Predictors of Short-Term Changes in Quality of Life after Cochlear Implantation

Abstract

Objective:

(1) To measure the impact of cochlear implantation on health-related quality of life (HR-QOL) using the Cochlear Implant Quality of Life (CIQOL) questionnaire. (2) To determine audiologic, demographic and non-CI/hearing-related QOL factors influencing the CIQOL.

Study Design:

Prospective observational study.

Setting:

Tertiary referral center.

Patients & Interventions:

Thirty-seven adult patients with sensorineural hearing loss undergoing cochlear implantation.

Main Outcome Measure(s):

CIQOL global score pre- and 6-months post-implantation. Physical function score as measured by the short form survey (SF-36), audiologic, and demographic variables.

Results:

CIQOL showed significant improvement from pre-implantation to 6-months post-activation with a mean difference of 14.9 points (95% confidence interval, 11.3 to 18.5; p < 0.0001). Improvement in CIQOL (ΔCIQOL) correlated linearly with age (r = −0.49; p = 0.001) and improvement in speech perception testing (r = 0.63; p < 0.0001). Multivariate modeling using age and change in CNC score explained 46% of the variability measured by the ΔCIQOL global score.

Conclusions:

Nearly all CI recipients achieve significant gains for all domains as measured by the CIQOL. However, younger patients and those with a greater improvement in speech perception performance (CNC) are more likely to achieve a greater CIQOL benefit. Results here suggest the importance of considering preoperative CIQOL and speech-perception measures when evaluating predictors of HR-QOL.

INTRODUCTION:

Cochlear implant (CI) technology was invented to improve speech recognition abilities and enhance communication when benefit from conventional amplification is no longer derived due to the extent of hearing loss severity.^1–3 CI technology has proven to fulfill this primary aim in most users, but outcomes evaluation is unfortunately still focused almost only on the improvement in speech-perception performance, while tools for evaluating improvements of hearing-related functions and activities that determine hearing-related quality-of-life (QOL) lag far behind.^{4, 5} Although most achieve significant improvements in speech recognition after implantation, there is significant variability in how CI recipients may experience indirect social and emotional benefits.^6–9 These benefits are not captured when focusing on speech-perception performance alone.

One approach to better understand these indirect outcomes after implantation is the use of quality-of-life (QOL) measures. Specifically, health-related QOL (HR-QOL) metrics can be used to evaluate a patient’s perception of health conditions and treatment on one’s well-being. Several disease-specific measures have been developed for hearing loss which include Speech, Spatial and Quality of Hearing Scale (SSQ)¹⁰, Hearing Handicap Inventory for Adults/Elderly (HHIA/E)^{11, 12}, and the Abbreviated Profile of Hearing Aid Benefit (APHAB)¹³. None of these measures incorporated CI users during development who often have more severe hearing loss; thus, these measures may not be reliable to understand the CI user population.^14–16 The motivation behind using a HR-QOL metric specific for CI recipients is that in previous studies on populations of CI users^{14, 17, 18} negligible to low positive correlations were found between non-CI-specific HR-QOL metrics and speech-perception scores. These findings raise the possibility that the hearing rehabilitation journey for CI users might be unique and called for evaluating it using a unique and specific HR-QOL metric. The Nijmegen Cochlear Implant Questionnaire (NCIQ) is the most commonly used CI-specific HR-QOL metric.¹⁹ Most studies have shown significant improvement in NCIQ scores after implantation on most domains.^{6, 19–21} It has been recently noted that the NCIQ may not contain domains—entertainment, environment, listening effort—that have been shown to be of importance to CI recipients.^{22, 23} This may in part be due to the items and domains in the NCIQ being selected by expert opinion rather than based on CI users’ experiences.

A new HR-QOL was recently developed, Cochlear Implant Quality of Life-35 (CIQOL) Profile instrument, to address the shortcomings of the NCIQ and incorporate the rigorous guidelines set by the NIH on patient-reported outcome measures (PROMs).²⁴ This metric was designed based on information gathered from focus groups and subsequent thematic analysis. By incorporating these additional domains, the CIQOL is thought to provide a more comprehensive evaluation of a user’s HR-QOL. The CIQOL has only been recently available and all the published work to date has focused on development and validation of the instrument.^23–25 Furthermore, no study has evaluated the change in preoperative to postoperative CIQOL (ΔCIQOL).

The first aim was to prospectively measure the impact of cochlear implantation on CIQOL in the short-term (i.e., 6-months post-implantation), thereby adding external validity to the novel HR-QOL metric and understanding the change from preoperative to post-activation in CI users. The second aim was to examine how change in CIQOL relates to patients’ demographics, audiologic factors, and a general QOL measure (the SF-36 surveys) that was not designed specifically for hearing loss or for evaluating treatment interventions related to hearing loss/CI. Understanding the short-term QOL measure is important as this is predictive of long-term QOL²⁶ and is critical in managing patient expectations.

MATERIALS AND METHODS:

Study Participants and Study Design

The participants (age, 23 – 96 years) were implanted consecutively from April 2020 to June 2021 (Table 1). Inclusion criteria included those patients who agreed to participate at candidacy and completed follow-up questionnaires and speech-perception testing. Exclusion criteria included those with single-sided deafness, prelingual onset of hearing loss, and known cognitive impairment. Individuals were recruited at the time of CI candidacy evaluation to respond to HR-QOL questionnaires. Audiologic testing was performed at candidacy and at 6-months post-activation. At the time of candidacy evaluation, the Montreal Cognitive Assessment (MoCA), a rapid cognitive screening assessment, was administered, and the comorbidity data in the form of the overall Adult Comorbidity Evaluation-27 (ACE-27) score²⁷ was calculated based on anesthesia preoperative assessment notes.

Table 1:

Demographic, cognitive health status, and physical health status of study sample.

Characteristic	No. (%) or Mean ± SD (Range, Min-Max)
Age (years) at implantation	68.6 ± 14.9 (23–96)
Male	25 (67.6)
Marital Status
Married/domestic partnership	31 (83.8)
Not married/no domestic partnership	6 (16.2)
Tobacco Use
Never	20 (54.1)
Former	12 (32.4)
Current	5 (13.5)
Alcohol Use
Yes	27 (73.0)
No	10 (27.0)
MoCA Scorea	24.3 ± 2.8 (19–29)
ACE-27 Comorbidity Scoreb
0	10 (27.0)
1	18 (48.6)
2	5 (13.5)
3	4 (10.8)

^aMontreal Cognitive Assessment Score, rapid cognitive screening instrument where scores of 26 and higher indicate normal cognitive, whereas scores of 25 and lower indicate possible cognitive impairment;

^bAdult Comorbidity Evaluation-27

Of the 37 patients included within the study, 67.6% (25/37) were male with a mean age of 68.6 ± 14.9 at the time of surgery. The average MoCA score for the cohort was 24.3 ± 2.8 and most patients received a perimodiolar device (31/37, 83.8%).

HR-QOL Measures

The CIQOL and short form survey (SF-36) were administered at the time of CI candidacy evaluation (preoperative) and at 6-months post-activation. Subjects were allowed to complete the questionnaires at home or in the office, which were self-administered with no time limit.

The CIQOL is a CI-specific QOL instrument designed to provide psychometrically sound and efficient measures that can be used to assess the QOL in adult CI users both in a clinical and research setting (https://medicine.musc.edu/departments/otolaryngology/research/cochlear-implant/instruments). The CIQOL Profile consists of 35 items in 6 psychometrically sound domains (communication, emotional, entertainment, environment, listening effort, and social) and a global domain.^{23, 24} Subjects respond to each item using a 5-point Likert scale, where higher scores are correlated with better HR-QOL. The global domain provides one score that is meant to represent a combination of all the domains. Scores are computed by calculating a sum of each domain’s response and using score conversion tables previously derived by item response theory (https://medicine.musc.edu/departments/otolaryngology/research/cochlear-implant/instruments).

The 36-item SF-36 surveys health status by assessing eight health domains: (1) limitations in social activities because of health problems, (2) limitations in social activities because of physical or emotional problems; (3) limitations in usual role activities because of physical health problems; (4) bodily pain; (5) general mental health; (6) limitations in usual role activities because of emotional problems, (7) vitality, and (8) general health perceptions. It was designed for use in clinical practice and research, health policy evaluations, and general population surveys.²⁸ It was not designed specifically for hearing loss or for evaluating treatment interventions related to hearing loss but has previously been evaluated in the context of CIs. It was chosen to be used in the present study to evaluate whether there was a change in health that could explain a change in CIQOL.^{8, 29, 30} The scores of the 36 questions (items) are coded and transformed into a scale from 0 to 100, where a higher score indicates a better self-perception of health.

Audiologic Measures

All study participants underwent comprehensive audiometric evaluation by a licensed audiologist prior to surgery in a double-walled sound attenuating booth. All test stimuli were presented through a loudspeaker (JBL; model LSR32) at ear-level at 0° azimuth and 1.5 meters from the center of the subject’s head. Preoperative audiometric evaluation included audiometric thresholds from 125- to 8000-Hz using both unaided pure tone audiometry and free field aided pure tone audiometry. Speech-perception testing was presented at 60 dB HL in the in the soundfield (0-degree azimuth) at candidacy and at 6-months of listening experience (CI-alone condition) using the Consonant-Vowel Nucleus-Consonant (CNC) word test in Quiet, AzBio in Quiet, and AzBio +10 dB signal-to-noise ratio (SNR) as previously described.^{31, 32} Preoperative word discrimination was assessed using the Northwestern University Auditory Test No. 6 (NU-6). The words are spoken by a male talker (recorded voice) with mid-western American dialect. The contralateral ear was masked using a headphone (DT48; Beyerdynamic) utilizing 40-dB suprathreshold hearing; this was done using narrowband noise and white noise for pure tone audiometry testing and speech-perception testing, respectively.

Data Analysis

The analysis here was exploratory in nature and corrections for multiple comparisons were not performed. To address the first aim (prospectively measure the impact of cochlear implantation on CIQOL), after confirmation of normality using Shapiro-Wilk test, a matched-pairs t-test was performed to compare preoperative to 6-months post-activation CIQOL scores for global and each subdomain. A Pearson correlation was determined between preoperative and 6-months post-activation CIQOL scores for only the global domain. To address the second aim (to examine how patient demographics and audiologic factors relate to the change in HR-QOL), the ΔCIQOL from preop to 6-months post-activation was compared to demographic and audiologic variables using bivariate analysis for continuous variables and one-way analysis of variance (ANOVA) for categorical variables. Significant variables were used to build a multivariable linear regression to evaluate both demographic and audiologic variables and their association with ΔCIQOL for each subdomain. A hierarchical model approach with each of the variables added one at a time was used to evaluate incremental role of each demographic or audiologic variable in predicting ΔCIQOL. A matched-pairs t-test was performed to compare preoperative to 6-months post-activation SF-36 scores. The ΔSF-36 was also included within the multivariate regression to determine if it explained any variance in ΔCIQOL. Analyses were performed with SPSS 27 for Windows (IBM Corp, Armonk, New York, USA). Alpha level for all statistical tests was set at 0.05 and were two-tailed.

The research protocol was approved by the IRB at Washington University (IRB #201911035).

RESULTS:

Thirty-seven patients that were implanted completed both preoperative and 6-month post-activation HR-QOL questionnaires between January 2020 and June 2021. Speech-perception performance improved significantly across all patients with a mean CNC score improvement of 45.1% (95% confidence interval, 34.1 to 56.1). The mean preoperative CNC score was 14.1 ± 13.4% and the mean 6-months post-implantation score was 55.6 ± 22.9%. Further hearing and cochlear-implanted related information are provided in Table 2.

Table 2:

Hearing and Cochlear Implant-Related Information for Sample Population

Characteristic	No. (%) or Mean ± SD (Range, Min-Max)
Duration of hearing loss (yrs)	23.6 ± 15.1 (2–58)
Duration of severe-to-profound hearing loss (yrs)	8.8 ± 13.6 (1–58)
Hearing aid use in ipsilateral ear
Yes	27 (73.0)
No	10 (27.0)
Hearing aid use in contralateral ear
Yes	30 (81.1)
No	7 (18.9)
Laterality
Left	16 (43.2)
Right	21 (56.8)
Electrode Type
Lateral wall	6 (16.2)
Perimodiolar	31 (83.8)
Preoperative PTA (dB HL)	83.6 ± 18.6 (48–120)
Postoperative FF CI-processor-attached PTA (dB HL)	23.7 ± 4.3 (16–33)
Preoperative CNC word score in quiet (%)	14.1 ± 13.4 (0–64)
Postoperative CNC word score in quiet – CI-only (%)	55.6 ± 22.9 (0–90)
Preoperative AzBio quiet score (%)	18.2 ± 19.7 (0–75)
Postoperative AzBio quiet score – CI-only (%)	66.6 ± 26.9 (0–98)
Preoperative word discrimination score, NU-6 (%)	17.3 ± 17.8 (0–60)
CI usage per day (hours)	11.0 ± 3.9 (3–21)
Hearing preserved (LFPTA < 80 dB HL)
Yes	11 (29.8)
No	26 (70.2)
Distance Traveled to Audiology (miles)	56.1 ± 60.4 (2–316)
Second side CI
Yes	2 (5.4)
No	35 (94.6)
Electro-acoustic use
Yes	5 (13.5)
No	32 (86.5)

PTA, pure tone average (500, 1000, 2000, 4000 Hz); FF, free field; CNC, consonant-nucleus-consonant; yrs, years; LFPTA, low frequency pure tone average (125, 250, 1000 Hz); NU-6, Northwestern University Auditory Test No. 6

Prospective evaluation of the impact of cochlear implantation on CIQOL (Aim 1)

Figure 1 shows the mean with 95% confidence interval for preoperative and 6-months post-activation scores for the CIQOL global score and six subdomains. The results revealed significantly higher scores at 6-months post-activation for all six domains including the global score: global, t(40) = 9.6, P < 0.0001; communication, t(40) = 11.9, P < 0.0001; emotional, t(40) = 7.1, P < 0.0001; entertainment, t(40) = 6.2, P < 0.0001; environment, t(40) = 9.2, P < 0.0001; listening effort, t(40) = 8.9, P < 0.0001; social, t(40) = 7.5, P < 0.0001. Correlation analyses using Pearson product moment correlations revealed a significant and positive correlation between preoperative and 6-months post-activation CIQOL-global score, r(40) = 0.42, P = 0.002 (Figure 2). Correlations between preoperative and 6-months post-activation for the other domains were not assessed.

Figure 1:

Average scores and 95% confidence intervals of global and subdomains of Cochlear Implant Quality of Life-35 (CIQOL) at the preoperative and 6-months post-activation test intervals. There was a statistically significant improvement from preoperative to postoperative across all domains (paired t-test).

Figure 2:

Preoperative Cochlear Implant Quality of Life-35 (CIQOL)-global has a moderate linear correlation with 6-months post-activation CIQOL-global. Line of equality is shown as dotted gray line, as all patients but two showed an improved CIQOL from preoperative to 6-months post-activation.

Relationship between CIQOL and patients’ demographics, audiologic factors, and a general QOL (non-CI/hearing specific) measure (Aim 2)

The following variables had significant correlations with ΔCIQOL: age, preoperative word discrimination (NU-6), ΔCNC (postop-preop), ACE-27 comorbidity score, and Montreal Cognitive Assessment (MoCA) score (Table 3). Table 4 summarizes the results of the best-fitting multivariable regression models for each domain. ACE-27 comorbidity score showed a significant main effect on ΔCIQOL, F(1, 38) = 3.05, P = 0.04. Both age and ΔCNC were both the most frequent and strongly correlated with ΔCIQOL across the various domains. There were strong linear correlations between ΔCIQOL and both younger age (r(40) = −0.49, P < 0.001) and ΔCNC (r(40) = 0.63, P < 0.0001; Figure 3). Two statistical outliers are denoted as blue triangles in Figure 3. One was the oldest patient (96 years old) in the cohort and had remarkable ΔCIQOL improvement despite a drop in speech-perception performance. Indeed, there were patients that had significant improvements in CNC after implantation with small changes in ΔCIQOL and vice versa, emphasizing that not all the variance in ΔCIQOL can be predicted by ΔCNC. The following variables were not significantly correlated with any ΔCIQOL domains, and were not included in the final multivariable models: marital status, gender, tobacco use, alcohol use, distance traveled to audiologic center, duration of hearing loss, duration of severe-to-profound hearing loss, hearing aid use in ipsilateral or contralateral ear, laterality, asymmetric sensorineural hearing loss, second side CI, electrode type, device manufacturer, electro-acoustic use, CI usage per day, hearing preservation, postoperative free field CI-processor-attached pure tone average (dB HL; 500, 1000, 2000, 4000 Hz), and ΔSF-36. There was an insignificant change in SF-36 scores from preop to postop (SF-36 total, t(40) = 0.69, P = 0.49).

Table 3:

Correlations of ΔCIQOL-35 scores and significant demographic and audiologic factors on bivariate analysis.

r values	Preop PTA (dB HL)	Preop Word Discrimination (%)	Δ CNC (postop-preop)	Age (years)	MoCA Score
Global	** 0.54	** −0.64	** 0.76	** −0.62	** −0.58
Communication	* 0.41	* −0.50	** 0.65	** −0.70	* −0.45
Emotional	** 0.43	** −0.58	** 0.57	−0.38	* −0.58
Entertainment	0.33	−0.36	0.25	−0.43	* −0.51
Environment	** 0.51	** −0.62	** 0.63	* −0.45	* −0.51
Listening Effort	** 0.59	* −0.53	** 0.82	** −0.59	** −0.73
Social	0.24	* −0.48	* 0.64	−0.28	* −0.55

^*Values have P value < 0.05

^**Values have P value < 0.01

PTA, pure tone average (500, 1000, 2000, 4000 Hz); CNC, consonant-nucleus-consonant; MoCA, Montreal Cognitive Assessment

Table 4:

Association of demographic and audiologic factors with the ΔCIQOL-35 domains using multivariate regression.

	βa (95% CI)
Measurement	Global	Communication	Emotional	Entertainment	Environment	Listening Effort	Social
R 2	0.46	0.32	0.13	NA	NA	0.29	0.18
Variable
Intercept	17.9 (−4.2 to 39.9)	32.9 (18.6 to 47.2)	8.1 (−0.7 to 17.0)	NA	NA	19.1 (−0.26 to 38.4)	10.6 (0.7 to 20.4)
Age	−0.2 (−0.3 to −0.05)	−0.2 (−0.4 to −0.1)	NA	NA	NA	−0.2 (−0.4 to 0.1)	NA
Preoperative Word Discrimination (%)	−0.1 (−0.4 to −0.02)	NA	NA	NA	NA	NA	−0.1 (−0.4 to 0.1)
Δ CNC (postop-preop)	0.1 (0.1 to 0.2)	NA	0.2 (0 to 0.4)	NA	NA	0.2 (0.1 to 0.3)	0.2 (0.0 to 0.3)
ACE-27	NA	−3.3 (−6.6 to −0.1)	NA	NA	NA	NA	NA
MoCA	NA	NA	NA	NA	NA	NA	NA

^aThe degree of change in ΔCIQOL score for each 1-unit change in predictor variable.

NA = not applicable as this variable was not included in final regression model

CNC, consonant-nucleus-consonant; ACE-27, Adult comorbidity evaluation-27; MoCA, Montreal Cognitive Assessment

Figure 3:

CIQOL-global change (ΔCIQOL-Global) from preoperative to 6-months post-activation plotted against (a) change in CNC word score from preoperative to 6-months post-activation and (b) age at implantation. Correlation coefficients are provided. Solid line represents the line of best fit. Blue triangles denote statistical outliers. Open circles in panel A represent patients older than 80 years old. CIQOL indicates Cochlear Implant Quality of Life; CNC, consonant-nucleus-consonant.

Despite including a significant number of demographic and audiologic variables into our analysis, only ΔCNC and age had the largest correlations across the various subdomains. The coefficients of multiple determination (R²) varied widely across the domains (Table 4) using multivariate modeling, where the proportion of the variance in ΔCIQOL scores predicted by the independent variables ranged from 12% (emotional domain) to 46% (global). For ΔCIQOL-global multivariate modeling, ΔCNC was first added as it had the greatest impact upon the prediction of ΔCIQOL-global (R² = 0.39, p < 0.0001). Next, the addition of age was able to slightly improve the model (R² = 0.46, p < 0.0001). Additionally, there was no significant multivariate model for the entertainment and environment domain.

There were seven patients age >80 yrs in this study (Figure 3). Two of these patients had a negative ΔCIQOL, or a decline in CIQOL at 6-months post-activation. Both of these patients lost their residual hearing and were only using their device 4–6 hours/day as they were gaining limited benefit, which was reflected in their ΔCNC scores (−28% and +2%). Four of the other patients age >80 yrs were also gaining limited benefit with the device, which was reflected with a low ΔCIQOL and below average ΔCNC compared to the study population. Only one subject older than 80 yrs had a large positive ΔCIQOL with a ΔCNC of 55%, which was above the average ΔCNC of all subjects.

DISCUSSION:

As the importance of QOL and using PROMs is becoming increasingly evident, a greater understanding of the factors that contribute to QOL is necessary to improve patient care. Although most patients achieve an improvement in HR-QOL after cochlear implantation, factors that affect HR-QOL in this population are largely unexplained. Prior studies have revealed certain demographic and audiologic factors that affect HR-QOL after implantation such as bilateral CIs, higher household income, being employed, and longer duration of hearing loss prior to implantation.^{7, 14, 17, 18, 25} Even with robust datasets using multivariate modeling and various HR-QOL metrics, prior studies have shown that these factors are only able to account for up to 19% of the variability in HR-QOL after implantation.^{7, 14, 17, 18, 25}

The present study was performed to understand the change in HR-QOL after implantation using a novel CI-specific metric, CIQOL, as well as to identify factors that could serve as predictors of improvement in CIQOL. To our knowledge, this study is the first to assess the change in CIQOL from preoperative and postoperative as prior studies have focused exclusively on understanding postoperative CIQOL.

Improvement in CIQOL after Implantation

This study showed an increase in all six domains including the global domain of the CIQOL after implantation (Figure 1). Previous studies have examined the change in NCIQ after implantation and have also shown increased scores across all domains using this disease-specific HR-QOL metric.^{6, 19–21} We then explored the relationship between preoperative and postoperative CIQOL-global and found these two variables to be highly correlated (r = 0.49). Patients with higher preoperative CIQOL scores are more likely to achieve better postoperative CIQOL scores. This was imperative to establish as much of the previous work has focused on predicting postoperative CIQOL without incorporating preoperative CIQOL scores.

Demographic and Audiologic Factors Predictive of ΔCIQOL

In the present study, only two factors were found to predict better ΔCIQOL scores for this group of adults with CIs: ΔCNC and younger age (Figure 3). This finding is in line with the fact that better speech perception should directly improve a CI user’s ability to interact and engage socially. These results are in contrast with the weak correlations shown between speech recognition ability after implantation and QOL in prior studies^{7, 14, 17, 18, 33} and similar to studies showing a positive correlation^{20, 34–36}. Of note, there were patients that had minimal improvement in CIQOL with large increases in CNC and vice versa. Including the CIQOL as a standard component for pre-implantation candidacy evaluation and post-implantation follow-up may show benefit or concerns in patients where CNC is not in line with the patient’s self-perception of their HR-QOL. The fact that younger patients have larger improvements in CIQOL is also in contrast to prior studies^{3, 25, 37} where age has not been shown as a significant predictor of HR-QOL. A few studies have shown a positive correlation between younger age and HR-QOL.^{20, 38–40} Younger age has often been identified as an important predictor of QOL in the general QOL literature, especially among the elderly who often have functional decline.⁴¹ The functional status, comorbidities, and socioeconomic conditions likely contribute to the declining QOL in the elderly. Therefore, it is not surprising that age likely serves as a surrogate measure of these conditions and is a strong predictor of ΔCIQOL. Of note, ACE-27 was not correlated with ΔCIQOL in the multivariate model which one would expect to be a better surrogate than age to assess the functional status at the time of implantation. Future studies will need to utilize different comorbidity indices with larger datasets to assess the impact of functional status on CIQOL.

While this study does show an improvement in CIQOL in patients who achieve better speech-perception scores and are younger in age, only 46% of the variance in ΔCIQOL-global is explained by demographic and audiologic variables. For the emotional, entertainment, environment, and social domains, ≤19% of the variance can be explained by patient-related or hearing-related factors as shown in this study and McRackan et al²⁵. These results suggest that a significant portion of the variability in a patient’s ΔCIQOL score is accounted for by variables that are not assessed as they remain unknown. Future studies will need to explore cochlear implantation outcome expectations, regret, cognition, among others, which may significantly contribute to ΔCIQOL. Understanding the variability in ΔCIQOL will be critical in developing interventions to improve CI outcomes.

Overall, ΔCNC is the most important predictor of ΔCIQOL, as it explains 39% of the variance. The addition of age improves the prediction of ΔCIQOL slightly, explaining 46% of the variability. The data here suggests that by understanding predictors of speech-perception performance using the CI, we may be able to more accurately predict which patients will obtain a larger ΔCIQOL. Recent studies have shown the utility of electrocochleography, cognitive testing, electrode location, and various other demographic and audiometric factors to predict postoperative CI performance.^42–47 A greater understanding of these factors may be critical for understanding which patients will receive a large ΔCIQOL.

CIQOL and SF-36

The comparison performed in the present study between the CIQOL and the SF-36 was not one of the primary study aims, but its results deserve attention. These findings include: an insignificant change in SF-36 scores from pre-implantation to 6-months post-implant activation (in line with previous studies^{29, 37}), significant positive correlation between pre-CI and 6-months post-CI activation CIQOL-global score (despite the very short post-CI follow-up time and the relatively large number of elderly individuals included in the study for whom a longer experience with CI is needed to gain benefit^{48, 49}), and lack of correlation between the pre- to post-implantation ΔCIQOL score and pre- to post-implantation ΔSF-36 scores. The SF-36 was not developed to evaluate functional impact of hearing loss. Yet, in the absence of specific hearing/CI specific tools, researchers have tended to use it (and other similar questionnaires) for evaluating pre-to post-implantation QOL evaluation^{50, 51}. This practice could mislead readers to believe that cochlear implantation has only negligible effect on cochlear implantees’ QOL. The accumulating clinical experience of CI professionals working with generic, post-implantation QOL improvement motivated the development of CI-focused QOL questionnaire, the NCIQ. To overcome the shortcomings of the NCIQ questionnaire, the CIQOL questionnaire was developed based on the experiences of cochlear implantees and incorporation of the rigorous guidelines set by the NIH on PROMs.^23–25 The results of the present study elucidate the advantage of the CIQOL questionnaire over previous, less CI-focused tools.

For the SF-36, there was no significant change from preoperative to postoperative across all the subdomains and total score. These results are consistent with previous studies^{29, 37} evaluating the SF-36 in cochlear implantation as this questionnaire was not developed to evaluate the functional impact of hearing loss and is likely to underscore the QOL benefits after implantation. This was also important in evaluating whether there was a significant difference in the patient’s perception of their health, as evaluated by the SF-36 from preoperative to 6-months post-implantation that could explain the ΔCIQOL.

Limitations

While this study can account for a significant portion of the variability in ΔCIQOL and overcomes some of the limitations of prior studies using QOL, there are several notable limitations. Since the study refers to a very short post-CI experience while re-acquirement of lost hearing-related functions and activities (many of them determine QOL) is a lifelong process, future studies must review the long-term impact of implantation on QOL. The limited sample size and age variability at implantation makes generalization of these findings challenging as larger numbers and greater age variability may have altered the results or at least would have provided greater confidence in the conclusions of these findings. The analysis presented here was exploratory and was not powered to test all significant predictors of CIQOL. Therefore, the variables that were found to not be significantly correlated with CIQOL may still be significant predictors if the study was adequately powered to detect them. Thus, the results and conclusions of the study must be viewed as preliminary.

CONCLUSIONS:

Similar to improvements in speech-perception performance, nearly all CI recipients obtain significant gains for HR-QOL as measured by CIQOL. HR-QOL has recently been noted as an important outcome measure in CI recipients to improve patient care, however, the factors that influence HR-QOL are largely understudied. There were strong linear correlations between improvement in CIQOL after implantation and both younger age at implantation and post-implantation improvement in CNC. Results here demonstrate the importance of considering both the baseline QOL score and speech-perception testing in future studies, as this is likely to impact understanding of predictors of HR-QOL. Post-implantation improvement in CNC score in quiet and younger age at implantation predict 46% of the variability in the CIQOL-global domain improvement after implantation.