The Ohio Department of Health (ODH) is utilizing the Ohio Data Analytics technology to develop, expand and enhance predictive models to determine characteristics of those at risk for infant mortality and design targeted interventions based on this data. For the first time ever, the State has been able to successfully and securely link 31 datasets across four state agencies. Ultimately, ODH and other state and local partners can use this data to design and administer targeted interventions for mothers at-risk, measure program efficacy, help more Ohio babies reach their first birthdays and eliminate racial and ethnic disparities in birth outcomes.
The project team is approaching the initiative with a Targeted Intervention Roadmap, designed to identify novel, actionable interventions to reduce and prevent Ohio’s infant mortality rate.
ODH’s project addresses four key questions:
- Characteristics of mothers and infants most at risk of infant death?
- Characteristics of families most likely to benefit from targeted interventions?
- Characteristics of families most likely to participate in targeted interventions?
- Which intervention programs yield the best return on investment?
The first phase of the project resulted in creation of the “G20 model” to predict characteristics of mothers most likely to benefit from interventions. It predicts Extreme Prematurity (~20 Weeks Gestation) based on facts the State knows a priori of a clinical encounter, to target vulnerable moms at risk of preterm birth/infant mortality. ODH and local partners are now planning for deployment of interventions via targeted data-driven pilots to reduce the infant mortality rate and preterm births in key Ohio metros, building on the data insights in the analytics, targeted to the highest risk mothers. Some examples include: targeted interventions delivered via home visitation programs, referrals to supplemental nutrition programs, reduction of risk behaviors (smoking, drinking, drugs), limited rollout of universal screening in birth hospitals, and more.