An ounce of prevention is worth a pound of cure: a cost-effectiveness analysis of incidentally detected aneurysms in functional MRI research

Abstract

Purpose: Despite ongoing debate about best practices for managing incidentally detected findings in brain research studies using magnetic resonance imaging (MRI), this issue has not been investigated from a health economics viewpoint. We applied a decision-analytic approach to assess the benefit of various strategies for functional MRI (fMRI) studies using intracranial aneurysms (IA) as a model.

Methods: A decision tree and Markov model were created to simulate the impact on the lifetime costs and quality-adjusted life-years (QALY) of four different strategies for review of scans for the presence of IA. To populate the decision model, we used current evidence from the literature and results from a survey of experts.

Results: Review of the anatomical scans by a nonspecialist is not cost-effective in any of the subgroups of participants. Full clinical examination of women with a positive family history before enrollment in a study is cost-effective. Cost-effectiveness of reviewing scans obtained from women without a family history and men with a positive family history of IA depends on the willingness-to-pay (λ) for a QALY: at λ of $50,000/QALY, review of scans by a specialist is cost-effective, whereas at λ of $100,000/QALY, a full clinical workup is the best option. Compared with not reviewing any scans, a customized strategy for each subgroup of participants results in an incremental cost-effectiveness ratio of $12,503 for λ=$50,000/QALY and $32,767 for λ=$100,000/QALY.

Conclusion: Tailored strategies based on the characteristics of research participants and λ for one QALY are needed to address the problem of incidental findings in research fMRI studies.

Figure 1

Schematic illustration of the decision tree (top, see text)* and Markov model simulating future life-years of hypothetical individuals with untreated aneurysm in the model (bottom). *For simplicity, stratification on baseline covariates and categorization of aneurysms by size are omitted from the illustration, and the full diagnostic workup is represented as a single step. Rectangles are states at which subjects remain for at least 1 year. Ovals are snapshot events. SAH, subarachnoid hemorrhage.

Figure 2

The incidence of true-positive aneurysm detections after initial review and lifetime incidence of SAH corresponding to each strategy. Strategy 1: no review; strategy 2: initial nonspecialist review followed by MR radiologist review; strategy 3: MR radiologist review; strategy 4: clinical-grade MR examination including MRA. IA, intracranial aneurysm; MR, magnetic resonance; MRA, magnetic resonance angiography; SAH, subarachnoid hemorrhage;WTP, willingness-to-pay.

Figure 3

Proportion of times each of the four review strategies was found to be the best option in the stochastic sensitivity analysis. Strategy 1: no review; strategy 2: initial nonspecialist review followed by MR radiologist review; strategy 3: MR radiologist review; strategy 4: clinical-grade MR examination including MRA. MR, magnetic resonance; MRA, magnetic resonance angiography.

Figure 4

Probability of the cost-effectiveness of different strategies based society’s willingness-to-pay for one QALY. Willingness-to-pay: society’s willingness to pay to gain one quality-adjusted life-year in an individual. The dashed lines correspond to the willingness-to-pay value of $50,000/QALY and $100,000/QALY.Strategy 1:no review;strategy 2: initial nonspecialist review followed by MR radiologist review; strategy 3: MR radiologist review; strategy 4: clinical-grade MR examination including MRA. MR, magnetic resonance; MRA, magnetic resonance angiography; QALY, quality-adjusted life-years.