Four Tips from States on Integrating Maternal and Child Health Data Systems

What happens when state health programs use separate data systems to serve the same population, such as mothers and children? State policymakers know that when those data systems do not “talk” to one another, states may waste resources on duplicative data entry and system maintenance; providers and state agencies may struggle to access information important to the health of mothers and children; and care may be uncoordinated or otherwise compromised.
Some states are tackling the data integration challenge head-on. In spring 2015, NASHP, in partnership with the Iowa Child and Family Policy Center, asked state officials in Illinois, New Jersey, Rhode Island, and Utah–and a state contractor in Connecticut—to share their experiences integrating maternal and child health data systems.
The table below shows maternal and child health data included in integrated data systems:

	Hearing Screening	Lead Screening	Immunizations	Medicaid /CHIP claims	Vital Records	Newborn Bloodspot Testing	WIC	Oral/Dental Health	Early Intervention
CT				X	X			X
IL	X	X	X	X	X	In process	X	X
NJ	X	X	X		X	X
RI	X	X	X		X	X	X		X
UT	X		X		X	X			X

Considering Data Integration? Tips from States
From our conversations with states, we distilled four recommendations that may be useful to others seeking to develop an integrated data system.

1. Start with a thorough needs assessment, and consider using a contractor who will not be involved in the development and implementation process.

In Utah, the state project team worked with a contractor to perform a needs assessment as part of their two-year planning and assessment process. At the conclusion of the process, Utah entered into a non-competitive, sole-source contract with a different contractor to develop and implement the CHARM data system.
Rhode Island also issued an RFP solely for the planning and requirements documentation process. The winning vendor was barred from submitting a proposal to construct the system, to prevent vendors from limiting the potential of the system by only proposing to build something within their own capabilities.
A strong needs assessment can also sharpen the focus on the intended uses for the system, which may include program monitoring and coordination, research and policy-making, and clinical use.

2. States may benefit from building a system that can easily accommodate future use by legislators, researchers, advocates or others.

While state agencies in Connecticut rely on Connecticut Voices for Children’s integrated data for program evaluation and Title V reporting, reports based on analyses of the Medicaid/CHIP data are also used by legislators, policymakers, researchers, providers, and advocates in Connecticut. States could maximize the utility of an integrated system by designing it with future evidence-based policy initiatives or research projects in mind.

3. States should consider whether to replace existing data systems with a new integrated system, or allow existing systems to remain in place.

Several states reported satisfaction with the integration of their maternal and child health data systems using a federated model that does not replace existing data systems. For example, the Illinois Departments of Public Health, Healthcare and Family Services, and Human Services all maintain their own systems but contribute data, pursuant to a multi-agency data sharing agreement, to a data warehouse maintained by the Medicaid agency.
The state agencies appreciate being allowed to keep their own data systems to generate required reports for federal agencies and other entities. Utah’s federated system also reduced duplicative tests and expedited appropriate referrals, services, and follow-up. Rhode Island’s integrated system, KIDSNET, is the sole database for many programs, although some programs still maintain their own data systems and use KIDSNET as their data warehouse to provide web-based user access to providers and other community partners.

4. States should approach vendor contracting with an eye toward the long term.

Utah’s contractor remained involved in the data system over the long term, and the state feels that this longstanding relationship has been beneficial. Similarly, Rhode Island reports that using the same vendor long-term helps ease staff transitions and provides helpful historical knowledge.
Some states emphasized the importance of anticipating changes to the agreement between the state and the vendor. Connecticut’s agreement included a process by which either the state or the vendor could initiate a request to conduct optional activities related to the contract’s scope of work.
Integrating maternal and child health data systems is a significant undertaking that requires a considerable investment of state resources. States embarking on a data integration project may be able to learn from the experiences of other states in order to maximize the potential of integrating data to better serve families, providers, and other stakeholders.