Acute respiratory distress syndrome (ARDS) is a severe form of hypoxemic respiratory failure often encountered in critically ill patients, with an overall in hospital mortality rate close to 40% (1 , 2). Although supportive care using lung protective ventilation remains the cornerstone of treatment for ARDS, there is keen interest in novel therapies that could improve survival (1). Extracorporeal membrane oxygenation (ECMO) has recently gained interest as a therapeutic option (3 , 4). However, both its clinical efficacy and its cost-effectiveness are not certain.
The evidence supporting the use of venovenous ECMO for severe ARDS is inconsistent. Early studies did not show a significant mortality benefit with the use of ECMO for ARDS (5 , 6). However, in 2009, the CESAR trial reported significant improvement in 6-month disability-free survival with referral to an ECMO center compared with conventional ventilation (7). Subsequently, following the 2009 influenza A (H1N1) pandemic, two observational studies were published that reported both neutral and positive outcomes with venovenous ECMO for ARDS (8 , 9). When the results of these three studies were combined in systematic reviews and meta-analyses, a mortality benefit was seen in favor of venovenous ECMO (1 , 10). More recently, the much-anticipated results of the ECMO for severe ARDS (EOLIA) trial, a randomized control trial comparing venovenous ECMO to lung protective ventilation for patients with severe ARDS, were published showing no significant difference in 60-day mortality between the two groups (relative risk, 0.76; 95% CI, 0.55–1.04) (11). However, interpretation of these results is complicated by significant crossover from the control group to venovenous ECMO therapy (28%) and by early termination of the study for futility (11–13). Although the evidence for venovenous ECMO is not conclusive, it continues to garner interest as a therapeutic option for these critically ill patients (4 , 14). In the absence of clear evidence to support the use of venovenous ECMO for all patients with ARDS, clinicians must carefully consider its appropriateness when making bedside decisions.
Policy-makers tasked with decisions related to the funding of venovenous ECMO within a health system must consider both its efficacy and its cost-effectiveness. Currently, there is little data on the cost-effectiveness of venovenous ECMO for ARDS. The CESAR trial included an economic evaluation and reported an incremental cost-effectiveness ratio (ICER) of £19,252 ($42,356 2016 Canadian dollars [CAD])/quality-adjusted life year (QALY) (7). The authors concluded that venovenous ECMO was cost-effective for ARDS patients. However, this study has been criticized as being an evaluation of a program of referral to a venovenous ECMO center for patients with severe ARDS rather than one of ECMO therapy itself and for the lack of consistent lung protective ventilation in the control group (7 , 15–17). These concerns limit the applicability of the results. We therefore undertook an economic evaluation of venovenous ECMO for ARDS using the best available efficacy data to estimate the treatment’s cost-effectiveness. We sought to determine the cost-utility of venovenous ECMO therapy for adult patients with severe ARDS compared with lung protective ventilation in terms of QALYs gained over a predicted lifetime.
We performed a model-based cost-utility analysis (CUA) of venovenous ECMO compared with standard lung protective mechanical ventilation. This type of economic evaluation expresses outcomes as the ICER, which is the incremental cost associated with a new therapy needed to generate one additional QALY. This methodology for economic evaluation is used by many health system payers and health technology assessment (HTA) organizations (18–21). In our study, the intervention of interest is venovenous ECMO as described in international guidelines (22 , 23), and the comparator is lung protective ventilation, defined by a plateau pressure of less than 30 cm H2O, and tidal volumes of 4–8 mL/kg of predicted body weight, which is considered the standard of care for patients with ARDS (4 , 24–26).
Cohort State Transition Model
We used a cohort state transition model to determine the life years, costs, and QALYs associated with venovenous ECMO and standard lung protective ventilation. These models are premised on a hypothetical cohort of patients who can exist in only one of a finite number of health states. Patients move between these health states according to transition probabilities, independent of previous or future transitions. Each health state is assigned a specific cost and a specific health state utility value representing its quality of life. The cohort of patients is simulated to move through the various health states, accumulating costs and health state utility values specific to their path through the model. It is then possible to compare the averages of the total costs and QALYs generated by the model for the cohorts of patients receiving different interventions. The results are presented as an ICER of costs/QALY gained. Sensitivity analyses are performed to explore uncertainty in the model and its outputs.
Our model had three health states: ARDS, recovered, and dead. Patients started in the ARDS health state and then transitioned to the recovered or dead health state. Death was an absorbing state. The age of patients at the start of the model was 45 years since this is the mean age of the patients in the meta-analysis describing venovenous ECMO therapy. Cycle length was 1 year, and the model was run until death. A graphical representation of the model is presented in Figure 1.
One-way deterministic sensitivity analyses (DSA) were performed to explore uncertainty in the model. A probabilistic sensitivity analysis (PSA) was performed to simultaneously assess the effects of uncertainty across all the stochastic model variables. Results of this analysis were plotted on the cost-effectiveness plane and the cost-effectiveness acceptability curve to graphically express the results across a range of acceptable cost-effectiveness thresholds (27 , 28).
We followed reference case recommendations endorsed by international HTA agencies (19 , 21 , 29). We adopted the perspective of the Canadian health system payer. The model used a lifetime time horizon, as recommended in guidelines for circumstances in which a mortality difference across strategies is present (21). All costs were converted to and are reported as 2016 CAD using best practices methods (21 , 27 , 30). In keeping with Canadian HTA guidelines, costs and QALYs beyond year 1 were discounted at 1.5%, reflecting society’s lower valuation of future costs and health outcomes (21). The cohort state transition model and all calculations and sensitivity analyses were constructed in Excel (Microsoft Corporation, Redmond, WA).
See Table 1 for model variable details. Transition probabilities from the ARDS health state were based on the results of a meta-analysis published by Munshi et al (1) that reports outcomes for patients with severe ARDS who had received venovenous ECMO or standard mechanical ventilation. Transition from the recovered health state to death was based on Canadian census life tables (31).
The model’s sensitivity to the efficacy of venovenous ECMO was explored in sensitivity analyses. A DSA was performed using the 95% CI values around the relative risk (RR) of death used to inform transition probabilities from the ARDS state. Further DSAs were performed using efficacy data for venovenous ECMO taken from a second systematic review and meta-analysis published by Zampieri et al (10). This meta-analysis used different methods of pooling and matching data from the primary studies than was used by Munshi et al (1). We also performed an additional sensitivity analysis using the results from the EOLIA trial that compared venovenous ECMO to mechanical ventilation for ARDS (11). The significant crossover of patients in this study from the control group to venovenous ECMO makes analysis and pooling of these data with previous meta-analyses challenging. We therefore chose not to pool these data. We instead performed a sensitivity analysis that used the reported primary outcome (60-d mortality) of the intention-to-treat analysis to inform transition probabilities from the ARDS health state for the two interventions.
Utility values and costs for the health states were obtained from the literature or calculated based on published data. Changes to utility values and costs were explored in sensitivity analyses. Additional sensitivity analyses using a discount rate of 0% and 3%, and using a 5- and 10-year time horizon were also performed (for comprehensive details regarding model design and inputs, see the Supplementary Appendix, Supplemental Digital Content 1, http://links.lww.com/CCM/E50). This study gained ethical approval from the London School of Hygiene and Tropical Medicine (LSHTM MSc Ethics Reference: 10148).
Short-term survival (< 1 yr) was higher amongst the cohort that received venovenous ECMO compared with the cohort that received lung protective ventilation (Table 2). Longer-term survival (beyond 1 yr) was similar between the two cohorts. Treatment with venovenous ECMO was associated with a gain of 5.2 additional life years and 4.05 additional QALYs compared with lung protective ventilation. The short-term survival effect drives the difference in long-term survival between the groups. Venovenous ECMO was associated with higher lifetime healthcare costs of $145,697. These additional costs are the result of the higher costs associated with the provision of venovenous ECMO, as well as the subsequent lifetime healthcare costs accrued during the additional life years that were gained with venovenous ECMO. The ICER for venovenous ECMO compared with lung protective ventilation was $36,001/QALY.
From the sensitivity analyses using different probabilities for mortality, we see that small changes in the inputs for the efficacy of venovenous ECMO or mechanical ventilation result in large changes in the value of the ICER (Fig. 2; and Table E3, Supplemental Digital Content 4, http://links.lww.com/CCM/E53). For example, in the sensitivity analysis using the high value of the 95% CI for the RR of mortality with venovenous ECMO, a 26% change from the base case for the probability for mortality with venovenous ECMO was associated with a 57% change in the ICER. Similarly, when the probabilities of mortality reported in the EOLIA trial were used, the probabilities of death for the two groups changed from the base case by only 6% and 11%, yet the ICER increased by 67%. These results indicate that the model is sensitive to the efficacy of venovenous ECMO. For the DSA exploring the effects of changes in costs, the ICER changed proportionally to changes in the variable. The ICER changed little with variations in utility values for 6 and 12 months after hospitalization and was therefore robust to changes in the utility values of ARDS survivors. Scenario analyses using a 5- and 10-year time horizon yield an ICER of $198,601 and $99,438, respectively. This suggests that it is the QALYs gained beyond 5 or 10 years of survival that drive the ICER into the cost-effective range.
The mean ICER for these simulations is $33,066, which is slightly lower than the ICER for the base-case scenario. Figure 3 shows the cost-effectiveness acceptability curves for the results of the PSA of venovenous ECMO and lung protective ventilation. This curve shows the probability that each of the therapies is cost-effective across a range of potential acceptable cost-effectiveness thresholds.
Our study shows that venovenous ECMO is associated with a lifetime gain of 5.2 life years and 4.04 QALYs when compared with lung protective ventilation for the treatment of severe ARDS in young adults 45 years old. It is associated with an incremental increase in costs of $145,697 and an ICER of $36,001/QALY. We believe that venovenous ECMO is likely cost-effective when used for patients with severe ARDS assuming an acceptable threshold of $50,000–100,000/QALY. This is the first analysis to define the CUA for venovenous ECMO using efficacy data from published meta-analyses and to incorporate the results from the EOLIA trial.
The results of our study can help decision-makers determine if venovenous ECMO represents a good value for money for patients with ARDS when compared with other potential interventions. This is not the first reported economic evaluation of a potential ARDS therapy. The OSCAR trial, which compared high frequency oscillation ventilation (HFOV) to standard care for ARDS patients, also included an economic evaluation as part of their analysis (32). This study used trial data to inform costs and QALYs generated up to 1 year post randomization, and then used a cohort state transition model, similar in design to ours, to generate the lifetime accumulated costs and QALYs (32). They reported an ICER of £78,260 ($148,312 CAD) per QALY gained with the use of HFOV compared with conventional ventilation when analyzed using a lifetime time horizon and from the perspective of the UK health system (32). There was no clinical benefit identified with the use of HFOV in this study (and HFOV is no longer recommended in treatment guidelines), and the economic evaluation revealed that HFOV was not cost-effective compared with standard treatment (4 , 32). The CESAR trial also conducted an economic evaluation as part of their study. It similarly used trial data to inform short-term economic outcomes and a cohort state decision model to generate lifetime costs and QALYs (33). The authors reported an ICER of £19,252 ($42,356 CAD)/QALY gained with the use of referral to an ECMO center compared with mechanical ventilation, using a lifetime time horizon and UK health system perspective (7). The results of the OSCAR study showed that HFOV does not have clinical or economic benefits for patients with ARDS. However, both the CESAR trial and our study suggest that venovenous ECMO has both clinical and economic benefits when used for patients with severe ARDS.
It must be stated that direct comparisons of ICERs are problematic as they are highly context specific (34 , 35). Methodological differences including costing methods, discount rate, time horizon, reference therapy, or perspective can all significantly change the value of the ICER (35). Ideally, ICERs should be interpreted in reference to a relevant acceptable cost-effectiveness threshold, with values below the threshold representing cost-effectiveness (34). In the United States, the threshold is often quoted at $50,000/QALY; however, the origins of this are dubious, and the Affordable Care Act prohibits the use of cost per QALY benchmarks (36 , 37). The World Health Organization has suggested a threshold equal to three times per capita gross domestic product (37). This would place the threshold well over $100,000/QALY for many high-income countries. The UK National Institute for Health and Care Excellence (NICE) identifies cost-effectiveness at an explicit threshold of £20,000–30,000 ($35,696–53,544 CAD) CAD)/QALY. This threshold has been criticized as being insufficient for therapies treating rare or terminal conditions, for which £100,000 ($178,480 CAD)/QALY has been proposed. Interestingly, NICE’s threshold has also been criticized as being too high. A threshold closer to £13,000 ($23,202 CAD)/QALY has been proposed that considers the opportunity cost associated with the disinvestment of previously funded therapies in order to fund new ones (38). Indeed, the acceptable threshold will vary based on the health system’s budget, the payer, as well as local societal and ethical values (34). These factors must be considered when interpreting the results of an economic evaluation like ours.
Generalizing the results of our study comes with caveats. The efficacy data for venovenous ECMO used in the model came from high-volume centers (1). Observational data suggest that less experienced centers with lower volumes of venovenous ECMO appear to have higher rates of mortality (39). Given our model’s sensitivity to its efficacy, venovenous ECMO may not be cost-effective when delivered at lower-volume centers where the outcomes may be poorer. We believe our study does not inform on the cost-effectiveness of the use of venovenous ECMO at satellite locations with lower volumes or less experience.
Second, we modeled a cohort 45 years old and used efficacy data largely based on patients who were previously healthy with influenza A (H1N1). The ICER for our sensitivity analysis using a 5-year time horizon exceeds the widely accepted willingness to pay threshold of $50,000/QALY, suggesting that it is the additional QALYs generated over a lifetime beyond 5 years that help to drive the ICER into the cost-effective range. This information, along with the growing body of evidence demonstrating the increased morbidity and mortality following critical illness among older and frailer patients, suggests that venovenous ECMO may not be cost-effective for critically ill patients with a lower likelihood of full recovery or long life-span (40–42). Therefore, it is important to note that without further evidence, our results are not generalizable to populations with different long-term predicted outcomes.
Our study has several methodological limitations. First, our results rely on the QALY, which has been criticized for being biased against older or disabled populations, and end of life care (36). However, no superior alternative has been proposed, and the QALY remains an important metric that incorporates both the length and quality of life as an outcome and has even been proposed as an outcome measure for critical care trials (36 , 43). Second, given limited published costing data, we had to incorporate values from different jurisdictions. The first 6 months in the ARDS cycle was populated by UK data, after which all costing data was from Canadian sources. Although not the ideal practice for economic modeling, we believe our results remain valid as the UK and Canadian medical systems are similarly funded and organized. Furthermore, we attempted to minimize jurisdictional differences in the costs by converting them to a common currency, taking into account differences in purchasing power and inflation. Like many economic evaluations, we were limited by the data available as inputs into the model. However, we feel that these shortcomings were adequately explored through sensitivity analyses and that our results are sufficiently robust despite these limitations.
Our analysis has several strengths. First, this CUA was conducted using modeling best practices endorsed by UK, Canadian, and international HTA organizations (19 , 21 , 29). Second, unlike the CESAR trial, which based its economic evaluation on efficacy data of referral to an ECMO center, this study used efficacy data specific to the use of venovenous ECMO (7). As a result, our study directly addresses the cost-effectiveness of venovenous ECMO therapy itself, which is important to decision-makers when they consider the funding of venovenous ECMO. Third, this study used efficacy data based on published meta-analyses and explored the results of the recently completed EOLIA trial. This enhances our estimate of the efficacy of venovenous ECMO and increases the confidence in our results. It is reassuring to see that while the results of the EOLIA trial were nonsignificant, when incorporated into our model, they still generated additional QALYs and life years at a reasonably acceptable incremental cost. We look forward to future analyses of these data that can be incorporated into future economic evaluations on this subject. Finally, we enhanced our model’s validity by incorporating the results of recent work that describes the reduced quality of life and increased healthcare costs of ARDS survivors (40).
We believe our results suggest that venovenous ECMO for ARDS is likely cost-effective for young adult patients with severe ARDS from the perspective of the healthcare system. Until additional data are available to inform on the cost-effectiveness in different patient populations, we believe it should be considered on a case-by-case basis, in patients who have a high likelihood of a good long-term functional outcome, and limited to expert, high-volume ECMO centers.
1. Munshi L, Telesnicki T, Walkey A, et al. Extracorporeal life support for acute respiratory failure. A systematic review and metaanalysis. Ann Am Thorac Soc 2014; 11:802–810
2. Bellani G, Laffey JG, Pham T, et al; LUNG SAFE Investigators; ESICM Trials Group: Epidemiology, patterns of care, and mortality for patients with acute respiratory distress syndrome in intensive care units in 50 countries. JAMA 2016; 315:788–800
3. Fan E, Gattinoni L, Combes A, et al. Venovenous extracorporeal membrane oxygenation
for acute respiratory failure: A clinical review from an international group of experts. Intensive Care Med 2016; 42:712–724
4. Fan E, Del Sorbo L, Goligher EC, et al; American Thoracic Society, European Society of Intensive Care Medicine, and Society of Critical Care
Medicine: An official American Thoracic Society/European Society of Intensive Care Medicine/Society of Critical Care
Medicine Clinical Practice Guideline: Mechanical ventilation in adult patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2017; 195:1253–1263
5. Zapol WM, Snider MT, Hill JD, et al. Extracorporeal membrane oxygenation
in severe acute respiratory failure. A randomized prospective study. JAMA 1979; 242:2193–2196
6. Morris AH, Wallace CJ, Menlove RL, et al. Randomized clinical trial of pressure-controlled inverse ratio ventilation and extracorporeal CO2
removal for adult respiratory distress syndrome. Am J Respir Crit Care Med 1994; 149:295–305
7. Peek GJ, Mugford M, Tiruvoipati R, et al; CESAR trial collaboration: Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation
for severe adult respiratory failure (CESAR): A multicentre randomised controlled trial. Lancet 2009; 374:1351–1363
8. Pham T, Combes A, Rozé H, et al; REVA Research Network: Extracorporeal membrane oxygenation
for pandemic influenza A(H1N1)-induced acute respiratory distress syndrome: A cohort study and propensity-matched analysis. Am J Respir Crit Care Med 2013; 187:276–285
9. Noah MA, Peek GJ, Finney SJ, et al. Referral to an extracorporeal membrane oxygenation
center and mortality among patients with severe 2009 influenza A(H1N1). JAMA 2011; 306:1659–1668
10. Zampieri FG, Mendes PV, Ranzani OT, et al. Extracorporeal membrane oxygenation
for severe respiratory failure in adult patients: A systematic review and meta-analysis of current evidence. J Crit Care 2013; 28:998–1005
11. Combes A, Hajage D, Capellier G, et al; EOLIA Trial Group, REVA, and ECMONet: Extracorporeal membrane oxygenation
for severe acute respiratory distress syndrome. N Engl J Med 2018; 378:1965–1975
12. Harrington D, Drazen JM. Learning from a trial stopped by a data and safety monitoring board. N Engl J Med 2018; 378:2031–2032
13. Hardin CC, Hibbert K. ECMO for severe ARDS. N Engl J Med 2018; 378:2032–2034
14. Cooper DJ, Hodgson CL. Extracorporeal membrane oxygenation
rescue for H1N1 acute respiratory distress syndrome: Equipoise regained. Am J Respir Crit Care Med 2013; 187:224–226
15. Jones A, Barrett N, Scales D, et al. Ventilatory support versus ECMO for severe adult respiratory failure. Lancet 2010; 375:550–551; author reply 551
16. Jainandunsing JS, Ismael F. Ventilatory support versus ECMO for severe adult respiratory failure. Lancet 2010; 375:549; author reply 551
17. Busch T, Laudi S, Kaisers U. Ventilatory support versus ECMO for severe adult respiratory failure. Lancet 2010; 375:549; author reply 551
18. Kyeremanteng K, Wan C, D'Egidio G, et al. Approach to economic analysis in critical care
. J Crit Care 2016; 36:92–96
19. National Institute for Health and Care Excellence: Guide to the Methods of Technology Appraisal 2013. 2013. Available at: nice.org.uk/process/pmg9
. Accessed February 2, 2015
20. Siebert U, Alagoz O, Bayoumi AM, et al; ISPOR-SMDM Modeling Good Research Practices Task Force: State-transition modeling: A report of the ISPOR-SMDM Modeling Good Research Practices Task Force–3. Value Health 2012; 15:812–820
21. Canadian Agency for Drugs and Technologies in Health: Guidelines for the Economic Evaluation of Health Technologies: Canada. 2017Fourth Edition. Ottawa, CADTH.
22. Peek GJ, Clemens F, Elbourne D, et al. CESAR: Conventional ventilatory support vs extracorporeal membrane oxygenation
for severe adult respiratory failure. BMC Health Serv Res 2006; 6:163
23. Extracorporeal Life Support Organization: ECMO Specialist Training Manual. 2010Third Edition. Ann Arbor, Extracorporeal Life Support Organization.
24. Slutsky AS, Ranieri VM. Ventilator-induced lung injury. N Engl J Med 2013; 369:2126–2136
25. Ferguson ND, Fan E, Camporota L, et al. The Berlin definition of ARDS: An expanded rationale, justification, and supplementary material. Intensive Care Med 2012; 38:1573–1582
26. Brower RG, Matthay MA, Morris A, et al; Acute Respiratory Distress Syndrome Network: Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342:1301–1308
27. Gray AM, Clarke PM, Wolstenholme JL, et al. Applied Methods of Cost-Effectiveness Analysis in Health Care. 2011New York, NY, Oxford University Press.
28. Briggs AH, Weinstein MC, Fenwick EA, et al; ISPOR-SMDM Modeling Good Research Practices Task Force: Model parameter estimation and uncertainty: A report of the ISPOR-SMDM Modeling Good Research Practices Task Force–6. Value Health 2012; 15:835–842
29. Caro JJ, Briggs AH, Siebert U, et al; ISPOR-SMDM Modeling Good Research Practices Task Force: Modeling good research practices–overview: A report of the ISPOR-SMDM Modeling Good Research Practices Task Force-1. Med Decis Making 2012; 32:667–677
30. Fox-Rushby J, Cairns J. Economic Evaluation. 2005London, Open University Press.
32. Lall R, Hamilton P, Young D, et al; OSCAR collaborators: A randomised controlled trial and cost-effectiveness analysis of high-frequency oscillatory ventilation against conventional artificial ventilation for adults with acute respiratory distress syndrome. The OSCAR (OSCillation in ARDS) study. Health Technol Assess 2015; 19:1–177, vii
33. Peek GJ, Elbourne D, Mugford M, et al. Randomised controlled trial and parallel economic evaluation of conventional ventilatory support versus extracorporeal membrane oxygenation
for severe adult respiratory failure (CESAR). Health Technol Assess 2010; 14:1–46
34. Cohen DJ, Reynolds MR. Interpreting the results of cost-effectiveness studies. J Am Coll Cardiol 2008; 52:2119–2126
35. Weintraub WS, Cohen DJ. The limits of cost-effectiveness analysis. Circ Cardiovasc Qual Outcomes 2009; 2:55–58
36. Neumann PJ, Cohen JT. QALYs in 2018—advantages and concerns. JAMA 2018; 1–2
37. Neumann PJ, Cohen JT, Weinstein MC. Updating cost-effectiveness–the curious resilience of the $50,000-per-QALY threshold. N Engl J Med 2014; 371:796–797
38. Claxton K, Martin S, Soares M, et al. Methods for the estimation of the National Institute for Health and Care Excellence cost-effectiveness threshold. Health Technol Assess 2015; 19:1–503, vvi
39. Barbaro RP, Odetola FO, Kidwell KM, et al. Association of hospital-level volume of extracorporeal membrane oxygenation
cases and mortality. Analysis of the extracorporeal life support organization registry. Am J Respir Crit Care Med 2015; 191:894–901
40. Herridge MS, Tansey CM, Matté A, et al; Canadian Critical Care
Trials Group: Functional disability 5 years after acute respiratory distress syndrome. N Engl J Med 2011; 364:1293–1304
41. Hill AD, Fowler RA, Pinto R, et al. Long-term outcomes and healthcare utilization following critical illness–a population-based study. Crit Care 2016; 20:76
42. Bagshaw SM, Stelfox HT, McDermid RC, et al. Association between frailty and short- and long-term outcomes among critically ill patients: A multicentre prospective cohort study. CMAJ 2014; 186:E95–E102
43. Ferguson ND, Scales DC, Pinto R, et al; Canadian Critical Care
Trials Group: Integrating mortality and morbidity outcomes: Using quality-adjusted life years in critical care
trials. Am J Respir Crit Care Med 2013; 187:256–261
44. Skinner EH, Denehy L, Warrillow S, et al. Comparison of the measurement properties of the AQoL and SF-6D in critical illness. Crit Care Resusc 2013; 15:205–212
45. Orpana H, Ross N, Feeny D, et al. The natural history of health-related quality of life: A 10-year cohort study. Health Report 2009; 20:29–44
46. Wodchis WP, Austin PC, Henry DA. A 3-year study of high-cost users of health care. CMAJ 2016; 188:182–188