Federal and state governments are continuing to vaccinate residents as quickly as possible, while working to ensure they reach populations experiencing barriers to vaccination. Access to high-quality data to track and identify under-vaccinated areas and populations is critical to this goal. States have a variety of data systems at their disposal, with vaccine registries at the center.
Tracking and reporting of COVID-19 vaccine distribution and administration data requires collaboration and integration across various systems that are administering and distributing the vaccine. To collect and track data for the COVID-19 vaccine, states are using their existing immunization information systems (IIS), adopting the CDC’s newly developed Vaccine Administration Management System (VAMS), creating a new system specifically designed for the COVID-19 vaccine, or some combination of these different approaches. The ability to connect IIS with other data sources, like Medicaid claims, and hospitals’ and health systems’ electronic health records (EHRs), is critical to identifying gaps and opportunities for improvement.
In addition to aggregating data across data systems, the completeness of patient records affects efforts to address the gaps. Because Black and LatinX communities have been disproportionately affected by the pandemic, complete patient data, including race and ethnicity data, can help target vaccination outreach to vulnerable communities.
Can you give an overview of the system you use to track and collect COVID-19 specific immunization data?
Minnesota: Minnesota uses an immunization information system (IIS), called Minnesota Immunization Information Connection (MIIC). MIIC-enrolled pharmacists are the only providers who are mandated to collect or upload data into the system. However, all health care staff at provider’s offices can access MIIC, so they have flexibility to input data. And, our health systems are excellent partners that share high quality data to MIIC via the electronic health record (EHR). Additionally, MIIC can capture patients’ vaccine refusal comments and has a reminder/recall function that allows providers to assess which patients are overdue for which vaccine. Minnesota also has data sharing laws that allow schools, childcare providers, purchasers, and community health boards to access MIIC. The ability for schools and childcare providers to review data in the system will be important as children get vaccinated. Our IIS has been working well during the pandemic; we can track and enter data into the system, and it can handle the volume of data without issue. We have been able to onboard new providers expediently and have created a more consolidated process that has been extremely useful.
North Carolina: North Carolina developed the COVID-19 Vaccine Management System (CVMS) instead of using our state’s IIS to create a single end-to-end system for COVID-19 vaccinations. CVMS gives us the ability to add or subtract fields at our discretion. For example, we were able to incorporate North Carolina’s vaccine eligibility determinations and include provider enrollment directly into the system. The system also has the ability to configure to meet provider’s needs, It does not have a reminder/recall system like North Carolina’s Immunization Registry (NCIR). However, CVMS does send proactive e-mail reminders to recipients to get their second doses. CVMS is Version 1 of an iterative software, so all enhancements can be developed in an agile manner.
How are you using data to track areas in need of targeted vaccination approaches?
Minnesota: We have pinpointed a growing list of individuals who have not yet received their second dose. We created this list using our data but have been grappling with how we ensure they actually get vaccinated. One solution we are working to implement is a pilot texting reminder/recall program through a partnership with one of our large health systems and are hoping to make this available more broadly across the state. We have also implemented a change in the data system to allow providers to set parameters to see who in a population needs a vaccine. They will be able to define age parameters and see who in that age bracket has not been vaccinated yet (such as seniors). They can also set product-specific parameters to see who in a county needs a second dose of a specific vaccine and do targeted outreach.
North Carolina: We have geospatial and demographic data for everyone who has received the vaccine, and demographic data and Social Vulnerability Index (SVI) data for all census tracts across the state, so we are able to see where vaccination rates are keeping up with the state average, and which regions are in need of more proactive engagement and partnership.
We are building equity into every aspect of vaccine distribution in order to close the vaccination gap between white populations and Black/African American, Hispanic/Latinx, and American Indian populations in North Carolina, including prioritizing data transparency. We require all vaccine providers to collect and report race and ethnicity data; provide a bi-weekly report to each vaccine provider on their vaccination rates by race and ethnicity; update a public dashboard daily that shows vaccine rates by race and ethnicity at the state and county level, and use this data to inform strategies.
How do COVID-19 immunization data systems interact with EHRs?
Minnesota: MIIC created a unique partnership with the state’s 10 largest EHRs through the EHR consortium. Through partner phone calls, we realized EHR systems collected race/ethnicity and other demographic and comorbidity data while MIIC collected individual patients’ full vaccine history. We partnered to share information across systems to create a full data set. MIIC also gets immunization data directly from the EHRs, which avoids double data entry. And, providers can also query MIIC to get vaccine history and forecast recommendations.
North Carolina: CVMS does not conform with HL7 message structure to exchange immunization information with health systems’ electronic health records and IIS but the platform enables imports of data from EHRs using a standardized file format, which prevents the need for double data entry. We are developing a system that will be able to push the COVID-19 vaccine data into the state IIS, which is critical to having one source of vaccine data for providers, schools, etc. The state IIS is also connected to EHRs, and allows providers that have been onboarded to check for vaccine status through the EHRs.
What are some challenges you have seen in accurately identifying areas of need?
Minnesota: Previously, vaccine supply and inconsistency with delivery had been an external factor that created challenges to accurately identify areas of need, though this is less of an issue now given more consistent supply. In terms of data, because we do not have a mandate to enter data into the IIS, we accept many different types of data, and we have heard from individuals that it has been a barrier to use full-scale EHRs in vaccine clinics because of the technology hurdles.
North Carolina: When supply was more limited, we set aside doses for vaccine providers and events focused on historically marginalized populations (HMP) and relied upon provider data of vaccinating these populations to determine allocation strategy. We track equity gaps – i.e., the difference between HMP vaccination rates and proportion of population – at the provider type, county, and individual provider level, and we share this information back with vaccine providers. We found that equity gaps have steadily declined across geographies as a result of this and other equity-focused vaccination strategies. External barriers like internet access, limited interpretation services, and transportation have also created challenges in ensuring access to vaccinations. We have invested in strategies for people to access information without having to go online – i.e., set up a call center with English- and Spanish-speaking agents who can answer common vaccine questions and help people find vaccine providers near them. We also have had to make it clear up front that identification and insurance are not required, and that data collection relies on self-attestation.
What are some “best data practices” you have found to ensure an equitable distribution of the vaccine?
Minnesota: We use data from MIIC to look at vaccine uptake by SVI. A Federal Emergency Management Agency (FEMA) site was placed in St. Paul which targets zip codes with high SVI. FEMA sites can distribute a small percentage of their allocated vaccine doses off-site, and have utilized some mobile vaccinations for the distribution processes.
North Carolina: We regularly review provider race and ethnicity data internally to evaluate progress and share externally. We promote accountability through data transparency and use of data; we share bi-weekly reports to vaccine providers on their race/ethnicity and publish public dashboards that are updated daily with vaccine rates by race/ethnicity at the state and county levels. We use the data to identify census tracts with high SVI and low vaccination coverage to recruit and allocate to new providers and inform micro-targeting of related resources, such as public communications/media or the support of community health workers. Our data platform is also flexible; it is able to handle new requirements over time.
How have you used federal funding to enhance your data capabilities and ensure full vaccination coverage?
Minnesota: We have a cooperative agreement through the CDC on the business and operational side of the IIS and technical funding comes from the HITECH 90/10 match. We used our previous funding to implement the reminder/recall function, as well as other IIS enhancements, like a COVID-19 assessment report, that will be available soon, improvements to geocoding, implementing COVID-19 vaccine ordering in MIIC, and automating our reporting to the CDC.
North Carolina: We fund CVMS through a variety of funding sources, but primarily through the CARES Act Coronavirus Relief funds. We plan to use American Rescue Plan Act (ARPA) funding to support continued vaccine implementation efforts, including strategies that ensure greater equity and access to the COVID-19 vaccine by those disproportionately affected by COVID-19. The new ARPA funding will also be used to support local communities through local health departments, community-based organizations, and current community vendors to provide mobile vaccination. In addition, we are planning to sponsor vendors to go into neighborhoods to provide vaccine education and administer vaccines to historically marginalized populations that have had challenges accessing vaccines.
What lessons have you learned from the pandemic that you will be able to use to improve vaccination rates (both for COVID-19 and for routine immunizations) moving forward?
Minnesota: There continues to be concern around the gap in childhood immunization rates that has developed as the result of children missing primary care visits, and the MN Immunization program is in the process of determing the best method to help close those gaps. In general, we’ve had new funding conversations that could not have happened without our strong partnerships with health systems and are hoping these partnerships will have built a foundation for immunizations that we can continue past the pandemic.
North Carolina: As we move from very limited supply to increased volume, our approach to using data to achieve vaccine equity is evolving. Moving forward, we are focusing even more intently on census tracts with low vaccination rates and high social vulnerability to determine tailored strategies for identifying providers (including state-sponsored vendors) who can vaccinate in those census tracts, paired with trusted community partners and community health workers to optimally establish mobile or fixed vaccination sites. It can be tricky to balance data sharing and transparency with the critical requirement (and value) of preserving privacy, but it is possible. Overall, our team has learned to be flexible and to openly communicate within the team and with partners.
Acknowledgements: This blog is supported by the Centers for Disease Control and Prevention (CDC) of the US Department of Health and Human Services (HHS) as part of a financial assistance award totaling $250,000 with 100 percent funded by CDC/HHS. The contents are those of the author and do not necessarily represent the official views of, nor an endorsement, by CDC/HHS or the US government. CDC General Terms and Conditions for Non-research Awards, Revised: February 2021.