Changes in outcomes expectations during the cochlear implant evaluation process

Abstract

Objectives:

Patient expectations are a critical factor in determining cochlear implant (CI) candidacy. However, minimal data are available on how potential CI recipients develop their expectations and if expectations can be modified by providers. In addition, there is little insight into the resources patients use to inform their decision to undergo implantation. This project aims to assess (1) the role of the CI evaluation (CIE) process on patients’ expectations, (2) the extent to which patients’ pre-CI outcome expectations can be modified, (3) the information patients use to inform their expectations, and patients’ preferences for the discussion/display of potential CI outcomes.

Design:

Prospective mixed methods study of 32 adult CI patients undergoing CIEs. Outcome measures included: pre-CI Cochlear Implant Quality of Life 35 Profile scores (CIQOL-35 Profile); pre-CIE/post-CIE/day of surgery CIQOL-Expectations scores; post-CIE/day of surgery Decisional Conflict Scale (DCS) scores; and pre-CI aided CNC word and AzBio sentence scores. Thematic analyses of key informant interviews with 19 potential CI recipients were also performed.

Results:

In aim 1, CI CIQOL-Expectation domain scores remained essentially unchanged following the CIE when averaged across all participants (d=0.01 to 0.17). However, changes in expectations were observed for many participants at the individual level. Regarding the second aim, participants with higher pre-CIE expectations showed a decrease in expectations following the CIE for all CIQOL domains except emotional and social (d=−0.27 to −0.77). In contrast, the only significant change in participants with lower expectations was an increase in expectations in the environment score from pre-CIE to the day of surgery (d= 0.76). Expectations remained essentially unchanged or continued to change in the same direction between the post-CIE and the day of surgery, narrowing the gap between participants with higher and lower expectations. Overall, participants demonstrated low overall conflict related to their decision to proceed with cochlear implantation (mean DCS 11.4 post-CIE and 14.2 time of surgery out of 100) but DCS scores were higher for participants with lower pre-CIE expectations (d= 0.71). In aim three, key informant interviews demonstrated no differences between the low and high expectation cohorts regarding resources used to develop their perception of CI outcomes. Potential CI recipients placed high value in talking with patients who had previously received a CI, and preferred discussing CI functional abilities via clinical vignettes described in the CIQOL Functional Staging System rather than by discussing speech recognition or CIQOL-35 Profile scores.

Conclusions:

The results of the current study suggest that, although overall expectations averaged across the cohort remained essentially unchanged, individual participants’ pre-CI expectations can be modified and there is value in measuring these expectations using the CIQOL-Expectations tool to determine if they are realistic. This information can then be utilized during personalized counselling to present a more accurate representation of likely CI outcomes for each patient. Discussions between potential CI recipients and current CI users may also provide valuable information to inform their expectations. In addition, communicating potential CI benefits using CIQOL functional stages and associated clinical vignettes may result in more realistic patient expectations and support shared decision-making related to CI surgery.

Level of Evidence - III

IRB Pro00113247, approved 8/21/2021, (Medical University of South Carolina)

INTRODUCTION

It is estimated that as many as 1.2 million Americans with hearing loss could benefit from a cochlear implant (CI) (Holder et al., 2018). Since its FDA approval nearly 40 years ago, cochlear implantation remains the gold standard for treatment of severe sensorineural hearing loss in patients who no longer benefit from hearing aids. To date, approximately half a million CIs have been performed in the United States (Zeng, 2022), with patients experiencing improved speech recognition, communication ability, and psychosocial function (Bosdriesz et al., 2018; Heutink et al., 2021).

Despite these known benefits, CI outcomes are not always effectively conveyed to potential recipients during the CI counseling process. This may result in the patient developing expectations that do not appropriately reflect what they are likely to experience functionally after implantation (Fabie et al., 2023). While research has focused on outcomes such as speech recognition and quality of life after CI activation (McRackan et al., 2018; Schafer et al., 2021), little attention has been given to the pre-implantation evaluation period and its critical components such as the factors individuals consider when making the decision to receive a CI or what resources they utilize to form their expectations (Illg et al., 2022).

Patient expectations play a crucial role in medical interventions, and in a variety of surgical procedures. It has been established that greater alignment between pre-treatment expectations and post-treatment outcomes is associated with higher levels of satisfaction, better compliance with prescribed treatment plans, and improved patient reported outcomes (Thompson & Sunol, 1995). However, as research in other medical interventions has shown, patient expectations often exceed realized outcomes leading to a discrepancy between expected and actual outcomes, which can result in both decreased satisfaction with an intervention as well as lower quality of life (Witiw et al., 2018).

Preliminary evidence from research into the role of expectations in CI patients suggests that patients’ expectations before implantation may influence their post-operative quality of life and functional outcomes (McRackan et al., 2021). This potential relationship between preoperative expectations and post-operative outcomes in CI patients is important to elucidate further because it presents an opportunity for providers to more effectively counsel patients before their implantation so they may have better results after implantation. Therefore, it is first necessary to assess the role of the CI evaluation (CIE) process on patients’ expectations and determine the extent to which patient expectations are modifiable and when during the CIE process they may be modified. We also aim to identify the sources of information patients use to inform their expectations and how they would like their potential outcomes discussed so providers can appropriately counsel patients and modify expectations to ensure they are reasonable. In addition, measuring a patient’s level of conflict with their decision to continue with implantation at multiple points in the evaluation process can provide a better understanding of where possible interventions can be made to ensure patient counseling is appropriate and provide the information needed to make a shared decision.

In order to achieve these goals, we tracked patient expectations and levels of decisional conflict at multiple stages of the CI evaluation process to gain a better understanding of how each metric changes over time leading up to surgery. The aim was to first quantitatively measure changes in potential CI recipients’ expectations over the course of the CI evaluation process and identify specific timepoints where the greatest changes occur. This allows for the opportunity to assess our hypothesis that patient expectations will differ before and after the CI evaluation. Secondly, in addition to these quantitative data, key informant interviews were conducted to gather qualitative information that provides a better understanding of the resources potential CI recipients consider when forming their expectations. This gives providers additional insight so they can direct patients to appropriate resources that will contribute to developing reasonable expectations. Finally, recognizing how patients would like their postoperative outcomes discussed is a key part of the shared decision-making framework that is necessary for optimizing care and maximizing positive results for CI patients.

MATERIALS AND METHODS

A convergent parallel mixed methods research design was utilized based on concurrent data collection and separate analysis of quantitative and qualitative data and comparison of the two analyses (Figure 1). Purposeful sampling of patients with high and low expectations based on the CIQOL-Expectations Global measure was employed for key informant interviews (Palinkas, 2015). Patients undergoing evaluation for cochlear implantation at the Medical University of South Carolina from October 2021 to May 2022 were approached for recruitment in the study. Patients 18 years of age or older, with post-lingual deafness who were found to be a CI candidate were eligible for inclusion. Exclusion criteria included: patients with single sided deafness, established cognitive disorders, known retrocochlear pathology, and patients receiving a second side surgery or revision surgery. Patients interested in participating completed informed consent, and the study was approved by our Institutional Review Board.

Figure 1:

Phases of convergent parallel mixed methods research design

Demographic and audiometric data were collected for enrolled participants including age at implantation, sex, preoperative hearing aid use, duration and etiology of hearing loss, and hearing data collected during a standard CI evaluation. Duration of hearing loss before CI was defined by self-reported number of years with hearing loss before implantation. Hearing aid use before CI was defined as the participants’ self-reported active hearing aid use for most of the day at the time of the CI evaluation (yes/no).

Pure-tone average (PTA) (average air conduction thresholds at 500, 1000, and 2000 Hz) (Campbell, 1998; Schapira & Byrne, 2007), speech recognition thresholds (SRT), and consonant-nucleus-consonant word scores (CNC-W) were measured with either insert earphones (Earlink 3A) or supraaural headphones (TDH-50) per clinician preference. Pure-tone thresholds, SRT, CNC-W, CNC phoneme scores (CNC-P), AzBio sentence scores in quiet (AzBio), and AzBio sentences presented in noise (multi-talker speech babble) at +10 dB SNR (AzBio +10) scores were measured under aided conditions (Spahr et al., 2012). Aided AzBio +10 measures were obtained when individuals scored >50% on AzBio quiet. Aided speech recognition was measured with speech presented at 0 degrees azimuth at 60 dB SPL in the sound field in a sound treated room. Each ear was tested independently. Hearing aid users were tested with their personal hearing aids, and subjects who did not have appropriately fit amplification used clinic-owned hearing aids during the evaluation process. All hearing aids (personal or clinic-owned aids) were considered appropriate if responses matched NAL-NL2 targets, which were verified using real ear measurements (Byrne & Dillon, 1986). Word and sentence material came from the MSTB-2 recorded materials. Per standard audiometric procedures the presentation level for CNC or AzBio words was presented at the loudest comfortable level in an effort to achieve PB-Max. The loudest comfortable levels was obtained using monitored live voice or recorded words to determine uncomfortable loudness level and then decreasing by 5 dB.

We tracked changes in CI candidate expectations during the pre-CI period (prior to CI evaluation (CIE), after CI evaluation, and prior to CI surgery; discussed below). We utilized a natural history strategy during participant visits, so CI clinicians did not use a specific script to discuss potential CI outcomes with the participants but provided their usual clinical counseling. During the CIE process patients first meet with 1 of 6 CI audiologists and then 1 of 5 neurotologists at our institution. Audiologists and physicians were blinded regarding which patients were in the study.

During the CIE, the patient’s case history, aided and unaided audiometry, speech recognition scores, Montreal Cognitive Assessment, CIQOL-35 Profile, and CIQOL-Expectation scores are reviewed with the patient. General information about cochlear implants is discussed, including how long it takes to reach full benefit of the CI (as measured on standard speech recognition measures performed in the sound booth) on average speech recognition scores at 6-12 months post activation (based on national research publications), initial sound quality, timeline of activation and follow up appointments, and addressing any specific concerns expressed by each patient. The discussion also includes an overview of the surgery and associated surgical risks. The patient is asked about general goals with a CI; for example, using the telephone, improving communication at work, or improved speech understanding in noise.

As a part of our standard CI evaluation, participants completed the Cochlear Implant Quality of Life-35 Profile (CIQOL-35 Profile) instrument (McRackan et al., 2019) and the CIQOL-Expectations instrument (McRackan, Hand, Chidarala, & Dubno, 2022) before their CI evaluation as part of their intake CI packet that they received prior to coming to clinic. For the study participants completed the CIQOL-Expectations instrument twice more: once immediately after undergoing audiological and speech recognition testing and counseling with a CI audiologist and surgeon; and again, on the day of their CI surgery. Participants also completed the Decisional Conflict Scale (DCS) at these two time points. The DCS is a validated patient reported outcome measure (PROM) that measures patients' uncertainty in making a health-related decision (O'Connor, 1995). A score below 25 is associated with someone implementing a decision, whereas a score greater than 37.5 is associated with a decision delay or feeling unsure about implementation of a choice (O’Connor, 1993). The CIQOL-Expectations and CIQOL-35 Profile instruments use the same five response choices for each individual item. Both instruments include a global measure and six domains (communication, emotional, entertainment, environment, listening effort, and social), and each is scored on a 0- to 100- point scale, with 0 indicating the lowest expectations/functional abilities and 100 indicating the highest expectations/functional abilities. The CIQOL-Expectations Global measure was utilized to create high and low expectation groups; the Global measure provides a single score derived from items in all 6 CIQOL domains. Those participants whose CIQOL-Expectations global score was below the mean score of the overall study cohort were classified as low expectations, while those whose scores were above the cohort's mean were classified as high expectations.

After completion of the CI evaluation and expectation / CIQOL instruments, key informant interviews were conducted with 19 participants. Purposeful sampling was employed to include 8 participants with low expectations and 11 participants with high expectations based on the same parameters above. The key informant interviews were conducted using best practices including the consolidated criteria for reporting qualitative research (COREQ-32) which provides a formal reporting checklist to improve comprehensiveness of key informant interviews (Knudsen et al., 2012; Tong et al., 2007). The development and execution of the interviews and analysis of participant responses were based on classical grounded theory, which was selected because we were developing conceptual frameworks from the collected data without any preconceived theories given the lack of available research on this topic. Grounded theory is an inductive approach to collect and analyze participant data to develop a theoretical framework that describes an individual’s experience (Chun Tie et al., 2019). Interviews were conducted in person in a private room using a semi-structured interview guide and were audio recorded. After each interview, audio files were transcribed using MAXQDA software and the transcriptions were checked for accuracy by the first author. Two separate individuals coded the transcriptions of the interviews while actively keeping memos regarding the data (MAXQDA 2022, Berlin, Germany). Using grounded theory each coder independently performed a thematic analysis and developed a coding tree by identifying minor themes. After each round of interviews, the interviewer explored the identified themes further with the subsequent participants (constant comparison). We ceased enrolling patients for additional interviews when the research team determined no additional new perspectives were being obtained (data saturation). Ultimately, the initial open codes were reviewed and linked using axial coding. Next, selective coding was used to distill the minor themes down to central themes by the senior author by reviewing the 2 independent thematic analyses. Four core topics were addressed throughout the interviews: the resources participants used to develop their expectations, the methods participants prefer to use to discuss their potential expectations, the impact of the CI evaluation process on the clarity of their expectations, and suggestions for improvements that can be made to the CI evaluation process. Three formats for discussing potential CI outcomes were presented to participants: speech recognition scores, CIQOL scores (0-100 for each domain), and the CIQOL functional staging system (McRackan et al., 2022). The CIQOL functional staging system translates CIQOL-35 Profile scores into clinical vignettes to improve patient understanding of current and possible future functional abilities. Participants could indicate their preference for utilizing multiple options in the discussion of their outcomes. See supplemental digital content for full interview outline.

Statistical Methods

Descriptive statistics such as frequency, percentage, mean, standard deviation, minimum, and maximum were calculated for all outcome variables as appropriate. Differences in demographics between cohorts were compared with Fisher’s exact tests and Chi-square tests for categorical variables and with unpaired T-tests for continuous variables. Differences in participant expectations and decisional conflict between timepoints were compared with paired T-tests. A two-way repeated measures of ANOVA was used to measure the change in scores at all three timepoints, as well as to compare the differences between the low and high Expectations groups for all CIQOL-Expectations domains at each of the three timepoints. Cohen’s d was used to measure the effect size of differences for all analyses. An effect size of 0.2 to 0.49 is considered small, 0.5 to 0.79 medium, 0.8 to 1.29 large, and above 1.3 very large.(Cohen, 1988) Statistical analyses were performed using SPSS version 28 (IBM Corp., Armonk, NY).

RESULTS

Demographics

A total of 32 participants were included in the study, split evenly into cohorts of 16 individuals in both the high and low expectations groups based on pre-CI evaluation CIQOL-Expectation Global scores. The demographics of the overall study are summarized in Table 1. There were no significant differences between the low and high expectations cohorts regarding age (d=0.57, 95% CI [−0.14-1.27] ), race (d= 0.0, 95% CI [−0.69-0.69]), sex (d= 0.0, 95% CI [−0.69-0.69]), duration of hearing loss (d=0.25, 95% CI [−0.45-0.94]), etiology of hearing loss (p=0.35), pre-implantation PTA (d=0.01, 95% CI [−0.69-0.70]), pre-implantation CNC-W (d=0.18, 95% CI [−0.52-0.87]), or pre-implantation AzBio quiet (d=0.27, 95% CI [−0.42-0.97]).

Table 1:

Participant demographics

	Overall Cohort	High Expectations	Low Expectations
Age at implantation, Years (mean ± SD; range)	63.7 ± 16.7 (27-88)	68.3 (14.9)	59.0 (17.6)
Sex
Male (N, %)	15 (46.9%)	7 (43.8%)	8 (50.0%)
Female (N, %)	17 (53.1%)	9 (56.2%)	8 (50.0%)
Race
Black (N, %)	5 (15.6%)	2 (12.5%)	3 (18.8%)
White (N, %)	27 (84.4%)	14 (87.5%)	13 (81.2%)
Duration of hearing loss prior to CI, Years (mean ± SD; range)	22.6 ± 18.2 (.5-65)	24.9 (19.7)	20.4 (17.0)
Active hearing aid user prior to CI, Yes (N, %)	26 (81%)	12 (75%)	14 (88%)
Bilateral (N, %)	20 (62.5%)	7 (43.8%)	13 (81.2%)
Unilateral (N, %)	6 (18.8%)	5 (31.2%)	1 (6.3%)
No hearing aid use (N, %)	6 (18.8%)	4 (25.0%)	2 (12.5%)
Pre-CI PTA, dB (mean, ± SD)	88.4 (17.6)	88.4 (17.4)	88.3 (17.4)
Pre-CI CNC-Word, % Correct (mean, ± SD)	11.8 (14.9)	13.1 (16.1)	10.4 (13.9)
Pre-CI AzBio quiet, % Correct (mean, ± SD)	14.0 (19.0)	11.4 (15.4)	16.7 (22.3)

Changes in Expectations over Time

Reporting of the average data obscures large individual changes that occurred in participants’ CIQOL-Expectations scores from pre-CI evaluation to the time of CI surgery counseling. Figure 2 presents the change in Expectation scores from pre-CI evaluation to the day of surgery scores for individual participants. This timeframe was selected because our analysis showed the majority of change in Expectations scores occurred between these two points in the pre-CI process. Here, a negative score means the Expectations decreased from pre-CI evaluation to the day of surgery. The individual results demonstrate that most of the participants in the high expectations group experienced decreases in all scores from pre-CIE to the day of surgery. In contrast, most of the participants in the low expectations group experienced either increased or no change in their scores during the same time period.

Figure 2: CIQOL-Expectation score changes for individual participants.

Blue bars represent participants in the low expectations group and orange bars represent participants in the high expectations group. Scores on the y-axis are differences between expectations at the day of surgery and expectations at the pre-CI evaluation. A positive score means expectations increased from before the first CI evaluation to the time of surgery. Each number on the x-axis represents a different participant. The order was determined by starting with the most positive change in the communication domain and continuing on to the most negative. The numbering system was kept consistent across all other domains and global score.

For the overall cohort, there were no significant changes in the CIQOL-Expectations global or domain scores between any of the three timepoints. For the high expectations group, there was a significant decrease in the global score, meaning their expectations decreased, as well as all domains except social from pre-CI evaluation to the day of surgery (range d= 0.55-1.14). The time between the CIE and surgery was, on average 53.5 days (SD±26.5). This variability is expected based on schedule availability and patients’ individual preferences and life circumstances. Specifically looking at changes from pre- to post-CI evaluation, the high expectations group showed a significant decrease in scores for the environment (d= 0.73 95% CI [0.15-1.36]) and listening effort domains (d= 0.59 95% CI [0.07-1.16]), as well as the global score (d= 0.48 95% CI [0.01-1.01]). Moreover, there was a significant decrease in the entertainment domain from post-CI evaluation to the day of surgery (d= 0.67 95% CI [0.14-1.24]). This demonstrates expectations may be altered during the CI evaluation and may continue to decrease after the CI evaluation in those who start with high expectations. In contrast, the only significant change in the low expectations group was an increase in the environment score from pre-CI evaluation to the day of surgery (d= 0.76 95% CI [0.23-1.36]). This demonstrates expectations are less likely to change over the course of the CI evaluation process in those who start with low expectations. Given the lack of differences in responses on the CIQOL-Expectations instrument, additional analyses were not performed. Figure 3 illustrates the trends in scores for each domain at all three timepoints for both the low- and high-expectation groups.

Figure 3:

The CIQOL-Expectation scores with standard error bars for each domain and global score at three timepoints for both the low and high expectations groups. The dashed blue line represents the low expectations group, and the boxed orange line represents the high expectations group.

Decisional Conflict

Decisional conflict was measured post-CI evaluation and on the day of surgery for each participant. No participants in the low expectations group met the threshold for significant decisional conflict post-CI evaluation (score >37.5). However, there were two participants (12.5%) from the low expectations group who met the threshold for significant decisional conflict at the day of surgery. In the high expectations group, no participants met the threshold for significant decisional conflict at either timepoint. Figure 4 shows the average level of decisional conflict for both the low and high expectation groups. The overall level of decisional conflict was 11.4 post-CI evaluation compared to 14.2 at day of surgery (d=0.21, 95% CI [0.14-0.29]). For each of the expectation cohorts, the scores post-CI evaluation and at day of surgery were 15.6 and 16.9 for the low expectations group (d=0.09, 95%CI [0.01-0.17]) and 7.2 and 11.4 for the high expectations group (d=0.39, 95%CI [0.23-0.56]). Finally, the low expectation cohort had higher decisional conflict than the high expectation cohort post-CI evaluation (d=0.69, 95%CI [−0.01-1.42]) and at the day of surgery (d=0.42, 95%CI [−0.27-1.13]). Although statistically significant changes were observed, both cohorts maintained low levels of decisional conflict at both periods and few participants in either group ever met the threshold of significant decisional conflict.

Figure 4:

Decisional conflict scores for each expectations group post-CI evaluation and on the day of surgery. A greater score on the y-axis indicates a participant’s increased level of decisional conflict. The dashed blue line represents the decisional conflict scores for the low expectations group, and the boxed orange line represents the decisional conflict scores for the high expectations group.

Qualitative Analysis

Nineteen participants of the original cohort of thirty-two participated in interviews, with 8 in the low expectations cohort and 11 in the high expectations cohort. The demographics of the interviewees can be found in Table 2. There were no significant differences as demonstrated by the confidence intervals between overall study cohort and those that participated in interviews in terms of age (d=0.12, 95% CI [−0.45-0.69] ), race (d= 0.14, 95% CI [−0.56-0.83]), sex (d= 0.10, 95% CI [−0.59-0.79]), duration of hearing loss (d=0.04, 95% CI [−0.52-0.61]), pre-implantation PTA (d=0.02, 95% CI [−0.55-0.59]), pre-implantation CNC-W (d=0.03, 95% CI [−0.53-0.60]), or pre-implantation AzBio quiet (d=0.17, 95% CI [−0.40-0.74]).

Table 2:

Demographics of interview participants

Participant Demographics
Age at implantation, Years (mean ± SD; range)	61.7 (± 17.0)
Sex
Male (n, %)	8 (42.0%)
Female (n, %)	11 (58.0%)
Race
Black (n, %)	2 (10.5%)
White (n, %)	17 (89.5%)
Duration of hearing loss prior to CI, Years (mean years ± SD; range)	23.4 (± 18.8)
Active hearing aid user prior to CI, Yes (n, %)	15 (78.9%)
Bilateral (N, %)	11 (57.9%)
Unilateral (N, %)	3 (15.8%)
No hearing aid use (N, %)	5 (26.3%)
Pre-CI PTA, dB (mean, ± SD)	88.0 (19.3)
Pre-CI CNC-Word, % Correct (mean, ± SD)	12.3 (15.0)
Pre-CI AzBio quiet, % Correct (mean, ± SD)	10.9 (15.7)

Throughout the interviews, participants were permitted to list any resources they utilized during the CI counseling process, and there were no significant differences between the high and low expectations cohorts in terms of resources used. Overall, the two most common resources participants identified in developing their expectations were providers involved in the evaluation process, including both audiologists and otolaryngologists (n=12, 63.2%), and current CI users (n=11, 57.9%). Participants from both the high and low expectations groups emphasized the importance of being able to speak with someone familiar with the process and have their questions answered.

“Most beneficial was [meeting someone] that actually got the implants because he was very, very informative about everything. So even though I could read about [CIs], him showing me everything made it easier for me to understand.”

In addition, members of both the high and low expectations specifically mentioned the benefits of using social media to communicate with active CI users.

“[CI users] talk about their surgeries, how their recovery is going, the different types of equipment, and if they are having equipment issues”

Over half of participants (10, 52.6%) also identified other online resources, such as YouTube videos, as a factor in the development of their expectations, though only participants from the low expectations group cited these as the most important. Other resources identified by multiple participants include speaking with trusted friends and family who are non-CI users (4, 21.1%), information mailed to participants from our CI center (3, 15.8%), speaking directly with CI manufacturer representatives (2, 10.5%), and magazine articles (2, 10.5%).

Methods for Discussing Potential Recipient Expectations

Three formats for discussing potential CI outcomes were presented to participants: the CIQOL functional staging system (McRackan et al., 2022), CIQOL scores (0-100 for each domain), and speech recognition scores. There were no significant differences between the high and low expectations cohorts in terms of preferences (d= 0.17 95% CI [−0.39-0.92]). Overall, the CIQOL functional staging system, along with the associated clinical vignettes for each stage, was by far the most popular, with 14 participants (73.6%) indicating they would prefer it be used in discussions of potential outcomes. It was cited as being the easiest to understand on multiple instances.

“It’s simple and direct and to the point. I mean, you’re here now and you want to get to there.”

The CIQOL scoring system was the second most popular (6, 31.5% indicating preference for its use as part of their outcomes discussion), though one participant mentioned the learning curve was steeper than the staging system and required extra instructions to understand how to interpret the scores. Speech recognition scores were the least popular option, with only 4 (21.1%) participants indicating a preference to have these as part of their discussion of potential outcomes, and one participant who cited this as their preference qualified their choice by saying they do not like the process of undergoing speech recognition testing. These results are summarized in Figure 5.

Figure 5:

Percentage of interview participants who indicated a preference for using each method in the discussion of their potential cochlear implant outcomes. Participants were allowed to select more than one option.

“So even though I really do hate the hearing test, I hate having to do the words and repeating, I know it’s necessary.”

In total, four participants (21.1%) indicated a preference for utilizing multiple options for discussing potential CI outcomes, with two participants selecting CIQOL staging and speech recognition, one selecting CIQOL staging and CIQOL scores, and one selecting CIQOL scores and speech recognition scores.

Impact of Evaluation Process on Clarity of Expectations

All but one participant said they have clear expectations of what to expect with their CI after going through the evaluation process, with multiple participants acknowledging they now understood their outcomes will not be perfect, but they do expect some improvement after implantation.

“I feel like what I should anticipate is gradual improvement. Now, I do not expect to have this done and I’m [hearing] like I was 12 years old, but over a period of time I expect improvement from where I am now.”

The one participant that did not state they have clear expectations after the evaluation process was in the high expectations group, stating:

“Well, I have an idea. I don’t know if I fully understand what I’m in for. You know on paper and for face value. But a part of it you can’t really know until you’re there.”

Improvements to the CI Evaluation Process

We focused on this topic to determine if changes in the CI evaluation process could result in more clear outcome expectations. Most participants stated they did not have any recommendations for improvements to the evaluation process. Participants often credited the providers they spoke with during the day for being clear, patient, and willing to answer all questions. They appreciated how much individualized attention they were given, and cited the staff as being experienced and professional.

“Nobody has rushed, everybody has taken the time to explain everything to us. It has been so different than most medical visits that we’ve ever had [where it is] like an assembly line rushing to get you in and out.”

The most common improvement that was suggested was the opportunity to speak with other CI users to understand their experience and be able to ask questions of someone who has been through the process and overcome challenges. Approximately one fifth of participants (4, 21%) stated either having the contact information of a current CI user or having them present during the evaluation process would have been beneficial to their understanding.

“If you actually had somebody here who had done it, because they know the emotions that you go through with hearing loss. So they could connect and explain, ‘look, this is what you’re going to experience. This is what you’re going to feel.’”

Other improvements that were recommended by multiple participants were digitizing the intake and patient-reported outcome forms to make them easier to fill out and shortening the length of time of the evaluation process. It was also mentioned that participants would prefer their CI evaluation be conducted over one day regardless of length instead of spreading over multiple visits because of some participants’ long commute.

DISCUSSION

Cochlear implants often result in major improvement in individuals’ lives, but unless someone undergoes this dramatic change themselves it is hard to know what to expect. The results of the current study support measuring patient expectations to direct outcome discussions with the goal of ensuring patients have an accurate depiction of living with a CI, leading to improved shared decision making. Similar to prior research (McRackan, Hand, Chidarala, & Dubno, 2022), considering only average changes in expectations for the entire cohort obscures large individual differences in changes in expectations. Moreover, the large individual differences following the CIE for many patients provide evidence that patient expectations can be modifiable. Personalized counselling can be directed to those with unrealistically high or low expectations, and the CIE appointment is an appropriate time to adjust patient expectations.

Currently there is no standardized approach to counselling CI candidates or discussing post-operative expectations. There are minimal data available to guide counselling other than average speech recognition outcomes, which can vary widely and poorly correlate with patients’ real-world experience (McRackan et al., 2018). Thus, clinicians often rely on general conversations regarding patient expectations and their personal experience with CI users without a standardized way to determine if their patients’ expectations are realistic or if their patients’ expectations are met following surgery. The CIQOL instrument suite provides a framework to understand patients’ expectations regarding their functional abilities following implantation and guide preoperative counselling. A key element of shared decision-making is patients having a clear understanding of potential outcomes of the proposed surgical intervention. This becomes especially important if there is a large mismatch between expectations and normative post-CI CIQOL-35 Profile data (McRackan, Hand, Chidarala, Velozo, et al., 2022).

Prior research has demonstrated that realistic patient expectations are one of the most important factors considered by clinicians when deciding whether or not a patient should proceed with a CI (Prentiss et al., 2020). Prior research has also shown that post-CI functional abilities fall short of pre-CI expectations for up to half of CI users (Fabie et al., 2023; Illg et al., 2022); however, the results of the current study demonstrate that patient expectations can be modified during the CIE. In a study by Harris et al, during preoperative counselling CI users were specifically interested in information regarding the quality of receptive communication after implantation, the time needed to observe improvement, and the post-implantation effort required to retrain the brain to understand speech. The majority of patients interviewed wished they had received more detailed preoperative counselling regarding what to expect as a new CI user (Harris et al., 2016). A recent study from our institution highlights the importance of improving preoperative counseling for CIs. We found that CI users with greater congruence between pre-CI expectations and post-CI real-world experiences had higher CI-related satisfaction and less regret surrounding the decision to undergo implantation. In contrast, speech recognition scores had minimal impact on satisfaction and decisional regret (Shannon et al., 2023).

The results in this study demonstrate that expectations can be modifiable during provider interactions leading up to surgery. Discussions with the audiologist and surgeon during the CI evaluation resulted in more realistic expectations for those participants with high or low expectations. For all domains but especially communication, emotion, and entertainment there was convergence of expectations by the time of surgery between high and low expectation cohorts. The largest changes in expectations occurred during the CIE, and afterward expectations continued to trend in the same direction or remained relatively stable until surgery. This demonstrates the importance of outcome discussions with providers in reducing the mismatch between pre-CI expectations and post-CI outcomes. Effective preoperative counseling may thereby improve satisfaction and minimize patient regret about the decision to undergo cochlear implantation (Shannon et al., 2023). Decisional conflict leading up to implantation was low overall across the study population but was slightly higher for participants with lower pre-CI evaluation expectations. Though non-significant, in both high and low expectation groups decisional conflict increased from the time of the CI evaluation to the day of surgery. This was likely attributable to efforts made by providers to better align overly high expectations with likely outcomes for a subset of the sample and discussing details and risks related to implantation the patient may not have considered previously.

There are currently minimal data available regarding the resources that adult patients use to inform their CI outcome expectations. Prior research in the pediatric CI population has demonstrated that health professionals, websites, social media, online CI communities, CI meeting groups, and friends or family members are the main resources for parents considering implantation for their child (Aloqaili et al., 2019). Key informant interviews in the present study revealed that potential CI recipients placed high value in speaking with other patients who had previously received a CI. Discussions with current CI users provide valuable end user information to CI candidates regarding potential CI benefits as well as challenges through real world experience. Likewise, participants found discussing CI functional abilities via clinical vignettes described in the CIQOL Functional Staging System more meaningful than speech recognition data or CIQOL-35 Profile scores. Vignettes based on ability level performing everyday tasks were reported to be more relatable and easier to gauge current hearing level and future progress than other metrics. Recently research by our department demonstrated that older patients and those with higher baseline quality of life were weakly associated with increased patient expectations (Shannon et al., submitted). However, the specific patient characteristics that may be associated with developing higher or lower expectations remains an area of ongoing investigation. Among other factors, the amount of time spent researching CIs and resources utilized, educational background, and socioeconomic status may all contribute in specific ways to raising or lowering pre-CIE expectations.

Limitations of the study include that it was performed at a single institution, a small sample size, and expectation results of the cohort were slightly higher than previously published patient data for the CIQOL-Expectations instrument (McRackan, Hand, Chidarala, & Dubno, 2022). This was thought to be related to the Hawthorn effect due to active involvement by researchers given the study’s prospective design (Landsberger, 1958). Additionally, 1 of 5 neurotologists and 2 of 6 audiologists performing adult CIEs at our institution during patient recruitment were involved in the study, which also may have introduced bias if their usual counselling was altered. However, clinicians were not made aware whether individual patients were study participants. Additionally, while several changes observed in this study were statistically significant, the clinical importance of these changes requires further investigation.

Taken together, the results of the collective analyses suggest that the CIQOL-Expectations instrument has adequate face, content, and construct validity and adequate ability to measure pre-CI outcome expectations. The CI evaluation represents a key timepoint where clinicians can identify patients with unrealistically low or high expectations and provide personalized counselling to present a more accurate representation of likely CI outcomes, with a goal of improving treatment satisfaction and minimizing decisional regret. Both audiologists and otolaryngologists play a critical role in adjusting patient expectations during the CI evaluation process. Discussions with current CI users may provide valuable end user information to CI candidates to inform their expectations and clinical vignettes from the CIQOL Functional Staging System may assist in providing real world examples to draw comparison between expectations and normative results for a given preoperative ability level. Measuring expectations and working to maximize patient understanding of likely outcomes are key steps in further advancing shared decision-making for CI surgery.