R/CH_FEVT_C_ADV.R

In surveyPrev: Mapping the Prevalence of Binary Indicators using Survey Data in Small Areas

##'CH_FEVT_C_ADV Children with fever for whom advice or treatment was sought
##' ch_fev_care in github ml_fev_care should produce the same data
##' KR
#' @param Rdata  data.frame from surveyPrev::getDHSdata
#'
#' @return A partially processed data.frame that will be used in surveyPrev::getDHSindicator. The whole function can be used as a parameter in surveyPrev::getDHSindicator
#'
#' @author Miaolei Bao, Yunhan Wu, Qianyu Dong
#' @examples
#' \dontrun{
#' dhsData <- getDHSdata(country = "Zambia", indicator = "CH_FEVT_C_ADV", year = 2018)
#' }
#'
#' @export
CH_FEVT_C_ADV<- function(Rdata){

  # weight variable
  KRdata <- Rdata %>%
    mutate(wt = v005/1000000)

  # ** ARI indicators ***
  # age of child. If b19 is not available in the data use v008 - b3
  if ("TRUE" %in% (!("b19" %in% names(KRdata))))
    KRdata [[paste("b19")]] <- NA
  if ("TRUE" %in% all(is.na(KRdata$b19)))
  { b19_included <- 0} else { b19_included <- 1}

  if (b19_included==1) {
    KRdata <- KRdata %>%
      mutate(age = b19)
  } else {
    KRdata <- KRdata %>%
      mutate(age = v008 - b3)
  }

  # //ARI symptoms
  # ARI definition differs by survey according to whether h31c is included or not
  if ("TRUE" %in% (!("h31c" %in% names(KRdata))))
    KRdata [[paste("h31c")]] <- NA
  if ("TRUE" %in% all(is.na(KRdata$h31c)))
  { h31c_included <- 0} else { h31c_included <- 1}

  if (h31c_included==1) {
    KRdata <- KRdata %>%
      mutate(ch_ari =
               case_when(
                 h31b==1 & (h31c==1 | h31c==3) & b5==1 ~ 1,
                 b5==1 ~ 0  )) %>%
      set_value_labels(ch_ari = c("Yes" = 1, "No"=0)) %>%
      set_variable_labels(ch_ari = "ARI symptoms in the 2 weeks before the survey")
  } else {
    KRdata <- KRdata %>%
      mutate(ch_ari =
               case_when(
                 h31b==1 & (h31==2) & b5==1 ~ 1 ,
                 b5==1 ~ 0 )) %>%
      set_value_labels(ch_ari = c("Yes" = 1, "No"=0)) %>%
      set_variable_labels(ch_ari = "ARI symptoms in the 2 weeks before the survey")
  }


  # survey specific changes
  # if survey is "IAKR23" or "PHKR31"
  # KRdata <- KRdata %>%
  #   mutate(ch_ari =
  #            case_when(
  #              h31b==1 & (h31==2|h31==1) ~ 1 ,
  #              b5==1 ~ 0 )) %>%
  #   set_value_labels(ch_ari = c("Yes" = 1, "No"=0)) %>%
  #   set_variable_labels(ch_ari = "ARI symptoms in the 2 weeks before the survey")
  #

  # //ARI care-seeking
  # This is country specific and the footnote for the final table needs to be checked to see what sources are included.
  # The code below only excludes traditional practitioner (usually h32t).
  # The variable for traditional healer may be different for different surveys (you can check this by checking all the h32* variables).
  # Some surveys also exclude pharmacies, shop, or other sources.
  # If you want to also remove pharmacy for example as a source of treatment (country specific condition) you can remove
  # h32k from the code below.

  KRdata <- KRdata %>%
    mutate(ch_ari_care =
             case_when(
               (ch_ari==1 &  b5==1) &
                 (h32a == 1 | h32b == 1 | h32c == 1 | h32d == 1 | h32e == 1 | h32f == 1 |
                    h32g == 1 | h32h == 1 | h32i == 1 | h32j == 1 | h32k == 1 | h32l == 1 |
                    h32m == 1 | h32n == 1 | h32o == 1 | h32p == 1 | h32q == 1 | h32r == 1 |
                    h32s == 1 |             h32u == 1 | h32v == 1 | h32w == 1 | h32x == 1 )  ~ 1 ,
               b5==1 & ch_ari==1 ~ 0)) %>%
    set_value_labels(ch_ari_care = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_ari_care = "Advice or treatment sought for ARI symptoms")

  # //ARI care-seeking same or next day
  # some surveys do not have the variable needed to code this indicator which is h46b

  if ("TRUE" %in% (!("h46b" %in% names(KRdata))))
    KRdata [[paste("h46b")]] <- NA
  if ("TRUE" %in% all(is.na(KRdata$h46b)))
  {h46b_included <- 0} else {h46b_included <- 1}

  if (h46b_included==1) {
    KRdata <- KRdata %>%
      mutate(ch_ari_care_day =
               case_when(
                 ch_ari==1 & h46b<2 & b5==1 ~ 1 ,
                 ch_ari==1 & b5==1 ~ 0 )) %>%
      set_value_labels(ch_ari_care_day = c("Yes" = 1, "No"=0)) %>%
      set_variable_labels(ch_ari_care_day = "Advice or treatment sought for ARI symptoms on the same or next day")
  } else {
    KRdata <- KRdata %>%
      mutate(ch_ari_care_day = NA)
  }

  # *** ARI treatment by source ***
  # Two population bases: 1. among children with ARI symptoms, 2. among children with ARI symptoms that sought treatment
  # This is country specific and needs to be checked to produce the specific source of interest.
  # Some sources are coded below and the same logic can be used to code other sources. h32a-z indicates the source.

  # //ARI treatment in government hospital
  KRdata <- KRdata %>%
    mutate(ch_ari_govh =
             case_when(
               ch_ari==1 & h32a==1 & b5==1 ~ 1 ,
               ch_ari==1 & b5==1 ~ 0 )) %>%
    set_value_labels(ch_ari_govh = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_ari_govh = "ARI treatment sought from government hospital among children with ARI")

  KRdata <- KRdata %>%
    mutate(ch_ari_govh_trt =
             case_when(
               ch_ari_care==1 & h32a==1 & b5==1 ~ 1 ,
               ch_ari_care==1 & b5==1 ~ 0 )) %>%
    set_value_labels(ch_ari_govh_trt = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_ari_govh_trt = "ARI treatment sought from government hospital among children with ARI that sought treatment")

  # //ARI treamtment in government health center
  KRdata <- KRdata %>%
    mutate(ch_ari_govcent =
             case_when(
               ch_ari==1 & h32b==1 & b5==1 ~ 1 ,
               ch_ari==1 & b5==1 ~ 0 )) %>%
    set_value_labels(ch_ari_govcent = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_ari_govcent = "ARI treatment sought from government health center among children with ARI")

  KRdata <- KRdata %>%
    mutate(ch_ari_govcent_trt =
             case_when(
               ch_ari_care==1 & h32b==1 & b5==1 ~ 1 ,
               ch_ari_care==1 & b5==1 ~ 0 )) %>%
    set_value_labels(ch_ari_govcent_trt = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_ari_govcent_trt = "ARI treatment sought from government health center among children with ARI that sought treatment")

  # //ARI treatment from a private hospital/clinic
  KRdata <- KRdata %>%
    mutate(ch_ari_pclinc =
             case_when(
               ch_ari==1 & h32j==1 & b5==1 ~ 1 ,
               ch_ari==1 & b5==1 ~ 0 )) %>%
    set_value_labels(ch_ari_pclinc = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_ari_pclinc = "ARI treatment sought from private hospital/clinic among children with ARI")

  KRdata <- KRdata %>%
    mutate(ch_ari_pclinc_trt =
             case_when(
               ch_ari_care==1 & h32j==1 & b5==1 ~ 1 ,
               ch_ari_care==1 & b5==1 ~ 0 )) %>%
    set_value_labels(ch_ari_pclinc_trt = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_ari_pclinc_trt = "ARI treatment sought from private hospital/clinic among children with ARI that sought treatment")

  # //ARI treatment from a private doctor
  KRdata <- KRdata %>%
    mutate(ch_ari_pdoc =
             case_when(
               ch_ari==1 & h32l==1 & b5==1 ~ 1 ,
               ch_ari==1 & b5==1 ~ 0 )) %>%
    set_value_labels(ch_ari_pdoc = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_ari_pdoc = "ARI treatment sought from private doctor  among children with ARI")

  KRdata <- KRdata %>%
    mutate(ch_ari_pdoc_trt =
             case_when(
               ch_ari_care==1 & h32l==1 & b5==1 ~ 1 ,
               ch_ari_care==1 & b5==1 ~ 0 )) %>%
    set_value_labels(ch_ari_pdoc_trt = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_ari_pdoc_trt = "ARI treatment sought from private doctor among children with ARI that sought treatment")

  # //ARI treatment from a pharmacy
  KRdata <- KRdata %>%
    mutate(ch_ari_pharm =
             case_when(
               ch_ari==1 & h32k==1 & b5==1 ~ 1 ,
               ch_ari==1 & b5==1 ~ 0 )) %>%
    set_value_labels(ch_ari_pharm = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_ari_pharm = "ARI treatment sought from a pharmacy among children with ARI")

  KRdata <- KRdata %>%
    mutate(ch_ari_pharm_trt =
             case_when(
               ch_ari_care==1 & h32k==1 & b5==1 ~ 1 ,
               ch_ari_care==1 & b5==1 ~ 0 )) %>%
    set_value_labels(ch_ari_pharm_trt = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_ari_pharm_trt = "ARI treatment sought from a pharmacy among children with ARI that sought treatment")

  # *** Fever indicators ***

  # //Fever
  KRdata <- KRdata %>%
    mutate(ch_fever =
             case_when(
               h22==1 & b5==1 ~ 1,
               b5==1 ~ 0  )) %>%
    set_value_labels(ch_fever = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(ch_fever = "Fever symptoms in the 2 weeks before the survey")

  # //Fever care-seeking
  # This is country specific and the footnote for the final table needs to be checked to see what sources are included.
  # The code below only excludes traditional practitioner (usually h32t).
  # The variable for traditional healer may be different for different surveys (you can check this by checking all the h32* variables).
  # Some surveys also exclude pharmacies, shop, or other sources.
  # If you want to also remove pharmacy for example as a source of treatment (country specific condition) you can remove
  # h32k from the code below.

  KRdata <- KRdata %>%
    mutate(CH_FEVT_C_ADV =
             case_when(
               (ch_fever==1 &  b5==1) &
                 (h32a == 1 | h32b == 1 | h32c == 1 | h32d == 1 | h32e == 1 | h32f == 1 |
                    h32g == 1 | h32h == 1 | h32i == 1 | h32j == 1 | h32k == 1 | h32l == 1 |
                    h32m == 1 | h32n == 1 | h32o == 1 | h32p == 1 | h32q == 1 | h32r == 1 |
                    h32s == 1 |             h32u == 1 | h32v == 1 | h32w == 1 | h32x == 1 )  ~ 1 ,
               b5==1 & ch_fever==1 ~ 0)) %>%
    set_value_labels(CH_FEVT_C_ADV = c("Yes" = 1, "No"=0)) %>%
    set_variable_labels(CH_FEVT_C_ADV = "Advice or treatment sought for fever symptoms")

  colnames(KRdata)[colnames(KRdata) == 'CH_FEVT_C_ADV'] <- 'value'






return(KRdata)
}