The economic impact of implementing a multiple inflammatory biomarker-based approach to identify, treat, and reduce cardiovascular risk

Abstract

Objectives:

To develop an economic model to estimate the change in the number of events and costs of non-fatal myocardial infarction (MI) and non-fatal ischemic stroke (IS) as a result of implementing routine risk-stratification with a multiple inflammatory biomarker approach.

Methods:

Reductions in the numbers of non-fatal MI and non-fatal IS events and in related per-member-per-month (PMPM) and 5-year costs (excluding test costs) due to biomarker testing were modeled for a US health plan with one million beneficiaries. Inputs for the model included literature-based MI and IS incidence rates, healthcare costs associated with MI and IS, laboratory results of biomarker testing, MI and IS hazard ratios related to biomarker levels, patient monitoring and intervention costs and use/costs of preventative pharmacotherapy. Preventative pharmacotherapy inputs were based on an analysis of pharmacy claims data. Costs savings (2013 USD) were assessed for patients undergoing biomarker testing compared to the standard of care. Data from MDVIP and Cleveland Heart Lab supported two critical inputs: (1) treatment success rates and (2) the population distribution of biomarker testing. Incidence rates, hazard ratios, and other healthcare costs were obtained from the literature.

Results:

For a health plan with one million members, an estimated 21,104 MI and 22,589 IS events occurred in a 5-year period. Routine biomarker testing among a sub-group of beneficiaries ≥35 years old reduced non-fatal MI and IS events by 2039 and 1869, respectively, yielding cost savings of over $187 million over 5 years ($3.13 PMPM), excluding test costs. Results were sensitive to changes in treatment response rates. Nonetheless, cost savings were observed for all input values.

Conclusions:

This study suggests that health plans can realize substantial cost savings by preventing non-fatal MI and IS events after implementation of routine biomarker testing. Five-year cost savings before test costs could exceed $3.13 PMPM.

Keywords::

• Inflammation biomarker
• Budget-impact analysis
• Cardiovascular risk
• Myeloperoxidase

Transparency

Declaration of funding

Funding for this research has been provided by Cleveland HeartLab, Inc.

Declaration of financial/other relationships

Research support was provided to Analysis Group, Inc. by Cleveland Heart Lab, Inc. MY, AKGC, JD and HB are employees of Analysis Group, Inc. MSP is the scientific and medical founder of Cleveland HeartLab, Inc. (OH). MSP also serves as the Chief Medical Officer of Cleveland HeartLab, Inc. SP and MB are employees of Cleveland HeartLab, Inc. AK is the Chief Medical Officer of MDVIP. The other authors have no conflicts of interest that are directly relevant to the content of this study. JME peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgements

The authors are grateful for the substantial research assistance provided by Todor Totev of Analysis Group, Inc. The authors also thank Jake Orville, CEO of Cleveland HeartLab, Inc. for his assistance during this project. Mr. Totev has no additional financial conflicts of interest to disclose. Mr. Orville is an employee and shareholder of Cleveland HeartLab, Inc.

Notes

*The model estimated numbers of events and costs among all beneficiaries age 25 and older under both scenarios. However, the impact of biomarker testing relative to the standard of care for patients aged 25–34 was not captured in the model’s estimated impact as the model default assumed that no patients aged 25–34 received routine biomarker testing under either scenario.

*Testing rates are capped at 90% over all age and gender strata. Sensitivity analyses of biomarker testing rates are conducted as percentage changes in all age- and gender-specific testing rates. Because testing rates among patients aged 25–34 equal 0% under the model default, the effect of testing patients aged 25–34 is not captured in any sensitivity.