaire, the Significant Other Assessment of Communication (SOAC), which were developed by Schow and Nerbonne in 1982.11 The purposes of SAC and SOAC include both identification of persons with hearing loss and use as outcome assessment tools.12

The SAC and SOAC were recently picked by the large, four-state Hear USA company personnel as their first choice out of 20 different self-report outcome measures they evaluated in the context of hearing aid fitting.13 The SAC and HHIE-S (a shorter version of HHIE proposed by Ventry and Weinstein14) were recommended in the most recent (1997) American Speech-Language-Hearing Association (ASHA) guidelines for use in hearing screening of adults.15 The popularity of various self-report tools was also determined in the cited national survey wherein SAC and SOAC were shown to be in the Top-5 used by audiologists.6 The apparent support of the SAC and SOAC was, in part, the motivation for developing new computer versions whose statistical properties are reported in the current study. The computer versions were based on the SAC and SOAC as found in Schow and Nerbonne16 and are nearly identical to the 1982 version.11

However, the current findings relate to the 2007 version (ie, they should be distinguished from the 1982 version). Comparison between the two versions was not the purpose of the following study. Rather, the study provides a statistical foundation for SAC and SOAC, and it shows comparisons to similar outcome measures and we provide important data to help improve the day-to-day fitting of hearing aids.

Several recent reports have reviewed a process for choosing an appropriate self-assessment measure for outcome purposes.17,18 One of the important practical qualities of such a measure is having a light “burden” for clinicians and clients. SAC and SOAC do have this quality, as indicated by their brevity. The computerized scoring approach (developed for the current study) has the potential to further decrease the clinician burden. In addition, benefit, use, and satisfaction are important outcome domains, and these are all measured by the 2007 SAC and SOAC (for information, visit www.isu.edu/csed/profile).

The near-identical SAC and SOAC have been optimized to make them useful tools for validation of treatment outcomes, because they are based on recommendations of the World Health Organization.19 The first questions (Questions 1-5) focus on Disability or Activity Limitation, and later questions (Questions 6-9) focus on Handicap or Participation Restriction (with two of these last four questions as social and two emotional). Most (eight of 10) of the original SAC/SOAC questions from 1982 continue to be included in the current paper-pencil and computerized versions. And with new items added the computerized SAC and SOAC allow measurement of several important areas of inquiry for an outcome measure, including benefit (pre- minus post-disability and handicap), residual-disability and handicap (post results), social issues (others bothered, negative effect on social life), life quality, use, and satisfaction. (Complete coverage of these issues is similar to the International Outcome Inventory for Hearing Aids (IOI-HA)20 and both are brief and comprehensive compared to other similar tools.)

Background on Current Study

The goal of the current report was to make these two self-report measures (SAC & SOAC) even more attractive for use, via quick computer scoring, as has been done with other self-report tools.21 Technical elements desirable in any outcome measure include the availability of norms, reliability data, and critical difference (CD) scores. All three of these items have been reported for the 1982 SAC22 (found at www.isu.edu/csed/profile). Strong correlations with other similar measures such as HHIE-S (as found in Schow et al23 and Frank, Bennett, and Blood24) help to demonstrate concurrent validity, another important technical feature. Statistical findings gathered here for the 2007 versions thus add support to the previous data, as based on well-supported psychometric standards.25

Number of asterisks corresponds to the number of participants in that Pure-Tone Group who wore hearing aids.
TABLE 1. Mean HFPTA by age and number of participants in HFPTA hearing loss groups, normal-hearing group, and Pure Tone Groups (PTGs) by Age Interval (N=60).

The current study follows the pattern of previous similar studies by including those with normal hearing and those with hearing loss, and because it uses subjects where there is no change in hearing status. As in previous studies, we found in our subjects that there was no reported or measured change in hearing for the subjects involved.21,26,27

In summary, the current study was designed to determine important statistical properties of new 2007 computerized versions of SAC and SOAC and provides comparisons with other commonly used self-report tools.

Study Methods

Participants. This study utilized 60 participants ages 25 to 84 years, an age range chosen that encompasses 90% of those who wear hearing aids in the United States.28 Participants were recruited to fill age groups as divided up into the following 10-year age intervals: 25-34, 35-44, 45-54, 55-64, 65-74, and 75-84. Ten participants (five males, five females) were included in each age interval. Participants were recruited in community and church settings in the communities of Meridian and Pocatello, Idaho, and are representative of adults in these communities who have a concern for and interest in having their hearing tested. Thus, they were judged to be typical of persons who fill out SAC and SOAC forms in community screening and clinical situations. The inclusion of equal numbers in each age interval (rather than a simple convenience sample) was made to ensure that the sample included strong representation for the entire age range for which it might be applied.

As individuals volunteered, about half of the 60 subjects had hearing within normal limits, while the remainder had some form of hearing loss. An attempt was made to distribute evenly the number of individuals with normal hearing and with hearing loss within each age group. Participants who volunteered were accepted until each of the desired cell groups was filled with an appropriate mix of individuals. There were eight people in the entire sample who were hearing aid users.

Post-hoc examination revealed that the average age of the participants overall was 54.3 years (SD=17.7; Range=26-84). The average age for males and females was 54.1 years and 54.4 years respectively (M:SD=17.6, range=27-83; F:SD=18.0, range=26-84).

Hearing status of each participant was measured using a Maico MA-40 portable audiometer, which was calibrated electroacoustically prior to data collection and biologically (ie, using listening checks) on a daily basis. Evaluation of the typical sound environments where it was most practical to test participants for this study (quiet rooms in the homes of participants) revealed noise levels 10-20 dB below the maximum permissible noise levels allowed for occupational noise test conditions.29 Measurements were made with a B&K 2203 sound level meter and associated octave filter (B&K 1613). Therefore, noise environments used in this study were judged to be acceptable for current purposes but were verified by listening tests on site.

Participants were categorized by hearing status as determined by the thresholds of the better ear from the initial test session. The determination of hearing loss or normal hearing was based on high-frequency pure-tone averages (HFPTA) similar to the Holcomb and Punch21 study. HFPTAs were based on the average of 1000 Hz, 2000 Hz, and 4000 Hz. Pure Tone Groups (PTGs) were also calculated. An individual was placed in a PTG based on two data points (threshold at 1000 Hz and the average of the thresholds at 2000 Hz and 4000 Hz). For details on PTG and SAC/SOAC norms by PTG, see www.isu.edu/csed/profile.

FIGURE 1. Thresholds (M and SD) in better ear for participants in current study.

Table 1 shows the number of participants with a hearing loss based on HFPTA along with hearing according to the PTG strategy. Using HFPTA, two fences were examined. Using a demarcation line of hearing loss at greater than 20 dBHL, the number of participants from the current study with a hearing loss was 32/60 (53%). This included 20 males and 12 females. With the fence greater than 25 dBHL, the number with loss was 24/60 (40%) with 18 males and 6 females. Using the PTG strategy, the number with hearing loss was 32 (20 males and 12 females). Table 1 also contains a summary of the total number of individuals in each PTG. Unless otherwise noted for all subsequent hearing loss comparisons throughout the study, the greater than 20 dB HFPTA method was used to define the hearing loss group.

Ideally, the study would have included an equal distribution of individuals with hearing loss across all age intervals. As seen in Table 1, this was achieved only in the 45-54 age interval. The larger number of hearing loss participants in the 75-84 year old group helped balance out the participants without hearing loss in the younger groups and made an almost equal distribution for hearing loss and normal hearing across the total age range.

Of the 60 participants in the current study, there were eight male and no female participants who regularly wore hearing aids, as indicated by asterisks in Table 1. The remainder of the participants either did not currently wear hearing aids or were those with normal hearing.

The mean hearing loss for the current participants is shown in Figure 1. This audiogram represents the combined thresholds of all participants using the better ear thresholds. The mean hearing status for the current study showed a mild hearing loss of 32 dBHL found only at 4000 Hz. The average hearing thresholds obtained by male as compared to female participants in the current study were very similar except that men, on average, had poorer 4000 Hz thresholds than did women (41 dBHL vs 22 dBHL).

Figure 2 shows the normal-hearing group (left audiogram) and hearing-impaired group (right audiogram) when average hearing thresholds of all participants were combined based on hearing status. Mean thresholds were within normal limits for the normal-hearing group, as would be expected. The average hearing thresholds from the 32 participants who had hearing loss show that averages were outside of normal limits (>25 dB) at 500 Hz (26 dBHL), 2000 Hz (30 dBHL), and 4000 Hz (53 dBHL). The SDs also were larger in the hearing loss group, especially at 2000 and 4000 Hz (SDs of 15 dBHL). Average thresholds of normal-hearing male participants and normal-hearing female participants in this study were 6-16 dBHL and were approximately the same in both groups.

FIGURE 2A & 2B. Thresholds (M and SD) for participants in the current study with normal hearing (left) and those with hearing loss (right).

Computerized SAC and SOAC. The computerized SAC and SOAC were run on a Dell Inspiron laptop computer. A face-to-face hearing-related case history of participants was taken using a standard clinical procedure. Participants were excluded from this study if they had any previous contact with the SAC or the SOAC, as determined by self-report.

The programming for the computerized SAC and SOAC was written in Macromedia Flash (Version 8.0). The large type size and large icon buttons allowed for minimal mistakes when clicking with a mouse (there is also a touch-screen option).

TABLE 2. Mean, SD, and range of self-report scores for normal-hearing and hearing-impaired groups based on a HFPTA >20 dB criterion for loss.

Two 9-item computer versions were evaluated in this study. Both these versions have an additional question (10 items total) dealing with hours of current hearing aid use and the amount of satisfaction from hearing aids. The 10-question versions are used after hearing aid fitting. This tenth question was not evaluated in the current study. All 10 questions, including the nine from which we derived the current statistical data, are shown in Figure 3, and computer versions can be found at www.isu.edu/csed/profile.

Because the current study focused on self-reported hearing status without the use of amplification (see instructions in the questionnaire in the online version of this article), the eight participants who had amplification were simply asked to report their hearing based on experiences when not using hearing aids as were significant others. In all eight cases, respondents felt they were able to do this. These instructions, involving the nonuse of hearing aids, of course, would not apply when SAC and SOAC are used a second time as outcome measures; the instruction in that case asks for reports based on hearing aid use.

For each item on the SAC and SOAC, there were five possible answers that could be selected: 1) Almost never (or never); 2)Occasionally (about one-fourth of the time); 3) About one-half of the time; 4) Frequently (about three-quarters of the time); and 5) Practically always (or always).

When the last question had been answered, the application displayed a summary score sheet showing all of the questions and the answers that had been provided by the participant. On the right-hand side of the score sheet was a summary based on the WHO definitions of Activity Limitation or AL (Disability)(Questions 1-5) and Participation Restriction or PR (Handicap)(Questions 6-9). There was also an entry for Health Related Quality of Life (HRQoL)(Question 9), which is considered to be one of the four PR items.

On the right side of each of these SAC/SOAC subsection labels was a calculation of the composite score (in %) for each subsection. Percentage scores for each subsection were derived by dividing the total raw score by the number of items (5, 4, or 1), subtracting -1, and multiplying by 25 (eg, 25/5=5-1=4×25=100%).

Below the WHO classifications was the total composite percentage score for all of the three sections. On the bottom right-hand side of the score sheet was a list of the composite scoring classifications in percent divided into four degree categories: (0-20%) no disability/handicap, (21-40%) slight disability/handicap, (41-70%) mild-to-moderate disability/handicap, and (71-100%) severe hearing disability/handicap.22,23,31,32 On the bottom of the page were PRINT, EXIT, and REPEAT buttons (the last button allowed the individual to take the same questionnaire again). The REPEAT button was also helpful if the individual was not satisfied with some of his/her answers and wished to change them.

Procedures. The procedures of this study closely followed the two-session strategy of Holcomb and Punch21 on a similar computerized tool, the MHHI. In the current study, the sessions were set at 1- to 2-week intervals. In the first session, there was an interview/case history performed, completion of the computerized SAC, and audiometric testing.

While the participant was filling out the case history or having a hearing evaluation, the significant other associated with the participant was completing the computerized SOAC, but this person had no knowledge of how the SAC items had been answered by the participant. This lack of knowledge was in full effect on the first test, but could not be totally controlled at retest (ie, there was no attempt made to control for participants discussing the procedures used with their significant other during the test-retest interval). Therefore, the most emphasis for SAC and SOAC comparisons was placed on the first testing session.

In the second session, there was again an interview/case history performed, completion of the computerized SAC, and audiometric retest. The significant other was also retested on SOAC. For the participant case history, a simple question of, “Has there been any significant change in your hearing, since the last time we met?” was asked along with any follow-up based on the response. We had planned to ask additional questions if any change in hearing had been reported, but this did not occur so these were not needed. In audiometric retest we found largely + or – 0 dB and + or – 5 dB (468/480 retests) and 12 times we found + or – 10 dB (no more than one 10 dB change per subject and no sign of systematic change in any person). Each participant again answered the questions on the computerized SAC based on unaided listening. If a participant in the study had no previous computer skills or knowledge, it was found that, after a brief introduction of how to use a computer mouse/scroll pad, none of these participants had any difficulty. Significant others were not the focus of this study and no data were gathered on them beyond SOAC scores, except the fact that 58/60 were spouses. The remaining two cases involved adult children (age 25 or older) and parent relationships.

TABLE 3. Test-retest correlations for a variety of self-report questionnaires (time interval from test to retest is shown in parenthesis—Cox and Rivera35 recommend 2-3 weeks).

Results and Discussion

Overall performance. Means and SDs for the total group on the initial administration of the SAC and SOAC were 16.5 (7.5) and 16.4 (8.3), respectively. The means and SDs of the differences between the initial and retest scores overall on the SAC and SOAC were 0.3 (2.8) and 0.02 (1.9), respectively. Scatterplots showed a direct linear correspondence between SAC and SOAC test and retest scores. Means and SDs for the SAC and SOAC at each test time for individuals with normal hearing and for those with hearing loss are shown in Table 2.

The mean percentage scores for normal-hearing persons in Table 2 are quite similar for both the SAC and SOAC (6 to 8%). Thus, they are well within the middle of the normal range previously established for the 1982 SAC and SOAC, which seems appropriate for the current version (0-20%). The mean scores for the hearing loss group were outside normal limits for both SAC and SOAC. A feasible interpretation of the difference between the groups (loss and no loss) is that both SAC and SOAC appeared to capture appropriately the impact of a hearing loss.

Internal consistency. Cronbach alpha was determined for the first SAC test and the first SOAC test. The findings were 0.94 for SAC and 0.96 for SOAC, which suggested strong internal consistency reliability. This is comparable to findings by Holcomb and Punch,21 where they ranged from .85 to 0.93. According to Hyde,25 values above 0.85 are considered “acceptable” in health-related fields.

Test-retest reliability. The Pearson test-retest correlation on all subjects in the current study was 0.94 for the computerized SAC and 0.97 for the computerized SOAC (p <.001). Pearson r test-retest correlations for Activity Limitation (Questions 1-5) were 0.95 for SAC and 0.73 for SOAC and for Participation Restriction (Questions 6-9) were 0.79 for SAC and 0.95 for SOAC (all ps <.001).

The current study also examined social and emotional scores separately. In order to obtain the social score for the current study, Questions 6 and 7 were used, and the correlations were 0.85 for SAC and 0.92 for SOAC. In order to obtain the emotional score for the current study, Questions 8 and 9 were used, and the correlations were 0.78 for SAC and 0.95 for SOAC (all ps < .001).

Test-retest correlations for those with normal hearing were 0.96 for SAC and 0.92 for SOAC, and for hearing loss it was 0.90 for SAC and 0.96 for SOAC. For males and females, the correlations were relatively high at 0.91 and 0.97, respectively, for SAC, and 0.97 and 0.95, respectively, for SOAC.

When we looked at individuals who were classified in the younger group (<55) and participants who were classified in the older group (≥ 55), we found both had equally high test-retest correlations. For SAC they were 0.92 and 0.94, respectively, and for SOAC they were 0.97 and 0.98, respectively (all ps < .001).

In addition to findings from the current study, Table 3 also includes findings for the earlier 1982 paper-and-pencil version of the SAC and SOAC,11 as well as test-retest findings for several similar self-report instruments. Findings clearly indicate a high degree of test-retest reliability for the 2007 computerized SAC and SOAC. When compared to the earlier paper-and-pencil version, the test-retest findings in the current version appear to be somewhat better. Correlation findings from the current study appear to compare quite favorably to other commonly used self-report questionnaires.

Standard error of measurement (SEM) and critical differences (CDs). Table 4 shows a derivation of SEMs and CDs. The CDs in the present study suggest that an improvement of 4-5 points in raw scores is necessary to indicate a real change (95% confidence level). For the SAC, this translates to a percentage improvement of 15% or more; for the SOAC, the improvement needed is 12% or more.

The current findings show very high correlations and little, if any, difference between those with and without hearing loss (rs = 0.90-0.96). One of the main reasons these high correlations are helpful is that reliability findings are used to derive SEMs (SEM=SDx square root of 1-r), which are used in the derivation of CDs. CDs, in turn, help in the determination of whether outcomes are truly better following treatment [CD=(2)(square root of 2) (SEM)]. As shown in Table 4, the CDs in the present study are similar to or smaller than findings on another computerized self-report tool, which were also derived on a sample of persons with and without hearing loss.21

Comparisons between SAC and SOAC total scores. A repeated-measures analysis of variance (ANOVA) was run on the scores of the SAC and SOAC tests and the retests for the current study. Results revealed that there was no significant difference between any of the SAC and SOAC scores (F1,59=.006, p=.941) (F1,59=.494, p=.485). There was also no interaction between any of the four test or retest situations (F1,59=.444, p=.508). In short, the current participants reported on the impact of their hearing problems in a similar way as compared to their significant others (spouses, etc).

When the various computerized SAC and SOAC tests were correlated with each other and the second tests were correlated with each other, all correlation coefficients (r) were 0.80 or higher (all ps < .001).

As noted earlier, comparisons between SAC and SOAC have previously demonstrated a close relationship in total score.11,23,24 The data show that participants in the current study reported no significant difference in their perceptions of these hearing problems when compared to their significant others. All of the tests and retests of SAC and SOAC were also highly correlated.

These findings suggest that mean data, variance data, and test-retest correlations for all SAC findings tend to reinforce the SOAC findings in terms of a magnitude estimation of hearing status.

Summary

The main goal of the current study was to determine important statistical properties of the 2007 computerized SAC and SOAC based on a sample of individuals who might typically complete such an instrument. We were able to show that the computerized SAC and SOAC have similar mean scores, SDs, and high Cronbach alpha and test-retest correlations. This means that an individual who is given the computerized SAC or SOAC would be expected to score approximately the same on both test and retest if given the questionnaire a second time within a relatively short period of time (a few weeks). This is, of course, assuming that their hearing status has not changed, as was the case with the current participants.

*Correlation is significant at the 0.01 level (2-tailed).
TABLE 4. Computerized SAC and SOAC mean (M), standard deviation (SD) correlation coefficients (r), standard error of measurement (SEM), and the critical differences (CDs) for the present group of participants, along with comparable Holcomb & Punch21 (H&P, 2006) data.

SAC and SOAC have often been used in community hearing screening to separate those who may have hearing problems from those who do not. SAC and SOAC have also been used as outcome measures after hearing aid fitting is completed in an attempt to improve hearing. The critical difference (CD) scores derived here were 15% for SAC and 12% for SOAC, meaning that an improvement of those amounts is required with hearing aid fitting before we can have 95% confidence that communication is better when the hearing aid is in use. A change in score, therefore, of 12% or 15% on SOAC and SAC, respectively, can be regarded as a true difference based on the current sample.

The current findings, along with previous findings on earlier versions of these questionnaires, suggest that normal-hearing people will have mean scores on computerized SAC and SOAC in the midrange between 0 and 20%. Those with hearing loss similar to the mean for the current sample of individuals with hearing loss will have scores averaging about 32%.

Since there is such high test-retest reliability, we can conclude that—when a person with hearing loss reports (and others report) difficulty of 32% and is treated with amplification, and then on SAC retest reports scores of 17% (or on SOAC 20%) or better—there is a 95% likelihood that their new scores represent a real improvement. This is perhaps the most important aspect of the current study.

Acknowledgments

This study was an AuD doctoral project by Michael Hodes as directed by Ronald Schow, with major support from Jeff Brockett. The authors thank Teri Peterson, Tom Longhurst, and Peter Flipsen Jr for their generous help with statistics and manuscript preparation, and all those participants who donated their time to make this study possible.

FIGURE 3. Paper-and-pencil forms of Self-Assessment of Communication (SAC) and Significant Other Assessment of Communication (SOAC)[images above are linked to downloadable pdfs]. Similar computer versions can be found at www.isu.edu/csed/profile.

Citation for this article:

Hodes M, Schow R, Brockett J. New support for hearing aid outcome measures: The computerized SAC and SOAC. Hearing Review. 2009;16(12):26-36.

References

  1. High W, Fairbanks G, Glorig A. Scale for assessment of hearing handicap. J Speech Hear Dis. 1964;29:215-230.
  2. Schow R, Gatehouse S. Fundamental issues in self-assessment of hearing. Ear Hear. 1990;11:6S-16S.
  3. Noble W. Self-Assessment of Hearing and Related Functions. London, UK: Whurr Publishers, Ltd; 1998.
  4. Alpiner J, Schow R. Rehabilitative evaluation of hearing-impaired adults. In: Alpiner JG, McCarthy PA, eds. Rehabilitative Audiology; Children and Adults. Baltimore, Md: Lippincott Williams & Wilkins; 2000:308-315.
  5. Janota J. ASHA [2006] Audiology Survey: Clinical Focus Patterns. Rockville, Md: American Speech-Language-Hearing Assn; 2007.
  6. Millington D. Audiologic rehabilitation practices of ASHA audiologists: Survey 2000. Masters Thesis, Idaho State University; 2001.
  7. Cox R, Alexander G. The Abbreviated Profile of Hearing Aid Benefit. Ear Hear. 1995;16:176-186.
  8. Dillon H, James A, Ginis J. Client Oriented Scale of Improvement (COSI) and its relationship to several other measures of benefit and satisfaction provided by hearing aids. J Am Acad Audiol. 1997;8:27-43.
  9. Ventry IM, Weinstein BE. The Hearing Handicap Inventory for the Elderly: a new tool. Ear Hear. 1982;3:128-134.
  10. Newman CW, Weinstein BE, Jacobson GP, Hug GA. The Hearing Handicap Inventory for Adults: psychometric adequacy and audiometric correlates. Ear Hear. 1990;11:430-433.
  11. Schow R, Nerbonne M. Communication Screening Profile: use with elderly clients. Ear Hear. 1982;3:135-147.
  12. Gailey CE. Self-assessment of communication ability before and after fitting of hearing aids. Masters Thesis. Logan, Utah: Utah State University; 1987.
  13. Nemes J. Despite benefits of outcomes measures, advocates say they’re underused. Hear Jour. 2003;56:19-25.
  14. Ventry IM, Weinstein BE. Identification of elderly people with hearing problems. ASHA. 1983;25:37-42.
  15. American Speech-Language-Hearing Assn (ASHA). Committee on Audiometric Evaluation. Guidelines for Audiologic Screening. Rockville, Md: ASHA; 1997:58-59.
  16. Schow R, Nerbonne M. Introduction to Audiologic Rehabilitation. 5th ed. Boston, Mass: Allyn and Bacon; 2007.
  17. Bentler RA, Kramer SE. Guidelines for choosing a self-report outcome measure. Ear Hear. 2000;21:7S-49S.
  18. Cox RM. Choosing a self-report measure for hearing aid fitting outcomes. Semin Hear. 2005;26(3):149-156.
  19. World Health Organization. International Classification of Functioning, Disability and Health. Geneva: WHO; 1980, 2001.
  20. Cox R, Hyde M, Gatehouse S, et al. Optimal outcome measures, research priorities, and international cooperation. Ear Hear. 2000;21[Suppl]:106S-115S.
  21. Holcomb S, Punch J. Multimedia Hearing Handicap Inventory: reliability and clinical utility. Am J Audiol. 2006;15:3-13.
  22. Schow R. The status and future of SAC & SOAC. Presented at: International Collegium of Rehabilitative Audiology; Goteberg, Sweden; 1995.
  23. Schow R, Smedley T, Longhurst TL. Self-assessment and impairment in adult/elderly hearing screening—recent data and new perspectives. Ear Hear. 1990;11:17S-27S.
  24. Frank T, Bennett S, Blood I. Relations between hearing handicap and impairment. Presented at: annual American Speech-Language-Hearing Association convention; St Louis; 1989.
  25. Hyde ML. Reasonable psychometric standards for self-report outcome measures in audiological rehabilitation. Ear Hear. 2000;21(4):24S-36S.
  26. Newman CW, Weinstein BE. Test-retest reliability of the Hearing Handicap Inventory for the Elderly using two administration approaches. Ear Hear. 1989;10:190-191.
  27. Newman CW, Weinstein BE, Jacobson GP, Hug GA. Test-retest reliability of the Hearing Handicap Inventory for Adults. Ear Hear. 1991;12:355-357.
  28. Kochkin S. MarkeTrak VII: Hearing loss population tops 31 million. Hearing Review. 2005;12(7):16-29.
  29. Occupational Safety and Health Administration (OSHA). Occupational noise exposure: Final rule. Federal Register. 1983;48:9738-9785.
  30. Brockett J, Schow R. Web site profiles common hearing loss patterns and outcome measures (www.isu.edu/csed/profile), announced as part of Page Ten. Hear Jour. 2001;54:8, 10-20.
  31. Schow R, Tannahill C. Hearing handicap scores and categories for subjects with normal and impaired hearing sensitivity. J Am Aud Soc. 1977;3:134-139.
  32. Schow R, Brockett J, Sturmak M, Longhurst T. Self-assessment of hearing in rehabilitative audiology: developments in the USA. Brit J Audiol. 1989;23:13-24.
  33. Short BJ. Communication ability assessment of an adult population. Masters Thesis, Idaho State University; 1982.
  34. Weinstein BE, Spitzer JB, Ventry IM. Test-retest reliability of the Hearing Handicap Inventory for the Elderly. Ear Hear. 1986;7:295-299.
  35. Cox RM, Rivera IM. Predictability and reliability of hearing aid benefit measured using the PHAB. J Am Acad Audiol. 1992;3:242-254.
  36. Gatehouse S. Glasgow Hearing Aid Benefit Profile: derivation and validation of a client-centered outcome measure of hearing aid services. J Am Acad Audiol. 1999;10:80-103.
  37. MRC Institute of Hearing Research. Glasgow Hearing Aid Benefit Profile. 2007. www.ihr.mrc.ac.uk/scottish.
  38. Cox R. Hearing Aid Research Laboratory. 1997. www.ausp.memphis.edu/harl/index.html. Accessed August 24, 2007.

Correspondence can be addressed to Ronald L. Schow, PhD, at .