Community Collaborators Digging Deeper to Solve High Infant Mortality
Mar 17, 2017
Between 2008 and 2012, 434 infants died in Allegheny County within their first year, ranking the county slightly worse than the national average. While blacks make up only 13.2% of the Allegheny County’s population, black infant deaths over a four-year period constituted 42% (225) of all infant deaths. This racial disparity in the infant mortality rate is 27% larger in Allegheny County compared to the national rate.
Despite historical investment and intense focus by various groups both locally and across the country, infant mortality has only been slightly ameliorated. Resolute toward an answer, the Richard King Mellon Foundation made multiple grants in June 2016 to Magee-Womens Research Institute, the University of Pittsburgh, and RAND Corporation to form a broad collaboration with experts and community partners to seek answers using data and predictive modeling.
Dr. Yoel Sadovsky, Director of Magee-Womens Research Institute, and lead investigator on the infant mortality project, makes this observation: “Realizing that the most common conditions associated with infant mortality occur before birth, it is clear that a healthy pregnancy is key to prevention of child death or disease by year one of life."
Specifically, Richard King Mellon Foundation granted $4 million to Magee-Womens Research Institute and Foundation (MWRIF) to support the growth and utility of the Magee Obstetrical Maternal Infant Database (MOMI), and $1 million to MWRIF to support basic research relating to infant mortality.
The University of Pittsburgh and RAND Corporation who received separate grants of $725,000 and $640,000, respectively, will enhance the MOMI data surrounding mothers and babies at risk, and create algorithms to predict or score mortality risk. Using Allegheny County as a microcosm of the infant mortality trends in various pockets of the country this collaboration – of grant recipients, data sources, and social agencies will identify those predictive factors that contribute to infant mortality. Later, interventions using this information will be tested at Magee clinical sites.
MOMI’s Play Book
The MOMI Database and Biobank is an electronic database comprised of patient information about deliveries at Magee-Women’s Hospital of UPMC. It contains more than 300 variables for nearly 190,000 deliveries from January 1, 1995 to the present. The Richard King Mellon Foundation grant will be used in part to upgrade the MOMI databank to collect up to 500 variables, and fund 10 competitive research projects chosen for their potential to reduce infant mortality in the short-term.
Magee’s database and approach to the problem of infant mortality are unique because of its preventive capabilities. With the MOMI longitudinal view of generational births and the DNA Biobank, Magee- Womens Research Institute and its collaborators expect to develop a predictive model for prevention of, and clinical interventions to reduce, infant mortality.
“The MOMI database provides more than 20 years of clinical data, and will grow to integrate new features including a Biobank. The immediate link to real time clinical care makes this platform unique and powerful,” observes Dr. Janet Catov, Associate Professor, Department of Obstetrics, Gynecology & Reproductive Sciences and the Department of Epidemiology, University of Pittsburgh. Dr. Catov oversees the MOMI database.
Risk factors such as weight gain, low birth weight, chromosomal defects, congenital diseases, extreme maternal age, and many other medical identifiers already collected in MOMI will be combined with other external data surrounding maternal lifestyle, educational status, employment status, home environments and many other social indicators to create an extremely robust tool that will be designed to provide a 360° view in to the lives of mothers and babies with risk of death before the age of 1.
“While Allegheny County has a robust health and social services network of support within the maternal and child healthcare system, the problem of high infant mortality rates remains,” said Rich Tsui, Ph.D., and Associate Professor in Pitt’s Department of Biomedical Informatics.
Dr. Tsui and his colleagues will create a program to cross-reference infant outcomes with a wide variety of information, including socio-economics, race, maternal health, education and age, vaccination records, parental smoking habits, health department air quality reports, medications, and ultrasounds, among others. The results will help the scientists create a tool that considers certain clinical and non-clinical characteristics of a patient and her pregnancy or baby, and determines the risk of the baby dying in its first year of life.
Evan Peet, Ph.D. and his team at RAND Corporation will evaluate the effectiveness of potential and existing interventions – ranging from social services to clinical procedures – by incorporating the risk predictions and information regarding mother’s and child’s program participation. The results will indicate what will work best to alleviate the negative effects of the various and specific risks that each mother and child face. “Our goal is to help clinicians refer the right patient to the right intervention at the right time,” says Peet. Once the tool is created, the data feeds established and the algorithms tested, the next phase will be to deploy a pilot with the physicians and patients in the outpatient clinic at Magee-Womens Hospital of UPMC, and other possible locations. The team will make adjustments to the tool based on feedback from the pilot tests.
Partnerships help communities find answers
More partnership with social agencies in Allegheny County yields valuable data to be used to populate the databases against which Pitt’s and RAND’s algorithms can be run. Dr. Karen Hacker, Director of the Allegheny County Health Department commented, “The Allegheny County Health Department has long been involved with addressing infant mortality and the disparities that exist. While we have had some success, there is work to do. This data collaboration will help us dig deeper and better understand the risk factors and the interventions that work so we can have an even greater impact.”
The Allegheny County Department of Human Services includes five programmatic offices including aging, behavioral health, children, youth and families, community services, and intellectual disability. Marc Cherna, director of this agency since 1997 is enthusiastic, “The Allegheny County Department of Human Services is eager to help forward this collaborative effort to reduce the scourge of infant mortality by leveraging our robust data and enlisting the assistance of agencies in our provider network. For more than two decades, our Department has dedicated itself to fostering healthy children, strong families and vibrant communities.”
Moving the Needle in Infant Mortality
So how do we remediate infant mortality using MOMI? How can we personalize outcomes for at-risk births? How does it translate to the woman who is high risk for infant mortality? What agencies need to be involved, and why? These questions and more will be answered in a later phase of this project. The first phase is to create the tool that will enable efficient and rapid risk identification. The second phase will implement the risk prediction tool and the interventions, and will continually evaluate intervention effectiveness.
“Interventions will be impactful and effective if patients are engaged,” says Dr. Robert Edwards, Chair, University of Pittsburgh’s Department of Obstetrics, Gynecology & Reproductive Sciences, “and patients will be engaged by providing infrastructure, resources including new technologies to promote engagement, and funding for obstetrical care in the county’s lower socio-economic sectors. This is exactly what we hope to do.”
Looking to the future success of the program, Dr. Tsui extends this offer. “If other hospitals would like to participate, they’re more than welcome,” said Dr. Tsui. “We want to get input from others involved in the problem of infant mortality so we can ultimately save lives.”
For more information on the Richard King Mellon Foundation grant, Magee-Womens Research Institute, or any other entity listed in this article, contact: Jane Rudov, Director of Communications, Magee-Womens Research Institute and Foundation (412) 641-8934 rudovjj@mwri.magee.edu