From the Beginning: How Data Collection and Use Can Improve Knowledge and Influence Change

Improving American Indian and Alaska Native Cancer and Mortality Surveillance Data

Authors:

Melissa Jim (Presenter)
Centers for Disease Control and Prevention

Donald Haverkamp, Centers for Disease Control and Prevention
Cheyenne Jim, Centers for Disease Control and Prevention
Stephanie Melkonian, Centers for Disease Control and Prevention
David Espey, Centers for Disease Control and Prevention

Public Health Statement: In 2010, an estimated 5.2 million people reporting American Indian/Alaska Native (AI/AN) ancestry alone or in combination with 1 or more races lived in the US, representing approximately 1.7% of the population. These communities have diverse languages, cultures, and histories. The Indian Health Service (IHS) provides primary health care to approximately 2.2 million enrolled members of 567 federally recognized tribes. Misclassification of AI/AN as non-AI/AN in cancer incidence and vital statistics data has resulted in the underestimation of the disease burden in these populations.

Purpose: We present background information on the Indian Health Service, describe our attempt to address racial misclassification through record linkage, and characterize patterns of cancer incidence among AI/AN by IHS region.

Methods/Approach: Linkages of IHS patient registration data and data from central cancer registries that are part of the Centers for Disease Control and Prevention’s National Program of Cancer Registries (NPCR) and the National Cancer Institute’s Surveillance, Epidemiology, and End Results Program (SEER) provided evidence that, when reporting national rates, the regional variations were masking the real burden of disease among AI/AN. Similarly, a linkage of US National Death Index records with IHS patient registration data showed that the disparity in death rates between AI/AN and non-Hispanic white populations in the US remain large for most causes of death.

Results: Cancer registry records from 55 central cancer registries submitted for the IHS Linkage included 118,905 AI/AN cases over the variable time period. The IHS matched against 34,700,335 cancer registry records. Of these, 29,096 did not have AI/AN race recorded by the cancer registry but linked to IHS to bring the total AI/AN cancer cases to 148,001. This resulted in an average misclassification (underreporting) rate of 19.7%, with a range of 0.2% to 51.6%.

Conclusions/Implications: The results from the IHS linkages have been used to provide the most accurate data available in numerous peer-review and tribal publications. Tribal linkages to further address race misclassification will also be presented.

Opening the Black Box: Secrets to EHR-Reported Cancer Screening Rates

Author:

Christen Lara (Presenter)
Colorado Department of Public Health and Environment

Public Health Statement: Technological innovations through EHR reporting enhancements.

Purpose: Since 2012, Colorado’s breast, cervical and colorectal cancer control programs have engaged the state health department’s informatics program to build capacity within health care organizations to report cancer screening rates using electronic health record data. Through this work, Colorado has encountered underlying reporting accuracy problems, challenges clinics face finding solutions, and issues around navigating the role of state health departments and other agencies in finding solutions.

Methods/Approach: Colorado collects chart review and EHR-reported rates to assess reporting capacity and accuracy among partner health care organizations. Health department informatics staff provide feedback on the accuracy of EHR-reported rates, empowering organizations to navigate internal or vendor IT structure to correct reporting inaccuracies.

Results: Colorado engaged health care organizations representing about 10% of Colorado’s eligible screening population. Many organizations lack capacity to validate their EHR rates and routinely submit inaccurate rates. Among organizations that have partnered with Colorado for three years, significant improvements in EHR reporting capacity are seen within the first year, with most organizations sustaining improvement over the following year. Underlying issues are often technical in nature, requiring an external organization to provide technical assistance to obtain accurate results.

Conclusions/Implications: It is important for statewide partners to engage in conversations about EHR enhancements from a place of technical expertise. This can lead to improvements in use of reports for day-to-day clinical operations and confidence that changes in cancer screening rates represent accurate population-level improvements.

What’s In It for Me? Utilizing Technology to Enhance the Wisconsin Cancer Council’s Value-Added Collaboration

Authors:

Sarah Kerch (Presenter)
Wisconsin Comprehensive Cancer Control Program

Courtney Harris, Wisconsin Comprehensive Cancer Control Program
Amy Conlon, Wisconsin Comprehensive Cancer Control Program
Noelle LoConte, University of Wisconsin Carbone Cancer Center

Public Health Statement: The Wisconsin Cancer Council (WCC) serves as the state’s CCC coalition. From 2006–2015, WCC membership grew from 40 to 119 organizations. Despite this growth, member engagement declined. Members were no longer able to see what was in it for them. Value-added Collaboration, one of the Nine Habits of Successful CCC Coalitions, needed to be enhanced.

Purpose: To enhance value-added collaboration, the WCC revised its membership application into an online form. The online form requires members to input their cancer plan implementation efforts which then generates an online, members-only directory.

Methods/Approach: The WCC Leadership Team, while addressing membership benefit needs identified by a 2015 member assessment, recommended transforming the member application into an online commitment form, collecting several data points, including: member representatives’ role; cancer plan implementation activities; populations served; and resources needed. Data then generates a searchable online member directory to encourage member collaboration. Commitment forms and the online directory were unveiled at October 2016 Regional Meetings.

Results: Regional Meeting evaluation and discussion results indicated the online member directory as the most useful new member benefit. As of February 2017, 79 organizations have completed the commitment form and are being added to the directory.

Conclusions/Implications: Data from the commitment forms will be used to conduct a gap analysis in summer 2017 and drive the development of the 2018–2019 WCC priority work plan to maximize member efforts in specific Wisconsin CCC Plan areas.

Online Mapping for Michigan’s Cancer Programs

Author:

Michael Carr (Presenter)
State of Michigan

Public Health Statement: State-managed cancer programs can map their own data to assist in locating hard to reach clients.

Purpose: The importance of client outreach in the Michigan Breast and Cervical Cancer Control Navigation Program (BCCCNP) has changed. From 2008 to 2013, an abundance of low-income, uninsured women sought screening services to the point that outreach was not necessary. After Medicaid expansion in the form of the Healthy Michigan Plan (HMP) began in April 2014, the number of program eligible women dropped dramatically. As a result, targeted client outreach to those not served by the HMP (women between 139% and 250% FPL) is now critical for the program’s survival.

Methods/Approach: In 2005, the Michigan BCCCNP began mapping client addresses and sub-county areas based on poverty and population estimates. After the state developed a template for online mapping applications, the BCCCNP collaborated with the pertinent department to create a custom mapping application called Cancer Mapper. Development took two years.

Results: Cancer Mapper went live in 2014. It has all of the tools associated with popular mapping applications, such as Google maps. It has custom search and report features. Custom layers show categories of race, ethnicity, poverty, cancer diagnosis information, and estimated eligible counts of women. Because Cancer Mapper shows protected health information, access is limited.

Conclusions/Implications: In 2014, local health departments used Cancer Mapper to develop targeted outreach strategies. As of March 2017, Cancer Mapper has new enhancements. Cancer Mapper could serve as a template application for other grant-funded programs.