The coverage gap in PREVENT: Why cardiovascular risk prediction needs a lab-free complement

The American Heart Association's PREVENT equations are now embedded in three major AHA/ACC guidelines governing hypertension, dyslipidemia, and cardiovascular-kidney-metabolic syndrome, including a specific risk threshold that gates initiation of GLP-1-based therapy and SGLT2 inhibitors. This rapid guideline uptake has outpaced a persistent limitation: PREVENT requires laboratory values, chiefly lipids and estimated glomerular filtration rate, that are frequently missing in the populations where risk prediction matters most, including patients seen in community health centers, emergency departments, and substance use treatment programs. Real-world validation studies across multiple health systems show that PREVENT underestimates cardiovascular risk most severely in these same subgroups, with the largest underestimation among Medicaid-insured, uninsured, and Black patients. Imputation, whether simple substitution or multiple imputation, does not resolve this problem because the probability that a laboratory value is missing is itself associated with the underlying risk the model is trying to estimate, a form of informative missingness that standard imputation methods are not designed to address. We argue that the field should invest in developing and validating a standalone cardiovascular risk prediction model that does not require laboratory measures. Such a model would function as a complement to PREVENT, not a replacement, providing risk estimates for patients without laboratory data and helping identify those who warrant expedited evaluation. As PREVENT becomes the gatekeeper for guideline-directed therapy across three clinical domains simultaneously, closing this coverage gap is no longer a methodological refinement but a clinical implementation imperative.