Measuring Hearing Aid Outcomes Using the Satisfaction with ... · PDF fileSurvey;HAUQ =...

J Am Acad Audiol 16:383–402 (2005)

383

*Communication Disability in Ageing Research Centre, School of Health and Rehabilitation Sciences, University of Queensland,Australia

Louise Hickson, Associate Professor, Communication Disability in Ageing Research Centre, School of Health and RehabilitationSciences, University of Queensland, St Lucia Q 4072, Australia; Phone: 61-7-3365-3096; Fax: 61-7-3365-1877; E-mail: [email protected]

Measuring Hearing Aid Outcomes Usingthe Satisfaction with Amplification in DailyLife (SADL) Questionnaire: AustralianData

Margaret Uriarte*Lauren Denzin*Amy Dunstan*Jillian Sellars*Louise Hickson*

Abstract

The aims of this study were to investigate hearing aid satisfaction for a groupof older Australians fitted with government-funded hearing aids using theSatisfaction with Amplification in Daily Life (SADL) questionnaire; to comparethe Australian data gathered with the provisional normative data reported byCox and Alexander (1999); and to investigate the relationship between SADLsatisfaction and several participant variables, hearing aid variables, and otheroutcome measures.The SADL questionnaire and a Client Satisfaction Survey (CSS) were dis-tributed by mail to 1284 adults fitted with government-funded hearing aidsthree to six months previously. 1014 surveys were returned. The mean age ofparticipants was 75.32 years; 54.4% of participants were male, and 54.8%were fitted binaurally. Participants were fitted primarily with digitally program-mable hearing aids of various styles (22.5% BTEs, 34.8% ITEs, 41.8% ITCs,0.9% nonstandard [NS] devices).Overall, participants reported a considerable level of satisfaction with theirdevices. SADL Global and subscale scores were significantly higher for theAustralian sample than the U.S. norms described by Cox and Alexander (1999).

Key Words: Hearing aids, older people, outcomes, satisfaction

Abbreviations: BTE = behind-the-ear hearing aid; CSS = Client SatisfactionSurvey; HAUQ = Hearing Aid Users Questionnaire; ITC = in-the-canal hearingaid; ITE = in-the-ear hearing aid; NS = nonstandard amplification; OHS =Office of Hearing Services; SADL = Satisfaction with Amplification in Daily Life;SHAPIE = Shortened Hearing Aid Performance Inventory for the Elderly

Sumario

El propósito de este estudio fue investigar la satisfacción generada por el auxiliarauditivo en un grupo de australianos mayores, a quienes se les adaptaronauxiliares pagados por el gobierno, utilizando el cuestionario Satisfaccióncon la Amplificación en la Vida Diaria (SADL), comparar los datos australianosrecolectadas de la información normativa reportada por Cox y Alexander(1999), e investigar la relación entre el SADL, la satisfacción y algunasvariables de los participantes, variables de los auxiliares auditivos y otrasmedidas de resultado.El cuestionario SADL y una Encuesta de Satisfacción del Cliente (CSS) fuerondistribuidos por correo a 1284 adultos, adaptados con audífonos pagados por

In the current consumer-driven era ofhealth care, health professionals need tobe able to demonstrate, to both the com-

munity and resource providers, that the serv-ices they provide have a positive impact ontheir clients’ functional status and quality oflife (Beck, 2000; Gagné, 2000; Weinstein,2000). In audiological rehabilitation settings,outcome measures have emerged as an effec-tive method for determining whether or notspecific interventions such as hearing aids areworking to achieve positive results for clients(Cox and Alexander, 1999; Kricos and Lesner,2000). Outcome measures in audiology alsoserve to identify areas of service or treat-ment that could be modified or enhanced tobetter suit client needs; provide clients withobjective information regarding the benefitsof certain interventions and technologies;promote data-driven decision making; eval-uate the performance of new and existinghearing aid technologies; provide hearing aidmanufacturers with quantitative informa-tion regarding clients’ hearing needs andconcerns with hearing aids; and track andcompare provider performance over time(Beck, 2000; Humes et al, 2001).

Although objective outcome measures ofhearing aid benefit, such as insertion gainand aided versus unaided speech recognition

scores, are essential for documenting theimprovement in a client’s hearing abilityresulting from amplification (i.e., verification),only the clients can determine how wellhearing aids have solved their hearing-relatedproblems in everyday life and the extent towhich their criteria for quality of service,convenience, and value-for-money have beenmet (i.e., validation) (Huch, 1999; Bentler andKramer, 2000; Cox and Alexander, 2000; Kricosand Lesner, 2000). Consequently, numeroussubjective inventories or self-report measureshave been developed for the purpose ofquantifying the client’s perspective on thesuccess of their hearing aids (Huch, 1999;Bentler and Kramer, 2000). Many of theseself-report measures focus on the benefit theclient has received from intervention (Coxand Alexander, 1999; Humes et al, 2001).Benefit typically refers to the improvement ina client’s ability to communicate with othersand/or carry out normal everyday activities aswell as the reduction in the psychosocialimpact caused by hearing loss followingamplification and/or aural rehabilitation (Coxand Alexander, 1999; Weinstein, 2000).

The reliance on benefit as an effectiveindex of service quality and value isunderstandable considering that prospectivehearing aid wearers regularly report that

el gobierno en los previos tres a seis meses. Se recibieron 1014 encuestas.La edad media de los participantes fue de 75.32 años; 54.4% de los participanteseran varones, y 54.8% fueron amplificados binauralmente. Los participantesrecibieron primariamente auxiliares auditivas digitalmente programables devarios estilos (22.55 retroauriculares, 34.8% intracanales, 41.8% canales,0.9% dispositivos (NS) no convencionales).Globalmente, los participantes reportaron un nivel considerable de satisfaccióncon sus dispositivos. Los puntajes globales y de sub-escala de SADL fueronsignificativamente mayores para la muestra australiana que para las normasU.S. descritas por Cox y Alexander (1999).

Palabras Clave: Auxiliares auditivos, gente mayor, resultados, satisfacción

Abreviaturas: BTE = auxiliar auditivo retro-auricular; CSS = Cuestionario deSatisfacción del Cliente; HAUQ = Cuestionario de Uso de Auxiliar Auditivo; ITC= auxiliar auditivo dentro del canal; ITE = auxiliar auditivo intra-auricular; NS= amplificación no estándar; OHS = Oficina de Servicios Auditivos; SADL =Satisfacción con la Amplificación en la Vida Diaria; SHAPIE = InventarioAcortado de Desempeño con el Auxiliar Auditivo para Ancianos

384

Journal of the American Academy of Audiology/Volume 16, Number 6, 2005

Measuring Outcomes Using the SADL: Australian Data/Uriarte et al

385

improved everyday communication is afundamental goal they wish to achievethrough amplification (Dillon et al, 1997;Kochkin, 1992). However, even though a clientmay demonstrate quantifiable benefit fromhearing aids, such as improved ability tohear others speak, this does not ensure thatthe client is fully satisfied with the overallservice received, nor does it guarantee thatthe client considers the hearing aid purchaseto be worthwhile (Cox and Alexander, 1995).Based on an extensive survey of hearing aidwearers, Kochkin (1992) investigated thosefactors considered by consumers to be criticalfor hearing aid satisfaction. It was revealedthat improved hearing ability, suitable soundquality, instrument reliability, usefulness inmultiple listening environments,postpurchase service, and fit/comfort werethe most important factors influencingconsumer satisfaction. It can thus be arguedthat quantification of hearing aid outcomesfrom the client’s perspective might best beaccomplished using a measure that is morecomprehensive than benefit (Cox andAlexander, 1999; Weinstein, 2000). The term“satisfaction” has been used to describe aglobal outcome variable that encompassesthe full spectrum of issues that are importantto the client (Cox and Alexander, 1999).

Satisfaction is a subjective phenomenonand can be defined as the client’s reaction tosalient aspects of the structure, processes,and outcomes of the service received (Kricosand Lesner, 2000; Weinstein, 2000).Quantifying satisfaction is important inrehabilitation in that it not only determineshow the client has responded personally to theintervention but also sheds light onimprovements that may be needed in areassuch as personal interaction between theclinician and client, accessibility andavailability of services, continuity of services,client convenience, physical settings, andfinancial considerations (Cox and Alexander,1999; Kricos and Lesner, 2000). Additionalclient-satisfaction studies conducted byKochkin (2000) showed that a client’s overallsatisfaction with hearing aids was stronglylinked with the likelihood of repurchasing aparticular brand of hearing aid,recommending hearing aids to friends andrelatives, recommending a particular clinicianor dispenser to others, and general quality-of-life ratings. Thus, high levels of consumersatisfaction are not only beneficial for the

client but also for indirectly promoting thepublic perception of hearing aids and hearingaid dispensers (Kochkin, 2000).

Many of the available measures of clientsatisfaction rely on a single item questionsuch as “Rate your overall satisfaction withyour hearing aid/s” (e.g., Humes et al, 1996;Dillon et al, 1999). However, Cox andAlexander (1999) report that a single globalsatisfaction score is of limited clinical value.A single score cannot be used by the clinicianto specifically determine why a client mightbe dissatisfied with their hearing aids andoffers no insight into areas of rehabilitationthat need to be modified or improved.

The Satisfaction with Amplification inDaily Life (SADL), a self-report measuredeveloped by Cox and Alexander (1999), fulfillsthe need for a clinically viable tool, whichassesses the multidimensional nature ofsatisfaction.The scale consists of 15 questionsrelated to aspects of hearing aid use, andprovides a global score indicating overallsatisfaction, as well as four subscale scoresprofiling satisfaction in the areas of “PositiveEffect” (improved psychoacoustic andpsychological functioning), “Service and Cost”(value for money, confidence in provider),“Negative Features” (undesirable effects ofhearing aids including background noise andfeedback), and “Personal Image” (appearanceand stigma) (Cox and Alexander, 1999; Coxand Alexander, 2001). Respondents arerequired to indicate their level of satisfactionon a scale of one (not at all) to seven(tremendously) (Cox and Alexander, 1999).

SADL items were determined based onkey satisfaction issues raised duringinterviews with hearing aid users (Cox andAlexander, 1999). Not surprisingly, goodconstruct validity for the SADL has beenindicated (Cox and Alexander, 2001; Hosford-Dunn and Halpern, 2000). Furthermore, test-retest reliability was demonstrated by theoriginal authors, who found a high Globalscore test-retest correlation coefficient (r =0.81) (Cox and Alexander, 1999). Hosford-Dunn and Halpern (2001) suggest that withits sound construct and psychometricproperties, the SADL may serve as a goldstandard for measuring satisfaction.

At present, little Australian data for theSADL has been reported in the literature. In2001, McLeod et al investigated thedifferences in SADL scores whenadministered at two weeks (N = 45) and


12–24 months postfitting (N = 75) for twogroups of Australian adults fitted withgovernment funded hearing aids. The aimwas to determine whether early responseswere representative of long-term satisfaction.The results of the study indicated thatapplication of the SADL two weeks afterfitting may not be appropriate, as Globalscores and all subscale scores, with theexception of the Personal Image subscale,were significantly lower at one to two yearspostfitting (McLeod et al, 2001). In their study,McLeod et al (2001) did not attempt toestablish Australian SADL norms.These areimportant, however, as the SADL wasoriginally designed for use in conjunctionwith normative data. The idea was thatcomparison of an individual’s scores withnorms allows the clinician to identify thoseclients reporting lower than averagesatisfaction levels (Cox and Alexander, 1999).This can then help the clinician to determinewhat areas need to be addressed in order toimprove the client’s overall satisfaction withthe hearing aid experience (Hosford-Dunnand Halpern, 2001). Interim norms wereprovided by Cox and Alexander (1999) forAmerican private-pay and third-party paypatients; however, these may not be applicablefor the Australian population. Thus the firstaim of this study was to provide SADL datafor a group of older Australians provided withgovernment-funded hearing aids.

In the study by McLeod et al (2001), theAustralian SADL results were compared toCox and Alexander’s (1999) provisional norms.They reported that the 12–24 monthpostfitting group produced similar SADLscores to the original American cohort withthe exception that satisfaction in the NegativeFeatures subscale was higher in theAustralian group. However, it is difficult todraw definitive conclusions from this studyas the Australian and American groups werenot compared statistically (McLeod et al,2001). Therefore, the second aim of thepresent study was to compare our Australiandata with the provisional norms described byCox and Alexander (1999).

Hosford-Dunn and Halpern (2001)identified some variables that have smalleffects on SADL results, namely hours ofhearing aid use per day, years of prior hearingaid experience, pure-tone average, self-perceived hearing difficulties, style of hearingaid, cost of hearing aid, and type of processingused in the aid. In addition to considering the

influence of audiological variables on a client’sperformance on the SADL, it is also importantto be aware of those nonaudiological variables,such as the person’s age, gender, personalitytype, race/ethnicity, self-efficacy, socialsupport, attitudes toward hearing aid use,socioeconomic status, and general health,which may potentially affect a client’s levelof satisfaction with their hearing aid/s. Todate, there have been only two studies thathave investigated such relationships usingthe SADL. First, a study by Hosford-Dunnand Halpern (2001) examined therelationship between various client variablesand overall performance on the SADL, andfound that older hearing aid wearers tendedto report less satisfaction than youngerclients. Second, a study by Cox and Alexander(2000) revealed that those clients withrelatively high prefitting expectationsgenerally demonstrated higher postfittingsatisfaction. Therefore, a further aim of thisstudy was to investigate the effect ofparticipant and hearing aid variables on theSADL results. The specific variablesexamined were severity of hearingimpairment (better ear three frequencyaverage); self-perceived hearing difficulty;age; gender; previous hearing aid experience;type of fitting (monaural/binaural); style ofhearing aid; and type of hearing aid funding(fully or partially government subsidized).

In order to observe whether the singleitem satisfaction measures correlated to theGlobal scores for the SADL, this studyinvolved a comparison between SADL GlobalScores and scores for a single item satisfactionmeasure (Cox and Alexander, 2001). Resultsof this comparison indicated a modestly strongrelationship between the two measures (Coxand Alexander, 2001). Presently this is theextent of such comparisons between the SADLand other outcome measures in the literature.Hence, the final aim of this study was toexamine the relationship between otheroutcome measures and SADL scores. Thespecific outcome measures examined includeself-reported hearing aid use, satisfactionwith hearing aid/s (single item), satisfactionwith practitioner (single item), hearing aidbenefit, and hearing aid difficulties.

386


387

METHOD

Participants

Participants for the present study weresampled from the population of new and returnclients, involved in Australia’s CommonwealthHearing Services Program (which requiresclients to be on a pension for eligibility), whohad been fitted with a government-fundedhearing aid three to six months previously atvarious clinics throughout Australia.Participants were recruited by the Office ofHearing Services (OHS) and asked to completethe Client Satisfaction Survey (CSS) inaddition to a SADL questionnaire. Allparticipants were fitted with hearing aidsaccording to the OHS Clinical Standards forService Providers as well as the protocols andprocedures employed by each clinic (Office ofHearing Services, 2002).

The hearing aid fitting process typicallyinvolves three appointments: (1) initialassessment (i.e., administering audiologicalassessments, providing feedback andcounseling to the client and family regardingassessment findings, discussing rehabilitationoptions, determining the client’s listeningdifficulties and everyday communicationneeds, assisting clients in choosing a style ofhearing aid that suits their type and severityof hearing loss, their physical and visualcapabilities, their daily activities, and theirpersonal preferences, explaining to clientsand families how the Commonwealth HearingServices Program operates, and taking earmould impressions); (2) fitting appointment(i.e., verifying that appropriate amplificationhas been achieved through the use of couplermeasurement, real ear insertion gain, and/orreal ear aided response measurements,educating the client in how to manage and carefor their hearing aids, and providing the clientwith an appropriate listening program to meetthe goals established in the prior appointment);

and (3) follow-up appointment (i.e.,determining any further complications theclient may be having with the fit, management,or acoustics of his or her hearing aid/s, carryingout speech perception assessments, andcompleting outcome measures).

An equal proportionate sampling methodwas used to ensure an even representation ofOHS clients throughout Australia. Eightpercent of the target population was randomlysampled from each Australian state andterritory. A total of 1284 Client SatisfactionSurveys were mailed out to the sample. Onethousand and fourteen (1014) surveys werereturned, equating to a response rate of 79%.No exclusion criteria were applied; however,some participants were excluded fromstatistical analysis, as surveys were notalways completed in full. Participantdemographics were supplied by the OHS.Participants ranged from 29 to 104 years ofage (M = 75.32, SD = 9.73). The majority ofparticipants were male (54.4%), primarilyfitted with digitally programmable aids andfitted binaurally (54.8%). Specific hearingaid details were not provided to theresearchers. The mean better ear threefrequency average was 40.21 dB HL (SD =14.46, range = 0 to 110 dB HL) (OHS recordthree frequency average not four frequencyaverage). A large proportion (89.7%, N = 910)of participants were using hearing aids fullysubsidized by the government, while theremaining 10.3% (N = 104) had been fittedwith top-up aids (which are devices that arefunded partially by the government with anadditional contribution by the hearing aidpurchaser). At the time of the survey, 48.1%of respondents had six weeks to 11 monthsof lifetime hearing aid experience (includingall old and current hearing aids); 34.8% ofparticipants had one to ten years hearingaid experience; 15.7% indicated they hadover ten years of hearing aid experience; and1.3% of participants had less than six weeksexperience with hearing aids.Table 1 provides

Table 1. Summary of Styles of Hearing Aid Fitted

BTE ITE ITC NS

Left Ear 22.2% 34.3% 42.9% 0.6%N = 177 N = 274 N = 342 N = 5

Right Ear 22.9% 35.3% 40.6% 1.2%N = 177 N = 273 N = 315 N = 9

Note: N = 1572 ears. BTE = behind-the-ear hearing aids; ITE = in-the-ear hearing aids; ITC = in-the-canal hearing aids; NS =nonstandard amplification (e.g., FM system).


details of the different styles of amplificationdevices fitted. The nature of specificnonstandard (NS) devices (e.g., FM systems)is not known.

Materials

Two questionnaires were mailed toparticipants for the present study, the SADL(Cox and Alexander, 1999) and the OHSClient Satisfaction Survey (CSS) (seeAppendix A), adapted from the Hearing AidUsers Questionnaire (HAUQ) (Dillon et al,1999). This study was conducted using theoriginal published version of the SADL andnot the more recent version with minorrevisions to the wording of two of the questions(available from www.ausp.memphis.edu/harl).In addition, Item 14 relating to hearing aidcost was omitted in the present study as nocost was incurred for the majority of hearingaids. This was in accordance with Cox andAlexander’s suggestion that “subjects whohave not paid for their hearing aids shouldomit the cost item and the Service and Costsubscale score should be computed on thetwo remaining items only” (1999, p. 313).

With regard to the SADL, respondentswere asked to rate their responses to each of14 questions using the seven-point descriptorscale. For positively geared questions (e.g.,Item 8: How content are you with theappearance of your hearing aids?), a higherscore was indicative of greater satisfaction(Cox and Alexander, 1999). Four items onthe SADL were negatively geared (i.e., Items2, 4, 7, and 13), such that a higher scoreindicated less satisfaction. For analysis ofthe negatively geared items, the scoring scalewas reversed so that high scores on allquestions could be equated with highsatisfaction. The score for each subscale wascomputed by averaging the responses to therelevant items for that subscale. For thesubscale score to be considered valid andused in the analysis, at least two thirds of thequestions within that subscale had to becompleted. Global scores were calculated asthe average of all SADL items. However,Global scores were only produced for thoserespondents for whom there were four validsubscale scores.

The OHS CSS features ten items relatingto hearing aid use, and hearing aid difficultiesand benefits, which are measures of hearingaid success that are considered independent

of, but related to, satisfaction (Hosford-Dunnand Halpern, 2001). Response formats varyand require the participant to give bothquantitative and qualitative information forsome items (e.g., “If you never wear yourhearing aid/s please tell us why”).Incorporated within the OHS CSS are twosingle-item questions relating to satisfactionwith hearing aids and satisfaction with thepractitioner. Additionally, the followingquestions were used as outcome measures inthe present study: Question 2 assigned ascore to how often hearing aids are used;Question 3 addressed how much help/benefithearing aids provide in six generalizedlistening situations (e.g., family, small groupconversation); and Question 4 identifieddifficulties with six potential managementproblems (e.g., positioning/removal of hearingaids, manipulating controls).

Procedure

OHS routinely collects outcome measuresfor clients under the Hearing ServicesProgram as an initiative toward improvingservice and satisfaction with auralrehabilitation. OHS typically uses mailsurveys to collect outcome data. This methodis supported by Dillon et al (1991b), whosuggests that clients tend to underrate theirlevel of difficulties when questioned on thephone or via interview, and that mail surveysmay elicit more information. Participantswere contacted by OHS via post,approximately three to six months followinghearing aid fitting, and asked to completeand return both the SADL and the OHS CSSforms in a postage paid envelope.A remindernotice was also forwarded to participantswho had not returned their questionnaires,to encourage a higher response rate.Additionally, a toll-free number was providedfor respondents in case of questions regardingthe survey. Thus, participants completed thequestionnaires independently, without inputfrom their audiologists.

Data Analysis

The data were de-identified by the OHSand entered into a database on an SPSSpackage (Version 11.0 for Windows) forstatistical analysis. This SPSS file wassubsequently provided to the researchers. Aspreviously mentioned, negatively

388


389

geared/reversed items on the SADL wereaccounted for. Additionally, the rating scaleson question 3, 4, 7, and 8 on the CSS werereversed when compared to the SADL so thatboth scales went in the same direction (e.g.,higher scores were associated with greatersatisfaction and greater aid use).

A number of tests were used to analyzethe data including descriptive statistics forthe Australian SADL data; single sample t-test statistical comparisons to examinedifferences between the Australian samplemean and the U.S. population mean obtainedfor the global and subscale scores;independent groups t-tests to compare themeans of two independent samples todetermine if those means differedsignificantly (i.e., gender, top-up/fullysubsidized devices, monaurally/binaurallyfit); Pearson’s r to observe any correlationsbetween global and subscale SADL scoresand a number of variables (i.e., age, better earaverage, overall benefit and problemsindicated by the OHS CSS, satisfaction withpractitioner and satisfaction with hearingaid/s); and one way analysis of variance(ANOVA) tests, were used to assess anddetermine the significance of the remaining

hearing aid variables (i.e., style of hearingaid/s and hearing aid use) for SADL scoremeans. Where rank data was present,ANOVAs were fitted with polynomialcontrasts, and Bonferroni post hoc comparisontests were performed where categorical datawas evident.

RESULTS

The N values for each test varied, accordingto how many participants received valid

SADL scores (criteria previously mentioned).

Australian SADL Data

Global Satisfaction Score

A summary of the mean SADL Globaland subscale scores obtained is reported inTable 2. The overall mean Global score was5.27 (SD = 0.81, range = 2.43 to 7.00),indicating a considerable level of participantsatisfaction.The distribution of Global scores,shown in Figure 1, indicates that more than70% of participants were very satisfiedoverall.

Table 2. Australian SADL Data

SADL Score N Mean Standard Deviation Range

Global 813 5.27 0.81 2.43–7.00

Positive Effect 961 4.98 1.21 1.00–7.00

Service and Cost 835 5.70 1.14 1.00–7.00

Negative Features 940 4.74 1.38 1.00–7.00

Personal Image 953 5.86 0.93 1.00–7.00

Figure 1. Distribution of SADL Global score (N = 813).


Positive Effect

The Positive Effect subscale is comprisedof Items 1, 3, 5, 6, 9, and 10 of the SADL. Themean level of participant satisfaction for thePositive Effect subscale was 4.98 (SD = 1.21,range = 1.00 to 7.00), consistent withconsiderable satisfaction. Additionally,participants indicated considerable satisfactionlevels for items relating to improvement inunderstanding familiar conversation (M = 5.09,SD = 1.46); to reducing the number of repetitionsrequested by the hearing aid user (M = 4.49,SD= 1.77); and to the naturalness of the soundderived from their hearing aids (M = 4.74, SD= 1.36).A medium level of satisfaction (M = 4.35,SD = 2.03) was reported for improved self,whileparticipants were greatly satisfied (M = 5.67,SD = 1.49) when indicating whether obtainingtheir hearing aids was in their best interest andwhether obtaining hearing aids was worth thetrouble (M = 5.51, SD = 1.64).

Service and Cost

The Service and Cost subscale is comprisedof Items 12 and 14 of the SADL. Participantsindicated that they were greatly satisfied (M= 5.70, SD = 1.14) with the services theyreceived when obtaining their hearing aids.Specifically, participants were greatly satisfied(M = 6.28, SD = 1.02) with the competency ofthe hearing aid provider and indicated aconsiderable level of satisfaction (M = 5.03,SD = 1.89) with the dependability of theirhearing aids.

Negative Features

Negative Features is comprised of Items2 (reversed item), 7 (reversed item), and 11 of

the SADL. Generally, a considerable level ofsatisfaction (M = 4.74, SD = 1.40) was observedwith regard to undesirable aspects of hearingaid use.Participants were considerably satisfied(M = 5.12, SD = 1.82) with extraneous soundsamplified by the hearing aids and greatlysatisfied (M = 5.66,SD = 1.76) with their abilityto turn hearing aids up loud enough withoutgetting feedback. However, participants wereonly somewhat satisfied (M = 3.27, SD = 2.00)with the benefit provided by their hearing aidsusing standard telephones (i.e., with noamplifier).

Personal Image

The Personal Image subscale is comprisedof Items 4 (reversed item), 8, and 13 (reverseditem) of the SADL. Participants indicated highlevels of satisfaction on this subscale (M =5.86, SD = 0.93). Furthermore, participantswere greatly satisfied (M = 5.85,SD = 1.60) thatpeople did not notice their hearing loss morewhen using their hearing aids.A considerablelevel of satisfaction (M = 5.14, SD = 1.63) wasnoted for aesthetic aspects of their hearingaids, and participants indicated that they weretremendously satisfied (M = 6.61, SD = 1.01)that wearing their hearing aids did not makethem appear less capable.

Comparison of Australian SADL Datawith U.S. Interim Norms

In Figure 2, the mean SADL Global andsubscale scores for the Australian sampleare presented with the corresponding U.S.interim norms described by Cox andAlexander (1999). The U.S. sample in theoriginal 1999 study was comprised of 257

390

Figure 2. Australian sample SADL subscale means and corresponding U.S. interim norms. Note: *p < .05; **p < .001.


391

participants who completed the SADL. Thisincluded a large group of Veteran’s Affairsclients as well as a number of private payingclients. Specific numbers in each group werenot specified.The SADL scores for the presentparticipant group were compared to thenormative Global score and Service and Costsubscale score provided by Cox and Alexander(1999) for their Veterans Affairs participants.These scores were calculated without the“reasonable cost” item as they did not pay fortheir hearing aids. For the remainingsubscales, the present Australian SADL datawas compared to the normative data thatincluded both Veterans Affairs and privatepaying hearing aid users.

A single sample t-test statisticalcomparison revealed a significant differencebetween the mean SADL Global scoreobtained for the Australian group and the U.S.group (Cox and Alexander, 1999) (t [812] =12.97, p < .001). Participants in the Australiangroup were significantly more satisfied (M =5.29) than individuals included in the U.S.group (M = 4.9). A significant difference wasalso observed for all means of both samples

for each subscale SADL score: Positive Effect(t [960] = 2.01, p < .05), with the Australiangroup being significantly more satisfied (M= 4.98) than their U.S. counterparts (M =4.9); Service and Cost (t[834] = 7.65, p < .001), such that theAustralian participants were significantlymore satisfied (M = 5.70) than U.S.participants (M = 5.4); Negative Features (t[939] = 25.35, p < .001), with the Australiangroup significantly more satisfied (M = 4.74)than the U.S. participants (M = 3.6); andPersonal Image (t [952] = 8.68, p < .001),Australian participants reporting greatersatisfaction (M = 5.86) than the U.S. group (M= 5.6).

Variable Effects on Australian SADLResults

Participant Variables

An independent groups t-test was usedto compare the mean SADL Global scoresfor male and female participants. Nosignificant difference was found (t [952] =8.68, p > .05). Furthermore, the correlation(Pearson’s r) between participant age andmean SADL Global scores was not significant(r [813] = 0.001, p > .05). However, a very weakbut significant positive correlation wasobserved between the degree of hearing loss(better ear three frequency average) andSADL Global scores (r [806] = 0.100,p < .01), (seeFigure 3), such that participants with higherbetter ear averages (more hearing loss) weremore likely to be satisfied with amplificationthan those participants with minimal betterear averages (i.e., better hearing).

One way analysis of variance (ANOVA)testing, fitted with polynomial contrasts, wasused to examine the effect of the remainingparticipant variables on mean SADL scores(see Table 3). A significant effect of priorexperience with hearing aids was observed forGlobal satisfaction, and Positive Effect and

Table 3. ANOVA (F) Results for Prior Experience and Degree of Difficulty without Hearing Aids

Global Positive Effect Service and Cost Negative Features Personal Image

Prior experience 8.60* 75.37* 3.27** 0.96 0.68

Degree of difficulty without hearing aid(s) 14.31* 45.43* 6.89* 1.70 0.20

*p < 0.001 **p < 0.05

Figure 3. Correlation between better ear averagehearing threshold and Global satisfaction (N = 813).


Service and Cost subscale scores,such that greaterexperience with hearing aids was associated withhigher satisfaction.Participants were also asked toindicate the degree of hearing difficulty withouthearing aids. A significant effect for this variablewas demonstrated for SADL Global,Positive Effect,and Service and Cost scores, such that the moredifficulty participants reported without their hearingaids,the greater the postfitting satisfaction levels.

Hearing Aid Variables

A number of hearing aid variables wereinvestigated to determine the relationship betweenamplification characteristics and self-reportedsatisfaction.An independent groups t-test revealedno significant difference between mean SADLGlobal satisfaction scores for participants whoreceived top-up or fully subsidized hearing aids (t[811] = 1.026, p > .05).Additionally, there was nosignificant difference between mean Globalsatisfaction for participants who were fittedmonaurally compared to those who were fittedbinaurally (t [811] = 1.305,p> .05).ANOVA revealeda significant effect of hearing aid style (BTE,ITE,ITC) on Global satisfaction (F2,794 = 3.353,p< .05),Personal Image (F2,540= 3.850,p< .05),and PositiveEffect (F2,938 = 13.195, p < .001). However, aBonferroni post hoc multiple comparisons analysisdid not suggest that one style of hearing aid wassignificantly better for overall Global satisfaction(p> .05 for all comparisons).For the Positive Effectsubscale, Bonferroni comparisons indicated thatsatisfaction was significantly greater for BTE andITE aids compared to ITC aids (p < .001). Withregard to the Personal Image subscale,Bonferronicomparisons demonstrated significantly highersatisfaction for ITC aids in comparison with BTEstyle aids (p < .001).There was no significant effectof hearing aid style for the Service and Cost (F2,818= 1.005,p > .05) and Negative Features subscales(F2,919 = 1.708, p > .05). Nonstandard (NS)amplification devices were excluded from theseanalyses due to the small sample size (N= 14),andonly participants who were either monaurally fitor binaurally fit with the same style of aid in bothears were included.

Relationship of Other Outcome Variablesto SADL Scores

Satisfaction with Practitioner

Responses on the CSS indicated that 95% ofparticipants were either satisfied or very satisfied

with the quality of services provided by theirpractitioner.Pearson’s rwas used to determine thepresence of a significant association betweensatisfaction with the service provided by theirpractitioner (as indicated by single item on theCSS) and the SADL Global and subscale scores.Weak, yet significant, correlations were observedbetween levels of satisfaction with the hearing aidpractitioner and SADL Global scores (r [813] =0.35,p< .01),Positive Effect (r [953] = 0.3,p< .01),Service and Cost (r [828] = 0.33,p < .01),NegativeFeatures (r [932] = 0.21,p< .01),and Personal Imagesubscale scores (r[945] = 0.18,-p< .01).This positiverelationship indicated that participants whoreported greater satisfaction with their hearingaid provider also reported greater satisfaction withamplification.

Satisfaction with Hearing Aids

Overall, 89% of respondents were eithersatisfied or very satisfied with the hearing aidsprovided. Satisfaction with hearing aids (asindicated by a single item on the CSS) wassignificantly correlated with satisfaction levelsin the SADL results. Specifically, significantweak correlations were observed betweenhearing aid satisfaction and SADL Global (r[802] = 0.54, p < .01), Positive Effect (r [945] =0.49, p < .01), Service and Cost (r [824] = 0.28,p < .01), Negative Features (r [924] = 0.41, p <.01), and Personal Image scores (r [937] = 0.35,p < .01).This relationship was positive such thatthe more satisfied respondents were with theirhearing aids,the higher the satisfaction reportedin all areas of the SADL.

Hearing Aid Use

Approximately 91% (N = 919) ofparticipants used their hearing aids on a dailybasis, that is, one or more hours per day. Only3% of participants reported using hearing aidsless than one hour per week or not at all (seeFigure 4). One-way ANOVA revealed asignificant effect of hearing aid use (F5,799 =35.00, p < .001) on Global satisfaction suchthat increased hearing aid use was associatedwith greater satisfaction. Furthermore, asignificant linear trend according to hearing aiduse was noted for mean satisfaction in thePersonal Image (F5,937 = 2.40,p < .05),NegativeFeatures (F5,923 = 5.93, p < .001), Service andCost (F5,821 = 6.39,p < .001),and Positive Effect

392


393

subscales (F5,945 = 75.37, p < .001), such thatgreater satisfaction levels within each of thesesubscales was associated with greater hearingaid use. See Figure 5 for an illustration of theobserved trends. Although a major proportionof the variation in satisfaction levels accordingto hearing aid use was explained by this lineartrend, a small but significant deviation wasobserved for Global,Negative Features,Serviceand Cost, and Positive Effect scores (p < .05),and hence,a quadratic function was performed.This revealed a significant quadraticrelationship for the Positive Effect (F1,945 =7.52, p < .05) and Negative Features subscales(F1,923 = 11.38, p = .001). Neither a quadraticfunction (F1,821 = 2.47,p > .05) nor cubic function(F1,821 = 1.93, p < .001) was significant for theService and Cost subscale in accounting forthis deviation from the linear trend.

Hearing Aid Benefit

Over 75% of participants found thathearing aids helped a moderate amount or a lot

in certain situations (see Figure 6). Overallhearing aid benefit was determined fromQuestion 3 on the CSS (see Appendix A),whichasked the participants to indicate the level ofassistance (on a scale of one to four) providedby their hearing aids in six different situations.Overall benefit for each participant wascalculated by averaging the benefit ratingsover the number of situations that werepersonally relevant to each participant. Thisoverall hearing aid benefit experienced byparticipants was significant and moderatelycorrelated to Global (r [807] = 0.53,p < .01) andPositive Effect (r [951] = 0.52, p < .01), andsignificant,but weakly,correlated to Service andCost (r [828] = 0.19,p < .01),Negative Features(r [929] = 0.39, p < .01), and Personal Imagesatisfaction (r [943] = 0.2, p < .01). Therelationship demonstrated was positive, suchthat greater reported benefit from amplificationwas associated with higher levels of satisfactionfor the SADL.

Figure 4. Distribution of hearing aid use (N = 1014).

Figure 5. Mean SADL satisfaction level by extent of hearing aid use. Please note: only 3% of participants used aids<1 hr. per week or never.


Hearing Aid Difficulties

Over 90% of participants reporteddifficulties with hearing aids some or none ofthe time, with only 4% of the study groupexperiencing difficulties all the time. Similarlywith overall benefits, overall difficulties withhearing aids were calculated from Question4 on the CSS (see Appendix A). Significant,yet relatively weak, correlations wereobserved between hearing aid difficulties andSADL scores as follows: Global score (r [813]= -0.36, p < .01), Positive Effect (r [961] = -0.31,p < .01), Service and Cost (r [835] = -0.25, p< .01), Negative Features (r [940] = -0.24, p< .01), and Personal Image (r [953] = -0.23,p < .01). This negative relationship indicatedthat high levels of reported hearing aiddifficulties were associated with reducedsatisfaction levels.

DISCUSSION

The first aim of this study was to provideSADL data for a group of older Australian

hearing aid users.The results indicated that,overall, there was a considerable level ofparticipant satisfaction with amplification.This is consistent with a number of otherstudies conducted with older Australianhearing aid users (e.g., Hickson et al, 1999;McLeod et al, 2001). Results for the presentstudy can be compared to the SADL dataobtained by McLeod et al (2001), whichincluded similar participants in terms of ageand gender representation. Table 4 showsthe mean SADL scores obtained for the presentstudy,the two week and 12–24 month postfittinggroups from the McLeod et al study, as well asCox and Alexander’s (1999) provisional norms.Overall, satisfaction levels reported in thepresent study are generally more comparablewith the two-week group.In their study,McLeodet al reported that satisfaction declines overtime, because while hearing aid benefit israpidly noticed by the hearing aid user,negativeaspects of hearing aid use take longer to becomeapparent, and thus the hearing aid user mayexperience a “honeymoon”effect.The results forthe present study seem to support this assertion.

394

Table 4. Comparison of SADL Scores across Three Studies

McLeod et al (2001) Present Study McLeod et al (2001) Cox and Alexander (1999)

2 weeks 3–6 months 12–24 months At least 12 months postfitting postfitting postfitting postfitting(N = 45) (N = 1014) (N = 75) (N = 257)

Global 5.5 5.27 4.9 4.9Positive Effect 5.1 4.98 4.6 4.9Service and Cost 5.9 5.70 5.4 5.4Negative Features 5.5 4.74 4.3 3.6Personal Image 6.0 5.86 5.8 5.6

Demographic Details:Mean Age 75.7 years 75.3 years 76.3 years Not Reported% Males 46.4% 54.4% 43.0% Not Reported% Binaural Fit 75.1% 54.8% 72.0% Not Reported

Mean SADLScores

Figure 6. Distribution of hearing aid benefit (N = 1014).


395

All SADL scores in the present study, whichwere collected at three to six months postfitting,were lower than McLeod et al’s two-week group.Conversely, SADL scores for the present studywere higher than McLeod et al’s (2001) 12–24month postfitting group, possibly because thepresent data were collected at an earlier stagepostfitting.

The second aim of the study was to comparethe Australian SADL data to the provisionalnorms described by Cox and Alexander (1999).A statistically significant difference was notedbetween the Global scores and all subscalescores obtained for the two groups, with theAustralian group being significantly moresatisfied than the U.S. group.The difference insatisfaction cannot be attributed to differencesin the cost of the hearing aids, as the majorityof participants in both studies had hearing aidsthat were fully subsidized; to the exclusion ofthe reasonable cost item, as it was omitted inboth studies; or to differences in participantage, which was similar between studies.

The level of hearing aid technology suppliedto participants in both studies may possiblyaccount for the difference. Participants in thepresent study had predominantly been fittedwith digitally programmable hearing aids,whileparticipants in Cox and Alexander’s (1999)study were fitted primarily with “conventional”aids (presumably analogue), with fewprogrammable aids included in the group (Coxand Alexander, 2001). As programmabletechnology is related to higher satisfactionlevels relative to nonprogrammable technology(Kochkin, 2000), this may explain the higherSADL scores observed in the present study.Additionally,Cox and Alexander (1999) reportedadministering the SADL with participants atleast one year postfitting,while participants inthe present study were surveyed only three tosix months postfitting. McLeod et al (2001)reported that satisfaction decreases over time,although it is unsure at this stage at whatpoint this decline occurs and what is the mostappropriate point postfitting to administer theSADL. It is possible that satisfaction for theparticipants in the present study, if measuredat a later stage postfitting, may more closelyresemble the normative data developed by Coxand Alexander (1999), as was the case whenMcLeod et al (2001) compared their 12–24month postfitting participant group with Coxand Alexander’s (1999) interim norms.

Figure 2 illustrates that the differencebetween the Negative Features subscale scoresfor the Australian and U.S. samples is by far

greater than the other SADL scores. Otherstudies have also reported higher satisfactionfor Negative Features subscales compared toCox and Alexander’s (1999) interim norms(Hosford-Dunn and Halpern, 2000; McLeod etal, 2001). The finding for the present studyadds further weight to Cox and Alexander’s(2001) belief that upwards adjustment of thenormative data for this subscale may benecessary.

However,having stated that the differencesbetween the Australian and the U.S. samplesare statistically significant, the mathematicaldifference is relatively small, which makes itdifficult to speculate how this difference wouldtranslate clinically.For example, the significantdifference in the Positive Effect score for theAustralian group (M = 4.98) and their U.S.counterparts (M = 4.90) equates to amathematical difference of 0.08. Given thatthe critical test-retest difference for this subscaleis 0.90 (Cox and Alexander, 1999), a differenceof 0.08 between the two groups, regardless ofstatistical significance,does not seem compellingin terms of clinical applicability.

The third aim of this study was to examinethe effect of a number of participant and hearingaid variables on SADL results.A significant butweak positive correlation (r = .100) was notedbetween SADL Global scores and degree ofhearing loss. This relationship was reportedby Hosford-Dunn and Halpern (2001);however,a study by Jerram and Purdy (2001) failed toreveal a significant relationship between degreeof hearing loss and satisfaction. The weaknature of the correlation in the present studysupports Jerram and Purdy’s findings andindicates that degree of hearing loss is notstrongly associated with satisfaction withhearing aids.A positive relationship was foundbetween perceived degree of hearing difficultywithout hearing aids and mean SADL Global,Positive Effect,and Service and Cost scores.Thefinding that greater perceived hearing difficultyinitially is associated with higher satisfactionscores postfitting is supported in the literature(Hosford-Dunn and Halpern, 2001; Kochkin,2000). Similarly, the positive relationshipevident between prior hearing aid experienceand mean SADL Global, Positive Effect, andService and Cost scores is consistent with thefindings of a previous study by Kochkin (2000).It may be that experienced hearing aid usershave more realistic expectations regardingamplification and thus view amplification morefavorably. Experienced hearing aid users alsoreported greater amounts of hearing aid use,

which is positively associated with SADLsatisfaction.

No relationship was found between SADLGlobal scores and either gender or age,a findingthat is generally consistent with the literature.Previous studies examining the relationshipbetween gender and satisfaction have reportedinsignificant results (Hosford-Dunn andHalpern, 2001; Jerram and Purdy, 2001).Previous results regarding age and satisfactionhave been more inconsistent. Cox andAlexander’s (1999) preliminary data indicatedno difference in SADL scores for subjects underand over 60 years of age,and Jerram and Purdy(2001) likewise found no effect on satisfactionfor age. However, both of these studies mayhave been constrained by methodologicalfactors.For instance,Cox and Alexander (1999)appeared to compare their older subject groupover 60 years of age (N = 257) with a small groupof younger hearing aid users (N = 50), andJerram and Purdy (2001) only included asomewhat restricted age range (31–88 years ofage). Hosford-Dunn and Halpern (2001) wereable to demonstrate a negative effect of age onsatisfaction, and attributed the relationshipprimarily to the deleterious effects of age-relatedcentral processing changes on speechunderstanding (Gordon-Salant and Fitzgibbons,1999).The present study examined a smaller agerange (29–104 years) compared to the study byHosford-Dunn and Halpern (2001) (6–101 yearsof age), which may explain why a significanteffect for age was not observed in the presentstudy.

The present study revealed a significanteffect of hearing aid style on SADL Global,Personal Image,and Positive Effect scores.Thiseffect was in favor of ITC aids for PersonalImage, which is consistent with the findings ofhigher satisfaction for smaller aids noted inprevious studies (Baumfield and Dillon, 2001;Kochkin, 2000). The result for Positive Effectwas in favor of larger aids (BTEs and ITEs) overITC style aids.Although on first appraisal thiswould seem inconsistent with previous studiesreporting greater satisfaction for smaller aids(Baumfield and Dillon, 2001; Kochkin, 2000),respondents wearing larger hearing aids haveboth greater average hearing losses and greaterperceived hearing difficulties without hearingaids, which are both factors related to highersatisfaction. It is plausible to suggest thatbecause ITE and BTE hearing aids usuallyfeature more gain within their specifications,these aids deliver better performance to caterfor clients with these larger hearing losses.

These most probably explain the highersatisfaction for the Positive Effect subscale.

The remaining hearing aid variablesinvestigated in the present study did not showsignificant effects in relation to SADL scores.No difference was found between SADLsatisfaction scores for fully subsidized and top-up (partially subsidized) participants;however,as there were relatively few top-up participants(N = 104) compared to fully subsidizedparticipants (N = 910), this result should beinterpreted with caution.This finding,however,was also evident in Cox and Alexander’s (2001)study whereby no significant difference in SADLscores was observed between “co-pay”participants and fully subsidized Veteran’sAffairs participants.Monaural/binaural statusdid not have a significant effect on meansatisfaction scores. Kochkin (2000) did findhigher satisfaction scores on the MarkeTraksurvey (Kochkin, 2000) with binaural fittingrelative to monaural fitting; however, thedifference was very small, and the finding hasnot been repeated in subsequent studies usingother measures of satisfaction (SADL inHosford-Dunn and Halpern, 2001; single-itemmeasure in Jerram and Purdy, 2001).

The fourth aim of this study was to examinethe relationship between the SADL and otheroutcome measures typically used for evaluatinghearing aid fitting success, specifically single-item satisfactions scales, measures of hearingaid use, hearing aid benefit, and hearing aiddifficulties. The purpose of investigating thecorrelation between Australian SADL scores(Global and subscale) and single-itemsatisfaction scores was to reinforce the validityof the SADL as a true measure of satisfactionand to determine those subscale areas thatcontribute most to a respondent’s overallsatisfaction. A weak correlation was observedbetween participants’ Global scores on theSADL and their single-item scores for overallhearing aid satisfaction, as measured byQuestion 8 on the CSS. This shows that asparticipants’ satisfaction on the single-itemscale improved, so too did their satisfaction onthe SADL. This reconfirms the validity of theSADL as a genuine measure of satisfaction.However, the relationship between the twomeasures is not as strong as we might expect(r = 0.54) given that both scores are intendedto be indicators of hearing aid satisfaction.Some reduction in correlation strength may bedue to random error inherent in both types ofmeasures or the relative insensitivity of thesingle item (Cox and Alexander, 2001).


396


397

Additionally, the SADL questionnaire taps intocontent areas that may not always be consideredby participants when responding to a single-item satisfaction scale. For example, a personmay not always consider issues such as howcompetent the clinician was or how the hearingaids have affected their self-image when askedto rate on a single-item scale how satisfied theyare with their hearing aids. Instead, they mayfocus more on the communication benefits ornegative effects produced by their hearing aids.This was evidenced in the current study whenlooking at the relationship between single-itemscores for overall hearing satisfaction andindividual subscale scores on the SADL.Although the single-item satisfaction scalecorrelated with all four SADL subscales,correlations were particularly weak for theCost and Service subscale and the PersonalImage subscale.This indicates that when clientsrate their satisfaction level using a single-itemscale, they do not consider issues relating toservice provision and self-image as much ascommunication benefits and negative features.

Kochkin’s study (1992) revealed that areduction in the disability or handicap causedby a hearing loss (i.e., benefit) was,by itself,notsufficient to ensure a client was completelyhappy with the amplification aspect of theirrehabilitation program. It was found thathearing aid users considered improved hearingability, suitable sound quality, instrumentreliability, usefulness in multiple listening environments,postpurchase service, and fit/comfort to be themost important factors influencing consumersatisfaction.This was supported in the currentstudy in that the single-item hearing aidsatisfaction scale correlated with all four SADLsubscales,not just the Positive Effect subscale,which primarily relates to hearing aid benefit.That is, issues relating to sound quality andhearing aid benefit (i.e.,Positive Effect subscale),difficult listening situations (i.e., NegativeFeatures subscale), clinician competency andhearing aid reliability (i.e., Service and Costsubscale),and the appearance of the hearing aid(i.e., Personal Image subscale) all contributedto how satisfied a person was overall with theirhearing aids. This further highlights theimportance of using outcome measures thatlook beyond benefit when evaluating the successof a rehabilitation program.

The importance of evaluating how a clientfeels about the competency of their clinicianwhen measuring hearing aid success wasfurther supported in the current study. A

relatively weak positive correlation (r = 0.35)was observed between participants’ Globalscores on the SADL and single-item satisfactionscores for the quality of service provided bythe practitioner (as measured by Question 7 onthe CSS). This shows that as participants’satisfaction with their clinician increased,so toodid their satisfaction on the SADL. Thissupports Kochkin’s (1997) theory that a client’ssatisfaction with amplification is a function ofthree important variables: satisfaction withthe physical features of the instrumentdispensed; satisfaction with the instrument’sperformance in a variety of listening situations;and satisfaction with the services provided bythe clinician.

In the literature,degree of hearing aid useis often reported as an outcome measure forassessing hearing aid fitting success (e.g.,Dillonet al, 1997; Humes, 1999). In a study by Dillonet al (1991b), satisfaction with hearing aids (asmeasured by the single-item hearing aidsatisfaction scale on the HAUQ) correlatedhighly with reported hearing aid use. Thisrelationship was also observed in the currentstudy in that greater hearing aid use wasassociated with increased satisfaction on theSADL. However, Humes (1999) cautions thatemploying hearing aid use as a measure ofhearing aid success can be misleading. Notonly do clients tend to overestimate orunderestimate their use time,some clients mayonly require use of their hearing aids in certainlistening conditions and may be very happywith how their hearing aids are performingwithin this limited space of time.Alternatively,some clients,particularly those who rely heavilyon their hearing aids for day-to-day functioning,may persevere in using their hearing aidsdespite being dissatisfied with them.This issuewas highlighted in the current study in that,although there was a significant linearrelationship between hearing aid use and allfour SADL subscales, significant deviationsfrom the linear trend were exhibited for thePositive Effect,Negative Features,and Serviceand Cost subscales. That is, an increase inhearing aid use was not always associated witha steady increase in satisfaction scores on thesesubscales.For example,as can be seen in Figure5,there was a slight decrease in satisfaction withService and Cost with an increase in hearingaid use from never to less than one hour per day,followed by an increase in satisfaction withfurther use.This suggests that people who usetheir hearing aids a little may require morerepairs than those who never use their aids and

are thus less satisfied with the dependabilityof their hearing aids and the competency ofthe service provider. Additionally, those whouse their hearing aids a lot may be more usedto hearing aid breakdowns and/or be morefamiliar with how to care for their aids to ensurethat they last longer.

Hearing aid benefit is broadly defined asthe reduction in disability or handicap causedby a hearing loss (e.g., improved communicationability, increased participation in socialactivities) following amplification and/or auralrehabilitation (Cox and Alexander, 1999). In astudy by Baumfield and Dillon (2001), asignificant correlation was observed betweenclients’ single-item satisfaction scores on theHAUQ and their scores on a measure of hearingaid benefit (i.e., the Shortened Hearing AidPerformance Inventory for the Elderly[SHAPIE]). This relationship was alsodemonstrated in the present study in that themore benefit participants received from theirhearing aids in certain situations (as measuredby Question 3 on the CSS), the higher theiroverall satisfaction level on the SADL.However,this correlation was only moderate (r = 0.53),which suggests that, although benefit is asignificant determinant of satisfaction, there areother factors that contribute to overallsatisfaction.This once again supports Kochkin’s(1992) findings that satisfaction is more thanjust benefit.

Although not a traditional outcomemeasure, the number of hearing aid difficultiesreported by a client may also be used by theclinician as measure of hearing aid success(Dillon et al,1997).Baumfield and Dillon (2001)found a significant correlation between clients’single-item satisfaction scores on the HAUQ andtheir degree of hearing aid difficulties. Thisrelationship was also demonstrated in thecurrent study in that the more difficultiesparticipants experienced with their hearingaids, the lower their overall satisfaction level onthe SADL.However, this correlation was weak(r = 0.36), which suggests that, although thenumber of hearing aid difficulties experiencedby a client is a significant determinant ofhearing aid satisfaction, there are other factorsthat contribute to a client’s overall satisfactionwith their hearing aids.Thus, when validatingthe rehabilitation process in follow-up sessions,the clinician must not only focus on addressingand resolving the client’s hearing aid difficulties,they must also address other issues consideredto be important to clients if high levels ofsatisfaction are to be achieved (Kochkin,1992).

As expected, degree of hearing aid difficulty(as measured by Question 4 on the CSS)correlated with the Negative Features subscaleof the SADL. Less anticipated was thesignificant correlation observed between degreeof hearing aid difficulty and the other threeSADL subscales (i.e.,Positive Effect,Service andCost, and Personal Image). This suggests thata client experiencing difficulties with a hearingaid may not only develop negative perceptionsabout the hearing aid itself but also thecompetency of the service provider, thedependability of hearing aids, the overall benefitand value of amplification, and their self-confidence and personal image.

Clinical Applications

Cox and Alexander (1999) envisioned thatthe SADL might be used to evaluate satisfactionin a normative sense. If SADL norms wereestablished for certain distinct groups (e.g.,different styles of hearing aids,degree of hearingloss), the clinician could compare a particularclient’s SADL results to these norms todetermine whether or not their level ofsatisfaction was acceptable given certainpersonal and/or hearing aid characteristics.The results of the present study offerpreliminary evidence that it is necessary toestablish separate population norms, based onthe specific variables found to have a significantinfluence on SADL scores, such as style of aidand perceived degree of hearing difficulty.However, if using the SADL in a normativesense, it would be critical for the clinician toadminister the SADL within the same timeframe postfitting as when the normative groupwas evaluated,because,as shown in the McLeodet al (2001) study, client satisfaction tends todecline over time. Additionally, given thesignificant difference between the data obtainedfor the present study and the U.S. study (Coxand Alexander, 1999), it is recommended thatdata be collected for other nations that intendto use the SADL in a normative manner, withcareful attention to methodological factors.

The SADL could also be implemented at theservice level as an outcome measure.The Officeof Hearing Services (OHS) could examineaverage overall satisfaction scores for variousclinics, to ensure that all accredited serviceproviders were delivering a satisfactory serviceto clients.If a particular service provider was notachieving a comparable level of client satisfactionas other centers, the OHS could look at the


398


399

procedures and policies of this center as well asthe practices of individual clinicians andrecommend appropriate changes. From amarketing point of view,this information wouldbe particularly valuable for service providers tobe aware of, as those clients who are moresatisfied are more likely to recommend theservice to others (Kochkin, 2000).

It is the researchers’ view that the SADLis particularly appropriate for use at the servicelevel. It is important to note, however, thatDillon et al (1997) stated that self-reportmeasures that consist of a fixed list of situationsthat the client must rate in terms of difficulty(or satisfaction), are disadvantaged in thatcertain situations may be irrelevant to theclient. Although the 15 questions containedwithin the SADL were included based onresearch by Cox and Alexander (1999) intothose areas considered to be most important toclients in terms of overall satisfaction withamplification, the questionnaire may fail tocapture all clients who are dissatisfied. It maybe the case that a client is dissatisfied withsome issue not included on the SADL (e.g.,difficulty managing the small controls on thehearing aid; long waiting time for anappointment).A questionnaire in addition to theSADL that offers the client the opportunity toadd their own comments through open-endedquestions may be advantageous. At a servicelevel, the authors believe that the SADL has aninvaluable application as it offers quantifiableresults, organized into neat category areas,which would make collation of results at aservice level quicker and more efficient.

Limitations of the Study

Some of the limitations of the present studyincluded restricted availability of participantdetails and amplification characteristics, theparticipant group used in the study, theprocedure used for administering the SADL,andmissing data. Participant details that wereunavailable to researchers in the present studyincluded type of hearing loss (i.e., conductive,sensorineural,or mixed), four frequency average,symmetry of hearing loss, attitude towardamplification,hearing aid expectations,personalgoals for rehabilitation, and other disabilities(e.g., visual impairment).Amplification detailsthat were unavailable included use of additionalhearing aid features (e.g., multiple memoryprogram, directional microphones, telecoil,remote control) and the prescriptive procedureused to fit the participants’ hearing aid(s).

Hosford-Dunn and Halpern (2001) recognizedthat many complex variables such aspersonality,psychosocial adjustment to hearingloss, and overall health, which are typicallyweighed in clinical decision making, wereomitted from their study. In this study, theauthors did not have access to this information.The relationship of many of these listedvariables with regard to hearing aid satisfactionhas yet to be established (Humes et al, 2001).

The participant group for the present studywas limited to people who had receivedgovernment-funded hearing aids and did notinclude people who had paid the full cost oftheir hearing aids. The preliminary SADLnorms established by Cox and Alexander (1999)showed a difference between nonpaying andpaying clients, with the latter groupdemonstrating less satisfaction on the Serviceand Cost subscale of the SADL. Furtherinvestigation in this area for the Australianpopulation is warranted.

The CSS and SADL questionnaire wassent to participants via mail. Although allparticipants were strongly encouraged tocomplete and return the questionnaire using theenclosed reply paid envelope, a percentage ofclients failed to respond,despite everyone beingsent a reminder notice. This may introduce aresponse bias that is frequently indeterminate(Stein and Cutler, 1996). Thus, the normativedata obtained in the present study may not bea valid representation of the majority ofAustralian hearing aid users (under theCommonwealth Hearing Services Program).Of those questionnaires that were returned tothe OHS,a proportion was not fully completed.Dillon et al (1991a) highlighted thatmisinterpretation of questions may occur insurveys that are mailed to and filled in solelyby the client. It may have been the case thatsome of the questions in the CSS and/or SADLwere considered irrelevant to certain clients.Dillon et al (1991b) emphasizes that for anoutcome measure to be a valid indicator of therehabilitation success, it must be applicable tothe majority of clients.

CONCLUSION

This study has provided SADL data for theolder Australian population. Significant

differences were identified between Australianand U.S. levels of satisfaction as measured bythe SADL, with Australian hearing aid usersbeing more satisfied than their U.S. clients

across all SADL subscales. Several variablesthat can influence satisfaction levels wereidentified. These include perceived degree ofhearing difficulty without hearing aids,amountof previous hearing aid experience,and style ofhearing aid. Satisfaction scores for the SADLwere significantly related to all outcomemeasures examined, including single-itemsatisfaction (practitioner and hearing aids),self-reported hearing aid use, hearing aidbenefits, and hearing aid difficulties.

REFERENCES

Baumfield A, Dillon H. (2001) Factors affecting theuse and perceived benefit of ITE and BTE hearingaids. Br J Audiol 35:247–258.

Beck LB. (2000) The role of outcome data in health-care resource allocation. Ear Hear 21:89S–96S.

Bentler RA, Kramer SE. (2000) Guidelines for choos-ing a self-report outcome measure. Ear Hear21(4):37S–49S.

Cox RM,Alexander GC. (1995) The Abbreviated Profileof Hearing Aid Benefit (APHAB). Ear Hear 16:176–186.

Cox RM,Alexander GC. (1999) Measuring Satisfactionwith Amplification in Daily Life: the SADL scale. EarHear 20(4):306–320.

Cox RM, Alexander GC. (2000) Expectations abouthearing aids and their relationship to fitting outcome.J Am Acad Audiol 11:368–382.

Cox RM,Alexander GC. (2001) Validation of the SADLquestionnaire. Ear Hear 22(2):151–160.

Dillon H, Birtles G, Lovegrove R. (1999) Measuringthe outcomes of a national rehabilitation program:normative data for the Client Oriented Scale ofImprovement (COSI) and the Hearing Aid User’sQuestionnaire (HAUQ). J Am Acad Audiol 10:67–79.

Dillon H, James A, Ginis J. (1997) Client OrientedScale of Improvement (COSI) and its relationship toseveral other measures of benefit and satisfaction pro-vided by hearing aids. J Am Acad Audiol 8(1):27–43.

Dillon H, Koritschoner E, Battaglia J, Lovegrove R,Ginis J, Mavrias G, Carnie L, Roy P, Forsythe L,TowersE, Goulias H, Macaskill F. (1991a) Rehabilitation effec-tiveness I: assessing the needs of clients entering anational hearing rehabilitation program. Aust J Audiol13(2):55–65.

Dillon H, Koritschoner E, Battaglia J, Lovegrove R,Ginis J, Mavrias G, Carnie L, Roy P, Forsythe L,TowersE, Goulias H, Macaskill F. (1991b) Rehabilitation effec-tiveness II: assessing the outcomes for clients of anational hearing rehabilitation program. Aust J Audiol13(2):68–82.

Gagné J. (2000) What is treatment evaluation research?What is its relationship to the goals of audiologicalrehabilitation? Who are the stakeholders of this typeof research? Ear Hear 21(4):60S–73S.

Gordon-Salant S, Fitzgibbons PJ. (1999) Profile of audi-tory temporal processing in older listeners. J SpeechLang Hear Res 42:300–311.

Hickson L,Timm M,Worrall L. (1999) Hearing aid fit-ting: outcomes for older adults. Aust J Audiol 21:19–23.

Hosford-Dunn H, Halpern J. (2000) Clinical applica-tion of the Satisfaction with Amplification in DailyLife scale in private practice I: statistical, content andfactorial validity. J Am Acad Audiol 11:523–539.

Hosford-Dunn H, Halpern J. (2001) Clinical applica-tion of the SADL scale in private practice II: predictivevalidity of fitting variables. J Am Acad Audiol 12:15–36.

Huch J. (1999) In documenting user benefit/satisfac-tion there are many tools to choose among. Hear J52(4):60–70.

Humes LE. (1999) Dimensions of hearing outcome. JAm Acad Audiol 10:26–39.

Humes LE, Garner CB,Wilson DL, Barlow NN. (2001)Hearing-aid outcome measures following one monthof hearing aid use by the elderly. J Speech Lang HearRes 44:469–486.

Humes LE, Halling D, Coughlin M. (1996) Reliabilityand stability of various hearing-aid outcome meas-ures in a group of elderly hearing aid wearers. J SpeechHear Res 39:923–935.

Jerram JC, Purdy SC. (2001) Technology, expectations,and adjustment to hearing loss: predictors of hearingaid outcome. J Am Acad Audiol 12:64–79.

Kochkin S. (1992) MarkeTrak III identifies key fac-tors in determining consumer satisfaction. Hear J45(8):39–44.

Kochkin S. (1997) MarkeTrak IV : what is the viablemarket for hearing aids? Hear J 50:31–39.

Kochkin S. (2000) MarkeTrak V: consumer satisfac-tion revisited. Hear J 53(1):38–55.

Kricos PB, Lesner SA. (2000) Evaluating the successof adult audiologic rehabilitation support programs.Semin Hear 20(3):267–279.

McLeod B, Upfold L, Broadbent C. (2001) An investi-gation of the applicability of the inventory, Satisfactionwith Amplification in Daily Life, at 2 weeks post hear-ing aid fitting. Ear Hear 22(4):342–347.

Office of Hearing Services. (2002) Clinical standards–-schedule 2. www.health.gov.au/pdf/spc5.pdf (accessedOctober 16, 2002).

Stein F, Cutler SK. (1996) Clinical Research in AlliedHealth and Special Education. 3rd ed. San Diego:Singular Publishing Group.

Weinstein B. (2000) Outcome measures in rehabili-tative audiology. In: Alpiner JG, McCarthy PA, eds.Rehabilitative Audiology: Children and Adults. 3rded. Philadelphia: Lippincott, Williams and Wilkins,575–594.


400


401

Appendix A: Client Satisfaction Survey

Instructions: Some questions require a tick in the answer box. Please tick only one box perquestion. For some other questions you will be asked to indicate your answer by circling a number.For these questions you will need to circle a number on each row unless otherwise directed.Other questions require a short written answer.

1. How many hearing aids do you usually wear?(Tick one box only)

One Hearing Aid 1

Two Hearing Aids 2

2. How often do you use your hearing aid/s?(Tick one box only)

More than 8 hours a day 1

Between 5 and 8 hours a day 2

Between 1 and 4 hours a day 3

Less than 1 hour a day 4

Less than 1 hour a week 5

Never 6

2a. If you never wear your hearing aid/s please tell us why.

3. How much has your hearing aid helped you with any of the following?(Circle the appropriate number for each statement—if you do not participate in any of the activities please leavethe row blank)

A lot A moderate A little Not at alamount

Family conversation 1 2 3 4

Small groups 1 2 3 4

Meetings/Church 1 2 3 4

Social situations 1 2 3 4

TV/Radio 1 2 3 4

Telephone 1 2 3 4

4. How often do you experience problems with any of the following when using your hearing aid?(Circle the appropriate number for each statement)

All of the Most of Some of None of time the time the time the time

Positioning the hearing aid or removing the hearing aid 1 2 3 4

Adjusting the controls of the hearing aid 1 2 3 4

The aid whistling when it is in the ear and set at a comfortable listening level 1 2 3 4

The hearing aid or ear mould causing you discomfort 1 2 3 4

The hearing aid making any sudden loud noises(unbearably loud not just annoying) 1 2 3 4

Your own voice sounding hollow or echoing 1 2 3 4

5. Have you reported these problems to your hearing practitioner?(Tick one box only)

Yes 1

No 2

If No, go to Question 7.

6. What action did your practitioner take after you reported the problem?

The next few questions are about your satisfaction with the service you received and with the hearing aid/s.

7. How satisfied were you with the quality of service provided by your chosen practitioner?(Tick one box only)

Very satisfied 1

Satisfied 2

Dissatisfied 3

Very dissatisfied 4

8. How satisfied are you with your hearing aid/s?(Tick one box only)

Very satisfied 1

Satisfied 2

Dissatisfied 3

Very dissatisfied 4

9. If you are dissatisfied with your hearing aid/s or with the service provided by your chosen practitioner,please explain why.(Please feel free to attach any additional written material that you feel may help to answer the question)

10. Do you want the Office of Hearing Services to follow up on this matter with your practitioner?

Yes 1

No 2

Not applicable 3


402

Measuring Hearing Aid Outcomes Using the Satisfaction with ... · PDF fileSurvey;HAUQ =...

Documents

Transcript of Measuring Hearing Aid Outcomes Using the Satisfaction with ... · PDF fileSurvey;HAUQ =...