Online Program

289679
Improving measures of AI/an mortality and cancer incidence through data linkages


Monday, November 4, 2013 : 8:50 a.m. - 9:10 a.m.

Melissa Jim, MPH, Division of Cancer Prevention and Control, National Center for Chronic Disease Prevention and Health Promotion, CDC, Albuquerque, NE
David Espey, MD, Division of Cancer Prevention and Control/NCCDPHP/CDC, CDC, Albuquerque, NM
Diana Roberts, MPH, Division of Cancer Prevention and Control/NCCDPHP/CDC, CDC, Albuquerque, NM
Donald Haverkamp, MPH, Division of Cancer Prevention and Control/NCCDPHP/CDC, CDC, Albuquerque, NM
Hannah Weir, PhD, Division of Cancer Prevention and Control, National Center for Chronic Disease Prevention and Health Promotion, CDC, Atlanta, GA
Background: Misclassification of American Indians/Alaska Natives (AI/AN) as non-AI/AN has been described in mortality and cancer incidence data resulting in the underestimation of disease burden in this population. Linkages of data from cancer registries in the CDC's National Program of Cancer Registries (NPCR) or the National Cancer Institute's Surveillance and End Results Program (SEER) with Indian Health Service (IHS) registration data have improved estimates of cancer incidence. We conducted linkages of death certificate data with IHS to also improve accuracy of health indicators and to inform program planning/resource allocation.

Methods: Data from the linkage of IHS registration records with all NPCR/SEER cancer registry records (variable years) and linkage with death certificate data from 1990 to 2008 in the National Death Index (NDI) were used to improve race classification.

Results: Cancer registry records revealed 106,033 AI/AN cases over the time period. Matching IHS records with cancer registry records indicated 21,273 additional cases coded by the registry as non-AI/AN to bring the total AI/AN cancer cases to 127,306, a misclassification (underreporting) rate of 17.1%. Similarly NDI records indicated 217,391 AI/AN deaths over the 19-year period. When IHS database was matched against 44,660,888 death records from NDI. Of these, 33,287 did not have AI/AN race recoded on the death certificate to bring the total AI/AN deaths to 250,678; a misclassification rate of 13.3%.

Conclusions: Routine linkages of death records and cancer registry records with IHS data improve data quality and allow more accurate descriptions of mortality patterns and cancer incidence in AI/AN populations.

Learning Areas:

Epidemiology
Program planning

Learning Objectives:
Describe the problem of race mis-classification for AI/AN in 2 key public health databases Discuss methods for addressing race mis-classification for AI/AN in mortality and cancer data.

Keyword(s): Mortality, American Indians

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I am very familiar with the topic and I have been involved in the planning, analysis, interpretation and development of the presentation.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.