Why conventional CV risk scores fail Native American patients

Cardiovascular disease (CVD) encompasses a range of conditions affecting the heart and blood vessels, including coronary artery disease, stroke, heart failure, and peripheral artery disease. These conditions are driven by a complex relationship between genetic predispositions, environmental factors, and lifestyle choices. Established risk factors such as hypertension, dyslipidemia, diabetes, obesity, and smoking are well-documented contributors to CVD development and progression. For the general population, risk prediction models like the pooled cohort equations (PCE) are widely used to estimate the 10-year risk of atherosclerotic cardiovascular disease (ASCVD) events, guiding decisions on statin therapy and other preventive measures.

But these models, while useful for many, are not universally applicable. Native American populations exhibit distinct epidemiological profiles, including a higher prevalence of type 2 diabetes, metabolic syndrome, and certain inflammatory markers, often at younger ages. These differences mean that a model calibrated on a different population may not accurately capture the true risk in Native American individuals. The consequence is either overtreatment, exposing patients to unnecessary medication and side effects, or, more critically, undertreatment, leaving high-risk individuals unprotected from preventable cardiovascular events. This unmet need for precise risk stratification is a significant barrier to equitable cardiovascular care.

The limitations of current risk assessment

The standard approach to cardiovascular risk assessment typically involves gathering demographic information, medical history, and laboratory values to calculate a composite risk score. These scores are then used to categorize individuals into low, intermediate, or high-risk groups, informing clinical guidelines for primary prevention. For instance, guidelines often recommend statin therapy for individuals with a 10-year ASCVD risk above a certain threshold. The underlying assumption is that the predictive power of these models holds across all ethnic and racial groups.

But this assumption does not hold true for Native American populations. Studies have consistently shown that models like the PCE can misclassify a substantial proportion of Native American individuals. This misclassification is not random; it often reflects systemic biases in the data used to develop and validate these models. If a model does not adequately account for the unique risk factor prevalence, genetic variations, or socioeconomic determinants of health within a specific population, its predictive accuracy will suffer. This is particularly relevant for conditions like diabetes, which has a higher prevalence and often an earlier onset in many Native American communities, significantly impacting cardiovascular risk.

The mechanisms behind these disparities are complex and multifactorial. Genetic factors play a role, with certain genetic variants being more prevalent in Native American populations that may influence lipid metabolism, inflammation, and glucose regulation. Environmental factors, including access to healthy foods, safe physical activity spaces, and quality healthcare, also contribute. Socioeconomic determinants of health, such as poverty, historical trauma, and systemic discrimination, further exacerbate these issues, creating a unique risk landscape, the market, that is not adequately captured by broad-based risk calculators. For clinicians, understanding these nuances is critical for providing effective care, and a comprehensive resource like the Oxford Handbook of Cardiology can offer valuable insights into population-specific considerations.

Developing tailored prediction tools

Addressing the shortcomings of existing risk prediction models requires a concerted effort to develop and validate tools specifically for Native American populations. This involves collecting comprehensive, longitudinal data from diverse Native American communities, ensuring that the models are built on a foundation that accurately reflects their unique risk profiles. Such models would incorporate traditional risk factors but also consider additional variables that may be more relevant to these populations, such as specific biomarkers, genetic markers, or social determinants of health.

The development process for these tailored tools must be community-engaged, involving Native American leaders, healthcare providers, and community members at every stage. This collaborative approach ensures that the models are not only scientifically sound but also culturally appropriate and acceptable to the populations they are intended to serve. Without this engagement, even a statistically robust model may fail to achieve clinical uptake and impact. The goal is to create tools that empower clinicians to make more informed decisions about preventive therapies, ultimately reducing the burden of cardiovascular disease in these underserved communities.

The open-label design of many observational studies in this area is an obvious caveat, as is the inherent difficulty in standardizing data collection across diverse tribal nations. The trial was not powered to detect differences in specific tribal subgroups, and that gap matters for truly personalized medicine. Future research needs to focus on larger, prospective cohorts with detailed phenotyping and genetic data to refine these models further. Only then can we move towards a truly equitable approach to cardiovascular risk prediction and prevention.