Feasibility of implementing a standardized preeclampsia risk screening tool into the electronic medical record

TL;DR

This study shows that adding a preeclampsia risk screening tool to electronic health records is feasible but finds disparities in screening rates based on race, language, and insurance.

Contribution

The study demonstrates the feasibility of integrating a standardized preeclampsia risk screening tool into the EHR and identifies sociodemographic disparities in screening rates.

Findings

55.3% of eligible patients were screened using the tool, with usage increasing from 46.2% to 67.4% over a year.

Hispanic/Latino patients had lower odds of being screened compared to Non-Hispanic White patients.

Non-English speakers and publicly insured patients were less likely to be screened than English speakers and privately insured patients.

Abstract

Despite the fact that low‐dose aspirin (LDA) is the only effective intervention for preeclampsia risk reduction, not all those at increased risk of preeclampsia are recommended it. This study examines feasibility of implementing a preeclampsia risk screening tool into the electronic health record (EHR) to prompt LDA recommendation when appropriate and assesses whether there are differences in screening rates by race/ethnicity, language, or insurance payor.

A standardized preeclampsia risk screening tool was launched in September 2022, which was intended to be used for all obstetrical patients to assess LDA eligibility. To assess feasibility of this tool, we conducted a retrospective cohort study of patients who delivered at an academic medical center from July 2023 to June 2024. The primary outcome of this was feasibility, defined as percentage of patients who delivered each month who had been screened using the tool. The secondary outcome was association between sociodemographic characteristics and screening. Chi‐square and Student's t‐tests were used to compare characteristics between screened status, and multivariable logistic regression determined strength of association. Logistic regression was used to determine differences in screening rates over each month.

A total of 4556 patients met inclusion criteria, and 2521 (55.3%) were screened using the tool. Tool use increased from 46.2% in the first month to 67.4% in the last month of the study, with the odds of the tool being used increasing on average 7% per month over 1 year (p < 0.05). Compared to patients who identify as Non‐Hispanic White, patients who identify as Hispanic/Latino had lower odds of being screened when adjusting for ethnicity/race, language, insurance, body mass index (BMI), and age (adjusted odds ration [aOR], 0.75; 95% confidence interval [CI], 0.62–0.89). In the same adjusted analysis, non‐English speaking patients were less likely to be screened than English‐speaking (aOR, 0.51; 95% CI, 0.36–0.71 and aOR, 0.51; 95% CI, 0.36–0.73) and publicly insured patients were less likely to be screened than privately insured (aOR, 0.83; 95% CI, 0.72–0.97).

Preeclampsia risk screening increased substantially after implementing this standardized tool into the EHR which demonstrates its feasibility. We observed disparities in screening rates by race/ethnicity, language, and insurance payor.

Figures