Expert opinion: Can different assessments be used to overcome current candidacy issues?

Abstract

Since the National Institute of Health and Care Excellence (NICE) review of cochlear implantation in 2009, (NICE, 2009) there have been a number of significant changes to our understanding of the impact of severe-to-profound hearing loss on quality of life and comorbidity with life limiting illness. There have been questions about the validity of current methods of assessing candidacy for cochlear implants. There have also been significant improvements in the effectiveness of implants, the age of successful implantation and a reduction in costs. Additionally, the costs to the health and welfare system of not addressing severe-to-profound hearing loss are often not considered when assessing costs and benefits of this technology and when assessing candidacy criteria. Consideration of these changes since the NICE review suggests the need for an urgent review of the current guidance.

Keywords:

• Cochlear Implants
• Candidacy
• NICE guidelines
• Quality of life measures

Introduction

Since the NICE guidance in (2009) and (2011) last reviewed the effectiveness of Cochlear Implants (CI) there are more adults and children with significant residual hearing, single-sided deafness, and of various ages who previously would not have met the criteria and are now being implanted in other countries (Arnoldner and Lin, 2013). There have also been significant improvements to CI technology coupled with a deepening understanding of patient benefit and a growing recognition of the limitations of current candidacy criteria and the measurement methods used to establish them (Sampaio et al., 2011). The cost–benefit analysis applied to current thinking about candidacy needs to be considered within the context of a better understanding of the real costs of hearing loss, patient benefit and more sophisticated measures of quality of life (Lamb et al., 2015).

What is the size of the current population who could benefit?

NICE (2009) identified that there are approximately 613 000 people older than 16 years with severe-to-profound deafness in England and Wales. With a population of 51.4 million over 15 years of age in the UK (Office of National Statistics, 2011) there is an estimated 100 000 people with a profound loss and 360 000 with a severe loss who might benefit from CI's. However recent figures show that at best only 6.7% of adults with profound hearing loss, the group most likely to meet the current criteria, are implanted (Lamb et al., 2015; Raine, 2013).

For children, it is estimated that the rate of implantation is 74% of estimated eligible children with severe to profound hearing loss for children in the 0–3 group and increasing to 94% of these children by the time they have reached 17 years of age. These estimates also suggest underutilization for children compared to current clinical criteria especially in the younger age group (Raine, 2013).

A number of problems have been identified with current measures including overreliance on audiological criteria and sentence testing.

Audiological criteria for cochlear implantation

The audiological criterion for cochlear implantation set by NICE is the ability to hear only sounds that are louder than 90 dB HL at frequencies of 2 and 4 kHz without acoustic hearing aids (i.e. the 90 dB HL criterion level). An additional criterion is the absence of adequate benefit from acoustic hearing aids which is defined as:

• for adults, a score of 50% or greater on Bamford–Kowal–Bench (BKB) sentence testing at a sound intensity of 70 dB SPL;

• for children, speech, language, and listening skills appropriate to age, developmental stage, and cognitive ability.

There is considerable variation at the International level (Vickers et al., 2016) with many countries with audiometric guidelines which are much less restrictive. In Australia they use 70 dB HL criteria, Germany, Italy and the USA are also less restrictive than the UK with the majority of clinics using a 75–80 dB HL cut off at frequencies greater than 1 kHz (Raine, 2013; Vickers et al., 2016). Further in Germany, Italy and Australia implant teams have a greater level of clinical discretion to determine appropriate candidacy using a number of criteria that clinicians find useful (Raine, 2013; Vickers et al., 2016). In the UK however there are only limited examples of obtaining funding for special cases.

The requirement for higher audiometric testing levels in CI candidacy in the UK than many other countries means that a lower proportion of adults are being provided with CIs. Audiometric tests have also come under scrutiny as pre-implant residual hearing is one of the important attributes contributing to the post-operative outcomes (Chundu and Flynn, 2014) and therefore criteria which favour a lower threshold would also indicate the prospect of greater post-operative benefit.

Adults who are deafened early have also been seen to have substantial gains from cochlear implantation and those who are also late implanted as well as early deafened are now showing significant levels of improvement across speech recognition and quality of life measures.

Further Leal et al. (2016) showed that by using the speech intelligibility index (SII) that children, and by implication adults, could benefit from a lower threshold candidacy level (80 dB HL at 2 and 4 kHz) and that the SII could be used as an additional resource to determine candidacy, especially for borderline candidates with atypical hearing loss configurations. This work supports other calls for a reduction of the audiometric criteria (Lovett et al., 2015)

It has also been proposed (Chundu and Flynn, 2014) that in areas such as ANSD, the average of the whole frequency range of pure-tone hearing threshold levels at 250, 500, 1000, 2000, and 4000 Hz ≥90 dB HL (British Society of Audiology, 2011), and BKB scores <50% need to be considered. Individuals who are not deriving benefit from their hearing aids should also be considered for cochlear implantation regardless of their scores on the assessments because there is currently no established measure to quantify the damage to the cochlea.

The ability to preserve residual hearing at low frequency thanks to improved implantation techniques and the ability to use EAS stimulation also indicates that we might need to reassess our views on candidacy as it is clear that more patients could benefit. Further as Verschuur et al. (2016) has shown more patients could be referred under current candidacy requirements if the impact of EAS was better understood.

Use of sentence testing

The type of testing used to establish candidacy is also crucial to determining the overall results. The use of sentence tests, rather than monosyllabic words, has become increasing questioned as it enables deafened adults to use their previous linguistic knowledge and cognitive ability to predict words and give a false impression of audibility.

A number of alternative measures or combination of measures has been proposed as an alternative. Raine (2013) has proposed that assessment of performance with monosyllabic words would be more appropriate. The use of monosyllables in testing for qualification purposes has been used in Germany for a number of years where patients achieve <30% correct for Freiberg monosyllables at 70 dB SPL in the best-aided condition (Aschendorff et al., 2007; Gifford et al., 2010). As people with greater residual hearing are being assessed they may score greater than 50% on BKB in quiet and should therefore be tested in noise as well to fully understand the difficulties that the individual faces. In Germany where people with greater levels of hearing are being assessed there has been a move towards testing in noise (Haumann et al., 2012).

Further Doran and Jenkinson (2016) has suggested that the Arthur Boothroyd (AB) word test would be a suitable supplement to BKB testing to detect candidates who have some residual hearing but would benefit and Vickers and Bradley (2016) also challenge the current guidelines arguing for a combination of tests which do not rely on a person's pre-existing linguistic knowledge including monosyllabic word tests.

Gifford et al. (2010) in a retrospective review of post-operative speech perception performance for 22 adult cochlear implant recipients who demonstrated preoperative Consonant Nucleus Consonant word recognition scores of 30% or higher in the best-aided condition found that non-traditional implant recipients, with higher levels of preoperative speech understanding than traditional patients, demonstrated significant benefit from cochlear implantation.

These considerations suggest that the BKB provides useful information and together with the AB words this would be a good combination of measures. However there is more work needed on the candidacy cut off point.

Children

There is evidence that the current criteria for children may also be too restrictive. Paediatric implant recipients that have not met the current criteria have been seen to derive significant benefit from cochlear implantation. Research by Carlson et al. (2015) concluded, after a retrospective case study of implanted children who had less severe hearing loss than specified in the current indications and had open-set word and/or sentence recognition scores greater than 30% for children who are able to participate in speech perception testing; that a large-scale reassessment of paediatric cochlear implant candidacy, including less severe hearing losses and higher preoperative speech recognition, should be undertaken.

In looking at post-operative benefit Sampath Kumar et al (2016) also found that audiometric tests were not good at capturing benefits for pre-lingually deaf young people arguing that improvements in speech discrimination scores may not be apparent in this group after implantation but more subtle benefits such as improved interpersonal skills, emotional wellbeing, greater satisfaction at work and greater personal satisfaction were reported.

Patient experience and outcomes

Patient experiences of candidacy, where a majority of those refused an implant, believed the decision was made on the basis of the audiological criteria alone which were too restrictive and did not take account of their inability to hear in real world situations (Althalye et al., 2015). Further patient surveys showed that those implanted highly value the impact the CI has had on their lives in improving confidence, ability to relate to others, maintain employment and improved wellbeing. To gauge the value patients might put on their implant researchers have also asked patients to put a financial value on their implants. Patients already fitted with one implant all but one said that if they had the choice between a second implant or £15 000 they would choose the second implant (Buhagiar and Lutman, 2011) Further, Lamb et al. (2015) found that patients put a high financial value on the benefit of their implant. When asked to give their implant a financial value on a monetary scale 60% chose the highest value of above £150.00 per month. However many also qualified this answer with statements which suggested that their implant was ‘priceless’. This suggests that real world effects and self-reported patient benefit are not being reflected in the way audiological criteria are being applied. There could be much greater use of Patient Reported Outcome Measures (PROMS) such as Speech, Spatial and Qualities of Hearing Scale (SSQ) and other tools in assessing candidacy (Chundu and Flynn, 2014; Vickers and Bradley, 2016)

While there is not one single consensus on the best alternative measure, these considerations suggest that a more flexible approach to candidacy criteria, for adults and children, using a mix of clinician led approaches may be more appropriate to assess suitability. A selection of carefully selected resources for assessing candidacy should be made available to the clinician so that they can develop a candidacy profile. Also as Vickers and Bradley (2016) suggest tests will be much more successful when combined or replaced with other measures such as patient administered questionnaire like the (SSQ) as such tools are more responsive to real world benefits for patients.

Quality of Life Criteria in measuring patient outcomes.

Quality of Life (QoL) measurement depends on ensuring that the instruments are sufficiently attuned to correctly assess the issues that the adult experiences. Compared to speech perception tests QoL scores allow a more comprehensive insight into how interventions impact on people's ability to benefit from the additional communication capacity that implants enable by measuring the benefit that obtain in real world situations. When combined with cost effectiveness criteria they can help with evaluating the effectiveness of an intervention.

As Loeffler et al. (2010) concluded after reviewing some of the established QoL instruments “QoL instruments are an essential addition to speech perception tests to quantify the outcome of cochlear implants. Compared to speech perception tests QoL scores allow a more comprehensive insight into patients’ daily life and activities.” NICE has confirmed that commonly used measures of generic health-related quality of life, such as the EuroQol 5 Dimensions (EQ-5D), which allow the quality of life changes from different interventions to be compared, will not always be appropriate for individual conditions. (NICE, 2013) and its suitability as a measure for hearing loss has also been challenged by Payakachat (2015)

When more disease specific measures are used a significant impact of quality of life is shown for CI users in most studies. The Nijmegen Cochlear Implantation Questionnaire (NCIQ) is proposed (Loeffler et al., 2010) as a more reliable and sensitive measure to reflect clinical changes in CI patients than more generic tests (Bess, 2000) such as SF36 which does not reflect issues relating to hearing loss and the Health Utilities Index which is partially sensitive to hearing loss. It is important to ensure that these instruments can fully account for the life experiences of users. For example, (Ramen et al., 2011) demonstrated a significant effect on disease specific functional and QoL scales for unilateral cochlear implantation in adults, but no effect according to generic scales (Gaylor et al., 2013). It is not clear that all the potential benefits are currently considered by such tests and we need to look at how these can be refined further (Ramen et al., 2011)

Candidacy requirements should encompass a more sophisticated understanding of the additional costs of not treating hearing loss by taking account of additional health care and social care costs related to earlier onset dementia, mental health, and reduction in independence, falls and reduction in economic activity (NHS, 2015). The UK is poor at taking account of these additional measures in candidacy criteria as the additional costs tend to get discounted as the cost is taken by another area of public expenditure. Future economic assessment studies should consider the other categories of non-medical direct costs and indirect costs (Turche et al., 2011). Making a fuller consideration of total costs across the health and social care system, which is possible to consider, and has been done for other conditions in developing health strategies, would more than balance the cost of expanding the current criteria (Archbold et al., 2014; Lamb et al., 2015).

A significant number of systematic reviews have raised issues about the overestimating of the cost of technology and surgery, which have both reduced substantially, in assessing the overall costs of implants. This has impacted negatively on assessments of the cost effectiveness of CI's as has the failure to fully account for how auditory test scores might translate into real world benefits; the additional potential real world and audiological gains from bilateral squelch contributions, and self-reported benefits from patients (Archbold et al., 2014; Lamb and Archbold, 2013). These considerations become even more relevant when considering cost effectiveness assessments for bilateral fitting. Without these considerations financial criteria for assessing Quality of Life Adjusted Years can be distorted against the interest of patients who could benefit.

Within the current criteria there is also a level of non-provision, measured against those who could benefit, which suggests that centres do not think that funding is available for all who meet the criteria. This underutilization also questions how well the current criteria are understood and applied by the Audiology profession. (Lamb et al., 2015; Raine, 2013).

Conclusion

The recommendation for a CI should be always based on functional hearing, taking into account the difficulties faced by the patients and their families in real-life situations rather than strictly adhering to the audiological criteria. Further, the current testing needs review and supplementing with more refined measures, including PROMS, and a lower threshold for testing where 80 dB HL has also been suggested given the evidence of gains outside of the current criteria and for specific groups of patients. Taken together this suggests that NICE needs to urgently review its current guidelines so that more people can benefit from this technology. A more sophisticated analysis would show that increased candidacy would save the NHS money if the wider costs of disability and illness are taken into account.

Disclaimer statements

Contributors None.

Funding None.

Conflict of interest None.

Ethics approval None.