Cost-Utility Analysis of Routine Hypercoagulable State Testing in Young Patients with Ischemic Stroke
Michael S. Freedman1, Natalie T. Cheng2, Anthony S. Kim3
1UCSF School of Medicine, 2New York-Presbyterian Brooklyn Methodist Hospital Department of Neurology, 3UCSF Department of Neurology
• Causes of stroke in the young (18-45) are diverse and testing for rare hypercoagulable disorders is routine1,2
• Unclear whether routine hypercoagulable state testing with targeted anticoagulation based on this testing or empiric antiplatelet therapy is preferred, given:
• Uncertain benefits, high costs of testing
• Bleeding risks associated with long-term anticoagulation3
1. To evaluate the cost-utility of hypercoagulable testing in first-time ischemic stroke/TIA in young adults
2. To determine the variables with the largest impact on the cost-utility of hypercoagulable testing in this setting
• Decision-analytic and Markov state-transition models were created with TreeAge 2015
• Baseline scenario is an 18-26 year old adult with first-time ischemic stroke
• Two treatment strategies were compared:
• Test strategy: Hypercoagulable state testing
• If positive: anticoagulation therapy
• If negative: antiplatelet therapy
• No Test strategy: empiric antiplatelet therapy
• Markov model includes 6 mutually exclusive annual transition states (recurrent stroke, major hemorrhage, minor hemorrhage, mRS 0-2, mRS 3-5, death) (Fig. 1)
• Model includes costs and utilities of the resulting disease states in a given year which are aggregated over 20 years
• All model inputs varied across plausible ranges to reflect input uncertainty
• First order Monte-Carlo simulation was performed to generate probabilistic sensitivity analyses
• Disease States
• Recurrent Stroke: includes ischemic stroke and TIA
• Major Hemorrhage: intracranial hemorrhage, or bleeding requiring hospitalization or transfusion.
• Minor Hemorrhage: bleeding that did not meet criteria for major hemorrhage
• Modified Rankin Scale (mRS) 0-2: no to mild disability
• mRS 3-5: moderate to severe disability
• mRS 6: death
• Cost-utility Parameters
• Cost: cost for care, as well as costs to society, measured in 2015 USD.
• Utility: ranges from 0 (death) to 1 (one year of perfect health), measured in quality-adjusted life years(QALYs)
• Death is irreversible
• Sensitivity and specificity of hypercoagulable testing assumed to be 100% as simplifying assumption
• No back-to-back major events
• Point estimates and sensitivity ranges and distributions derived from the medical literature (Table 1)
• Test strategy resulted in an increase of $2,669 and a loss of 0.04 QALYs compared to No
Test strategy (Table 2)
• Test strategy is not cost-effective in 98% of outcomes (Fig. 2)
• Preferred strategy is not affected by willingness-to-pay (Fig. 3)
• Three variables have potential to change the directionality of incremental cost or
incremental benefit (Fig. 4)
• Test strategy may not be cost-effective
• Routine hypercoagulable testing in young patients with ischemic stroke may result in increased cost without any increase in benefit
• In the absence of increased clinical suspicion of underlying hypercoagulable state, hypercoagulable testing may not be warranted
I would like to acknowledge Naomi Bardach for her thoughtful comments in the revision process. Support provided by NIH StrokeNet (NIH/NINDS Award Number U01NS086872).
1Lisovoski et al. J Neurol Neurosurg Psychiatry 1991;54(7):576–9.
2Carolei et al. Stroke 1993;24(3):362–7.
3Kernan et al. Stroke 2014;45(7):2160-236.
4Goeggel et al. Stroke 2015;84:1941-7.
5Mohr et al. NEJM 2001;345(20):1444-51.
6Synhaeve et al. Stroke 2014;45(4):1157–60.
7Saloheimo et al. Stroke 2006;37(1):129-33.
8Hart et al. Stroke 1995;26(8):1471-7.
9Kochanek et al. National Vital Statistics Reports, CDC 2012;61(6):1-51.
10Kim et al. Stroke 2011;42(7):2013-8.
11The Facts about Social Security’s Disability Program. SSA, Publication No. 05-10570. June 2016. https://www.ssa.gov/disabilityfacts/materials/pdf/factsheet.pdf.
12Ji et al. Arch Neurol 2013;70(1):51-57
13Luengo-Fernandez et al. Neurology 2013;81(18):1588-1595.
14Hsu et al. PLoS One 2015;10(4):e0124806
15Shah et al. Circulation 2011;123(22):2562-2570.
16Baeten et al. Cost Eff Resour Alloc 2010;8(1):1-21.
17Somma et al. Am J Clin Pathol 2006;126(1):120-127.
18HCUP NIS, ICD9 434. HCUP 2013. AHRQ, Rockville, MD. http://www.hcupus.ahrq.gov/nisoverview.jsp.
19HCUP NIS, ICD9 431. HCUP 2013. AHRQ, Rockville, MD. http://www.hcupus.ahrq.gov/nisoverview.jsp.
20Auerbach et al. Am J Med 2004;116(12):816-28.
21Ecotrin Safety Coated Aspirin Tablets Low Strength - 365 CT. https://www.walmart.com/ip/Ecotrin-Low-Strength-Safety-Coated-Aspirin-Pa....