data-raw/global_summary.R

In pepfar-datim/daa-analytics: Utilities for Compiling and Analyzing PEPFAR Data Alignment Activity Data

# Code to prepare the global summary spreadsheet goes here
library(magrittr)
library(dplyr)

if(!exists("combined_data")){ combined_data <- readRDS("support_files/combined_data.rds") } #nolint
if(!exists("import_history")){ import_history <- readRDS("support_files/import_history.rds") } #nolint

output_folder <- Sys.getenv("OUTPUT_FOLDER")

process_data <- function(combined_data, import_history, output_folder) {
  summary_data <- daa.analytics::global_summary(combined_data)

  result_data <- summary_data %>%
    group_by(period, indicator) %>%
    mutate(
      total_pepfar_summation = sum(PEPFAR_Results_FacilitiesReportedByBoth, na.rm = TRUE),
      weight = PEPFAR_Results_FacilitiesReportedByBoth / total_pepfar_summation,
      ou_contribution_to_global = Concordance * weight
    ) %>%
    ungroup() %>%
    group_by(period, indicator) %>%
    mutate(
      summation_global_concordance = sum(ou_contribution_to_global, na.rm = TRUE),
      Global_Concordance_Percent = round(summation_global_concordance * 100, 2)
    ) %>%
    ungroup()
  result_data <- result_data %>% select(-total_pepfar_summation)

  summary_data <- result_data |>
    dplyr::filter(!is.na(OU)) |>
    dplyr::left_join(dplyr::select(import_history,
                                   OU,
                                   period,
                                   indicator,
                                   CourseOrFine = has_disag_mapping,
                                   DataOrMapping = has_mapping_result_data),
                     by = c("OU", "period", "indicator")) |>

    dplyr::mutate(DataOrMapping = ifelse((is.na(MOH_Results_Total) | MOH_Results_Total == "None") & CourseOrFine == "Coarse" & period < 2022, "Mapping Coarse",
                                         ifelse((is.na(MOH_Results_Total) | MOH_Results_Total == "None") & CourseOrFine == "Fine" & period < 2022, "Mapping Fine",
                                                ifelse((is.na(CourseOrFine) | CourseOrFine == "None" | CourseOrFine == "NA") & period < 2022, "No Mapping",
                                                       ifelse(!is.na(MOH_Results_Total) & !is.na(PEPFAR_Results_Total) & CourseOrFine == "Fine" & period < 2022, "Data Fine",
                                                              ifelse(!is.na(MOH_Results_Total) & CourseOrFine == "Fine" & period < 2022, "Data Fine",
                                                                     ifelse(!is.na(MOH_Results_Total) & CourseOrFine == "Coarse" & period < 2022, "Data Coarse",
                                                              ifelse(!is.na(MOH_Results_Total) & !is.na(PEPFAR_Results_Total) & CourseOrFine == "Coarse" & period < 2022, "Data Coarse", DataOrMapping))))))))
  summary_data <- dplyr::select(summary_data, -CourseOrFine)

  write.csv(summary_data, paste0(output_folder, "global_summary.csv"))
}
process_data(combined_data, import_history, output_folder)