Remap the raw NCHS detail-age field to age in completed years

The NCHS "detail age" field (`age`) is unit-coded, and its encoding changed at data year 2003: 1979-2002 use a 3-digit code, 2003 onward a 4-digit code. In both eras a leading digit gives the unit (years, months, weeks, days, hours, minutes, or "not stated"). This dispatches on the data year and returns age in completed years in a new `age_years` column: sub-year ages (months/weeks/days/ hours/minutes) collapse to `0`, and not-stated ages become `NA`.

Usage

remap_age(df, year = NULL)

Arguments

df

an MCOD data frame for a single data year, with the raw `age` column

year

data year; if `NULL`, extracted from the data frame

Value

`df` with an added numeric `age_years` column

Details

The era boundary is 2003 (the death-certificate revision), NOT the ICD-9/ ICD-10 boundary at 1999. `age_years` is single completed years, so it is not directly comparable to the pre-binned `ager27` recode consumed by [convert_ager27()] / [categorize_age_5()]; to bin it into 5-year groups use, e.g., `pmin(floor(age_years / 5) * 5, 85)`.

Examples

df <- data.frame(year = 2019, age = c(1037, 2006, 1999, 9999))
remap_age(df)$age_years          # 37 (years), 0 (months), NA, NA
#> [1] 37  0 NA NA