layout.fixed.naaccr.doc.naaccr16.countyDxGeocode2010.html Maven / Gradle / Ivy
                 Go to download
                
        
                    Show more of this group  Show more artifacts with this name
Show all versions of layout Show documentation
                Show all versions of layout Show documentation
Framework that allows defining file formats (layouts) and use them to read and write data files.
                
            
    
        Alternate Name 
        Item # 
        Length 
        Source of Standard 
        Implemented Year 
        Implemented Version 
        Retired Year 
        Retired Version 
        Column # 
     
    
         
        96 
        3 
        NAACCR 
        2016 
        16 
         
         
        470 - 472 
     
Description
Code for the county of the patient’s residence at the time the tumor was diagnosed is a derived (geocoded) variable based on Census Boundary files from 2010 Decennial Census.
    This code should be used for county and county-based (such as CHSDA) rates and analysis for all cases diagnosed in 2010-2019.
Rationale
County of diagnosis is essential for investigating epidemiologic pattern at both the county and the census tract level. Census tracts are nested within counties and designated with a 6-digit number code. A given census tract code is commonly repeated within a state in different counties, making census tract numbers unique only when paired with the state and the county. Example from Massachusetts: Rural Franklin County contains a tract 040600 with 2010 population 4,612 people, and urban Suffolk County contains a tract 040600 with 2,444 people.
County borders occasionally change over time. For instance, part of Adams County in Colorado become a new county, Broomfield County, in 2001. Because the historic NAACCR county code, County at DX [90], is not associated with any particular date, census tract codes can be paired with the wrong county if these changes occur. Thus, county-level rates may be generated using temporally incorrect county populations. Further, this county code is derived (through geocoding) and, therefore, generally more accurate than those provided manually. Finally, some variables, like the Tract-Level Poverty Indicator (Census Tr Poverty Indictr [145]), requires census information closest to year of diagnosis and not the decade of diagnosis. Having separate geocoded county codes for each decennial census, as we do for census tract, enables registries to retain the original, reported county as needed but will ensure the correct pairing by county and tract occur.
Recording a county at diagnosis that reflects the relevant date (decade) and relies on geocoded data will improve the accuracy of county and census tract assignments and of links with geographic data (i.e., population, poverty category, urban/rural designation).
Codes
    
        
            001-997 
            County at diagnosis. Valid FIPS code. 
         
        
            998 
            Outside state/county code unknown. Known town, city, state, or country of residence but county code not known AND a resident outside of the state of reporting institution (must meet all criteria). 
         
        
            999 
            County unknown. The county of the patient is unknown, or the patient is not a United States resident. County is not documented in the patient's medical record. 
         
        
             
         
    
    Note: For U.S. residents, historically, standard codes are those of the FIPS publication “Counties and Equivalent Entities of the United States, Its Possessions, and Associated Areas.” These FIPS codes (FIPS 6-4) have been replaced by INCITS standard codes, however, there is no impact on this variable as the codes align with the system the Census used for each decennial census and will automatically be accounted for during geocoding.
    © 2015 - 2025 Weber Informatics LLC | Privacy Policy