microdatasus: Download and preprocess DataSUS files

Documented in process_sinan_malaria

#' Process SINAN Malaria variables from DataSUS
#'
#' \code{process_sinan_malaria} processes SINAN Malaria variables retrieved by \code{fetch_datasus()}.
#'
#' This function processes SINAN Malaria variables retrieved by \code{fetch_datasus()}, informing labels for categoric variables including NA values.
#'
#' @param data \code{data.frame} created by \code{fetch_datasus()}.
#' @param municipality_data optional logical. \code{TRUE} by default, creates new variables in the dataset informing the full name and other details about the municipality of residence.
#'
#' @examples \dontrun{
#' df <- fetch_datasus(year_start = 2016, year_end = 2016,
#' uf = "RJ", information_system = "SINAN-MALARIA-FINAL")
#' df_a <- process_sinan_malaria(df)
#' df_b <- process_sinan_malaria(df, municipality_data = FALSE)
#' }
#' @export

process_sinan_malaria <- function(data, municipality_data = TRUE){
  # Variables names
  variables_names <- names(data)

  # Use dtplyr
  data <- dtplyr::lazy_dt(data)

  # TP_NOT
  if ("TP_NOT" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(TP_NOT = dplyr::case_match(
        .data$TP_NOT,
        "1" ~ "Negativa",
        "2" ~ "Individual",
        "3" ~ "Surto",
        "4" ~ "Agregado",
        .default = .data$TP_NOT
      )) %>%
      dplyr::mutate(TP_NOT = as.factor(.data$TP_NOT))
  }

  # DT_NOTIFIC
  if ("DT_NOTIFIC" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(DT_NOTIFIC = as.Date(.data$DT_NOTIFIC))
  }

  # SG_UF_NOT
  if ("SG_UF_NOT" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(SG_UF_NOT = dplyr::case_match(
        .data$SG_UF_NOT,
        "0" ~ "Ignorado",
        "99" ~ "Ignorado",
        "11" ~ "Rond\u00f4nia",
        "12" ~ "Acre",
        "13" ~ "Amazonas",
        "14" ~ "Roraima",
        "15" ~ "Par\u00e1",
        "16" ~ "Amap\u00e1",
        "17" ~ "Tocantins",
        "21" ~ "Maranh\u00e3o",
        "22" ~ "Piau\u00ed",
        "23" ~ "Cear\u00e1",
        "24" ~ "Rio Grande do Norte",
        "25" ~ "Para\u00edba",
        "26" ~ "Pernambuco",
        "27" ~ "Alagoas",
        "28" ~ "Sergipe",
        "29" ~ "Bahia",
        "31" ~ "Minas Gerais",
        "32" ~ "Esp\u00edrito Santo",
        "33" ~ "Rio de Janeiro",
        "35" ~ "S\u00e3o Paulo",
        "41" ~ "Paran\u00e1",
        "42" ~ "Santa Catarina",
        "43" ~ "Rio Grande do Sul",
        "50" ~ "Mato Grosso do Sul",
        "51" ~ "Mato Grosso",
        "52" ~ "Goi\u00e1s",
        "53" ~ "Distrito Federal",
        .default = .data$SG_UF_NOT
      )) %>%
      dplyr::mutate(SG_UF_NOT = as.factor(.data$SG_UF_NOT))
  }

  # IDADE
  if ("NU_IDADE_N" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(NU_IDADE_N = dplyr::case_match(.data$NU_IDADE_N,
                                                   999 ~ NA,
                                                   .default = .data$NU_IDADE_N)) %>%
      # Codigo e valor
      dplyr::mutate(idade_cod = substr(.data$NU_IDADE_N, 1, 1),
                    idade_value = as.numeric(substr(.data$NU_IDADE_N, 2, 3)),) %>%
      dplyr::mutate(IDADEminutos = dplyr::case_match(.data$idade_cod,
                                                     "0" ~ idade_value,
                                                     .default = NA)) %>%
      dplyr::mutate(IDADEhoras = dplyr::case_match(.data$idade_cod,
                                                   "1" ~ idade_value,
                                                   .default = NA)) %>%
      dplyr::mutate(IDADEdias = dplyr::case_match(.data$idade_cod,
                                                  "2" ~ idade_value,
                                                  .default = NA)) %>%
      dplyr::mutate(IDADEmeses = dplyr::case_match(.data$idade_cod,
                                                   "3" ~ idade_value,
                                                   .default = NA)) %>%
      dplyr::mutate(
        IDADEanos = dplyr::case_match(
          .data$idade_cod,
          "4" ~ idade_value,
          "5" ~ idade_value + 100,
          .default = NA
        )
      ) %>%
      dplyr::select(-"idade_cod", -"idade_value")
  }

  # CS_SEXO
  if ("CS_SEXO" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(CS_SEXO = dplyr::case_match(
        .data$CS_SEXO,
        "M" ~ "Masculino",
        "F" ~ "Feminino",
        "I" ~ "Ignorado",
        .default = .data$CS_SEXO
      )) %>%
      dplyr::mutate(CS_SEXO = as.factor(.data$CS_SEXO))
  }

  # CS_GESTANT
  if ("CS_GESTANT" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(CS_GESTANT = dplyr::case_match(
        .data$CS_GESTANT,
        "1" ~ "1o trimestre",
        "2" ~ "2o trimestre",
        "3" ~ "3o trimestre",
        "4" ~ "Idade gestacional ignorada",
        "5" ~ "N\u00e3o",
        "6" ~ "N\u00e3o se aplica",
        "9" ~ "Ignorado",
        .default = .data$CS_GESTANT
      )) %>%
      dplyr::mutate(CS_GESTANT = as.factor(.data$CS_GESTANT))
  }

  # CS_RACA
  if ("CS_RACA" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(CS_RACA = dplyr::case_match(
        .data$CS_RACA,
        "1" ~ "Branca",
        "2" ~ "Preta",
        "3" ~ "Amarela",
        "4" ~ "Parda",
        "5" ~ "Ind\u00edgena",
        "9" ~ "Ignorado",
        .default = .data$CS_RACA
      )) %>%
      dplyr::mutate(CS_RACA = as.factor(.data$CS_RACA))
  }

  # CS_ESCOL_N
  if ("CS_ESCOL_N" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(CS_ESCOL_N = dplyr::case_match(
        .data$CS_ESCOL_N,
        "1" ~ "1a a 4a s\u00e9rie incompleta do EF",
        "2" ~ "4a s\u00e9rie completa do EF (antigo 1o grau)",
        "3" ~ "5a \u00e0 8a s\u00e9rie incompleta do EF (antigo gin\u00e1sio ou 1o grau)",
        "4" ~ "Ensino fundamental completo (antigo gin\u00e1sio ou 1o grau)",
        "5" ~ "Ensino m\u00e9dio incompleto (antigo colegial ou 2o grau)",
        "6" ~ "Ensino m\u00e9dio completo (antigo colegial ou 2o grau)",
        "7" ~ "Educa\u00e7\u00e3o superior incompleta",
        "8" ~ "Educa\u00e7\u00e3o superior completa",
        "9" ~ "Ignorado",
        "10" ~  "N\u00e3o se aplica",
        .default = .data$CS_ESCOL_N
      )) %>%
      dplyr::mutate(CS_ESCOL_N <- as.factor(.data$CS_ESCOL_N))
  }

  # SG_UF
  if ("SG_UF" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(SG_UF = dplyr::case_match(
        .data$SG_UF,
        "0" ~ "Ignorado",
        "99" ~ "Ignorado",
        "11" ~ "Rond\u00f4nia",
        "12" ~ "Acre",
        "13" ~ "Amazonas",
        "14" ~ "Roraima",
        "15" ~ "Par\u00e1",
        "16" ~ "Amap\u00e1",
        "17" ~ "Tocantins",
        "21" ~ "Maranh\u00e3o",
        "22" ~ "Piau\u00ed",
        "23" ~ "Cear\u00e1",
        "24" ~ "Rio Grande do Norte",
        "25" ~ "Para\u00edba",
        "26" ~ "Pernambuco",
        "27" ~ "Alagoas",
        "28" ~ "Sergipe",
        "29" ~ "Bahia",
        "31" ~ "Minas Gerais",
        "32" ~ "Esp\u00edrito Santo",
        "33" ~ "Rio de Janeiro",
        "35" ~ "S\u00e3o Paulo",
        "41" ~ "Paran\u00e1",
        "42" ~ "Santa Catarina",
        "43" ~ "Rio Grande do Sul",
        "50" ~ "Mato Grosso do Sul",
        "51" ~ "Mato Grosso",
        "52" ~ "Goi\u00e1s",
        "53" ~ "Distrito Federal",
        .default = .data$SG_UF
      )) %>%
      dplyr::mutate(SG_UF = as.factor(.data$SG_UF))
  }

  # ID_PAIS
  if ("ID_PAIS" %in% variables_names) {
    data$ID_PAIS <- dplyr::left_join(data, microdatasus::paisnet, by = c("ID_PAIS" = "ID_PAIS"))$NM_PAIS
  }

  # ID_OCUPA_N
  if ("ID_OCUPA_N" %in% variables_names) {
    data$ID_OCUPA_N <- factor(dplyr::left_join(data, microdatasus::tabCBO, by = c("ID_OCUPA_N" = "cod"))$nome)
  }

  # CLASSI_FIN
  if ("CLASSI_FIN" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(CLASSI_FIN = dplyr::case_match(
        .data$CLASSI_FIN,
        "1" ~ "Confirmado",
        "2" ~ "Descartado",
        .default = .data$CLASSI_FIN
      )) %>%
      dplyr::mutate(CLASSI_FIN = as.factor(.data$CLASSI_FIN))
  }

  # AT_ATIVIDA
  if ("AT_ATIVIDA" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(AT_ATIVIDA = dplyr::case_match(
        .data$AT_ATIVIDA,
        "1" ~ "Agricultura",
        "2" ~ "Pecu\u00e1ria",
        "3" ~ "Dom\u00e9stica",
        "4" ~ "Turismo",
        "5" ~ "Garimpagem",
        "6" ~ "Explora\u00e7\u00e3o vegetal",
        "7" ~ "Ca\u00e7a/Pesca",
        "8" ~ "Construtor de estradas/barragens",
        "9" ~ "Minera\u00e7\u00e3o",
        "10" ~ "Viajante",
        "11" ~ "Outros",
        "99" ~ "Ignorado",
        .default = .data$AT_ATIVIDA
      )) %>%
      dplyr::mutate(AT_ATIVIDA = as.factor(.data$AT_ATIVIDA))
  }

  # AT_LAMINA
  if ("AT_LAMINA" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(AT_ATIVIDA = dplyr::case_match(
        .data$AT_LAMINA,
        "1" ~ "BP",
        "2" ~ "BA",
        "3" ~ "LVC",
        .default = .data$AT_LAMINA
      )) %>%
      dplyr::mutate(AT_LAMINA = as.factor(.data$AT_LAMINA))
  }

  # AT_SINTOMA
  if ("AT_SINTOMA" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(AT_SINTOMA = dplyr::case_match(
        .data$AT_SINTOMA,
        "1" ~ "Com sintomas",
        "2" ~ "Sem sintomas",
        .default = .data$AT_SINTOMA
      )) %>%
      dplyr::mutate(AT_SINTOMA = as.factor(.data$AT_SINTOMA))
  }

  # TPAUTOCTO
  if ("TPAUTOCTO" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(TPAUTOCTO = dplyr::case_match(
        .data$TPAUTOCTO,
        "1" ~ "Sim",
        "2" ~ "N\u00e3o",
        "3" ~ "Indeterminado",
        .default = .data$TPAUTOCTO
      )) %>%
      dplyr::mutate(TPAUTOCTO = as.factor(.data$TPAUTOCTO))
  }

  # COUFINF
  if ("COUFINF" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(COUFINF = dplyr::case_match(
        .data$COUFINF,
        "0" ~ "Ignorado",
        "99" ~ "Ignorado",
        "11" ~ "Rond\u00f4nia",
        "12" ~ "Acre",
        "13" ~ "Amazonas",
        "14" ~ "Roraima",
        "15" ~ "Par\u00e1",
        "16" ~ "Amap\u00e1",
        "17" ~ "Tocantis",
        "21" ~ "Maranh\u00e3o",
        "22" ~ "Piau\u00ed",
        "23" ~ "Cear\u00e1",
        "24" ~ "Rio Grande do Norte",
        "25" ~ "Para\u00edba",
        "26" ~ "Pernambuco",
        "27" ~ "Alagoas",
        "28" ~ "Sergipe",
        "29" ~ "Bahia",
        "31" ~ "Minas Gerais",
        "32" ~ "Esp\u00edrito Santo",
        "33" ~ "Rio de Janeiro",
        "35" ~ "S\u00e3o Paulo",
        "41" ~ "Paran\u00e1",
        "42" ~ "Santa Catarina",
        "43" ~ "Rio Grande do Sul",
        "50" ~ "Mato Grosso do Sul",
        "51" ~ "Mato Grosso",
        "52" ~ "Goi\u00e1s",
        "53" ~ "Distrito Federal",
        .default = .data$COUFINF
      )) %>%
      dplyr::mutate(COUFINF = as.factor(.data$COUFINF))
  }

  # COPAISINF
  if ("COPAISINF" %in% variables_names) {
    data$COPAISINF <- dplyr::left_join(data, microdatasus::paisnet, by = c("COPAISINF" = "COPAISINF"))$NM_PAIS
  }

  # RESULT
  if ("RESULT" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(RESULT = dplyr::case_match(
        .data$RESULT,
        "1" ~ "Negativo",
        "2" ~ "F",
        "3" ~ "F + FG",
        "4" ~ "V",
        "5" ~ "F + V",
        "6" ~ "V + FG",
        "7" ~ "FG",
        "8" ~ "M",
        "9" ~ "F + M",
        "10" ~ "O",
        .default = .data$RESULT
      )) %>%
      dplyr::mutate(RESULT = as.factor(.data$RESULT))
  }

  # PCRUZ
  if ("PCRUZ" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(PCRUZ = dplyr::case_match(
        .data$PCRUZ,
        "1" ~ "Menor que meia cruz",
        "2" ~ "Meia cruz",
        "3" ~ "Uma cruz",
        "4" ~ "Duas cruzes",
        "5" ~ "Tr\u00eas cruzes",
        "6" ~ "Quatro cruzes",
        .default = .data$PCRUZ
      )) %>%
      dplyr::mutate(PCRUZ = as.factor(.data$PCRUZ))
  }

  # TRA_ESQUEM
  if ("TRA_ESQUEM" %in% variables_names) {
    data <- data %>%
      dplyr::mutate(TRA_ESQUEM = dplyr::case_match(
        .data$TRA_ESQUEM,
        "1" ~ "Infec\u00e7\u00f5es por Pv com Cloroquina em 3 dias e Primaquina em 7 dias",
        "2" ~ "Infec\u00e7\u00f5es por Pf com Quinina em 3 dias + Doxiciclina em 5 dias + primaquina no 6o dia",
        "3" ~ "Infec\u00e7\u00f5es mistas por Pv + Pf com Mefloquina em dose \u00fanica e primaquina em 7 dias",
        "4" ~ "Infec\u00e7\u00f5es por Pm com cloroquina em 3 dias",
        "5" ~ "Infec\u00e7\u00f5es por Pv em crian\u00e7as apresentando v\u00f4mitos, com c\u00e1psulas retais de artesunato em 4 dias e Primaquina em 7 dias",
        "6" ~ "Infec\u00e7\u00f5es por Pf com Mefloquina em dose \u00fanica e primaquina no segundo dia",
        "7" ~ "Infec\u00e7\u00f5es por Pf com Quinina em 7 dias",
        "8" ~ "Infec\u00e7\u00f5es por Pf de crian\u00e7as com c\u00e1psulas retais de artesunato em 4 dias e dose \u00fanica de Mefloquina no 3o dia e Primaquina no 5o dia",
        "9" ~ "Infec\u00e7\u00f5es mistas por Pv + Pf com Quinina em 3 dias, doxiciclina em 5 dias e Primaquina em 7 dias",
        "10" ~ "Preven\u00e7\u00e3o de reca\u00edda da mal\u00e1ria por Pv com Cloroquina em dose \u00fanica semanal durante 3 meses",
        "11" ~ "Mal\u00e1ria grave e complicada",
        "99" ~ "Outro esquema utilizado (por m\u00e9dico)",
        .default = .data$TRA_ESQUEM
      )) %>%
      dplyr::mutate(TRA_ESQUEM = as.factor(.data$TRA_ESQUEM))
  }

  # From data.table to tibble
  data <- tibble::as_tibble(data)

  # Purge levels
  data <- droplevels(data)

  # Unescape unicode characters
  data <- suppressWarnings(tibble::as_tibble(lapply(X = data, FUN = stringi::stri_unescape_unicode)))

  # Return
  return(data)
}