Stepwise approach to screen high-risk individuals using the non-laboratory-based and laboratory-based CVD risk scoring.

Abstract

Objective: We compared non-laboratory models' efficacy with standard laboratory-based model in identifying high-risk populations for cardiovascular disease (CVD) in resource-limited settings.

Methods: A national sample of 121 672 individuals aged 40-70 from the PERSIAN cohort was analyzed. Non-laboratory models, including the World Health Organization (WHO) and Iranian pooled-cohort CVD mortality models, were compared with the WHO laboratory-based model. Intra-class correlation coefficient (ICC) and concordance correlation coefficient (CCC) were utilized. Sensitivity and specificity of non-laboratory models were evaluated against the laboratory-based one at various risk thresholds. The number of reduced tests in the stepwise approach was calculated considering the Iranian census.

Results: Both non-laboratory and laboratory-based models showed similar trends in predicting CVD risks across age groups. Strong correlations and concordance were observed in both men (ICC: 94.4%, CCC:0.893) and women (ICC: 93.8%, CCC:0.883). Utilizing a 5% risk threshold for WHO non-laboratory and 2% for the Iranian pooled-cohort CVD mortality model as the initial step achieved high sensitivity (99.6%) and moderate specificity (52%) for identifying candidates for the second-step laboratory test. This approach effectively reduced the number of tests by 16 807 982.

Conclusion: Non-laboratory models, in a stepwise approach, offer a promising strategy to alleviate strain on financial resources and enhance healthcare system efficiency in resource-limited countries.