The Laryngoscope

© 2020 American Laryngological,

Rhinological and Otological Society Inc,
“The Triological Society” and American
Laryngological Association (ALA)

Long-term Impact of Hearing Aid Provision or Cochlear Implantation

on Hearing Handicap

Alexander S. Kim, MTM ; Joshua F. Betz, MS; Carrie L. Nieman, MD, MPH ; Matthew R. Hoyer, MD;
Jeremy Applebaum, MD; Frank R. Lin, MD, PhD; Adele M. Goman, PhD

Objectives/Hypothesis: Previous research has shown hearing handicap to be reduced following hearing aid use or
cochlear implantation in short-to-medium follow-up periods, yet the impact of interventions for hearing loss on hearing handi-
cap in the long term remains understudied. This article reports hearing handicap at 6 months, 12 months, and 5 years after
either hearing aid provision or cochlear implantation.
Study Design: Observational study.
Methods: A study of 115 participants from the Studying Multiple Outcomes after Aural Rehabilitative Treatment
(SMART) study cohort assessed self-reported hearing handicap using the Hearing Handicap Inventory for the Elderly Screening
version (HHIE-S) at baseline, 6 months, 12 months, and 5 years. Generalized estimating equations (GEE) were used to estimate
the population mean HHIE-S score over time, accounting for the correlated nature of repeated measures data, and multiple
imputation with chained equations was performed to impute missing data.
Results: Compared to baseline, mean HHIE-S scores after hearing aid provision were significantly reduced at 6 months
(mean = −7.96, 95% confidence interval [CI]: −10.40, −5.53), 12 months (mean = −6.58, 95% CI: −9.26, −3.90), and 5 years
(mean = −4.58, 95% CI: −7.87, −1.30). After cochlear implantation, mean hearing handicap scores were also significantly lower
compared to baseline at 6 months (mean = −8.18, 95% CI: −11.07, −5.30), 12 months (mean = −10.04, 95% CI: −12.92,
−7.16), and 5 years (mean = −8.97, 95% CI: −12.92, −7.16).
Conclusions: This study found short-term benefits from hearing aids and cochlear implantation on hearing handicap
were maintained over 5 years.
Key Words: Hearing loss, hearing handicap, hearing aids, cochlear implant, long-term outcomes.
Laryngoscope, 00:1–5, 2020

INTRODUCTION of hearing loss interventions on hearing handicap in the

Hearing loss affects over 38 million Americans, long term remains understudied.
which includes nearly two-thirds of adults over the age of One of the established measures of hearing in every-
70 years.1 Previous research has shown the perceived dif- day life is the Speech, Spatial, and Qualities of Hearing
ficulty in everyday situations attributable to hearing loss, Scale, which broadly assesses the range of hearing func-
termed hearing handicap, to be reduced following hearing tions that are deployed in monitoring and attending to the
aid use2–7 or cochlear implantation8–10 in short-to- auditory scene, including components of direction, distance,
medium (<1 year) follow-up periods. However, the impact movement, sound quality, and sound-source segregation.11
In contrast, the Hearing Handicap Inventory for the
Elderly (HHIE) was developed to address not only the situ-
ational difficulties but also the psychosocial effects of hear-
From the Johns Hopkins University School of Medicine (A.S.K., C.L.N., ing loss.12 The 25-item HHIE scale has a score range of
M.R.H., J.A., F.R.L.),
Baltimore, Maryland, U.S.A.; Cochlear Center for Hearing
and Public Health (A.S.K., J.F.B., C.L.N., M.R.H., J.A., F.R.L., A.M.G.), Johns
0 (no handicap) to 100 (worse handicap), whereas a 10-item
Hopkins Bloomberg School of Public Health, Baltimore, Maryland, U.S.A.; version, the HHIE Screening (HHIE-S), has a score range
and the Department of Biostatistics (J.F.B.), Johns Hopkins Bloomberg from 0 (no handicap) to 40 (worse handicap).13 Hearing
School of Public Health, Baltimore, Marland, U.S.A.
handicap scores have been shown to reduce following hear-
This manuscript was supported in part by grants R01AG055426:
NIH/NIA (PI: Lin), R33DC015062: NIH/NIDCD (PI: Lin), and ing aid use for those with mild–moderate hearing loss.3–5 A
K23AG059900: NIH/NIA (PI: Nieman). F.R.L. is a consultant to randomized controlled trial of 104 older adults with moder-
Boeringher-Ingelheim and Autifony Inc., and a scientific advisory board
member for Autifony Inc. F.R.L. and C.L.N. are board members of the non-
ate high-frequency hearing loss demonstrated that HHIE
profit Access HEARS. scores were significantly reduced from baseline to 4 months
Editor’s Note: This Manuscript was accepted for publication on and were maintained over a 12-month period.3 Significant
September 29, 2020.
The authors have no other funding, financial relationships, or con- reductions in hearing handicap have also been shown in
flicts of interest to disclose. more recent observational studies using newer technol-
Accepted for presentation at the Triological Society 123rd Annual
Meeting at COSM, Atlanta, Georgia, U.S.A., April 24, 2020 (meeting
ogy.4,5 For instance, Vuorialho et al. administered the
canceled). HHIE-S to 98 first-time hearing aid users with moderate
Send correspondence to Adele M. Goman, PhD, 2024 East Monu- hearing loss, and HHIE scores reduced from 28.7 (out of a
ment Street, Baltimore, MD 21205. E-mail: [email protected]
possible 40) at baseline to 12.7 6 months later.5 Benefits
DOI: 10.1002/lary.29175 have also been shown with cochlear implantation for

Laryngoscope 00: 2020 Kim et al.: Five-Year Aural Rehab for Hearing Handicap
1
individuals with severe–profound hearing loss in small instructed not to skip a question if they avoid a situation because
observational studies.8,10 Vermeire et al. found hearing of a hearing problem. In addition, participants were instructed
handicap was significantly reduced at 4 months after to respond with the way they hear when using the hearing aid
or cochlear implant.
cochlear implantation among a group of 29 profoundly
deafened adults.10 Amoodi et al. similarly found reduced
hearing handicap among a group of 27 adults, with the
Other Variables
mean HHIE score reducing from 68 out of 100 before
Self-reported demographic (age, sex, race/ethnicity, educa-
implantation to 35 out of 100 1 year after implantation.8 tion) and health history data (hypertension, diabetes, and
Although common clinical interventions, little is smoking), and pure-tone audiometric hearing thresholds were
known regarding whether short-to-medium reductions in gathered during the baseline session.
hearing handicap from hearing aid use or cochlear
implantation are sustained in the long term. One study
found that reductions in hearing handicap with hearing Procedure
aid provision were maintained after 5 years, whereas Baseline evaluations of hearing handicap took place either
non–hearing aid users experienced worsening hearing before or within 1 to 2 weeks of receiving a hearing aid or under-
handicap after 5 and 11 years, consistent with the pro- going cochlear implantation, and in-person follow-up evaluations
gression of age-related hearing loss.14 Another study took place at 6 and 12 months following intervention.15 For the
observed reduced hearing handicap at 1 year following 5-year assessment of hearing handicap, participants completed
hearing aid provision that was sustained at 2 years, but the HHIE-S either on paper or electronically. From the 156 partici-
substantial attrition occurred, with just 36% of partici- pants consented to the SMART study, one did not receive treat-
ment, 22 had not consented to be contacted about future follow-up
pants completing the HHIE at the 2-year time point.6
studies, and Johns Hopkins medical records indicated that four
Currently, there are no known long-term studies reported participants were known to be deceased. As such, 129 participants
for hearing handicap after cochlear implantation. The who had received either cochlear implantation or hearing aids
present study recontacted participants from a prospective were contacted for the 5-year follow-up via email or mail according
cohort receiving hearing aids or cochlear implants to to preferred contact methods indicated during the baseline assess-
assess their respective intervention’s short-to-medium ment; contact details were updated using Johns Hopkins records
and long-term impact on hearing handicap. Based on pre- as needed. At 1-week, participants who had provided an email
vious research, we hypothesized that hearing handicap address and who had not yet responded were sent an email
would be reduced in the short-term and would be reminder. Mailed reminders were sent at 2 weeks to all partici-
maintained over the medium and long term. pants who had not yet responded. Telephone reminders were
made at 4 weeks to all participants who had not responded. Mean
duration since baseline for the 5-year follow-up was 4.8 years
(standard deviation = 0.7). The Johns Hopkins Medicine Institu-
MATERIALS AND METHODS tional Review Board reviewed and approved the study.
Participants
One hundred fifty-six patients from the Johns Hopkins
Department of Otolaryngology were consented to the Studying Analyses
Multiple Outcomes after Aural Rehabilitation Treatment Generalized estimating equations (GEEs) were used to esti-
(SMART) study.15 The focus of the study was to evaluate the mate the population mean HHIE-S score over time, accounting for
impact of hearing treatment on a broad range of functional the correlated nature of repeated measures data. Responses were
domains in older adults. All participants were oral–aural com- assumed to be approximately normally distributed after factoring
municators, English speaking, ≥50 years old, and were receiving in demographic (age, sex, race/ethnicity, education) and health fac-
a cochlear implant or hearing aid for the first time (or had less tors (hypertension, diabetes, smoking), with an independence work-
than 1 hour a day prior hearing aid use). Full details of study ing correlation structure. GEEs provide consistent estimates of the
recruitment and design of the SMART study is described else- population average even when the working correlation structure is
where.15 A follow-up study initiated in 2017 gathered long-term misspecified. GEE estimates are only valid under the assumption
outcomes at approximately 5 years after baseline. The analytic that data are missing completely at random. Therefore, multiple
sample (N = 115) consisted of individuals who had completed a imputation with chained equations was performed to impute miss-
baseline session, received treatment, and completed at least one ing data. A total of 50 imputed datasets were created, each having
follow-up session either at 6 months, 12 months, or 5 years. 20 iterations of the imputation process to achieve convergence. No
adjustment was made to confidence intervals (CIs) to account for
multiple comparisons. All analyses were performed in R version
3.4.1 (R Foundation for Statistical Computing, Vienna, Austria).
Hearing Handicap Inventory for the Elderly
Screening Version
The HHIE-S is a validated 10-item questionnaire assessing
the perceived emotional and situational impact of hearing loss RESULTS
among older adults.13,16,17 For each question, participants
Demographic Data
respond “yes” (four points), “no” (zero points), or “sometimes”
(two points). Scores range from zero to 40, and higher scores indi- Table I shows baseline summary demographic and
cate greater handicap. A referral to an audiologist or hearing health data for the two groups of participants. The two
specialist is recommended for scores greater than nine9, as groups were similar in age, race/ethnicity, and health his-
scores from 10 to 24 reflect a mild-to-moderate handicap, and tory, though baseline HHIE scores were significantly lower
26 to 40 represent severe handicap.13 Participants were in the hearing aid group. The hearing aid group reported

Laryngoscope 00: 2020 Kim et al.: Five-Year Aural Rehab for Hearing Handicap
2
 TABLE I.
Summary Self-Reported Baseline Demographic and Health Information, and Objective Hearing Thresholds for the Cochlear Implant and
Hearing Aid Groups.
Cochlear Implant, N = 51 Hearing Aid, N = 64 Total, N = 115 P Value

Age, yr [IQR] 69 [64.2–78.5] 71 [63.5–75.8] 69.6 [63.6–77.2] .931

Female 21 (41.2%) 26 (40.6%) 47 (40.9%) 1.000
Race: White 46 (90.2%) 58 (90.6%) 104 (90.4%) 1.000
Education
High school or less 12 (23.5%) 4 (6.2%) 16 (13.9%) .012
College/associates 22 (43.1%) 26 (40.6%) 48 (41.7%)
Postgraduate 16 (31.4%) 33 (51.6%) 49 (42.6%)
Missing data 1 (2%) 1 (1.6%) 2 (1.7%)
Smoking status
Current/former smoker 30 (58.8%) 26 (40.6%) 56 (48.7%) .087
Nonsmoker 20 (39.2%) 36 (56.2%) 56 (48.7%)
Missing data 1 (2%) 2 (3.1%) 3 (2.6%)
History of hypertension
Hypertension 27 (52.9%) 35 (54.7%) 62 (53.9%) 1.000
Missing data 2 (3.9%) 1 (1.6%) 3 (2.6%)
History of diabetes
Diabetes 7 (13.7%) 17 (26.6%) 24 (20.9%) .163
Missing data 2 (3.9%) 1 (1.6%) 3 (2.6%)
Baseline HHIE scores (SD) 27.7 (8.9) 18.7 (9.8) 22.9 (10.4) <.001
PTA [IQR]* 69.4 [64.4–81.6] 36.2 [27.5–42.8] 46.9 [35.0–65.3] <.001

Parentheses indicate percent unless otherwise stated.

*
PTA indicates the better-ear four-frequency (0.5, 1, 2, 4 kHz) PTA threshold (decibels hearing level).
HHIE = Hearing Handicap Inventory for the Elderly, IQR = interquartile range; PTA = pure-tone average; SD = standard deviation.

higher levels of education than the cochlear implant group. group as calculated from multiple imputation. Partici-
An analysis of associations between baseline HHIE scores pants with missing data were compared against those
and demographic factors was performed for potential with data from all four study assessments, and no signifi-
covariates but was not significant for any variable. cant differences were found except when comparing
cochlear implants and hearing aids, as a higher percent-
age of participants in the cochlear implant group com-
Missing Data pleted all three follow-ups compared to the hearing aid
Among the 115 participants, 83.5% of participants group. For the cochlear implant group, participants were
(n = 96) completed the baseline session and at least two measured on average to have a severe handicap that
follow-up assessments. Of the participants, 40.9% (56.9% of reduced to a mild to moderate handicap at all follow-up
the cochlear implant group and 28.1% of the hearing aid periods. The mean HHIE-S score decreased from a base-
group) completed all study assessments (baseline, 6 months, line of 28.30 (95% CI: 22.84, 33.76) to 20.11 (95% CI:
12 months, and 5 years). At 6 months, 12 months, and 14.62, 25.60) at 6 months (mean difference = −8.18, 95%
5 years, hearing aid group response rates were 94%, 58%, CI: −11.07, −5.30) and to 18.26 (95% CI: 12.73, 23.78) at
and 42%, respectively, whereas the cochlear implant group 12 months (mean difference = −10.04, 95% CI: −12.92,
response rates were 94%, 80%, and 61%. −7.16), and to 19.33 (95% CI: 13.56, 25.11) at 5 years
(mean difference = −8.97, 95% CI: −12.92, −7.16). For
the hearing aid group, participants were measured on
Hearing Aid Usage average to have a mild-to-moderate handicap, which was
Table II displays the hearing aid usage patterns as maintained at all follow-up periods. HHIE-S scores
reported at 6 months, 12 months, and 5 years. Though decreased from a baseline of 19.24 (95% CI: 12.85, 25.64)
usage >10 hours increased from 40.6% at 6 months to to 11.28 (95% CI: 4.83, 17.73) at 6 months (mean difference
62.7% at 5 years, there was a high degree of missingness = −7.96, 95% CI: −10.40, −5.53), and to 12.66 (95% CI:
at 12 months (42.2%) and 5 years (27.5%). 6.00, 19.33) at 12 months (mean difference = −6.58, 95%
CI: −9.26, −3.90), and to 14.66 (95% CI: 7.99, 21.32) at
5 years (mean difference = −4.58, 95% CI: −7.87, −1.30).
Hearing Handicap Though wider CIs were observed at 5 years, differences in
Figure 1 plots the mean HHIE-S score differences HHIE scores were significant and negative at all time
from baseline at 6 months, 12 months, and 5 years by points in both groups.

Laryngoscope 00: 2020 Kim et al.: Five-Year Aural Rehab for Hearing Handicap
3
 TABLE II.
Hearing Aid Usage Patterns.
Hearing Aid Usage Per Day
Follow-up <1 Hour 1–5 Hours 6–10 Hours >10 Hours Missing

6 months 5 (7.8%) 13 (20.3%) 17 (26.6%) 26 (40.6%) 3 (4.7%)

12 months 1 (1.6%) 9 (14.1%) 12 (18.8%) 15 (23.4%) 27 (42.2%)
5 years 1 (2%) 1 (2%) 3 (5.9%) 32 (62.7%) 14 (27.5%)

and Vuorialho et al. observed significant reductions in

hearing handicap among samples with greater degrees of
hearing loss (mean four-frequency pure tone aver-
age = 70.7 dB and 48 dB HL, respectively, versus 36.2 dB
HL in the present study) and poorer hearing handicap
scores at baseline.4,5 For instance, whereas 6-month
scores observed in the current study were similar to those
observed by a similarly aged cohort at 4 months after
hearing aid provision, the participants in the current
study reported substantially less handicap at baseline
(18.99 compared to 31.1 reported by Chang et al. and
28.7 reported by Vuorialho et al.).4,5 Therefore, as the
participants in the current study were experiencing less
handicap at baseline, they had less room for reduction
following hearing aid use compared to other studies.
Though the data in this article present novel find-
ings in the long-term follow-up for hearing handicap
research, notable limitations of this study include the
Fig. 1. Mean Hearing Handicap Inventory for the Elderly (HHIE)- lack of an untreated cohort, qualitative satisfaction data,
Screening score differences from baseline for hearing aid (white)
and cochlear implant (gray) groups at 6 months, 12 months and
and complete hearing aid usage patterns over time. Thus,
5 years. Error bars show 95% confidence intervals. this observational study cannot comment on the contribu-
tion of normal aging on the trends observed, including
benefit from hearing aids and cochlear implantation rela-
DISCUSSION tive to a population without these interventions. Addi-
Hearing aids and cochlear implants are mainstays in tional data regarding hearing aid usage patterns may
the clinical management of hearing loss, though rela- elucidate whether compliance was a factor in the possible
tively little is known regarding the long-term impact of attenuation of the benefits seen by the 5th year of these
these interventions on hearing handicap. We found that respective participants. A more robust cohort with more
hearing aids and cochlear implants were associated with frequent measurements between the 1st and 5th year
sustained improvements in hearing handicap up to since intervention may further strengthen the trend
5 years postintervention. Among participants undergoing seen in this article, though our group was limited by the
cochlear implantation, perceived hearing handicap was feasibility of conducting such a trial.
significantly reduced at 6 months and 12 months. This This study also experienced a high degree of missing
observation is consistent with previous smaller-sampled data, with 59% of participants having missing data for at
observational studies looking at short-term outcomes.8,10 least one follow-up visit, though this is consistent with pre-
The present study additionally observed that the benefit vious longitudinal research with clinical populations.6 Com-
following cochlear implantation was maintained after pared to participants who received hearing aids, a higher
5 years. For hearing aid users, reductions in HHIE-S percentage of participants in the cochlear implant group
scores were also maintained at 5-year follow-up compared completed all three follow-ups. This may be due to partici-
to baseline, though their reductions were less pronounced pants in the cochlear implant group having more routine
compared to those receiving cochlear implantation. How- clinical appointments separate from our study in the first
ever, we do not know whether these reductions are clini- year following cochlear implantation. To account for the
cally meaningful from the perspective of the participants. high degree of missingness, we utilized multiple imputa-
Previous research has found significant benefits in hear- tion to reduce potential bias in the study. This study bal-
ing handicap following hearing aid use in the short anced the demands of a more resource-intensive study
term.2–7 Potential reasons for differences in the strength design to address the long-term impact of hearing loss on
of findings between studies include differences in analytic hearing handicap, and future studies will require rigorous
approach and sample characteristics. Both Chang et al. study designs to further strengthen the present findings.

Laryngoscope 00: 2020 Kim et al.: Five-Year Aural Rehab for Hearing Handicap
4
CONCLUSION 7. Chisolm TH, Abrams HB, McArdle R. Short- and long-term outcomes of
adult audiological rehabilitation. Ear Hear 2004;25:464–477.
This study presents short-term benefits of cochlear 8. Amoodi HA, Mick PT, Shipp DB, et al. Results with cochlear implantation
implantation and hearing aid provision on hearing handi- in adults with speech recognition scores exceeding current criteria. Otol
Neurotol 2012;33:6–12.
cap that were maintained over 5 years. Though random- 9. Summerfield Q, Barton GR, Toner J, et al. Self-reported benefits from suc-
ized control trial designs are ideal to establish the impact cessive bilateral cochlear implantation in post-lingually deafened adults:
of interventions, this study is one of the first to explore randomised controlled trial. Int J Audiol 2006;45:99–107.
10. Vermeire K, Brokx JPL, Wuyts FL, Cochet E, Hofkens A, Van de
the long-term association of interventions for hearing loss Heyning PH. Quality-of-life benefit from cochlear implantation in the
with hearing handicap. elderly. Otol Neurotol 2005;26:188–195.
11. Gatehouse S, Noble W. The Speech, Spatial and Qualities of Hearing Scale
(SSQ). Int J Audiol 2004;43:85–99.
12. Ventry IM, Weinstein BE. The hearing handicap inventory for the elderly: a
BIBLIOGRAPHY new tool. Ear Hear 1982;3:128–134.
13. Ventry IM, Weinstein BE. Identification of elderly people with hearing prob-
1. Goman AM, Lin FR. Prevalence of hearing loss by severity in the United lems. ASHA 1983;25:37–42.
States. Am J Public Health 2016;106:1820–1822. 14. Dawes P, Cruickshanks KJ, Fischer ME, Klein BEK, Klein R, Nondahl DM.
2. Mulrow CD, Aguilar C, Endicott JE, et al. Quality-of-life changes and hear- Hearing-aid use and long-term health outcomes: hearing handicap, men-
ing impairment a randomized trial. Ann Intern Med 1990;113:188–194. tal health, social engagement, cognitive function, physical health, and
3. Mulrow CD, Tuley MR, Aguilar C. Sustained benefits of hearing aids. mortality. Int J Audiol 2015;54:838–844.
J Speech Hear Res 1992;35:1402–1405. 15. Li L, Blake C, Shpritz B, et al. The studying multiple outcomes after aural
4. Chang W-H, Tseng H-C, Chao T-K, Hsu C-J, Liu T-C. Measurement of hear- rehabilitative treatment (SMART) study: study design and baseline
ing aid outcome in the elderly: comparison between young and old elderly. results. J Am Geriatr Soc 2014;62:S99.
Otolaryngol Head Neck Surg 2008;138:730–734. 16. Lichtenstein MJ, Bess FH, Logan SA. Diagnostic performance of the hear-
5. Vuorialho A, Karinen P, Sorri M. Effect of hearing aids on hearing disability ing handicap inventory for the elderly (screening version) against differing
and quality of life in the elderly. Int J Audiol 2006;45:400–405. definitions of hearing loss. Ear Hear 1988;9:208–211.
6. Humes LE, Wilson DL, Barlow NN, Garner C. Changes in hearing-aid bene- 17. Sindhusake D, Mitchell P, Smith W, et al. Validation of self-reported hear-
fit following 1 or 2 years of hearing-aid use by older adults. J Speech Lang ing loss. The Blue Mountains Hearing Study. Int J Epidemiol 2001;30:
Hear Res 2002;45:772–782. 1371–1378.

Laryngoscope 00: 2020 Kim et al.: Five-Year Aural Rehab for Hearing Handicap
5