Using package functions and local data to calculate TB incidence rates in England

Sam Abbott

2019-07-11

Clean and Create Variables from the Enhanced Tuberculosis Surveillance Data Set

library(tbinenglanddataclean)

## rebuild existing data sets
rebuild <- FALSE

## Analysis running in docker or locally? if local set ~ if in docker set ..
run_envir <- ".."

if (!file.exists(paste0(run_envir, "/data/tb_data/tbinenglanddataclean/clean_ets_2016.rds")) || rebuild) {
  clean_munge_ets_2016(data_path = paste0(run_envir, "/data/tb_data/ETS/ETS_2016_extract/SAbbott_BCG dataset_version2_final140217.dta"),
                     return = FALSE,
                     save = TRUE,
                     save_name = "clean_ets_2016",
                     save_path = paste0(run_envir, "/data/tb_data/tbinenglanddataclean"),
                     save_format = c("rds", "csv"),
                     verbose = TRUE)

}

Clean the Office of National Statistics Demographic Data to be used as a Reference Data set

if (!file.exists(paste0(run_envir,"/data/tb_data/tbinenglanddataclean/E_demo_2000_2015.rds")) || rebuild) {
  clean_demographics_uk(data_path = paste0(run_envir, "/data/tb_data/UK_demographics"),
                      demo_2000 = "UK_2000_age.csv",
                      demo_2001_2015 = "UK_2001_2015_age.csv",
                      countries = c("E"),
                      return = FALSE,
                      save = TRUE,
                      save_name = "E_demo_2000_2015",
                      save_path = paste0(run_envir, "/data/tb_data/tbinenglanddataclean"),
                      save_format = c("rds", "csv"),
                      verbose = TRUE)
}

Clean and Combine the Labour Force Survey Data set

if (!file.exists(paste0(run_envir, "/data/tb_data/tbinenglanddataclean/formatted_LFS_2000_2016.rds")) || rebuild) {
clean_labour_force_survey(data_path = paste0(run_envir, "/data/tb_data/LFS"),
                          years = 2000:2016,
                          years_var = list( '2000' = c('age', 'sex', 'cry', 'govtof', 'pwt07'),
                                            '2001' = c('age', 'sex', 'cry01', 'country', 'pwt07'),
                                            '2002' = c('AGE', 'SEX', 'CRY01', 'COUNTRY', 'PWT14'),
                                            '2003' = c('AGE', 'SEX', 'CRY01', 'COUNTRY', 'PWT14'),
                                            '2004' = c('AGE', 'SEX', 'CRY01', 'COUNTRY', 'PWT14'),
                                            '2005' = c('AGE', 'SEX', 'CRY01', 'COUNTRY', 'PWT14'),
                                            '2006' = c('AGE', 'SEX', 'CRY01', 'COUNTRY', 'PWT14'),
                                            '2007' = c('AGE', 'SEX', 'CRY01', 'COUNTRY', 'PWT14'),
                                            '2008' = c('AGE', 'SEX', 'CRY01', 'COUNTRY', 'PWT14'),
                                            '2009' = c('AGE', 'SEX', 'CRY01', 'COUNTRY', 'PWT14'),
                                            '2010' = c('AGE', 'SEX', 'CRY01', 'COUNTRY', 'PWT14'),
                                            '2011' = c('AGE', 'SEX', 'CRY01', 'COUNTRY', 'PWT14'),
                                            '2012' = c('AGE', 'SEX', 'CRY12', 'COUNTRY', 'PWT14'),
                                            '2013' = c('AGE', 'SEX', 'CRY12', 'COUNTRY', 'PWT16'),
                                            '2014' = c('AGE', 'SEX', 'CRY12', 'COUNTRY', 'PWT16'),
                                            '2015' = c('AGE', 'SEX', 'CRY12', 'COUNTRY', 'PWT16'),
                                            '2016' = c('AGE', 'SEX', 'CRY12', 'COUNTRY', 'PWT16')),
                          return = FALSE,
                          save = TRUE,
                          save_name = "formatted_LFS_2000_2016",
                          save_path = paste0(run_envir, "/data/tb_data/tbinenglanddataclean"),
                          save_format = c("rds", "csv"),
                          verbose = TRUE)
}

Combine the LFS and ONS datasets; check for issues

if (!file.exists(paste0(run_envir, "/data/tbinenglanddataclean/E_ons_lfs_2000_2016.rds")) || rebuild) {
combine_ons_with_lfs(data_path = paste0(run_envir, "/data/tb_data/tbinenglanddataclean"),
                     ons_name = "E_demo_2000_2015.rds",
                     lfs_name = "formatted_LFS_2000_2016.rds",
                     countries = "England",
                     return = FALSE,
                     save = TRUE,
                     save_name = "E_ons_lfs_2000_2016",
                     save_path = paste0(run_envir, "/data/tb_data/tbinenglanddataclean"),
                     save_format = c("rds", "csv"),
                     verbose = TRUE) 
}

Clean and Munge Changes in English Population

if (!file.exists(paste0(run_envir, "/data/tb_data/tbinenglanddataclean/england_births.rds")) || rebuild) {
  clean_and_munge_england_births(birth_path = paste0(run_envir, "/data/tb_data/UK_demographics/annual_reference_table.xls"),
                               projected_birth_path = paste0(run_envir, "/data/tb_data/UK_demographics/england_population_projections.xls"),
                               return = FALSE,
                               save = TRUE,
                               save_name = "england_births",
                               save_path = paste0(run_envir, "/data/tb_data/tbinenglanddataclean"),
                               save_format = c("rds", "csv"),
                               verbose = TRUE)
}

if (!file.exists(paste0(run_envir, "/data/tb_data/tbinenglanddataclean/england_mortality_rate.rds")) || rebuild) {
clean_and_munge_england_life_expectancy(life_tables_path = paste0(run_envir, "/data/tb_data/UK_demographics/england_life_tables.xls"),
                                                sheets = paste0(seq(2013,1998, -1), "-", seq(2015,2000, -1)),
                                                return = FALSE,
                                                save = TRUE,
                                                save_name = "england_mortality_rates",
                                                save_path = paste0(run_envir, "/data/tb_data/tbinenglanddataclean"),
                                                save_format = c("rds", "csv"),
                                                verbose = TRUE)
}

Calculate TB Incidence Rates for England

if (!file.exists(paste0(run_envir, "/data/tb_data/tbinenglanddataclean/cases_demo_incidence.rds")) || rebuild) {
  calculate_incidence_ets_lfs_ons(data_path = paste0(run_envir, "/data/tb_data/tbinenglanddataclean"),
                                ets_name = "clean_ets_2016.rds",
                                demo_name = "E_ons_lfs_2000_2016.rds",
                                return = FALSE,
                                save = TRUE,
                                incidence_name = "incidence",
                                grouped_incidence_name = "age_grouped_incidence",
                                condensed_grouped_incidence_name = "condensed_age_group_incidence",
                                cases_demo_incidence_name = "cases_demo_incidence",
                                save_path = paste0(run_envir, "/data/tb_data/tbinenglanddataclean"),
                                save_format = c("rds", "csv"),
                                verbose = TRUE)
}