R/generate_fu_logs.R
In Thaliehln/timci: Data management and visualisation for the TIMCI project
Defines functions
generate_ltfu_log
generate_physical_fu_log_csv
generate_fu_log_csv
generate_fu_log
generate_fu_logs
Documented in
generate_fu_log
generate_fu_log_csv
generate_fu_logs
generate_ltfu_log
generate_physical_fu_log_csv
#' General follow-up logs
#'
#' This function runs follow-up Rmarkdown files to generate standardised follow-up logs for the Tools for Integrated Management of Childhood Illnesses (TIMCI) project.
#'
#' @param rctls_pid numeric ID of the RCT/LS ODK Central project (mandatory parameter)
#' @param rctls_pp passphrase (mandatory parameter)
#' @param research_facilities dataframe that contains the research facilities (mandatory parameter)
#' @param fu_dir path to the output folder for the follow-up exports (mandatory parameter)
#' @param start_date start date (optional parameter)
#' @param end_date end date (optional parameter)
#' @param for_today boolean that enables to generate the follow-up for today (defaut set to `FALSE`)
#' @param hospitfu_ok boolean that enables to generate the hospital follow-up for today (defaut set to `TRUE`)
#' @param is_pilot Boolean, default set to `FALSE`
#' @import rmarkdown ruODK
#' @export
generate_fu_logs <- function(rctls_pid,
rctls_pp,
research_facilities,
fu_dir,
start_date = NULL,
end_date = NULL,
for_today = FALSE,
hospitfu_ok = TRUE,
is_pilot = FALSE) {
###########################
# Set up current language #
###########################
timci::set_language_settings()
################
# Set up ruODK #
################
ruODK::ru_setup(
svc = Sys.getenv("ODKC_SVC"),
un = Sys.getenv("ODKC_UN"),
pw = Sys.getenv("ODKC_PW"),
tz = Sys.getenv("TZ"),
verbose = TRUE # Can be switched to TRUE for demo or debugging
)
# List of projects visible with the credentials `ODKC_UN` and `ODKC_PW`
odkc_project_list <- ruODK::project_list()$id
wd_fid <- Sys.getenv("TIMCI_WD_FID")
# RCT / LS environment variables
crf_facility_fid <- Sys.getenv("TIMCI_CRF_FACILITY_FID")
crf_day7_fid <- Sys.getenv("TIMCI_CRF_DAY7_FID")
crf_hospit_fid <- Sys.getenv("TIMCI_CRF_HOSPIT_FID")
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania" || Sys.getenv('TIMCI_COUNTRY') == "India") {
crf_day28_fid <- Sys.getenv("TIMCI_CRF_DAY28_FID")
}
thres_date <- Sys.Date() - 3000# 45
filter <- paste0("__system/submissionDate ge ", thres_date)
#######################
# Load TIMCI ODK data #
#######################
# List of forms visible in the RCT / LS project
rct_ls_form_list <- ruODK::form_list(pid = rctls_pid)$fid
# Load facility data
print("Load facility data")
raw_facility_zip <- ruODK::submission_export(local_dir = tempdir(),
pid = rctls_pid,
fid = crf_facility_fid,
pp = rctls_pp,
filter = filter,
delfields = FALSE,
group = TRUE,
split = FALSE,
media = FALSE)
raw_facility_data <- timci::extract_data_from_odk_zip(raw_facility_zip,
paste0(crf_facility_fid,".csv"),
start_date,
end_date)
write("Generate follow-up logs", stderr())
if (Sys.getenv('TIMCI_COUNTRY') == 'Tanzania') {
facility_data <- timci::process_tanzania_facility_data(raw_facility_data, is_pilot)
facility_data <- timci::correct_day0_all(facility_data)
} else if (Sys.getenv('TIMCI_COUNTRY') == 'Kenya') {
facility_data <- timci::process_facility_data(raw_facility_data, is_pilot)
facility_data <- timci::correct_day0_all(facility_data)
} else {
facility_data <- timci::process_facility_data(raw_facility_data, is_pilot)
facility_data <- timci::correct_day0_all(facility_data)
}
pii <- timci::extract_enrolled_participants(facility_data, is_pilot)[[2]]
# Load day 7 follow-up data
print("Load day 7 follow-up data")
raw_day7fu_data <- NULL
if (crf_day7_fid %in% rct_ls_form_list) {
raw_day7fu_zip <- ruODK::submission_export(local_dir = tempdir(),
pid = rctls_pid,
fid = crf_day7_fid,
pp = rctls_pp,
filter = filter,
delfields = FALSE,
group = TRUE,
split = FALSE,
media = FALSE)
raw_day7fu_data <- timci::extract_data_from_odk_zip(raw_day7fu_zip,
paste0(crf_day7_fid,".csv"),
start_date,
end_date)
}
# Load hospital visit follow-up data
print("Load hospital visit data")
raw_hospit_data <- NULL
if (crf_hospit_fid %in% rct_ls_form_list) {
raw_hospit_zip <- ruODK::submission_export(local_dir = tempdir(),
pid = rctls_pid,
fid = crf_hospit_fid,
pp = rctls_pp,
delfields = FALSE,
group = TRUE,
split = FALSE,
media = FALSE)
raw_hospit_data <- timci::extract_data_from_odk_zip(raw_hospit_zip,
paste0(crf_hospit_fid,".csv"),
start_date,
end_date)
}
# Load day 28 follow-up data
raw_day28fu_data <- NULL
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania" || Sys.getenv('TIMCI_COUNTRY') == "India") {
write("Load day 28 follow-up data", stderr())
if (crf_day28_fid %in% rct_ls_form_list) {
raw_day28fu_zip <- ruODK::submission_export(local_dir = tempdir(),
pid = rctls_pid,
fid = crf_day28_fid,
pp = rctls_pp,
filter = filter,
delfields = FALSE,
group = TRUE,
split = FALSE,
media = FALSE)
raw_day28fu_data <- timci::extract_data_from_odk_zip(raw_day28fu_zip,
paste0(crf_day28_fid,".csv"),
start_date,
end_date)
}
}
# Load widthdrawal data
write("Load withdrawal data", stderr())
raw_withdrawal_data <- NULL
if (wd_fid %in% rct_ls_form_list) {
raw_withdrawal_zip <- ruODK::submission_export(local_dir = tempdir(),
pid = rctls_pid,
fid = wd_fid,
pp = rctls_pp,
delfields = FALSE,
group = TRUE,
split = FALSE,
media = FALSE)
raw_withdrawal_data <- timci::extract_data_from_odk_zip(raw_withdrawal_zip,
paste0(wd_fid,".csv"),
start_date,
end_date)
}
#######################
# Day 7 follow-up log #
#######################
write("Generate the Day 7 follow-up log", stderr())
day7fu_dir <- file.path(fu_dir, "day7_log")
dir.create(day7fu_dir, showWarnings = FALSE)
fu7all <- timci::generate_fu_log(pii, raw_day7fu_data, 0, 12, 7, 10, ext = TRUE, deidentify = FALSE)
timci::export_df2xlsx(fu7all, day7fu_dir, "02_timci_day7_fu_weekly_log_all")
# Daily log
fu_start <- 6
physicalfu_start <- 8
physicalfu_end <- 14
if (Sys.getenv('TIMCI_COUNTRY') == "Kenya") {
fu_end <- 15
} else {
fu_end <- 12
}
if (for_today) {
fu_start <- fu_start + 1
physicalfu_start <- physicalfu_start + 1
fu_end <- fu_end + 1
physicalfu_end <- physicalfu_end + 1
}
params <- list(output_dir = fu_dir,
rct_ls_form_list = rct_ls_form_list,
facility_data = facility_data,
rctls_pid = rctls_pid,
fu_fid = crf_day7_fid,
raw_fu_data = raw_day7fu_data,
raw_withdrawal_data = raw_withdrawal_data,
fu_start = fu_start,
fu_end = fu_end,
fu_vstart = 7,
fu_vend = 10)
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania") {
physical_params <- list(physicalfu_start = physicalfu_start,
physicalfu_end = physicalfu_end)
params <- append(params, physical_params)
}
generate_pdf_report(day7fu_dir, "fu_daily_log.Rmd", "01_timci_day7_fu_daily_log", params)
# # Weekly log
# for (i in 1:nrow(research_facilities)) {
# fid <- research_facilities[[i, 'facility_id']]
# fname <- research_facilities[[i, 'facility_label']]
# params <- list(pii = pii,
# fu_fid = crf_day7_fid,
# raw_fu_data = raw_day7fu_data,
# raw_withdrawal_data = raw_withdrawal_data,
# facility_id = fid,
# facility_label = fname,
# fu_start = 0,
# fu_end = 12,
# fu_vstart = 7,
# fu_vend = 10)
# generate_word_report(day7fu_dir, "fu_weekly_log.Rmd", paste0(fid, "_", fname, "_timci_day7_fu_weekly_log"), params)
# }
#################################
# Hospitalisation follow-up log #
#################################
if (hospitfu_ok){
hospitfu_dir <- file.path(fu_dir, "hospitalisation_log")
dir.create(hospitfu_dir, showWarnings = FALSE)
params <- list(output_dir = fu_dir,
rct_ls_form_list = rct_ls_form_list,
pii = pii,
day0_data = facility_data,
rctls_pid = rctls_pid,
raw_day7fu_data = raw_day7fu_data,
raw_hospit_data = raw_hospit_data,
raw_withdrawal_data = raw_withdrawal_data,
fu_end = 12)
generate_word_report(hospitfu_dir, "hospit_fu_log.Rmd", "timci_hospit_fu_log", params)
}
#######################
# Day 28 follow-up log #
#######################
# Day 28 follow-up log is only generated for India and Tanzania
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania" || Sys.getenv('TIMCI_COUNTRY') == "India") {
day28fu_dir <- file.path(fu_dir, "day28_log")
dir.create(day28fu_dir, showWarnings = FALSE)
fu28all <- timci::generate_fu_log(pii, raw_day28fu_data, 0, 32, 28, 32)
timci::export_df2xlsx(fu28all, day28fu_dir, "02_timci_day28_fu_weekly_log_all")
# Daily log
fu_start <- 27
physicalfu_start <- 29
fu_end <- 35
physicalfu_end <- 40
if (for_today) {
fu_start <- fu_start + 1
physicalfu_start <- physicalfu_start + 1
fu_end <- fu_end + 1
physicalfu_end <- physicalfu_end + 1
}
params <- list(output_dir = fu_dir,
rct_ls_form_list = rct_ls_form_list,
facility_data = facility_data,
rctls_pid = rctls_pid,
fu_fid = crf_day28_fid,
raw_fu_data = raw_day28fu_data,
raw_oth_fu_data = raw_day7fu_data,
raw_withdrawal_data = raw_withdrawal_data,
fu_start = fu_start,
fu_end = fu_end,
fu_vstart = 28,
fu_vend = 32)
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania") {
physical_params <- list(physicalfu_start = physicalfu_start,
physicalfu_end = physicalfu_end)
params <- append(params, physical_params)
}
generate_pdf_report(day28fu_dir, "fu_daily_log.Rmd", "01_timci_day28_fu_daily_log", params)
# # Weekly log
# for (i in 1:nrow(research_facilities)) {
# fid <- research_facilities[[i, 'facility_id']]
# fname <- research_facilities[[i, 'facility_label']]
# params <- list(pii = pii,
# fu_fid = crf_day28_fid,
# raw_fu_data = raw_day28fu_data,
# raw_withdrawal_data = raw_withdrawal_data,
# facility_id = fid,
# facility_label = fname,
# fu_start = 21,
# fu_end = 35,
# fu_vstart = 28,
# fu_vend = 32)
# generate_word_report(day28fu_dir, "fu_weekly_log.Rmd", paste0(fid, "_", fname, "_timci_day28_fu_weekly_log"), params)
# }
}
}
#' Generate follow-up log (TIMCI-specific function)
#'
#' Generate a list of participants to be called in a time window after baseline between wmin and wmax
#' @param pii dataframe containing personally identifiable data
#' @param fudf dataframe containing the processed follow-up data
#' @param wmin numerical, start of the follow-up period (in days)
#' @param wmax numerical, end of the follow-up period (in days)
#' @param vwmin numerical, start of the follow-up period valid for the statistical analysis (in days)
#' @param vwmax numerical, end of the follow-up period valid for the statistical analysis (in days)
#' @param ext boolean, allow to export more information
#' @param deidentify boolean, enable the generation of a format with personally identifiable information (PII) for upload on ODK Central (if set to FALSE) or of a deidentified dataframe (if set to TRUE)
#' @return This function returns a dataframe.
#' @export
#' @import magrittr dplyr
generate_fu_log <- function(pii,
fudf,
wmin,
wmax,
vwmin,
vwmax,
ext = FALSE,
deidentify = FALSE) {
fu_log <- pii
fu_log$min_date <- as.Date(fu_log$date_visit) + wmin
fu_log$max_date <- as.Date(fu_log$date_visit) + wmax
if(Sys.getenv('TIMCI_COUNTRY') != 'Tanzania'){
fu_log$label <- paste(fu_log$fs_name, fu_log$ls_name)
} else{
fu_log$label <- paste(fu_log$fs_name, fu_log$ms_name, fu_log$ls_name)
}
fu_log$caregiver <- paste(fu_log$cg_fs_name, fu_log$cg_ls_name)
fu_log$mother <- paste(fu_log$mother_fs_name, fu_log$mother_ls_name)
fu_log$sex <- ifelse(fu_log$sex == 1, "male", ifelse(fu_log$sex == 2, "female", "other"))
# Exclude children who already underwent successful follow-up
if (!is.null(fudf)) {
if (nrow(fudf) > 0) {
fudf <- fudf %>% dplyr::filter(proceed == 1)
fu_log <- fu_log[!(fu_log$child_id %in% fudf$"a1-pid"),]
}
}
# Exclude children who are outside of the follow-up window period
fu_log <- fu_log[fu_log$min_date <= Sys.Date() & fu_log$max_date >= Sys.Date(),]
# Order columns
if (deidentify) {
col_order <- c('fid',
'child_id',
'sex',
'date_visit',
'main_cg_lbl')
} else{
col_order <- c('fid',
'child_id',
'label',
'sex',
'date_visit',
'caregiver',
'main_cg_lbl',
'mother',
'phone_nb',
'phone_nb2',
'location',
'cmty_contact')
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania") {
col_order <- c(col_order, 'physical_fu_instructions')
}
}
# If ext is set to TRUE, also add the district and facility name
if (ext) {
col_order <- c('district',
'facility',
col_order)
}
# Select only a subset of columns
fu_log <- fu_log[, col_order]
# Order entries by date
fu_log <- fu_log %>%
dplyr::arrange(fid, date_visit = as.Date(date_visit, "%Y-%m-%d"))
# Add valid window in export
if ( is_not_empty(fu_log) ) {
fu_log$date_visit <- paste0("From ", as.Date(fu_log$'date_visit', "%Y-%m-%d") + vwmin, " to ", as.Date(fu_log$'date_visit', "%Y-%m-%d") + vwmax, " [enrolled on ", fu_log$date_visit, "]")
}
# Add a first generic row
# If ext is set to TRUE, do not add
if (!ext & !deidentify) {
first_row <- data.frame('fid' = 'F0000',
'child_id' = 'X-F0000-P0000',
'label' = 'CHILD NAME',
'sex' = 'SEX',
'date_visit' = 'VALID WINDOW [ENROLMENT DATE]',
'caregiver' = 'CAREGIVER NAME',
'main_cg_lbl' = 'RELATIONSHIP',
'mother' = 'MOTHER NAME',
'phone_nb' = 'PHONE NB 1',
'phone_nb2' = 'PHONE NB 2',
'location' = 'LOCATION',
'cmty_contact' = 'CONTACT')
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania") {
first_row$physical_fu_instructions <- 'INSTRUCTIONS'
}
fu_log <- rbind(first_row,
fu_log)
}
if (!deidentify) {
fu_log <- fu_log %>% dplyr::rename('name' = 'child_id',
'enroldate' = 'date_visit',
'relationship' = 'main_cg_lbl',
'phonenb1' = 'phone_nb',
'phonenb2' = 'phone_nb2',
'contact' = 'cmty_contact')
}
fu_log
}
#' Generate follow-up log CSV for upload on ODK Central (TIMCI-specific function)
#'
#' Generate a list of participants to be called in a time window after baseline between wmin and wmax
#' @param pii dataframe containing personally identifiable data
#' @param fudf dataframe containing the processed follow-up data
#' @param wmin numerical, start of the follow-up period (in days)
#' @param wmax numerical, end of the follow-up period (in days)
#' @param vwmin numerical, start of the follow-up period valid for the statistical analysis (in days)
#' @param vwmax numerical, end of the follow-up period valid for the statistical analysis (in days)
#' @return This function returns a dataframe.
#' @export
#' @import magrittr dplyr
generate_fu_log_csv <- function(pii,
fudf,
wmin,
wmax,
vwmin,
vwmax) {
fu_log <- pii
fu_log$min_date <- as.Date(fu_log$date_visit) + wmin
fu_log$max_date <- as.Date(fu_log$date_visit) + wmax
if(Sys.getenv('TIMCI_COUNTRY') != 'Tanzania'){
fu_log$child_name <- paste(fu_log$fs_name, fu_log$ls_name)
fu_log$label <- paste0(fu_log$child_name,
" (",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmin,
" to ",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmax,
" )")
}
else{
fu_log$child_name <- paste(fu_log$fs_name, fu_log$ms_name, fu_log$ls_name)
fu_log$label <- paste0(fu_log$child_name,
" (",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmin,
" to ",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmax,
" ) - ",
fu_log$ra_name)
}
fu_log$caregiver <- paste(fu_log$cg_fs_name, fu_log$cg_ls_name)
fu_log$mother <- paste(fu_log$mother_fs_name, fu_log$mother_ls_name)
fu_log$sex <- ifelse(fu_log$sex == 1, "male", ifelse(fu_log$sex == 2, "female", "other"))
# Exclude children who already underwent successful follow-up
if (!is.null(fudf)) {
if (nrow(fudf) > 0) {
fudf <- fudf %>% dplyr::filter(proceed == 1)
fu_log <- fu_log[!(fu_log$child_id %in% fudf$"a1-pid"),]
}
}
# Exclude children who are outside of the follow-up window period
fu_log <- fu_log[fu_log$min_date <= Sys.Date() & fu_log$max_date >= Sys.Date(),]
if (nrow(fu_log) > 0) {
fu_log$population <- "all"
fu_log$type <- "1"
if(Sys.getenv('TIMCI_COUNTRY') != 'Tanzania'){
fu_log$physical_fu_instructions <- ""
}
# Order columns
col_order <- c('fid',
'device_id',
'child_id',
'child_name',
'sex',
'date_visit',
'caregiver',
'main_cg_lbl',
'mother',
'phone_nb',
'phone_nb2',
'location',
'physical_fu_instructions',
'cmty_contact',
'label',
'sys_submit_id',
'population',
'type')
fu_log <- fu_log[, col_order]
# Order entries by date
fu_log <- fu_log %>%
dplyr::arrange(fid, sys_submit_id, date_visit = as.Date(date_visit, "%Y-%m-%d"))
# Add a first generic row
fu_log <- rbind(data.frame('fid' = 'F0000',
'device_id' = 'none',
'child_id' = 'X-F0000-P0000',
'child_name' = 'CHILD NAME',
'sex' = 'SEX',
'date_visit' = 'ENROLMENT DATE',
'caregiver' = 'CAREGIVER NAME',
'main_cg_lbl' = 'RELATIONSHIP',
'mother' = 'MOTHER NAME',
'phone_nb' = 'PHONE NB 1',
'phone_nb2' = 'PHONE NB 2',
'location' = 'LOCATION',
'physical_fu_instructions' = 'INSTRUCTIONS',
'cmty_contact' = 'CONTACT',
'label' = 'CHILD NAME (VALID WINDOW)',
'sys_submit_id' = 'SUBMITTER',
'population' = 'POPULATION',
'type' = 'MANUAL'),
fu_log)
fu_log <- fu_log %>% dplyr::rename('name' = 'child_id',
'enroldate' = 'date_visit',
'relationship' = 'main_cg_lbl',
'phonenb1' = 'phone_nb',
'phonenb2' = 'phone_nb2',
'contact' = 'cmty_contact',
'submitter' = 'sys_submit_id')
}
fu_log
}
#' Generate physical follow-up log CSV for upload on ODK Central (TIMCI-specific function, Tanzania only)
#'
#' Generate a list of participants to be called in a time window after baseline between wmin and wmax
#' @param pii dataframe containing personally identifiable data
#' @param fudf dataframe containing the processed follow-up data
#' @param wmin_nophone numerical, start of the follow-up period if the caregiver has no phone number (in days)
#' @param wmin_phone numerical, start of the follow-up period if the caregiver has a phone number (in days)
#' @param wmax numerical, end of the follow-up period (in days)
#' @param vwmin numerical, start of the follow-up period valid for the statistical analysis (in days)
#' @param vwmax numerical, end of the follow-up period valid for the statistical analysis (in days)
#' @return This function returns a dataframe.
#' @export
#' @import magrittr dplyr
generate_physical_fu_log_csv <- function(pii,
fudf,
wmin_nophone,
wmin_phone,
wmax,
vwmin,
vwmax) {
fu_log <- pii
fu_log$min_nophone_date <- as.Date(fu_log$date_visit) + wmin_nophone
fu_log$min_phone_date <- as.Date(fu_log$date_visit) + wmin_phone
fu_log$max_date <- as.Date(fu_log$date_visit) + wmax
fu_log$child_name <- paste(fu_log$fs_name, fu_log$ms_name, fu_log$ls_name)
fu_log$caregiver <- paste(fu_log$cg_fs_name, fu_log$cg_ls_name)
fu_log$mother <- paste(fu_log$mother_fs_name, fu_log$mother_ls_name)
fu_log$sex <- ifelse(fu_log$sex == 1, "male", ifelse(fu_log$sex == 2, "female", "other"))
fu_log$label <- paste0(fu_log$child_name,
" (",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmin,
" to ",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmax,
" )")
# Exclude children who already underwent successful follow-up
if (!is.null(fudf)) {
if (nrow(fudf) > 0) {
fudf <- fudf %>% dplyr::filter(proceed == 1)
fu_log <- fu_log[!(fu_log$child_id %in% fudf$"a1-pid"),]
}
}
#########################################################################
# Select children for whom the caregiver did not provide a phone number #
#########################################################################
fu_log_nophone <- fu_log %>%
dplyr::filter(phone_nb_avail == 0) %>% # Include only children for whom the caregiver did not provide a phone number
dplyr::filter(min_nophone_date <= Sys.Date() & max_date >= Sys.Date()) # Exclude children who are outside of the physical follow-up window period
if (nrow(fu_log_nophone)){
fu_log_nophone$population <- "no phone number available"
fu_log_nophone$type <- "2"
} else {
fu_log_nophone$population <- character()
fu_log_nophone$type <- character()
}
#######################################################################################################
# Select children for whom the caregiver provided a phone number and agreed with a physical follow-up #
#######################################################################################################
# Add a constraints on having at least 1 recorded failed attempt
fu_log_phone <- fu_log %>%
dplyr::filter(phone_nb_avail == 1 & physical_fu == 1) %>% # Include only children for whom the caregiver provided a phone number and agreed with physical follow-up
dplyr::filter(min_phone_date <= Sys.Date() & max_date >= Sys.Date()) # Exclude children who are outside of the physical follow-up window period
if (nrow(fu_log_phone)) {
fu_log_phone$population <- "phone number available"
fu_log_phone$type <- "2"
} else {
fu_log_phone$population <- character()
fu_log_phone$type <- character()
}
#########################################################
# Combine both selections of children in one single log #
#########################################################
#if (!is.null(fu_log_nophone)) {
# if (nrow(fu_log_nophone) > 0) {
# if (!is.null(fu_log_phone)) {
# fu_log <- dplyr::bind_rows(fu_log_nophone, fu_log_phone)
# } else{
# fu_log <- fu_log_nophone
# }
# }
#}
fu_log <- dplyr::bind_rows(fu_log_nophone,fu_log_phone)
# Order columns
col_order <- c('fid',
'device_id',
'child_id',
'child_name',
'sex',
'date_visit',
'caregiver',
'main_cg_lbl',
'mother',
'phone_nb',
'phone_nb2',
'location',
'location_lvl1',
'location_lvl2',
'location_lvl3',
'location_lvl4',
'physical_fu_instructions',
'cg_common_name',
'cmty_contact',
'label',
'sys_submit_id',
'population',
'type')
fu_log <- fu_log[, col_order]
# Order entries by date
fu_log <- fu_log %>%
dplyr::arrange(date_visit = as.Date(date_visit, "%Y-%m-%d"))
# Add a first generic row
fu_log <- rbind(data.frame('fid' = 'F0000',
'device_id' = 'none',
'child_id' = 'X-F0000-P0000',
'child_name' = 'CHILD NAME',
'sex' = 'SEX',
'date_visit' = 'ENROLMENT DATE',
'caregiver' = 'CAREGIVER NAME',
'main_cg_lbl' = 'RELATIONSHIP',
'mother' = 'MOTHER NAME',
'phone_nb' = 'PHONE NB 1',
'phone_nb2' = 'PHONE NB 2',
'location' = 'LOCATION',
'location_lvl1' = 'LOCATION_DIVISION',
'location_lvl2' = 'LOCATION_WARD',
'location_lvl3' = 'LOCATION_LVL3',
'location_lvl4' = 'LOCATION_LVL4',
'cmty_contact' = 'CONTACT',
'physical_fu_instructions' = 'INSTRUCTIONS',
'cg_common_name' = 'CAREGIVER COMMON NAME',
'label' = 'CHILD NAME (VALID WINDOW)',
'sys_submit_id' = 'SUBMITTER',
'population' = 'POPULATION',
'type' = 'MANUAL'),
fu_log)
fu_log %>% dplyr::rename('name' = 'child_id',
'enroldate' = 'date_visit',
'relationship' = 'main_cg_lbl',
'phonenb1' = 'phone_nb',
'phonenb2' = 'phone_nb2',
'contact' = 'cmty_contact',
'instructions' = 'physical_fu_instructions',
'submitter' = 'sys_submit_id')
}
#' Generate lost to follow-up log (TIMCI-specific function)
#'
#' Generate a list of participants who are lost to follow-up after wmax
#' @param df dataframe
#' @param fudf dataframe containing the processed follow-up data
#' @param end_date numerical, end of the follow-up period (in days)
#' @param raw boolean flag for selecting raw (TRUE) or processed (FALSE) data
#' @return This function returns a dataframe.
#' @export
#' @import magrittr dplyr
generate_ltfu_log <- function(df,
fudf,
end_date,
raw = TRUE) {
ltfu_log <- df
ltfu_log$date_maxfu <- as.Date(ltfu_log$date_visit) + end_date
# If raw data
if (raw) {
fudf <- fudf %>%
dplyr::rename(child_id = "a1-pid",
proceed_day7 = proceed)
}
# Exclude children who already underwent successful follow-up
if (!is.null(fudf)) {
if (nrow(fudf) > 0) {
suc_fudf <- fudf %>%
dplyr::filter(proceed_day7 == 1)
un_fudf <- fudf %>%
dplyr::filter(proceed_day7 == 0)
# Count the number of unsuccessful attempts
attempt_df <- un_fudf %>%
group_by(child_id) %>%
count
ltfu_log <- ltfu_log[!(ltfu_log$child_id %in% suc_fudf$child_id),]
ltfu_log <- merge(ltfu_log, attempt_df, by = 'child_id', all.x = TRUE) %>%
dplyr::rename(fu_attempts = n)
}
}
# If fu_attempts does not exist, create a column and fill out values with zeroes
if(!("fu_attempts" %in% colnames(ltfu_log))) {
ltfu_log$fu_attempts <- 0
}
# Exclude children who are still within or before the follow-up window period
ltfu_log <- ltfu_log[ltfu_log$date_maxfu < Sys.Date(),]
# Order entries by date
ltfu_log <- ltfu_log %>%
dplyr::arrange(date_visit = as.Date(date_visit, "%Y-%m-%d"))
# Order columns
ltfu_log$date_maxfu <- as.Date(ltfu_log$date_maxfu, "%Y-%m-%d")
if(Sys.getenv('TIMCI_COUNTRY') == 'India'){
col_order <- c('child_id',
'date_visit',
'date_maxfu',
'fu_attempts',
'age_mo',
'yg_infant',
'who_age_ctg',
'fid',
'facility',
'district',
'referral_cg',
'referral_hf',
'urg_referral_hf',
'ref_facility',
'ref_facility_oth',
'device_id',
'sys_submit_id')
} else{
col_order <- c('child_id',
'date_visit',
'date_maxfu',
'fu_attempts',
'age_mo',
'yg_infant',
'who_age_ctg',
'fid',
'facility',
'district',
'phone_nb_avail',
'referral_cg',
'referral_hf',
'urg_referral_hf',
'ref_facility',
'ref_facility_oth',
'device_id',
'sys_submit_id')
}
ltfu_log <- ltfu_log[, col_order]
}
Thaliehln/timci documentation built on April 8, 2024, 3:38 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions generate_ltfu_log generate_physical_fu_log_csv generate_fu_log_csv generate_fu_log generate_fu_logs
Documented in generate_fu_log generate_fu_log_csv generate_fu_logs generate_ltfu_log generate_physical_fu_log_csv
#' General follow-up logs
#'
#' This function runs follow-up Rmarkdown files to generate standardised follow-up logs for the Tools for Integrated Management of Childhood Illnesses (TIMCI) project.
#'
#' @param rctls_pid numeric ID of the RCT/LS ODK Central project (mandatory parameter)
#' @param rctls_pp passphrase (mandatory parameter)
#' @param research_facilities dataframe that contains the research facilities (mandatory parameter)
#' @param fu_dir path to the output folder for the follow-up exports (mandatory parameter)
#' @param start_date start date (optional parameter)
#' @param end_date end date (optional parameter)
#' @param for_today boolean that enables to generate the follow-up for today (defaut set to `FALSE`)
#' @param hospitfu_ok boolean that enables to generate the hospital follow-up for today (defaut set to `TRUE`)
#' @param is_pilot Boolean, default set to `FALSE`
#' @import rmarkdown ruODK
#' @export
generate_fu_logs <- function(rctls_pid,
rctls_pp,
research_facilities,
fu_dir,
start_date = NULL,
end_date = NULL,
for_today = FALSE,
hospitfu_ok = TRUE,
is_pilot = FALSE) {
###########################
# Set up current language #
###########################
timci::set_language_settings()
################
# Set up ruODK #
################
ruODK::ru_setup(
svc = Sys.getenv("ODKC_SVC"),
un = Sys.getenv("ODKC_UN"),
pw = Sys.getenv("ODKC_PW"),
tz = Sys.getenv("TZ"),
verbose = TRUE # Can be switched to TRUE for demo or debugging
)
# List of projects visible with the credentials `ODKC_UN` and `ODKC_PW`
odkc_project_list <- ruODK::project_list()$id
wd_fid <- Sys.getenv("TIMCI_WD_FID")
# RCT / LS environment variables
crf_facility_fid <- Sys.getenv("TIMCI_CRF_FACILITY_FID")
crf_day7_fid <- Sys.getenv("TIMCI_CRF_DAY7_FID")
crf_hospit_fid <- Sys.getenv("TIMCI_CRF_HOSPIT_FID")
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania" || Sys.getenv('TIMCI_COUNTRY') == "India") {
crf_day28_fid <- Sys.getenv("TIMCI_CRF_DAY28_FID")
}
thres_date <- Sys.Date() - 3000# 45
filter <- paste0("__system/submissionDate ge ", thres_date)
#######################
# Load TIMCI ODK data #
#######################
# List of forms visible in the RCT / LS project
rct_ls_form_list <- ruODK::form_list(pid = rctls_pid)$fid
# Load facility data
print("Load facility data")
raw_facility_zip <- ruODK::submission_export(local_dir = tempdir(),
pid = rctls_pid,
fid = crf_facility_fid,
pp = rctls_pp,
filter = filter,
delfields = FALSE,
group = TRUE,
split = FALSE,
media = FALSE)
raw_facility_data <- timci::extract_data_from_odk_zip(raw_facility_zip,
paste0(crf_facility_fid,".csv"),
start_date,
end_date)
write("Generate follow-up logs", stderr())
if (Sys.getenv('TIMCI_COUNTRY') == 'Tanzania') {
facility_data <- timci::process_tanzania_facility_data(raw_facility_data, is_pilot)
facility_data <- timci::correct_day0_all(facility_data)
} else if (Sys.getenv('TIMCI_COUNTRY') == 'Kenya') {
facility_data <- timci::process_facility_data(raw_facility_data, is_pilot)
facility_data <- timci::correct_day0_all(facility_data)
} else {
facility_data <- timci::process_facility_data(raw_facility_data, is_pilot)
facility_data <- timci::correct_day0_all(facility_data)
}
pii <- timci::extract_enrolled_participants(facility_data, is_pilot)[[2]]
# Load day 7 follow-up data
print("Load day 7 follow-up data")
raw_day7fu_data <- NULL
if (crf_day7_fid %in% rct_ls_form_list) {
raw_day7fu_zip <- ruODK::submission_export(local_dir = tempdir(),
pid = rctls_pid,
fid = crf_day7_fid,
pp = rctls_pp,
filter = filter,
delfields = FALSE,
group = TRUE,
split = FALSE,
media = FALSE)
raw_day7fu_data <- timci::extract_data_from_odk_zip(raw_day7fu_zip,
paste0(crf_day7_fid,".csv"),
start_date,
end_date)
}
# Load hospital visit follow-up data
print("Load hospital visit data")
raw_hospit_data <- NULL
if (crf_hospit_fid %in% rct_ls_form_list) {
raw_hospit_zip <- ruODK::submission_export(local_dir = tempdir(),
pid = rctls_pid,
fid = crf_hospit_fid,
pp = rctls_pp,
delfields = FALSE,
group = TRUE,
split = FALSE,
media = FALSE)
raw_hospit_data <- timci::extract_data_from_odk_zip(raw_hospit_zip,
paste0(crf_hospit_fid,".csv"),
start_date,
end_date)
}
# Load day 28 follow-up data
raw_day28fu_data <- NULL
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania" || Sys.getenv('TIMCI_COUNTRY') == "India") {
write("Load day 28 follow-up data", stderr())
if (crf_day28_fid %in% rct_ls_form_list) {
raw_day28fu_zip <- ruODK::submission_export(local_dir = tempdir(),
pid = rctls_pid,
fid = crf_day28_fid,
pp = rctls_pp,
filter = filter,
delfields = FALSE,
group = TRUE,
split = FALSE,
media = FALSE)
raw_day28fu_data <- timci::extract_data_from_odk_zip(raw_day28fu_zip,
paste0(crf_day28_fid,".csv"),
start_date,
end_date)
}
}
# Load widthdrawal data
write("Load withdrawal data", stderr())
raw_withdrawal_data <- NULL
if (wd_fid %in% rct_ls_form_list) {
raw_withdrawal_zip <- ruODK::submission_export(local_dir = tempdir(),
pid = rctls_pid,
fid = wd_fid,
pp = rctls_pp,
delfields = FALSE,
group = TRUE,
split = FALSE,
media = FALSE)
raw_withdrawal_data <- timci::extract_data_from_odk_zip(raw_withdrawal_zip,
paste0(wd_fid,".csv"),
start_date,
end_date)
}
#######################
# Day 7 follow-up log #
#######################
write("Generate the Day 7 follow-up log", stderr())
day7fu_dir <- file.path(fu_dir, "day7_log")
dir.create(day7fu_dir, showWarnings = FALSE)
fu7all <- timci::generate_fu_log(pii, raw_day7fu_data, 0, 12, 7, 10, ext = TRUE, deidentify = FALSE)
timci::export_df2xlsx(fu7all, day7fu_dir, "02_timci_day7_fu_weekly_log_all")
# Daily log
fu_start <- 6
physicalfu_start <- 8
physicalfu_end <- 14
if (Sys.getenv('TIMCI_COUNTRY') == "Kenya") {
fu_end <- 15
} else {
fu_end <- 12
}
if (for_today) {
fu_start <- fu_start + 1
physicalfu_start <- physicalfu_start + 1
fu_end <- fu_end + 1
physicalfu_end <- physicalfu_end + 1
}
params <- list(output_dir = fu_dir,
rct_ls_form_list = rct_ls_form_list,
facility_data = facility_data,
rctls_pid = rctls_pid,
fu_fid = crf_day7_fid,
raw_fu_data = raw_day7fu_data,
raw_withdrawal_data = raw_withdrawal_data,
fu_start = fu_start,
fu_end = fu_end,
fu_vstart = 7,
fu_vend = 10)
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania") {
physical_params <- list(physicalfu_start = physicalfu_start,
physicalfu_end = physicalfu_end)
params <- append(params, physical_params)
}
generate_pdf_report(day7fu_dir, "fu_daily_log.Rmd", "01_timci_day7_fu_daily_log", params)
# # Weekly log
# for (i in 1:nrow(research_facilities)) {
# fid <- research_facilities[[i, 'facility_id']]
# fname <- research_facilities[[i, 'facility_label']]
# params <- list(pii = pii,
# fu_fid = crf_day7_fid,
# raw_fu_data = raw_day7fu_data,
# raw_withdrawal_data = raw_withdrawal_data,
# facility_id = fid,
# facility_label = fname,
# fu_start = 0,
# fu_end = 12,
# fu_vstart = 7,
# fu_vend = 10)
# generate_word_report(day7fu_dir, "fu_weekly_log.Rmd", paste0(fid, "_", fname, "_timci_day7_fu_weekly_log"), params)
# }
#################################
# Hospitalisation follow-up log #
#################################
if (hospitfu_ok){
hospitfu_dir <- file.path(fu_dir, "hospitalisation_log")
dir.create(hospitfu_dir, showWarnings = FALSE)
params <- list(output_dir = fu_dir,
rct_ls_form_list = rct_ls_form_list,
pii = pii,
day0_data = facility_data,
rctls_pid = rctls_pid,
raw_day7fu_data = raw_day7fu_data,
raw_hospit_data = raw_hospit_data,
raw_withdrawal_data = raw_withdrawal_data,
fu_end = 12)
generate_word_report(hospitfu_dir, "hospit_fu_log.Rmd", "timci_hospit_fu_log", params)
}
#######################
# Day 28 follow-up log #
#######################
# Day 28 follow-up log is only generated for India and Tanzania
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania" || Sys.getenv('TIMCI_COUNTRY') == "India") {
day28fu_dir <- file.path(fu_dir, "day28_log")
dir.create(day28fu_dir, showWarnings = FALSE)
fu28all <- timci::generate_fu_log(pii, raw_day28fu_data, 0, 32, 28, 32)
timci::export_df2xlsx(fu28all, day28fu_dir, "02_timci_day28_fu_weekly_log_all")
# Daily log
fu_start <- 27
physicalfu_start <- 29
fu_end <- 35
physicalfu_end <- 40
if (for_today) {
fu_start <- fu_start + 1
physicalfu_start <- physicalfu_start + 1
fu_end <- fu_end + 1
physicalfu_end <- physicalfu_end + 1
}
params <- list(output_dir = fu_dir,
rct_ls_form_list = rct_ls_form_list,
facility_data = facility_data,
rctls_pid = rctls_pid,
fu_fid = crf_day28_fid,
raw_fu_data = raw_day28fu_data,
raw_oth_fu_data = raw_day7fu_data,
raw_withdrawal_data = raw_withdrawal_data,
fu_start = fu_start,
fu_end = fu_end,
fu_vstart = 28,
fu_vend = 32)
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania") {
physical_params <- list(physicalfu_start = physicalfu_start,
physicalfu_end = physicalfu_end)
params <- append(params, physical_params)
}
generate_pdf_report(day28fu_dir, "fu_daily_log.Rmd", "01_timci_day28_fu_daily_log", params)
# # Weekly log
# for (i in 1:nrow(research_facilities)) {
# fid <- research_facilities[[i, 'facility_id']]
# fname <- research_facilities[[i, 'facility_label']]
# params <- list(pii = pii,
# fu_fid = crf_day28_fid,
# raw_fu_data = raw_day28fu_data,
# raw_withdrawal_data = raw_withdrawal_data,
# facility_id = fid,
# facility_label = fname,
# fu_start = 21,
# fu_end = 35,
# fu_vstart = 28,
# fu_vend = 32)
# generate_word_report(day28fu_dir, "fu_weekly_log.Rmd", paste0(fid, "_", fname, "_timci_day28_fu_weekly_log"), params)
# }
}
}
#' Generate follow-up log (TIMCI-specific function)
#'
#' Generate a list of participants to be called in a time window after baseline between wmin and wmax
#' @param pii dataframe containing personally identifiable data
#' @param fudf dataframe containing the processed follow-up data
#' @param wmin numerical, start of the follow-up period (in days)
#' @param wmax numerical, end of the follow-up period (in days)
#' @param vwmin numerical, start of the follow-up period valid for the statistical analysis (in days)
#' @param vwmax numerical, end of the follow-up period valid for the statistical analysis (in days)
#' @param ext boolean, allow to export more information
#' @param deidentify boolean, enable the generation of a format with personally identifiable information (PII) for upload on ODK Central (if set to FALSE) or of a deidentified dataframe (if set to TRUE)
#' @return This function returns a dataframe.
#' @export
#' @import magrittr dplyr
generate_fu_log <- function(pii,
fudf,
wmin,
wmax,
vwmin,
vwmax,
ext = FALSE,
deidentify = FALSE) {
fu_log <- pii
fu_log$min_date <- as.Date(fu_log$date_visit) + wmin
fu_log$max_date <- as.Date(fu_log$date_visit) + wmax
if(Sys.getenv('TIMCI_COUNTRY') != 'Tanzania'){
fu_log$label <- paste(fu_log$fs_name, fu_log$ls_name)
} else{
fu_log$label <- paste(fu_log$fs_name, fu_log$ms_name, fu_log$ls_name)
}
fu_log$caregiver <- paste(fu_log$cg_fs_name, fu_log$cg_ls_name)
fu_log$mother <- paste(fu_log$mother_fs_name, fu_log$mother_ls_name)
fu_log$sex <- ifelse(fu_log$sex == 1, "male", ifelse(fu_log$sex == 2, "female", "other"))
# Exclude children who already underwent successful follow-up
if (!is.null(fudf)) {
if (nrow(fudf) > 0) {
fudf <- fudf %>% dplyr::filter(proceed == 1)
fu_log <- fu_log[!(fu_log$child_id %in% fudf$"a1-pid"),]
}
}
# Exclude children who are outside of the follow-up window period
fu_log <- fu_log[fu_log$min_date <= Sys.Date() & fu_log$max_date >= Sys.Date(),]
# Order columns
if (deidentify) {
col_order <- c('fid',
'child_id',
'sex',
'date_visit',
'main_cg_lbl')
} else{
col_order <- c('fid',
'child_id',
'label',
'sex',
'date_visit',
'caregiver',
'main_cg_lbl',
'mother',
'phone_nb',
'phone_nb2',
'location',
'cmty_contact')
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania") {
col_order <- c(col_order, 'physical_fu_instructions')
}
}
# If ext is set to TRUE, also add the district and facility name
if (ext) {
col_order <- c('district',
'facility',
col_order)
}
# Select only a subset of columns
fu_log <- fu_log[, col_order]
# Order entries by date
fu_log <- fu_log %>%
dplyr::arrange(fid, date_visit = as.Date(date_visit, "%Y-%m-%d"))
# Add valid window in export
if ( is_not_empty(fu_log) ) {
fu_log$date_visit <- paste0("From ", as.Date(fu_log$'date_visit', "%Y-%m-%d") + vwmin, " to ", as.Date(fu_log$'date_visit', "%Y-%m-%d") + vwmax, " [enrolled on ", fu_log$date_visit, "]")
}
# Add a first generic row
# If ext is set to TRUE, do not add
if (!ext & !deidentify) {
first_row <- data.frame('fid' = 'F0000',
'child_id' = 'X-F0000-P0000',
'label' = 'CHILD NAME',
'sex' = 'SEX',
'date_visit' = 'VALID WINDOW [ENROLMENT DATE]',
'caregiver' = 'CAREGIVER NAME',
'main_cg_lbl' = 'RELATIONSHIP',
'mother' = 'MOTHER NAME',
'phone_nb' = 'PHONE NB 1',
'phone_nb2' = 'PHONE NB 2',
'location' = 'LOCATION',
'cmty_contact' = 'CONTACT')
if (Sys.getenv('TIMCI_COUNTRY') == "Tanzania") {
first_row$physical_fu_instructions <- 'INSTRUCTIONS'
}
fu_log <- rbind(first_row,
fu_log)
}
if (!deidentify) {
fu_log <- fu_log %>% dplyr::rename('name' = 'child_id',
'enroldate' = 'date_visit',
'relationship' = 'main_cg_lbl',
'phonenb1' = 'phone_nb',
'phonenb2' = 'phone_nb2',
'contact' = 'cmty_contact')
}
fu_log
}
#' Generate follow-up log CSV for upload on ODK Central (TIMCI-specific function)
#'
#' Generate a list of participants to be called in a time window after baseline between wmin and wmax
#' @param pii dataframe containing personally identifiable data
#' @param fudf dataframe containing the processed follow-up data
#' @param wmin numerical, start of the follow-up period (in days)
#' @param wmax numerical, end of the follow-up period (in days)
#' @param vwmin numerical, start of the follow-up period valid for the statistical analysis (in days)
#' @param vwmax numerical, end of the follow-up period valid for the statistical analysis (in days)
#' @return This function returns a dataframe.
#' @export
#' @import magrittr dplyr
generate_fu_log_csv <- function(pii,
fudf,
wmin,
wmax,
vwmin,
vwmax) {
fu_log <- pii
fu_log$min_date <- as.Date(fu_log$date_visit) + wmin
fu_log$max_date <- as.Date(fu_log$date_visit) + wmax
if(Sys.getenv('TIMCI_COUNTRY') != 'Tanzania'){
fu_log$child_name <- paste(fu_log$fs_name, fu_log$ls_name)
fu_log$label <- paste0(fu_log$child_name,
" (",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmin,
" to ",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmax,
" )")
}
else{
fu_log$child_name <- paste(fu_log$fs_name, fu_log$ms_name, fu_log$ls_name)
fu_log$label <- paste0(fu_log$child_name,
" (",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmin,
" to ",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmax,
" ) - ",
fu_log$ra_name)
}
fu_log$caregiver <- paste(fu_log$cg_fs_name, fu_log$cg_ls_name)
fu_log$mother <- paste(fu_log$mother_fs_name, fu_log$mother_ls_name)
fu_log$sex <- ifelse(fu_log$sex == 1, "male", ifelse(fu_log$sex == 2, "female", "other"))
# Exclude children who already underwent successful follow-up
if (!is.null(fudf)) {
if (nrow(fudf) > 0) {
fudf <- fudf %>% dplyr::filter(proceed == 1)
fu_log <- fu_log[!(fu_log$child_id %in% fudf$"a1-pid"),]
}
}
# Exclude children who are outside of the follow-up window period
fu_log <- fu_log[fu_log$min_date <= Sys.Date() & fu_log$max_date >= Sys.Date(),]
if (nrow(fu_log) > 0) {
fu_log$population <- "all"
fu_log$type <- "1"
if(Sys.getenv('TIMCI_COUNTRY') != 'Tanzania'){
fu_log$physical_fu_instructions <- ""
}
# Order columns
col_order <- c('fid',
'device_id',
'child_id',
'child_name',
'sex',
'date_visit',
'caregiver',
'main_cg_lbl',
'mother',
'phone_nb',
'phone_nb2',
'location',
'physical_fu_instructions',
'cmty_contact',
'label',
'sys_submit_id',
'population',
'type')
fu_log <- fu_log[, col_order]
# Order entries by date
fu_log <- fu_log %>%
dplyr::arrange(fid, sys_submit_id, date_visit = as.Date(date_visit, "%Y-%m-%d"))
# Add a first generic row
fu_log <- rbind(data.frame('fid' = 'F0000',
'device_id' = 'none',
'child_id' = 'X-F0000-P0000',
'child_name' = 'CHILD NAME',
'sex' = 'SEX',
'date_visit' = 'ENROLMENT DATE',
'caregiver' = 'CAREGIVER NAME',
'main_cg_lbl' = 'RELATIONSHIP',
'mother' = 'MOTHER NAME',
'phone_nb' = 'PHONE NB 1',
'phone_nb2' = 'PHONE NB 2',
'location' = 'LOCATION',
'physical_fu_instructions' = 'INSTRUCTIONS',
'cmty_contact' = 'CONTACT',
'label' = 'CHILD NAME (VALID WINDOW)',
'sys_submit_id' = 'SUBMITTER',
'population' = 'POPULATION',
'type' = 'MANUAL'),
fu_log)
fu_log <- fu_log %>% dplyr::rename('name' = 'child_id',
'enroldate' = 'date_visit',
'relationship' = 'main_cg_lbl',
'phonenb1' = 'phone_nb',
'phonenb2' = 'phone_nb2',
'contact' = 'cmty_contact',
'submitter' = 'sys_submit_id')
}
fu_log
}
#' Generate physical follow-up log CSV for upload on ODK Central (TIMCI-specific function, Tanzania only)
#'
#' Generate a list of participants to be called in a time window after baseline between wmin and wmax
#' @param pii dataframe containing personally identifiable data
#' @param fudf dataframe containing the processed follow-up data
#' @param wmin_nophone numerical, start of the follow-up period if the caregiver has no phone number (in days)
#' @param wmin_phone numerical, start of the follow-up period if the caregiver has a phone number (in days)
#' @param wmax numerical, end of the follow-up period (in days)
#' @param vwmin numerical, start of the follow-up period valid for the statistical analysis (in days)
#' @param vwmax numerical, end of the follow-up period valid for the statistical analysis (in days)
#' @return This function returns a dataframe.
#' @export
#' @import magrittr dplyr
generate_physical_fu_log_csv <- function(pii,
fudf,
wmin_nophone,
wmin_phone,
wmax,
vwmin,
vwmax) {
fu_log <- pii
fu_log$min_nophone_date <- as.Date(fu_log$date_visit) + wmin_nophone
fu_log$min_phone_date <- as.Date(fu_log$date_visit) + wmin_phone
fu_log$max_date <- as.Date(fu_log$date_visit) + wmax
fu_log$child_name <- paste(fu_log$fs_name, fu_log$ms_name, fu_log$ls_name)
fu_log$caregiver <- paste(fu_log$cg_fs_name, fu_log$cg_ls_name)
fu_log$mother <- paste(fu_log$mother_fs_name, fu_log$mother_ls_name)
fu_log$sex <- ifelse(fu_log$sex == 1, "male", ifelse(fu_log$sex == 2, "female", "other"))
fu_log$label <- paste0(fu_log$child_name,
" (",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmin,
" to ",
as.Date(fu_log$date_visit, "%Y-%m-%d") + vwmax,
" )")
# Exclude children who already underwent successful follow-up
if (!is.null(fudf)) {
if (nrow(fudf) > 0) {
fudf <- fudf %>% dplyr::filter(proceed == 1)
fu_log <- fu_log[!(fu_log$child_id %in% fudf$"a1-pid"),]
}
}
#########################################################################
# Select children for whom the caregiver did not provide a phone number #
#########################################################################
fu_log_nophone <- fu_log %>%
dplyr::filter(phone_nb_avail == 0) %>% # Include only children for whom the caregiver did not provide a phone number
dplyr::filter(min_nophone_date <= Sys.Date() & max_date >= Sys.Date()) # Exclude children who are outside of the physical follow-up window period
if (nrow(fu_log_nophone)){
fu_log_nophone$population <- "no phone number available"
fu_log_nophone$type <- "2"
} else {
fu_log_nophone$population <- character()
fu_log_nophone$type <- character()
}
#######################################################################################################
# Select children for whom the caregiver provided a phone number and agreed with a physical follow-up #
#######################################################################################################
# Add a constraints on having at least 1 recorded failed attempt
fu_log_phone <- fu_log %>%
dplyr::filter(phone_nb_avail == 1 & physical_fu == 1) %>% # Include only children for whom the caregiver provided a phone number and agreed with physical follow-up
dplyr::filter(min_phone_date <= Sys.Date() & max_date >= Sys.Date()) # Exclude children who are outside of the physical follow-up window period
if (nrow(fu_log_phone)) {
fu_log_phone$population <- "phone number available"
fu_log_phone$type <- "2"
} else {
fu_log_phone$population <- character()
fu_log_phone$type <- character()
}
#########################################################
# Combine both selections of children in one single log #
#########################################################
#if (!is.null(fu_log_nophone)) {
# if (nrow(fu_log_nophone) > 0) {
# if (!is.null(fu_log_phone)) {
# fu_log <- dplyr::bind_rows(fu_log_nophone, fu_log_phone)
# } else{
# fu_log <- fu_log_nophone
# }
# }
#}
fu_log <- dplyr::bind_rows(fu_log_nophone,fu_log_phone)
# Order columns
col_order <- c('fid',
'device_id',
'child_id',
'child_name',
'sex',
'date_visit',
'caregiver',
'main_cg_lbl',
'mother',
'phone_nb',
'phone_nb2',
'location',
'location_lvl1',
'location_lvl2',
'location_lvl3',
'location_lvl4',
'physical_fu_instructions',
'cg_common_name',
'cmty_contact',
'label',
'sys_submit_id',
'population',
'type')
fu_log <- fu_log[, col_order]
# Order entries by date
fu_log <- fu_log %>%
dplyr::arrange(date_visit = as.Date(date_visit, "%Y-%m-%d"))
# Add a first generic row
fu_log <- rbind(data.frame('fid' = 'F0000',
'device_id' = 'none',
'child_id' = 'X-F0000-P0000',
'child_name' = 'CHILD NAME',
'sex' = 'SEX',
'date_visit' = 'ENROLMENT DATE',
'caregiver' = 'CAREGIVER NAME',
'main_cg_lbl' = 'RELATIONSHIP',
'mother' = 'MOTHER NAME',
'phone_nb' = 'PHONE NB 1',
'phone_nb2' = 'PHONE NB 2',
'location' = 'LOCATION',
'location_lvl1' = 'LOCATION_DIVISION',
'location_lvl2' = 'LOCATION_WARD',
'location_lvl3' = 'LOCATION_LVL3',
'location_lvl4' = 'LOCATION_LVL4',
'cmty_contact' = 'CONTACT',
'physical_fu_instructions' = 'INSTRUCTIONS',
'cg_common_name' = 'CAREGIVER COMMON NAME',
'label' = 'CHILD NAME (VALID WINDOW)',
'sys_submit_id' = 'SUBMITTER',
'population' = 'POPULATION',
'type' = 'MANUAL'),
fu_log)
fu_log %>% dplyr::rename('name' = 'child_id',
'enroldate' = 'date_visit',
'relationship' = 'main_cg_lbl',
'phonenb1' = 'phone_nb',
'phonenb2' = 'phone_nb2',
'contact' = 'cmty_contact',
'instructions' = 'physical_fu_instructions',
'submitter' = 'sys_submit_id')
}
#' Generate lost to follow-up log (TIMCI-specific function)
#'
#' Generate a list of participants who are lost to follow-up after wmax
#' @param df dataframe
#' @param fudf dataframe containing the processed follow-up data
#' @param end_date numerical, end of the follow-up period (in days)
#' @param raw boolean flag for selecting raw (TRUE) or processed (FALSE) data
#' @return This function returns a dataframe.
#' @export
#' @import magrittr dplyr
generate_ltfu_log <- function(df,
fudf,
end_date,
raw = TRUE) {
ltfu_log <- df
ltfu_log$date_maxfu <- as.Date(ltfu_log$date_visit) + end_date
# If raw data
if (raw) {
fudf <- fudf %>%
dplyr::rename(child_id = "a1-pid",
proceed_day7 = proceed)
}
# Exclude children who already underwent successful follow-up
if (!is.null(fudf)) {
if (nrow(fudf) > 0) {
suc_fudf <- fudf %>%
dplyr::filter(proceed_day7 == 1)
un_fudf <- fudf %>%
dplyr::filter(proceed_day7 == 0)
# Count the number of unsuccessful attempts
attempt_df <- un_fudf %>%
group_by(child_id) %>%
count
ltfu_log <- ltfu_log[!(ltfu_log$child_id %in% suc_fudf$child_id),]
ltfu_log <- merge(ltfu_log, attempt_df, by = 'child_id', all.x = TRUE) %>%
dplyr::rename(fu_attempts = n)
}
}
# If fu_attempts does not exist, create a column and fill out values with zeroes
if(!("fu_attempts" %in% colnames(ltfu_log))) {
ltfu_log$fu_attempts <- 0
}
# Exclude children who are still within or before the follow-up window period
ltfu_log <- ltfu_log[ltfu_log$date_maxfu < Sys.Date(),]
# Order entries by date
ltfu_log <- ltfu_log %>%
dplyr::arrange(date_visit = as.Date(date_visit, "%Y-%m-%d"))
# Order columns
ltfu_log$date_maxfu <- as.Date(ltfu_log$date_maxfu, "%Y-%m-%d")
if(Sys.getenv('TIMCI_COUNTRY') == 'India'){
col_order <- c('child_id',
'date_visit',
'date_maxfu',
'fu_attempts',
'age_mo',
'yg_infant',
'who_age_ctg',
'fid',
'facility',
'district',
'referral_cg',
'referral_hf',
'urg_referral_hf',
'ref_facility',
'ref_facility_oth',
'device_id',
'sys_submit_id')
} else{
col_order <- c('child_id',
'date_visit',
'date_maxfu',
'fu_attempts',
'age_mo',
'yg_infant',
'who_age_ctg',
'fid',
'facility',
'district',
'phone_nb_avail',
'referral_cg',
'referral_hf',
'urg_referral_hf',
'ref_facility',
'ref_facility_oth',
'device_id',
'sys_submit_id')
}
ltfu_log <- ltfu_log[, col_order]
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.