Objectives: To review evidence regarding the health-related quality of life (HRQoL) and cost-effectiveness of unilateral and bilateral cochlear implantation (CI) among children and adults with severe-to-profound hearing loss. Study Design: Narrative review. Methods: Publications related to quality of life (QoL) and costs of care in CI were acquired through searches in English-language databases. Studies were included if they had identified the HRQoL attainment, cost of care, cost-utility, or cost-effectiveness associated with CI. Results: 57 studies were critically reviewed. The QoL outcome metrics used in these articles were divided into 2 categories - generic and condition specific. In studies investigating children, many reported no significant difference in QoL attainment between CI recipients and normal-hearing peers. In adults, significant improvements in QoL after implantation and higher QoL than in their nonimplanted (hearing-aided) peers were frequently reported. Studies involving an older adult cohort reported significant improvement in QoL after implantation, which was often independent of audiological performance. Overall, the calculated cost-utility ratios consistently met the threshold of cost acceptance, indicating acceptable values for expenditures on CI. Conclusions: Considerable work has been done on the QoL attainment and health economic implications of CI. Unilateral CI across all age groups leads to reported sustained benefits in the recipients' overall and disease-specific QoL. With increased cost associated with bilateral CI, further study is needed to characterize its costs and benefits with respect to the recipients' health, well-being, and contributions to society.
Since its introduction into clinical practice, hearing care clinicians have increasingly utilized cochlear implantation (CI) to restore auditory stimulation in selected patients with advanced sensorineural hearing loss. Over the last 15 years, cochlear implant manufacturers have consistently expanded their technologies and device sophistication. In this same period, criteria for implantation in candidates of all ages have been broadened based on more refined methods of diagnosis and heightened levels of benefit observed with implantation.
The considerable direct and indirect costs associated with this intervention remain a significant barrier to patients and their families, and present an obstacle to a more widespread adoption of CI technology. However, CI also results in economically valuable quality of life (QoL) improvement [Lindemark et al., 2014; Semenov et al., 2012], enhanced academic achievement [Semenov et al., 2013], and improved vocational outcomes [Emmett and Francis, 2015; McKinnon, 2014] leading to potentially even greater clinical and nonclinical cost savings among CI recipients. As a result, the measurement of the economic benefits of implantation has become essential in making policy decisions. In the United States, one of the first economic analyses investigating CI cost-utility helped convince the Californian MediCal program to include CI in their funding schedule [Harris and Anderson, 1999]. Economic evaluation of health interventions requires the consideration of factors beyond clinical effectiveness, such as the intervention's effects on a patient's level of physical and emotional function, vitality, interpersonal communication, independence in daily living, overall satisfaction with life, and mental health.
Our objective was to perform a narrative review of the literature on the QoL attainment among CI recipients of all ages, and examine cost-effectiveness measures as related to CI. Specifically, we reviewed (1) the characteristics of the studies reporting on QoL attainment and health economic implications of CI, (2) the types of QoL measures used, (3) key results, conclusions, and limitations of the included studies, and (4) outcome differences between individuals implanted with unilateral and bilateral implants.
We designed a narrative literature review to assess the health-related QoL (HRQoL) attainment and cost-effectiveness of CI in recipients of all ages with severe to profound hearing loss. A secondary goal was to compare the health economic outcomes between unilateral and bilateral implantation. Multiple criteria were agreed upon and used in identifying studies for this review prior to inclusion or exclusion.
Original contributions and review articles from peer-reviewed publications were identified. Both prospective and retrospective study designs, and cross-sectional and longitudinal trials were included.
Studies that enrolled participants with either prelingual or postlingual onset of hearing loss in 3 age categories included children (aged less than 18 years), adults (18-65 years), and older adults (older than 65 years). No exclusions of etiology of hearing loss, duration of hearing loss, timing of CI, or duration of follow-up were made.
Identified studies with primary outcome measures reported noted:
1. HRQoL metrics utilized in CI and overall QoL associated with CI
2. Costs and savings associated with CI
3. Cost-effectiveness, cost-benefit, or cost-utility of CI
Secondary outcomes that were also identified:
1. Types of QoL metrics
2. Consideration of bilateral versus unilateral CI
Search Methods for Identification of Studies
In order to capture the breadth of contemporary literature on all relevant studies investigating CI, the main English language electronic biomedical publication databases were utilized to identify source articles. Specifically, we mined PubMed and EMBASE for classic biomedical literature, periodicals covering audiology and communication sciences (CINAHL), psychology periodicals (PsycINFO and Web of Science), and economic periodicals (EconLit). We used variations of the term “cochlear implantation” and “cost-effectiveness” to identify potential articles of interest. While there were no restrictions on the articles' country of origin, only abstracts available in English were screened for inclusion in the review.
Identified studies using the criteria above were reviewed in 2 stages. In the first stage, titles and abstracts of all retrieved articles were reviewed for relevance. Inclusion criteria were: (1) original peer-reviewed research article or literature review published in a peer-reviewed journal, (2) study abstract available in English, (3) study participants had cochlear implants or were candidates for CI, and (4) study participant ages or age at implantation were clearly identified. After article title and abstracts had been screened, the full-length articles were reviewed. Articles were accepted for review if they met the above 4 criteria and all of the following: (1) study reported on any of the primary outcomes of this review, or (2) study included outcomes of bilateral versus unilateral implantation.
For each included article, the basic characteristics of each study were extracted including details of study design, characteristics of participants, interventions, and the primary and secondary outcomes. Only data explicitly reported in the articles were included in the current review. No attempt was made to contact the investigators of included trials for any missing data. When a statistic was provided, we also examined the degree of overlap in the reported confidence intervals where appropriate. The generation of weighted averages of incremental cost-effectiveness ratios was completed by using literature-reported gains and the number of patients included in the pertinent study. If there was no overlap amongst reported statistics, this was taken to indicate the presence of heterogeneity.
Study Selection and Common Characteristics
After subjecting our initial database search to our primary exclusion criteria, 68 articles were included in the final results. Studies used both cross-sectional and longitudinal design to capture QoL and health economic outcomes following CI. The sample size of the 57 studies ranged from 8 to 908 CI recipients (Tables 1, 2, 3, 4). Additionally, there was considerable variance in the ages at the time of CI within the 3 age categories considered in this review and the duration of implant use. Most studies did not report age ranges at implantation, but some did report on the standard deviation of age at CI.
The QoL outcome metrics used in these articles could be divided into 2 categories - generic and disease specific. They were then further divided into population-validated and ad hoc QoL instruments designed for the specific use of the given trial. Fifteen of the studies focused on presenting metrics for assessing the QoL attainment of children with cochlear implants (Table 1).
Generic Instruments in Children
Eight studies reported on generic health utility metrics, which included KINDL® [Huber, 2005; Loy et al., 2010; Warner-Czyz et al., 2009, 2011], visual analog scale (VAS) [Clark et al., 2012; Sach and Barton, 2007], Quality of Well-Being Scale (QWB) [Clark et al., 2012], Child Health and Illness Profile (CHIP) child edition [Clark et al., 2012; Meserole et al., 2014], European Quality of Life Questionnaire in 5 domains (EQ5D) [Clark et al., 2012; Sach and Barton, 2007], Satisfaction with Life Scale (SWLS) [Spencer et al., 2012], and EuroQoL VAS [Sach and Barton, 2007]. A comprehensive review of these surveys is beyond the subject of the current paper and may be found elsewhere [Froberg and Kane, 1989; Zullig et al., 2010]. KINDL® is a 24-item QoL questionnaire developed in Germany and validated for use in children from 8 to 16 years of age [Ravens-Sieberer and Bullinger, 1998]. It is composed of 6 scales: physiological and psychological well-being, self-esteem, family, friends, and functioning. Both child report and parent-proxy versions are available, and normative values have been established for comparison. As is the case for all questionnaires administered to children, results are limited by the fact that very young children cannot respond, and parents serve as proxy respondents.
The CHIP child edition is a population-validated self-reported QoL instrument for children 6-11 years old [Riley et al., 2004]. It is comprised of 5 domains: satisfaction (with self and health), comfort (emotional and physical symptoms and limitations), resilience (positive activities that promote health), risk avoidance (risky behaviors that influence future health), and achievement (of social expectations in school and with peers). The QWB is a 71-item questionnaire and measures the respondent's overall health status and well-being over the previous 3 days across 4 domains: physical activities, social activities, mobility, and symptom/problem complexes [Seiber et al., 2008].
The EQ5D is a self-administered questionnaire that assesses function in 5 socially relevant domains: mobility, self-care, usual activities, pain-discomfort, and anxiety-depression. The Euro-QoL VAS is a linear visual analog scale that accompanies the EQ5D [Gusi et al., 2010]. The VAS presents a line with grid marks from 0 (“death”) to 100 and asks the respondent or a respondent's proxy to draw a line corresponding to the respondent's QoL. Lastly, the SWLS is a short 5-item instrument designed to measure global cognitive judgments of satisfaction with one's life and is not a health-specific metric [Kobau et al., 2010].
Disease-Specific Instruments in Children
Six studies reported on disease-specific QoL metrics, which include youth QoL deaf and hard of hearing edition (YQoL-DHH) [Meyer et al., 2013], Deaf Identity Scale [Spencer et al., 2012], Children with Cochlear Implants: Parental Perspectives questionnaire [Fortunato-Tavares et al., 2012; Huttunen et al., 2009], and the Nijmegen Cochlear Implant Questionnaire (NCIQ) [Necula et al., 2013]. The YQoL-DHH is a 32-item questionnaire that assesses hearing loss-related QoL across 3 domains: self-acceptance/advocacy, perceived stigma, and participation [Patrick et al., 2011]. The Deaf Identity Scale is composed of 3 subscales, including hearing identification, deaf identification, and dual identification. Each subscale consisted of 5 statements regarding the individual's desire to associate with and the individual's assumed similarity with deaf, hearing, or both groups [Weinberg and Sterritt, 1986]. The Glasgow Children's Benefit Inventory (GCBI, [Sparreboom et al., 2012]; Table 2) is used to measure and evaluate a child's health benefit retrospectively, after an otological intervention. Answers are provided on a 5-point Likert scale and converted to a 200-point scale which ranges from -100 (maximal harm) to +100 (maximal benefit). The GCBI can also be analyzed according to 4 domains, including emotion, physical health, learning, and vitality.
The CCIPP is a cochlear implant-specific closed-set questionnaire that assesses the following dimensions of a child's experience with the implant: decision to implant, process of implantation, positive effect of the implant, communication, supporting the child, self-reliance, well-being and happiness, social relationships, education, and pre- and postoperative services provided by the implant center [O'Neill et al., 2004]. The NCIQ is another CI-specific QoL instrument covering 6 subdomains: basic sound perception, advanced sound perception, speech production, self-esteem, activity and social interactions [Hinderink et al., 2000].
Ad hoc Instruments in Children
Four articles used ad hoc instruments, which are designed for the purposes of an individual study [Edwards et al., 2012; Schorr et al., 2009; Stacey et al., 2006; Warner-Czyz et al., 2011]. These questionnaires provide considerable information on a child's experience with an implant and use one or a combination of open-ended questionnaires, semistructured interviews, or quantitative instruments. Three studies used quantitative QoL questionnaires [Edwards et al., 2012; Schorr et al., 2009; Stacey et al., 2006; Warner-Czyz et al., 2011] and one used a combination qualitative and quantitative instruments [Schorr et al., 2009]. While providing considerable detail and additional insights that are not often assessed in more established QoL instruments, these questionnaires have not been population validated and are not generalizable across different study groups.
Summary of Results for Children
Overall, there was no significant difference in QoL attainment between CI recipients and their normal-hearing peers [Clark et al., 2012; Huber, 2005; Meserole et al., 2014; Meyer et al., 2013]. Factors associated with poorer QoL outcomes included use of hearing aids (rather than a CI) [Meyer et al., 2013; Necula et al., 2013], shorter experience with an implant [Loy et al., 2010], older age at implantation [Necula et al., 2013; Schorr et al., 2009; Stacey et al., 2006; Warner-Czyz et al., 2011], additional developmental comorbidities [Sach and Barton, 2007; Edwards et al., 2012], male gender [Sach and Barton, 2007], lower auditory perception [Sach and Barton, 2007], lower parental educational level [Sach and Barton, 2007], and reduced oral communication by the child [Fortunato- Tavares et al., 2012].
QoL Outcomes in Adults
Fourteen of the 57 studies focused on presenting metrics for assessing the QoL attainment of adults with cochlear implants. The studies were comprised of participants from 9 countries with a sample size ranging from 8 to 283 study participants. The mean age at implantation ranged from 43 to 67.2 years (Table 1).
Generic Instruments in Adults
Ten studies reported on generic health utility metrics, which included the Short Form-36 (SF-36) [Chung et al., 2012; Damen et al., 2007; Mo et al., 2005], Ontario Health Utilities Index (HUI) Marks II and III [Damen et al., 2007; Faber and Grøntved, 2000; Klop et al., 2007, 2008; Sparreboom et al., 2012], the EQ5D, the VAS from the EQ5D, the time trade-off technique (TTO), the Australian Quality of Life (AQoL) instrument [Hawthorne et al., 2004], the Hopkins Symptom Check List 25 (HSCL-25) [Mo et al., 2005], the Psychological General Well-Being (PGWB) Index [Rembar et al., 2009], and the Glasgow Benefit Inventory (GBI) [Lassaletta et al., 2006]. Although none of these measures are specific to hearing or the effects of CI, some do incorporate hearing ability as part of the measure, as discussed below.
The SF-36 health questionnaire is a generic, validated, 36-item tool developed to evaluate HRQoL of a medical or surgical intervention. The multi-item scale assesses 8 different health domains, including physical functioning, role limitations due to physical problems, social functioning, bodily pain, general mental health, role limitations due to emotional problems, vitality, and general health perceptions [Chung et al., 2012]. The PGWB is a questionnaire that has been used to compare general psychological well-being of CI recipients with the general population. It has also been used to study general psychological well-being in other patient groups treated in audiology practices [Rembar et al., 2012].
The HUI3 is the latest version of the Canadian multiattribute utility instrument. It scores in 8 dimensions, including vision, speech, hearing, ambulation, dexterity, emotion, cognition, and pain. Although a generic instrument, it does include an assessment of the impact of sensory impairments [Kuthubutheen et al., 2015]. The EQ5D and associated VAS are described above. The TTO technique requires subjects to hypothetically trade years of life with normal hearing while maintaining the same health status in other aspects. In one example, subjects were instructed to choose between living in a state of hearing loss for 30 years or forfeiting a portion of the next 30 years beyond their current age for normal hearing [Chung et al., 2012]. The TTO is noted to be a difficult task for subjects to complete. Of the 4 instruments - HUI3, EQ5D, VAS, and TTO - that have been used in many recent studies, HUI3 is the only tool that measures health utility directly and has been shown to be the most conservative measure of health utility [Chung et al., 2012].
The AQoL measure is a multiattribute utility HRQoL instrument that comprises 5 dimensions measuring illness, social independence, social relationships, sensory abilities, and psychological well-being [Hawthorne et al., 2004]. The HSCL-25 is a self-report, psychological inventory with 58 items, scoring on the 5 underlying dimensions of somatization, obsessive-compulsive, interpersonal sensitivity, anxiety and depression. It has been shown to be a good indicator of emotional distress [Mo et al., 2005]. The GBI is a questionnaire developed to assess communication abilities and QoL after otolaryngology procedures [Lassaletta et al., 2006]. It has been validated over a wide range of procedures and found to be sensitive to changes in health after the procedure. QoL is measured in 3 domains, social, general, and physical.
Disease-Specific Instruments in Adults
Nine studies reported on disease-specific QoL instruments in adults, most specific to the benefits of CI. These include the Nijmegen Cochlear Implant Questionnaire (NCIQ) [Cohen et al., 2004; Damen et al., 2007; Fuller et al., 2013; Klop et al., 2007; Looi et al., 2011; Olze et al., 2011 ], the Cochlear Implant Function Index (CIFI) [Fuller et al., 2013], the Speech, Spatial and Qualities Questionnaire (SSQ) [Fuller et al., 2013], Patient Quality of Life Form (PQLF) [Mo et al., 2005], Index Relative Questionnaire Form (IRQF) [Mo et al., 2005], the Hearing Participation Scale (HPS) [Hawthorne et al., 2004], and the Complete Intelligibility Spatiality Quality (CISQ) [Looi et al., 2011].
The NCIQ is a validated, CI-specific HRQoL questionnaire, composed of 3 categories in 6 domains: physical functioning: sound perception - basic, sound perception - advanced, speech production; social functioning: activity, social functioning; and psychological functioning: self-esteem. The CIFI is a tool developed to assess the auditory related function of CI users, in 6 fields of auditory functioning: reliance on visual assistance, telephone use, communication at work, “hearing” in noise, hearing in groups, and hearing in large room settings. The SSQ is a validated environmental and spatial hearing questionnaire that was developed to quantify the abilities of hearing-impaired people and CI users, particularly for speech perception and spatial hearing [Fuller et al., 2013].
The PQLF and IRQF are related instruments, both developed at the House Ear Institute as disease-specific instruments to assess patients' ability to cope with their hearing loss, adaptation to the CI, and emotional alterations since implantation. The PQLF is completed by the hearing-impaired individual, while the IRQF reflects a relative's experience with that individual, including the effect of the hearing handicap on their daily activities, and the hearing-impaired individual's adaptation to the implant [Mo et al., 2005].
The HPS is an 11-item instrument that measures self-esteem, social handicap, and hearing handicap [Hawthorne et al., 2004]. The CISQ was created to enable a rapid and simple instrument to evaluate the benefits of a hearing aid. This 36-item questionnaire assesses an individual's hearing abilities relative to spatiality and quality of signal, intelligibility in silence, background noise intelligibility, averseness, and reverberation [Looi et al., 2011].
Ad hoc Instruments in Adults
Four reports included ad hoc QoL instruments [Fuller et al., 2013; Lassaletta et al., 2006; Rembar et al., 2009; Zhao et al., 2008]. Of these, 2 studies used quantitative QoL questionnaires [Fuller et al., 2013; Zhao et al., 2008], 1 used a qualitative instrument [Rembar et al., 2009], and 1 used a combination qualitative and quantitative instrument [Lassaletta et al., 2006].
Summary of Findings for Adults
Overall, implanted adults showed significant improvement in QoL after implantation [Chung et al., 2012; Hawthorne et al., 2004; Klop et al., 2007, 2008; Mo et al., 2005; Olze et al., 2011; Zhao et al., 2008] and attained significantly higher QoL than their nonimplanted (hearing-aided) peers [Cohen et al., 2004; Looi et al., 2011]. There were no differences in QoL between implanted adults and normal-hearing peers [Rembar et al., 2012]. Most of the QoL benefit of CI was observed within the first 4 months following surgery [Klop et al., 2007, 2008], with the gain sustained for at least 6 years after CI [Damen et al., 2007]. Younger age at implantation was associated with improved QoL outcomes [Chung et al., 2012].
QoL Outcomes in Older Adults
Six of the 57 studies focused on presenting metrics for assessing the QoL attainment of older adults with cochlear implants. The studies were comprised of participants from 6 countries with a sample size ranging from 20 to 283 study participants. The mean age at implantation ranged from 52.9 to 74.4 years (Table 1).
Generic Instruments in Older Adults
All 6 studies reported on generic health utility metrics, which included the SF-36, GBI, and the Glasgow Health Status Inventory (GHSI) questionnaire. The SF-36 and GBI are described above. The GHSI questionnaire assesses the effect of a hearing problem on overall QoL, including physical health and social support [Orabi et al., 2006].
Disease-Specific Instruments in Older Adults
Two studies reported on disease-specific QoL instruments in adults, including the Specific Questionnaire (SQ) [Ramos et al., 2013], the NCIQ [Olze et al., 2012], and the Hearing Handicap Inventory - Adults (HHIA) [Vermeire et al., 2005]. The SQ evaluates 6 different aspects related to CI: speech recognition, social interaction, telephone use, confidence, family life, and satisfaction [Faber and Grøntved, 2000; Ramos et al., 2013]. The NCIQ is described above. The HHIA is a well-validated questionnaire that assesses hearing handicap in adult populations. The HHIA is a 25-item, hearing-specific, QoL scale with 2 components that measure the emotional and situational impact of hearing loss [Vermeire et al., 2005].
Ad hoc Instruments in Older Adults
Four reports included ad hoc QoL instruments [Di Nardo et al., 2014; Olze et al., 2012; Orabi et al., 2006; Vermeire et al., 2005]. Of these, 3 studies used quantitative QoL questionnaires, and 1 [Olze et al., 2012] used a qualitative instrument.
Summary of Results for Older Adults
Overall, implanted older adults showed significant improvement in QoL after implantation [Chung et al., 2012; Di Nardo et al., 2014; Olze et al., 2012; Orabi et al., 2006; Ramos et al., 2013; Vermeire et al., 2005]. Improvements in QoL were independent of audiological performance. While all studies demonstrated significant auditory benefit of CI in all age groups, some studies [Di Nardo et al., 2014; Ramos et al., 2013; Vermeire et al., 2005] showed poorer auditory performance in the elderly group, and others showed a similar benefit of CI regardless of age [Olze et al., 2012; Orabi et al., 2006]. Regardless, nearly all studies showed significant improvement in QoL in this age group, most often reported in multiple dimensions.
The objective of this narrative review was to comprehensively assess the literature on the QoL determination and attainment among CI recipients of all ages, and to examine cost-effectiveness measures. We were able to access a large compendium of research conducted over the past 2 decades and around the world to assess the impact of CI in deafness from a perspective that combines economic considerations with measured outcome of implantation offered by recipients or proxies. We observed a consistently high impact of implantation on perceived QoL, and a general trend towards favorable economic impact of CI that is subject to single- versus bilateral- implantation considerations.
Studies that combine clinical and economic evaluation are increasingly important in guiding comprehensive assessments and payer decision-making. Such decision-making is critical to the acceptance of new technologies; however, there is often limited, timely evidence available to payers on the effectiveness and cost-effectiveness of emerging and evolving technologies [Steiner et al., 1996]. As a result, health care stakeholders are at a disadvantage when attempting to justify interventions based not only on safety and efficacy, but also on cost-effectiveness. Economic analyses, when based on a broad range of HRQoL domains that permit comparison of outcomes across interventions, diseases, and populations, can provide guidance in programmatic initiatives designed to enhance performance and contain costs while prioritizing expenditures.
The results of this review suggest that an approach to the question “Are cochlear implants cost-effective?” is nuanced and that a meaningful answer must incorporate multiple factors that offer objective measures and reflect priorities of hearing-impaired individuals, professionals with relevant expertise, and society in general. We observed that even when priorities are recognized, the answer varies, depending on the economic measures used in the analysis.
Economic Evaluation Measures
Changes in quality-adjusted life years that are based on HRQoL measures can vary based on several factors, including the measurement tool used (whether direct or indirect measure, disease-specific or generic measurement), and the population studied (e.g., whether those surveyed are hearing-impaired CI candidates, individuals with unilateral or bilateral implants, hearing health care professionals, parents serving as proxy respondents for their children, or members of society). We noted consistency across studies using the various instruments; however, the HUI3 is generally regarded as the most conservative and consistent measure of HRQoL for studies of CI cost utility [Kuthubutheen et al., 2015]. In most studies, the HUI3 does not show significant gain in utility from the unilateral to bilateral CI condition though it was reported clearly sensitive to unilateral CI. For this reason, the cost-utility analysis of bilateral intervention is highly driven by the increase in cost (up to double the cost of unilateral CI) with modest commensurate, enhanced utility produced by bilateral implantation.
CI analysts have struggled to understand why the impact of a technology that patients consistently value as a life-changing intervention, such as bilateral CI, is not reflected in significant increments in utility scores. It is possible that the communicative advantages produced are simply not captured in the instruments used, and even disease-specific metrics often fail to capture the nuanced ways in which access to binaural auditory inputs may confer enhanced connectivity. We noted that part of the challenge is adequately probing the benefit of a second implant when a first implant has been placed. Weighted-average QoL gains, reflected in generic QoL instruments, were observed in our literature survey to be 0.04 in children and 0.08 in adults (Table 5). These same weighted-average QoL gains are, when considered in the context of bilateral implantation as compared with the baseline state, 0.26 for children and 0.24 for adults - gains in utility that are substantial in comparison to many established medical and surgical interventions.
The neurobiology of the central auditory system informs us that the system relies on bilateral stimulation to develop normal optimal function. While significant gains in localization and the ability to understand speech in the presence of background noise are realized in bilateral CI recipients [Kral and O'Donoghue, 2010], such benefits are not robust by current QoL measures. It must be appraised whether this discrepancy is a failure of CI to adequately restore an important sensory deficit, or whether current measurement tools are simply inadequate to demonstrate a meaningful gain in function. When QoL gains are modest, costs of the intervention become more important. For this reason, efforts to control incremental costs of bilateral CI, which are limited primarily to the cost of the second implant and the episode of care, are of great importance.
Sensitivity analyses are particularly useful to describe the variety of costs and other assumptions that drive the determination of cost-utility analysis. Costs can vary widely both within and between countries, and aggressive cost-cutting measures can improve financial sustainability [McKinnon, 2014]. Factors that can be examined in sensitivity analyses may include: numbers of pre- and post-CI assessments, failure rates of external devices out of warranty, discount rates (for costs as well as benefits over time), time horizon, utility gains, timing of surgeries such as bilateral sequential and simultaneous implants, and time horizon of the intervention. Cost-utility of bilateral versus unilateral CI is highly dependent upon cost of the second implant. In some health care settings, the cost of the second implant, particularly if placed simultaneously, has been discounted 50% from the first device.
While HRQoL is traditionally used in economic analyses, there are other measures of outcome that can be considered and have not been typically measured in cost-utility analyses of CI. These measures reflect financial well-being and contributions to society by the hearing-impaired individual. Work by Emmett and Francis  using the US National Health and Nutrition Examination Survey information from 1999 to 2002 demonstrated that individuals with hearing loss (World Health Organization definition of bilateral pure-tone average of 0.5, 1, 2, and 4 kHz of greater than 25 dB) were 3.21 times more likely to have low educational attainment (not complete high school), 1.58 times more likely to be of low income (family income less than USD 20,000 per annum), and 1.98 times more likely to be unemployed or underemployed (work less than 35 h per week). Other societal benefits may include lower rates of depression and dementia in elderly CI recipients. It is not yet clear if individuals with hearing loss and CI perform as well as normal-hearing people with respect to these important measures of health, well-being, and contributions to society, but several studies do suggest that income levels for adults are improved following CI and that eventual educational and employment attainment for children with CI may be on par with those with normal hearing [McKinnon, 2014]. Hence, as difficult as it is to define all possible costs of an intervention, it is equally challenging to fully reflect all possible associated benefits to society. These benefits can be, but have not yet been, measured in financial terms.
The final step in the economic analysis requires that incremental cost-utility be determined to reflect the incremental cost per HRQoL benefit from CI. Thresholds are set based on the incremental cost-utility ratio: if it falls below this set level, an intervention is determined to be “cost-effective.” This varies by country in the developed world. This threshold is approximately USD 50,000 in the USA, and an equivalent-currency cost in the UK. Here, we observed that despite variance in utility associated with different methodologies, the calculated cost-utility ratios for the USA and Canada consistently met this threshold, indicating high value for expenditures on CI (Table 6). Some of the studies, particularly those performed in the UK, revealed higher cost-utility ratios; the reason for this is not known, but should be explored further. Such data can also be assessed in developing countries, wherein the threshold for cost-effectiveness suggested by the World Health Organization as a factor of gross domestic product can be developed using a methodology that drives CI feasibility in lower income countries [Emmett et al., 2015].
Range of Outcomes Observed (Unilateral)
Results observed in this survey demonstrated mean health-utility gains in children receiving unilateral implants ranging on average from 0.27 with generic measures to 0.09 with disease-specific measures (single study). In adults, utility gains in generic QoL measures ranged on average from 0.21 with generic measures to 0.28 with disease-specific measures. While there is variability in these pooled results, mean gains in utility that were observed to consistently exceed 0.20 are impressive and rarely observed in health utility studies of interventions commonly covered, for example, within the US health care system.
Range of Outcomes (Bilateral)
Health utility gains following bilateral CI may be compared either to no intervention, in which case gains are robust (average 0.26 and 0.24 using generic measures for children and adults, respectively, and 0.62 using disease-specific measures for adults), or to unilateral implantation, in which case benefits, as mentioned previously, are more nuanced. Absolute gains in health utility for 2 versus 1 CI average from 0.04 to 0.1 in children for generic versus disease-specific measures, and from 0.08 to 0.1 in adults for the same. In each case, there is a small tendency for the disease-specific measures to be more reflective of the benefits of bilateral CI. This makes sense, as bilateral CI is less likely to show incremental benefits in overall QoL than to benefit the individual in listening in complex auditory environments, particularly with adequate time for follow-up and refinement of auditory skills.
Overall Valuation and Related Ethical Considerations
The overall economic value of CI is generally regarded as greater in children who are implanted early in life than in adults, both because of the greater benefit over longer mean time horizons, and because of the fact that young children with significant hearing loss do not develop normal speech and language skills. This raises an ethical consideration: What if lack of spoken language is considered acceptable, “normal” and even desirable for a segment of the population? For example, this is the often discussed case with a deaf community stance. How should their valuation of utility of intervention for deafness, which they do not regard as a disability, be considered? We note that this situation is quite different from the stroke victim for whom health care providers are considering a costly intervention, or for a potential transplant recipient. One argument is that, while a dialogue around issues related to health valuation can include a choice of perspective (patients or able-bodied individuals) as well as the perimeter of assessment (medical or nonmedical care), the community finance of treatment such as bilateral CI for children is justified once it is shown that the disability has a significant impact on the fundamental social achievements of the individual and that such treatment can reduce inequalities effectively [Thebaut, 2013]. Evidence of the first point has been advanced [Emmett and Francis, 2014], and evidence of the second point continues to accumulate.
Considerable work has been done on the QoL attainment and health economic implications of CI. Unilateral CI across all age groups leads to reported sustained benefits in the recipients' overall and disease-specific QoL. With increased cost associated with bilateral CI, further study is needed to characterize its costs and benefits with respect to the recipients' health, well-being, and contributions to society.
There was no conflict of interest.
There was no funding/support.
Matthew G. Crowson and Yevgeniy R. Semenov are co-first authors.