Cost-effectiveness of Out-of-Hospital Continuous Positive Airway Pressure for Acute Respiratory Failure.

STUDY
OBJECTIVE: We determine the cost-effectiveness of out-of-hospital continuous positive airway pressure (CPAP) compared with standard care for adults presenting to emergency medical services with acute respiratory failure.
METHODS: We developed an economic model using a United Kingdom health care system perspective to compare the costs and health outcomes of out-of-hospital CPAP to standard care (inhospital noninvasive ventilation) when applied to a hypothetical cohort of patients with acute respiratory failure. The model assigned each patient a probability of intubation or death, depending on the patient's characteristics and whether he or she had out-of-hospital CPAP or standard care. The patients who survived accrued lifetime quality-adjusted life-years (QALYs) and health care costs according to their age and sex. Costs were accrued through intervention and hospital treatment costs, which depended on patient outcomes. All results were converted into US dollars, using the Organisation for Economic Co-operation and Development purchasing power parities rates.
RESULTS: Out-of-hospital CPAP was more effective than standard care but was also more expensive, with an incremental cost-effectiveness ratio of £20,514 per QALY ($29,720/QALY) and a 49.5% probability of being cost-effective at the £20,000 per QALY ($29,000/QALY) threshold. The probability of out-of-hospital CPAP's being cost-effective at the £20,000 per QALY ($29,000/QALY) threshold depended on the incidence of eligible patients and varied from 35.4% when a low estimate of incidence was used to 93.8% with a high estimate. Variation in the incidence of eligible patients also had a marked influence on the expected value of sample information for a future randomized trial.
CONCLUSION: The cost-effectiveness of out-of-hospital CPAP is uncertain. The incidence of patients eligible for out-of-hospital CPAP appears to be the key determinant of cost-effectiveness.

Introduction

Background

Acute respiratory failure is a common but life-threatening medical emergency, especially in elderly patients with respiratory and cardiac diseases.1, 2, 3 Inhospital noninvasive ventilation is widely used to treat acute respiratory failure that is refractory to initial medical therapy.4, 5, 6, 7, 8, 9, 10, 11, 12 However, the delay in providing noninvasive ventilation until arrival at the hospital may be one factor explaining why the risk of death in patients with respiratory problems increases markedly with distance traveled to the hospital. It has been argued that noninvasive ventilation is more likely to be effective if used early in the course of respiratory failure, before fatigue develops. Recent reviews have indicated that out-of-hospital noninvasive ventilation is feasible and beneficial in selected patients with acute respiratory failure.15, 16, 17, 18 We have undertaken a meta-analysis suggesting that out-of-hospital continuous positive airway pressure (CPAP) has the best evidence of effectiveness. CPAP is also the most practical way of providing out-of-hospital noninvasive ventilation.

What is already known on this topic

Emergency medical services (EMS) often use continuous positive airway pressure (CPAP) to treat out-of-hospital acute respiratory failure.

What question this study addressed

Is out-of-hospital CPAP cost-effective?

What this study adds to our knowledge

In this cost-effectiveness analysis using a United Kingdom perspective, the incremental cost-effectiveness ratio (£20,514 [$29,720] per quality-adjusted life-year gained) was insufficient by British standards to support widespread CPAP implementation. Interpretation of cost-effectiveness may vary by country, including the US.

How this is relevant to clinical practice

Although not directly impacting clinical practice, these findings highlight factors that should be considered in the selection and implementation of EMS therapies.

Importance

In the United States, the National Association of Emergency Medical Services Physicians stated that noninvasive ventilation is an important treatment modality for the out-of-hospital management of acute dyspnea. The recent UK Ambulance Services Clinical Practice Guidelines 2013 recommended (for the first time) the use of CPAP in the out-of-hospital environment on the basis of expert consensus. However, use of out-of-hospital CPAP in the United Kingdom remains limited in practice, probably reflecting the significant costs of establishing this treatment. The decision to establish out-of-hospital CPAP depends on weighing the benefits of improved outcomes against the additional costs incurred by establishing the service and is fundamentally an issue of cost-effectiveness.

Goals of This Investigation

We aimed to estimate the incremental cost per quality-adjusted life-year (QALY) of out-of-hospital CPAP compared with standard care and determine whether out-of-hospital CPAP should be recommended for funding according to accepted thresholds for cost-effectiveness.

Materials and Methods

Theoretical Model of the Problem

The cost-effectiveness of a more effective and expensive treatment, such as out-of-hospital CPAP, can be estimated by comparing the outcomes and costs associated with the treatment to an appropriate alternative, such as inhospital noninvasive ventilation. Modeling is used to estimate how better effectiveness leads to improved patient outcomes relative to the cost of the care, expressed in terms of QALYs gained by using the intervention. A QALY is equivalent to a year of life spent in full health and incorporates both quality of life and survival. If outcomes are measured as QALYs, then the incremental cost-effectiveness ratio, or cost per QALY gained, can be calculated and compared with alternative uses of health care funding. The incremental cost-effectiveness ratio can be crudely understood as the amount needed for the intervention to “buy” each additional QALY compared with standard care. In the United Kingdom, the National Institute for Health and Care Excellence typically recommends in favor of funding interventions with an incremental cost-effectiveness ratio below a threshold, widely accepted to be between £20,000 and £30,000 per QALY (ie, interventions that can be used to buy QALYs for less than £20,000 [$29,000] or £30,000 [$43,500] each) and recommends against funding interventions with an incremental cost-effectiveness ratio above these thresholds. Other countries use different thresholds; one of $50,000 is conventionally used in the United States. We used the National Institute for Health and Care Excellence threshold because our analysis took a UK perspective and was based on UK practice and unit costs. Appendix E1 (available online at http://www.annemergmed.com) describes the general principles of cost-effectiveness analysis.

Study Design

We developed an economic model with the statistical software program R (version 3.0.3) to explore the costs and health outcomes associated with the use of out-of-hospital CPAP to treat acute respiratory failure compared with standard care, and to calculate the incremental cost per QALY gained. We based the analysis on a UK health care system perspective using a lifetime horizon. We converted the results into US dollars, using the Organisation for Economic Co-operation and Development purchasing power parities rates (£1=$1.45).

The structure of the model is shown in Figure 1 and the parameters used in the model are reported in Table 1. Patients in acute respiratory failure may receive either out-of-hospital CPAP or standard care, so these alternative treatments were applied to a hypothetical cohort of patients with acute respiratory failure who were eligible for noninvasive ventilation, ie, all patients receive out-of-hospital CPAP in the intervention group and standard care in the comparator group. Each patient in the model could have one of the following outcomes: hospitalized without intubation, hospitalized with intubation, or death in the ambulance or hospital. The probability of death and probability of intubation depended on whether the patient received out-of-hospital CPAP or standard care. The patients who survived accrued lifetime QALYs and health care costs according to their life expectancy. Costs were also accrued through costs of intervention (ie, out-of-hospital CPAP) and hospital treatment costs, which depended on whether the patient needed intubation. Details of each of these processes are outlined below.

Figure 1

Model structure. CPAP, Continuous positive airway pressure.

Table 1

Summary of model parameters.∗

Parameter	Mean	Distribution	Source
Baseline risks
General population mean 30-day mortality probability	0.118	Beta (79, 589)	Nicholl et al13
Risk of intubation	0.029	Beta (4.45, 150)	3CPO,24 clinical opinion
OR for out-of-hospital CPAP
Mortality	0.43	Samples	Meta-analysis19
Intubation	0.32	Samples	Meta-analysis19
Life expectancy of patients
Lifetime years	2.67	Normal (2.67, 0.16)	3CPO,24 clinical opinion
Health-related quality of life
Utility	0.6	Beta (640, 425)	3CPO,24 clinical opinion
Costs, £ ($)
Out-of-hospital CPAP	1,212 (1,740)	1,500–1,000×β (2, 5)	Clinical input
Hospitalization	2,250 (3,260)	Gamma (75, 30)	NHS reference costs30
Intubation	3,500 (5,075)	Gamma (70, 50)	NHS reference costs 2011–201230
Annual	5,360 (7,685)	Gamma (53, 100)	NHS reference costs 2011–201230

3CPO, Three Interventions in Cardiogenic Pulmonary Oedema; NHS, National Health Service.

Beta (a,b) Distribution is a statistical distribution defined between 0 and 1; a and b parameters in the Beta distribution can be thought of as counts of the event of interest versus its complement, eg, Beta (79,589) for mortality represents 79 deaths in a population of 668 (ie, 79+589). Normal distribution is represented with mean and SD, with 95% of the values in the distribution lying between 2 SDs on either side of the mean, eg, normal (2.67, 0.16) implies that 95% of the samples lie between 2.35 and 2.99. Gamma (a,b) Distribution, where a is the shape parameter and b is the scale parameter, is typically used for skewed distributions and has a mean expected value of a×b, eg, the average value of the samples of distribution Gamma (75,30) is 2,250 (75×30).

We estimated the baseline risks of intubation and death (ie, for patients receiving standard care) with data from published studies. We modeled the mortality risk of emergency admissions with respiratory illness with a large cohort data set of 668 patients presenting with “respiratory disease” across 4 English ambulance services during a 4-year period, from 1997 to 2001. There were 79 deaths in 668 patients, which resulted in a mean mortality rate of 11.8%. This baseline mortality rate is similar to that in the more recent Three Interventions in Cardiogenic Pulmonary Oedema (3CPO) study.

We modeled the risk of intubation for respiratory illness with the data from 3CPO study, a multicenter, open, prospective, randomized, controlled trial of 1,069 patients presenting with severe acute cardiogenic pulmonary edema at 26 emergency departments (EDs) in the UK. The study reported a mean intubation rate of 2.9%, which is similar to the intubation rates of 2.7% reported in British Thoracic Society national respiratory audit program annual report 2011/12.

We estimated the effectiveness of out-of-hospital CPAP in reducing mortality and intubation as odds ratios (ORs) from a meta-analysis of 6 randomized trials, which are summarized in Table 1. The effectiveness estimates from our meta-analysis are similar to those of Williams et al but greater than those of Mal et al because the latter authors estimated effectiveness for all forms of out-of-hospital noninvasive ventilation together, whereas we estimated effectiveness of out-of-hospital CPAP and bi-level inspiratory positive airway pressure separately and used only the estimate for CPAP.

The patients who survived (ie, who avoided the short-term mortality risk) accrued QALYs, and these lifetime QALYs were estimated according to patients’ life expectancy and their utilities. The life expectancy of patients with acute respiratory failure and admitted to the hospital were captured from the 3CPO trial and parameterized as a normal distribution with a mean of 2.67 and SD of 0.16, after discussions with a clinical expert group (S.G., M.W., J.P.-A., and G.D.P.). The 3CPO study reported that the mean utility value (quality of life) was 0.6. The estimated QALYs for patients with acute respiratory failure were estimated by multiplying the life-years by the lifetime quality of life shown in Table 1. There was no evidence that patients who survived after receiving out-of-hospital CPAP experienced better health-related quality of life or lived longer compared with patients given standard care. We assumed that the lifetime QALYs were same for all survivors, irrespective of whether they were in standard care or the out-of-hospital CPAP arm.

The costs included in the model are for of out-of-hospital CPAP, intubation, hospitalization, and lifetime care for patients. Hubble et al reported a mean additional 5 days in length of stay associated with intubation compared with that for patients without intubation. We assumed that the additional 5 hospital days spent by the intubated patients would be in the ICU, according to the suggestions by the clinical expert group, at a cost of £700 ($1,015) per day. We estimated lifetime costs of survivors by using the annual costs and the discounted life expectancy of patients captured from the 3CPO trial. The study reported that mean costs in months 4 to 6 were £1,341 ($1,944.50), which resulted in mean annual costs of £5,360 ($7,685). The costs of standard care were not included in the model because the analysis is based on incremental costs, ie, we assumed that all initial treatment costs were the same in both arms, regardless of whether the patient received out-of-hospital CPAP.

We estimated the costs of out-of-hospital CPAP at an ambulance service level and converted these into a cost per patient according to a 5-year depreciation period. These costs included those for initial equipment, implementation, and ongoing maintenance. Although the costs of setting up the service are largely the same, there are substantially different estimates of incidence reported in different sources.25, 26, 27, 28, 29 The cost of out-of-hospital CPAP per patient in an ambulance service is estimated to be £189.93+£202,446/N ($275.50+$293,546/N), where N is the number of patients per ambulance service in a year. This information was synthesized into a mean cost of £1,212 ($1,740) per patient, according to our clinical expert input. Because there are different estimates of incidence, ranging from approximately 175 to 2,000 patients per ambulance service in a year, scenario analysis was also conducted for 3 different estimates for unit cost for performing out-of-hospital CPAP per patient (for different estimates of the eligible population), ie, a high-cost scenario with a unit cost of £1,400 ($2,030), a low-cost scenario with a unit cost of £745 ($1,080), and a lower-cost scenario with a unit cost of £300 ($435). Full details are provided in Appendix E2 (available online at http://www.annemergmed.com).

Primary Data Analysis

Probabilistic analysis incorporated uncertainty in the parameter estimates to provide a measure of precision and confidence in the estimates of the mean costs and QALYs. Additionally, we calculated the probability that each strategy would be the most cost-effective at different thresholds for willingness to pay for health gain. We constructed cost-effectiveness acceptability curves by plotting the probability of each strategy’s being cost-effective against willingness to pay. Furthermore, expected value of perfect information was estimated to identify whether the expected cost of future research would be valuable. Expected value of partial perfect information and expected value of sample information techniques were also used to identify the critical areas of uncertainty in which future research would produce the most benefit. Value-of-information analyses (expected value of perfect information, expected value of partial perfect information, and expected value of sample information) provide an estimate of the monetary value of further research to reduce uncertainty and, in particular, an estimate of how much we should be prepared to pay for a trial to reduce uncertainty.

Results

The total costs of out-of-hospital CPAP are higher than those of usual care (£16,895 versus £14,863, or $24,497 versus $21,551), but 0.099 QALYs are gained per patient treated (1.513 versus 1.414). The mean incremental cost-effectiveness ratio of out-of-hospital CPAP compared with standard care in the base case analysis is £20,514 per QALY ($29,720/QALY). It therefore costs the health service £20,514 ($29,720) to buy each additional QALY with out-of-hospital CPAP. Figure 2 shows the uncertainty associated with this estimate by plotting samples of mean incremental costs and QALYs. There is substantial uncertainty, with samples falling equally on either side of the red line, indicating the £20,000 per QALY ($29,000/QALY) threshold and a 49.5% probability of out-of-hospital CPAP’s being cost-effective at this threshold.

Figure 2

Cost-effectiveness plane for the base-case economic analysis. ICER, Incremental cost-effectiveness ratio; QALY, quality-adjusted life-years.

This is also observed in the cost-effectiveness analysis curve in Figure 3, which shows the proportion of model runs for which each strategy is cost-effective over a range of potential thresholds for willingness to pay. The more we are willing to pay for health gain (ie, the more we are willing to spend to buy a QALY), the more likely it is that out-of-hospital CPAP will be cost-effective, but there is substantial uncertainty between the thresholds of £20,000 per QALY ($29,000/QALY) and £30,000 per QALY ($43,500/QALY).

Figure 3

Cost-effectiveness acceptability curve for the base-case economic analysis.

Scenario analysis was also conducted for 3 different estimates for unit cost for providing out-of-hospital CPAP per patient, ie, a high-cost scenario with a unit cost of £1,400 ($2,030), a low-cost scenario with a unit cost of £745 ($1,080), and a lower-cost scenario with a unit cost of £300 ($435). These estimates reflect variation in our estimates of the incidence of appropriate patients, with high estimates of incidence resulting in lower estimated costs. Table 2 compares each scenario with the base case and shows that out-of-hospital CPAP is more likely to be cost-effective (93.8% probability) if the incidence of appropriate patients is high and the resulting cost per patient low. Results from threshold analysis suggested that CPAP is unlikely to be cost-effective at £30,000 per QALY ($43,500/QALY) in an ambulance service is greater than if it costs more than £2,170 ($3,150) for out-of-hospital CPAP per patient.

Table 2

Results for different cost scenarios.

Scenario Type, £ ($)	Standard Care		Out-of-Hospital CPAP		Differences Between Out-of-Hospital CPAP and Standard Care		ICER (per QALY), £ ($)	Probability of being Cost-effective
	Total Costs, £ ($)	Total QALYs	Total Costs, £ ($)	Total QALYs	Costs, £ ($)	QALYs
Base case	14,863 (21,551)	1.414	16,895 (24,498)	1.513	2,032 (2,946)	0.099	20,514 (29,720)	0.495
High cost, 1,400 (2,030)∗	14,863 (21,551)	1.414	17,078 (24,763)	1.513	2,216 (3,213)	0.099	22,368 (32,434)	0.354
Low cost, 745 (1,080)∗	14,863 (21,551)	1.414	16,421 (23,810)	1.513	1,558 (2,259)	0.099	15,728 (22,805)	0.798
Lower cost, 300 (435)∗	14,863 (21,551)	1.414	15,977 (23,166)	1.513	1,114 (1,615)	0.099	11,248 (16,309)	0.938

Scenario analysis was conducted for different estimates for unit cost for performing out-of-hospital CPAP per patient (for different estimates of the eligible population). See Appendix E2 (available online at http://www.annemergmed.com) for more details.

The incidence of appropriate patients was also an important determinant of the expected value of information. The population expected value of perfect information at the threshold of £20,000 per QALY ($29,000/QALY) is £1.9 million ($2.75 million) at a low estimate of incidence of 3.5 per 100,000 population per year and £22.5 million ($32.5 million) at a higher incidence of 40.8 per 100,000 population per year. This value is defined as the maximum investment a decisionmaker would be willing to pay to eliminate all parameter uncertainty from the decision problem and reflects the amount we should be willing to pay for research to reduce uncertainty. Expected value of partial perfect information analysis suggested baseline mortality, out-of-hospital CPAP mortality OR, and costs of out-of-hospital CPAP as the key parameters driving uncertainty. The population expected value of partial perfect information for the 3 parameters together at the threshold is estimated as £1.83 million ($2.65 million) at the low incidence and £21.3 million ($30.8 million) at the higher one, both of which are very close to the population expected value of perfect information values, suggesting that most of the uncertainty in the decision problem is from these 3 parameters. The population expected value of sample information value for baseline mortality and out-of-hospital CPAP mortality OR parameters, assuming a randomized controlled trial with 100 patients in each arm, is estimated as £1.08 million ($1.56 million) at low incidence and £12.67 million ($18.37 million) at the higher one. It is cost-effective to conduct the trial to address the uncertainty if the population expected value of sample information of a proposed trial at a given sample size is greater than the costs of the trial.

Limitations

This model is generally based on robust data sources, with the estimates of effectiveness of out-of-hospital CPAP being derived from a meta-analysis of randomized trials and most cost estimates being derived from UK National Health Service reference costs. However, there are some limitations to the data. The trials in the meta-analysis were generally small and had potentially selected study populations, which may not have compared out-of-hospital CPAP with best alternative care. The model parameters were estimated from different sources, which may include different patient populations; for example, the baseline mortality risk was estimated for patients with respiratory disease, whereas the intubation risk was based on data for patients with a diagnosis of severe acute cardiogenic pulmonary edema.

The perspective of the analysis was the English health service, UK cost estimates were used, key model parameters were estimated from UK sources, and the thresholds for cost-effectiveness were those used in the United Kingdom by the National Institute for Health and Care Excellence. Costs may differ in other countries; for example, the unit cost of the CPAP system may be lower in the United States. The cost-effectiveness of out-of-hospital CPAP may be more certain in health services with different parameters of willingness to pay for health gain. In particular, cost-effectiveness appears to be more certain when compared against a US threshold of $50,000 per QALY. However, our analysis used UK estimates of costs and resource use. If costs and resource use are higher in the United States, the cost-effectiveness will be less certain.

The cost per patient of providing out-of-hospital CPAP was calculated by dividing the total cost of setting up and running the service by the total number of patients treated, which means that the cost per patient was determined by the incidence of patients who were likely to benefit from out-of-hospital CPAP. We identified a number of sources for our estimate of this parameter, but these estimates varied markedly. Sensitivity analysis showed that cost per patient is an important determinant of cost-effectiveness, so an accurate estimate of the incidence of patients likely to benefit from out-of-hospital CPAP is required to accurately estimate cost-effectiveness.

We have assumed in the analysis that the ambulance service has a 1-tiered response. However, some services may have different tiers of response that may allow more efficient use of equipment and trained staff. We assumed that out-of-hospital CPAP had a constant effect on mortality and intubation rate, according to estimates from meta-analysis. Effectiveness may depend on distance traveled to the hospital, being more effective in settings with long distances to the hospital. Unfortunately, distance to the hospital was not consistently collected in primary studies, so this factor could not be explored in the individual patient data meta-analysis.

Discussion

The economic analysis showed that out-of-hospital CPAP was more effective than standard care, with 0.099 QALYs gained per patient treated, but was more expensive, with an additional cost of £2,032 ($2,934) per patient treated. The incremental cost-effectiveness ratio for out-of-hospital CPAP was £20,514 per QALY ($29,720/QALY) compared with standard care, with 49.5% probability of being cost-effective at the £20,000 per QALY threshold. These findings suggest that even if the apparent effectiveness of out-of-hospital CPAP reported by recent meta-analysis17, 18 were confirmed, the cost-effectiveness of this treatment is uncertain when compared with a UK cost-effectiveness threshold. It is therefore unlikely that out-of-hospital CPAP would be recommended for widespread implementation in the United Kingdom on the basis of this analysis.

In developing the economic model, we identified marked variation between estimates from different sources of the incidence of patients likely to benefit from out-of-hospital CPAP. Sensitivity analysis showed that this parameter was an important determinant of cost-effectiveness, with the probability of out-of-hospital CPAP’s being cost-effective at the £20,000 per QALY ($29,000/QALY) threshold varying from 35.4% to 93.8%. Most of the costs of providing out-of-hospital CPAP are incurred in setting up the service. If only a small number of patients will benefit from out-of-hospital CPAP, then the cost per patient will be high and cost-effectiveness is unlikely.

We identified only 1 previous economic analysis of out-of-hospital CPAP. This was undertaken in the United States and estimated that out-of-hospital CPAP would save 0.75 additional lives per 1,000 patients, at a cost of $490 per life saved. This analysis had a number of limitations. Treatment effect estimates were based on trial of inhospital rather than out-of-hospital CPAP, so the analysis effectively compared CPAP with no noninvasive ventilation, rather than comparing out-of-hospital CPAP with inhospital noninvasive ventilation. Outcomes were valued as lives saved rather than QALYs, and the model used only a 1-year time horizon. One-way sensitivity analysis was performed, but the authors did not perform a probabilistic sensitivity analysis. The estimate that out-of-hospital CPAP would be used 4 times per 1,000 patients seems high compared with our estimates of patient eligibility. As a consequence, although this analysis suggested that out-of-hospital CPAP is cost-effective, it is unlikely to convince purchasers of health care.

Expected value of information analysis was undertaken to explore uncertainty and determine the value of further research. It showed that the value of undertaking a trial depends on the estimated incidence of eligible patients. The maximum cost at which it would be cost-effective to conduct a trial with 100 patients in each arm is only £1.08 million ($1.56 million) if there were a low estimated incidence (of 3.5 per 100,000 population per year) of eligible patients, but would be £12.67 million ($18.37 million) if there were a high estimated incidence (of 40.8 per 100,000 population per year). A more precise estimate of the incidence of eligible patients is therefore required to determine the cost-effectiveness of a future trial of out-of-hospital CPAP. Of course, the feasibility of a future trial would also depend on the incidence of eligible patients.

Our model can be used to determine whether out-of-hospital CPAP is likely to be cost-effective in a particular system, given the estimate of the incidence of eligible patients. Our analysis indicates that out-of-hospital CPAP has uncertain cost-effectiveness. Establishing out-of-hospital CPAP as a standard treatment of acute respiratory failure will require substantial resources, and there is a relatively high probability that outcomes improvements from out-of-hospital CPAP may not be justified by the financial investment. A large pragmatic randomized trial could improve our estimates of effectiveness and cost-effectiveness, but this would also require substantial funding, with uncertain value. Our analysis suggests that the incidence of patients eligible for out-of-hospital CPAP is an important determinant of the cost-effectiveness of out-of-hospital CPAP itself and of a future trial of it. Health systems considering implementing out-of-hospital CPAP should first estimate the incidence of potentially eligible patients. If the incidence is low, then implementation of out-of-hospital CPAP is not likely to be cost-effective. If the incidence is high, then implementation or further research with a large pragmatic randomized trial may be appropriate. If incidence varies between health care systems, then out-of-hospital CPAP may be cost-effective in some systems but not others.

Table E1

Breakdown of out-of-hospital CPAP costs.

	Number of Devices			Source	Unit Cost, £ ($)		Source	Total Cost, £ ($)
Device costs
Out-of-hospital CPAP device	Number of ambulances that need the CPAP device (420)			Expert advisory input	513.49 (744.50)		Vygon: hospital CPAP kit	513.49×420
	Assuming 10% new CPAP devices during 5-y usage (42)			Expert advisory input	513.49 (744.50)		Vygon: hospital CPAP kit	513.49×42
Total cost of the device								237,232
Setup/implementation costs
	Resource Usage
Initial training	1,500 paramedics for 2 days each			Expert advisory input	150 (217.50) per day		Expert advisory input	450,000 (652,500)
Service reconfiguration	1-off cost for reconfiguration						Expert advisory input	100,000 (145,000)
Total setup/implementation costs								550,000 (797,500)
Maintenance costs
		Resource Usage
Consumables		Number of patients during 5 y=5×N	Expert advisory input		189.93 (275) per use	Vygon: facial mask, oxygen tubing, and valve		189.93×5×N (275×5×N)
Ongoing training		1,500 paramedics for 1 day each	Expert advisory input		150 (217.50) per day	Expert advisory input		225,000 (326,500)
Total maintenance costs								225,000+949.65×N(326,500+1,375×N)
Total costs of out-of-hospital CPAP								1,012,232+949.65×N
Total number of patients (N patients per year times depreciation period of 5 y (ie, assuming new out-of-hospital CPAP equipment will be required in 5 y)								5×N
Cost of out-of-hospital CPAP per patient								189.93+202,446/N(275.50+293,546/N)

Table E2

Scenarios for unit costs of out-of-hospital CPAP.

Source	Incidence of Eligible Patients per 100,000	Annual Eligible Patients in an Ambulance Service	Unit Cost of Out-of-Hospital CPAP, £ ($)∗
Spijker et al1	3.5	175	1,346.76 (1,952.80)
Aguilar et al2	7.3	365	744.58 (1,096.64)
Luhr et al3	17.8	890	417.40 (605.20)
Hubble et al5	34.2	1,700	309.02 (448.08)
BTS audit4	36.1	1,800	302.40 (438.48)
STH ED data	40.8	2,000	291.15 (422.15)

BTS, British Thoracic Society; STH, Sheffield Teaching Hospital.

Using the formula unit cost=£189.93+£202,446/N ($275.5+$293,546/N), where N is the number of patients per year.

Table E3

Summary of costs.

Scenario	Mean Value, £ ($)	Distribution, £ ($)
Baseline	1,212 (1,740)	1,500–1,000×β (2, 5)
High cost	1,400 (2,030)	Normal (1,400, 100) [normal (2,030, 145)]
Low cost	745 (1,080)	Normal (745, 100) [normal (1,080, 145)]
Lower cost	300 (435)	Normal (300, 50) [normal (435, 72.50)]

Economic evaluation∗

Section/Item	Item No.	Recommendation	Reported on Page No.
Title and abstract
Title	1	Identify the study as an economic evaluation or use more specific terms such as “cost-effectiveness analysis,” and describe the interventions compared.	1
Abstract	2	Provide a structured summary of objectives, perspective, setting, methods (including study design and inputs), results (including base case and uncertainty analyses), and conclusions.	1
Introduction
Background andobjectives	3	Provide an explicit statement of the broader context for the study.Present the study question and its relevance for health policy or practice decisions.	2
Methods
Target population and subgroups	4	Describe characteristics of the base-case population and subgroups analyzed, including why they were chosen.	2
Setting and location	5	State relevant aspects of the system(s) in which the decision(s) need(s) to be made.	3
Study perspective	6	Describe the perspective of the study and relate this to the costs being evaluated.	3
Comparators	7	Describe the interventions or strategies being compared and state why they were chosen.	3
Time horizon	8	State the time horizon(s) over which costs and consequences are being evaluated and say why appropriate.	3
Discount rate	9	Report the choice of discount rate(s) used for costs and outcomes and say why appropriate.	4
Choice of health outcomes	10	Describe what outcomes were used as the measure(s) of benefit in the evaluation and their relevance for the type of analysis performed.	3
Measurement of effectiveness	11a	Single-study-based estimates: Describe fully the design features of the single effectiveness study and why the single study was a sufficient source of clinical effectiveness data.	4
Measurement and valuation of preference-based outcome	12	If applicable, describe the population and methods used to elicit preferences for outcomes.	NA
Estimating resources and costs	13a	Single-study-based economic evaluation: Describe approaches used to estimate resource use associated with the alternative interventions. Describe primary or secondary research methods for valuing each resource item in terms of its unit cost. Describe any adjustments made to approximate to opportunity costs.	4, 5Appendix
Currency, price date, and conversion	14	Report the dates of the estimated resource quantities and unit costs. Describe methods for adjusting estimated unit costs to the year of reported costs if necessary. Describe methods for converting costs into a common currency base and the exchange rate.	3, 4, 5
Choice of model	15	Describe and give reasons for the specific type of decision analytical model used. Providing a figure to show model structure is strongly recommended.	3
Assumptions	16	Describe all structural or other assumptions underpinning the decision-analytical model.	3, 4, 5
Analytical methods	17	Describe all analytical methods supporting the evaluation. This could include methods for dealing with skewed, missing, or censored data; extrapolation methods; methods for pooling data; approaches to validate or make adjustments (such as half cycle corrections) to a model; and methods for handling population heterogeneity and uncertainty.	3, 4, 5Appendix
Results
Study parameters	18	Report the values, ranges, references, and, if used, probability distributions for all parameters. Report reasons or sources for distributions used to represent uncertainty where appropriate. Providing a table to show the input values is strongly recommended.	Table 1
Incremental costs and outcomes	19	For each intervention, report mean values for the main categories of estimated costs and outcomes of interest, as well as mean differences between the comparator groups. If applicable, report ICERs.	Table 2
Characterizing uncertainty	20a	Single-study-based economic evaluation: Describe the effects of sampling uncertainty for the estimated incremental cost and incremental effectiveness parameters, together with the impact of methodological assumptions (such as discount rate and study perspective).	Figure 2, Figure 3
Characterizing heterogeneity	21	If applicable, report differences in costs, outcomes, or cost-effectiveness that can be explained by variations between subgroups of patients with different baseline characteristics or other observed variability in effects that are not reducible by more information.	Table 2
Discussion
Study findings, limitations, generalizability, and current knowledge	22	Summarize key study findings and describe how they support the conclusions reached. Discuss limitations and the generalizability of the findings and how the findings fit with current knowledge.	s 6,7
Other
Source of funding	23	Describe how the study was funded and the role of the funder in the identification, design, conduct, and reporting of the analysis. Describe other nonmonetary sources of support.	tbc
Conflicts of interest	24	Describe any potential for conflict of interest of study contributors in accordance with journal policy. In the absence of a journal policy, we recommend authors comply with International Committee of Medical Journal Editors recommendations.	tbc

Husereau D, Drummond M, Petrou S, et al. Consolidated Health Economic Evaluation Reporting Standards (CHEERS)—explanation and elaboration: a report of the ISPOR Health Economic Evaluations Publication Guidelines Good Reporting Practices Task Force. Value Health. 2013;16:231-250.