R/smk_status.R

In STAPM/hseclean: Health Survey Data Wrangling

#' Cigarette smoking status
#'
#' Categorise cigarette smoking into current, former and never regular cigarette smoker.
#'
#' Note that if some smokes either regularly or occasionally, then they are classified as a current regular
#'  cigarette smoker. People who used to smoke regularly or occasionally are classified as former smokers, but
#'  people who have only tried a cigarette once or twice are classified as never smokers.
#'
#'  This variable is created for adults aged >= 16 years, and children aged 8-15 years.
#'
#'  The data is processed from the Health Survey for England and the Scottish Health Survey.
#'  The Scottish Health Survey only has data on people >= 16 years old.
#'
#' @param data Data table - the health survey dataset.
#' @importFrom data.table :=
#' @return Returns two smoking variables:
#' \itemize{
#' \item cig_smoker_status - current, former, never
#' \item cig_ever - ever_smoker, never_smoker, based on current and former smokers from the variables above.
#' }
#' @export
#'
#' @examples
#'
#' \dontrun{
#'
#' data_2016 <- read_2016()
#'
#' data_2016 <- smk_status(data = data_2016)
#'
#' }
#'
#'
smk_status <- function(
    data
) {

  country <- unique(data[ , country][1])
  year <- unique(data[ , year][1])

  ######################################################################
  # Regular cigarette smoking status (age >= 16)

  if (country %in% c("Wales")){

    if(year > 2015) {

      # NSW
      data[smoke == 5, cig_smoker_status := "never"]
      data[smoke %in% 3:4, cig_smoker_status := "former"]
      data[smoke %in% 1:2, cig_smoker_status := "current"]

    }

    if(year >= 2009 & year <= 2015) {

      # WHS
      # smokec - Currently smoke either daily or occasionally - use this

      data[smokec == 1, cig_smoker_status := "current"]
      data[smokex == 1, cig_smoker_status := "former"]
      data[smok == 5, cig_smoker_status := "never"]

      data[is.na(cig_smoker_status) & smokstat == 1, cig_smoker_status := "current"]
      data[is.na(cig_smoker_status) & smokstat == 2, cig_smoker_status := "former"]
      data[is.na(cig_smoker_status) & smokstat == 3, cig_smoker_status := "never"]

      data[ , `:=`(smokee = NULL, smokec = NULL, smokex = NULL, smok = NULL, smoked = NULL, smokstat = NULL)]


    }
  }

  if (country %in% c("England", "Scotland")){

    ##
    # Never smokers (age >= 16)

    # If never smoked, or never smoked cigarettes
    data[smkevr == 2 | cigevr == 2, cig_smoker_status := "never"]

    # If have ever smoked, but only tried once or twice
    data[cigreg == 3, cig_smoker_status := "never"]

    ##
    # Former smokers (age >= 16)

    # If used to smoke regularly or occasionally
    data[cigreg %in% 1:2, cig_smoker_status := "former"]

    ##
    # Current smokers (age >= 16)

    # If currently smoke cigarettes
    data[cignow == 1, cig_smoker_status := "current"]


    ######################################################################
    # Regular cigarette smoking status (age 8-15)

    ##
    # Never smokers (age 8-15)
    if(country == "England"){

      # If never smoked cigarettes
      data[kcigevr == 2, cig_smoker_status := "never"]

      # If have ever smoked, but only tried once or twice
      data[kcigreg %in% 1:2, cig_smoker_status := "never"]

      ##
      # Former smokers (age 8-15)

      # If used to smoke regularly or occasionally
      data[kcigreg == 3, cig_smoker_status := "former"]

      ##
      # Current smokers (age 8-15)

      # If currently smoke cigarettes
      data[kcigreg %in% 4:6 | kcigweek == 1 | kcignum > 0, cig_smoker_status := "current"]


      # If less than age 8, assume never smoker
      data[age < 8, cig_smoker_status := "never"]

      data[ , kcigevr := NULL]
    }

    # no data on under 16s in Scotland
    if(country == "Scotland"){

      data[age < 16, cig_smoker_status := NA_real_]

    }

    # for England - the above over estimates former smoking and under estimates never smoking
    # compared to the HSE's reporting tables, which specifically define former smoking
    # as former *regular* cigarette smokers
    # to align with these tables use the "cigsta3" derived variable where present

    # review data processing for Scotland and Wales to be as consistent as possible with regular smoking

    if(country == "England" & "cigsta3" %in% names(data)) {
      data[cigsta3 == 1, cig_smoker_status := "current"]
      data[cigsta3 == 2, cig_smoker_status := "former"]
      data[cigsta3 == 3, cig_smoker_status := "never"]
    }

    # Remove variables no longer needed
    remove_vars <- c("smkevr", "cigevr", "cigreg", "cignow", colnames(data)[stringr::str_detect(colnames(data), "cigst")])
    data[ , (remove_vars) := NULL]
  }

  ######################################################################
  # Ever regular cigarette smoker

  # derive the ever-smoking variable to be consistent with the above
  data[cig_smoker_status %in% c("current", "former"), cig_ever := "ever_smoker"]
  data[cig_smoker_status == "never", cig_ever := "never_smoker"]


  return(data[])
}