Cost-effectiveness Analysis of Anatomic vs Functional Index Testing in Patients With Low-Risk Stable Chest Pain

Key Points Question Are first-line anatomic approaches to low-risk stable chest pain evaluation cost-effective compared with functional testing? Findings In this cost-effectiveness analysis using an individual-based Markov microsimulation model based on 10 003 participants in a randomized clinical trial, anatomic approaches were cost-effective compared with functional testing across a wide range of variations in clinical care and patient characteristics. Adding fractional flow reserve to coronary computed tomography angiography resulted in modest improvements after the initially increased costs of care were offset by fewer and more targeted coronary revascularizations. Meaning These findings suggest that anatomic strategies may present a favorable initial diagnostic option in the evaluation of low-risk stable chest pain compared with functional testing.

This supplemental material has been provided by the authors to give readers additional information about their work.

Baseline model assumptions: The key drivers of decision making
1) The true underlying CAD status defined based on PROMISE data defines the health status (no CAD/nonobstructive CAD/obstructive CAD) of each individual entering the model. 2) The index diagnostic tests (coronary CTA, functional testing, and coronary CTA+FFRCT) are modeled to identify the true underlying CAD status with a diagnostic accuracy reported in the ESC Guidelines 1 3) Preventive treatment is simulated for those above 7.5% ASCVD risk score and for those with diagnosed CAD. 4) Patients are referred to ICA as per the following diagnostic test findings: a. Functional testing: reversible myocardial ischemia (supplemental table 2); b. Coronary CTA: 70% in more than one vessel or 50% CAD in left main; 30-69% luminal stenosis CAD based on individual cardiovascular risk evaluation (supplemental table 2); c. CTA+FFRCT: 70% in more than one vessel or 50% CAD in left main and 30-69% luminal stenosis CAD with an FFRCT value <0.80. 5) Patients are revascularized with the following invasive diagnostic findings: a. One-or two-vessel disease: PCI; b. Left main disease or three-vessel disease: CABG. 6) MACE risk associated with a given health state (no CAD/non-obstructive CAD/obstructive CAD) is modeled based on age and sex specific cross-sectional data 2 , which is modeled for each individual in monthly cycles until the end of life taking into account the potential change in health status. All-cause death risk is modeled in each monthly cycle for each patient based on US life tables. 7) Treatment effects: 8) Cost of FFRCT CMS ($1450).

2.
Underlying CAD status: the underlying disease status for patients of the CTA arm was derived from the CT core laboratory reads and ICA results. For patients of the CT arm but without this information available and for patients allocated to the functional testing arm, therefore patients without anatomical information on the underlying disease status, we assumed that the underlying prevalence and extent of CAD was similar compared to the anatomical testing arm, because of the randomized controlled trial design. Hence, to derive CAD status for those without this information available, we used a multivariable ordered logistic regression approach as imputation method. First, we applied multivariable ordered logistic regression to patients with available underlying disease status including various risk factors and demographics. Then based on the resulting regression parameters we predicted the underlying disease status in patients without known CAD status.
3. Progression of CAD is modeled as a function of age, gender, disease status and National Cholesterol Education Program (NCEP) risk score from a cohort of stable chest pain patients using a simulated annealing approach. 1,2 In the model, each cycle started with the cohort divided among a set of mutually exclusive health states, where health states refer to a specific underlying CAD status. In each different health state, we modeled the events and progression that could occur within a cycle. Across a lifetime, transition to different health states was modeled in monthly cycles. From cycle to cycle the CAD status of a patient could remain the same or progress, and patients could suffer myocardial infarctions or could die from either cardiovascular disease or other causes. Therefore, the model was likely to start each cycle with a different distribution among those health states.

Definitions of cost, QALY, cost-effectiveness analysis, ICER and life years gained.
Cost: direct medical costs such as cost for diagnostic testing, interventions, and subsequent medication associated with each clinical strategy in US.
QALYs: measures quality adjusted life years (QALYs) by combining the quality and the quantity of life lived: the time spent in a particular health state measured in years multiplied by an utility weight ranging from 0 to 1, where 1 reflects perfect health and 0 the worst health state (in this study the worst health state is assumed to be death).
Cost-effectiveness analysis: estimates the costs (in this study measured in US$) and health gains (in this study measured in QALYs) of alternative strategies. A strategy is called dominant compared to another strategies if is both, cost saving and more effective (in this study this equals a higher number of QALYs). In general, 'new' strategies are more effective than 'old' strategies but at higher cost. In this case, a strategy is called cost effective if the cost per additional QALY, i.e. the ICER, is below $100,000 (see also next ICER paragraph).
ICER: measures the incremental cost-effectiveness ratio (ICER). It is defined by the cost difference between two interventions divided by the difference in healthcare effect and reflects the additional (incremental) cost associated with one unit increase of the measured healthcare effect (here QALYs).

= − −
When calculating the ICER, each strategy is compared with the next best alternative, based on the economic concept of opportunity costs. In this study the ICER shows the additional cost per QALY gained. We consider a diagnostic strategy cost-effective, when the ICER is below the cost-effectiveness threshold of 100,000$/QALY. A strategy is cost-saving, if additional QALY is gained at a lower cost compared to the other strategy.