Hearing Handicap Ratings Among Different Profiles of Adult Cochlear Implant Users

Hearing Handicap Ratings Among Different Profilesof Adult Cochlear Implant Users

William Noble,1,2 Richard Tyler,2 Camille Dunn,2 and Navjot Bhullar1

Objective: The aim was to compare outcomes in thedomain of self-reported hearing handicap acrossgroups of patients fit with one versus two cochlearimplants (CI, CI � CI), or with an implant and ahearing aid (HA) in the nonimplanted ear (CI � HA).

Design: The design was retrospective, and a prelim-inary step was to factor analyze the two measuresused, namely, the Hearing Handicap Inventory forthe Elderly (HHIE) and the Hearing Handicap Ques-tionnaire (HHQ). Longer versus shorter-term expe-rience with a single implant profile was compared,and further analysis confined to patients fit for lessthan 100 mo across the three profiles. Pre- versuspostimplant self-report and performance (speechtest, localization) data were also compared.

Results: Three factors were identified in the HHIE,labeled Emotional Distress (HHIE), Difficulty inHearing, and Social Restriction (HHIE). Highesthandicap score for Emotional Distress (HHIE) wasobserved in the CI � HA group. There were signifi-cantly lower scores for Difficulty in Hearing in theCI � CI group than in the CI (p � 0.02) or CI � HA(p � 0.001) groups. On the Social Restriction (HHIE)subscale, the CI � CI group reported significantlylower rating than the CI (p � 0.009) or CI � HA (p �0.006) groups. Two factors were identified in theHHQ, labeled Emotional Distress (HHQ) and SocialRestriction (HHQ). Significantly higher EmotionalDistress (HHQ) score was observed in the CI � HAgroup than in the CI � CI group (p � 0.002); signif-icantly lower Social Restriction (HHQ) score wasfound in the CI � CI group than in the CI (p � 0.02)or CI � HA (p < 0.001) groups. Pre-post speech testperformance showed least contrast in the CI � HAgroup.

Conclusions: Outcomes demonstrate an evident re-duction from single or bilateral implantation in thearea of emotional distress and a further advantagefrom bilateral implantation in the areas of hearingdifficulty and social restriction.

(Ear & Hearing 2008;29;1–000)

It is increasingly recognized (Humes, Wilson,Barlow, Garner, & Amos, 2002; Summerfield et al.,2006) that assessment of the benefit from any inter-vention designed to reduce disability, including fit-

ting of devices in the case of impaired hearing, needsto take into account the experience of the personunder treatment. To this end, in recent years, theUniversity of Iowa cochlear implant (CI) programhas augmented its clinical testing protocol by addinga range of measures to assess self-reported outcomesin the domains of device performance, residual dis-abilities, and residual handicaps—the last ex-pressed in terms of emotional stress and limitationson social interaction. The focus of the present paperis to report results from two measures of handicap:The Hearing Handicap Inventory for the Elderly(HHIE; Ventry & Weinstein, 1982), and the HearingHandicap Questionnaire (HHQ; Gatehouse & Noble,2004).

The central purpose of the present report is tocompare results from the foregoing handicap mea-sures across three implant profiles: people fit withone implant, those fit with two, and those fit with animplant and who retain an acoustic hearing aid (HA)in the nonimplanted ear. Because these profilesprovide quite different hearing and listening expe-riences there is good reason to appraise outcomes forpatients treated in these different ways. In additionto the intrinsic value of such examination, largerquestions are raised about the properties beingassessed by the measures we have applied. Weprovide summary descriptions of pre-implant self-report and performance test data as an aid to inter-preting the postimplant self-report findings.

One larger question about what is being assessedby the scales we have used, provoked especially bythe use of the HHIE, concerns the concept of hand-icap, the term featured in the titles of both question-naires referred to here. The term handicap does nothave universally agreed meaning and this has im-plications for assessment (Noble, 1998, p. 15–16). Ifdifferent investigators take the term to refer todifferent domains, or define those domains differ-ently, they will end up assessing different proper-ties, yet under the same supposed rubric.

As regards hearing impairment, the matter ofdefinition could be said to have been settled by theWorld Health Organization (1980) that distin-guished the terms disability and handicap in thefollowing way: disability refers to the effects of animpairment on abilities in real-world contexts, suchas to hear and discriminate speech and nonspeech

1School of Psychology, University of New England, Australia; and2Department of Otolaryngology, University of Iowa, Iowa.

balt5/zau-aud/zau-aud/zau00108/zau3111-08z xppws S�1 11/12/07 11:58 Art: AUD200208 Input-js

0196/0202/08/2901-0001/0 • Ear & Hearing • Copyright © 2008 by Lippincott Williams & Wilkins • Printed in the U.S.A.

1

sounds in the social and physical environments;handicap refers to nonauditory consequences for theindividual that flow from any disabilities, such asexperiences of emotional distress and restrictions onsocial engagement, directly because of hearing dis-abilities.

The World Health Organization produced a fur-ther scheme (2001), which introduced a somewhatdifferent distinction, between activity and participa-tion. Activity refers to “executing a task or action,”participation refers to “involvement in a life situa-tion.” Although the first term can be aligned with(dis)ability, as defined in the previous scheme, thesecond captures only the “restriction” element ofhandicap, neglecting the personal stress that is somuch a feature of the limiting of activity, especiallycommunicative activity. Our preference is to rely onthe earlier WHO terms when approaching an under-standing of handicap, and its contrast with disability.

The results reported in the present paper offer anopportunity to test the reality of a “restriction-distress” construction of handicap. They also offeropportunity to examine the reality of the contrastbetween handicap and disability as defined in WHO(1980). That contrast is not reflected in the assumedcontent of the HHIE. The claim by the authors ofthis scale (Ventry & Weinstein, 1982) is that itcomprises two subscales: one on social restrictionand one on emotional distress. Noble (1998, p. 16)has argued that a number of items taken to belong tothe “social” subscale could be interpreted as belong-ing to the disability domain. Thus, “Does a hearingproblem cause you difficulty when attending aparty?” could be understood as addressing the em-barrassment and potential withdrawal that mightoccur in that context (i.e., a handicap item), or itcould be read, more straightforwardly we argue, asan example of reduced ability to hear speech.

How to test these points uses the technique offactor analysis, a statistical procedure that allowsdiscovery of subsets of items that belong with eachother and, in content, form interpretable groupings.The data set accumulated in the Iowa implantsample is sufficiently large to permit factor analysesto be undertaken as a key methodological step togain understanding of the structure of the measuresin question, before the comparison of interest, thatis, across the different implant profiles. The outcomefrom the factor analysis of the HHIE, described indetail in one section of the Results and in theAppendix to the present paper, bears out the argu-ment that this scale indeed seems to assess aspectsof disability as well as handicap.

We are aware of three reports in the publishedliterature using developed measures of hearinghandicap that were applied to (unilateral) implant

patients (Hawthorne et al., 2004; Vermeire et al.,2005, 2006). Hawthorne et al. used a scale derivedby Hawthorne and Hogan (2002) that addressesaspects of emotional distress and social restriction.This was used with 34 patients pre- and postimplan-tation, and a decline in rated handicap of 42% wasobserved. Vermeire et al. (2005) used the HearingHandicap Inventory for Adults, a variant of theHHIE (Newman, Weinstein, Jacobson, & Hug,1990), pre- and postoperatively, among 29 CI pa-tients and observed a 30% decrease in overall hand-icap ratings. Vermeire, et al. (2006) also applied theHHIA to a group of seven patients with genetichearing impairment (DFNA9), pre- and postimplan-tation, and observed a similar outcome. We can thusexpect that rated handicap in the unilateral CIpatients will have improved. Bilateral implantationmay be expected to show incremental further hand-icap reduction, because of the improved self-ratedspatial and speech hearing associated with thatprofile (Summerfield, et al., 2006). Because self-rated abilities of the CI � HA profile are unknown,no prediction in the handicap domain is made re-garding that profile.

METHOD

Participants

The total number of implant patients providingdata for parts of this report is 183. Of these, 106have one implant, 35 have two, and 42 have oneimplant and a HA in the nonimplanted ear. Table 1shows age (in years) at the time of initial postim-plant self-report, and time (in months) since implantfor patients from the above three groups. This infor-mation is for the subsets of patients whose datawere used in the comparative analysis (i.e., im-planted up to 100 mo before initial self-report).Because this is a retrospective investigation there isless control over variation in a factor such as timesince implanting. Historically, unilateral implanta-tion has given way in more recent years to increased

TABLE 1. Gender distribution, average age and average postim-plantation time (SDs in brackets) for the one cochlear implant(CI), two cochlear implant (CI � CI), and cochlear implant andhearing aid (CI � HA) groups, for those implanted <100 mopreviously

CI(n � 71)

CI � CI(n � 35)

CI � HA(n � 40)

Male (%); Female (%) 50.7%;49.3%

42.9%;57.1%

30%;70%

Age (in years) 60.6 (15.1) 64.3 (15.5) 61.9 (15.9)Postimplantation time

(in months)42.2 (26.6) 34.1 (24.8) 27.7 (14.7)


NOBLE ET AL. / EAR & HEARING, VOL. 29, NO. 1, 0–02

T1

incidence of bilateral implantation; hence, on aver-age, those with one implant have had their devicefor longer. Patients who continue to use an acousticHA in the nonimplanted ear have made that as adeliberate decision, as against others who persist fora while with this profile but then discard one orother hearing device. There are no significant gen-der differences between CI and CI � CI groups;however, there is a greater incidence of females thanmales in the CI � HA group (�2

(1) � 6.40, p � 0.05).Patients whose involvement in the implant pro-

gram is via one or another funded research projectact as subjects across a wide range of clinical,performance and psychometric tests. All have signedtheir consent to participate and the program isapproved by the Institutional Review Board. Thetwo scales of particular interest in the present reportform part of a set of self-administered scales thatpatients complete subsequent to implantation. Insome cases, these scales were completed before im-plantation, and we report on pre- versus postimplantchange in scores within the comparative analysis.

Measures of Handicap

Self-report measures were introduced to the as-sessment protocol several years after the implantprogram itself was implemented at Iowa. Conse-quently, more recently implanted patients are rat-ing their abilities and problems soon after receivingtheir implant/s; those implanted earlier in the pro-gram are necessarily providing these ratings afterlengthier experience. An initial analysis comparesself-report average scores for those fit more than 100mo before initial self-report, versus those fit up to100 mo before.

The HHIE is a widely used questionnaire com-prising 25 items in two subscales, labeled social andemotional handicap. It is designed for self-adminis-tration as a paper-and-pencil test. Items are re-sponded to using a three-point scale (yes, some-times, no), with higher scores associated withgreater handicap. Response on the HHIE has beenshown to be a sensitive index of overall quality of life(Dalton, et al., 2003). A slight variant, the HearingHandicap Inventory for Adults (Newman, et al.,1990), has been successfully applied among implantpatients (Vermeire, et al., 2005, 2006). The items ofthe HHIE are general in reference; the changes toform the HHIA introduced items referring to work-place settings, potentially reducing its relevance forelements of a clinic population like the one at Iowa.

The HHQ is a 12-item self-administered question-naire, designed and confirmed to be unifactorial(Gatehouse & Noble, 2004), though nonetheless ad-dressing two topics: social limitation and emotional

distress. Responses are on a five-point scale (almostalways, often, sometimes, rarely, never) with higherscores indicating greater handicap.

Performance Measures

The performance measures are a single wordspeech recognition task and a directional hearingtask. The speech test is the recorded form of theConsonant-Nucleus-Consonant monosyllabic words(Tillman & Carhart, 1966). It was conducted in quietsound-field conditions (large test booth), with thesource at 0° azimuth, and signals presented at 70dB(C). Scoring is based on percent-correct perfor-mance at the word level. Testing was undertakenseparately for left, right, and both ears; the scorereported here is the average over these three condi-tions.

The directional hearing test uses everydaysounds (e.g., dog barking, buzzer, telephone ring)presented from any of eight numbered loudspeakersin the frontal horizontal plane. Sounds are pre-sented at 70 dB(C). The loudspeakers are 15.5°apart forming a 108° arc. The listener sits facing theloudspeaker arc, at 0° azimuth, and at a distance of1.4 m. Listeners are instructed not to move theirhead while sounds are played (each one lastingabout 1.5 sec); it has been shown that listeners cansuccessfully maintain such a posture for this type oftask (Noble, Byrne, & Ter-Horst, 1997). Each of 16different sounds is presented six times in a test run,with equal distribution over the eight sources, giv-ing a total of 96 trials (sounds and sources randomlyselected). The task is to identify the source, but notthe sound as such. Score on the test is expressed asthe RMS deviation between actual source azimuthand nominated azimuth; thus the higher the scorethe less accurate the performance. Pre-implant per-formance testing is typically within a few weeksbefore implanting; postimplant testing occurs soonafter implanting, but may be up to 6 mo postim-plant.

RESULTS

Preliminary Analysis: Long-Term VersusShort-Term Implanting

All results are based on the first self-report eventfor any patient. There are one or two cases wherethat initial rating is only 1 mo postimplantation,and these have not been included in the analyses.The impression gained from inspection of individualresults was that more recently implanted patients(within the previous 5 or so yr) were recordingslightly higher handicap scores than those im-planted earlier than that. A preliminary analysis


NOBLE ET AL. / EAR & HEARING, VOL. 29, NO. 1, 0–0 3

was undertaken, therefore, to compare those im-planted up to 100 mo before they provided their firstself-rating with those implanted more than 100 mopreviously. The breakdown of implant profile, withthat time division, was such that almost all the moreremote cases were unilateral CI, whereas all threeprofiles were represented in those implanted up to100 mo previously. In this preliminary analysis, tomake the comparison between long-term and morerecent implanting less confounded, it is limited tounilateral CI cases only. Table 2 shows the HHIEand HHQ scores as a function of time since implant-ing. The scores are shown for a set of subscales thatwere generated in the course of the present investi-gation, as the result of factor analyses, described inthe next section.

Univariate analysis of variances (ANOVAs) wereperformed to compare the two groups based onpostimplantation time (�100 mo; �100 mo). Thedependent variables, explained in detail in the nextsection, are three subscales of the HHIE: EmotionalDistress (HHIE), Difficulty in Hearing, and SocialRestriction (HHIE), and two subscales of the HHQ,Emotional Distress (HHQ) and Social Restriction(HHQ). As may be seen, the longer the time intervalsince implantation, the lower the handicap rating,and in the case of the Emotional Distress (HHIE)subscale, the difference is statistically significant(p � 0.05). In scale terms, the differences are small.Nonetheless, for the comparative analyses, we usedata only from those who completed the question-naires within 100 mo after receiving their CIs.

Factor Analyses

Separate principal component factor analyses ofthe items comprising the HHIE and the HHQ wereundertaken using data of all implant patients (183in the case of the HHIE, 181 for the HHQ). Thetechnical details and factor loadings of the itemscomprising the resulting subscales are shown in theAppendix, as is the wording of the items themselves.

Only 16 items of the HHIE were resolvable intothe three unique factors derived from that scale. Thefirst factor, labeled Emotional Distress (HHIE), isreadily interpretable, consisting of seven items ask-ing about specific feelings (embarrassment, irritabil-ity, frustration, feeling “dumb,” nervous, feeling up-set, depressed). These are all from the “emotional”subscale of the original questionnaire. The secondfactor, also comprising seven items, is somewhatless clearly interpretable. We have labeled it Diffi-culty in Hearing because a consistent theme, repre-sented in six of the seven items, is to do withproblems of audibility in various settings (telephone,whispered speech, at a restaurant, at a party, listen-ing to TV/radio). The wording of the telephone item,and one of the TV items, is in terms of limitation; weassume the reason the items remain in the Difficultysubscale is because the respondents are stronglyconscious of problems with these types of signals, forwhich there is, for example, no or often limitedvisual support. The remaining Difficulty item (limi-tation on personal/social life) less obviously has to dowith the Difficulty theme, and could be seen asbelonging, or also belonging, to the Social Restric-tion (HHIE) factor. Perhaps inclusion of the word“personal” in the item’s wording is what has tippedit into the Difficulty arena.

The third HHIE factor (two items) we have la-beled Social Restriction (HHIE). It has to do moreobviously with limitations on engagement in valuedsocial activity, such as attending religious services,visiting friends and neighbors. There may be anargument that attending religious service is a per-sonal matter, but it is legitimate also to recognizethe significant social element that attaches to thisactivity. Whereas the average scores on the Emo-tional Distress(HHIE) and Social Restriction (HHIE)subscales are just in excess of 1.0 (Table 2) out of amaximum of 4.0, the average for the Difficulty inHearing subscale is nearly 1.5 scale points higher.There is a significant difference among the HHIEsubscales [F(2, 210) � 21.82, p � 0.001]. Post hoccomparison (Games-Howell, equal variance not as-sumed) shows that the scores on the Difficulty inHearing subscale are significantly higher than thescores on the Emotional Distress (HHIE) (p � 0.001)and Social Restriction (HHIE) subscales (p � 0.001).We see the Emotional Distress (HHIE) and SocialRestriction (HHIE) subscales generated by thepresent analysis as located in the WHO (1980)handicap domain, and the Difficulty in Hearingsubscale as in the disability domain.

All 12 items of the HHQ were resolved into twounique factors. We call the first factor EmotionalDistress (HHQ), the second Social Restriction(HHQ). As can be seen from the item wordings in the

TABLE 2. Comparison between two groups based on postim-plantation time on the three subscales of HHIE and two sub-scales of the HHQ for CI implant group (SDs in brackets)

Month �100(n � 71)

Month �100(n � 35)

HHIE subscalesEmotional Distress (HHIE) 1.30* (1.1) 0.90† (0.7)Difficulty in Hearing 2.47* (1.1) 2.25* (1.0)Social Restriction (HHIE) 1.21* (1.5) 0.86* (1.1)

HHQ subscalesEmotional Distress (HHQ) 2.49* (0.9) 2.26* (0.7)Social Restriction (HHQ) 2.87* (0.9) 2.62* (0.8)

Means in the same row with different superscripts are significantly different at p � 0.05.



T2AQ:A

Appendix, the theme of the first subscale is similarto that of the Emotional Distress (HHIE) subscale ofthe HHIE, and the theme of the second addressessimilar issues to the Social Restriction (HHIE) sub-scale.

Comparative Analyses

In the light of the comparison of longer versusshorter time since implantation, the comparisonacross the three profiles has been confined to thoseimplanted up to 100 mo before first handicap rating.Tables 3 and 4 show the results of that comparisonfor the HHIE and HHQ subscales, respectively. Theorder of self-rated handicap, from least to most, forall subscales, is consistently from the CI � CI to CIto CI � HA groups. With regard to the HHIEsubscales, there is a significant group effect [F(2,143) � 6.14, p � 0.003]. Post hoc comparison(Tukey’s HSD) shows a significantly higher ratingfor Emotional Distress (HHIE) in the CI � HA groupcompared with the CI � CI group (p � 0.002); the CIgroup does not differ significantly from either of theother two. On the Difficulty in Hearing subscale, theCI � CI group has significantly lower rating thanboth the CI (p � 0.02) and CI � HA (p � 0.001)groups, which in turn do not differ significantly fromeach other. On the Social Restriction (HHIE) sub-scale, the CI � CI group has significantly lowerrating than the CI (p � 0.009) and CI � HA (p �0.006) groups, which in turn do not differ signifi-cantly from each other.

This pattern is repeated in responses on the HHQ[F(2, 141) � 6.31, p � 0.002]. There is no equivalent

to the Difficulty in Hearing subscale of the HHIE,because the HHQ was designed so as only to addresshandicap as defined in WHO (1980). The EmotionalDistress (HHQ) subscale shows significantly greaterhandicap rating in the CI � HA group comparedwith the CI � CI group (p � 0.002); the CI groupdoes not differ significantly from either of the othertwo. The Social Restriction (HHQ) subscale showsthe same pattern as the Social Restriction (HHIE)subscale; significantly lower handicap in the CI � CIgroup compared with the CI group (p � 0.02) and theCI � HA group (p � 0.001), and no significantdifference between the latter pair.

Pre-Post Self-Report and Performance DataAcross Profiles

To analyze the foregoing outcome in more detail, wehave turned to other components of the data, in par-ticular looking at pre-implant versus postimplant self-ratings and performance across the three profiles.Because this is essentially a retrospective study,pre-post data are more limited than the postimplan-tation data reported to this point; furthermore,while the point in time at which performance testswere undertaken coincides with self-report comple-tion in some cases, in others there may be a gap ofseveral months between these events.

There were 14 CI cases, 13 CI � CI cases, and 16CI � HA cases with pre- as well as postimplantHHIE scores, all �100 mo postimplantation. Thereare no differences in the pre-implant scores acrossthe three profiles, whereas the postimplant patternfor these more limited samples is similar to thatreported above. There were 14, 19, and 16 cases,respectively, in the CI, CI � CI, and CI � HA groupswith pre- and postimplant HHQ ratings. There wereno differences across profiles in pre- or postimplantscores, although the trend was consistent with thatreported above, namely, that the greatest reductionin rated handicap is observed in the CI � CI group,and there are lesser, similar reductions in the CIand CI � HA groups. There is no indication in thesedata that pre-implant levels are higher in one grouprelative to another, hence, explanation for what isobserved postimplant in the CI � HA group is notbecause they may have been showing reduction froma higher initial disability/handicap status.

There were 30 CI, 13 CI � CI, and 17 CI � HAcases for which we have pre- and postimplant speechtest data. The average percent correct scores on theCNC test are shown in Table 5. A one-way ANOVAfor pre-implant scores shows a significant groupeffect [F(2, 57) � 12.70, p � 0.001], and post hoctesting confirms that the CI � HA group has signif-icantly higher pre-implant scores than the CI (p �

TABLE 3. Comparison of mean handicap ratings (SDs in brack-ets) across three implant groups on the three subscales of theHHIE

HHIE subscalesCI

(n � 71)CI � CI(n � 35)

CI � HA(n � 40)

Emotional Distress (HHIE) 1.31*† (1.1) 0.85* (0.9) 1.67† (1.0)Difficulty in Hearing 2.47* (1.1) 1.84† (1.2) 2.84* (0.9)Social Restriction (HHIE) 1.21* (1.5) 0.37† (0.9) 1.35* (1.4)

Means in the same row with different superscripts are significantly different at p � 0.05using Tukey’s HSD. Thus, for example, the CI � CI group is significantly different from theCI � HA group on all three subscales, whereas the CI group is not significantly differentfrom the CI � CI or CI � HA groups on the Emotional Distress (HHIE) subscale.

TABLE 4. Comparison across three implant groups on the twosubscales of the HHQ

HHQ subscalesCI

(n � 70)CI � CI(n � 35)

CI � HA(n � 39)

Emotional Distress 2.49*† (0.9) 2.12* (0.8) 2.82† (0.8)Social Restriction 2.87* (0.9) 2.38† (0.8) 3.15* (0.8)

Means in the same row with different superscripts are significantly different at p � 0.05using Tukey’s HSD.



T3–4

T5

0.001) and CI � CI (p � 0.001) groups. One-wayANOVA shows no significant differences in postim-plant speech scores. A 2 � 3 repeated measuresANOVA on speech test pre- and postimplantationscores across the three implant profiles shows asignificant main effect for pre- and postimplantationscores [F(1,114) � 249.61, p � 0.001], and an inter-action on the borderline of significance [F(2,114) �3.52, p � 0.03]; a criterion of 0.01 is recommendedgiven a violation of the assumption of homogeneityof variance). The interaction arises because, startingfrom a higher pre-implant base, the improvement inCI � HA speech performance is less than in theother two groups.

The pre-implant hearing threshold levels (aver-age over 0.5 and 2 KHz) of the CI and CI � CI groupsare 103 and 101 dB, respectively, with near identicallevels in left and right ears. The pre-implant thresh-olds of the CI � HA group are 98 dB in the im-planted ear and 89 dB in the nonimplanted (acous-tically aided) ear. The lesser, though still very large,loss in the nonimplanted ear can explain the slightlybetter pre-implant speech performance in the CI �HA group.

The localization test is normally administered topatients either fit with two implants (CI � CI) orwho retain a HA (CI � HA). The numbers of peopleengaged in such testing in this report were verylimited. There are 11 CI � CI cases and 3 CI � HAcases for whom we have pre- and postimplant local-ization data. Separate one-way ANOVAs showedsignificant reduction in RMS error for the CI � CIgroup [from 38° to 19°, F(1,20) � 13.48, p � 0.002],and a nonsignificant increase in RMS error in theCI � HA group (from 24° to 34°). A 2 � 2 repeatedmeasures ANOVA on localization test pre- andpostimplantation scores across two implant groups(CI � CI and CI � HA) showed a significant inter-action, F(1, 27) � 6.39, p � 0.02.

DISCUSSION

The primary purpose of this study was to compareoutcomes from three CI profiles, specifically in thedomain of handicap experience, where the concept of

handicap is understood to refer to nonauditory con-sequences of hearing disability. This is a criticalarea of inquiry in that it goes beyond performanceon standard clinical tests, focusing instead on thequality of life of the person living with severe-profound loss of hearing. Measured performancedata have also been referred to for purposes ofinterpreting certain outcomes using the handicapmeasures.

As preparation for the comparative analysis, thehandicap measures were factor analyzed and it wasfound that the Hearing Handicap Inventory for theElderly (HHIE) was optimally resolvable into threefactors, which were labeled Emotional Distress(HHIE), Difficulty in Hearing, and Social Restric-tion (HHIE). To our knowledge, this is the first suchanalysis of the HHIE, and the present result isconsistent with the argument in Noble (1998) thatthis scale assesses in the disability domain, as wellas in the domain of handicap. The fact that averagescores on the Difficulty in Hearing subscale weresignificantly higher than on the other two subscalesalso supports this point. People are responding tothe items identified as uniquely constituting theDifficulty subscale differently from how they re-spond to the items in the other subscales. Thisoutcome also supports the distinctiveness of theconcepts and domains of disability and handicap asdefined by WHO (1980).

The Hearing Handicap Questionnaire (HHQ) re-solved into two factors, labeled Emotional Distress(HHQ) and Social Restriction (HHQ), with bothsubscales yielding similar scores, and similar to theequivalent subscales of the HHIE. That result indi-cates that the HHQ is assessing in similar domainsas the two handicap-oriented subscales of the HHIE.The outcome is different from the unifactorial struc-ture reported by Gatehouse and Noble (2004) and, ifreliable, suggests that the scale performs differentlyin populations with severe-profound hearing loss(typical CI patients) than in populations with mild-moderate loss (typical hearing clinic outpatients).The factor analysis results overall indicate also thevalidity of construing handicap because of hearingimpairment as at least including emotional distressalong with social limitation.

The outcomes with respect to CI versus CI � CIshow that bilateral CI fitting is associated withlower disability and less social restriction than asingle CI, but there is no significant difference in thepostimplant level of emotional distress between thetwo groups. This suggests that the change in qualityof life felt at a personal level, occasioned by a singleimplant, is almost as powerful as being fit bilater-ally. We note that the average changes in speechtest scores for CI and CI � CI groups are almost

TABLE 5. Pre- and Postimplant scores on the CNC speechperformance measure in the three implant groups (values are %correct at whole word level)

CNCCI

(n � 30)CI � CI(n � 13)

CI � HA(n � 17)

Preimplant 3.10* (4.0) 7.90* (6.6) 16.45† (14.5)Postimplant 57.47* (21.7) 62.18* (20.2) 53.72* (19.2)

Means in the same row with different superscripts are significantly different at p � 0.05using Tukey’s HSD.



identical. But improvement in disability and lesssocial restriction among CI � CI patients, comparedwith unilateral CI, indicates gain in quality of life inother areas deriving from that second implant; forexample, because of improved spatial hearing abil-ity. A recent report (Summerfield, et al., 2006) onthe incremental benefit of a second CI concludesthat there is further reduction in disability, andsome selective improvement in quality of life, be-cause of a second implant, a finding not dissimilar towhat we report here. The disability measure used inthat study was the Speech, Spatial and Qualities ofHearing Scale (Gatehouse & Noble, 2004).

As to comparative outcomes across the threeprofiles, the initial assumption was that the CI �HA profile would fall somewhere between CI andCI � CI. Although limited, there is some perfor-mance test evidence in the literature to support thatexpectation (Ching, Incerti, & Hill, 2004; Dunn,Tyler, & Witt, 2005). To our knowledge, there is noprevious comparison of the profiles analyzed here,based on self-report measures. Probably most strik-ing is the greater level of residual emotional distressreported by users of CI � HA, an outcome observ-able on both handicap scales. We note that there aresigns of lesser improvement in measured speechhearing, and poorer localization performancepostimplant in that group. People comprising theCI � HA group are those who previously had bilat-eral HAs, hence they are transferring from onebilateral hearing device profile to a different bilat-eral profile; furthermore, they are transferring froma fully acoustic profile that, though nonoptimal interms of deliverable signal level, is at least the samein structure at each ear. There may be greaterpotential for binaural interference when the signalin one ear is acoustic, and the signal in the other earis electrical. There is an evident need to examine theconsequences of the CI � HA profile in more detail.

We note, finally, that length of time since im-planting may have an influence on experiencedhandicap, but the effect is small. It may reflectgreater initial handicap in more long-term cases, orit may represent a gradual improvement in use ofthe implant over the very long term.

In conclusion, our results for unilateral CI are inline with some previous reports in the literature.The present results for bilateral compared withunilateral CI in the handicap domain indicate thereis significant incremental benefit from the secondimplant in the area of social restriction, but not inthe area of emotional distress (although the trend isin the direction of greater incremental benefit). Thearea we have labeled Difficulty in Hearing, derivedfrom our factor analysis of the HHIE, also showssignificantly more advantage for bilateral implanta-

tion. That area we identify as oriented more in theWHO (1980) disability domain, specifically, disabil-ity hearing speech, than in the domain of handicap.We reason that significantly greater reduction inspeech hearing disability would then be associatedwith significantly less social restriction.

ACKNOWLEDGMENTS

This research was supported in part by research grant P50DC000242-21 from the National Institutes on Deafness andOther Communication Disorders, National Institutes of Health;grant RR00059 from the General Clinical Research CentersProgram, Division of Research Resources, National Institutes ofHealth; the Lions Clubs International Foundation; and the IowaLions Foundation. We thank Haihong Ji and Shelley Witt fortheir help in data retrieval and analysis.

The background to the present report was established whileWilliam Noble held a Levitt Visiting Professorship at the Univer-sity of Iowa. Thanks to the Levitt family for the generosity of theirendowment to the University.

Address for correspondence: Miss Navjot Bhullar, School of Psy-chology, University of New England, Australia. E-mail: ●●●.

Received December 20, 2006; accepted August 2, 2007.

APPENDIX

Factor Analysis: HHIE

Exploratory factor analysis was performed us-ing principal component analysis with direct ob-

TABLE A1. Oblimin rotated communalities (h2), factor loadings(Cronbach’s � in brackets) for three-component structure ofHHIE scores (N � 183)

Scale h2

Components

1Emotional Distress(HHIE) (� � 0.88)

2Difficulty in

Hearing(� � 0.86)

3Social Restriction(HHIE) (� � 0.66)*

Item 7 0.65 0.84Item 2 0.68 0.73Item 22 0.62 0.71Item 12 0.58 0.71Item 17 0.57 0.62Item 4 0.54 0.56Item 5 0.56 0.40Item 15 0.55 0.79Item 21 0.64 0.74Item 1 0.53 0.73Item 20 0.62 0.68Item 8 0.45 0.67Item 23 0.46 0.63Item 6 0.60 0.61Item 11 0.60 0.77Item 13 0.76 0.59Mean 1.21 2.38 1.01SD 1.0 1.1 1.3

Following the recommendations of Tabachnick and Fidell (2001), pattern matrix loadings(overlapping variance among factors has been partialled out) are reported. Items 3, 9, 10,14, 16, 18, 19, 24 and 25 are excluded because of their cross-loadings of �0.30.* Cronbach’s � � 0.70 can be attributed to the fact that only 2 items make up this factor.



AQ:1

AQ:2

TA1–4

limin rotation because it was expected the factorswould correlate. Relevant statistical assumptionswere checked as suggested by Tabachnick andFidell (2001). There were no univariate outliers,and six multivariate outliers. Analysis was runtwice, with and without these outliers; no differ-ence was observed, thus outliers were retained inthe analysis.

The maximum number of factors to be extractedwas determined using Cattell’s (1966) scree plot,Kaiser’s (1960) rule (eigenvalues greater than 1),and Velicer’s (1976) minimum average partial test.Together, these indicated one to three factors shouldbe retained. All possible solutions were tried, and athree-factor solution was identified for ease of inter-pretability and the best model possible. The finalversion of the HHIE for present purposes yielded 16items. Only those items were retained that load 0.30or above and did not load on more than one factor.Factor 1 (Emotional Distress (HHIE)), factor 2 (Dif-ficulty in Hearing), and factor 3 (Social Restriction(HHIE)) explained 59% of the total variance. Asummary of the pattern matrix loadings is pre-sented below (item numbers are from the originalpublished inventory), followed by the items orderedinto three subscales (our numbering).

Factor Analysis: HHQ

The same steps were taken to check relevantstatistical assumptions. Univariate and multivari-ate outliers were found. No difference was observedwhen analysis was run with and without these, sooutliers were retained. The criteria used in theprevious analysis suggested one to two factors

should be retained. A two-factor solution was re-tained for ease of interpretability and the best modelpossible.

The same criteria as before were applied in theprincipal component analysis. No item cross-loaded.Factor 1 (Emotional Distress) and factor 2 (SocialRestriction) explained 78% of the total variance. Asummary of the pattern matrix loadings from theprincipal component analysis is given below (itemnumbers from original), followed by items constitut-ing two subscales (our numbering).

TABLE A2. HHIE subscales

HHIE item Anchors

Subscale 1: Emotional Distress (HHIE)1 Does a hearing problem cause you to feel embarrassed when meeting new people? Yes–Sometimes–No2 Does a hearing problem make you irritable? Yes–Sometimes–No3 Does a hearing problem cause you to feel frustrated when talking to members of your family? Yes–Sometimes–No4 Does a hearing problem cause you to feel “stupid” and “dumb”? Yes–Sometimes–No5 Does a hearing problem cause you to be nervous? Yes–Sometimes–No6 Does any problem or difficulty with your hearing upset you at all? Yes–Sometimes–No7 Does a hearing problem cause you to feel depressed? Yes–Sometimes–No

Subscale 2: Difficulty in Hearing8 Does a hearing problem cause you difficulty when listening to TV or radio? Yes–Sometimes–No9 Does a hearing problem cause you difficulty when in a restaurant with relatives or friends? Yes–Sometimes–No

10 Does a hearing problem cause you to use the phone less often than you would like? Yes–Sometimes–No11 Do you feel that any difficulty with your hearing limits or hampers your personal or social life? Yes–Sometimes–No12 Do you have difficulty hearing when someone speaks in a whisper? Yes–Sometimes–No13 Does a hearing problem cause you to listen to TV or radio less often than you would like? Yes–Sometimes–No14 Does a hearing problem cause you difficulty when attending a party? Yes–Sometimes–No

Subscale 3: Social Restriction (HHIE)15 Does a hearing problem cause you to attend religious services less often than you would like? Yes–Sometimes–No16 Does a hearing problem cause you to visit friends, relatives, or neighbors less often than you would like? Yes–Sometimes–No

TABLE A3. Oblimin rotated communalities (h2), factor loadingsand Cronbach’s � for two-component structure of the HHQscores (N � 181)

Scale items h2

Components

1Emotional Distress

(� � 0.95)

2Social Restriction

(� � 0.93)

HHQ6 0.82 0.97HHQ5 0.84 0.94HHQ4 0.76 0.87HHQ7 0.78 0.85HHQ2 0.80 0.85HHQ3 0.74 0.84HHQ12 0.73 0.74HHQ1 0.80 0.92HHQ8 0.71 0.91HHQ9 0.81 0.89HHQ11 0.81 0.73HHQ10 0.79 0.72Mean 2.44 2.78SD 0.8 0.9

Following the recommendations of Tabachnick and Fidell (2001), pattern matrix loadings(overlapping variance among factors has been partialled out) are reported.



REFERENCES

Cattell, R. B. (1966). The scree test for the number of factors.Multivariate Behavioral Research, 1, 245–276.

Ching, T. Y. C., Incerti, P., & Hill, M. (2004). Binaural benefits foradults who use a hearing aid and a cochlear implant in oppositeears. Ear and Hearing, 25(1), 9–21.

Dalton, D. S., Cruickshanks, K. J., Klein, B. E. K., Klein, R.,Wiley, T. L., & Nondahl, D. M. (2003). The impact of hearingloss on quality of life in older adults. The Gerontologist, 43(5),661–668.

Dunn, C. C., Tyler, R. S., & Witt, S. A. (2005). Benefit of wearinga hearing aid on the unimplanted ear in adult users of acochlear implant. Journal of Speech, Language, and HearingResearch, 48(3), 668–680.

Gatehouse, S., & Noble, W. (2004). The Speech, Spatial andQualities of Hearing Scale (SSQ). International Journal ofAudiology, 43(2), 85–99.

Hawthorne, G., & Hogan, A. (2002). Measuring disability-specificpatient benefit in cochlear implant programs: developing ashort form of the Glasgow Health Status Inventory, and theHearing Participation Scale. International Journal of Audiol-ogy, 41, 535–544.

Hawthorne, G., Hogan, A., Giles, E., Stewart, M., Kethel, L.,White, K., et al. (2004). Evaluating the health-related qualityof life effects of cochlear implants: a prospective study of anadult cochlear implant program. International Journal of Au-diology, 43, 183–192.

Humes, L. E., Wilson, D. L., Barlow, N. N., Garner, C. B., & Amos,N. (2002). Longitudinal changes in hearing aid satisfaction andusage in the elderly over a period of one or two years afterhearing aid delivery. Ear and Hearing, 23(5), 428–438.

Kaiser, H. F. (1960). The application of electronic computers tofactor analysis. Educational and Psychological Measurement,20, 141–151.

Newman, C. W., Weinstein, B. E., Jacobson, G. P., & Hug, G. A.(1990). The hearing handicap inventory for adults: Psychomet-ric adequacy and audiometric correlates. Ear and Hearing,11(6), 430–433.

Noble, W. (1998). Self-assessment of hearing and related func-tions. London: Whurr.

Noble, W., Byrne, D., & Ter-Horst, K. (1997). Auditory localiza-tion, detection of spatial separateness, and speech hearing innoise by hearing impaired listeners. Journal of the AcousticalSociety of America, 102(3), 2343–2352.

Summerfield, A. Q., Barton, G. R., Toner, J., McAnallen, C.,Proops, D., Harries, C., et al. (2006). Self-reported benefitsfrom successive bilateral cochlear implantation in postlin-gually deafened adults: randomised controlled trial. Interna-tional Journal of Audiology, 45(Suppl. 1), S99–S107.

Tabachnick, B. B. & Fidell, L. S. (2001). Using multivariate statis-tics. (4th ed.). New York: HarperCollins College Publishers.

Tillman, T. W., & Carhart, R. (1966). An expanded test for speechdiscrimination utilizing CNC monosyllabic words. Northwest-ern University Auditory Test No. 6: Technical Report No.SAM-TR-66–55. USAF School of Aerospace Medicine, BrooksAir Force Base, Texas.

Velicer, W. F. (1976). Determining the number of componentsfrom the matrix of partial correlations. Psychometrika, 41,321–327.

Ventry, I. M., & Weinstein, B. E. (1982). The Hearing HandicapInventory for the Elderly: A new tool. Ear and Hearing, 3(3),128–134.

Vermeire, K., Brokx, J. P., Wuyts, F. L., Cochet, E., Hofkens, A.,De Bodt, M., et al. (2006). Good speech recognition and quality-of-life benefit from cochlear implantation in patients withDFNA9. Otology and Neurotology, 27(1), 44–49.

Vermeire, K., Brokx, J. P., Wuyts, F. L., Cochet, E., Hofkens, A.,& Van de Heyning, P. H. (2005). Quality-of-life benefit fromcochlear implantation in the elderly. Otology and Neurotology,26(2), 188–195.

WHO. (1980). International classification of impairments, dis-abilities, and handicaps. Geneva: World Health Organiza-tion.

WHO. (2001). International classification of functioning, disabil-ity and health. Geneva: World Health Organization.

TABLE A4. HHQ subscales

HHQ item Anchors

Subscale 1: Emotional Distress (HHQ)1 How often does any difficulty with your hearing make you feel self-conscious? Never–Almost always*2 How often does your hearing difficulty make you feel nervous or uncomfortable? Never–Almost always3 How often is your self-confidence affected by your hearing difficulty? Never–Almost always4 How often does any difficulty with your hearing affect the way you feel about yourself? Never–Almost always5 How often do you feel worried or anxious because of your hearing difficulty? Never–Almost always6 As a result of your hearing difficulty, how often do you feel embarrassment when in the

company of other people?Never–Almost always

7 How often do you feel tense and tired because of your hearing difficulty? Never–Almost alwaysSubscale 2: Social Restriction (HHQ)8 How often does your hearing difficulty restrict the things you do? Never–Almost always9 How often are you inconvenienced by your hearing difficulty? Never–Almost always

10 How often do you feel inclined to avoid social situations because of your hearing difficulty? Never–Almost always11 How often does your hearing difficulty restrict your social or personal life? Never–Almost always12 How often do you feel cut-off from things because of your hearing difficulty? Never–Almost always

* Responses are on a five-point scale (never, rarely, sometimes, often, almost always).



JOBNAME: AUTHOR QUERIES PAGE: 1 SESS: 1 OUTPUT: Mon Nov 12 11:58:56 2007/balt5/zau�aud/zau�aud/zau00108/zau3111�08z

1—Please provide footnotes for * and † for table 2-5

2—Please note that footnote text linked to author “William Noble” has been placed in theAcknowledgement section. Kindly check whether this is OK as given.

3—Please provide e-mail address for the corresponding author.

AUTHOR QUERIES

AUTHOR PLEASE ANSWER ALL QUERIES 1

Hearing Handicap Ratings Among Different Profiles of Adult Cochlear Implant Users

Documents