Prediction of Long-Term Incidence of Chronic Cardiovascular and Cardiopulmonary Diseases in Primary Care Patients for Population Health Monitoring: The Intermountain Chronic Disease Model (ICHRON)

Abstract

Objective: To apply the practical parsimonious modeling method of the Intermountain Mortality Risk Score in a primary care environment to predict chronic disease (ChrD) onset.

Patients and methods: Primary care patients free of ChrD (women: n=98,711; men: n=45,543) were evaluated to develop (70% [n=95,882] of patients) and validate (the other 30% [n=48,372]) the sex-specific Intermountain Chronic Disease Risk Score (ICHRON) if seen initially between January 1, 2003, and December 31, 2005. The sex-specific ICHRON was composed of comprehensive metabolic profile and complete blood count components and age. The primary outcome was the first diagnosis of coronary artery disease, myocardial infarction, heart failure, atrial fibrillation, stroke, diabetes, renal failure, chronic obstructive pulmonary disease, peripheral vascular disease, or dementia within 3 years of baseline.

Results: At 3 years, 9.0% of men (mean age, 44±16 years) and 6.6% of women (mean age, 42±16 years) received a diagnosis of ChrD. In the derivation population, C-statistics were 0.783 (95% CI, 0.774-0.791) for men and 0.774 (95% CI, 0.767-0.781) for women. In the validation population, C-statistics were 0.774 (95% CI, 0.762-0.786) for men and 0.762 (95% CI, 0.752-0.772) for women. Evaluation of 10-year outcomes for ICHRON and analysis of its association with each outcome individually at 3 years revealed similar predictive ability.

Conclusion: An augmented intelligence clinical decision tool for primary care, ICHRON, is developed using common laboratory parameters, which provides good discrimination of ChrD risk at 3 and 10 years.