Revolutionizing Maternal Health Prediction: The Launch of PregPoRT

In a groundbreaking advancement for Canadian public health, researchers from the University of Toronto have published the study protocol for the Adverse Pregnancy Outcomes Population Risk Tool, known as PregPoRT. This innovative predictive model aims to estimate the population-level risk of adverse pregnancy outcomes (APOs) like gestational diabetes, preeclampsia, and placental abruption using routinely collected national data. Published today in Diagnostic and Prognostic Research, PregPoRT stands out as the first Canadian tool of its kind to integrate biomedical, behavioral, social, and environmental factors, moving beyond traditional clinical models.

Led by PhD candidate Sabrina Chiodo under the supervision of Professor Laura C. Rosella, the tool leverages linked data from the Canadian Community Health Survey (CCHS) and Discharge Abstract Database (DAD), ensuring national representativeness. This development addresses a critical gap in maternal health surveillance, where social determinants often go unaccounted for in risk assessment.

What Are Adverse Pregnancy Outcomes and Why Do They Matter in Canada?

Adverse pregnancy outcomes (APOs) encompass conditions such as gestational diabetes mellitus (GDM), preeclampsia (high blood pressure disorder with organ damage), and placental abruption (premature separation of the placenta from the uterus). These complications contribute significantly to maternal and fetal morbidity, with long-term health implications including increased cardiovascular disease risk for mothers.

In Canada, the prevalence is notable: rates of hypertensive disorders of pregnancy (HDP), including preeclampsia, have risen from 6.1% to 8.5% over the past decade, while GDM affects around 5-9% of pregnancies, varying by province. Placental abruption occurs in about 1% of deliveries but carries high risks of preterm birth and perinatal death. Collectively, APOs impact nearly one in three pregnancies, straining healthcare resources and exacerbating inequities.

These outcomes are not evenly distributed. Indigenous women, recent immigrants, and those in low-income neighborhoods face higher burdens due to intersecting social determinants like housing instability, food insecurity, and limited access to prenatal care. PregPoRT seeks to illuminate these patterns for targeted interventions.

The Limitations of Current APO Prediction Models

Existing tools for predicting APOs predominantly rely on clinical data obtained during pregnancy, such as blood pressure readings or glucose tests. While effective for individual care, they overlook pre-pregnancy social and environmental risks, limiting their utility for population-level planning.

For instance, models like those using biomarkers fail to capture modifiable factors such as neighborhood marginalization or air pollution exposure, which are crucial in Canada's diverse urban and rural contexts. This gap hinders equity-focused public health strategies, particularly for vulnerable groups where APO rates are elevated.

PregPoRT fills this void by drawing on a health equity framework from Kramer et al., incorporating multilevel determinants across the life course.

How PregPoRT is Developed: Data Sources and Methodology

PregPoRT utilizes a retrospective cohort of over 13,000 female-identifying individuals aged 15-49 from CCHS cycles (2000-2017) linked to DAD delivery records within two years. Predictors span biomedical (e.g., pre-existing hypertension, obesity), behavioral (e.g., smoking, physical activity), social (e.g., income, immigration status via CAN-Marg index), and environmental domains (e.g., air pollution from CANUE, walkability via Can-ALE).

The model employs a Weibull accelerated failure time approach with restricted cubic splines for continuous variables and LASSO for selection, accounting for survey weights and complex design. Validation includes split-sample, bootstrap, and temporal methods to ensure robustness. For more on the protocol, see the full open-access publication.

• Primary outcome: Composite APO identified by validated ICD codes.
• Follow-up: Up to 2 years post-survey.
• Exclusions: Multifetal pregnancies, Quebec data (due to DAD coverage).

This rigorous, transparent methodology positions PregPoRT for real-world scalability.

Spotlight on the Research Team at University of Toronto

Sabrina Chiodo, the lead author and PhD candidate in Epidemiology at U of T's Dalla Lana School of Public Health, drives PregPoRT's development. Supported by CIHR awards and the Leong Centre Studentship, her work builds on prior perinatal pharmacoepidemiology research.

Senior author Laura C. Rosella, Professor and Canada Research Chair in Population Health Analytics, brings expertise from pioneering the Diabetes Population Risk Tool (DPoRT), widely used for diabetes surveillance. Co-authors Sonia M. Grandi (SickKids) and Jessica Gronsbell (U of T Statistics) provide methodological strength.

Affiliations span U of T, ICES, and Trillium Health Partners, exemplifying interdisciplinary collaboration. Aspiring researchers can explore opportunities in epidemiology via research jobs or faculty positions in Canadian higher ed.

Photo by Anna Civolani on Unsplash

Prioritizing Health Equity: Social Determinants in PregPoRT

PregPoRT's equity lens distinguishes it, explicitly modeling social gradients like material deprivation, ethnic concentration, and dependency ratios from CAN-Marg. Early analyses reveal higher APO risks in deprived neighborhoods with recent immigrants and visible minorities.

This approach aligns with Canada's commitments to reduce maternal health disparities, particularly for Indigenous and Black communities facing 2-4 times higher preterm birth rates. By quantifying these influences, the tool supports upstream interventions like community nutrition programs or housing supports. Learn more about social determinants of health from Public Health Agency of Canada.

Potential Impacts on Canada's Healthcare System

Unlike clinical tools, PregPoRT enables population surveillance, forecasting APO burden for resource allocation. Public health units could use it to prioritize high-risk regions, potentially averting thousands of cases annually given APOs' 7.6% baseline prevalence in the cohort.

Integrated with existing platforms like ICES or provincial perinatal registries, it could inform policies reducing HDP rates, which climbed 40% recently. Long-term, lowering APOs could cut future cardiovascular admissions, easing system pressures amid rising maternal morbidity.

Validation and Next Steps for PregPoRT Implementation

The protocol outlines comprehensive validation: C-statistic for discrimination, calibration plots, and Brier scores, stratified by equity markers. Temporal validation with post-2017 CCHS ensures transportability. Open-source R code will facilitate adoption.

Future phases include external validation across provinces and extension to preterm birth. Knowledge translation via workshops targets health ministries. For similar predictive analytics careers, check higher ed career advice.

Building on U of T's Legacy of Population Risk Tools

PregPoRT extends Rosella's DPoRT and PreMPoRT, tools revolutionizing chronic disease prediction. DPoRT, validated internationally, predicts diabetes using survey data, mirroring PregPoRT's approach. This lineage underscores U of T's leadership in precision public health.

Collaborations with StatsCan and ICES highlight higher ed's role in national data infrastructure. Explore U of T's contributions via Canadian academic opportunities.

Stakeholder Perspectives and Real-World Applications

Obstetricians praise the equity focus, noting it could guide preconception counseling in primary care. Policymakers see value in burden forecasting amid rising GDM. For patients, it empowers community-level advocacy.

In Ontario, where HDP rose sharply, local units anticipate using PregPoRT for targeted screening. See the Leong Centre update for early insights.

Photo by Febe Vanermen on Unsplash

Future Outlook: Transforming Maternal Health in Canada

PregPoRT heralds a new era of equity-driven maternal health, potentially reducing APOs through proactive planning. As validation progresses, its integration could save lives and costs.

For those passionate about public health innovation, opportunities abound in rate my professor, higher ed jobs, career advice, university jobs, and post a job at AcademicJobs.com. Stay informed on Canadian research breakthroughs.