Positive Predictive Value of Universal Cardiovascular Disease Risk Assessment for Pregnant and Postpartum Patients.

BACKGROUND: Cardiovascular disease (CVD) accounts for over one-fourth of pregnancy-related deaths in the United States. Timely recognition is critical for an appropriate response that has the potential to reduce maternal morbidity and mortality. OBJECTIVES: The purpose of this study was to evaluate the predictive value of the CMQCC CVD risk assessment algorithm among a diverse population of pregnant and postpartum women. We hypothesize that the algorithm will demonstrate superior predictive value compared to any individual element within the algorithm alone. METHODS: We conducted a retrospective chart review from 2020 to 2024 at 4 major health networks in the United States. The positive predictive value of each element and physical exam (PE) findings included in the CMQCC CVD risk assessment algorithm was calculated, and a multivariable logistic regression model was fitted. A C-statistic was calculated to measure the algorithm's accuracy in classifying CVD cases, and the Hosmer and Lemeshow goodness-of-fit test was used to assess model calibration. RESULTS: CVD risk assessment was performed in 19,238 pregnant and postpartum patients using CMQCC algorithm. Of these patients, 2.0% were identified as being "at risk" for CVD, and 2.3% of patients were subsequently confirmed to have CVD, yielding an overall positive predictive value of 15.6% (95% CI: 0.1278-0.1887). PE findings were the most predictive elements of the risk assessment algorithm. CONCLUSIONS: The CMQCC CVD Risk Assessment algorithm effectively identifies pregnant and postpartum patients with previously unrecognized CVD. While the overall predictive value is moderate, the PE findings drive the algorithm's strongest predictive performance.