OMNI Research Group - Obstetrics & Maternal Newborn Investigations

Research

Ensuring the best possible health for both a mother and her child begins before conception, pregnancy and at birth. Many conditions or complications of pregnancy can be prevented, or their severity reduced through the management of risk-factors during the pre-pregnancy and pregnancy period. Research in maternal child health is vital to answering the many important questions that will change care and improve maternal and child health outcomes. 

Immunization in pregnancy

Lead investigators

In the early stages of the COVID-19 pandemic, experts across disciplines recognized that data collection and research on COVID-19 in pregnancy were urgently needed. In April 2021, the OMNI Research Group established the COVID-19 Pregnancy Event (COPE) Network — a collaboration of 13 obstetrical hospitals in 6 of Ontario’s largest cities, which collectively handle 75 per cent of hospital births in Ontario each year. The COPE Network includes experts in maternal-fetal-medicine, perinatal epidemiology, public health, molecular virology and infectious diseases.

Over the last 2 years we have launched a series of CIHR-funded seroprevalence, vertical transmission and surveillance investigations designed to generate rapid, rigorous evidence specific to the impacts of COVID-19 on maternal, fetal and newborn health and the role of COVID-19 vaccines for protecting mothers and their newborns. Our findings will inform strategies to optimize processes in care and patient counselling and improve health systems management of COVID-19 in pregnancy.

The purpose of this study is to find out what effects COVID-19 vaccines have on the immune systems of pregnant women/individuals and their babies. We will measure immune responses in vaccinated participants and their babies after they are born, and document vaccine-related reactions and health outcomes that may occur after vaccination.    

 In addition, the data from this study will be compared with data from other COVID-19 research studies. We will measure differences in the mothers’ and babies’ immune responses between individuals who received a COVID-19 vaccination in pregnancy and those who had COVID-19 in pregnancy. We will also measure differences in the immune responses to COVID-19 vaccination between pregnant and non-pregnant women/individuals.   

Methods and modes of delivery

Lead investigators

Caesarean sections are the most common inpatient surgical procedures in North America. Caesarean sections may be medically indicated for reasons including complications during pregnancy or the labour process, and issues compromising fetal growth and wellbeing. Pregnant individuals living with obesity and those with diabetes during pregnancy are also more likely to be submitted for caesarean section. Further, some pregnant individuals without medical indications may request Caesearan deliveries for reasons including scheduling convenience, anxiety about labour pain and process, and fear surrounding possible pelvic floor damage and sexual dysfunction after vaginal delivery.  

Thus, whereas unique populations and scenarios may require caesarean sections for the health and wellbeing of the birthing parent and the fetus/neonate, others receive the procedure in the absence of medical indications. This research program explores the optimal timing, mode and method of delivery across obstetrical sub-populations, and examines the risk factors and short and long-term maternal and offspring outcomes associated with caesarean sections compared to vaginal deliveries.  

Substance use in pregnancy

Lead investigators

Cannabis use has been increasing in Canada, including among pregnant Canadian women. Although we anticipate further increases because of greater availability of cannabis and low perceptions of harm, there is a lack of conclusive evidence on the short-term outcomes and long-term sequelae of exposed children.    

Using provincial data from BORN Ontario and ICES, we are conducting population-based cohort studies to examine pregnancy, neonatal and childhood health outcomes following exposure to cannabis and other substances. We are using these data to evaluate patterns in cannabis and other substance use in pregnant women/individuals in Ontario, and the association between substance use in pregnancy and newborn outcomes including preterm birth, birthweight and newborn admission to neonatal intensive care units and re-hospitalization. We are also exploring the longer-term health outcomes of children exposed to cannabis and other substances in pregnancy and through breastfeeding and evaluating the knowledge, attitudes and practices regarding cannabis use in pregnancy among Canadians.  

The findings from this research program will yield important results to assist in shaping comprehensive policy and public health messaging for Canadian women and their health care providers.  

Diabetes in pregnancy

Lead investigators

Gestational diabetes mellitus (GDM) is a form of glucose intolerance characterized by onset or detection during pregnancy. The incidence of GDM is increasing across all provinces and territories. Both pre-existing diabetes and GDM are associated with an increased risk of obstetrical complications and adverse fetal outcomes. In addition, a history of GDM is associated with increased risk of GDM in future pregnancies and development of type-2 diabetes and cardiovascular disease in later life.  

We are characterizing this unique population of women, the patterns of care that they receive during pregnancy, as well as their pregnancy, neonatal and breastfeeding outcomes. We are also examining how the persistence or development of postpartum risk factors are associated with adverse outcomes in later life and in subsequent pregnancies. With this data we will develop predictive models capable of identifying populations at highest risk of GDM and other outcomes associated with hyperglycemia in pregnancy.   

Maternal connections to specialized perinatal care

Lead investigator

The rising trend in non-communicable diseases (NCDs), such as endocrine disorders, cardiovascular diseases, cancer, and diabetes, are a public health priority that is increasingly affecting those of reproductive age, including women during pregnancy. Currently, one in five women in Canada do not achieve adequate prenatal care, and women with NCDs may require additional specialized care than what is required by the standard prenatal care schedule.

The optimal perinatal management for women with NCDs includes adequate prenatal care, care from specialists during pregnancy, and coordinated care during the pre-conception, pregnancy and post-partum period. However, understanding and defining what is ‘appropriate’ care is complex and currently unclear. With limited evidence, there is a need to evaluate the standard prenatal care guidelines including frequency and timing of visits, review patterns of access to specialists, and evaluate multidisciplinary care health utilization for women with NCDs. Of importance, the experience of patients with NCDs is central to understanding the barriers to care that exist and what drivers contribute to a healthy pregnancy, however, this perspective is often missing.

This program is designed to evaluate care pathways of women with non-communicable diseases who require multidisciplinary care during pregnancy in order to identify gaps/inconsistencies in the provision of prenatal and specialist care including the social inequities perpetuating these gaps. 

Predictive modelling and artificial intelligence

Lead investigators

Artificial intelligence is increasingly being used to predict health outcomes from large, complex datasets. Machine learning, a branch of AI, and deep learning, a particularly powerful type of machine learning modeled loosely on the human brain, are especially well-suited to this task: these models can be continuously refined as new data becomes available, making them valuable tools for supporting medical diagnostics and clinical decision-making. Ontario's extensive health information systems offer a unique opportunity to develop, test, and validate these prediction algorithms in maternal, newborn, and child health.

This research program applies machine learning to clinical and administrative datasets to build prediction models that identify populations at high risk of adverse obstetrical, infant, and child health outcomes, as well as those most likely to benefit from specialized care, pharmacological treatment, or surgical intervention. A particular focus of our work is the use of deep learning models to analyze obstetrical imaging, enabling earlier identification of fetal anomalies and maternal complications. Together, these findings will help identify gaps in care, direct treatments and interventions to those who stand to benefit most, and ultimately improve outcomes for mothers and children.

A related focus of our research program is the use of large language models (LLMs), a type of deep learning model trained on vast amounts of text to understand and generate human-like language and reasoning. To ensure their outputs are accurate and grounded in reliable clinical evidence, we pair LLMs with retrieval-augmented generation (RAG), a technique that allows the model to reference trusted external sources, such as clinical guidelines or peer-reviewed literature, in real time, rather than relying solely on information learned during training. Using this combined approach, our team is developing patient-facing educational materials, evaluating the quality of medical information, and exploring clinical diagnostic, quality and safety support tools, with the goal of making accurate, evidence-based information more accessible to both patients and clinicians.