Resource Efficient Screening for Primary Prevention of Coronary Heart Disease: A Proof‐of‐Concept Test in the MESA Cohort

Figures

• 

  Figure 1. Outline of current guideline‐based strategy using PCE and CAC scoring vs the novel self‐report–based screening strategy. A, Current guidelines are based on the PCE with or without subsequent CT screening. If PCE‐estimated risk is high (≥20%), a person is eligible for lipid‐lowering therapy. If PCE‐estimated risk is intermediate (PCE ≥7.5% to <20%) and CACS is ≥100, treatment is also warranted. If these rules are followed, 852 of 4564 individuals (18.7%) are deemed eligible for lipid‐lowering therapy. B, In the self‐report screening strategy, we first estimate the risk of having a high CACS (≥100) based on self‐reported data. A CT scan for CAC scoring is then performed in those ranked at the highest risk. Treatment decision is then based on the measured CACS only. CAC indicates coronary artery calcification; CACS, CAC score; CT, computed tomography; and PCE, pooled cohort equation.

• 

  Figure 2. Flow diagram of the study inclusion. Individuals aged >75 years were excluded to optimize the performance of the PCE‐based strategy. Individuals with already ongoing lipid‐lowering therapy or having diabetes were excluded because the model will not change their need for lipid‐lowering therapy. MESA indicates Multi‐Ethnic Study of Atherosclerosis; and PCE, pooled cohort equation.

• 

  Figure 3. CACS distribution in groups eligible for lipid‐lowering therapy as defined by either strategy. A, CACS distribution in the group deemed eligible for lipid‐lowering therapy as defined by the current guideline‐based PCE strategy. In the subgroup deemed eligible for lipid‐lowering therapy based on a PCE >20%, 123 individuals had a CACS score of 0. B, CACS distribution in the group deemed eligible for lipid‐lowering therapy by the self‐report strategy. The lowest CACS score in the self‐report group was 18. CACS indicates coronary artery calcification score; and PCE, pooled cohort equation.

• 

  Figure 4. Health care resources required by the 2 strategies and their capacity to identify individuals who develop CHD. A, Health care resources required for the 2 screening strategies. The self‐report–based strategy requires fewer initial health care visits and, by design, the same number of CT scans and treatments. Note that the similarity in proportions of CT scans and treatments between strategies is intentional for comparison; however, the specific individuals identified being at high risk differ. B, The self‐report screening strategy shows a higher capacity to identify individuals who later develop CHD compared with the PCE‐based strategy (PCE strategy, 47.7%; self‐report strategy, 55.6%; P value for difference=0.017). CHD indicates coronary heart disease; CT, computed tomography; and PCE, pooled cohort equation.

• 

  Figure 5. Three examples of real‐world screening scenarios using self‐report with fixed cutoffs. The risk cutoff for being eligible for CT screening and the CAC cutoff for treatment can be tuned according to preferences for screening/treatment availability. We selected strategies in which the risk of having a high CACS (≥100) based on self‐reported data varied between >15% and >20%. The CACS score at which lipid‐lowering therapy was recommended varied between >0 AU and >20 AU. This created a conservative (A), intermediate (B), and liberal (C) strategy when compared with our benchmarked model. The resulting number of health care visits, CT screenings, treatments, and event rate is presented in Figure 6. AU indicates Agatston unit; CAC, coronary artery calcification; CACS, CAC score; and CT, computed tomography.

• 

  Figure 6. Resources required and treatment accuracy of all 5 screening strategies presented. A, Resources required for each screening strategy. The self‐report–based strategies require fewer health care visits and a comparable number of CT scans and treatments. B, The self‐report screening strategies show equal or greater ability to predict CHD events compared with the PCE‐based strategy (PCE strategy, 47.7%; self‐report conservative strategy, 51.9%; self‐report intermediate strategy, 56.9%; liberal strategy, 63.0%). CHD indicates coronary heart disease; CT, computed tomography; and PCE, pooled cohort equation.