Import MCOD fixed-width data (restricted or public tier)

Reads a raw NCHS Multiple Cause of Death fixed-width file into a data frame using the byte-verified column dictionary for the given year and tier. The restricted and public files share the within-record layout; the public file suppresses (blanks) sub-state geography from 2005 and the certifier-entered items (certifier, tobacco, pregnancy) from 2003. Those suppressed columns are returned as all-NA so the public output is column-compatible with the restricted output. Data year 2013 is a separate case: certifier and tobacco_use are blank in the restricted 2013 file, which is an upstream NCHS source gap rather than a suppression (2012 and 2014 are populated, and pregnancy_status is populated in 2013). Both fields read as NA from either tier that year, but for different reasons – the public tier blanks them in every year as a standing suppression.

Usage

import_mcod_fwf(file, year, tier = c("restricted", "public"))

Arguments

file: path to the raw MCOD plaintext (or unzipped) fixed-width file
year: year of the MCOD data (integer)
tier: "restricted" (default) or "public"

Value

a tibble with one row per death and columns in the restricted layout order for that year, plus a canonical year column (1979-1995 files also retain their original datayear column)

Examples

if (FALSE) { # \dontrun{
df <- import_mcod_fwf("MULT2020.USAllCnty.txt", 2020, tier = "restricted")
pub <- import_mcod_fwf("mort2020us.dat", 2020, tier = "public")
} # }