# Comparison of framingham risk model, atherosclerotic cardiovascular disease risk model, and assign risk model in detecting sub-clinical atherosclerosis among DIMAMO residents, Limpopo province, South Africa

## TL;DR

This study compared three cardiovascular risk models in South Africa to see if they can detect early signs of atherosclerosis, but found they were not effective.

## Contribution

The study evaluates the performance of three CVD risk models in a South African population for detecting subclinical atherosclerosis.

## Key findings

- The Framingham, ASCVD, and ASSIGN risk models showed no significant association with increased CIMT.
- The models had poor discriminative ability to distinguish individuals with subclinical atherosclerosis.
- Risk scores were not significantly different between normal and increased CIMT groups.

## Abstract

Metabolic and Cardiovascular risk factors affect the outcome of an individual's cardiovascular risk scores. There are several cardiovascular disease (CVD) risk models developed to predict CVD risk in individuals, although most of the CVD risk models are not validated, or their performance is understudied in some populations. The study aims to evaluate the performance ability of these CVD risk models in distinguishing individuals with increased Carotid Intima Media Thickness (CIMT).

The study was retrospective and involved 245 participants' data. Three CVD risk models were evaluated and compared for the ability to determine the association between baseline risk scores and a cross-sectional marker of subclinical atherosclerosis. The data were analyzed using the Statistical Package for SPSS, version 30. A T-test was used to compare continuous CVD risk variables between groups, a Chi-square was used to compare the proportion of categories for the Framingham risk scores, the Atherosclerotic, and chi-square test was used to compare the proportion of categories for the Framingham risk scores, the Atherosclerotic Cardiovascular Disease (ASCVD) risk score, and the ASSIGN risk scores between groups. Logistic regression and Receiver Operating Characteristic (ROC) analyses were used to determine the model's accuracy in terms of sensitivity and specificity. A p-value of less than 0.05 was considered statistically significant.

The mean age for individuals with high risk was 60 years. The proportion of high Framingham risk score (FRS), intermediate ASCVD, high ASCVD, very high ASCVD, and high ASSIGN risk scores were statistically not significant between normal CIMT and increased CIMT participants. The Framingham risk model, the ASCVD risk model, and the ASSIGN risk model all showed no significant association with CIMT.

In this study, the CVD risk models' performance and association were poor, with poor discriminative ability to distinguish individuals with increased CIMT.