QALYs and Value Assessment

QALYs and Value Assessment

Mayvis Rebeira PhD, Canadian Health Policy Institute

ABSTRACT

This literature review highlights the limitations of using Quality-Adjusted Life-Years or QALYs, in assessing the value of therapies through its use in cost-effectiveness analysis. Objective: To highlight key issues and limitations with the use of QALYs and present alternative methods to assess value-for-money for evaluating innovative therapies: Methods and Results: Key considerations highlighted include ethical considerations, use of QALYs for resource allocation, underestimating impact of therapy, ageist bias and double jeopardy with respect to people with disabilities or permanent ill-health. Methodological and other limitations covered include lack of established threshold in cost-effectiveness analysis and QALY derivation. Recommendations: Several recommendations are proposed that can either be used in conjunction with QALYs or replace the current QALY metric in value assessment. Conclusion: QALYs are deeply embedded in health technology assessments. For a long period, it represented the only measure that attempts to quantify the impact of a therapy on an individual’s health and wellbeing. Given the limitations highlighted here, use of QALYs in value assessments can lead to sub-optimal decisions and impact health outcomes of patients. The gradual inclusion of alternate measures can lead to better evaluation of value-for money of new interventions whilst enabling a more just and fair system for all patients.

ACKNOWLEDGMENTS: N/A

CITATION: Rebeira, Mayvis (2022). QALYs and Value Assessment. Canadian Health Policy, NOV 2022. Toronto: Canadian Health Policy Institute. ISSN 2562-9492, https://doi.org/10.54194/DFUL2957, www.canadianhealthpolicy.com.

CORRESPONDENCE:  [email protected]

DISCLAIMER: N/A

DISCLOSURE: No conflicts were declared.

OPEN ACCESS: N/A Not sponsored.

REVIEW: This article was subject to peer review.

SUBMITTED: 2 NOV 2022 | PUBLISHED: 24 NOV 2022

INTRODUCTION

Health technology assessments (HTAs) are used in many countries to evaluate the cost-effectiveness of new medical devices, diagnostics, biologics and pharmaceuticals. HTAs are separate from the processes used by regulatory agencies such as Health Canada or the Food and Drug Administration to evaluate new therapies for marketing approval based on efficacy and safety. HTAs undertake evaluations of value-for-money to guide Insurance formulary inclusion and reimbursement recommendations or decisions.

HTA generally involves measuring two components of a new technology: clinical effects and health effects. Clinical effects are measured using observational data preferably from Phase III randomized, placebo-controlled trials. The health effects are estimated using Quality-Adjusted Life-Years (QALYs). QALYs are the most common metric for measuring the effects of a therapy on a patient’s overall health and wellbeing. QALYs take into account both the extension of life and the quality of life as a result of a new therapy.

Many theoretical and practical limitations have been identified regarding the application of QALYs in HTA and remain unresolved (1,2). This paper reviews and highlights some of the key concerns. The discussion is focused on the limitations of QALYs which can lead to sub-optimal decisions or recommendations regarding funding and the use of new health technologies. Alternatives to QALYs are also reviewed and discussed. The purpose of this paper is to review the limitations and practical implications arising from the use of QALYs and to propose alternative methods for consideration.

LIMITATIONS OF QALY

Some of the main limitations of QALYs are summarized below.

Use of QALYs to Determine Cost-Effectiveness

Cost-effectiveness analysis (CEA) is based on a principle of the need to optimize resource allocation via rational and explicit methodologies. Cost-effectiveness analysis is utilized when where both costs and health benefits need to be considered and compared between the new therapy and its comparator or existing therapy.

The metric that is used to evaluate if a technology is cost-effective is the incremental costs-effectiveness ratio or ICER. ICER is the ratio of the incremental costs of using the new technology over the incremental benefits gained from the new technology over existing technology or status quo. The cost is expressed in monetary terms and the effects are expressed as quality-adjusted life years (QALYs). Hence ICER is reported as incremental costs per QALY (3,4).

The ICER is compared to a threshold to determine if an intervention is cost-effective. Interventions with ICERs above the particular cost-effectiveness threshold are deemed not to show value-for-money and can be excluded from public reimbursement or insurance coverage.  Thresholds can have a wide range based on past reimbursement decisions. The current threshold range that is generally used is not supported empirically and to-date, there remains a lack of research conducted both in Canada and elsewhere as to the determination of an appropriate scientifically derived threshold value for cost-effectiveness that may vary over time. In addition, several questions remain unresolved around the thresholds that are currently used:

• Should there be different thresholds for different drugs based on disease areas (e.g. drugs for cancer, drugs for rare diseases), different phases of life (e.g. end of life drugs, different age-groups (e.g. paediatrics drugs) or different geographic areas (rural verses urban)?
• Should the cost-effectiveness threshold be the same for all therapies respective of the mechanism of action or the innovative nature of the therapy?
• What is the appropriate cost-effectiveness threshold, or the estimate of the health forgone as public money is reallocated from its existing use towards the new technology?

Thresholds currently being used for decision-making are not objectively derived metrics and vary significantly between jurisdictions. It is generally assumed, based on previous HTA decisions, that the threshold can range between C$100,000 to $150,000 per QALY in Canada. In the UK, the threshold ranges between £20,000 and £30,000 per QALY for regular drugs and £50,000 per QALY for drugs used at the end of life (5). Braithwaite et al showed that in the US, the threshold is closer to approximately US$400K/ QALY in the high-end range (6). Other studies have shown that the threshold can be closer to US$300K/QALY to be considered cost-effective (7).

Ethical Considerations – QALYs and Resource Allocation

Several ethical issues have been raised regarding the distribution of health resources from the use of QALYs (8). One concern is that the metric can lead to resource allocation scenarios where one person’s life is valued higher or lower than another. A patient can exhibit a lower quality of life but that does not imply that that patient’s value of life is in any way lower that someone else who demonstrates a higher quality of life. QALYs further attempt to measure and determine a patient’s overall quality of life which extends far beyond what the intended main key outcome of a therapy in trying to improve a particular clinical outcome of the patient.

This issue has been described clearly by Rawles (9): “Distribution of resources according to best value for money, assessed as QALYs per unit cost leads to absurd anomalies. In the calculation of QALYs the implied value of life is no more than the absence of suffering. The use of QALYs for the comparison of treatments that are symptomatic or life-saving therefore leads to serious undervaluation of life and treatments that prolong it. Moreover, distribution of resources by best value for money, however assessed, is inequitable since for a given degree of suffering those whose illnesses happen to be cheaper to treat will be treated in preference to those whose treatments are more expensive”

This can also be viewed as not only contrary to the idea of universal and equitable health care coverage but goes against the ability of clinicians to select and provide the best and most suitable treatment for patients based on their professional judgement.  As illustration, consider the case of an oncology drug that extends a person’s life by one year. If ten people are treated by this drug, there is ten life-years to be gained in total. If there was another therapy that restores ten years of life to one in ten people who would have died otherwise, then the gain would also have been ten life years. Patients who require the most urgent care will generally be given priority by the clinician. The clinician determines this based on the severity of the health condition of the patient and the perceived degree of suffering of the patient. However, if the resource allocation is made using metrics such as cost per QALY gained, then the top priority may end up going towards patients with health conditions that can be treated at a lower price compared to a similar patient who requires more expensive treatments. Similarly, using these principles, a patient who has a shorter life expectancy but who requires major costly treatment could be given lower priority. This can be viewed as treating patients inequitably as it is not based on need but rather on ‘value for money’.

Hence, attempting to ration healthcare in line with ‘value for money’ through using QALY in CEA may result in decisions that tend to favour one group over the other rather than the medical need for therapy. As such, those who have rare diseases or ultra-rare diseases and whose treatments tend to be more costly face numerous hurdles to get access to the much-needed therapy. Similarly, patients who have conditions that can be treated relatively inexpensively is more likely to get an easier route for public reimbursement (and on private insurance formularies).

Ageist bias

QALYS have also been criticized as having an ageist bias. When the prime objective is maximizing QALYs with all other things being equal, this implies that there can be an ageist bias. This is because any additional life-years generated using a particular therapy or medical device is based on the current life-expectancy of that patient. The older a patient is, naturally the less life-years can be achieved by any intervention. A therapy that targets both younger and older patients can yield more life-years than a therapy that treats and targets only older patients (10).

Double Jeopardy – Disability and Permanent ill-health

In addition, a common criticism that has been raised with QALY is that it tends to put less value on preserving the lives of people with either a disability or permanent ill-health than it does for people who are either not disabled or who tend to generally consider themselves generally healthy. Disabled patients not only suffer from their disability, but they can also be given a lower priority to health care interventions and therapies that can possibly improve their health and prolong their lives (11, 12). Hence, we have the concept of double jeopardy. For disabled patients, they not only suffer from their disability, but they can also be given a lower priority to health care interventions and therapies that can possibly improve their health and prolong their lives.

Utilitarian

Critics have noted that the use of QALYs in decision-making makes it an approach that is too utilitarian since it is a highly quantitative approach to valuing life. This in turn does not necessarily always lead to optimal health outcomes for the patient or the population. The phrase ‘A QALY is a QALY is a QALY’ is often cited as a way to show the focus nature of this approach. It also reinforces the fact that the approach considers all QALYs equal regardless of a specific patient’s situation or any other factors that can usually play a pivotal role in a patient’s overall wellbeing

Underestimating impact of therapy

QALYs tend to underestimate the impact of a drug (13). The undervaluing of drugs can have far-reaching implications for patients that critically require the therapy. In complex interventions, there can be several overlapping measures where ‘gain of health’ is not the only measure. QALYs are therefore not appropriate in these situations, and they simplify complex scenarios into one measure that can lead to flawed decision-making (14). In addition, QALY is an aggregate metric that do not capture the full-extent of patient-level heterogeneity. Being an aggregate metric, it can underestimate the value of therapies for select population that is targeted by specialized therapies. This becomes bigger issue in an era of personalized medicine where genetic markers are now more commonly used for specialized therapies and drug discovery is increasingly focused on finding cures for targeted populations. Also, the use of QALYS can present more issues as clinical cases get more complex. Take for example the case of oncology. A cancer patient’s quality of life may deteriorate in the beginning of the treatment due to side-effects of the drug (e.g., chemotherapy). QALYs are then gained much later in the pathway of care as the patient improves in health over the longer-term.

Methodological Issues of QALY Derivation

QALY values are derived from Quality of Life (QoL) survey instruments that attempt to measure patient’s perceptions of quality of life and can include health, physical, psychological and social dimensions. In order to use the information to derive value-for-money analysis through a cost-effectiveness model, the health effects data from these surveys need to be translated into a quantitative measure called utilities reflecting the many different dimensions of health such as mobility, self-care, usual activities, discomfort and anxiety. Many of the current quality of life scales that are used to measure health benefits of a new intervention involve using a Quality of Life (QoL) survey that patients complete during the trial period. Unlike end points of a clinical study that is targeted towards the focussed clinical benefits of the intervention, QoL instruments were developed to gather information that tend to go beyond the scope of the intended clinical benefit of a therapy.

These surveys reflect respondents’ own interpretation of his or her overall quality of life. It is recognized that measuring quality of life is a challenging process with the existence of many different quality of life scales. Different utility elicitation instruments can bring about different and wide-ranging results, producing different utility values for the same health outcome. Further, utility values do not account for the heterogeneity of the patient population or the initial health state of the patient and this further confounds its use in HTA (15). The use of QoL and utility elicitation instruments to derive QALYs may have important implications for its use in cost-effectiveness analysis of new healthcare interventions (16, 17).

Measuring quality of life is itself a challenging process with the existence of many different quality of life scales. Selecting the right instrument in a clinical trial is itself a complex and strategic process. Further, different utility elicitation instruments can bring about different and wide-ranging results.

The concern with the use of QALYs derived from surveys of patient preferences is that the metric is inevitably a reflection of subjective data which is prone to variability related to factors like survey population characteristics and other contextual issues. This is compounded by the lack of any empirically derived CEA thresholds, which can reduce confidence that the ICER metric produces meaningful results. The ratios are also highly sensitive to QALYs and small differences in QALYs can produce large variations in ICER.

Assessing value without QALY

Globally, different approaches have emerged to reward the value added on by new and innovative drugs. In countries such as the Netherlands, Ireland, Sweden and the UK, the price of new drugs and the access to a new drug has to be assessed and justified by the health gain it delivers (i.e. QALY gain) compared with existing approved therapy. On the other hand, in countries such as France and Germany, the assessment of added benefit is expressed on an ordinal scale. This ordinal scale is based on an assessment of the clinical outcomes compared with existing care. It is this assessment (and not QALYs gain) that ultimately influences price negotiations and access to the therapy by patients within these countries. It demonstrates that it is possible to undertake value assessments of new therapies without the need to use QALYs and cost-effectiveness results.

Productivity

Productivity costs are normally omitted from economic evaluations and health technology assessments even though they have significant impact on cost-effectiveness outcomes. One reason for this is that HTAs are normally just focused on a narrow perspective – that of the budget of therapies and health budget rather than looking at a broader societal perspective. Further, there is currently a lack of standardization regarding the methodology to be used to estimate productivity costs. Productivity losses should incorporate losses related to absenteeism, reduced productivity at paid work and potentially losses related to unpaid work. More accurate and sophisticated methods that exists currently such as human capital approach and friction costs can be used to calculate productively costs so that these costs can also be considered for decision-makers. The exclusion of productivity costs and the use of QALYs together increase the uncertainty of the ICER result.

OPTIONS

This paper has elaborated some pertinent issues with the use of QALYs. There exists alternative forms of metrics and assessments (18,19) that can be used in place of QALYs or in conjunction with QALYs and these are described below.

Life-Years Gained (LYG)

Life-Years Gained (LYG) is a simpler measure as it addresses some of the fundamental issues with the use of QALYs. LYG measures any gains in length of life evenly regardless of the ability of the therapy to improve the quality of life. Hence, it removes the challenge of trying to identify how a therapy improves overall quality of life that spans across different health conditions of patients, different severities of illness, different therapies and importantly, a patient’s own interpretation of his or her own overall wellbeing and quality of life.

LYG also has its own limitations. For example, in cases where a new drug reduces adverse events or side effects compared to existing therapy, quality of life can differ for patients that have access to this drug and this improvement may not be considered if quality improvements of life is not captured. In addition, the difference in cost-effectiveness analysis results using QALYS or LYGs may not vary much for many conditions. Some studies have reported that cost per QALY and cost per LY assessments tend to have similar results and the differences are small (20). Though LYG does have its own limitations, it still presents an important metric that address many of the fundamental flaws of QALYs.  As such, for ethical reasons highlighted earlier in this paper, it’s important for decision-makers to consider both metrics and consider placing a higher weight on LYG for populations that can be impacted negatively by QALYs.

Quality and Risk-Adjusted Life-Years (QRALYs)

Alternate approaches have been developed notably to account for the fact that QALYs underestimate the value that a drug brings to those with severe illnesses (19). One consideration is to include patient preferences through risk adjusting the QALY. This new metric sometimes referred to as Quality and Risk-Adjusted Life-Years (QRALYs) can replace QALYs and be used in the same manner in cost-effectiveness analysis. The two key components of risk adjustment include incorporating:

• Value of hope: This occurs in cases when a therapy may not help a majority of patients but only a smaller minority of patients. This can be important in therapeutic areas such as oncology where there are cases where the median survival does not improve but the therapy does improve outcomes for a smaller portion of patients.
• Value of Risk Protection: The inclusion of value of risk protection is pertinent in cases where there is a severe illness and QALYs have underestimated the value of an innovative therapy for these conditions. The more severe these conditions are, the greater the value can be under-assessed using QALYs. The value of risk protection therefore will be higher for diseases with higher burden of illnesses.

Healthy Year Equivalent (HYE)

Healthy Year Equivalent (HYE) has also been suggested as a method superior to QALYs. In HYE, patients’ preferences are measured over the entire path of health states through which a patient goes through instead of a static preference in just one health state. The way this is done is by measuring the utility for a specific path and then to determine the health years that would provide that same utility. Though conceptually it represents a better alternative, it has been acknowledged that in practical terms, it may be challenging to implement (21).

Use of real-world data to augment QALYs

There is a lot of value in incorporating real-world data and evidence in arriving at decisions on funding. Decisions that are based solely on incremental cost/QALY at the point of market entry may ignore critical information on the effectiveness of the drug in real-world. Real-world evidence based on patient outcomes and clinical practices in jurisdictions that have usage of these drugs can provide important new information to decision-makers on the value of a therapy to its population. Incorporating real-world data enables new patient outcomes information to be re-evaluated post marketing approval of a drug.

Cost-effectiveness analysis relies on clinical data from pivotal trials. These trials may have homogenous participants that may not reflect the demographics or diversity of individuals in the broad patient population. In addition, practice patterns in real-life may be different from that in the clinical trials. In some cases, for example, some of the therapies for managing side effects may not be available in Canada. Comorbidities and complexities of cases in patients could be different than that of participants in the trails and patients could be managed for a wide arrays of health issues concurrently in addition to the condition being managed by the therapy. Subgroup analysis of trial participants can be one way of determining value-for-money for critical groups of patients and the ability to return for re-assessments with real-world data should be encouraged.

Perspective

New health technologies can have a financial impact that goes beyond the dedicated budget for medical devices or pharmaceuticals. There can be longer-term savings and cost avoidance to the health system due to these technologies, e.g. avoidance of emergency surgeries, fewer hospital readmissions or shorter length of stay in a hospital, less physician visits, more minor side effects which all can impact other areas of the healthcare system. Similarly, there can be efficiencies gained outside the healthcare system such as lower absenteeism, increased productivity of workers and less out of pocket costs for patients and caregivers. Current use of cost/QALY analysis narrows the focus on just the health system perspective or in some cases even on a narrower scale. The need to incorporate societal perspectives could be included as a required and necessary secondary analysis. It is recognized that such quantification can be challenging and time-consuming but a high-level thorough understanding of the impact to the broader system as a whole is an important dimension to consider for decision-makers.

Multi-criteria Decision Analysis (MCDA) Framework

Multi-criteria Decision Analysis (MCDA) framework is a more comprehensive way to undertake evaluations as it recognizes the multiple facets, criteria and objectives that need to be considered when making these complex decisions that have huge impact on patients. MCDA frameworks captures the multi-faceted value of therapies that extend beyond a single cost/QALY criteria. It was originally developed by Keeney and Raiffa (22) to enable decisions in the face of multiple objectives instead of a single objective of maximizing QALYs. The use of MCDA enables considerations of therapies that have larger societal benefits and can ensure far better optimal allocation of public resources in the longer-term. This is especially useful in the face of complex interventions and targeted therapies as we enter into an era personalized medicine. There is however a higher degree of complexity to undertake MCDA. There are jurisdictions have used this in a timely manner (e.g. EMA Committee for Medical Products for Human Use) and lessons can be learnt on how it is used effectively in these committees in order to widely expand the use of this framework for reimbursement decisions. Decision-makers would also benefit from a simpler, more transparent, structured and more easily understood framework. MCDA decision making will also be more aligned with the values of society, funding agencies and of the patients, caregivers and health care providers.

CONCLUSION

QALYs are very embedded in health technology assessments. However, currently QALYs as used in cost-effectiveness analysis to determine value-for-money by funding agencies can lead to sub-optimal decisions and impact health outcomes of specific populations due to the inherent flaws of the metric.  It can also result in decisions that could discriminate on care if that is the only metric used for funding decisions. Some important limitations in using QALYs have been highlighted and several alternate methods have been proposed. One of the best ways to mitigate some of the challenges of QALYs is to incorporate alternate methods such as LYG and QRALY in an overall MCDA framework using a societal perspective. This can lead to a more fair and representative way of assessing the value of therapies.

REFERENCES

1. Neumann, P. J., Cohen, J. T., & Weinstein, M. C. (2014). Updating cost-effectiveness—the curious resilience of the $50,000-per-QALY threshold. N Engl J Med, 371(9), 796-797.
2. Rebeira, Mayvis (2016). Limitations of Economic Evaluations for Health Technology Assessments in Canada. Canadian Health Policy, October 17, 2016. Toronto: Canadian Health Policy Institute. URL: www.canadianhealthpolicy.com
3. NICE UK (National Institute for Health and Care Excellence) (https://www.nice.org.uk/); GRADE website (www.gradeworkinggroup.com)
4. Drummond, M. F., Sculpher, M. J., Claxton, K., Stoddart, G. L., & Torrance, G. W. (2015). Methods for the economic evaluation of health care programmes. Oxford university press.
5. McCabe, C., Claxton, K., & Culyer, A. J. (2008). The NICE cost-effectiveness threshold. Pharmacoeconomics, 26(9), 733-744.
6. Braithwaite, R. S., Meltzer, D. O., King Jr, J. T., Leslie, D., & Roberts, M. S. (2008). What does the value of modern medicine say about the $50,000 per quality-adjusted life-year decision rule? Medical care, 349-356.
7. Kaura, S., Nanavaty, M., Seetasith, A., Nyandege, A., & Khan, Z. M. (2015). Literature review of the use of ICER thresholds in Healthcare decision-making. Value in Health, 18(3), A90.
8. Pettitt, D. A., Raza, S., Naughton, B., Roscoe, A., Ramakrishnan, A., Ali, A., … & Brindley, D. A. (2016). The limitations of QALY: a literature review. Journal of Stem Cell Research and Therapy, 6(4).
9. Rawles, J. (1989). Castigating QALYs. Journal of medical ethics, 15(3), 143-147.
10. Kappel, K., & Sandøe, P. (1992). QALYs, age and fairness. Bioethics, 6(4), 297-316.
11. Singer, P., McKie, J., Kuhse, H., & Richardson, J. (1995). Double jeopardy and the use of QALYs in health care allocation. Journal of medical ethics, 21(3), 144.
12. Broome, J. (1994). Fairness versus doing the most good. The Hastings Center Report, 24(4), 36-39.
13. Neumann, P. J., & Cohen, J. T. (2018). QALYs in 2018—advantages and concerns. Jama, 319(24), 2473-2474.
14. Payne, K., McAllister, M., & Davies, L. M. (2013). Valuing the economic benefits of complex interventions: when maximising health is not sufficient. Health economics, 22(3), 258-271.
15. Torrance, G. W. (1986). Measurement of health state utilities for economic appraisal: a review.Journal of health economics,5(1), 1-30.
16. Torrance, G. W. (1970). A generalized cost-effectiveness model for the evaluation of health programs.
17. Harris, J. (1987). QALYfying the value of life. Journal of medical ethics, 13(3), 117-123.
18. Whitehead, S. J., & Ali, S. (2010). Health outcomes in economic evaluation: the QALY and utilities. British medical bulletin, 96(1), 5-21.
19. Lakdawalla, D. N., Doshi, J. A., Garrison Jr, L. P., Phelps, C. E., Basu, A., & Danzon, P. M. (2018). Defining elements of value in health care—a health economics approach: an ISPOR Special Task Force report [3]. Value in Health, 21(2), 131-139.
20. Chapman, R. H., Berger, M., Weinstein, M. C., Weeks, J. C., Goldie, S., & Neumann, P. J. (2004). When does quality‐adjusting life‐years matter in cost-effectiveness analysis?. Health economics, 13(5), 429-436.
21. Mehrez, A., & Gafni, A. (1991). The healthy-years equivalents: how to measure them using the standard gamble approach. Medical Decision Making, 11(2), 140-146.
22. Keeney, R. L., Raiffa, H., & Meyer, R. F. (1993). Decisions with multiple objectives: preferences and value trade-offs. Cambridge university press.