REVIEW URRENT C OPINION
Cost effectiveness of cochlear implants Brian J. McKinnon
Purpose of review Health professionals would be well served to have as good an understanding of cost effectiveness as clinical effectiveness, as both are critical to their patients having access to better health care and achieving better health outcomes. Cost-effectiveness evaluations allow decision makers a means of comparing different interventions when deciding resource allocation. It is a powerful tool, but like any analysis, not understanding the processes and assumptions involved leads to misinterpretation. Recent findings Cost effectiveness is an economic evaluation of cost and benefit. The threshold at which an intervention is considered cost effective is reflected by the payer’s ‘‘willingness to pay’’, which can vary considerably from country to country. These evaluations are complex and can involve the use of incomplete financial data, and subjective impressions of benefit, while excluding broader social and economic benefits. Summary Pediatric unilateral and simultaneous bilateral cochlear implantation, and adult unilateral cochlear implantation are felt to be cost effective in the United States. Pediatric sequential cochlear implantation, adult bilateral cochlear implantation, implantation in the aged and the long deaf are not. However, costeffectiveness economic evaluations are only part of broader assessment of social and economic benefit when determining resource allocation. Keywords cochlear implants, cost effectiveness, cost–utility analysis, quality adjusted life year
INTRODUCTION Cochlear implantation is understood to be a well established means of addressing severe-to-profound sensorineural hearing loss in patients who do not benefit from conventional amplification. It has been estimated that over 324 200 patients worldwide have received this medical device [1]. Unilateral cochlear implantation is considered to be both clinically effective and cost effective [2–4]. Nevertheless, this evidence has not led to consistent interest in or support for unilateral cochlear implantation in the United States [5 ]. Further, although there is evidence supporting the clinical benefits of bilateral cochlear implantation, cost effectiveness evaluations of bilateral cochlear implantation has been far less favorable [6–9,10 ]. This is certainly confusing. If something is cost effective, what does that mean? If one cochlear implant is cost effective, why are not two cochlear implants even more cost effective? &
effectiveness should be interpreted is needed. This cannot be understated. In a review by Kezirian and Yueh [11] in 2001 of the otolaryngology literature, the authors found that over half of the terms, such as ‘‘cost effective’’, were used incorrectly, with 60% of the reviewed articles confusing ‘‘charge’’ and ‘‘cost’’ data. Only 11% of papers specified the perspective of their analysis, about half reported a summary measure, such as a cost-effectiveness ratio, and only one third performed sensitivity analyses [11]. Cost effectiveness is not the same as clinical effectiveness, although cost effectiveness can be very dependent upon the perceived and/or actual benefit derived from clinical effectiveness. As an economic evaluation, cost effectiveness is
&
CALCULATING COST EFFECTIVESS To address these questions, a greater understanding of how cost effectiveness is determined, and how cost www.co-otolaryngology.com
Shea Ear Clinic & Department of Otolaryngology-Head and Neck Surgery, University of Tennessee Health Sciences Center, Memphis, Tennessee, USA Correspondence to Brian J. McKinnon, MD, FACS, Associate, Neurotology/Otology, Shea Ear Clinic, 6133 Poplar Pike, Memphis, TN 38119, USA. Tel: +1 901 761 9720; fax: +1 901 415 6640; e-mail: brian.mcki [email protected] Curr Opin Otolaryngol Head Neck Surg 2014, 22:344–348 DOI:10.1097/MOO.0000000000000091 Volume 22 Number 5 October 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Cost effectiveness of cochlear implants McKinnon
KEY POINTS Cost effectiveness is an economic evaluation of cost and benefit for the comparison of different interventions; the evaluation uses a process known as a CUA. WTP is the level of spending a country is willing to commit toward a health benefit and is the threshold that is used to describe an intervention as cost effective or not cost effective. HRQoL is the measure of health utility (subjective or perceived benefit) obtained from an intervention; the QALY is a common form in which the HRQoL is expressed. In the United States where the WTP is less than $50,000/DQALY, pediatric unilateral and simultaneous bilateral cochlear implantation, and adult unilateral cochlear implantation are felt to be cost effective, although in other countries with lower WTP thresholds, this is may not be the case. As a CUA does not include broader social and economic benefits, such as improved educational, and employment attainment, and increased income, CUA should not be the exclusive means to determining resource allocation.
concerned with economic efficiency, and although this evaluation does not necessarily include patient wants and desires, the declared goal is an equitable allocation of resources. The equitable allocation of resources is referred to as welfare economics [12]. Economics is a social science that evaluates behavior on economic wants and needs, and that behavior is both rational and irrational. Welfare economics, a branch of economics, holds two concepts or theorems; first, a competitive equilibrium is economically efficient, and second, redistribution need not conflict with economic efficiency. Four common forms of equity that inform the economic discussions of resource allocation are egalitarian (all members of society receive equal amounts of goods), Rawlsian (maximize the utility of the least well off person), utilitarian (maximize the total utility of all members of a society), and market-oriented (the market outcome is the most equitable) [13,14]. Cost-effectiveness evaluations are performed using a cost–utility analysis (CUA), which is an attempt to assign a value to a change in health that is achieved with or by an intervention [8]. To calculate this, the health changes are incorporated into a measure known as the health-related quality of life (HRQoL), and from this measure a utility can be obtained, the quality-adjusted life year (QALY) [8]. Although QALY is not the only HRQoL measure
used in discussions of cost effectiveness, it is the one most commonly used, and permits comparisons of relative value between different treatments for different disease states [12,15]. As the CUA formula is represented as cost (in the appropriate currency) divided by the change in QALY (DQALY) between the two compared interventions, cost calculations are as equally important as QALY calculations. QALY values are preferred or desired state of health over a particular period of time, and can involve a trade-off of time for a particular state of health [12,15]. A simplified example of such comparison questions can be shown by asking the hypothetical patient the following: would he or she prefer a perfect state of health, but live only one more year, or would he or she accept some reduced or diminished state of health and live 3 years. The hypothetical patient would be signaling his or her thoughts of the utility of the state of health (quality of life), for either more time, and poorer health, or more health and less time. These preferences can then be taken for individuals and calculated for the group, and when costs of the intervention are included, the intervention’s QALY can be derived. This QALY allows comparisons of various interventions by their HRQoL outcomes [12,15]. There are several methods to discovering a patient’s HRQoL; direct elicitation methods, generic preference-based methods, and condition-specific measures [12]. Common direct elicitation methods include the visual analog scale, the time trade-off, and the standard gamble. The visual analog scale is a very simple method, as it uses a linear rating scale, with one end of the scale indicating the state of death and the opposite indicating the best possible state of health. The time trade-off is not unlike the simplified example given above of having to choose between two possible states of health, each for a given period. The standard gamble involves choosing between states of certainty and uncertainty, usually involving a choice between a current state of health and the risk of death or the risk of severe decline in health from an intervention to achieve a possible greater state of health [12]. Generic preference-based methods and condition-specific measures are forms of indirect elicitation methods, and are often preferred, as direct elicitation methods can be difficult and time consuming. A commonly used questionnaire such as the Health Utilities Index is an example of a generic preference-based measure that uses the standard gamble in its valuation [12]. These questionnaires vary depending on the states of health being measured, the population being assessed, the detail and complexity of the information being collected, and the valuation method being used. In the case of
1068-9508 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-otolaryngology.com
345
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Otology and neuro-otology
the Health Utilities Index, the Health Utilities Mark 3 version has eight health attributes, each having five to six levels, and is capable of describing 972 000 health states [12]. Condition-specific measures are used when generic preference-based methods may not be sensitive to states of health unique to a condition, although a drawback is that conditionspecific measures may not be suitable for QALY calculation, and may not incorporate important facets of the disease state into the CUA [12]. There are several weaknesses associated with the use of QALY for HRQoL determinations. QALY calculations do not incorporate considerations of equity or social value; the impact of a health intervention rating a low health utility on an economically disadvantaged child or working age adult may yield life-long benefits, such as improved educational and employment attainment, where an intervention with a higher health utility in an affluent geriatric patient would not [12]. QALY calculations are dependent on an individual making a decision on a preferred health state, a decision that is highly personal. The individual may overrate a poor state of health, underrate a good state of health, or have little or no experience with the health condition being rated, all leading to a skewing of the health utility calculation [12]. These concerns regarding the shortcomings of health utility calculations are particularly important in view of the results of the WHO’s recent survey on the global burden of disease. Hearing impairment is defined by the WHO as a hearing loss of greater than 40 dB in the better hearing ear [16]. Hearing impairment is a leading cause of years lived with disability, impacting approximated 1.3 billion persons worldwide [17]. In the United States, it is one of the top 20 contributors to years lived with disability, at nearly 560 000 in 2010 [18]. The survey uses the disability-adjusted life year (DALY) as a measure of disease burden (disability-weighted measure of loss of functioning). In the previous surveys, health professionals rated the disability impact of disease burden, and the calculated DALY of hearing loss was fairly high. In 2010, the survey was conducted with members of the general public in Bangladesh, Indonesia, Peru, Tanzania, and the USA using household and open-access web-based surveys, and the calculated DALY for hearing impairment was one tenth that of the previous survey [17]. Puzzled as to the cause of this dramatic change, on investigation it was found that the survey used lay definitions of hearing loss focused on the hearing impairment itself, excluding other possible outcomes that might accompany severe levels of hearing loss, such as depression or learning disabilities [17]. It was concluded that the dramatic decrease in DALY did not 346
www.co-otolaryngology.com
so much reflect a true decrease in the burden of hearing loss, but rather a flawed survey instrument [17]. The costs used in CUA typically include both indirect and direct costs [19–21]. Indirect costs are those associated with lost wages, loss productivity, and travel costs, and as these are more difficult to quantify, are often estimated, or are based on standard, accepted assumptions [19,20]. Direct costs are those that are from materials and labor such the cost of the device, and health professionals’ time; direct costs are usually obtained directly from billing and other financial records [21]. Although there is consistency in what is considered a direct cost, there is marked variability in what is included as indirect costs, which contributes significantly to differences in reported cost effectiveness [19,20]. Variations in direct costs can occur; one cochlear implant program reported that an aggressive cost management program reduced direct costs by 22% [22]; if a similar cost containment effort was included in the calculations of cost, this could significantly change CUA calculations.
INTERPRETING COST EFFECTIVE CALCULATIONS So how to address the following two questions. First, if something is cost effective, what does that mean? Second, if one cochlear implant is cost effective, why are not two cochlear implants even more cost effective? An intervention is ‘‘cost effective’’ when it falls under a certain Cost/DQALY threshold, a threshold that is determined by what the purchaser is willing to pay. The willingness to pay (WTP) threshold is usually set at a number that reflects what percentage of a county’s gross domestic product the country is willing to spend and is stated in that country’s currency. For example in the United States, this is $50 000 [16]; in the United Kingdom, £30 000 (roughly $48 300) [6]; and in the Netherlands, s20 000 (roughly $28 600) [6]. If the Cost/DQALY is below the country’s WTP threshold, then it is considered ‘‘cost effective’’; as can be seen, what may be considered cost effective in one country, may be deemed as not cost effective in another. Reviews comparing pediatric cochlear implants, unilateral and bilateral, and adult unilateral cochlear implantation to no cochlear implantation [2–4,6–9] have found their Cost/DQALY within the United States WTP (
Cost effectiveness of cochlear implants Brian J. McKinnon
Purpose of review Health professionals would be well served to have as good an understanding of cost effectiveness as clinical effectiveness, as both are critical to their patients having access to better health care and achieving better health outcomes. Cost-effectiveness evaluations allow decision makers a means of comparing different interventions when deciding resource allocation. It is a powerful tool, but like any analysis, not understanding the processes and assumptions involved leads to misinterpretation. Recent findings Cost effectiveness is an economic evaluation of cost and benefit. The threshold at which an intervention is considered cost effective is reflected by the payer’s ‘‘willingness to pay’’, which can vary considerably from country to country. These evaluations are complex and can involve the use of incomplete financial data, and subjective impressions of benefit, while excluding broader social and economic benefits. Summary Pediatric unilateral and simultaneous bilateral cochlear implantation, and adult unilateral cochlear implantation are felt to be cost effective in the United States. Pediatric sequential cochlear implantation, adult bilateral cochlear implantation, implantation in the aged and the long deaf are not. However, costeffectiveness economic evaluations are only part of broader assessment of social and economic benefit when determining resource allocation. Keywords cochlear implants, cost effectiveness, cost–utility analysis, quality adjusted life year
INTRODUCTION Cochlear implantation is understood to be a well established means of addressing severe-to-profound sensorineural hearing loss in patients who do not benefit from conventional amplification. It has been estimated that over 324 200 patients worldwide have received this medical device [1]. Unilateral cochlear implantation is considered to be both clinically effective and cost effective [2–4]. Nevertheless, this evidence has not led to consistent interest in or support for unilateral cochlear implantation in the United States [5 ]. Further, although there is evidence supporting the clinical benefits of bilateral cochlear implantation, cost effectiveness evaluations of bilateral cochlear implantation has been far less favorable [6–9,10 ]. This is certainly confusing. If something is cost effective, what does that mean? If one cochlear implant is cost effective, why are not two cochlear implants even more cost effective? &
effectiveness should be interpreted is needed. This cannot be understated. In a review by Kezirian and Yueh [11] in 2001 of the otolaryngology literature, the authors found that over half of the terms, such as ‘‘cost effective’’, were used incorrectly, with 60% of the reviewed articles confusing ‘‘charge’’ and ‘‘cost’’ data. Only 11% of papers specified the perspective of their analysis, about half reported a summary measure, such as a cost-effectiveness ratio, and only one third performed sensitivity analyses [11]. Cost effectiveness is not the same as clinical effectiveness, although cost effectiveness can be very dependent upon the perceived and/or actual benefit derived from clinical effectiveness. As an economic evaluation, cost effectiveness is
&
CALCULATING COST EFFECTIVESS To address these questions, a greater understanding of how cost effectiveness is determined, and how cost www.co-otolaryngology.com
Shea Ear Clinic & Department of Otolaryngology-Head and Neck Surgery, University of Tennessee Health Sciences Center, Memphis, Tennessee, USA Correspondence to Brian J. McKinnon, MD, FACS, Associate, Neurotology/Otology, Shea Ear Clinic, 6133 Poplar Pike, Memphis, TN 38119, USA. Tel: +1 901 761 9720; fax: +1 901 415 6640; e-mail: brian.mcki [email protected] Curr Opin Otolaryngol Head Neck Surg 2014, 22:344–348 DOI:10.1097/MOO.0000000000000091 Volume 22 Number 5 October 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Cost effectiveness of cochlear implants McKinnon
KEY POINTS Cost effectiveness is an economic evaluation of cost and benefit for the comparison of different interventions; the evaluation uses a process known as a CUA. WTP is the level of spending a country is willing to commit toward a health benefit and is the threshold that is used to describe an intervention as cost effective or not cost effective. HRQoL is the measure of health utility (subjective or perceived benefit) obtained from an intervention; the QALY is a common form in which the HRQoL is expressed. In the United States where the WTP is less than $50,000/DQALY, pediatric unilateral and simultaneous bilateral cochlear implantation, and adult unilateral cochlear implantation are felt to be cost effective, although in other countries with lower WTP thresholds, this is may not be the case. As a CUA does not include broader social and economic benefits, such as improved educational, and employment attainment, and increased income, CUA should not be the exclusive means to determining resource allocation.
concerned with economic efficiency, and although this evaluation does not necessarily include patient wants and desires, the declared goal is an equitable allocation of resources. The equitable allocation of resources is referred to as welfare economics [12]. Economics is a social science that evaluates behavior on economic wants and needs, and that behavior is both rational and irrational. Welfare economics, a branch of economics, holds two concepts or theorems; first, a competitive equilibrium is economically efficient, and second, redistribution need not conflict with economic efficiency. Four common forms of equity that inform the economic discussions of resource allocation are egalitarian (all members of society receive equal amounts of goods), Rawlsian (maximize the utility of the least well off person), utilitarian (maximize the total utility of all members of a society), and market-oriented (the market outcome is the most equitable) [13,14]. Cost-effectiveness evaluations are performed using a cost–utility analysis (CUA), which is an attempt to assign a value to a change in health that is achieved with or by an intervention [8]. To calculate this, the health changes are incorporated into a measure known as the health-related quality of life (HRQoL), and from this measure a utility can be obtained, the quality-adjusted life year (QALY) [8]. Although QALY is not the only HRQoL measure
used in discussions of cost effectiveness, it is the one most commonly used, and permits comparisons of relative value between different treatments for different disease states [12,15]. As the CUA formula is represented as cost (in the appropriate currency) divided by the change in QALY (DQALY) between the two compared interventions, cost calculations are as equally important as QALY calculations. QALY values are preferred or desired state of health over a particular period of time, and can involve a trade-off of time for a particular state of health [12,15]. A simplified example of such comparison questions can be shown by asking the hypothetical patient the following: would he or she prefer a perfect state of health, but live only one more year, or would he or she accept some reduced or diminished state of health and live 3 years. The hypothetical patient would be signaling his or her thoughts of the utility of the state of health (quality of life), for either more time, and poorer health, or more health and less time. These preferences can then be taken for individuals and calculated for the group, and when costs of the intervention are included, the intervention’s QALY can be derived. This QALY allows comparisons of various interventions by their HRQoL outcomes [12,15]. There are several methods to discovering a patient’s HRQoL; direct elicitation methods, generic preference-based methods, and condition-specific measures [12]. Common direct elicitation methods include the visual analog scale, the time trade-off, and the standard gamble. The visual analog scale is a very simple method, as it uses a linear rating scale, with one end of the scale indicating the state of death and the opposite indicating the best possible state of health. The time trade-off is not unlike the simplified example given above of having to choose between two possible states of health, each for a given period. The standard gamble involves choosing between states of certainty and uncertainty, usually involving a choice between a current state of health and the risk of death or the risk of severe decline in health from an intervention to achieve a possible greater state of health [12]. Generic preference-based methods and condition-specific measures are forms of indirect elicitation methods, and are often preferred, as direct elicitation methods can be difficult and time consuming. A commonly used questionnaire such as the Health Utilities Index is an example of a generic preference-based measure that uses the standard gamble in its valuation [12]. These questionnaires vary depending on the states of health being measured, the population being assessed, the detail and complexity of the information being collected, and the valuation method being used. In the case of
1068-9508 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-otolaryngology.com
345
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Otology and neuro-otology
the Health Utilities Index, the Health Utilities Mark 3 version has eight health attributes, each having five to six levels, and is capable of describing 972 000 health states [12]. Condition-specific measures are used when generic preference-based methods may not be sensitive to states of health unique to a condition, although a drawback is that conditionspecific measures may not be suitable for QALY calculation, and may not incorporate important facets of the disease state into the CUA [12]. There are several weaknesses associated with the use of QALY for HRQoL determinations. QALY calculations do not incorporate considerations of equity or social value; the impact of a health intervention rating a low health utility on an economically disadvantaged child or working age adult may yield life-long benefits, such as improved educational and employment attainment, where an intervention with a higher health utility in an affluent geriatric patient would not [12]. QALY calculations are dependent on an individual making a decision on a preferred health state, a decision that is highly personal. The individual may overrate a poor state of health, underrate a good state of health, or have little or no experience with the health condition being rated, all leading to a skewing of the health utility calculation [12]. These concerns regarding the shortcomings of health utility calculations are particularly important in view of the results of the WHO’s recent survey on the global burden of disease. Hearing impairment is defined by the WHO as a hearing loss of greater than 40 dB in the better hearing ear [16]. Hearing impairment is a leading cause of years lived with disability, impacting approximated 1.3 billion persons worldwide [17]. In the United States, it is one of the top 20 contributors to years lived with disability, at nearly 560 000 in 2010 [18]. The survey uses the disability-adjusted life year (DALY) as a measure of disease burden (disability-weighted measure of loss of functioning). In the previous surveys, health professionals rated the disability impact of disease burden, and the calculated DALY of hearing loss was fairly high. In 2010, the survey was conducted with members of the general public in Bangladesh, Indonesia, Peru, Tanzania, and the USA using household and open-access web-based surveys, and the calculated DALY for hearing impairment was one tenth that of the previous survey [17]. Puzzled as to the cause of this dramatic change, on investigation it was found that the survey used lay definitions of hearing loss focused on the hearing impairment itself, excluding other possible outcomes that might accompany severe levels of hearing loss, such as depression or learning disabilities [17]. It was concluded that the dramatic decrease in DALY did not 346
www.co-otolaryngology.com
so much reflect a true decrease in the burden of hearing loss, but rather a flawed survey instrument [17]. The costs used in CUA typically include both indirect and direct costs [19–21]. Indirect costs are those associated with lost wages, loss productivity, and travel costs, and as these are more difficult to quantify, are often estimated, or are based on standard, accepted assumptions [19,20]. Direct costs are those that are from materials and labor such the cost of the device, and health professionals’ time; direct costs are usually obtained directly from billing and other financial records [21]. Although there is consistency in what is considered a direct cost, there is marked variability in what is included as indirect costs, which contributes significantly to differences in reported cost effectiveness [19,20]. Variations in direct costs can occur; one cochlear implant program reported that an aggressive cost management program reduced direct costs by 22% [22]; if a similar cost containment effort was included in the calculations of cost, this could significantly change CUA calculations.
INTERPRETING COST EFFECTIVE CALCULATIONS So how to address the following two questions. First, if something is cost effective, what does that mean? Second, if one cochlear implant is cost effective, why are not two cochlear implants even more cost effective? An intervention is ‘‘cost effective’’ when it falls under a certain Cost/DQALY threshold, a threshold that is determined by what the purchaser is willing to pay. The willingness to pay (WTP) threshold is usually set at a number that reflects what percentage of a county’s gross domestic product the country is willing to spend and is stated in that country’s currency. For example in the United States, this is $50 000 [16]; in the United Kingdom, £30 000 (roughly $48 300) [6]; and in the Netherlands, s20 000 (roughly $28 600) [6]. If the Cost/DQALY is below the country’s WTP threshold, then it is considered ‘‘cost effective’’; as can be seen, what may be considered cost effective in one country, may be deemed as not cost effective in another. Reviews comparing pediatric cochlear implants, unilateral and bilateral, and adult unilateral cochlear implantation to no cochlear implantation [2–4,6–9] have found their Cost/DQALY within the United States WTP (
