Skip to contents

Calculate ACT Treatment Indicators from DHS Data

Source: R/indicator_act.R
calc_act_dhs.Rd

Computes the full set of (World Malaria Report) ACT treatment indicators from DHS Children's Recode (KR) data. Returns survey-weighted proportions with logit confidence intervals in standardized long format.

Usage

calc_act_dhs(
  dhs_kr,
  dhs_kr_raw = NULL,
  survey_vars = list(cluster = "v021", weight = "v005", stratum = "v022", age = "hw1",
    fever = "h22", alive = "b5", act = "ml13e", test = "ml13a"),
  region_var = NULL,
  gps_data = NULL,
  gps_vars = list(cluster = "DHSCLUST", lat = "LATNUM", lon = "LONGNUM"),
  shapefile = NULL,
  admin_level = NULL,
  join_nearest = TRUE,
  dhs_pr = NULL,
  indicators = NULL,
  ci_method = "logit"
)

Arguments

dhs_kr

DHS Children's Recode (KR) dataset (data.frame or tibble). If read via dhs_read(), haven labels may be standardised. Supply dhs_kr_raw for accurate ACT variable detection.

dhs_kr_raw

Optional un-standardised KR dataset (e.g., from arrow::read_parquet()) with original survey-specific haven labels. When provided, ACT and antimalarial variables are detected from its labels, and any extra variables (e.g., ml13aa, ml13da) are copied into the analysis dataset. This is the "two-pass" approach needed when dhs_read() strips country-specific drug names.

survey_vars

Named list mapping DHS variable names. Required keys:

  • cluster: Cluster/PSU ID (default: "v021")

  • weight: Survey weight (default: "v005")

  • stratum: Stratum variable (default: "v022")

  • age: Child's age in months (default: "hw1")

  • fever: Had fever in last 2 weeks (default: "h22")

  • alive: Child alive (default: "b5")

  • act: ACT variable(s) (default: "ml13e"; auto-detected from labels)

  • test: Test-positive filter variable (default: "ml13a")

region_var

Optional column name for subnational grouping (e.g., "v024").

gps_data

Optional DHS GE (Geographic) dataset with cluster coordinates.

gps_vars

Named list for GE variables: cluster, lat, lon.

shapefile

Optional sf object with administrative boundaries.

admin_level

Character vector of admin columns from shapefile.

join_nearest

Logical; if TRUE, assigns unmatched clusters to nearest admin unit. Default: TRUE.

dhs_pr

Optional DHS Person Recode (PR) for febrile RDT indicators.

indicators

Character vector of indicator names to compute. If NULL (default), computes all indicators from act_dictionary().

ci_method

Method for confidence intervals. Default: "logit".

Value

Named list of tibbles, one per admin level:

adm0

National-level estimates (always present)

adm1

Admin-1 estimates (when region_var or shapefile used)

adm2

Admin-2 estimates (when shapefile with adm2 used)

Each tibble contains columns:

  • survey_id: Survey identifier (e.g., "TG2017DHS")

  • iso3: ISO 3166-1 alpha-3 country code (e.g., "TGO")

  • iso2: DHS 2-letter country code (e.g., "TG")

  • survey_type: Survey type ("DHS", "MIS", or "AIS")

  • survey_year: Survey year (integer)

  • adm0: Country name in UPPERCASE (e.g., "TOGO")

  • adm1, adm2: Admin names in UPPERCASE (subnational tabs only)

  • type: Analysis type ("survey_weighted")

  • geo_source: Source of admin names ("survey" or "gps")

  • point: Survey-weighted proportion (0-1 scale)

  • ci_l, ci_u: Lower and upper 95\

  • numerator: Unweighted numerator count (has_act == 1 in filtered subgroup)

  • denominator: Unweighted denominator count (condition-filtered subgroup size)

  • indicator: Indicator name in Title Case (e.g., "Act Antimalarial")

  • indicator_code: Short indicator code (e.g., "act_am")

  • numerator_description: Description of numerator

  • denominator_description: Description of denominator

  • denominator_code: Short code for the denominator subpopulation

Details

Computes up to 12 ACT indicators following DHS methodology. Each indicator measures the proportion of febrile U5 children receiving ACT within a specific subpopulation defined by care-seeking behaviour and antimalarial receipt. See act_dictionary() for the full indicator list.

The function uses three internal helpers:

  • .prepare_act_data() for ACT variable detection and febrile U5 filtering

  • .classify_csb_from_h32() for care-seeking behaviour classification

  • Antimalarial composite built from ml13/h37 drug series

See also

act_dictionary() for indicator definitions, calc_act_mbg() for cluster-level MBG inputs

Examples

if (FALSE) { # \dontrun{
# Two-pass approach (recommended): read standardised + raw
kr <- sntmethods::dhs_read(path, file_type = "KR", ...)
kr_raw <- arrow::read_parquet(parquet_path)
act <- calc_act_dhs(dhs_kr = kr, dhs_kr_raw = kr_raw)

# By region
act <- calc_act_dhs(dhs_kr = kr, dhs_kr_raw = kr_raw, region_var = "v024")

# With GE spatial join
act <- calc_act_dhs(
  dhs_kr = kr, dhs_kr_raw = kr_raw,
  gps_data = ge_data,
  shapefile = admin_sf,
  admin_level = "adm1"
)

# Subset of indicators
act <- calc_act_dhs(
  dhs_kr = kr, dhs_kr_raw = kr_raw,
  indicators = c("ACT_ANTIMALARIAL", "ACT_PUBLIC_ANTIMALARIAL")
)
} # }