transfer from old repo

2 Longterm/260315/_targets.R

@@ -0,0 +1,285 @@
+# Created by use_targets().
+# Follow the comments below to fill in this target script.
+# Then follow the manual to check and run the pipeline:
+#   https://books.ropensci.org/targets/walkthrough.html#inspect-the-pipeline
+
+# Load packages required to define the pipeline:
+library(targets)
+library(tarchetypes) # Load other packages as needed.
+
+# Set target options:
+tar_option_set(
+  seed = 142,
+  packages = c("tibble"), # packages that your targets need to run
+  format = "qs" # , # Optionally set the default storage format. qs is fast.
+  #
+  # For distributed computing in tar_make(), supply a {crew} controller
+  # as discussed at https://books.ropensci.org/targets/crew.html.
+  # Choose a controller that suits your needs. For example, the following
+  # sets a controller with 2 workers which will run as local R processes:
+  #
+  #   controller = crew::crew_controller_local(workers = 2)
+  #
+  # Alternatively, if you want workers to run on a high-performance computing
+  # cluster, select a controller from the {crew.cluster} package. The following
+  # example is a controller for Sun Grid Engine (SGE).
+  #
+  #   controller = crew.cluster::crew_controller_sge(
+  #     workers = 50,
+  #     # Many clusters install R as an environment module, and you can load it
+  #     # with the script_lines argument. To select a specific verison of R,
+  #     # you may need to include a version string, e.g. "module load R/4.3.0".
+  #     # Check with your system administrator if you are unsure.
+  #     script_lines = "module load R"
+  #   )
+  #
+  # Set other options as needed.
+)
+
+# tar_make_clustermq() is an older (pre-{crew}) way to do distributed computing
+# in {targets}, and its configuration for your machine is below.
+options(clustermq.scheduler = "multiprocess")
+
+# tar_make_future() is an older (pre-{crew}) way to do distributed computing
+# in {targets}, and its configuration for your machine is below.
+future::plan(future.callr::callr)
+
+# Run the R scripts in the R/ folder with your custom functions:
+tar_source()
+source(here::here("R/glmnet-reg.R")) # Source other scripts as needed.
+
+# Replace the target list below with your own:
+list(
+  tar_target(
+    name = pop_df,
+    command = get_clinical()
+  ),
+  tar_target(
+    name = reg_list,
+    command = get_reg_ls()
+  ),
+  tar_target(
+    name = df_deaths,
+    command = get_deaths(reg_list)
+  ),
+  tar_target(
+    name = df_events,
+    command = get_events(reg_list)
+  ),
+  tar_target(
+    name = df_events_deaths,
+    command = merge_events(list(events = df_events, deaths = df_deaths, clinical = pop_df))
+  ),
+  tar_target(
+    name = df_treated,
+    command = get_treated(reg_list)
+  ),
+  tar_target(
+    name = df_dst,
+    command = get_dst(reg_list, pop_df)
+  ),
+  tar_target(
+    name = list_filtered,
+    command = list(all_events = df_events_deaths, clinical = pop_df, dst = df_dst)
+  ),
+  tar_target(
+    name = df_all_data,
+    command = collectall(list_filtered)
+  ),
+  tar_target(
+    name = df_all_data_formatted,
+    command = data_formatting(df_all_data)
+  ),
+  tar_target(
+    name = ls_all_events,
+    command = all_events(list(events = df_events, deaths = df_deaths, clinical = pop_df))
+  ),
+  tar_target(
+    name = df_event_data,
+    command = events_ready(df_all_data_formatted)
+  ),
+  tar_target(
+    name = df_talos_data,
+    command = talos_ready(df_all_data_formatted)
+  ),
+  tar_target(
+    name = df_talos_data_imp,
+    command = talos_imp(df_talos_data)
+  ),
+  tar_target(
+    name = tbl_events_summary,
+    command = events_tblone(df_event_data)
+  ),
+  tar_target(
+    name = tbl_events_cox_regression,
+    command = show_table_regression(df_event_data, use.mice = FALSE)
+  ),
+  tar_target(
+    name = tbl_events_cox_regression_uv,
+    command = uv_cox_table(df_event_data)
+  ),
+  tar_target(
+    name = df_event_data_small,
+    command = events_ready_small(df_all_data_formatted)
+  ),
+  tar_target(
+    name = tbl_events_summary_small,
+    command = events_tblone(df_event_data_small)
+  ),
+  tar_target(
+    name = tbl_events_cox_regression_small,
+    command = show_table_regression(df_event_data_small, use.mice = FALSE)
+  ),
+  tar_target(
+    name = df_events_mids,
+    command = events_dataset(df_event_data, impute = TRUE)
+  ),
+  tar_target(
+    name = df_events_complete,
+    command = complete_preds_data(df_event_data)
+  ),
+  tar_target(
+    name = tbl_events_mids_cox_regression,
+    command = show_table_regression(df_event_data, use.mice = TRUE)
+  ),
+  tar_target(
+    name = plot_events_survival_smooth,
+    command = df_event_data |> events_dataset(impute = FALSE) |> cox_regression(include_formula=TRUE) |> plot_survival_smooth()
+  )#,
+  # tar_target(
+  #   name = plot_events_survival_smooth_mids,
+  #   command = df_events_mids |> cox_regression() |> plot_survival_smooth()
+  # ),
+  # tar_target(
+  #   name = df_pred_data,
+  #   command = prediction_ready(df_all_data_formatted)
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary,
+  #   command = preds_tblone(df_pred_data)
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_true,
+  #   command = df_pred_data |>
+  #     dplyr::select(pase_0, pase_4, soc_status_nowork, fam_indk_hl, edu_level_hl) |>
+  #     true_pred_sum_plot()
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_exp,
+  #   command = df_pred_data |>
+  #     dplyr::select(pase_0, pase_4, soc_status_nowork, fam_indk_hl, edu_level_hl) |>
+  #     preds_tblone()
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_exp_sex,
+  #   command = df_pred_data |>
+  #     dplyr::select(reg_female, soc_status_nowork, fam_indk_hl, edu_level_hl) |>
+  #     summary_tblone()
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_exp_pase,
+  #   command = df_pred_data |> sum_pase_tables()
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_exp_missing_edu,
+  #   command = df_pred_data |>
+  #     dplyr::select(age,reg_female, reg_trombolyse, reg_trombektomi, reg_hyperten, reg_diabetes, nihss_0, pase_0, pase_4, soc_status_nowork, fam_indk_hl, edu_level_hl) |>
+  #     who_is_missing(var="edu_level_hl")
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_exp_missing_bmi,
+  #   command = df_pred_data |>
+  #     dplyr::select(age,reg_female, reg_trombolyse, reg_trombektomi, reg_hyperten, reg_diabetes, nihss_0, pase_0, pase_4, reg_bmi, soc_status_nowork, fam_indk_hl, edu_level_hl) |>
+  #     who_is_missing(var="reg_bmi")
+  # ),
+  # tar_target(
+  #   name = ls_pred_models,
+  #   command = pred_models(df_pred_data,auto.l = TRUE,weighted = FALSE)
+  # ),
+  # tar_target(
+  #   name = ls_pred_models_anyupdown,
+  #   command = pred_models(df_pred_data,auto.l = TRUE,weighted = FALSE,split.type="anyupdown")
+  # ),
+  # tar_target(
+  #   name = ls_pred_models_relupdown_20,
+  #   command = pred_models(df_pred_data,auto.l = TRUE,weighted = FALSE,split.type="relupdown",rel.bin=20)
+  # ),
+  # tar_target(
+  #   name = ls_pred_models_relupdown_50,
+  #   command = pred_models(df_pred_data,auto.l = TRUE,weighted = FALSE,split.type="relupdown",rel.bin=50)
+  # ),
+  # tar_target(
+  #   name = df_pred_mids,
+  #   command = fun_impute(data = df_pred_data, outcome.vars = c("pase_0", "pase_4"))
+  # ),
+  # tar_target(
+  #   name = ls_pred_mids_reg,
+  #   command = mids_regularisation(df_pred_mids)
+  # ),
+  # tar_target(
+  #   name = ls_pred_mids_reg_anyupdown,
+  #   command = mids_regularisation(df_pred_mids,split.type="anyupdown")
+  # ),
+  # tar_target(
+  #   name = ls_pred_mids_reg_relupdown_20,
+  #   command = mids_regularisation(df_pred_mids,split.type="relupdown",rel.bin=20)
+  # ),
+  # tar_target(
+  #   name = ls_pred_mids_reg_relupdown_50,
+  #   command = mids_regularisation(df_pred_mids,split.type="relupdown",rel.bin=50)
+  # ),
+  # tar_target(
+  #   name = ls_pred_summary,
+  #   command = multi_summary(ls_pred_models)
+  # ),
+  # tar_target(
+  #   name = ls_pred_mids_summary,
+  #   command = multi_summary(ls_pred_mids_reg)
+  # ),
+  # tar_target(
+  #   name = tbl_preds_log_reg,
+  #   command = pred_log_reg(df_pred_data)
+  # ),
+  # tar_target(
+  #   name = tbl_preds_lin_reg,
+  #   command = pred_lin_reg(df_pred_data)
+  # ),
+  # tar_target(
+  #   name = tbl_preds_lin_imp_reg,
+  #   command = pred_lin_reg(df_pred_mids)
+  # ),
+  # tar_target(
+  #   name = list_pred_clusters,
+  #   command = get_clusters(df_events_complete,rm.out = TRUE)
+  # ),
+  # tar_target(
+  #   name = list_pred_clusters_seq,
+  #   command = get_clusters(df_events_complete,n.cl=4)
+  # )#,
+  # tar_target(
+  #   name = list_df_multi_grouping,
+  #   command = multi_grouping_df_list(df_all_data_formatted,args.list=df_mega_list())
+  # ),
+  # tar_target(
+  #   name = list_multi_group_results,
+  #   command = multi_results_list(list_df_multi_grouping)
+  # ),
+  # tar_target(
+  #   name = list_multi_group_results_direct,
+  #   command = df_all_data_formatted |> multi_grouping_df_list(args.list=df_mega_list()) |> multi_cox_performance_test()
+  # )
+  
+  
+  #,
+  # tar_quarto(
+  #   name = report,
+  #   path = "index.qmd"
+  # ),
+  # tar_target(
+  #   name = pa_change_export_files,
+  #   command = quarto::quarto_render(here::here("doc/pa_change.qmd"), output_format = "docx")
+  # )
+)
+
+## TODO
+## - modify pipeline to only perform imputation once and have the rest use this single object.

2 Longterm/260315/alternative_grouping.R

@@ -0,0 +1,268 @@
+source("R/functions.R")
+
+### FUNCTIONALISE --> DONE
+
+df <- targets::tar_read(df_all_data_formatted) |>
+  group_format(binning.fun = bin_anyupdown) |>
+  dplyr::mutate(pase_change = factor(pase_change, levels = c("high", "up", "low", "down")))
+
+
+df |> gtsummary::tbl_summary(by = pase_change)
+
+cox <- df |>
+  # dplyr::select(-pase_0,-pase_4,-pase_rel_dif) |>
+  cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change")
+
+
+df |>
+  # dplyr::select(-pase_0,-pase_4,-pase_rel_dif) |>
+  cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change") |>
+  tbl_regression_standard()
+
+df |>
+  # dplyr::select(-pase_0,-pase_4,-pase_rel_dif) |>
+  cox_regression() |>
+  ggsurvfit::survfit2() |>
+  ggsurvfit::ggsurvfit() + ggsurvfit::add_confidence_interval() +
+  ggsurvfit::scale_ggsurvfit() +
+  ggsurvfit::add_risktable()
+
+
+###
+
+
+data <- targets::tar_read(df_all_data_formatted)
+
+
+df <- targets::tar_read(df_all_data_formatted) |>
+  group_format(binning.fun = bin_relupdown, rel.bin = 50)
+
+
+df |> gtsummary::tbl_summary(by = pase_change)
+
+df |>
+  # dplyr::select(-pase_0,-pase_4,-pase_rel_dif) |>
+  cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change") |>
+  tbl_regression_standard()
+
+df |>
+  # dplyr::select(-pase_0,-pase_4,-pase_rel_dif) |>
+  cox_regression() |>
+  ggsurvfit::survfit2() |>
+  ggsurvfit::ggsurvfit() + ggsurvfit::add_confidence_interval() +
+  ggsurvfit::scale_ggsurvfit() +
+  ggsurvfit::add_risktable()
+
+
+
+targets::tar_read(list_df_multi_grouping) |> purrr::map(\(.x)  summary(.x[["pase_change"]]))
+
+
+
+
+
+## BMI out
+
+### Univariable models
+
+
+
+ls_df <- targets::tar_read(df_all_data_formatted) |> 
+  multi_grouping_df_list(args.list=df_mega_list(strategies=c(
+  # "bin_original", 
+  # "bin_anyupdown", 
+  "bin_relupdown", 
+  "bin_absupdown",
+  "bin_quantile", 
+  # "bin_clusterlcm",
+  "bin_percentage")),remove="pase_0")
+
+
+performance_overview_uni <- ls_df |> multi_cox_performance_test(all.vars = FALSE, use.strata = FALSE, outcome.var = "pase_change") 
+
+# performance_overview_uni
+
+# Comparing the best and the original
+
+
+cox_models_uni <- ls_df|>
+  purrr::map(\(.x){
+    .x |>
+      cox_regression(all.vars = FALSE, use.strata = FALSE, outcome.var = "pase_change")
+  }) 
+
+ls_df[c(pick_non_duplicated(performance_overview_uni,"Name","AIC_wt",1:3),"lowest_percentage_25")]|>
+  purrr::map(\(.x){
+    .x |> dplyr::select(pase_change) |> 
+      gtsummary::tbl_summary()
+  }) |> tbl_merged_named()
+
+
+lapply(best_models_uni, performance::model_performance) |> 
+  (\(.x){
+    dplyr::bind_cols(model=names(.x), dplyr::bind_rows(.x))
+  })()
+ 
+
+models_ls_uni <- best_models_uni |> lapply(\(.x){
+  .x |> gtsummary::tbl_regression(exponentiate=TRUE) |> gtsummary::bold_p()
+}) 
+
+
+models_ls_uni |> tbl_merged_named()
+
+
+## Multivariable models
+
+performance_overview_multi <- ls_df |> multi_cox_performance_test(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change")
+
+# performance_overview_multi
+
+# Comparing the best and the original
+# best_models_multi <- head(performance_overview_multi,10)
+
+cox_models_multi <- ls_df |>
+  purrr::map(\(.x){
+    .x |>
+      cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change")
+  }) 
+
+ls_df[c(pick_non_duplicated(performance_overview_multi,"Name","AIC_wt",1:3),"lowest_percentage_25")]|>
+  purrr::map(\(.x){
+    .x |> dplyr::select(pase_change) |> 
+      gtsummary::tbl_summary()
+  }) |> tbl_merged_named()
+
+lapply(best_models_multi, performance::model_performance) |> 
+  (\(.x){
+    dplyr::bind_cols(model=names(.x), dplyr::bind_rows(.x))
+  })()
+
+
+models_ls_multi <- best_models_multi |> lapply(\(.x){
+  .x |> gtsummary::tbl_regression(exponentiate=TRUE) |> gtsummary::bold_p()
+}) 
+
+
+models_ls_multi |> tbl_merged_named()
+
+
+
+## Handle for mids objects
+# targets::tar_read(df_events_mids)
+# targets::tar_read(df_event_data)
+
+ls_df_mids <- targets::tar_read(df_events_mids) |> multi_grouping_df_list(args.list=df_mega_list(strategies=c(
+  # "bin_original", 
+  # "bin_anyupdown", 
+  "bin_relupdown", 
+  "bin_absupdown",
+  "bin_quantile", 
+  # "bin_clusterlcm",
+  "bin_percentage")),remove="pase_0")
+
+cox_models_mids <- ls_df_mids|>
+  purrr::map(\(.x){
+    .x |>
+      cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change")
+  }) 
+
+performance_overview_mids <- lapply(cox_models_mids,mids_model_aic)|>
+  (\(.x){
+    dplyr::bind_cols(Name=names(.x), AIC=Reduce(c,.x))
+  })() |> dplyr::arrange(AIC) |> 
+  dplyr::mutate(rank=rank(AIC,ties.method = "min"))
+
+# performance_overview_mids|> head(20)
+
+# Comparing the best and the original
+
+best_models_mids <- head(performance_overview_mids,10)
+
+# best_models_mids <- cox_models_mids[c(pick_non_duplicated(performance_overview_mids,"Name","AIC",1:3),"quantile_change_6_2_ANY","lowest_percentage_25")]
+
+models_ls_mids <- best_models_mids |> lapply(\(.x){
+  .x |> gtsummary::tbl_regression(exponentiate=TRUE) |> gtsummary::bold_p() |> gtsummary::add_glance_source_note()
+}) 
+
+models_ls_mids |> tbl_merged_named()
+
+## Difference between mids-results
+# targets::tar_read(df_event_data)
+
+## NOTE on investigation
+# In the original grouping, PASE-cutting was performed based only on patients without prior events.
+# In the new multi-grouping strategies evaluation, PASE grouping is based on all TALOS-patients.
+# We already did export some data, so we should probably stick with this. On the other hand, 
+# this is probably a minor thing, and we should perform analyses based on all patients like intended.
+# MAYBE
+# 
+
+## 
+
+
+
+
+best_models <- list("uni"=performance_overview_uni,
+                    "multi"=performance_overview_multi,
+                    "mids"=performance_overview_mids) |> 
+  lapply(\(.x){
+    .x |> dplyr::select(Name,rank,AIC)
+  }) |> dplyr::bind_rows()
+
+models_overall_rank <- split(best_models,best_models$Name) |> 
+  purrr::imap(\(.x,.i){
+    tibble::tibble(name=.i,rank_sum=sum(.x$rank), median_AIC=median(.x$AIC))
+  })|> dplyr::bind_rows() |> 
+  dplyr::arrange(rank_sum)
+
+models_overall_rank[models_overall_rank$name %in% c("quantile_change_10_1","quantile_change_8_2","quantile_change_8_2_ANY"),]
+
+all_cox_models <- list(
+  "Univariable"=cox_models_uni,
+    "Multivariable"=cox_models_multi,
+    "Imputed multivariable"=cox_models_mids
+)
+
+names_best <- models_overall_rank$name[c(1:6)]
+
+best_sum_tables <- names_best|> 
+  lapply(\(.x){
+    ls_df[[.x]] |> 
+      dplyr::select(pase_change) |> gtsummary::tbl_summary()
+  }) |> 
+  setNames(glue::glue("{rank(models_overall_rank$rank_sum,ties.method = 'min')[c(1:6)]}_{names_best}")) 
+
+best_cox_tables <- names_best |> 
+  lapply(\(.x){
+    list("Group counts"=ls_df[[.x]] |> 
+        dplyr::select(pase_change) |> gtsummary::tbl_summary() |> fix_labels(),
+      all_cox_models |> 
+      lapply(\(.y){
+      .y[[.x]] |> tbl_regression_standard() |> gtsummary::modify_table_styling(columns=tidyselect::starts_with("p.value"),hide = TRUE) |> 
+          gtsummary::remove_row_type(variables=-pase_change,type="all")
+    })) |> purrr::list_flatten() |> 
+      tbl_merged_named()
+  }) |> 
+  setNames(glue::glue("{rank(models_overall_rank$rank_sum,ties.method = 'min')[c(1:6)]}_{names_best}"))
+
+## Counts also for multivariable analyses could be added as well
+
+best_stack <- best_cox_tables |> tbl_stack_named()
+
+best_stack |> gtsummary::as_gt() |> 
+  gt::gtsave(filename = here::here("out/sens_cox.docx"))
+
+## Checking on a specific model
+ls_df_fun[[24]] |>
+  # dplyr::select(-pase_0,-pase_4,-pase_rel_dif) |>
+  cox_regression() |>
+  ggsurvfit::survfit2() |>
+  ggsurvfit::ggsurvfit() + ggsurvfit::add_confidence_interval() +
+  ggsurvfit::scale_ggsurvfit() +
+  ggsurvfit::add_risktable()
+
+ls_df_fun[[24]] |>
+  # dplyr::select(-pase_0,-pase_4,-pase_rel_dif) |>
+  cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change") |>
+  tbl_regression_standard()

2 Longterm/260315/combined_script.R

@@ -0,0 +1,740 @@
+# Code from: cont_pase_sens.qmd
+targets::tar_config_set(store = here::here("_targets"))
+source(here::here("R/functions.R"))
+source(here::here("R/glmnet-reg.R"))
+library(targets)
+library(tidyverse)
+tbl_cont_0 <- targets::tar_read("df_all_data_formatted")|> 
+  dplyr::filter(!is.na(pase_0),!is.na(pase_4)) |> 
+  # pase_cutter(drop.nas = TRUE) |>
+  events_ready(v.groups=c("clin","lifestyle.events","ses", "assess.events","quartiles"))  |>
+  dplyr::select(dplyr::all_of(c("pase_0", "age", "reg_female", "nihss_0", "reg_trombolyse", 
+"reg_trombektomi", "rtreat_placebo", "reg_alone", "reg_smoker", 
+"reg_more_alc", "reg_hyperten", "reg_diabetes", "reg_atriefli", 
+"reg_ami", "soc_status_nowork", "fam_indk_hl", "edu_level_hl", 
+"who_4", "mdi_4", "mfi_gen_4", "mrs_4_above1", "time", "status"
+))) |>
+  (\(data){
+    # c("pase_0","pase_4") |> 
+    #   purrr::map(\(exp){
+        list("Univariable"=cox_regression(data=data,all.vars = FALSE, use.strata = FALSE, outcome.var = "pase_0"),
+             "Multivariable"=cox_regression(data=data,all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_0")) |> 
+          purrr::map(\(.x){
+            .x |> gtsummary::tbl_regression(exponentiate=TRUE) |> gtsummary::add_nevent()|> 
+  fix_labels()
+          }) |> 
+          tbl_merged_named()
+      # })
+    })()   |>
+      gtsummary::modify_table_body( ~ filter(.x, variable == "pase_0"))
+tbl_cont_4 <- targets::tar_read("df_all_data_formatted")|> 
+  dplyr::filter(!is.na(pase_0),!is.na(pase_4)) |> 
+  # pase_cutter(drop.nas = TRUE) |>
+  events_ready(v.groups=c("clin","lifestyle.events","ses", "assess.events","quartiles"))  |>
+  dplyr::select(dplyr::all_of(c("pase_4", "age", "reg_female", "nihss_0", "reg_trombolyse", 
+"reg_trombektomi", "rtreat_placebo", "reg_alone", "reg_smoker", 
+"reg_more_alc", "reg_hyperten", "reg_diabetes", "reg_atriefli", 
+"reg_ami", "soc_status_nowork", "fam_indk_hl", "edu_level_hl", 
+"who_4", "mdi_4", "mfi_gen_4", "mrs_4_above1", "time", "status"
+))) |>
+  (\(data){
+    # c("pase_0","pase_4") |> 
+    #   purrr::map(\(exp){
+        list("Univariable"=cox_regression(data=data,all.vars = FALSE, use.strata = FALSE, outcome.var = "pase_4"),
+             "Multivariable"=cox_regression(data=data,all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_4")) |> 
+          purrr::map(\(.x){
+            .x |> gtsummary::tbl_regression(exponentiate=TRUE) |> gtsummary::add_nevent()|> 
+  fix_labels()
+          }) |> 
+          tbl_merged_named()
+      # })
+    })()   |>
+      gtsummary::modify_table_body( ~ filter(.x, variable == "pase_4"))
+list("PASE 0"=tbl_cont_0,
+     "PASE 6"=tbl_cont_4) |> tbl_stack_named()
+
+
+
+# Code from: events_type_sens.qmd
+targets::tar_config_set(store = here::here("_targets"))
+source(here::here("R/functions.R"))
+source(here::here("R/glmnet-reg.R"))
+library(targets)
+library(tidyverse)
+df <- targets::tar_read(df_all_data_formatted) |> 
+  dplyr::mutate(
+  status.all=dplyr::if_else(
+    startsWith(as.character(event),"death"),"death","cve",missing = NA_character_)|> factor()
+  )|>
+  get_vars(vars.groups = c("clin","lifestyle.events","ses", "ssri")) |> 
+  dplyr::rename(status=status.all,
+                time=time.all)
+df$status |> summary()
+tbl <- df |>
+  pase_cutter(drop.nas = TRUE, drop.pase = TRUE) |>
+  (\(.x) {
+    list(
+      "death" = .x |> dplyr::mutate(status = dplyr::if_else(status == "death", 1, 0, 0)),
+      "cve" = .x |> dplyr::mutate(status = dplyr::if_else(status == "cve", 1, 0, 0))
+    )
+  })() |>  lapply(\(.x) {
+    ls <- list(
+      "Univariate" = .x |> dplyr::select(-rtreat) |>
+        standard_multi_cox_table(all.vars = FALSE) |> gtsummary::add_nevent(),
+      "Multivariate" = .x |> dplyr::select(-rtreat) |>
+        standard_multi_cox_table(all.vars = TRUE) |> gtsummary::add_nevent()
+    ) |>
+      purrr::map(gtsummary::bold_p) |>
+      
+      tbl_merged_named()
+    ls |>
+      gtsummary::modify_table_body( ~ filter(.x, variable == "pase_change"))
+  })   |>
+  gtsummary::tbl_stack(group_header = c("death","cve"))# |>
+  # gtsummary::as_gt() |>
+  # gt::gtsave(here::here("out/trial_strat_sens.docx"))
+     
+tbl
+
+names(tbl)
+
+
+
+# Code from: mrs_sensitivity.qmd
+source(here::here("R/functions.R"))
+ls_mrs0 <- targets::tar_read(df_all_data_formatted) |> 
+  dplyr::filter(mrs_0==0)|>
+    dplyr::filter(!is.na(pase_0),!is.na(pase_4)) |> 
+  events_ready() |> 
+  (\(.x){
+    list(
+      std=.x,
+      imp=.x |> events_dataset(impute = TRUE)
+    )
+  })() |> 
+  purrr::map(\(.x){
+    .x |>
+      pase_cutter(drop.pase = TRUE,drop.nas = TRUE)
+  }) 
+
+# targets::tar_read(df_all_data_formatted) |> 
+#   dplyr::filter(mrs_0==0) |>
+#     dplyr::filter(!is.na(pase_0),!is.na(pase_4)) |> events_ready() |>
+#   pase_cutter(drop.pase = TRUE,drop.nas = TRUE) |>
+#     cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change") |>
+#     tbl_regression_standard()
+
+ls <- list(
+  "Univariable"= ls_mrs0$std |>
+    dplyr::select(pase_change,dplyr::everything()) |> 
+    splitdf4uvcox(include=c("time","status")) |> 
+    purrr::map(\(.x){
+      .x |> tbl_regression_standard()
+    }) |> 
+    gtsummary::tbl_stack(),
+  "Multivariable" = ls_mrs0$std |>
+    cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change") |>
+    tbl_regression_standard() |> gtsummary::add_glance_source_note(),
+  "Multivariable Imputed"= ls_mrs0$imp |>
+    cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change") |>
+    tbl_regression_standard()
+) |> purrr::map(\(.x){
+  .x|> 
+    gtsummary::modify_table_styling(column = p.value,
+                                    hide=TRUE)
+}) |> tbl_merged_named()
+
+ls
+# |> gtsummary::as_gt() |>
+#   gt::gtsave(filename = here::here(glue::glue("out/sens_subset_mrs0_0.docx")))
+
+# ls_mrs0$std |>
+#   cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change") |> performance::check_model()
+targets::tar_read(df_all_data_formatted)|>
+    dplyr::filter(!is.na(pase_0),!is.na(pase_4)) |> 
+  dplyr::filter(mrs_0==0) |>
+  dplyr::select(pase_0,pase_4) |> 
+  summary()
+# ls_mrs0$std |> gtsummary::tbl_summary(by = pase_change) |> fix_labels()
+ls_mrs0_change <- targets::tar_read(df_all_data_formatted) |>
+  get_vars(vars.groups =c("clin","lifestyle.events","ses", "assess.events.pre")) |>
+  dplyr::filter(event.include) |>
+  dplyr::select(-tidyselect::all_of("event.include")) |> 
+  (\(.x){
+    list(
+      std=.x,
+      imp=.x |>
+      dplyr::select(-tidyselect::any_of("reg_bmi"))|>
+      fun_impute(ignore = c("pase_0","pase_4"),pase.mod = FALSE) 
+    )
+  })() |> 
+  purrr::map(\(.x){
+    .x |>
+      pase_cutter(drop.pase = TRUE,drop.nas = TRUE)
+  }) 
+
+# targets::tar_read(df_all_data_formatted) |> 
+#   dplyr::filter(mrs_0==0) |>
+#     dplyr::filter(!is.na(pase_0),!is.na(pase_4)) |> events_ready() |>
+#   pase_cutter(drop.pase = TRUE,drop.nas = TRUE) |>
+#     cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change") |>
+#     tbl_regression_standard()
+
+ls_change <- list(
+  "Univariable"= ls_mrs0_change$std |>
+    dplyr::select(pase_change,dplyr::everything()) |> 
+    splitdf4uvcox(include=c("time","status")) |> 
+    purrr::map(\(.x){
+      .x |> tbl_regression_standard()
+    }) |> 
+    gtsummary::tbl_stack(),
+  "Multivariable" = ls_mrs0_change$std |>
+    cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change") |>
+    tbl_regression_standard() |> gtsummary::add_glance_source_note(),
+  "Multivariable Imputed"= ls_mrs0_change$imp |>
+    cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change") |>
+    tbl_regression_standard()
+) |> purrr::map(\(.x){
+  .x|> 
+    gtsummary::modify_table_styling(column = p.value,
+                                    hide=TRUE)
+}) |> tbl_merged_named()
+
+ls_change
+#| eval: false
+coll <- list(
+  mrs0_0=ls_mrs0$std|>
+    cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change"),
+  pase_change_mrs0 = ls_mrs0_change$std|>
+  cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change"),
+  pase_change=df |>
+    get_vars(vars.groups =c("clin","lifestyle.events","ses", "assess.events")) |>
+    dplyr::filter(event.include) |>
+    dplyr::select(-tidyselect::all_of("event.include")) |> 
+    pase_cutter(drop.pase = TRUE,drop.nas = TRUE)|>
+    cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_change"),
+  prestroke_pase=df |>
+    pase_cutter(drop.nas = FALSE) |>
+    get_vars(vars.groups = c("clin", "lifestyle.events", "ses", "assess.pred", "quartiles"), vars.vec = c("inc_time",
+                                                                                                          "time",
+                                                                                                          "status")) |>
+    dplyr::mutate(time = time + inc_time / 365) |>
+    dplyr::select(-dplyr::any_of(c("pase_0", "pase_4", "pase_change", "inc_time", "event.include", "pase_4_quartile")))|>
+    cox_regression(all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_0_quartile")) |> 
+    purrr::map(performance::check_collinearity)
+
+coll |> purrr::imap(\(.x,.i){
+  .x |> dplyr::as_tibble() |>  gt::gt() |> gt::tab_header(.i)|> 
+  gt::gtsave(filename = here::here(glue::glue("out/coll_{.i}.docx")))
+})
+
+
+
+# Code from: pa_event_plots.qmd
+targets::tar_config_set(store = here::here("_targets"))
+source(here::here("R/functions.R"))
+source(here::here("R/glmnet-reg.R"))
+library(targets)
+library(tidyverse)
+# targets::tar_read(plot_events_survival_smooth)
+p1 <- targets::tar_read(df_event_data) |>
+  events_dataset(impute = FALSE) |>
+  dplyr::mutate(pase_change = factor(pase_change, 
+                                     levels = c("Persistently high", "Decrease", "Increase", "Persistently low")),
+                pase_change = dplyr::recode(pase_change,
+                                            "Persistently high"="Consistently above lowest", 
+                                            "Decrease"="Decrease to lowest", 
+                                            "Increase"="Increase from lowest", 
+                                            "Persistently low"="Consistently lowest")
+                ) |>
+  cox_regression() |>
+  plot_survival_smooth(line.w = 1) +
+  ggplot2::labs(
+    fill = "PA level change group",
+    color = "PA level change group",
+    linetype = "PA level change group"
+  ) +
+  ggplot2::scale_x_continuous(limits = c(0, 9.5), breaks = c(0, 3, 6, 9))
+
+
+ggplot2::ggsave(here::here("out/smooth_surv.png"), 
+                p1, 
+                dpi = 600, 
+                units = "cm", 
+                height = 8, 
+                width = 15)
+p1
+surv.data <- targets::tar_read(df_event_data) |>
+  events_dataset(impute = FALSE) |>
+  cox_regression(all.vars = FALSE) |> # Minimal model to just give risk table
+  ggsurvfit::survfit2() |>
+  ggsurvfit::tidy_survfit(times = c(0, 3, 6, 9))
+
+
+risk_event <- c("n.risk", "cum.event") |>
+  purrr::map2(c("Numbers at risk", "Events"), \(.x, .y){
+    surv.data |>
+      tidyr::pivot_wider(id_cols = strata, names_from = time, values_from = {{ .x }}) |>
+      mask_micro_table(col.sel = -strata) |> # Masking columns
+      tidyr::pivot_longer(cols = -strata) |>
+      setNames(c("strata", "time", .x)) |>
+      dplyr::mutate(dplyr::across(time, ~ as.numeric(.x))) |>
+      dplyr::mutate(strata = factor(strata, levels = rev(c("Persistently high", "Decrease", "Increase", "Persistently low")))) |>
+      ggplot2::ggplot(ggplot2::aes(x = time, y = strata, label = get(.x))) +
+      ggplot2::geom_text() +
+      ggplot2::labs(y = NULL, title = .y) +
+      ggplot2::theme_minimal() +
+      ggsurvfit::theme_risktable_default()
+  })
+
+p2 <- ggsurvfit::ggsurvfit_align_plots(list(p1, risk_event[1]) |> purrr::list_flatten()) |>
+  patchwork::wrap_plots(ncol = 1, heights = c(2, 1), guides = "collect")
+
+ggplot2::ggsave(here::here("out/smooth_surv_tables.png"), 
+                p2, 
+                dpi = 600, 
+                units = "cm", 
+                height = 9, 
+                width = 15)
+p2
+#| include: false
+targets::tar_read(df_event_data) |>
+  events_dataset(impute = FALSE) |>
+  dplyr::mutate(pase_change = factor(pase_change, levels = c("Persistently high", "Decrease", "Increase", "Persistently low"))) |>
+  cox_regression(all.vars = FALSE) |> # Minimal model to just give risk table
+  ggsurvfit::survfit2() |>
+  ggsurvfit::ggsurvfit() +
+  ggplot2::scale_y_continuous(
+    limits = c(0, 1.02),
+    breaks = seq(0, 1, .25),
+    labels = scales::percent,
+    expand = c(0.01, 0)
+  ) +
+  ggplot2::scale_x_continuous(breaks = c(0, 4, 8.5), expand = c(0.02, 0)) +
+  ggsurvfit::add_risktable()
+#| include: false
+targets::tar_read(df_event_data_small) |>
+  events_dataset(impute = FALSE) |>
+  dplyr::mutate(pase_change = factor(pase_change, levels = c("Persistently high", "Decrease", "Increase", "Persistently low"))) |>
+  cox_regression() |>
+  plot_survival_smooth() +
+  ggplot2::labs(
+    fill = "PA change group",
+    color = "PA change group",
+    linetype = "PA change group"
+  )
+#| include: false
+targets::tar_read(df_event_data) |>
+  events_dataset(impute = FALSE) |>
+  dplyr::mutate(pase_change = factor(pase_change, levels = c("Persistently high", "Decrease", "Increase", "Persistently low"))) |>
+  cox_regression() |>
+  plot_survival_smooth() +
+  ggplot2::labs(
+    fill = "PA change group",
+    color = "PA change group",
+    linetype = "PA change group"
+  )
+
+
+
+# Code from: pa_events_analyses.qmd
+targets::tar_config_set(store = here::here("_targets"))
+source(here::here("R/functions.R"))
+source(here::here("R/glmnet-reg.R"))
+library(targets)
+library(tidyverse)
+#| include: false
+list("Univariate"=targets::tar_read(tbl_events_cox_regression_uv),
+"Multivariate"=targets::tar_read(tbl_events_cox_regression),
+"Imputed Multivariate"=targets::tar_read(tbl_events_mids_cox_regression)) |> 
+  # purrr::map(gtsummary::modify_table_styling,column=p.value,hide=TRUE) |> 
+  purrr::map(gtsummary::bold_p) |> 
+  tbl_merged_named()
+#| include: true
+list("Univariate"=targets::tar_read(tbl_events_cox_regression_uv),
+"Multivariate"=targets::tar_read(tbl_events_cox_regression),
+"Imputed Multivariate"=targets::tar_read(tbl_events_mids_cox_regression)) |> 
+  purrr::map(gtsummary::modify_table_styling,column=p.value,hide=TRUE) |> 
+  tbl_merged_named()
+#| echo: true
+targets::tar_read(df_event_data) |> 
+  # dplyr::select(-reg_bmi) |>
+  na.omit() |> 
+  nrow()
+#| include: false
+targets::tar_read(tbl_events_cox_regression_small)
+
+targets::tar_read(df_event_data_small)
+#| include: false
+targets::tar_read(tbl_events_mids_cox_regression)
+targets::tar_read("df_all_data_formatted") |> 
+  events_ready() |> 
+  dplyr::filter(!is.na(pase_0),!is.na(pase_4)) |> 
+  (\(data){
+    c("pase_0","pase_4") |> 
+      purrr::map(\(exp){
+        with(data,survival::coxph(as.formula(glue::glue("survival::Surv(time, status)~{exp}")))) |> 
+  gtsummary::tbl_regression(exponentiate=TRUE)|> 
+  fix_labels()
+      })
+    })()  |> 
+  gtsummary::tbl_stack()
+targets::tar_read("df_all_data_formatted") |> 
+  pase_cutter(drop.nas = TRUE) |> 
+  events_ready(v.groups=c("clin","lifestyle.events","ses", "assess.events","quartiles"))  |> 
+  dplyr::select(-dplyr::any_of(c("pase_0","pase_4","pase_change"))) |> 
+  (\(data){
+    c("pase_0_quartile","pase_4_quartile") |> 
+      purrr::map(\(exp){
+        list("Univariable"=cox_regression(data=data,all.vars = FALSE, use.strata = FALSE, outcome.var = exp),
+             "Multivariable"=cox_regression(data=data,all.vars = TRUE, use.strata = FALSE, outcome.var = exp)) |> 
+          purrr::map(\(.x){
+            .x |> gtsummary::tbl_regression(exponentiate=TRUE)|> 
+  fix_labels()
+          }) |> 
+          tbl_merged_named()
+      })
+    })()  |> 
+  gtsummary::tbl_stack()
+targets::tar_read("df_all_data_formatted") |> 
+  pase_cutter(drop.nas = TRUE) |>
+    get_vars(vars.groups = c("clin","lifestyle.events","ses", "assess.events","quartiles"),vars.vec = c("inc_time"))  |> 
+  dplyr::mutate(time=time+inc_time/365) |> 
+  dplyr::select(-dplyr::any_of(c("pase_0","pase_4","pase_change","inc_time","event.include","pase_4_quartile"))) |> 
+  (\(data){
+    list(
+        with(data,survival::coxph(as.formula(glue::glue("survival::Surv(time, status)~{exp}")))) |> 
+  gtsummary::tbl_regression(exponentiate=TRUE)|> 
+  fix_labels() |> gtsummary::add_glance_source_note()
+      )
+    })()  |> 
+  gtsummary::tbl_stack()
+
+
+
+# Code from: pa_events_summaries.qmd
+targets::tar_config_set(store = here::here("_targets"))
+source(here::here("R/functions.R"))
+source(here::here("R/glmnet-reg.R"))
+library(targets)
+library(tidyverse)
+#| include: false
+targets::tar_read(tbl_events_summary)
+#| include: true
+targets::tar_read(tbl_events_summary) |> mask_micro_summary(micro.n = 5)
+#| include: false
+ls <- targets::tar_read(df_event_data)|>
+    pase_cutter(drop.pase = TRUE) |>
+    dplyr::select(-tidyselect::all_of(c("status", "time"))) |>
+    dplyr::filter(!is.na(pase_change)) |> 
+    labelling_data() |>
+    gtsummary::tbl_summary(
+      missing = "ifany",
+      by = pase_change,
+      # value = list(where(is.logical) ~ TRUE),
+      missing_text = "Missing"
+    ) |>
+    gtsummary::add_overall() |>
+    gtsummary::add_n()
+
+ls |> add_missing_stats() |>  mask_micro_summary(micro.n = 5)
+targets::tar_read(df_event_data)|>
+    pase_cutter(drop.pase = TRUE) |>
+    dplyr::select(pase_change, time) |>
+    dplyr::filter(!is.na(pase_change)) |> 
+    labelling_data() |>
+    gtsummary::tbl_summary(
+      missing = "ifany",
+      by = pase_change,
+      type = gtsummary::all_continuous() ~ "continuous2",
+      statistic = list(gtsummary::all_continuous() ~ c("{median} ({p25}, {p75})","{mean} ({sd})")),
+      # value = list(where(is.logical) ~ TRUE),
+      
+      missing_text = "Missing"
+    ) |>
+    gtsummary::add_overall()
+#| include: true
+targets::tar_read(df_all_data_formatted) |>
+  (\(.x)summary(.x$pase_0))()
+
+# targets::tar_read(df_all_data_formatted) |>
+#    dplyr::filter(!is.na(pase_0),!is.na(pase_4))|>
+# (\(.x)summary(.x$pase_0))()
+#| include: false
+targets::tar_read(df_all_data_formatted) |>
+  pase_cutter(drop.pase = TRUE) |>
+  dplyr::select(mrs_4, pase_change) |>
+  dplyr::mutate(pase_change = forcats::fct_relevel(pase_change, c("Persistent high", "Increase", "Decrease", "Persistent low"))) |>
+  na.omit() |>
+  (\(.x){
+    table(mrs = .x$mrs_4, pase = .x$pase_change)
+  })() |>
+  rankinPlot::grottaBar(groupName = "pase", scoreName = "mrs")
+targets::tar_read(df_event_data)|>
+    pase_cutter(drop.pase = FALSE) |>
+    dplyr::filter(!is.na(pase_change))|>
+    dplyr::select(pase_0,pase_4)  |> 
+    labelling_data() |> 
+  gtsummary::tbl_summary()
+targets::tar_read(df_all_data_formatted)|>
+    pase_cutter(drop.pase = FALSE) |>
+    dplyr::filter(!is.na(pase_change))|>
+    dplyr::count(pase_0_quartile,pase_4_quartile) |> 
+  write_csv(here::here("out/event_sankey_data.csv"))
+ds <- targets::tar_read(df_all_data_formatted) |>
+  get_vars(vars.groups = c("clin", "lifestyle", "ses", "assess.events", "extra")) |>
+  dplyr::select(-time, -status, -soc_status) |>
+  pase_cutter(drop.pase = TRUE) |>
+  labelling_data()
+
+ls <- list(
+  pase_out = ds |>
+    dplyr::mutate(event.filter = factor(
+      dplyr::case_when(is.na(pase_change) ~ "pase_incomplete",
+        !event.include ~ "early_event",
+        .default = "included"
+      ),
+      levels = c("pase_incomplete", "early_event", "included")
+    )),
+  event_out = ds |>
+    dplyr::mutate(event.filter = factor(
+      dplyr::case_when(!event.include ~ "early_event",
+        is.na(pase_change) ~ "pase_incomplete",
+        .default = "included"
+      ),
+      levels = c("early_event", "pase_incomplete", "included")
+    ))
+) |>
+  purrr::map(dplyr::select, -event.include, -pase_change, -event)
+
+ls_tbl <- ls |>
+  purrr::map(\(.x){
+    .x |>
+      dplyr::filter(event.filter != "pase_incomplete") |>
+      dplyr::mutate(event.filter = factor(event.filter))
+  }) |>
+  (\(.x) list(.x, purrr::pluck(ls, 1) |> dplyr::mutate(event.filter = event.filter == "included")))() |>
+  purrr::list_flatten()
+#| include: false
+ls_tbl |>
+  purrr::map(\(.x){
+    .x |>
+      gtsummary::tbl_summary(
+        by = event.filter,
+        missing = "ifany"
+      ) |>
+      gtsummary::add_p()
+  }) |>
+  (\(.x)gtsummary::tbl_merge(.x, c(names(.x)[1:2], "all_out")))()
+targets::tar_read(df_all_data_formatted) |>
+  get_vars(vars.groups = c("clin", "lifestyle", "ses", "assess.events", "extra", "assess.pred")) |>
+  dplyr::select(-time, 
+                # -status, 
+                -soc_status) |>
+  pase_cutter(drop.pase = FALSE) |>
+  labelling_data() |>
+    dplyr::mutate(event.filter = factor(
+      dplyr::case_when(
+        !event.include | is.na(pase_change) ~ "excluded",
+        .default = "included"
+      )
+    )) |>
+  dplyr::select(-who_4, -mdi_4, -mrs_4_above1, -mfi_gen_4) |>
+      gtsummary::tbl_summary(
+        by = event.filter,
+        missing = "ifany"
+      ) |>
+      gtsummary::add_p()
+ds |>
+  dplyr::mutate(pase_incomplete=is.na(pase_change),
+                excluded=pase_incomplete | !event.include,
+                early_event=!event.include) |> 
+  dplyr::select(early_event,pase_incomplete,excluded) |> 
+  gtsummary::tbl_summary(by=excluded)
+#| include: true
+ls_tbl |>
+  purrr::pluck(3) |>
+  dplyr::select(-who_4, -mdi_4, -mrs_4_above1, -mfi_gen_4) |>
+  (\(.x){
+    .x |>
+      gtsummary::tbl_summary(
+        by = event.filter,
+        missing = "no"
+      ) |>
+      gtsummary::add_p()
+  })() |> mask_micro_summary(micro.n = 5)
+#| include: true
+targets::tar_read(df_event_data)|>
+    pase_cutter(drop.pase = FALSE)|>
+    dplyr::filter(!is.na(pase_change)) |>
+    dplyr::select(-time, -status, -pase_0_quartile, -pase_4_quartile, -pase_change)  |> 
+    labelling_data() |> 
+  dplyr::mutate(reg_female=ifelse(reg_female,"Female","Male")) |>
+    gtsummary::tbl_summary(
+      missing = "ifany",
+      by = reg_female,
+      value = list(where(is.logical) ~ TRUE)
+    )|> gtsummary::add_p() |> 
+  mask_micro_summary()
+#| include: false
+events <- targets::tar_read(ls_all_events) |>
+  dplyr::bind_rows() |>
+  dplyr::left_join(targets::tar_read(df_all_data_formatted) |>
+    dplyr::select(c("event.include", "rdate", "enddate", "PNR")), by = c("CPR" = "PNR")) |>
+  dplyr::mutate(
+    date.event = as.Date(date.event),
+    event.trial = !date.event > enddate,
+    event.itt = !date.event > (lubridate::dmonths(6) + rdate)
+  ) |>
+  dplyr::mutate(event.type = dplyr::if_else(grepl("^death", event.type), "death", event.type))
+
+ls <- list(
+  # Events during inclusion and during first 6 months after inclusion (Intention to treat)
+  events |>
+    dplyr::select(event.type, event.trial, event.itt) |>
+    tidyr::pivot_longer(cols = c("event.trial", "event.itt")) |>
+    dplyr::filter(value) |>
+    dplyr::select(-value) |>
+    gtsummary::tbl_summary(by = name),
+
+  # All registred events
+  events |>
+    dplyr::select(event.type) |>
+    gtsummary::tbl_summary(),
+
+  # All events in the selected group
+  events |>
+    dplyr::select(event.type, event.include) |>
+    # tidyr::pivot_longer(cols = c("event.trial","event.itt")) |>
+    dplyr::filter(event.include) |>
+    dplyr::select(-event.include) |>
+    gtsummary::tbl_summary(),
+
+  # All considered events (first event)
+  targets::tar_read(df_events_deaths)|>
+    dplyr::mutate(event.type = dplyr::if_else(grepl("^death", event.type), "death", event.type)) |>
+    dplyr::select(event.type) |>
+    gtsummary::tbl_summary(),
+  
+   # All included events (first event)
+  targets::tar_read(df_all_data_formatted)|> 
+  pase_cutter(drop.pase = TRUE) |>
+  dplyr::filter(!is.na(pase_change))|>
+    dplyr::filter(event.include) |>
+    # dplyr::select(-event.include)|>
+    dplyr::mutate(event = dplyr::if_else(grepl("^death", event), "death", event)) |>
+    dplyr::select(event) |>
+    dplyr::filter(!is.na(event)) |> 
+    gtsummary::tbl_summary()
+) |>
+  setNames(c("During trial", "All events", "Selected events", "Considered", "Included"))
+
+ls[4] |> 
+  purrr::map(\(.x) .x |>
+               mask_micro_summary(micro.n = 5)) |>
+  tbl_merged_named()
+targets::tar_read(df_all_data_formatted)|> 
+  pase_cutter(drop.pase = TRUE) |>
+  dplyr::filter(!is.na(pase_change))|>
+    dplyr::filter(event.include) |>
+    # dplyr::select(-event.include)|>
+    dplyr::mutate(event = dplyr::case_when(grepl("^death", event)~"Mors",
+                                           grepl("^DI6", event)~ "DI61-4",
+                                           .default = event)) |>
+    dplyr::select(event) |>
+    dplyr::filter(!is.na(event)) |> 
+    gtsummary::tbl_summary() |> mask_micro_summary(micro.n = 5)
+targets::tar_read(df_event_data) |>
+  pase_cutter(drop.pase = TRUE) |>
+  dplyr::filter(!is.na(pase_change)) |> 
+  events_table(by="pase_change")
+
+
+
+# Code from: sex_events.qmd
+targets::tar_config_set(store = here::here("_targets"))
+source(here::here("R/functions.R"))
+source(here::here("R/glmnet-reg.R"))
+library(targets)
+library(tidyverse)
+
+df <- targets::tar_read(df_all_data_formatted) |> 
+  get_vars(vars.groups = c("clin","lifestyle.events","ses", "ssri")) |> 
+  dplyr::rename(status=status.all,
+                time=time.all)
+df |>
+  pase_cutter(drop.nas = TRUE,drop.pase = TRUE) |>
+  dplyr::mutate(reg_female=factor(ifelse(reg_female,"Female","Male"))) |> 
+  (\(.x) {
+    split(.x, .x$reg_female)
+  })() |> lapply(\(.x) {
+    ls <- list("Univariate"=.x |> dplyr::select(-reg_female,-rtreat) |>
+      standard_multi_cox_table(all.vars = FALSE) |> gtsummary::add_nevent(),
+"Multivariate"=.x |> dplyr::select(-reg_female,-rtreat) |>
+      standard_multi_cox_table(all.vars = TRUE) |> gtsummary::add_nevent())|> 
+  purrr::map(gtsummary::bold_p) |> 
+  tbl_merged_named() 
+    ls |> 
+       gtsummary::modify_table_body(~filter(.x, variable == "pase_change"))
+  })   |> 
+  (\(.x) {
+    gtsummary::tbl_stack(.x,group_header=names(.x)) 
+  })()
+  
+
+
+
+
+
+# Code from: ssri_events.qmd
+targets::tar_config_set(store = here::here("_targets"))
+source(here::here("R/functions.R"))
+source(here::here("R/glmnet-reg.R"))
+library(targets)
+library(tidyverse)
+
+df <- targets::tar_read(df_all_data_formatted) |> 
+  get_vars(vars.groups = c("clin","lifestyle.events","ses", "ssri")) |> 
+  dplyr::rename(status=status.all,
+                time=time.all)
+df |> 
+  dplyr::select(#-event.include,
+                -rtreat_placebo)|>
+  labelling_data() |> 
+  gtsummary::tbl_summary(by=rtreat) |> 
+  gtsummary::add_overall() #|> 
+  # mask_micro_summary()
+df |> 
+  dplyr::select(
+    -rtreat_placebo,
+                -pase_4#,
+    # -reg_bmi
+                )  |> 
+  cox_regression(outcome.var = "rtreat",use.strata = FALSE) |> 
+  gtsummary::tbl_regression(exponentiate = TRUE, 
+                                    add_estimate_to_reference_rows = TRUE,
+                                    show_single_row = where(is.logical)) |> 
+  gtsummary::bold_p() |> 
+  fix_labels()
+df |> 
+  dplyr::select(-rtreat_placebo) |> 
+  cox_regression(outcome.var = "rtreat",all.vars = FALSE, use.strata = TRUE,include_formula = TRUE) |> 
+  plot_survival_smooth()
+df |>
+  pase_cutter(drop.nas = TRUE,drop.pase = TRUE) |>
+  (\(.x) {
+    split(.x, .x$rtreat)
+  })() |> lapply(\(.x) {
+    ls <- list("Univariate"=.x |> dplyr::select(-rtreat_placebo, -rtreat) |>
+      standard_multi_cox_table(all.vars = FALSE),
+"Multivariate"=.x |> dplyr::select(-rtreat_placebo, -rtreat) |>
+      standard_multi_cox_table(all.vars = TRUE))|> 
+  purrr::map(gtsummary::bold_p) |> 
+  tbl_merged_named() 
+    ls |> 
+       gtsummary::modify_table_body(~filter(.x, variable == "pase_change"))
+  })   |> tbl_stack_named()
+  # gtsummary::tbl_stack(group_header=levels(factor(df$rtreat)))
+
+
+

2 Longterm/260315/glmnet-reg.R

@@ -0,0 +1,340 @@
+## ItMLiHSmar2022
+## regular_fun.R, child script
+## Regularisation model building function
+## Andreas Gammelgaard Damsbo, agdamsbo@clin.au.dk
+##
+## Now modified to use in publication
+##
+
+regular_fun <- function(X, y, K, lambdas, alpha) {
+  n <- nrow(X)
+  set.seed(321)
+
+  # Using caret function to ensure both levels represented in all folds
+  c <- caret::createFolds(y = y, k = K, list = FALSE, returnTrain = TRUE)
+
+  B <- yhatTestProbKeep <- list()
+  accTrain <- accTest <- err_train <- err_test <- auc_train <- auc_test <- matrix(nrow = K, ncol = length(lambdas))
+
+  TrainProb <- TestProb <- list()
+  
+  catinfo <- levels(y)
+
+  cMatTrain <- cMatTest <- table(true = factor(c(0, 0), levels = catinfo), pred = factor(c(0, 0), levels = catinfo))
+
+
+  ## Iterate over partitions
+  for (idx1 in 1:K) {
+    # Status
+    cat("Processing fold", idx1, "of", K, "\n")
+
+    # idx1=1
+    # Get training- and test sets
+    I_train <- c != idx1 ## Creating selection vector of TRUE/FALSE
+    I_test <- !I_train
+
+    Xtrain <- X[I_train, ]
+    ytrain <- y[I_train]
+    Xtest <- X[I_test, ]
+    ytest <- y[I_test]
+
+
+    ## Model matrices for glmnet
+    ## Using the complicated approach not to include first level.
+    # Xmat.train<-model.matrix(~ .-1, data=Xtrain,
+    #              contrasts.arg = lapply(Xtrain[,sapply(Xtrain, is.factor)],
+    #                                     contrasts, contrasts=T))
+    # Xmat.test<-model.matrix(~ .-1, data=Xtest,
+    #                contrasts.arg = lapply(Xtest[,sapply(Xtest, is.factor)],
+    #                                       contrasts, contrasts=T))
+
+    # Xmat.train<-model.matrix(~.-1,Xtrain)
+    # Xmat.test<-model.matrix(~.-1,Xtest)
+
+    # Weights
+    ytrain_weight <- as.vector(1 - (table(ytrain)[ytrain] / length(ytrain)))
+    # ytest_weight<-as.vector(1 / (table(ytest)[ytest] / length(ytest)))
+
+    # Fit regularized linear regression model
+    mod <- glmnet::glmnet(Xtrain, ytrain,
+      alpha = alpha, ## Alpha = 1 for lasso
+      lambda = lambdas, ## Setting lambdas
+      standardize = TRUE, ## Scales and centers
+      weights = ytrain_weight,
+      family = "binomial"
+    )
+
+    # Keep coefficients for plot
+    B[[idx1]] <- as.matrix(coef(mod))
+
+    # TrainProb[[idx1]] <- list()
+    TestProb[[idx1]] <- list()
+    
+    # Iterate over regularization strengths to compute training- and test
+    # errors for individual regularization strengths.
+    for (idx2 in 1:length(lambdas)) {
+      # idx2=1
+
+      # Predict
+      yhatTrainProb <- predict(mod,
+        s = lambdas[idx2],
+        newx = data.matrix(Xtrain),
+        type = "response"
+      )
+
+      yhatTestProb <- predict(mod,
+        s = lambdas[idx2],
+        newx = data.matrix(Xtest),
+        type = "response"
+      )
+      
+      # Compute training and test error
+      yhatTrain <- round(yhatTrainProb)
+      yhatTest <- round(yhatTestProb)
+
+      TestProb[[idx1]][[idx2]] <- dplyr::bind_cols(data.matrix(Xtest),y=ytest,pred=yhatTestProb,.name_repair = "unique_quiet")
+      
+      # Make predictions categorical again (instead of 0/1 coding)
+      yhatTrainCat <- factor(round(yhatTrainProb), levels = c("0", "1"), labels = catinfo, ordered = TRUE)
+      yhatTestCat <- factor(round(yhatTestProb), levels = c("0", "1"), labels = catinfo, ordered = TRUE)
+
+      # Evaluate classifier performance
+      # Accuracy
+      # accTrain[idx1,idx2] <- sum(yhatTrainCat==ytrain)/length(ytrain)
+      # accTest [idx1,idx2] <- sum(yhatTestCat==ytest)/length(ytest)
+      # #
+      # # Error rate
+      # err_train[idx1,idx2] =  1 - accTrain[idx1,idx2]
+      # err_test [idx1,idx2] =   1 - accTest[idx1,idx2]
+
+      # AUROC
+      suppressMessages(
+        auc_train[idx1, idx2] <- pROC::auc(ytrain, yhatTrainCat)
+      )
+      suppressMessages(
+        auc_test[idx1, idx2] <- pROC::auc(ytest, yhatTestCat)
+      )
+
+      # Compute confusion matrices
+      cMatTrain <- cMatTrain + table(true = ytrain, pred = yhatTrainCat)
+      cMatTest <- cMatTest + table(true = ytest, pred = yhatTestCat)
+    }
+  }
+  list(mod = mod, B = B, auc_train = auc_train, auc_test = auc_test, cMatTrain = cMatTrain, cMatTest = cMatTest, 
+       TrainProb=TrainProb, 
+       TestProb=TestProb)
+}
+
+
+## ItMLiHSmar2022
+## regularisation_steps.R, child script
+## Regularised model building and analysation for assignment
+## Andreas Gammelgaard Damsbo, agdamsbo@clin.au.dk
+##
+## Now modified to use in publication
+##
+
+
+
+#' Title
+#'
+#' @param data
+#' @param outcome.var
+#' @param weighted
+#'
+#' @return
+#' @export
+#'
+#' @examples
+#' data <- targets::tar_read(df_pred_data) |> 
+#' pred_ls_split(excluded.vars = "reg_bmi")
+#' 
+#' data <- data[[1]]
+#' mod <- data |> regularisation_steps(auto.l=TRUE)
+regularisation_steps <- function(data, outcome.var = "pase_bin", weighted = FALSE, auto.l = FALSE) {
+  n <- nrow(data)
+
+  y <- data |> dplyr::select({{ outcome.var }})
+  X <- data |> dplyr::select(-{{ outcome.var }})
+
+  ## ====================================================================
+  ## Step 0: data import and wrangling
+  ## ====================================================================
+
+  # setwd("/Users/au301842/PhysicalActivityandStrokeOutcome/1 PA Decline/")
+
+  # source("data_format.R")
+  y1 <- factor(as.integer(y[[1]])) ## Outcome is required to be factor of 0 or 1.
+  # summary(y1)
+
+  ## ====================================================================
+  ## Step 1: settings
+  ## ====================================================================
+
+  ## Folds
+  K <- 10
+  set.seed(3)
+  c <- caret::createFolds(
+    y = y1,
+    k = K,
+    list = FALSE,
+    returnTrain = TRUE
+  ) # Fold IDs for tuning
+
+  ## Defining tuning parameters
+  if (auto.l){
+    lambdas <- NULL
+  } else {
+    lambdas <- 2^seq(-10, 20, 1)
+  }
+  
+  alphas <- seq(0, 1, .1)
+
+  ## Weights for models
+  if (weighted) {
+    wght <- as.vector(1 - (table(y1)[y1] / length(y1)))
+  } else {
+    wght <- rep(1, length(y1))
+  }
+
+
+  ## Standardise numeric
+  ## Centered and
+
+
+
+  ## ====================================================================
+  ## Step 2: all cross validations for each alpha
+  ## ====================================================================
+
+  # library(furrr)
+  # library(purrr)
+  # library(doMC)
+  # registerDoMC(cores=6)
+
+  future::plan(strategy = "multisession", workers = 2)
+
+  # Nested CVs with analysis for all lambdas for each alpha
+  #
+  set.seed(3)
+  cvs <- furrr::future_map(alphas, .options = furrr::furrr_options(seed = 3), function(a) {
+    glmnet::cv.glmnet(model.matrix(~ . - 1, X),
+      y1,
+      weights = wght,
+      lambda = lambdas,
+      type.measure = "deviance", # This is standard measure and recommended for tuning
+      foldid = c, # Per recommendation the folds are kept for alpha optimisation
+      alpha = a,
+      standardize = TRUE,
+      family = quasibinomial, # Same as binomial, but not as picky
+      keep = TRUE
+    )
+  })
+
+  ## ====================================================================
+  # Step 3: optimum lambda for each alpha
+  ## ====================================================================
+
+
+  # For each alpha, lambda is chosen for the lowest meassure (deviance)
+  each_alpha <- sapply(seq_along(alphas), function(id) {
+    each_cv <- cvs[[id]]
+    alpha_val <- alphas[id]
+    index_lmin <- match(
+      each_cv$lambda.min,
+      each_cv$lambda
+    )
+    c(
+      lamb = each_cv$lambda.min,
+      alph = alpha_val,
+      cvm = each_cv$cvm[index_lmin]
+    )
+  })
+
+  if (auto.l){
+    # Best (min) lambda
+    best_lamb <- min(each_alpha["lamb", ])
+    
+    # Alpha is chosen for best lambda with lowest model deviance, each_alpha["cvm",]
+    best_alph <- each_alpha["alph", ][each_alpha["cvm", ] == min(each_alpha["cvm", ]
+                                                                 [each_alpha["lamb", ] %in% best_lamb])]
+    
+    # BEst lamb is set to NULL to allow glm.net to use the optimal method, which is bult in.
+    # best_lamb <- NULL
+    
+  } else {
+    # Best (min) lambda
+    best_lamb <- min(each_alpha["lamb", ])
+    
+    # Alpha is chosen for best lambda with lowest model deviance, each_alpha["cvm",]
+    best_alph <- each_alpha["alph", ][each_alpha["cvm", ] == min(each_alpha["cvm", ]
+                                                                 [each_alpha["lamb", ] %in% best_lamb])]
+  }
+  
+
+  ## https://stackoverflow.com/questions/42007313/plot-an-roc-curve-in-r-with-ggplot2
+  # df_roc <- glmnet::roc.glmnet(cvs[[match(best_alph, alphas)]]$fit.preval, newy = y1)[match(best_lamb, lambdas)]# |> # Plots performance from model with best alpha
+  #
+  #   df_roc |> plot_roc_curve()
+
+  ## ====================================================================
+  # Step 4: Creating the final model
+  ## ====================================================================
+
+  # source(here::here("R/regular_fun.R")) # Custom function
+  optimised_model <- regular_fun(X = X, y = y1, K = K, lambdas = best_lamb, alpha = best_alph)
+  # With lambda and alpha specified, the function is just a k-fold cross-validation wrapper,
+  # but keeps model performance figures from each fold.
+  
+  # list2env(optimised_model, .GlobalEnv)
+  # Function outputs a list, which is unwrapped to Env.
+  # See source script for reference.
+
+  ## ====================================================================
+  # Step 5: creating table of coefficients for inference
+  ## ====================================================================
+
+  # reg_coef_tbl <- optimised_model$B |> purrr::reduce(cbind)
+
+  # Bmatrix <- optimised_model$B |> purrr::reduce(cbind)
+  # Bmedian <- apply(Bmatrix, 1, median)
+  # Bmean <- apply(Bmatrix, 1, mean)
+  #
+  # reg_coef_tbl <- dplyr::tibble(
+  #   name = rownames(Bmatrix),
+  #   medianX = round(Bmedian, 5),
+  #   ORmed = round(exp(Bmedian), 5),
+  #   meanX = round(Bmean, 5),
+  #   ORmea = round(exp(Bmean), 5)
+  # ) # |>
+  # arrange(desc(abs(medianX)))%>%
+  # gt::gt()
+
+  ## ====================================================================
+  # Step 6: plotting predictive performance
+  ## ====================================================================
+
+  # reg_cfm <- caret::confusionMatrix(optimised_model$cMatTest)
+  # reg_cfm <- optimised_model$cMatTest
+  # reg_auc_sum <- optimised_model$auc_test[, 1]
+
+  ## ====================================================================
+  # Step 7: Packing list to save in loop
+  ## ====================================================================
+
+  list(
+    "IncludedN" = n,
+    "model" = optimised_model,
+    "alphas" = alphas,
+    "bestA" = best_alph,
+    "lambdas" = lambdas,
+    "bestL" = best_lamb,
+    # "TestTable" = reg_cfm,
+    # "AUROC" = reg_auc_sum,
+    # "ROC curve" = df_roc,
+    "y1" = y1,
+    "X" = X,
+    "data" = data,
+    "cvs" = cvs
+  )
+}

2 Longterm/260315/last_sensitivity.R

@@ -0,0 +1,172 @@
+targets::tar_read("df_all_data_formatted") |>
+  events_ready() |>
+  dplyr::filter(!is.na(pase_0), !is.na(pase_4)) |>
+  (\(data){
+    c("pase_0", "pase_4") |>
+      purrr::map(\(exp){
+        list(
+          "Univariable" = cox_regression(data = data, all.vars = FALSE, use.strata = FALSE, outcome.var = exp),
+          "Multivariable" = cox_regression(data = data, all.vars = TRUE, use.strata = FALSE, outcome.var = exp)
+        ) |>
+          purrr::map(\(.x){
+            .x |>
+              gtsummary::tbl_regression(exponentiate = TRUE) |>
+              fix_labels()
+          }) |>
+          tbl_merged_named()
+      })
+  })() |>
+  gtsummary::tbl_stack()
+
+
+targets::tar_read("df_all_data_formatted") |>
+  pase_cutter(drop.nas = TRUE) |>
+  events_ready(v.groups = c("clin", "lifestyle.events", "ses", "assess.events", "quartiles")) |>
+  dplyr::select(-dplyr::any_of(c("pase_0", "pase_4", "pase_change"))) |>
+  (\(data){
+    c("pase_0_quartile", "pase_4_quartile") |>
+      purrr::map(\(exp){
+        list(
+          "Univariable" = cox_regression(data = data, all.vars = FALSE, use.strata = FALSE, outcome.var = exp),
+          "Multivariable" = cox_regression(data = data, all.vars = TRUE, use.strata = FALSE, outcome.var = exp)
+        ) |>
+          purrr::map(\(.x){
+            .x |>
+              gtsummary::tbl_regression(exponentiate = TRUE) |>
+              fix_labels()
+          }) |>
+          tbl_merged_named()
+      })
+  })() |>
+  gtsummary::tbl_stack()
+
+
+targets::tar_read("df_all_data_formatted") |>
+  pase_cutter(drop.nas = TRUE) |>
+  get_vars(vars.groups = c("clin", "lifestyle.events", "ses", "assess.events", "quartiles"), vars.vec = c("inc_time")) |>
+  dplyr::mutate(time = time + inc_time / 365) |>
+  dplyr::select(-dplyr::any_of(c("pase_0", "pase_4", "pase_change", "inc_time", "event.include", "pase_4_quartile"))) |>
+  (\(data){
+    list(
+      "Univariable" = cox_regression(data = data, all.vars = FALSE, use.strata = FALSE, outcome.var = "pase_0_quartile"),
+      "Multivariable" = cox_regression(data = data, all.vars = TRUE, use.strata = FALSE, outcome.var = "pase_0_quartile")
+    ) |>
+      purrr::map(\(.x){
+        .x |>
+          gtsummary::tbl_regression(exponentiate = TRUE) |>
+          fix_labels()
+      }) |>
+      tbl_merged_named()
+  })()
+
+ls_sens <- list(
+  pase_0_all_quartile = list(
+    data = targets::tar_read("df_all_data_formatted") |>
+      pase_cutter(drop.nas = FALSE) |>
+      get_vars(vars.groups = c("clin", "lifestyle.events", "ses", "assess.pred", "quartiles"), vars.vec = c("inc_time",
+                                                                                                            "time",
+                                                                                                            "status")) |>
+      dplyr::mutate(time = time + inc_time / 365) |>
+      dplyr::select(-dplyr::any_of(c("pase_0", "pase_4", "pase_change", "inc_time", "event.include", "pase_4_quartile"))),
+    main.exp = "pase_0_quartile"
+  ),
+  # pase_0_all_contin = list(
+  #   data = targets::tar_read("df_all_data_formatted") |>
+  #     get_vars(vars.groups = c("clin", "lifestyle.events", "ses", "assess.pred", "quartiles"), vars.vec = c("inc_time",
+  #                                                                                                           "time",
+  #                                                                                                           "status")) |>
+  #     dplyr::mutate(time = time + inc_time / 365) |>
+  #     dplyr::select(-dplyr::any_of(c("pase_4", "pase_change", "inc_time", "event.include", "pase_0_quartile", "pase_4_quartile"))),
+  #   main.exp = "pase_0"
+  # ),
+  pase_0_excl_quartile = list(
+    data = targets::tar_read("df_all_data_formatted") |>
+      pase_cutter(drop.nas = TRUE,drop.pase = FALSE) |>
+      events_ready(v.groups = c("clin", "lifestyle.events", "ses", "assess.pred", "quartiles"), vars.vec = c("inc_time",
+                                                                                                             "time",
+                                                                                                             "status")) |>
+      dplyr::select(-dplyr::any_of(c("pase_0","pase_4", "pase_change", "inc_time", "event.include", "pase_4_quartile"))),
+    main.exp = "pase_0_quartile"
+  ),
+  # pase_0_excl_contin = list(
+  #   data = targets::tar_read("df_all_data_formatted") |>
+  #     events_ready(v.groups = c("clin", "lifestyle.events", "ses", "assess.pred", "quartiles"), vars.vec = c("inc_time",
+  #                                                                                                            "time",
+  #                                                                                                            "status")) |>
+  #     dplyr::filter(!is.na(pase_0), !is.na(pase_4)) |>
+  #     dplyr::select(-pase_4),
+  #   main.exp = "pase_0"
+  # ),
+  pase_4_quartile = list(
+    data = targets::tar_read("df_all_data_formatted") |>
+      dplyr::mutate(pase_4_quartile=cut(x = pase_4, breaks = quantile(pase_4, na.rm = TRUE), labels = 1:4, include.lowest = TRUE))  |>
+      get_vars(vars.groups = c("clin", "lifestyle.events", "ses", "assess.events", "quartiles"))|>
+      dplyr::filter(!is.na(pase_4_quartile),event.include) |>
+      dplyr::select(-dplyr::any_of(c("pase_0","pase_4", "pase_change", "inc_time", "event.include", "pase_0_quartile"))),
+    main.exp = "pase_4_quartile"
+  ),
+  # pase_4_contin = list(
+  #   data = targets::tar_read("df_all_data_formatted") |>
+  #     events_ready() |>
+  #     dplyr::filter(!is.na(pase_0), !is.na(pase_4)) |>
+  #     dplyr::select(-pase_0),
+  #   main.exp = "pase_4"
+  # ),
+  pase_4_change_late_cut = list(
+    data = targets::tar_read("df_all_data_formatted") |>
+      events_ready() |>
+      dplyr::filter(!is.na(pase_0), !is.na(pase_4)) |>
+      pase_cutter(drop.pase = TRUE),
+    main.exp = "pase_change"
+  # ),
+  # pase_4_change_earliest_cut = list(
+  #   data = targets::tar_read("df_all_data_formatted") |>
+  #     pase_cutter(drop.pase = TRUE)|>
+  #     events_ready(vars.vec = c("pase_change")) |>
+  #     dplyr::filter(!is.na(pase_change)) ,
+  #   main.exp = "pase_change"
+  )
+)
+
+
+
+f <- function(data, main.exp, ...) {
+  set.seed(3023)
+  imp <- fun_impute(data = data, ignore = main.exp)
+
+  list(
+    "Univariable" = cox_regression(data = data, all.vars = FALSE, use.strata = FALSE, outcome.var = main.exp),
+    "Multivariable" = cox_regression(data = data, all.vars = TRUE, use.strata = FALSE, outcome.var = main.exp),
+    "Multivariable Imputed" = cox_regression(data = imp, all.vars = TRUE, use.strata = FALSE, outcome.var = main.exp)
+  )
+}
+
+
+ls_out <- purrr::map(ls_sens, \(.x){
+  do.call(f, .x)
+})
+
+ls_stack <- ls_out |>
+  purrr::imap(\(.x, .i){
+    list(
+      "Group counts" = ls_sens[[.i]][["data"]] |>
+        dplyr::select(ls_sens[[.i]][["main.exp"]]) |> 
+        gtsummary::tbl_summary(statistic =  list(gtsummary::all_continuous() ~ "{N_nonmiss} ({p_nonmiss}%)", gtsummary::all_categorical() ~ "{n} ({p}%)")) |> 
+        fix_labels(),
+      .x |>
+        lapply(\(.y){
+          .y |>
+            tbl_regression_standard() |>
+            # gtsummary::modify_table_styling(columns = tidyselect::starts_with("p.value"), hide = TRUE) |>
+            gtsummary::remove_row_type(variables = -dplyr::any_of(ls_sens[[.i]][["main.exp"]]), type = "all") |> 
+            gtsummary::bold_p()
+        })
+    ) |>
+      purrr::list_flatten() |>
+      tbl_merged_named()
+  }) |>
+  tbl_stack_named()
+
+ls_stack <- ls_stack|> gtsummary::as_gt() |> gt::tab_style(style = gt::cell_text(weight="bold"),locations = gt::cells_row_groups(dplyr::everything()))
+
+ls_stack |> gt::gtsave(filename = here::here("out/pase_extra_cox2.docx"))

2 Longterm/DDV 241031/event_sankey_data.csv

@@ -0,0 +1,17 @@
+pase_0_quartile,pase_4_quartile,n
+1,1,56
+1,2,35
+1,3,14
+1,4,7
+2,1,38
+2,2,41
+2,3,22
+2,4,22
+3,1,15
+3,2,29
+3,3,46
+3,4,39
+4,1,10
+4,2,16
+4,3,34
+4,4,74

2 Longterm/Project B - DAG.Rmd

@@ -0,0 +1,114 @@
+---
+title: "Project B - DAG"
+author: "AGDamsbo"
+date: "`r Sys.Date()`"
+output: html_document
+---
+
+```{r setup, include=FALSE}
+knitr::opts_chunk$set(echo = TRUE)
+```
+
+# Hypothesis
+
+```{r}
+labels_all<-list(rtreat~"Trial treatment",
+                 pase_0~"PASE score",
+                 age~"Age",
+                 sex~"Sex",
+                 smoker~"History of smoking",
+                 civil~"Cohabitation",
+                 diabetes~"Known diabetes", 
+                 hypertension~"Known hypertension", 
+                 afli~"Known Atrialfibrillation", 
+                 ami~"Previos myocardial infarction", 
+                 tci~"Previos TIA",
+                 pad~"Known peripheral artery disease",
+                 nihss_0~"Acute NIHSS score", 
+                 thrombolysis~"Thrombolytic therapy", 
+                 thrombechtomy~"Endovascular treatment",
+                 SES~"Socio economic status",
+                 education~"Education",
+                 ad_treat~"Anti depression treatment",
+                 vasc_event~"Vascular event")
+```
+
+```{r}
+vars <- c(
+  "pase_0",
+  "age",
+  "sex",
+  "civil",
+  "smoker",
+  "rtreat",
+  "alc",
+  "afli",
+  "hypertension",
+  "diabetes",
+  "mrs_0",
+  "nihss_c",
+  "thrombolysis",
+  "pad",
+  "thrombechtomy",
+  "ami",
+  "tci",
+  "compliant",
+  "SES",
+  "education"
+)
+```
+
+```{r}
+library(ggdag)
+library(ggplot2)
+```
+
+```{r}
+
+dag <-
+  dagify(
+    vasc_event ~ pase_0 + age + sex + civil + smoker + rtreat + alc + afli +
+      hypertension + diabetes + mrs_0 + nihss_0 + thrombolysis + pad + thrombechtomy +
+      ami + tci + SES + education + ad_treat + svd,
+    pase_0 ~ sex + civil + alc + pad + SES + education + hypertension + diabetes,
+    diabetes ~ civil,
+    svd~hypertension + diabetes + alc,
+    mrs_0 ~ tci + ami + hypertension + diabetes + svd,
+    age ~ smoker+SES+education,
+    civil ~ sex + age + SES + civil,
+    # smoker~ ,
+    alc ~ sex+SES+education,
+    # afli~ ,
+    hypertension~ alc,
+    nihss_0 ~ hypertension + diabetes + pase_0 + age,
+    thrombolysis ~ mrs_0+nihss_0,
+    # pad~ ,
+    thrombechtomy ~ mrs_0+nihss_0,
+    # ami~ ,
+    # tci~ ,
+    # compliant~ ,
+    # SES~ ,
+    # education,
+    ad_treat~education+SES,
+    # labels = labels_all,
+    latent = c("svd"),
+    exposure = "pase_0",
+    outcome = "vasc_event"
+  )
+```
+
+```{r}
+dag |> ggdag(text = TRUE) + theme_dag(6) + geom_dag_edges_arc()
+```
+
+```{r}
+dag |> ggdag_parents("svd")
+```
+
+```{r}
+dag |> ggdag_paths(text = TRUE, shadow = TRUE)
+```
+
+```{r}
+dag |> ggdag_adjustment_set(text = TRUE, shadow = TRUE,stylized = TRUE)
+```

2 Longterm/assigndata.csv

@@ -0,0 +1,643 @@
+"pase_0","age","sex","civil","smoke_ever","rtreat","alc","afli","hypertension","diabetes","mrs_0","nihss_c","thrombolysis","pad","thrombechtomy","ami","tci","pase_6","mrs_1","mfi_gen_1","mfi_phys_1","mfi_act_1","mfi_mot_1","mfi_men_1","mdi_1","who5_score_1"
+"356.44","76","male","partner","never","Placebo","guideline","no","yes","no","0","2","yes","no","no","yes","no","260.52","0","10","4","4","4","8","6","84"
+"277","49","male","partner","never","Placebo","guideline","no","no","no","0","4","no","no","no","no","no","113.11","2","12","15","5","6","11","11","64"
+"192.6","43","male","alone","never","Placebo","guideline","no","yes","yes","0","2","yes","no","no","no","no","123.05","4","12","17","7","4","4","3","76"
+"30","89","female","alone","ever","Placebo","guideline","yes","no","no","0","3","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"44.14","80","male","partner","ever","Active","guideline","no","no","no","2","4","no","no","no","no","no","115.15","3","16","20","17","10","6","11","64"
+"128.76","72","male","partner","never","Placebo","guideline","no","yes","no","0","1","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"224.84","71","female","partner","ever","Placebo","guideline","no","yes","no","0","1","no","no","no","no","no","198.08","1","8","8","12","7","9","1","80"
+"100","66","female","alone","never","Active","guideline","no","no","no","0","4","no","no","yes","no","no","32.36","4","17","16","12","12","15","14","52"
+"144.8","64","male","partner","never","Placebo","guideline","no","no","no","0","3","no","no","no","no","no","171.84","1","14","13","14","6","14","3","84"
+"136.8","64","male","partner","never","Placebo","guideline","no","yes","no","0","3","no","yes","no","no","no","9.7","2","10","17","18","9","15","7","36"
+"134.33","65","female","alone","ever","Placebo","guideline","no","yes","yes","0","2","no","no","no","no","no","178.97","2","10","10","6","4","4","4","80"
+"118.2","63","male","partner","never","Active","guideline","no","yes","no","0","9","yes","no","no","no","no","33.14","2","4","16","13","9","14","3","48"
+"99.28","62","male","partner","never","Active","more","no","yes","no","0","1","no","no","no","no","no","209.68","1","15","11","13","12","12","9","48"
+"101.3","73","male","partner","never","Placebo","more","no","no","no","1","16","yes","no","yes","no","no","34.6","1","14","15","15","12","8","12","0"
+"75.5","59","female","alone","never","Placebo","guideline","no","no","no","0","5","no","no","no","no","no","123.97","2","20","20","19","4","20","22","44"
+"81","73","male","partner","never","Active","guideline","no","yes","no","0","2","no","no","no","yes","no","184.6","1","16","15","19","10","9","10","44"
+"79.08","70","male","partner","never","Active","guideline","yes","no","no","0","5","yes","no","no","no","no","235.75","2","17","14","14",NA,NA,"9","60"
+"27.2","83","male","partner","never","Active","guideline","no","yes","no","0","3","no","no","no","no","no","81","0","6","7","8","6","4","3","68"
+"173.25","80","male","partner","ever","Active","guideline","no","yes","no","0","5","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"255.8","63","male","alone","never","Placebo","guideline","no","no","no","0","9","no","no","no","no","no","101.4","1","4","8","8","6","4","5","84"
+"27.2","82","male","alone","never","Active","guideline","no","yes","no","0","3","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"309.51","72","male","partner","ever","Placebo","guideline","yes","no","no","1","11","yes","no","no","no","no","486.35","2","11","4","11","6","15","2","80"
+"199.91","58","female","alone","ever","Active","guideline","no","yes","no","0","0","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"218.09","64","male","partner","ever","Active","guideline","no","no","no","0","2","no","no","no","no","no","117.25","0","4","4","5","6","5","1","92"
+"166.8","49","female","partner","ever","Placebo","guideline","no","no","no","0","5","no","no","no","no","no","183.58","0","5","8","9","7","7","6","76"
+"25","99","female","alone","ever","Placebo","guideline","yes","no","no","0","8","yes","no","no","no","no","9.03","2","12","16","14","6","7","14","52"
+"90","79","female","alone","never","Placebo","guideline","no","yes","no","0","2","yes","no","no","yes","yes","116.4","0","16","4","8","4","12","11","96"
+"232.91","60","male","partner","never","Active","guideline","no","yes","no","0","5","no","no","no","no","no","228.25","2","12","12","12","10","9","9","72"
+"208.24","51","male","alone","never","Placebo","guideline","no","no","no","0","10","no","no","no","no","no","98.35","2","12","11","10","11","6","4","76"
+"28.11","77","male","alone","never","Placebo",NA,"yes","yes","no","0","9","no","no","no","no","no","8.6","4","10","14","16","10","9","14","52"
+"116","54","female","partner","never","Active","guideline","no","yes","no","0","6","no","no","no","no","no","111.28","0",NA,NA,NA,NA,NA,NA,"0"
+"271.5","71","female","partner","ever","Placebo","guideline","no","yes","no","0","16","yes","no","no","no","no","244.88","3","13","4","5","4","10","10","72"
+"155.8","67","female","alone","never","Placebo","more","no","yes","no","0","1","no","yes","no","no","no","187.4","0","6","10","9","4","4","0","100"
+"96.4","31","male","alone","never","Active","guideline","no","no","no","0","5","no","no","no","no","no","81.32","1","4","6","7","6","9","5","92"
+"88.99","72","female","partner","never","Active","guideline","yes","yes","no","1","5","no","no","no","no","no","106.65","1","11","13","14","4","4","8","64"
+"78.92","71","female","partner","never","Active","guideline","no","no","no","1","15","yes","no","no","no","no","152.73","0","4","6","6","7","6","4","80"
+"98.68","75","female","partner","ever","Active","guideline","no","no","no","0","4","no","no","no","no","no","60.83","2","9","11","10","14","6","3","84"
+"183.3","78","male","partner","never","Active","guideline","no","no","no","0","2","no","no","no","no","no","220.04","1","12","7","9","6","4","1","76"
+"192.5","74","male","partner","never","Active","guideline","no","no","no","0","3","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"218.05","62","female","partner","ever","Placebo","guideline","no","yes","no","1","3","yes","no","no","no","no","185.18","1","11","13","11","13","11","6","64"
+"144.7","86","male","partner","ever","Active","guideline","no","yes","no","0","2","yes","no","no","no","no","130.37","1","5","6","9","9","5","5","84"
+"85","82","male","partner","never","Active","guideline","no","yes","no","0","3","yes","no","no","no","no","62.8","0","20","19","19","16","9","18","48"
+"168.87","60","female","partner","never","Active","guideline","no","yes","no","0","0","no","no","no","no","no","230.14","2","15","16","11","6","12","14","40"
+"115.56","63","female","partner","never","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","168.73","2","13","15","12","13","9","12","60"
+"126","53","male","partner","never","Placebo","guideline","no","no","no","0","3","yes","no","no","no","no","136","1","17","8","8",NA,"14","6","92"
+"58.53","86","female","alone","never","Active","guideline","no","yes","no","0","0","no","no","no","no","no","246.47","0","14","14","13","7","9","4","52"
+"75.8","75","male","partner","never","Placebo","guideline","yes","no","yes","0","2","yes","no","no","no","no","108.2","2","11","13","16","10","7","14","64"
+"133.01","70","female","alone","never","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","228.24","2","7","8","4","4","4","3","88"
+"114.93","63","male","partner","never","Placebo","guideline","no","no","yes","0","3","no","no","no","no","no","188.05","2","18","18","12","6","12","5","72"
+"243.33","57","male","alone","never","Placebo",NA,"no","no","no","0","6","no","no","yes","no","no","200.09","3","7","5","6","4","10","5","8"
+"218.89","77","male","alone","ever","Placebo","guideline","no","no","no","0","3","no","no","no","no","no","32.09","1","5","13","16","11","5","11","88"
+"116.05","76","female","alone","ever","Active","guideline","no","no","no","0","7","no","no","no","no","no","66.52","4","8","9","4","5","5","6","88"
+"126.31","65","male","partner","never","Active","guideline","yes","yes","no","0","2","yes","no","no","no","no","259.37",NA,NA,NA,NA,NA,NA,NA,"0"
+NA,"82","female","alone","ever","Placebo","guideline","yes","no","no","1","16","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"81.72","77","male","partner","never","Active","more","yes","yes","no","1","1","no","yes","no","no","no","35.91","2",NA,NA,NA,NA,NA,NA,NA
+"155.83","63","female","alone","never","Placebo","guideline","no","no","no","0","0","no","no","no","yes","no","221.5","0","10","8","9","7","4","5","56"
+"136","55","male","alone","never","Placebo","guideline","no","no","no","0","7","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"247.12","53","female","partner","never","Placebo","guideline","no","no","no","0","8","yes","no","no","no","no","203.62","1","17","10","15","6","4","6","60"
+NA,"73","male","alone","never","Active","guideline","no","no","no","0","7","no","no","no","no","no","42.63","3","20","20","16","8","8","29","4"
+"33.6","62","female","alone","never","Active","guideline","no","no","no","1","8","no","no","no","no","no","192.4",NA,NA,NA,NA,NA,NA,NA,NA
+"296","60","male","partner","never","Active","guideline","no","no","no","0","12","no","no","no","no","no","112.82","4","8","12","15","4","4","3","92"
+"59.2","87","female","alone","never","Active","guideline","yes","yes","no","2","4","no","no","no","no","no","99.89","2","14","17","20","10","4","12","72"
+"214.48","74","male","alone","never","Placebo","guideline","no","no","yes","0","4","no","no","no","no","no","337.61","1","6","4","11","5","11","6","80"
+"418.9","54","female","alone","ever","Active","guideline","no","yes","yes","0","3","no","no","no","no","no","295.91","1","18","14","12",NA,"16","8","48"
+"114.5","69","male","partner","never","Placebo","guideline","no","yes","no","0","10","yes","no","no","yes","no","105.8","1","4","6","4","4","8","1","88"
+"56","67","male","partner","never","Active","guideline","no","no","no","0","4","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"292.24","63","male","partner","ever","Active","guideline","no","no","no","0","2","yes","no","no","yes","no","52.5","3","4","8","8","4","4","4","96"
+"155.83","45","male","partner","ever","Active","guideline","no","yes","no","0","4","no","no","no","no","no","335.69","1","9","7","9","8","4","3","92"
+"106","78","male","alone","never","Active","guideline","no","no","no","0","4","no","no","no","no","no",NA,"1","4","5","5","5","8","3","96"
+"114.6","52","male","partner","never","Active","guideline","no","yes","no","0","2","yes","no","no","yes","no","31.4","1","19","13","11","8",NA,"22","44"
+"0","86","female","alone","never","Placebo","guideline","no","no","no",NA,"2","yes","no","no","no","no","35","3","4","20","16","8","4","0","88"
+"55","67","male","partner","never","Active","guideline","no","yes","no","0","7","no","yes","no","no",NA,"3.3","2","10","7","14","10","5","2","88"
+"0","76","male","partner","never","Active","guideline","no","yes","no","0","3","yes","no","no","yes","no",NA,"1","16","19","17","18","4","15","44"
+"158.5","67","male","partner","ever","Active","guideline","no","yes","no","0","16","yes","no","no","no","no","222.07","1","14","8","5","4","16","4","80"
+"246.85","70","male","partner","ever","Placebo","guideline","no","yes","no","0","4","no","no","no","no",NA,"0","1","10","11","8","4","10","10","72"
+"196","81","male","alone","never","Placebo","guideline","no","no","no","0","2","yes","no","no","no","no","33.6","3","4","4","11","4","11","5","88"
+"249.9","62","male","partner","ever","Active","guideline","no","yes","no","0","1","yes","no","no","no","yes",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"61","64","male","partner","never","Active","guideline","no","yes","no","0","3","no","no","no","no","no","35.4","3","17","18","18","4","4","10","72"
+"0","60","male","alone","ever","Active","guideline","no","yes","no","2","3","no","no","no","no","no",NA,"4",NA,NA,NA,NA,NA,"16","52"
+"156.21","68","male","partner","never","Active","guideline","no","no","no","0","2","no","no","no","no","no","158.5","1","12","14","10","5","12","12","72"
+"200.76","69","male","partner","never","Active","guideline","no","yes","no","0","3","yes","no","no","no","no","231.5","0","5","6","9","9","6","12","76"
+"270.6","63","male","partner","never","Placebo","guideline","no","yes","no","0","3","no","no","no","no","no","86.11","3","11","13","11","8","14","9","76"
+"268.72","45","male","alone",NA,"Active","guideline","no","no","no","0","0","no","no","no","no","no","245.77","1","13","10","7","4","5","2","72"
+"215.8","64","male","partner","ever","Placebo","guideline","no","no","no","0","6","yes","no","no","no","no","250.3","1","13","12","15","9","6","5","60"
+"187.4","51","female","partner","ever","Active","more","no","yes","no","0","4","no","no","no","no","no","140.61","2","16","13","12","6","6","5","80"
+"66.04","77","female","alone","ever","Active","guideline","no","no","no","0","2","yes","no","no","no","no","50","0","7","4","4","4","6","2","100"
+"53.97","83","male","partner","never","Placebo","guideline","no","yes","no","0","1","no","no","no","no","no","66.5","2","10","17","18","10","4","4","72"
+"133.8","63","male","partner",NA,"Placebo",NA,"no","yes","yes","0","2","no","no","no","no","no","111","1","8","10","11","8","5","0","80"
+"54.51","71","male","alone","never","Placebo","guideline","no","yes","yes","0","5","no","no","no","no","no","50.02","1","11","13","12","13","11","6","60"
+"327.72","57","male","alone","never","Active","more","no","no","no","0","0","yes","no","no","no","no","206.05","1","16","17","9","5","15","24","44"
+"198.89","81","male","partner","ever","Active","guideline","no","yes","no","1","7","yes","no","no","no","no","164.51","1","10","6","5","5","8","5","80"
+"138.2","76","male","alone","ever","Placebo","guideline","yes","yes","no","1","6","no","no","no","no","no","185.92","2","10","11","4","7","9","2","100"
+"91.9","74","male","alone","never","Placebo","guideline","no","yes","no","0","8","yes","yes","no","no","no",NA,"6","8","9","14","5","4","2","80"
+"68.6","54","female","partner","never","Placebo","guideline","no","no","no","1","1","no","no","no","no","no","117.2","1","14","12","14","12","11","5","68"
+"137","44","male","partner","ever","Placebo","guideline","no","no","no","0","1","yes","no","no","no","no","255.14","0","8","14","16","10","8","5","72"
+"214.2","64","male","partner","never","Active","guideline","no","no","no","0","17","yes","no","yes","no","no","176.57","2","11","14",NA,"9","11","20","40"
+"85.8","52","male","partner","never","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","211","2","15","11","8","4","12","10","64"
+"216.47","48","female","partner","never","Placebo","guideline","no","yes","no","0","0","no","no","no","no","no","256.12","1","10","16","13","8","4","5","76"
+"207.53","85","male","partner","never","Placebo","guideline","yes","yes","no","0","6","yes","no","no","no","no","136","3","9","13","7","6","6","4","80"
+"27.2","61","male","alone","never","Placebo","guideline","no","yes","no","0","2","yes","no","no","no","no","65","1","13","17","15","4","6","1","68"
+"189.71","49","female","alone","never","Active","guideline","no","yes","no","0","0","no","no","no","no","no","52.2","2","20","16","16","7","18","18","24"
+"197","63","male","alone","ever","Active","guideline","no","no","no","0","2","no","no","no","no","no","231.73","2","15","7","11","10","11","11","52"
+"116","70","male","partner","never","Active","more","no","no","no","0","4","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"157.4","64","male","partner","ever","Active","guideline","no","yes","no","0","0","no","no","no","no","no","205","2","16","11","12","4","4","9","80"
+"88.2","63","male","partner","never","Placebo","guideline","no","no","no","1","28","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"50","85","female","alone","never","Active","guideline","yes","no","no","0","27","yes","no","no","no","no",NA,"6",NA,NA,NA,NA,NA,NA,NA
+"114.6","83","male","alone","ever","Active","guideline","no","no","yes","0","7","no","no","no","yes","no","106.8","1","8","6","8","8","4","4","80"
+"148.96","75","female","partner","never","Placebo","guideline","no","yes","no","0","1","yes","no","no","no","no","95.8","1","14","8","11","10","16","9","52"
+"100","72","male","partner","never","Active","guideline","no","yes","no","0","4","yes","no","yes","no","no","52.53","3","4","6","9","13","8","6","92"
+"195.92","64","female","alone","ever","Placebo","guideline","no","no","no","0","2","yes","no","no","no","no","158.51","1","6","7","7","4","6","5","80"
+"180.22","69","male","partner","never","Placebo","guideline","yes","no","no","0","5","yes","no","no","no","no","190","1","4","8","5","4","4","0","100"
+"116.12","65","male","partner","never","Active","guideline","no","yes","no","0","17","yes","no","yes","no","no","199.16","2","16","14","19","7","8","9","68"
+NA,"73","male","alone","never","Placebo","guideline","no","yes","yes",NA,NA,"no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"22.2","66","male","partner","ever","Active","guideline","yes","yes","yes","1","4","no","no","no","no","no","64.11","3","14","14","6","8","10","8","84"
+"29.73","79","male","alone","never","Active","guideline","no","yes","no","0","1","no","no","no","no","no","31.4","1","16","6","8","8","4","4","48"
+"216.86","69","male","partner","ever","Active","guideline","yes","no","no","1","6","yes","no","yes","no","no","252.43","1","6","4","4","5","4","3","76"
+"226.8","55","male","partner","never","Active","guideline","no","no","yes","0","0","no","no","no","no","no","142.4","1","10","12","15","8","8","6","92"
+"166","81","male","alone","never","Active","more","no","no","no","0","3","yes","no","no","no","yes","79.7","4","4","9","10","4","5","5","92"
+"131.8","69","male","alone","never","Placebo","guideline","no","yes","no","0","16","yes","no","yes","no","no","9.03","4","17","17","16","8","12","5","100"
+"161.5","53","female","partner","ever","Placebo","guideline","yes","no","no","0","2","no","no","no","no","no","302.77","0","16","10","7","4","6","8","80"
+"256","74","male","partner","ever","Active","guideline","yes","yes","no","0","0","no","no","no","no","no","311.8","1","4","8","6","6","8","1","100"
+"238.6","62","male","partner","ever","Active","more","no","no","no","0","12","yes","no","no","no","no","288.93","1","12","14","13","6","6","7","64"
+"252.8","51","male","partner","never","Placebo","guideline","no","no","no","0","5","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"84.6","45","male","partner","never","Placebo","guideline","no","yes","yes","0","3","no","no","no","no","no","73.61","2","16","17","17","12","10","14","60"
+"341.4","54","male","partner","never","Placebo","more","no","no","no","0","3","no","no","no","no","no",NA,"0","20","14","17","11","12","29","16"
+"117.8","87","male","partner","never","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","131.8","2","11","9","13","7","4","4","72"
+"113.21","79","female","partner","ever","Active","more","no","no","no","0","5","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"82.36","69","male",NA,"never","Active","guideline","no","yes","no","0","2","no","no","no","no","no","254.51","0","12","6","9","8","8","2","76"
+"30","62","male","partner","never","Placebo","more","no","no","no","0","2","yes","no","no","no","yes","95","2","10","14","11","9","7","13","60"
+"241.3","45","male","partner","ever","Placebo","guideline","no","no","no","0","32","yes","no","yes","no","no","315.97","2","11","7","8","4","7","14","80"
+"137.4","83","female","partner","ever","Active","guideline","no","no","no","1","4","no","no","no","no","no","155.01","3","13",NA,"9","7",NA,NA,"52"
+"260.5","50","male","partner","ever","Active","guideline","no","no","no","0","4","no","no","no","no","no","238.05","1","11","7","11","10","6","2","76"
+"289","63","male","partner","never","Placebo","more","yes","no","no","0","6","yes","no","no","no","no","253.4","1","14","17","15","9","14","10","64"
+"30","77","female","partner","ever","Active","guideline","no","no","no",NA,"7","no","no","no","no","no","31.72","4","15","20","16","4","4","9","76"
+"25","77","male","alone","never","Active","guideline","yes","yes","yes","2","2","no","no","no","no","no","0","2","8","11","13","12","17","6","92"
+"130.56","37","female","alone","ever","Placebo","guideline","no","no","no","0","1","no","no","no","no","no","288.72","1","16","14","10","8","10","13","32"
+"59.8","81","male","alone","never","Active","more","no","yes","no","0","2","no","no","no","no","no","154.84","1","4","6","6","6","4","0","100"
+"117.13","71","male","partner","ever","Active","guideline","no","yes","no","0","3","yes","no","no","no","no","89.6","2","12","12","10","4","8","1","100"
+"286.47","50","male","partner","never","Active","guideline","no","no","no","0","3","yes","no","no","no","no","306.55","1","12","9","16","11","10","5","56"
+"117.63","72","female","partner","ever","Active","guideline","no","yes","no","1","4","yes","no","no","no","yes","191.4","1","9","8","8","4","4","3","68"
+"27.2","80","female","alone","never","Active","guideline","no","no","no","1","14","no","no","no","no","no","4.84","4","20","19","20","4","4","14","64"
+"211.4","51","male","partner","ever","Active","guideline","no","no","no","0","1","yes","no","no","no","no","288.4","2","14","19","18","8","13","11","52"
+"100","73","female","partner","ever","Active","guideline","no","yes","no","0","3","no","no","no","no","no","102.89","2","15","16","20","12",NA,"19","56"
+"299.19","52","male","partner","ever","Placebo","guideline","no","yes","no","1","7","yes","no","no","no","no","149.05","1","11","10","14","11","8","7","64"
+"75.8","64","male","partner","never","Placebo","more","no","no","no","0","4","no","no","no","no","no","224.62","1","8","4","4","5","8","10","88"
+"25","77","male","partner","never","Active","more","no","yes","no","2","6","no","yes","no","no","no","116.59","3","13","15","9","11","13","16","60"
+"90","58","male","alone","never","Active","guideline","yes","yes","yes","0","0","no","no","no","no","no","55","1","15","18","12","10","9","13","60"
+"147.13","74","male","partner","never","Placebo","more","no","yes","no","0","5","no","no","no","no","no","135.1","2","15","14","11","7","4","5","72"
+"131.8","69","male","partner","never","Placebo","guideline","yes","no","no","2","3","no","no","no","no","no","167.45","2","5","12","10","5","4","1","92"
+"104.16","65","female","partner","ever","Placebo","guideline","yes","yes","no","0","12","yes","no","yes","yes","no","156.58","1","18","14","20","14","4","16","44"
+"54.51","85","male","partner","never","Placebo","more","no","yes","yes","0","2","no","no","no","no","no",NA,"2","20","16","19","20","17","14","16"
+"50","81","male","alone","never","Placebo","guideline","no","no","no","0","2","no","no","no","no","no","115.43","1","14","14","13","11","5","6","72"
+"297.4","71","male","partner","never","Active","guideline","no","no","no","1","5","no","no","no","no","no","149.4","2","9","15","9","6","12","5","88"
+"131.8","67","male","partner","never","Placebo","guideline","no","yes","yes","0","5","yes","no","no","no","no","244.23","0","10","5","6","4","11","2","84"
+"190","61","male","alone","ever","Placebo","guideline","no","no","no","0","2","no","no","no","no","no","230.37","1","10","8","12","8","11","3","88"
+"191.81","58","male","alone","never","Active","more","no","no","no","0","1","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"248.3","52","male","alone","never","Active","guideline","no","no","no","0","1","no","no","no","no","no","161.8","1","14","9","10","7","11","6","64"
+"348.97","49","male","partner","ever","Placebo","guideline","no","no","no","1","1","yes","no","no","no","no","255.66","0","7","4","6","5","5","3","80"
+"252","60","male","partner","ever","Active","guideline","no","yes","no","0","1","no","no","no","no","no","226.08","2","14","9","8","9","14","4","68"
+"85","67","male","partner","never","Placebo","guideline","no","yes","no","0","2","yes","no","no","no","no","225.11","1","4","5","7","7","6","0","92"
+"161","49","male","partner","never","Placebo","guideline","no","no","no","0","2","no","no","no","no","no","259.5","1","5","10","4","4","5","6","84"
+"87.47","45","male","partner","never","Active","guideline","no","yes","no","0","2","no","no","no","no","no","167.07","1","4","7","5","5","4","3","96"
+"167.8","83","male","partner","never","Active","guideline","yes","yes","no","0","5","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"86.72","72","female","partner","ever","Active","guideline","no","no","no","0","2","no","no","no","no","no","109.25","1","10","6","8","9","8","3","64"
+"116.75","67","male","alone","never","Placebo","guideline","no","yes","yes","0","4","no","no","no","yes","no","62.36","3","17","18","20","16","8","15","52"
+"183.91","78","male","partner","never","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","72.91","1","5","5","5","4","5","0","92"
+"269.28","63","male","partner","never","Placebo","guideline","no","yes","no","0","13","yes","no","no","no","no","184.72","0","6","6","6","5","7","12","72"
+"215.8","51","female","partner","ever","Active","guideline","yes","no","no","0","2","no","no","no","no","no",NA,"2","17","16","15","15","9","20","48"
+"91","72","female","partner","never","Active","guideline","no","no","no","1","7","yes","no","no","no","no","76.4","0","10","6","10","4","12","3","88"
+"221","44","female","alone","never","Active","guideline","no","no","yes","0","3","yes","no","no","no","no","146","2","18",NA,"16","9",NA,"17","44"
+"49.73","66","female","partner","never","Placebo","more","yes","no","no","0","18","yes","no","yes","no","no","87.09","2","12","9","19","8","8","13","80"
+"184.6","48","female","partner","never","Active","guideline","no","no","no","0","5","yes","no","yes","no","no","146.66","1","14","20","13","8","8","19","60"
+"64.35","76","female","partner","never","Active","guideline","yes","yes","no","2","17","yes","no","yes","no","no","71.48","2","18","18","17","11","6","20","48"
+"132.32","76","female","alone","ever","Active","guideline","no","yes","no","0","3","no","no","no","no","no","70.93","0","9","7","7","4","4","1","88"
+"190.67","54","female","partner","never","Placebo","guideline","no","yes","no","0","4","no","no","no","no","no","243.56","1","12","9","9","5","4","5","56"
+"58.6","83","female","partner","ever","Active","guideline","no","yes","no","0","10","yes","no","no","no","no","19.45","4","11","17","16","4","4","7","88"
+"316.76","44","male","partner","never","Placebo","guideline","no","yes","no","0","4","no","no","no","no","no","281.83","1","4","4","7","4","4","1","100"
+NA,"84","female","partner","ever","Active","guideline","no","yes","yes","0","2","no","no","no","no","no","58.6","2","12","16","20","8","7","7","80"
+"147.1","37","male","partner","ever","Active","guideline","no","no","no","0","2","yes","no","no","no","no","227.83","1","14","9","9","11","12","9","72"
+"132.4","69","male","partner","never","Active","guideline","no","yes","no","0","4","no","no","no","no","no","88.3","1","11","16",NA,"6","12","18","68"
+"192.4","81","male","partner","never","Placebo","guideline","no","no","no","1","3","no","no","no","no","no","108.34","0","10","12","6","5","5","1","92"
+"237.2","80","male","partner","never","Placebo","guideline","no","no","no","0","2","no","no","no","no","no","322.4","1","7","5","4","4","10","4","84"
+"85","75","female","alone","never","Placebo","guideline","no","yes","no","0","7","no","no","no","no","no","60.91","2","14","7","4","4","4","12","88"
+"106.93","84","male","partner","never","Active","more","yes","no","no","0","3","no","no","no","no","no","207.86","2","12","20","4","4","16","5","76"
+"221","57","female","alone","never","Placebo","guideline","no","yes","no","0","11","yes","no","no","yes","no","205.72","3","18","13","6","14","13","7","28"
+"25.8","48","male","partner","ever","Active","guideline","no","no","no","0","1","no","no","no","no","no","281.41","0","9","9","11","6","8","3","68"
+"30","87","female","alone","never","Active","guideline","no","yes","no","1","2","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"179.73","57","female","partner","never","Active","guideline","no","yes","no","0","2","no","no","no","no","no","240.38","1","17","14","16","13","4","15","36"
+"293.12","46","male","partner","ever","Placebo","guideline","yes","no","yes","0","1","yes","no","no","no","no","335.46","1","8","14","10","8","9","7","72"
+"60","65","female","partner","ever","Active","guideline","no","yes","no","0","4","no","no","no","no","no","98.31","1","12","12","14","12","11","5","72"
+"88.77","61","male","partner","never","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","243.98","2","10","15","10","5","12","12","84"
+"202.72","45","female","partner","ever","Placebo","guideline","no","no","no","0","1","yes","no","no","no","no","248.13","1","9","4","4","4","7","5","88"
+"147.35","76","male","alone","never","Placebo","guideline","yes","yes","no","0","2","yes","yes","no","yes","no","245.8","2","10","8","9","4","4","13","80"
+"172.2","63","male","alone","never","Placebo","guideline","yes","no","no","0","19","no","no","no","no","no",NA,"6",NA,NA,NA,NA,NA,NA,NA
+"282.7","48","male","partner","ever","Active","guideline","no","no","no","0","18","yes","no","yes","no","no","67.27","3","20","5","14","8","15","14","20"
+"254.6","64","female","partner","never","Placebo","guideline","no","no","no","0","11","yes","no","no","no","no","125.38","4","11","6","5","4","4","1","96"
+"149.71","64","female","alone","never","Placebo","guideline","no","no","no","0","1","no","no","no","no","no",NA,"0","4","8","8","4","4","0","96"
+"141.8","60","male","alone","never","Active","guideline","no","no","yes","1","6","no","no","no","no","no","118.06","4","7","10","4","4","4","2","100"
+"256","54","male","partner","never","Active","guideline","no","yes","no","0","5","yes","no","no","no","no","158.43","0","5","8","10","7","5","7","72"
+"163.55","68","male","partner","never","Active","guideline","yes","yes","no","0","6","no","no","no","yes","no","232.27","1","16","13","13","8","4","3","76"
+"40","66","female","partner","ever","Placebo","guideline","no","yes","no","0","4","yes","no","no","no","no","85","0","9","13","12","9","4","5","64"
+"37.84","84","female","alone","never","Placebo","guideline","no","yes","no","0","1","no","no","no","no","no","98.13","1","12","9","8","7","6","2","80"
+"106.2","72","male","partner","ever","Placebo","guideline","yes","yes","no","0","11","yes","no","no","no","no","93.5","1","12","10","18","8","6","7","40"
+"142.15","33","female","alone","ever","Placebo","guideline","no","no","no","0","2","no","no","no","no","no","56.87","1","19","16","14","14","18","15","44"
+NA,"70","male","partner","ever","Active","guideline","no","no","no","0","19","yes","no","no","no","no","62.87","3","11","9","11","9","10","7","84"
+"52.36","88","male","alone","ever","Active","more","yes","no","no","0","12","no","no","no","no","no","25.8","3","7","18","15","12","10","4","72"
+"191.8","84","male","partner","ever","Placebo","guideline","no","no","no","1","1","no","no","no","no","no","64.6","4","20","20","20","10","16","19","4"
+"75.8","45","male","partner","never","Placebo","guideline","no","no","no","0","3","yes","no","no","yes","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"252","74","male","partner","never","Placebo","guideline","no","yes","no","0","0","no","yes","no","no","no","224.64","0","8","5","4","4","5","3","80"
+"195.5","63","male","alone","never","Active","guideline","no","no","no","0","5","no","no","no","no","no","175.58","3","11","11","6","4","6","5","80"
+"151.4","62","male","partner","never","Placebo","guideline","no","no","yes","0","9","no","no","no","no","no","194.9","3","10","8","8","4","9","3","84"
+"123.29","67","female","partner","never","Active","guideline","no","no","no","0","3","yes","no","no","no","no","223.92","1","8","6","10","8","8","5","76"
+"110.8","66","male","alone","never","Placebo","guideline","no","yes","no","0","3","yes","no","no","no","no","38.27","2",NA,NA,NA,NA,NA,NA,NA
+"0","65","male","alone","never","Placebo","guideline","no","yes","yes",NA,NA,"no","no","no","yes","no","25","4","10","16","10","4","6","2","72"
+NA,"63","male","alone","never","Active","guideline","no","yes","no","0","16","no","yes","no","yes",NA,"21.35","5","16","14","12","7","14",NA,"20"
+"222.8","58","male","partner","never","Active","guideline","no","yes","no","0","2","no","no","no","no","no","377.58","2","12","10","5","8","9","6","80"
+"184.49","79","female","partner","ever","Placebo","guideline","no","yes","no","0","3","yes","no","no","no","no",NA,"0","7","6","6","4","4","1","96"
+"191","60","male","alone","never","Active","guideline","no","yes","no","1","3","yes","no","no","no","no","141.08","2","4","6","6","4","4","1","96"
+"184.47","54","female","partner","ever","Active","guideline","no","yes","no","0","1","no","no","no","no","no","277.76","2","15","18","15","10","4","12","52"
+"225.6","65","female","partner","ever","Active","guideline","no","no","no","0","19","yes","no","yes","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+NA,"52","male",NA,NA,"Placebo",NA,NA,NA,"no","0","10","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA,NA
+"92.19","77","female","partner","never","Active","guideline","yes","yes","no","0","0","no","no","no","yes","no","56.56","2","9","13","10","8","10","24","40"
+"53.11","72","female","alone",NA,"Placebo","guideline","no","no","no","0","4","yes","no","no","no","no","277.5","1","4","4","5","4","8","0","100"
+"161","61","male","partner","never","Placebo","guideline","yes","yes","no","0","1","no","no","no","no","no","140.83","0","8","11","12","10","8","2","68"
+"75","62","female","partner","never","Placebo",NA,"no","no","no","0","24","no","no","no","no","no","27.09","4","14",NA,"15","9","9","10","64"
+"143.2","66","female","partner","ever","Placebo","guideline","no","no","no","0","9","yes","no","no","no","yes","111.4","2","13","8","8","7","11","1","88"
+"61.72","50","male","alone","never","Active","guideline","no","yes","no","0","0","no","no","no","no","no",NA,"2","12","20","13","16","6","17","60"
+"33.6","82","female","partner","never","Placebo","guideline","no","yes","no","0","19","no","no","no","no","no",NA,"4","15","20","18","10","12","28","4"
+"75","70","female","partner","ever","Placebo","guideline","no","no","no","0","3","no","no","no","no","no","93.14","1",NA,"13","11","11","9","5","76"
+"123","69","male","partner",NA,"Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","58.6","1","7","14","10","6","6","3","84"
+"145","58","male","partner","ever","Placebo","guideline","yes","no","no","0","14","no","no","no","no","no","276.47","2","10","12","6","4","6","7","44"
+"52.2","72","female","alone","never","Active","guideline","no","no","no","0","4","yes","no","no","no","no","31.4","1","4","4","7","5","4","0","100"
+"141.2","79","male","partner","ever","Placebo","guideline","no","no","no","0","4","yes","no","no","no","no","107.3","2","12","16","9","6","10","13","72"
+"144.6","79","male","partner","ever","Placebo","guideline","no","no","no","2","9","no","no","no","no","no",NA,"6","18","14","20","11","16","11","40"
+"65","67","male","alone","never","Placebo","guideline","no","yes","yes","2","2","no","no","no","no","no","97.13","2","16","14","14","12","12","7","56"
+"110.8","87","male","partner","never","Placebo","guideline","no","no","no","0","4","yes","no","no","no","no","33.6","1","10","14","9","12","4","5","68"
+"131.8","67","male","partner","never","Placebo","guideline","yes","yes","no","2","3","yes","no","no","yes","no","70.8","0","7","10","10","13","4","4","76"
+"52.2","19","female","alone","never","Placebo","guideline","no","no","no","0","19","yes","no","yes","no","no","78.85","1","8","8","5","4","5","3","84"
+"106.8","83","female","alone","ever","Active","guideline","no","yes","no",NA,"16","yes","no","no","no","no","27.2","3",NA,NA,NA,NA,NA,NA,NA
+"27.53","72","female","alone","never","Placebo","guideline","no","yes","no",NA,"1","yes","no","no","no","no","39.4","1","14","13","10","6","11","3","64"
+"258.4","72","male","partner","ever","Active","guideline","yes","yes","yes","0","4","yes","no","no","no","no","65","2","6","6","9","4","4","3","84"
+"168.37","76","male","partner","never","Placebo","guideline","no","yes","no","0","1","no","no","no","no","no","164.8","2","11","13","8","8","11","2","100"
+"17.63","83","female","alone","never","Active","guideline","no","yes","no",NA,"2","no","no","no","no","no","8.6","3","12","12","17","8","8","9","72"
+"52.2","91","female","alone","ever","Placebo","guideline","no","yes","no","1","5","no","no","no","no","no","63.13","2","12","14","12","6","7","6","72"
+"51.4","67","male","partner","never","Placebo","guideline","no","no","no","0","5","yes","no","no","no","no","68.43","1","10","12","11","9","4","3","88"
+"144.6","72","male","partner","never","Active","guideline","no","no","no","1","20","yes","no","yes","no","no",NA,"6",NA,NA,NA,NA,NA,NA,NA
+"109.25","89","male","partner","ever","Active","guideline","yes","no","no","0","3","yes","no","no","no","no","33.6","2","11","12","13","8","10","4","84"
+"95.42","64","male","alone","never","Placebo","guideline","no","yes","no","0","1","no","no","no","no","no","85.82","1","15","15","16","8","10","16","60"
+"146","57","female","alone","never","Active","guideline","no","no","no","0","7","no","no","no","no","no","252.8","1","10","7","5","8","9","3","80"
+"179.2","80","male","partner","never","Active","guideline","no","yes","no","0","8","no","no","no","no","no","86","2","13","14","20","11",NA,NA,"48"
+"229.9","76","male","partner","ever","Active","guideline","no","no","no","0","3","yes","no","no","no","no","195.8","1","7","5","6","6","6","2","100"
+"241","57","male","partner","never","Placebo","guideline","no","no","no","0","8","no","no","yes","no","no","282","0","4","9","4","4",NA,"1","92"
+"186.8","60","male","alone","never","Active","guideline","no","yes","no","0","2","no","no","no","no","no","154.18","2","11","8","12","10","9","7","52"
+"163.09","81","female","alone","never","Active","guideline","no","yes","no","0","5","yes","no","no","no","no","121","1","12","7","5","4","4","2","88"
+"152.53","48","male","alone","never","Placebo","guideline","no","no","no","0","5","no","no","no","no","no","108.76","3","15","17","16","10","12","16","52"
+"98.63","86","male","alone","never","Active","guideline","no","yes","yes","2","3","yes","no","no","no","no","41.38","2","12","13","16",NA,"19","34","24"
+"50.8","74","male",NA,NA,"Active",NA,NA,NA,"no","0","1","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA,NA
+"84.83","67","female","partner","ever","Placebo","guideline","no","no","no","0","6","yes","no","no","no","no","98.85","0","8","8","4","4","9","6","92"
+"254.15","64","male","partner","never","Placebo","guideline","no","no","yes","0","5","yes","no","no","no","no","107.31","2","7","6","4","4","5","2","88"
+"221.4","61","male","partner","never","Active","guideline","no","yes","no","0","3","no","no","no","no","no","117.4","2","9","15",NA,"6","7","5","60"
+"291.47","58","male","partner","never","Placebo","more","no","no","no","0","7","no","no","no","no","no","484.8","1","9","7","4","4","4","3","88"
+"27.2","83","male","partner","never","Placebo","guideline","no","yes","no","1","1","no","no","no","no","no","121.8","2","13","12","7","6","5","2","88"
+"208","68","male","partner","never","Placebo","guideline","no","no","yes","0","2","yes","no","no","yes","no","207.5","0","7","14","8","6","9","4","92"
+"178.6","53","male","alone","never","Active","more","no","no","no","0","14","no","no","no","no","no","63.27","4","12","14","11","8","10","10","32"
+NA,"52","male","partner","never","Active","guideline","no","yes","no","0","5","no","no","no","no","no",NA,"6",NA,NA,NA,NA,NA,NA,NA
+"197.3","84","female","alone","ever","Active","guideline","yes","yes","no","0","0","no","no","no","no","no","80","1","13","16","15","12","7","7","64"
+"331.59","63","male","partner","never","Placebo","guideline","yes","no","no","0","15","yes","no","no","no","no","312","1","6","4","4","5","10","2","88"
+"111","58","male","alone","ever","Placebo","guideline","no","no","no","0","4","no","no","no","no","no","219.54","2","10","8","9","8","7","1","100"
+"30","89","female","alone","never","Placebo","more","no","yes","no","0",NA,"no","no","no","no","no",NA,"6",NA,NA,NA,NA,NA,NA,NA
+"78.33","84","female","alone","never","Active","guideline","no","yes","no","2","0","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"14.21","79","male","partner","never","Active","guideline","yes","no","no","2","6","no","yes","no","no","no","135.42","4","7","11","12","9","4","0","100"
+"95.8","86","female","alone","never","Active","guideline","no","yes","no","0","2","no","no","no","no","no","50.8","2","15","15","15","9","8","14","68"
+"75.39","73","male","partner","ever","Placebo","guideline","yes","yes","no","1","8","no","no","no","no","no",NA,"1","4","8","4","4","4","2","100"
+"170.25","69","female","partner","ever","Placebo","guideline","no","no","no","0","4","yes","no","no","no","yes","193.3","0","12","10","10","9","4","1","92"
+"61","70","male","partner","never","Placebo","guideline","no","no","no","0","3","yes","no","yes","no","no","220.8","0","6","6","7","7","5","6","80"
+"236.8","49","male","partner","never","Placebo","guideline","no","no","no","0","3","no","no","no","no","no","227.05","1","9","12","10","4","7","7","76"
+"163.2","53","male","partner","never","Active","guideline","no","no","no","0","5","no","no","no","no","no","179.49","1","13","20","19","8","7","14","68"
+"29.51","77","female","alone","never","Active","guideline","yes","yes","no","0","3","yes","no","no","no","no","61.96","0","10","16","11","13","11","3","76"
+"206.4","51","male","partner","never","Placebo","guideline","no","no","yes","0","7","yes","no","no","yes","no",NA,"2","17","15","16","14","14","30","20"
+"116","74","male","partner","never","Placebo","guideline","yes","yes","no","0","7","yes","no","no","yes","no","123.81","0","4","4","6","5","4","1","84"
+"323.17","58","male","partner","ever","Placebo","guideline","no","no","no","0","0","no","no","no","no","no","383","1","10","7","6","8","6","3","80"
+"107.81","79","male","partner","never","Placebo","guideline","no","no","no","0","3","no","no","no","no","no","163.08","1","16","15","19","10","13","5","68"
+"211.4","48","male","partner","never","Active","guideline","no","no","no","0","2","no","no","no","no","no","210.16","1","13","16","16","12","11","4","60"
+"157.4","78","male","partner","ever","Placebo","guideline","no","no","no","0","4","no","no","no","no","no","117.15","0","6","12","8","8","10","10","64"
+"126.64","52","male","partner","ever","Placebo","guideline","no","yes","no","0","4","no","no","no","no","no","251.15","2","6","12","9","4","4","2","100"
+"69.36","78","female","alone","never","Active","guideline","no","no","no","0","5","yes","no","no","no","no","128.04","1","20","17","20","18","15","23","32"
+"107.52","78","male","partner","never","Placebo","more","yes","yes","no","2","2","no","no","no","no","no","88.78","2","14","6","10","7","4","5","68"
+"60","69","male","partner","never","Active","guideline","no","yes","yes","0","8","yes","no","no","yes","no","107.16","1","13","4","16","6","16","10","60"
+"169.6","60","female","alone","ever","Placebo","guideline","no","no","yes","0","3","no","no","no","no","no","196.32","2","5","5","13","5","4","6","76"
+"106","68","male","partner","never","Active","guideline","yes","no","no","0","12","yes","no","yes","no","no","127","0","6","6","8","7","4","0","88"
+"132.5","68","female","partner","never","Placebo","guideline","no","no","no","0","2","no","no","no","no","no","131.16","0","4","5","5","5","4","0","92"
+"141.55","73","female","alone","ever","Active","guideline","no","yes","no","0","2","no","no","no","no","no","106.55","2","16","14","14","12","8","8","48"
+"247.23","37","female","partner","never","Active","guideline","no","no","no","0","1","no","no","no","no","no","172.14","1",NA,NA,"9","9","14","5","64"
+"71.72","44","male","alone","never","Active","guideline","no","no","no","0","12","yes","no","yes","no","no","50.8","2","19","17","17","15","15","24","16"
+"217.05","79","male","partner","never","Active","guideline","no","yes","no","0","3","no","no","no","no","no","147.74","2","10","11","12","10","7","3","60"
+"140.75","56","male","partner","ever","Placebo","guideline","no","yes","no","0","5","no","no","no","no","no","120.56","0","12","9","6","6","7","4","76"
+"95.8","84","female","alone","ever","Active","guideline","no","yes","no","0","3","no","no","no","no","yes",NA,"0","12","12","12","12","12","50","100"
+"140.85","73","male","partner","never","Active","guideline","no","no","no","0","2","yes","no","no","no","no","290.77","1","7","9","4","4","4","6","80"
+"2.2","71","male","partner","never","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","50.8","2","11","10","16","12","6","10","84"
+"56.4","67","male","partner","ever","Active","guideline","yes","no","no","1","2","no","no","no","no","no","103.12","2","9","10","12","8","9","4","84"
+"59.24","80","male","partner","never","Active","guideline","no","yes","no","0","3","no","no","no","yes",NA,NA,"3","18","16","20","20","12","29","8"
+"77","77","female","partner",NA,"Placebo","guideline","no","yes","no","0","9","yes","yes","no","yes","no","100","1","12","10","10","6","8","11","80"
+"188.8","49","female","partner","never","Placebo","guideline","no","yes","no","0","3","no","no","no","no","no","78.65","1","12","16","10","8","4","4","76"
+"163.2","63","female","partner","never","Placebo","guideline","no","no","no","0","3","no","no","no","no","no","231.54","2","14","9","12","9","15","20","56"
+"0","88","female","partner","ever","Placebo","guideline","no","yes","no","0","2","yes","no","no","yes","no","25","0","7","5","7","6","4","0","96"
+"50","86","female","alone","ever","Placebo","guideline","no","no","no","2","4","no","no","no","no","no","52.2","2","13","15","11","15","12","5","72"
+"249.9","67","female","partner","never","Placebo","guideline","no","yes","no","0","3","no","no","no","no","no","111.1","2","12","11","12","8","8","10","64"
+"170.72","75","male","alone","never","Active","guideline","no","yes","no","0","4","no","no","no","no","no","170.12","1","6","4","4","4","8","2","92"
+"70","79","female","alone","never","Active","guideline","no","no","no","0","5","no","no","no","no","no","55","3","17","8","12","10","10","13","48"
+"144.13","80","female","alone","never","Placebo","guideline","no","no","no","0",NA,"no","no","no","yes","no","215.43","0","4","4","4","6","4","0","96"
+"315.91","68","male","partner","never","Placebo","guideline","no","no","no","0","1","no","no","no","no","no","241.05","1","5","4","9","4","6","4","88"
+"50.8","72","male","partner","ever","Active","guideline","no","no","yes","1","22","yes","no","no","no","no","44.6","4","6","7","6","10","18","9","88"
+"80.75","69","female","partner","ever","Placebo","guideline","no","yes","no","0","0","no","no","no","no","no","161.07","0","6","6","5","6","7","0","96"
+"25","97","female","alone",NA,"Active","guideline","yes","yes","no","2","24","no","no","no","no","no",NA,"5",NA,NA,NA,NA,NA,NA,"0"
+"277.11","54","male","partner","never","Placebo","guideline","no","no","no","0","3","yes","no","no","no","no",NA,"0","14","7","14","11","10","11","36"
+"112.4","47","female","partner","never","Placebo","guideline","no","no","no","0","2","no","no","no","no","no","52.2","0",NA,NA,NA,NA,"4","3","100"
+"105","64","female","partner","never","Active",NA,"yes","no","no","0","23","no","no","yes","no","no","112.57","4","7","7","4","4","6","1","92"
+"181.32","54","female","partner","never","Placebo","guideline","no","no","no","2","5","yes","no","no","no","no","158.84","2","13","12","10","8","4","1","68"
+"27.31","52","male","alone","never","Placebo","guideline","no","yes","no","1","2","no","no","no","no","no","151.92","2","7","15","11","5","8","8","52"
+"37.8","76","female","alone","never","Placebo","guideline","no","yes","yes","0","4","no","yes","no","no","no","25","1","12","13","13","13","12","7","56"
+"108.76","79","male","partner",NA,"Active","guideline","no","no","no","0","2","no","no","no","yes","no","300.85","2","4","8","9","6","5","2","92"
+"152.03","71","female","partner","ever","Active","guideline","no","yes","no","0","7","yes","no","no","no","no","222.55","1","7","5","6","4","12","8","100"
+"55","66","male","partner","never","Active","guideline","no","yes","no","0","20","yes","no","yes","yes","no","139.51","1","11","10","14","7","5","7","64"
+"80.25","80","male","partner","never","Placebo","guideline","yes","yes","no","2","9","no","no","no","no","no",NA,"6","6","13","20","18","15","13","32"
+"50","72","male","partner","never","Placebo","more","no","no","no","0","2","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"188.47","70","male","partner","never","Active","guideline","no","yes","yes","0","0","no","yes","no","yes","no","158.89","1",NA,NA,NA,NA,NA,"0","96"
+"236.9","64","female","partner","ever","Placebo","guideline","no","no","no","0","2","no","no","no","no","no","297.6","0","7","10","4","4","4","1","100"
+"300.72","57","male","partner","ever","Active","guideline","no","no","no","0","1","no","no","no","no","no","280.13","0","4","4","5","4","4","1","100"
+"116.28","77","male","partner","ever","Placebo","guideline","yes","yes","no","1","9","yes","no","no","no","no","170.74","4","9","10","12","4","5","8","76"
+"156.31","65","female","alone","never","Active","guideline","no","yes","no","0","6","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"50","73","male","alone","never","Placebo","guideline","no","no","no","2","0","no","no","no","no","no","27.2","1","19","19","17","18","8","20","28"
+"27.2","85","female","alone","ever","Active","guideline","no","no","no","1","1","no","no","no","no","no",NA,"1","19","13","18","11","4","26","28"
+NA,"79","female","alone","never","Placebo","guideline","yes","no","no",NA,"9","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"129.63","75","male","partner","never","Placebo","guideline","no","no","no","0","18","no","no","yes","yes","no",NA,"6","15","16","11","8","12","26","24"
+"82.63","75","female","partner","ever","Active","guideline","no","yes","no","0","4","no","no","no","no","no","119.68","2","16","16","14","7","12","10","76"
+"114.59","70","female","alone","ever","Active","guideline","no","no","yes","0","7","yes","no","no","no","yes","83.16","1","13","13","15","8","16","18","56"
+"117","70","male","partner","never","Placebo","guideline","yes","no","no","0","5","no","no","no","no","no","102.03","2","13","12","8","8","4","3","0"
+"99.92","82","male","alone","never","Placebo","guideline","no","no","no","0","4","no","yes","no","no","no","98.63","0","14","15","11","6","8","8","72"
+"323.68","56","male",NA,NA,"Active",NA,NA,NA,"no","0","3","no","no","no","no",NA,"448.38","2","5","8","4","4","4","1","96"
+"76.4","84","female","alone","ever","Placebo","guideline","no","yes","yes","0","0","no","no","no","no","no","56.4","1","16","17","12","14","12","7","68"
+"95.75","69","female","alone","ever","Active","guideline","no","yes","no","0","13","yes","no","no","no","no","82.09","4","8","7","6","4","4","3","88"
+"160.56","67","male","partner","never","Active","more","no","no","no","0","2","no","no","no","no","yes","349.48","2","17","17","13","9","17","9","72"
+"162.6","59","male","partner","ever","Placebo","guideline","no","no","no","0","1","no","no","no","no","no","249.67","1","4","9","4","5","4","1","88"
+"151","69","male","partner","never","Placebo","guideline","no","yes","no","0","1","yes","no","yes","no","no","254.61","1","13","10","12","8","10","11","56"
+"98.2","40","male","partner","never","Placebo","more",NA,"yes","no","0","2","yes","no","no","no","no","315.11","0","7","12","13","5","4","9","80"
+"148.15","72","male","partner","never","Active","guideline","no","yes","no","0","1","no","no","no","no","no","136","1","12","16","20","12","4","8","68"
+"111.8","73","male","partner","ever","Placebo","guideline","no","yes","yes","0","1","no","no","no","no","no","121","1","10","4","12","8","4","3","88"
+"179.23","50","male","partner","ever","Placebo","guideline","no","no","no","0","0","no","no","no","no","no","336.27","0","5","4","4","5","4","0","92"
+"107","44","female","alone","never","Placebo","guideline","no","yes","yes","1","5","no","no","no","no","no","70","1","12","13","11","10","10","6","68"
+"229.5","72","male","partner","never","Placebo","guideline","yes","yes","no","0","5","no","no","no","no","no","58.6","2","10","4","12","5","8","18","48"
+"365.28","64","male","partner","never","Placebo","more","no","no","no","0","9","yes","no","no","no","no","142.93","2","8","8","8","4","10","2","100"
+"412.9","55","male","partner","never","Active","guideline","no","no","no","0","5","yes","no","no","no","no","222.84","3","10","7","10","8","4","3","100"
+"108.31","67","female","partner","never","Active","guideline","no","yes","no","0","3","no","no","no","no","no","116.25","1","15","12","13","9","9","8","72"
+"153.63","86","male","partner","never","Active","guideline","yes","no","no","0","4","no","no","no","no","no","108.2","1","4","4","6","6","5","0","92"
+"111.8","78","female","alone","ever","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"101.16","74","male","alone","ever","Active","guideline","no","no","no","0","3","no","no","no","no","no","96","3","6","10","11","8","10","2","96"
+"157","73","male","partner","never","Active","guideline","no","yes","no","1","3","no","no","no","no","no",NA,"0","8","6","5","5","4","1","88"
+"290.84","61","male","partner","ever","Active","guideline","no","no","no","0","2","no","no","no","no","no","262","1","9","8","9","4","4","0","88"
+"88.2","74","female","alone","ever","Placebo","guideline","no","yes","no","0","1","no","no","no","no","no","130","1","17","10","16","12","12","16",NA
+"170.89","45","male","alone","ever","Active","guideline","yes","yes","no","0","13","no","no","no","no","no","170.84","1","6","10","10","8","4","0","52"
+"107.72","82","male","partner","ever","Placebo","guideline","no","yes","no","0","10","yes","no","yes","no","no","138.38","1","9","10","11","9","9","4","64"
+"225.6","61","male","partner","never","Placebo","guideline","no","no","no","0","1","no","no","no","no","no","219.55","0","9","7","8","9","11","6","60"
+"235.16","57","male","partner","never","Active","guideline","yes","no","no","0","0","no","no","no","no","no","255.33","0","9","6","8","7","6","2","72"
+"50","82","female","partner","never","Active","guideline","no","no","no","0","17","yes","no","yes","no","no","30.5","2","15","15","13","4","4","8","84"
+"99.55","74","female","alone","never","Active","guideline","no","yes","no","0","7","no","no","no","no","no","59.03","4","13","10","14","8","4","8","76"
+"105.61","80","female","alone","never","Active","guideline","no","yes","yes","1","0","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"29.46","78","male","partner","never","Placebo","more","no","yes","no","0","2","yes","no","no","no","no","167.32","1","10","17","4","4","4","4","92"
+"179.5","71","male","alone","ever","Placebo","guideline","no","yes","no","0","5","yes","no","no","yes","no","157.34","2","12","10","10","11","7","8","76"
+"245.11","56","male","alone","never","Active","guideline","no","no","no","0","2","no","no","no","no","no","92.4","1","7","6","5","7","4","8","84"
+"315.8","45","male","partner","never","Active","guideline","no","no","no","0","3","yes","no","no","no","no","378.79","0","6","6","13","7","5","5","72"
+"120","70","female","partner","ever","Active","guideline","yes","no","no","0","11","no","no","yes","no","no","78.6","1","17","17","18","9","5","26","56"
+"58.6","84","female","partner","never","Placebo","guideline","yes","yes","no","0","4","yes","no","no","no","no","204.55","2","17","15","13","5","10","7","56"
+"56.4","66","female","alone","never","Active","guideline","yes","no","no","0","12","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"159.11","54","male","partner","never","Active","guideline","no","yes","no","0","1","yes","no","no","yes","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"156.53","52","male","partner","ever","Active","guideline","no","yes","no","0","17","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"186.8","62","female","partner","ever","Active","guideline","no","yes","yes","0","8","yes","no","no","no","no","165.6","2","16","16","18","9","20","18","24"
+"142.36","67","male","partner","never","Active","more","no","no","no","0","2","no","no","no","no","no","213.2","2","4","4","8","8","4","5","100"
+"222.4","64","male",NA,"never","Active",NA,"no","no","no","0","0","no","no","no","no","no","192.4","2","9","15","9","7","4","2","96"
+"58.6","93","male","alone","never","Placebo","guideline","no","yes","no","1","11","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"62.79","89","male","partner","never","Placebo","guideline","no","yes","no","2","3","yes","no","no","no","no","122.81","2","14","16","18","14","12","10","36"
+"105.72","74","male","partner","never","Placebo","guideline","yes","yes","no","0","2","no","yes","no","no","no","140.26","1","10","9","6","8","5","8","76"
+"258.2","66","male","partner","never","Placebo","more","no","yes","no","0","2","no","no","no","no","no","138.25","1","8","8","11","9","7","6","76"
+"256","51","female","alone","never","Active","guideline","no","no","no","0","4","yes","no","yes","no","no","77.25","2","15","17","7","7","20","29","60"
+"52.2","69","male","alone","ever","Placebo","guideline","yes","yes","yes","0","1","no","no","no","no","no","108.2","2","4","9","11","12","6","3","100"
+"50","93","female","alone","ever","Placebo","guideline","no","yes","yes","2","1","no","no","no","no","no","25","4","15","19","16","7","5","5","76"
+"103.49","66","male","partner","never","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","21.93","4","13","19","19","13","13","40","56"
+"229.73","64","male","alone","never","Active","guideline","no","yes","no","0",NA,"no","no","no","yes","no","29.51","4","5","8","12","7","4","1","88"
+"196.8","71","male","partner","never","Placebo","guideline","no","yes","no","1","1","yes","no","no","no","no","146.8","1","12","13","10","6","6","2","76"
+"163.4","66","male","alone","ever","Active","guideline","no","no","no","0","2","no","no","no","no","no","140.8","1","7","5","4","7","4","1","88"
+"76.4","73","female","partner","never","Active","guideline","no","yes","no","0","4","no","no","no","no","no",NA,"2","18","8","15","4","19","3","88"
+"33.6","75","male","alone","never","Active","guideline","yes","no","no","0","2","no","no","no","no","no","106.8","1","7","9","10","7","4","1","84"
+"75","69","female","partner","never","Active","guideline","no","yes","no","0","1","no","no","no","no","no","25","2","16","19","20","12","9","12","72"
+"65","78","male","partner","never","Active","guideline","no","no","no","2","2","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"185.15","74","female","partner","never","Active","guideline","no","no","no","0","4","no","no","no","no","no","183.79","1","4","6","4","4","4","3","84"
+"155.52","68","male","partner","ever","Active","guideline","yes","yes","no","0","3","no","no","no","no","no","134.8","0","7","7","7","6","5","3","76"
+"34.56","78","female","alone","never","Placebo","guideline","no","yes","no","0","1","yes","no","no","no","no","103.7","2","10","12","11","10","8","2","96"
+"76.21","65","male","partner","never","Placebo","guideline","no","no","no","0","2","no","no","no","no","no","67.76","0","13","14","15","9","10","27","24"
+"65","70","male","alone","never","Active","more","no","yes","no","0","1","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"133.89","79","male","partner","never","Active","guideline","no","yes","yes","2","19","yes","no","yes","no","no","82.76","3","12","20","20","4","4","11","48"
+"50","75","female","partner","ever","Active","guideline","yes","yes","no","1","1","no","no","no","no","no","109.72","2","7","9","5","6","4","2","80"
+"111","72","male","partner","never","Active","guideline","no","yes","no","0","1","no","no","no","no","no","113.53","0","20","16","19","16","12","29","16"
+"213.6","56","male","partner","never","Placebo","guideline","no","yes","no","0","1","no","no","no","no","no","182.9","2","17","10","13","12","19","18","24"
+"50.8","56","male","alone","never","Placebo","guideline","no","yes","no","1","12","yes","no","no","no","no","68.3","2","18","19","15","14","12","21","28"
+"213.25","55","male","partner","never","Active","guideline","no","yes","no","0","1","no","no","no","no","no","224.53","1",NA,"8","8","7","11","8","76"
+"145.56","80","male","partner","never","Active","guideline","yes","no","no","0","6","no","no","no","no","no","150.03","2","7","9","16","7","4","1","84"
+"50","80","male","alone","ever","Active","guideline","yes","no","no","0","8","yes","no","no","no","no","79.03","2","17","19","18","14","16","9","52"
+"76.4","66","male","partner","never","Active","guideline","yes","no","no","0","2","no","no","no","no","no","161","0","13","14","15","13","10","9","60"
+"136","63","female","partner","ever","Active","guideline","no","yes","yes","0","2","no","no","no","no","no","239.04","1","11","8","4","4","4","6","68"
+"230.9","54","male","partner","ever","Active","guideline","no","yes","no","0","2","no","no","no","no","no","143.53","1","8",NA,"11","7","7","3","72"
+"248.25","45","male","partner","ever","Active","guideline","yes","no","no","0","0","no","no","no","no","no","153.25","1","12","6","8","13","7","8","88"
+"101","88","female","alone","ever","Active","guideline","no","yes","no","0","4","no","no","no","no","no",NA,"3",NA,NA,NA,NA,NA,NA,NA
+"113.4","71","male","partner","never","Active","guideline","no","yes","no","0","10","yes","no","no","no","no","125.4","0","4","9",NA,"9","8","0","96"
+"113.52","73","male","partner","never","Placebo","guideline","no","no","yes","2","19","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"141.8","67","male","partner","never","Active","guideline","no","yes","no","0","7","yes","no","no","yes","no","218.39","1","18","14","16","15","17","37","48"
+"170.8","64","male","partner","never","Placebo","guideline","no","yes","yes","0","5","yes","no","no","no","no","210.24","1","12","8","16","12","4","2","80"
+"50.8","58","female","partner","never","Active","guideline","no","yes","no","1","4","no","no","no","no","no","35","3","16","13","19","10","15","37","40"
+"78.33","71","male","alone","never","Active","more","no","no","no","1","4","no","no","no","no","no","4.51","4","20","16","20","17","19","38","20"
+"146.8","71","female","partner","ever","Active","guideline","no","no","no","0","20","yes","no","no","no","no",NA,"3","14","6","4","8","12","7","88"
+"108.2","87","male","alone","never","Active","guideline","no","yes","no","0","2","no","no","no","no","no","67.63","2","5","7","11","4","4","4","88"
+"180","61","male","partner","never","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","281.28","2","10",NA,"12","13","6","6","76"
+"130.8","76","male","partner","ever","Placebo","guideline","yes","no","no","0","0","no","no","no","no","no","402.45","2","4","11","4","4","4","1","92"
+"135.63","70","female","alone","never","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","94.6","1","5","8","8","7","4","6","80"
+"201.8","69","male","partner","never","Placebo","guideline","no","yes","yes","0","2","no","no","no","no","no","204.55","2","11","9","8","5","11","6","44"
+"52.2","67","male","alone","never","Placebo","more","no","no","no","1","5","no","no","no","no","no","74.25","1","11","5","6","6","11","9","72"
+"282","60","male","partner","never","Placebo","guideline","no","yes","no","0","2","no","no","no","yes","no","177.4","2","10","13","7","4","4","7","80"
+"89.61","82","male","alone","never","Placebo","guideline","no","yes","yes","0","4","no","no","no","no","no","158.52","0","4","10","6","5","4","0","96"
+"77.31","78","female","alone","ever","Active","guideline","no","yes","no","0","2","no","no","no","no","no","189.9","2","16","17","20","18","13",NA,"44"
+"204.67","71","female","partner","never","Active","guideline","no","yes","no","0","4","no","no","no","no","no","91","2","4","4","4","4","4","0","100"
+"200.9","59","male","partner","never","Active","guideline","no","no","no","1","3","yes","no","no","no","no","174.43","2","13","15","12","11","13","9","40"
+"233.73","56","male","alone","never","Placebo","guideline","no","no","no","0","1","yes","no","no","no","no","144.7","1",NA,NA,NA,NA,NA,NA,NA
+"325.6","46","male","partner","ever","Placebo","guideline","no","no","no","0","3","yes","no","no","no","no","177.87","2","16","14","13","14","17","19","44"
+"113","76","female",NA,"never","Placebo","guideline","no","no","no","1","4","no","yes","no","no","no","79.45","2","20","18","19","16","14","36","12"
+"91.91","64","male","partner","never","Active","more","no","yes","yes","0","5","yes","no","no","no","no",NA,"1",NA,NA,NA,NA,NA,NA,NA
+"195.8","58","female","partner","never","Active","guideline","no","yes","no","0","2","no","no","no","no","no","111","1","11","13","6","4","7","11","76"
+"196.95","49","female","partner","ever","Active","guideline","no","no","no","0","5","yes","no","no","no","no","193.8","0","13","11","6","4","10","8","76"
+"207.31","52","male","alone","ever","Placebo","guideline","no","no","no","0","1","no","no","no","no","no","208.5","2","12","13","10","8","8","2","68"
+"53.6","88","male","partner","never","Placebo","guideline","no","yes","no","2","3","no","yes","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"45","82","male","partner","never","Placebo","guideline","no","no","no","0","7","no","no","no","no","no",NA,"4","7","18","16","7","12","15","48"
+"25","82","female","alone","never","Placebo","guideline","no","no","no","1","6","no","no","no","no","no","25.8","3","16","19","19","13","10","9","72"
+"147.4","34","female","partner","ever","Active","guideline","no","no","no","0","10","yes","no","no","no","no","139.05","2","20","14","15","16","18","36","24"
+"85","78","female","partner","ever","Active","guideline","no","no","no","0","3","no","no","no","no","no","131.8","1","12","8","4","12","4","1","92"
+"27.2","74","female","alone","never","Active","more","no","yes","no","2","2","no","no","no","no","no","36.23","2","10","11","15","9","7","3","84"
+"15","64","female","partner","never","Active","guideline","no","no","no","0","0","no","no","no","no","no","30.75","0","13","13","15","11","12","9","76"
+"225.6","51","male","partner","ever","Active","guideline","no","no","no","0","4","yes","no","no","no","no","211","0","7","4","4","5","7","1","88"
+"75.8","85","male","alone","never","Placebo","guideline","no","yes","no","0","3","no","no","no","no","no","75.8","1","17","16","13","10","12","12","36"
+"196.2","59","male","partner","never","Active","guideline","no","no","no","0","7","yes","no","no","no","no","122","1","11","13","11","10","4","3","84"
+"173.73","70","male","partner","ever","Placebo","guideline","no","no","no","0","4","yes","no","no","no","no","131.8","0","4","4","4","4","4","0","100"
+"40.32","77","male","partner","never","Placebo","guideline","yes","yes","no","0","1","yes","no","no","no","no","224.49","2","13","13","14","12","12","2","92"
+"106","82","male","partner","ever","Placebo","guideline","yes","no","no","0","19","yes","no","yes","no","no",NA,"2",NA,NA,NA,NA,NA,NA,NA
+"63.59","73","female","partner","never","Active","guideline","no","no","no","0","0","no","no","no","no","no","153.35","1","11","4","6","4","4","4","72"
+NA,"42","male","partner","ever","Placebo","guideline","no","no","no","0","7","yes","no","no","no","no","202.67","2","5","7","6","9","9","5","84"
+"155.11","62","male","partner","never","Active","guideline","no","yes","no","0","26","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"314.76","58","male","partner","never","Placebo","guideline","no","yes","no","0","0","yes","no","no","no","yes",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"196.47","84","male","partner","ever","Placebo","guideline","no","yes","no","0","6","no","no","no","no","no","7.31","4","16","17","9","5","9","15","64"
+NA,"71","female","alone","never","Active","guideline","yes","yes","no",NA,NA,"no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"247","54","male","partner","never","Active","more","no","yes","no","1","3","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"81.87","70","female","partner","ever","Active","guideline","no","yes","yes","0","3","yes","no","no","no","no","59.89","1","12","11","9","8","4","5","88"
+"132.12","87","male","partner","ever","Active","more","no","no","no","0","6","no","no","no","no","no","36.72","3","14","20","7","10","4","5","84"
+"50","84","male","alone","never","Active","guideline","yes","yes","no","2",NA,"no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+NA,"76","female","alone","never","Active","guideline","no","no","yes",NA,NA,"no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"247","67","male","partner","never","Active","guideline","no","yes","no","0","3","no","no","no","no","no","278.9","1","14","19","20","16","12","16","16"
+"166.72","52","male","partner","ever","Active","guideline","no","no","no","0","13","yes","no","yes","yes","no","161.4","1","13","12","14","10","15","5","64"
+"81.9","80","female","alone","ever","Active","guideline","no","yes","no","0","7","no","no","no","no","no","164.61","2","17","13","12","10","10","17","48"
+"148","63","female","alone","never","Placebo","guideline","no","yes","no","0","7","no","no","no","no","no","133","0","4","4","4","4","4","0","100"
+"43.8","67","male","partner","never","Placebo","guideline","no","yes","no","1","9","yes","no","yes","no","no","54.11","2","12","19","18","13","11","14","32"
+"143.31","67","male",NA,NA,"Placebo",NA,NA,NA,"no","0","0","no","no","no","no",NA,"27.2","1","18","20","20","17","8","27","40"
+"14.72","52","male","partner","never","Active","more","no","yes","yes","0","3","yes","no","no","no","no","97.4","0","14","19","16","9","6","6","68"
+"155.15","69","female","alone","never","Placebo","guideline","yes","yes","no","0","5","no","no","no","no","no","203.78","1","12","10","15","5","16","13","64"
+"247","71","male","partner","never","Active","guideline","yes","no","no","0","11","yes","no","no","no","no","61.2","2","8","7","4","4","15","5","96"
+"74.1","79","female","alone","never","Placebo","guideline","no","no","no","0","18","yes","no","yes","no","no","131.3","2","9","12","10","9","7","4","72"
+"122.55","75","male","partner","never","Active","guideline","no","no","no","0","3","no","yes","no","yes","no","81.59","0","7","6","9","11","11","4","76"
+"65","74","female","partner","never","Placebo","guideline","no","yes","no","0","6","no","no","no","no","no","75.8","1","16","12","8","8","12","7","72"
+"75","76","female","partner","ever","Active","guideline","no","yes","yes","0","2","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"64.67","75","female","alone","never","Placebo","guideline","no","no","no","0","1","yes","no","no","no","no","125.61","0","11","8","12","7","9","5","76"
+"115.48","72","female","alone","ever","Active","guideline","yes","no","no","0","1","no","no","no","no","no","108.8","1","7","5","12","7","4","0","96"
+"106.27","65","female","alone","ever","Placebo","guideline","no","no","no","0","1","no","no","no","no","no","92.09","1","5","4","10","4","12","5","84"
+"232.4","48","male","alone","never","Placebo","more","no","yes","no","0","2","no","no","no","no","no","282.48","1","7","8","9","4","7","7","76"
+"107.5","63","male","partner","ever","Active","guideline","no","yes","no","0","1","yes","no","no","no","no","210.17","1","8","6","9","6","6","2","96"
+"109.61","74","male","partner","never","Placebo","guideline","yes","yes","no","1","2","yes","no","no","no","no","111","2","4","4","8","8","4","3","40"
+"116.8","73","male","partner","never","Placebo","guideline","yes","no","no","0","2","yes","no","no","no","no",NA,"0",NA,NA,NA,NA,NA,NA,NA
+"114.92","78","male","partner","ever","Active","guideline","no","no","no","0","2","no","no","no","no","no","99.77","5","13","20","12","7","10","9","52"
+"85.5","45","male","partner","never","Placebo","guideline","no","no","no","0","5","no","no","no","no","yes","203.12","1","8","7","10","7","4","3","80"
+"124.27","66","male","alone","never","Active","guideline","no","no","no","0","3","no","no","no","yes","no","50.8","3","13","13","19","11","10","1","84"
+"263.33","37","male","partner","ever","Active","guideline","no","no","no","0","1","no","no","no","no","no","148.05","0","8","4","13","6","4","9","60"
+"166.8","54","male","partner","never","Placebo","guideline","no","no","no","0","3","no","no","no","no","no","236.8","0","4","8","4","4","4","1","100"
+NA,"83","male","partner","ever","Active","guideline","yes","no","no",NA,"27","no","no","yes","yes","yes",NA,"6",NA,NA,NA,NA,NA,NA,NA
+"124.11","66","female","alone","never","Placebo","guideline","no","no","no","0","6","no","no","no","no","no","238.43","2","16","16","4","4","8","7","44"
+"150","55","male","partner","never","Active","guideline","yes","yes","yes","0","5","no","no","no","no","no",NA,"6",NA,NA,NA,NA,NA,NA,NA
+"39.43","92","female","alone","ever","Placebo","guideline","no","yes","no","2","9","no","no","no","no","no","2.2","4","18","19","12","10","7","25","20"
+"79.03","88","female","partner","never","Active","guideline","yes","yes","no","0","21","yes","no","no","yes","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"270.93","57","male","partner","ever","Placebo","guideline","no","no","no","0","3","no","no","no","no","no","276.72","0","12","10","13","8","6","8","72"
+"431.8","55","male","partner","never","Active","guideline","yes","yes","no","0","1","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"65","79","female","alone","never","Active","guideline","no","no","no","0","8","no","no","no","no","no",NA,"6",NA,NA,NA,NA,NA,NA,NA
+"173.85","78","female","partner","ever","Active","guideline","yes","no","no","0","0","yes","no","no","no","no","210.58","1","8","4","6","5","4","5","72"
+"196.8","48","male","partner","ever","Active","guideline","no","no","no","0","2","yes","no","no","no","no","144.16","1","11","11","11","11","12","14","100"
+"58.6","80","male","alone","never","Placebo","guideline","no","no","no","2","2","no","no","no","no","no","78.6","3","14","16","13","5","14","10","80"
+"40","49","male","partner","never","Placebo","guideline","no","yes","no","0","2","no","yes","no","no","no","122.98","0","14","11","12","11","11","30","60"
+"90.55","75","male","partner","ever","Placebo","guideline","no","yes","yes","0","5","yes","no","no","yes","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"55.2","74","male","alone","never","Placebo","guideline",NA,"yes","no","0","5","no","yes","no","yes","no",NA,"2","18","20","20","8","10","12","68"
+"102.31","85","female","alone","never","Placebo","guideline","yes","no","no","0","1","no","no","no","no","no","171.28","0","11","12","13","6","6","8","76"
+"60.69","82","male","alone","ever","Active","guideline","no","yes","no","0","5","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"27.2","79","male","partner","never","Placebo","guideline","yes","yes","yes","2","3","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"68.6","75","male","partner","never","Active","guideline","no","yes","yes","1","5","no","no","no","no","yes","2.2","2","16","20","20","20","4","12","48"
+"221.47","70","male","partner","never","Placebo","guideline","no","yes","no","0","2","yes","no","no","yes","no","187.3","2","5","8","10","7","13","7","76"
+"272.8","57","male","partner","never","Active","guideline","no","no","no","0","4","no","no","no","no","no","259.84","2","9","8","9","8","14","2","100"
+"35.25","85","female","alone","ever","Active","guideline","no","yes","no",NA,"23","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"68.09","83","female","alone",NA,"Placebo",NA,"no","yes","no","0","5","no","no","no","no","no","92","1",NA,"10","13","12","10","8","84"
+"137.52","77","female","alone","ever","Active","guideline","no","no","no","0","5","yes","no","no","no","no","241.65","1","10","13","10","7","5","4","64"
+NA,"76","female","alone","never","Active","guideline","no","yes","no","1","7","no","no","no","no","no","30","3","16","14","20","10","6","11","28"
+"227.91","54","male","partner","ever","Active","guideline","no","no","no","0","4","yes","no","no","no","no","206.06","2","15","12","12","9","8","5","72"
+"56.4","78","female","alone","ever","Active","guideline","yes","yes","no","0","11","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"16","85","male","partner","never","Active","guideline","no","yes","no",NA,"2","no","no","no","no","no","4.51","3","4","11","17","6","5","4","84"
+"135.1","82","male","partner","ever","Placebo","more","no","no","no","1","3","yes","no","no","no","no","144.77","1","12","11","6","5","4","4","72"
+NA,"80","male","partner","never","Placebo","more","yes","yes","no","0","12","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"42.63","91","female","alone","never","Placebo","guideline","no","yes","yes","0","0","no","yes","no","no","no","42.63","1","20","16","19","12","4","7","36"
+"52.2","78","male","alone","never","Placebo","guideline","no","yes","no","0","4","no","no","no","no","no","171.56","3","10","11","13","14","6","6","64"
+"0","78","male","partner","never","Placebo","guideline","no","yes","no","1","2","no","no","no","no","no","14.86","4","11","19","17","7","11","1","84"
+"155.19","77","male","partner","ever","Placebo","guideline","no","yes","no","1","2","yes","no","no","no","no","107.5","2","18","19","18","9","13","12","40"
+"38.2","71","male","partner","never","Placebo","guideline","no","yes","yes","0","1","no","no","no","no","no","95","1","20","16","16","16","12","24","64"
+"135.69","66","male","partner","never","Active","guideline","no","no","yes","0","2","no","no","no","no","no","139.86","2","9","13","8","10","9","12","68"
+"205.6","55","male","alone","never","Placebo","guideline","no","yes","yes","0","2","no","no","no","no","no","60","1","17","20","20","14","9","8","40"
+"81.67","71","female","alone","never","Placebo","guideline","no","yes","no","0","5","no","no","no","no","no","44.16","3","16","20","17","4","17","20","20"
+"85.05","84","male","partner","never","Placebo","more","no","no","no","1","2","yes","no","no","no","no","128.11","1","5","5","7","8","8","5","88"
+"71.6","80","male","alone",NA,"Placebo","guideline","yes","yes","yes","0","3","yes","no","yes","no","no","151","1","16","12","14","6","10","5","80"
+"101.74","71","female","alone","never","Active","guideline","no","yes","no","0","0","no","no","no","no","no","27.9","1","12","14","14","13","13","6","88"
+"290.56","69","male","partner","never","Active","more","no","no","no","0","1","no","no","no","yes","no","254.35","1","7","9","10","7","6","3","80"
+"251.2","62","male","partner","never","Active","guideline","no","no","no","0","2","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"274.11","57","male","partner",NA,"Placebo",NA,"yes","yes","yes","0","7","yes","yes","yes","no","no","358.76","0","9","6","12","7","6","3","64"
+"226","61","male","alone","ever","Placebo","guideline","no","yes","no","0","5","yes","no","no","no","no","275.76","0","4","8","9","8","4","0","92"
+"59.56","75","male","alone","never","Active","guideline","no","yes","no","0","2","no","no","no","no","no","135.72","2","6","6","5","4","8","6","92"
+"53.11","93","female","alone","never","Placebo","guideline","no","yes","no","1","2","no","no","no","no","no","39.03","2","18","18","16","9","9","9","56"
+"150.6","86","male","partner","never","Placebo","guideline","no","no","no","0","3","no","yes","no","no","no","111","1","10","12","12","11","11","5","64"
+"119.76","84","male","alone","ever","Placebo","guideline","no","yes","no","0","2","yes","no","no","no","no","74.76","2","12","14","10","4","5","11","72"
+"101.4","80","male","partner","never","Placebo","guideline","no","yes","yes","2","4","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"186.56","70","male","partner","never","Placebo","guideline","no","yes","no","0","2","yes","no","no","yes","no","249.46","2","14","15","12","10",NA,"20","72"
+"337.8","68","female","partner",NA,"Active","guideline","no","no","no","2","1","yes","no","no","no","no","181.72","2","5","4","5","5","4","0","92"
+"170.05","40","female","partner","never","Placebo","guideline","no","no","no","0","15","no","no","yes","no","no","27.2","4","14","10","12","7","6","17","76"
+"95.8","77","female","partner","ever","Placebo","guideline","no","no","no","1","7","yes","no","no","no","no","131.8","1","15","9","5","7","4","12","100"
+"120.65","70","female","partner","never","Placebo","guideline","no","no","no","0","6","yes","no","no","no","no",NA,"4","12","20","16","8","18","14","72"
+"108.2","84","male","partner","ever","Active","guideline","yes","yes","yes","0","9","no","no","no","no","no","112.08","4","8","16","18","10","11","5","40"
+"188.16","71","male","partner","never","Active","guideline","yes","yes","no","0","9","no","no","no","no","no","183.93","4","5","9","11","5","7","3","80"
+"362.13","44","male","partner","never","Placebo","guideline","no","no","no","0","2","yes","no","no","no","no","241.4","2","14","14","13","9","10","16","68"
+"228.61","80","male","partner","never","Active","guideline","no","no","no","0","3","no","no","no","no","no","313.27","2","14","9","10","9","10","8","76"
+"65","65","female","alone","never","Active","guideline","no","yes","no","0","4","no","no","no","no","no","153.05","3","17","16","12","7","9","4","76"
+"259.53","60","male","partner","never","Placebo","more","no","yes","no","0","4","no","no","no","no","no","169.75","3","4","6","5","5","8","4","92"
+"272.72","69","female","alone","never","Placebo","guideline","no","yes","no","0","1","no","no","no","no","no","27.2","2","15","13","7","4","6","9","92"
+"169.6","47","male","partner","never","Placebo","guideline","no","no","no","0","1","yes","no","no","no","no","139.57","2","9","11","10","7","12","11","92"
+"88.31","63","male","partner","never","Placebo","more","no","yes","no","0","3","yes","no","no","no","no","81.41","2",NA,NA,NA,NA,NA,NA,NA
+"236","57","male","alone","never","Placebo","more","no","no","no","0","1","no","no","no","no","yes","217","1","4","12","14","8","9","2","88"
+"78.44","78","female","partner","never","Active","guideline","yes","yes","no","0","12","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"249.6","33","male",NA,NA,"Active",NA,NA,NA,"no","0","1","no","no","no","no",NA,"205.02","1","4","6","6","5","7","1","84"
+"94.6","81","female","alone","ever","Active","guideline","no","yes","no","1","2","no","no","no","yes","no","35.91","3","18","19","20","16","10","23","48"
+"194.12","70","male","partner","never","Placebo","guideline","no","no","no","0","0","no","no","no","no","no","191.8","1","16","14","15","10","7","11","60"
+"205.6","68","female","alone","ever","Placebo","guideline","no","no","no","0","3","yes","no","no","no","no","258.55","1","10","5","5","4","4","5","92"
+"139.3","67","female","alone","never","Placebo","guideline","yes","no","no","0","12","yes","no","yes","no","no","245.73","2","13","17","13","4","6","6","68"
+"217.87","73","male",NA,NA,"Active",NA,NA,NA,"no","0","4","no","no","no","no",NA,"169.58","1","9","8","13","6","4","2","96"
+"219.76","36","male","partner","ever","Active","guideline","no","no","no","0","0","no","no","no","no","no","161.4","2","18","20","17","7","11","8","56"
+"106.98","78","female","partner","never","Active","guideline","no","yes","no","0","4","no","no","no","no","no","199.56","2","6","16","12","9","13","5","76"
+"50","71","female","partner","never","Active","guideline","no","yes","no","0","1","no","no","no","no","no","52.2","3","17","20","19","11","8","14","48"
+"52.2","38","male","partner","ever","Placebo","guideline","no","no","no","0","5","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"85.45","79","male","partner","never","Active","guideline","no","yes","no","0","6","no","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"59.56","78","female","partner","ever","Active","more","no","yes","no","0","7","yes","no","no","no","no","110.2","4","16","17","17","8","12","12","68"
+"168.6","77","female","partner","ever","Active","guideline","no","yes","no","0","16","no","no","no","no","no","0","4","16","18","19","7","16","18","52"
+"232","63","male","partner","ever","Placebo","guideline","no","yes","yes","0","3","no","no","no","no","no","162.08","2","14","19","19","8","17","15","68"
+"116.73","77","male","partner","never","Placebo","guideline","no","yes","yes","0","8","yes","no","yes","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"89.12","75","male","partner","never","Active","guideline","no","yes","no","0","6","no","no","no","no","no","159.71","1","13","10","16","10","6","13","64"
+"50","68","female","alone","never","Active","guideline","no","no","no","0","2","no","no","no","no","no","129.1","0","12","15","11","10","6","4","72"
+"199.72","56","male","partner","never","Active","guideline","no","no","no","0","2","yes","no","no","no","no","172.25","1","18","13","10","4","9","7","76"
+"176.89","83","female","alone","never","Active","guideline","no","yes","no","0","5","no","no","no","no","no","138.53","0","13","8","7","8","10","2","92"
+"143.71","71","male","partner","never","Placebo","more","no","no","no","0","5","no","no","no","no","no","68.25","1","20","17","20","13","8","19","16"
+"148.11","68","male","partner","never","Active","more","no","yes","no","2",NA,"no","no","no","no","no","58.6","2","20","20","20","20","16","45","4"
+"91.4","74","male","partner","ever","Placebo","guideline","no","yes","no","0","5","no","no","no","yes","no","170.79","1","13","11","8","5","6","3","76"
+"406.8","79","female","alone",NA,"Placebo",NA,"no","yes","no","0","21","no","yes","no","no","no","0","4","10","16","10","12","4","15","20"
+"58.6","69","male","partner","ever","Placebo","guideline","yes","yes","no","0","17","yes","no","no","no","no","149.77","1","14","13","15","4","4",NA,"56"
+"162.99","67","female","partner","never","Active","guideline","no","no","no","0","4","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"108.2","76","female","alone","never","Active","guideline","yes","no","no","2","2","no","no","no","no","no","27.2","2","11","12","13","8",NA,"12","52"
+"119.51","48","female","partner","never","Active","guideline","no","no","no","0","15","yes","no","yes","no","no","167.11","2","14","5","14","6","12","6","68"
+"475.61","67","male","partner","never","Active","guideline","no","yes","no","0","1","no","no","no","yes","no","250","0","5","6","5","6","4","3","72"
+"31.4","82","female","alone","never","Placebo","guideline","no","yes","no","0","5","no","no","no","no",NA,NA,"4",NA,NA,NA,NA,NA,NA,NA
+"161.86","70","female","partner","never","Placebo","guideline","no","no","no","1","7","yes","no","no","no","no","96.86","2",NA,"5","4","4",NA,"5","84"
+"25","94","female","alone","ever","Placebo","guideline","yes","yes","no",NA,"12","no","no","no","no","no",NA,"4","13","19","19","4","4","32","80"
+"54.51","93","female","alone","never","Placebo","guideline","yes","yes","no","0","7","no","no","no","no","no",NA,"5","15","15","7","8","4","18","28"
+"144.6","53","male","partner","never","Placebo","guideline","no","yes","no","0","3","no","no","no","no","no","60.4","3","4","6","8","4","8","0","100"
+"55","42","male","alone","never","Active","guideline","no","no","no","0","2","yes","no","no","no","yes","58.6","1","18","19","15","16","7","8","48"
+"294.3","56","male","partner","never","Placebo","guideline","no","yes","no","0","5","yes","no","no","no","no",NA,"1","6","6","6","4","5","5","88"
+"110.8","61","female","partner","never","Active","guideline","no","yes","yes","0","22","yes","no","yes","no","no",NA,"6",NA,NA,NA,NA,NA,NA,NA
+"256.8","59","female","alone","never","Placebo","guideline","no","no","no","0","0","no","no","no","no","no","95.8","1","14","10","12","4","10","10","64"
+"232.91","69","female","partner","ever","Active","guideline","no","yes","no","2","5","yes","no","no","no","no","240.34","2","16","15","17","9","16","10","44"
+"264.65","47","male","partner","ever","Active","guideline","no","no","no","0","3","yes","no","no","no","no","305.28","1","16","15","14","10","9","4","68"
+"93.44","44","male","alone","never","Active","more","no","no","no","0","3","no","no","no","no","no","161.94","3","12","11","12","11","13","7","64"
+"132.5","69","male","partner","ever","Placebo","guideline","no","no","no","0","2","yes","no","no","no","no","179.53","0","12","6","10","10","11","6","60"
+"78.08","77","female","partner","never","Placebo","guideline","no","yes","no","0","15","yes","no","yes","no","no",NA,"1",NA,NA,NA,NA,NA,NA,NA
+"108.6","75","male","alone","ever","Placebo","guideline","no","no","no","0","1","yes","no","no","no","no",NA,"2","14","15","15","10","9","6","36"
+"33.93","86","female","alone","ever","Placebo","guideline","yes","no","no","2","24","yes","no","no","no","no","0","4","10","20","11","8","11","2","68"
+"574.26","70","male","partner","never","Placebo","guideline","no","yes","no","0","2","no","no","no","no","no","225.16","1","8","7","6","6","6","3","80"
+"309.5","56","male","partner","never","Active","guideline","no","no","no","0","1","yes","no","no","no","no","415.76","1","13","8","11","7","11","5","76"
+"247","48","female","partner","never","Placebo","guideline","no","no","no","0","12","yes","no","no","no","no","138.2","2","15","13","17","9","5","3","68"
+"136","55","male","partner","never","Active","guideline","no","no","no","0","13","yes","no","yes","no","no","172.2","0","13","7","13","7","7","6","52"
+"302.8","45","male","alone","never","Active","guideline","no","yes","no","0","0","no","no","no","no","no","101.4","2","12","17","13","9","7","4","92"
+NA,"79","male","partner","ever","Active","guideline","yes","yes","no","0","19","yes","no","yes","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"173.2","81","male","partner","never","Placebo","guideline","no","no","no","0","5","no","no","no","no","no","111.4","1","10","13","12","8","8","7","84"
+"129.11","79","male","alone","ever","Placebo","guideline","no","yes","no","0","1","no","no","no","yes","no","395.56","0","10","6","9","12","6","7","96"
+"218.23","57","male","alone","ever","Placebo","guideline","no","no","no","0","12","no","no","no","no","no","216.36","2","8","16","4","4","20","1","44"
+"229.16","83","male","alone","ever","Active","guideline","no","no","no","0","11","no","no","no","no","no","256.23","4","12","9","9","5","4","1","100"
+"136","82","male","alone","never","Placebo","guideline","no","yes","yes","1","18","no","no","no","no","no",NA,"4","17","20","20","4","4","18","16"
+"61","79","male","partner","never","Placebo","guideline","no","yes","no","2","2","no","yes","no","yes","no","27.2","2",NA,NA,NA,NA,NA,NA,NA
+"221.8","77","male","alone","never","Active","guideline","yes","yes","no","0","22","no","no","no","no","no","41.55","4",NA,NA,NA,NA,NA,"4","52"
+"149.16","49","male","partner","never","Placebo","guideline","no","no","no","1","2","no","no","no","no","no","256.55","2","10","7","6","6","5","1","80"
+"373.58","47","male","partner","ever","Active","guideline","no","no","no","0","6","yes","no","no","no","no","176.61","2","19","14","16","13","18","24","28"
+"204.6","81","female","alone","never","Active","guideline","no","no","no","0","6","no","no","no","no","no","115.03","1","17","9","13","5","5","9","60"
+"173.2","24","female","partner","ever","Active","guideline","no","no","no","0","1","no","no","no","no","no","171.2","2","10","13","8","8","6","6","64"
+"121","70","male","partner","never","Active","more","no","no","no","0","2","no","no","no","yes","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"0","68","male","partner","ever","Placebo","guideline","yes","yes","yes","0","8","no","no","no","no","no",NA,"2","7","13","12","7","4","4","16"
+"121.4","80","male","partner","never","Active","guideline","no","no","no","0","5","no","no","no","no","no","4.51","2","5","6","7","7","4","0","88"
+"46","67","male","partner","never","Active","guideline","no","yes","no","1","18","yes","yes","no","no","no",NA,"4","18","20","20","9","4","18","68"
+"206","55","male","alone","ever","Active","guideline","no","no","no","0","1","no","no","no","no","no",NA,"1","9","10","8","4","4","6","88"
+"160.11","77","male","partner","ever","Placebo","guideline","no","no","no","0","4","no","no","yes","no","no","124.44","1","9","13","13","8","9","6","68"
+"60.61","71","female","partner","never","Active","guideline","no","yes","no","0","5","no","no","no","no","no","58.94","1","20","13","15","11","4","22","24"
+"128.49","63","female","partner","never","Placebo","guideline","no","yes","no","0","3","no","no","no","no","no","95.05","0",NA,NA,NA,NA,NA,NA,NA
+"5","62","male","partner","ever","Placebo","guideline","no","yes","yes","2","19","yes","no","no","yes","no","103.93","2","18","15","18","12","16","36","36"
+"98.17","61","male","partner","never","Active","more","no","yes","no","0","2","no","no","no","no","yes","157.12","3","5","10","5","8","9","9","80"
+"282.4","77","female","alone","never","Placebo","guideline","yes","yes","no","0","11","yes","no","no","no","no","180.6","3","11","9","9","7","5","9","72"
+"110.8","77","male","partner","never","Placebo","guideline","no","no","no","0","1","no","no","no","no","no","58.6","1","11","7","7","7","5","7","72"
+"166.8","90","male","alone","never","Placebo","more","yes","yes","no","0","10","no","no","no","no","no","58.6","3","13","10","13","10","15","6","80"
+"162.45","71","female","alone","ever","Placebo","guideline","no","yes","no","0","1","yes","no","no","no","no","160.89","1","5","6","5","5","4","1","92"
+"156","68","female","alone","never","Active","more","no","yes","no","0","12","yes","no","no","no","no","85.69","2","4","4","4","8","11","1","100"
+"390.27","62","male","partner","ever","Active","guideline","no","no","yes","0","2","no","no","no","no","no","240.19","2","8","15","8","8","4","12","100"
+"155.96","57","male","partner","never","Placebo","guideline","no","no","no","0","1","no","no","no","no","no","151.06","1","10","7","12","8","6","8","64"
+"58.6","72","male","alone","never","Placebo","more","no","no","no","0","7","no","no","no","no","no",NA,"2","15","18","15","6","4","5","56"
+"179.87","79","male","partner","never","Placebo","guideline","no","no","no","0","3","no","no","no","yes","no",NA,NA,NA,NA,NA,NA,NA,NA,NA
+"75.8","76","female","partner","never","Active","guideline","no","yes","no","0","3","no","no","no","no","no","59.7","2","19","14","18","8","4","8","64"
+"0","67","male","partner","never","Active","guideline","no","no","no","0","3","no","no","no","no","no","119.05","2","6","14","10","8","10","5","80"
+"131.8","66","male","partner","never","Placebo","guideline","no","no","no","0","1","no","no","no","no","no","101.4","1","10","10","8","7","5","4","76"
+"95.8","83","female","partner","never","Active","more","no","yes","no","1","2","no","no","no","no","no","56.4","2","10","12","20","9","5","15","24"
+"90","71","male","alone","never","Placebo","guideline","no","no","no","0","6","no","no","no","no","no","55","1","4","4","4","8","8","4","100"
+"67.63","68","male","alone","never","Placebo","guideline","no","yes","no","1","2","yes","no","no","no","no","137.89","2","8","11","9","5","8","3","84"
+"162.72","52","male","partner","never","Placebo","more","no","yes","no","0","3","yes","no","no","no","no","347.78","0","6","6","7","6","6","3","76"
+"227.33","80","male","partner","never","Active","guideline","no","yes","no","0","2","no","no","no","no","no","284.33","1","6","4","8","4","4",NA,"100"
+"168.48","84","male","partner","never","Placebo","guideline","yes","yes","yes","1","1","no","no","no","no","no","152.72","1","10","9","8","8","8","6","76"
+"229.2","50","male","partner","never","Placebo","guideline","no","no","no","0","3","no","no","no","no","no",NA,"2","11","10","14",NA,"15","16","76"
+"197.15","44","male","partner","ever","Placebo","guideline","no","no","no","0","7","yes","no","no","no","no","245.75","1","8","4","9","5","7","6","76"
+NA,"63","male","partner","never","Placebo","guideline","no","no","no","0","2","yes","no","no","no","no",NA,NA,NA,NA,NA,NA,NA,NA,NA

2 Longterm/dags.R

@@ -0,0 +1,145 @@
+## Everything
+
+dag <- 'dag {
+  bb="-4.79,-6.035,4.818,5.504"
+  SES [pos="-4.463,-2.925"]
+  ad_treat [pos="2.007,-1.567"]
+  afli [pos="0.028,-1.851"]
+  age [pos="-1.242,-0.765"]
+  alc [pos="-3.716,-0.111"]
+  ami [pos="-3.137,3.986"]
+  civil [pos="-4.099,0.703"]
+  diabetes [pos="-3.838,3.579"]
+  education [pos="-3.632,-3.678"]
+  gen_PA [pos="-2.820,-1.247"]
+  hypertension [pos="-4.034,2.468"]
+  mrs_0 [pos="-1.746,2.715"]
+  mrs_1 [pos="1.326,-0.469"]
+  nihss_0 [pos="0.271,1.296"]
+  pase_0 [exposure,pos="-0.542,0.173"]
+  revasc [pos="0.980,2.332"]
+  rtreat [pos="2.334,2.036"]
+  sex [pos="-4.538,-0.778"]
+  smoker [pos="-1.877,-2.160"]
+  stroke [adjusted,pos="0.766,0.321"]
+  svd [latent,pos="-2.400,1.493"]
+  tci [pos="-2.418,4.467"]
+  vasc_event [outcome,pos="3.754,0.358"]
+  SES -> ad_treat
+  SES -> age
+  SES -> alc
+  SES -> gen_PA
+  SES -> stroke
+  SES -> vasc_event
+  ad_treat -> vasc_event
+  afli -> pase_0
+  afli -> stroke
+  afli -> vasc_event
+  age -> afli
+  age -> civil
+  age -> mrs_1
+  age -> nihss_0
+  age -> vasc_event
+  alc -> hypertension
+  alc -> pase_0
+  alc -> stroke
+  alc -> svd
+  alc -> vasc_event
+  ami -> afli
+  ami -> mrs_0
+  ami -> stroke
+  ami -> vasc_event
+  civil -> pase_0
+  civil -> stroke
+  civil -> vasc_event
+  diabetes -> mrs_0
+  diabetes -> nihss_0
+  diabetes -> stroke
+  diabetes -> svd
+  diabetes -> vasc_event
+  education -> SES
+  education -> ad_treat
+  education -> age
+  education -> alc
+  education -> gen_PA
+  education -> stroke
+  education -> vasc_event
+  gen_PA -> ami
+  gen_PA -> diabetes
+  gen_PA -> hypertension
+  gen_PA -> pase_0
+  gen_PA -> smoker
+  gen_PA -> tci
+  hypertension -> afli
+  hypertension -> mrs_0
+  hypertension -> nihss_0
+  hypertension -> stroke
+  hypertension -> svd
+  hypertension -> vasc_event
+  mrs_0 -> pase_0
+  mrs_0 -> revasc
+  mrs_1 -> ad_treat
+  mrs_1 -> vasc_event
+  nihss_0 -> revasc
+  pase_0 -> ad_treat
+  pase_0 -> nihss_0
+  pase_0 -> stroke
+  revasc -> mrs_1
+  rtreat -> mrs_1
+  rtreat -> vasc_event
+  sex -> ad_treat
+  sex -> alc
+  sex -> education
+  sex -> mrs_1
+  sex -> pase_0
+  sex -> stroke
+  sex -> vasc_event
+  smoker -> age
+  smoker -> mrs_1
+  smoker -> vasc_event
+  stroke -> mrs_1
+  stroke -> nihss_0
+  stroke -> rtreat
+  stroke -> vasc_event
+  svd -> mrs_0
+  svd -> pase_0
+  svd -> vasc_event
+  tci -> mrs_0
+  tci -> stroke
+  tci -> vasc_event
+}'
+
+
+## Simplified
+
+dag <- 'dag {
+bb="-4.79,-6.035,4.818,5.504"
+"Higher SES" [pos="-1.671,-1.740"]
+"U: higher svd score" [latent,pos="-3.091,2.172"]
+"active treat" [pos="1.914,3.209"]
+"higher PA" [exposure,pos="-0.999,0.555"]
+"lower mrs_0" [pos="-2.437,1.259"]
+ad_treat [pos="-0.448,-0.358"]
+cardio_vasc [pos="-4.202,0.839"]
+male [pos="-3.119,-0.926"]
+vasc_event [outcome,pos="3.754,0.358"]
+"Higher SES" -> "higher PA"
+"Higher SES" -> ad_treat
+"Higher SES" -> vasc_event
+"U: higher svd score" -> "lower mrs_0"
+"U: higher svd score" -> vasc_event
+"active treat" -> vasc_event
+"higher PA" -> ad_treat
+"higher PA" -> vasc_event
+"lower mrs_0" -> "higher PA"
+ad_treat -> vasc_event
+cardio_vasc -> "U: higher svd score"
+cardio_vasc -> "higher PA" [pos="-2.876,-0.185"]
+cardio_vasc -> "lower mrs_0"
+cardio_vasc -> vasc_event [pos="-4.762,5.060"]
+male -> "higher PA"
+male -> vasc_event [pos="-2.540,-5.134"]
+}'
+
+dag |> ggdag::ggdag_adjustment_set(node_size = 14, text_col = "black") + 
+  theme(legend.position = "bottom")

2 Longterm/data.R

@@ -0,0 +1,260 @@
+## 
+## Data pull and export for Forskermaskinen
+## 
+## Exports are made in REDCap instead
+## 
+## Data set is made here, not in REDCap, as data here is better organised.
+## 
+
+library(haven)
+library(dplyr)
+library(purrr)
+
+## All data
+# source("2 Longterm/data_import_files.R")
+source("/Users/au301842/PAaSO/2 Longterm/data_import_api.R")
+
+## Helpers
+source("/Users/au301842/PAaSO/2 Longterm/funs.R")
+
+# Streamlining missings for easier handling
+
+# NA values
+## Empty fields ("") are not included as NA, to ease later corrections.
+nas <- c("9.", "9. NA", "Not available", "Not relevant (filter/fold)")
+
+# Replaces all NAs in factors, and rejoins data frame
+# Factors keeps old levels; doesn't matter when exported as .csv and reimported...
+ls_nas <- lapply(seq_along(ls_sel), function(i) {
+  ds <- ls_sel[[i]] |> select(starts_with("talos_") 
+                              # & where(is.factor)
+                              ) |> 
+    data.frame()
+
+  ds_n <- lapply(seq_len(ncol(ds)),function(j){
+    as.character(if_else(ds[,j] %in% nas, NA, ds[,j]))
+  }) |> bind_cols()
+  
+  colnames(ds_n) <- colnames(ds)
+  
+  ds_sub <-  ls_sel[[i]] |> select(-colnames(ds_n)) 
+
+  ds_fin <- tibble(ds_sub,ds_n)
+  
+  ds_fin |> select(colnames(ls_sel[[i]])) |> select(-grep("[0-9]x$",colnames(ls_sel[[i]])))
+})
+
+names(ls_nas) <- names(ls_sel)
+
+
+
+## MFI domain scores
+
+# MFI variables to reverse
+# -> Create MFI function
+mfi_rev <- tolower(c("TALOS_MFI02","TALOS_MFI05","TALOS_MFI09","TALOS_MFI10","TALOS_MFI13","TALOS_MFI14","TALOS_MFI16","TALOS_MFI17","TALOS_MFI18","TALOS_MFI19"))
+
+match(mfi_rev,colnames(ls_nas$mfi|> select(matches(paste0("talos_mfi",stRoke::add_padding(1:20))))))
+
+domain_scores <- ls_nas$mfi |> select(matches(paste0("talos_mfi",stRoke::add_padding(1:20)))) |> mfi_domains(var = mfi_rev)
+
+colnames(domain_scores) <- paste0("talos_mfi_",colnames(domain_scores))
+
+ls_nas$mfi <- tibble(ls_nas$mfi,domain_scores)
+
+## SDMT score correction
+
+# Correction table
+# 
+# Holds manually determined 90 second times, additionally, "2015-02-18" is used 
+# as cut date for change from 60 seconds to 90 seconds
+# 
+# 
+
+if (FALSE){
+sdmt_time<-openxlsx::read.xlsx("/Volumes/Data/source/tid.sdmt.xlsx") |> 
+  tidyr::pivot_longer(cols = c(tid.1md, tid.6md)) |> mutate(deltager=as.character(deltager))
+sdmt_time$name <- as.double(as.character(factor(sdmt_time$name,labels = c("2","4"))))
+
+# Setting cut date
+sdmt_cut<-as.Date("2015-02-18")
+
+# Joining datasets to have date of visit
+sdmt_corr <- left_join(ls_nas$sdmt,sdmt_time |> mutate(name=as.character(name)), by = c("rnumb"="deltager","instance"="name"))
+
+# Modyfying old correction table to be complete
+sdmt_corr$value <- if_else(sdmt_corr$talos_sdmt00<sdmt_cut|sdmt_corr$value==60,60,90,missing = 90)
+
+# Write for database upload and easier future handling
+# write.csv(select(sdmt_corr, rnumb, instance, value),"2 Longterm/sdmt_time_correction.csv")
+
+# Multiplying meassure by correction valued turned weight and rounded
+sdmt_corr$talos_sdmt01a <- round(as.numeric(sdmt_corr$talos_sdmt01a)*(90/sdmt_corr$value),0)
+
+
+
+ls_nas$sdmt <- sdmt_corr
+}
+
+## PASE sum score
+
+pase_index <- stRoke::str_extract(colnames(stRoke::pase),"[0-9]{2}.*$")
+
+## Sourcing the newest pase_calc()
+source("/Users/au301842/stRoke/R/pase_calc.R")
+
+pase_scores <- ls_nas$pase |> select(ends_with(pase_index)) |> 
+  pase_calc(adjust_work = FALSE)
+
+ensure_prefix <- function(vec,prefix){
+  ifelse(!grepl(paste0("^",prefix),vec),paste0(prefix,vec),vec)
+}
+
+colnames(pase_scores) <- colnames(pase_scores)# |> ensure_prefix(prefix="pase_")
+
+last_cols <- function(ds, n){
+  ds[,(ncol(ds)-n+1):ncol(ds)]}
+
+pase_scores_work <- ls_nas$pase |> select(ends_with(pase_index)) |> 
+  pase_calc(adjust_work = TRUE) |> last_cols(4) %>%
+  rename_with(~ paste0(., "_w"))
+
+colnames(pase_scores_work) <- colnames(pase_scores_work) #|> ensure_prefix(prefix="pase_")
+
+ls_nas$pase <- tibble(ls_nas$pase,pase_scores,pase_scores_work)
+
+## Registration correction
+
+# Mutate all ends_with "00" as.Date in original data import file
+# source("2 Longterm/funs.R")
+
+ls_corr <- lapply(ls_nas,time_reg_correction, ref=subjects)
+
+run_eval=FALSE
+if (run_eval){
+library(compareDF)
+
+ls_comp <- lapply(seq_along(ls_nas),function(i){
+  compare_df(ls_corr[[i]],ls_nas[[i]],group_col = c("rnumb","instance"),stop_on_error = FALSE) 
+})
+
+ls_comp[[1]] |> 
+  create_output_table()
+}
+
+ls_wide <- ls_corr |>  longlist2wide()
+
+df_wide <- ls_wide |> purrr::reduce(full_join,by="rnumb")
+
+## The final assembly
+
+df_ddv <- subjects |> 
+  mutate(rnumb=as.character(rnumb)) |> 
+  left_join(df_wide %>% 
+              ## Keep only first "SITE" occurance
+              ## Using magrittr pipe for easy placeholder use
+              select(-grep("_site_",colnames(.))[-1])
+            )
+
+# write.csv(df_ddv,"/Volumes/Data/REDCap/DDV/talos_ddv.csv",row.names = FALSE)
+
+
+## Generate data description
+attr_files <- list.files("/Users/au301842/PAaSO/REDCap/attr",pattern = ".csv$",full.names = TRUE)
+attr_files_short <- list.files("/Users/au301842/PAaSO/REDCap/attr",pattern = ".csv$",full.names = FALSE)
+
+nms <- do.call(c,lapply(strsplit(attr_files_short,"_"),"[[",2)) |> gsub("['.']csv","",x=_) |> tolower()
+
+attr_lst <- lapply(attr_files,read.csv)
+names(attr_lst) <- nms
+
+attr_df <- lapply(seq_along(attr_lst),function(i){
+  
+  data.frame(name = tolower(ifelse(
+    !grepl("^TALOS|cpr|record_id", attr_lst[[i]][, 1]),
+    paste0(nms[i], "_", attr_lst[[i]][, 1]),
+    attr_lst[[i]][, 1]
+  )),
+  attr = attr_lst[[i]][, 2],
+  instr = paste(toupper(nms[i]),"instrument"))
+  
+}) |> bind_rows()
+
+attr_df$attr[grepl("sys_date$",attr_df$name)] <- "System date"
+attr_df$attr[grepl("sys_site$",attr_df$name)] <- "Trial site"
+attr_uni <- attr_df[!duplicated(attr_df$name),]
+
+# data.frame(variabel=colnames(df_ddv),
+#            visit=stRoke::str_extract(colnames(df_ddv),"[0124]$"),
+#            attr_df[match(gsub("_[0124]$","",colnames(df_ddv)),attr_df$name),-1]) |> 
+#   write.csv("REDCap/ddv_variabelbeskrivelse_raw.csv",row.names = FALSE)
+# 
+# read.csv("REDCap/ddv_variabelbeskrivelse.csv") |> 
+#   filter_at(1,all_vars(.%in%colnames(df_ddv))) |> 
+#   write.csv("REDCap/ddv_variabelbeskrivelse_mod.csv",row.names = FALSE)
+
+
+is.na(ds$talos_pase01_0) |> summary()
+
+
+
+## Cumulated data
+
+# dta<-read.csv("/Volumes/Data/exercise/source/background.csv",colClasses = "character", na.strings = c("NA","","unknown"))
+# 
+# export<-dta[,c("pase_0",
+#                "age",
+#                "sex",
+#                "civil",
+#                "smoker", 
+#                "rtreat",
+#                "alc",
+#                "afli",
+#                "hypertension",
+#                "diabetes",
+#                "mrs_0",
+#                "nihss_c",
+#                "thrombolysis",
+#                "pad", 
+#                "thrombechtomy", 
+#                "ami", 
+#                "tci",
+#                "rdate",
+#                "cpr",
+#                "rnumb",
+#                "height",
+#                "weight",
+#                "weight_est",
+#                "inc_time",
+#                "compliant",
+#                "mrs_1",
+#                "mrs_6",
+#                "pase_6",
+#                "visit_1",
+#                "visit_6")]
+
+# export$diabetes[is.na(export$diabetes)]<-"no"
+# export$hypertension[is.na(export$hypertension)]<-"no"
+# export$thrombolysis[is.na(export$thrombolysis)]<-"no"
+# export$thrombechtomy[is.na(export$thrombechtomy)]<-"no"
+# export$pad[is.na(export$pad)]<-"no"
+# export$ami[is.na(export$ami)]<-"no"
+# export$inc_time[export$inc_time<0]<-0
+# export$compliant<-as.numeric(factor(export$compliant))
+# export$rdate<-as.Date(export$rdate)
+# export$mrs_0[export$mrs_0==3]<-NA
+
+# export <- export|>
+#   mutate(any_rep=factor(ifelse(thrombolysis=="yes"|thrombechtomy=="yes","yes","no")), # If not noted, no therapy was received
+#          weight=ifelse(is.na(weight),weight_est,weight))|>
+#   select(-c(weight_est))
+# 
+# dput(names(export))
+
+# write.csv(export|>select(c(rnumb,rtreat)),"/Volumes/Data/SDS upload/study_treatment.csv",row.names = FALSE)
+# write.csv(export,"/Volumes/Data/SDS upload/data_all.csv",row.names = FALSE)
+# write.csv(export|>select(-c(rtreat)),"/Volumes/Data/SDS upload/background.csv",row.names = FALSE)
+
+# max(export$visit_6[!is.na(export$visit_6)])
+
+

2 Longterm/data_import_api.R

@@ -0,0 +1,110 @@
+
+token=keyring::key_get("TALOS_REDCAP_API")
+
+## Primary data set of recorded data
+
+inst <- REDCapR::redcap_instruments(redcap_uri = "https://redcap.au.dk/api/",token = token)$data
+
+tools <- tolower(c(
+  "PASE",
+  "MFI",
+  "mdi",
+  "mmse",
+  "mrs",
+  "SDMT",
+  "who",
+  "ham",
+  "bi",
+  "grad", 
+  "nihss"
+))
+
+vec_starts_with <- function(d,v){
+  filt <- paste0("^(", paste(v, collapse="|"), ")")
+  
+  d[grep(filt,d)]
+}
+
+vec_ends_with <- function(d,v){
+  filt <- paste0("(", paste(v, collapse="|"), ")$")
+  # sub(paste0(".*?_(",paste(v,collapse = "|"),")$"),"\\1",colnames(d))
+  d[grep(filt,d)]
+}
+
+# vec_starts_with(inst$instrument_name,tools)
+
+data <- REDCapR::redcap_read(redcap_uri = "https://redcap.au.dk/api/",token = token,
+                     forms = vec_starts_with(inst$instrument_name,tools),
+                     fields = "record_id")$data
+
+ds <- data |> select(!ends_with("complete"))
+
+talos2long <- function(ds,exclude=c("talos","record"),instance_vals=c("0","1","2","4"),exclude_col_ends=c("user","lock")){
+  nms <- unique(do.call(c,lapply(strsplit(colnames(ds),"_"),"[[",1)))
+  nms <- nms[!nms %in% exclude]
+  
+  ls <- lapply(seq_along(nms),function(i){
+    cols <- c("record_id",vec_starts_with(colnames(ds),c(paste0("talos_",nms[i]),nms[i])))
+    
+    d <- select(ds,all_of(cols))
+    
+    ends <- unique(do.call(c,lapply(strsplit(colnames(d),"_"),function(i){
+      i[length(i)]
+      })))
+    
+    ins <- colnames(d) %in% vec_ends_with(colnames(d),paste0("_",instance_vals))
+    
+    ls_ins <- split.default(d[-1],factor(stRoke::str_extract(colnames(d[ins]),paste0("[",paste(instance_vals,collapse=""),"]$"))))
+    
+    dat <- lapply(seq_along(ls_ins),function(i){
+      dr <- cbind(d[1],instance=names(ls_ins)[i],ls_ins[[i]])
+      colnames(dr) <- unique(gsub(pattern = paste0("_[",paste(instance_vals,collapse=""),"]$"),"",colnames(dr)))
+      dr |> select(record_id,contains("_sys_"),instance,everything())
+      }) |> bind_rows() |> rename(rnumb=record_id) |> mutate(rnumb=as.character(rnumb))
+    
+    ## Col name ending on 00 is date col of measure in all tools
+    dat[,vec_ends_with(colnames(dat),"00")] <- as.Date(dat[,vec_ends_with(colnames(dat),"00")])
+    
+    dat |> select(!ends_with(exclude_col_ends))
+  })
+  
+  names(ls) <- nms
+  ls
+  
+}
+
+ls_sel <- data |> select(!ends_with("complete"))|> talos2long() 
+
+## Data set of baseline and DAP data
+
+subjects_raw <-
+  REDCapR::redcap_read(
+    redcap_uri = "https://redcap.au.dk/api/",
+    token = token,
+    forms = vec_starts_with(inst$instrument_name, c("end", "inkl", "reg", "basis")),
+    fields = "record_id"
+  )$data 
+
+subjects <- subjects_raw |>
+  select(
+    record_id,
+    talos_end00,
+    talos_end01,
+    rdate,
+    rtreat,
+    cpr,
+    basis_kon,
+    starts_with("reg_")
+  ) |> select(!ends_with("complete")) |> 
+  rename(rnumb = record_id,
+         enddate = talos_end00,
+         eos_early = talos_end01,
+         sex=basis_kon) |>
+  mutate(
+    enddate = as.Date(enddate),
+    rdate = as.Date(rdate),
+    inc_time = as.numeric(difftime(enddate, rdate, units = "days")),
+    age = stRoke::age_calc(as.Date(stRoke::cpr_dob(cpr,"%Y-%m-%d")),enddate=rdate)
+  )
+
+# colnames(subjects)  <- gsub("^reg_","",colnames(subjects))

2 Longterm/data_import_files.R

@@ -0,0 +1,49 @@
+# Specifying needed data collection files
+data_source <- c(
+  "PASE_rev_v13.dta",
+  "MFI_rev_v13.dta",
+  "mdi_rev_v13.dta",
+  "mmse_rev_v13.dta",
+  "mrs_rev_v13.dta",
+  "SDMT_rev_v13.dta",
+  "who_rev_v13.dta"
+)
+
+## Cumulated data
+dta<-read.csv("/Volumes/Data/exercise/source/background.csv",colClasses = "character", na.strings = c("NA","","unknown"))
+
+# Getting full filenames
+file_nms <- list.files("/Volumes/Data/STATA13",full.names = TRUE)[match(data_source,list.files("/Volumes/Data/STATA13"))]
+
+# Loading datafiles
+
+dta<-read.csv("/Volumes/Data/exercise/source/background.csv",colClasses = "character", na.strings = c("NA","","unknown"))
+
+ls <- lapply(file_nms,function(i){
+  d <- read_dta(i)
+  colnames(d) <- tolower(gsub("instance","INSTANCE",colnames(d))) #in the sdmt dataset, instance column is lower case
+  d
+})
+
+# Selecting desired variables
+ls_sel <- lapply(seq_along(ls), function(i) {
+  ls[[i]] |> select(cpr,
+                    SYS_SITE,
+                    INSTANCE,
+                    starts_with("TALOS_")) |> as_factor() |> 
+    full_join(select(dta,cpr, rnumb)) |> select(rnumb,everything())
+})
+
+# Naming lists according to file names
+names(ls_sel) <- tolower(unlist(lapply(data_source,function(x){strsplit(x,"_")[[1]][1]})))
+
+## Screening list and EOS data
+subjects <- read_dta("/Volumes/Data/STATA13/inkl_rev_v13.dta") |>
+  select(c("cpr", "rnumb", "rdate", "rtreat")) |>
+  filter(rnumb != 999) |>
+  left_join(read_dta("/Volumes/Data/STATA13/end_rev_v13.dta") |>
+              select(c("cpr", "TALOS_end00", "TALOS_end01"))
+            ) |>
+  rename(enddate = TALOS_end00,
+         eos_early = TALOS_end01) |>
+  as_factor()

2 Longterm/flowchart.R

@@ -0,0 +1,33 @@
+data <- list(a=1000, b=800, c=600, d=400)
+
+
+DiagrammeR::grViz("
+digraph graph2 {
+
+graph [layout = dot]
+
+# node definitions with substituted label text
+node [shape = rectangle, width = 4, fillcolor = Biege]
+a [label = '@@1']
+b [label = '@@2']
+c [label = '@@3']
+node [shape = rectangle, width = 2, fillcolor = Biege]
+d [label = '@@4']
+e [label = '@@5']
+
+a -> b -> c [dir=s]
+a -> d [dir=e]
+b -> e [dir=e]
+# c -> f
+
+{rank=same; a -> b -> c [dir=s]}
+
+}
+
+[1]: paste0('All patients in TALOS (n = ', data$a, ')')
+[2]: paste0('Patients with PASE (n = ', data$b, ')')
+[3]: paste0('Patients with first event after follow-up (n = ', data$c, ')')
+[4]: paste0('Excluded due to missing PASE (n = ', data$d, ')')
+[5]: paste0('Excluded due to early event (n = ', data$d, ')')
+# [6]: paste0('Excluded due to missing PASE (n = ', data$d, ')')
+")

2 Longterm/funs.R

@@ -0,0 +1,248 @@
+
+## Utils
+
+multi_rev <- function(ds, var){
+  ndx <- match(var,colnames(ds))
+  for (i in seq_along(ndx)){
+    j <- ndx[i]
+    ds[[j]]<-as.character(factor(ds[[j]],labels = c(rev(levels(factor(ds[[j]]))))))
+  }
+  ds
+}
+
+reg2score <- function(ds){
+  # Removes all padding from registred scores
+  l <- c()
+  for (i in seq_len(ncol(ds))){
+    l[[i]] <- gsub("[^a-zA-Z0-9]","",ds[[i]])
+  }
+  
+  d <- do.call(cbind,l) |> data.frame()
+  
+  colnames(d) <- colnames(ds)
+  d |> tibble()
+}
+
+
+
+## MFI modding
+
+mfi_domains <- function(ds, reverse=TRUE, var){
+  
+  # Subscore indexes
+  indexes <- list(
+    data.frame(grp="gen", ndx=c(1, 5, 12, 16)),
+    data.frame(grp="phy", ndx=c(2, 8, 14, 20)),
+    data.frame(grp="act", ndx=c(3, 6, 10, 17)),
+    data.frame(grp="mot", ndx=c(4, 9, 15, 18)),
+    data.frame(grp="men", ndx=c(7, 11, 13, 19))
+  ) |> bind_rows() |> arrange(ndx)
+  
+  # Assumes reverse scores are not correctly reversed
+  if (reverse){ds <- ds |> multi_rev(var)}
+  
+  # Removes padding and converts to numeric
+  d <- ds |> reg2score() |> 
+    mutate_if(is.character, as.numeric)
+  
+  split.default(d, factor(indexes$grp)) |> 
+    lapply(function(x){
+      apply(x, MARGIN = 1, sum)
+    }) |> bind_cols()
+  
+}
+
+## Registration correction
+
+# 1.   If the inc_time is 38 days or less MDI 6 scores are moved to MDI 1 and visit 6 is defined as visit 1.
+# 2.   If both visit 1 and 6 dates are NA, use enddate as visit 1 date. This is the case if patients were excluded early.
+# 3.   If visit 6 is recorded later than enddate, use enddate instead.  MDI 6 score is dropped.
+# 4.   If visit delay is 7 days or less, and inclusion time is more than 38, MDI 1 is moved to MDI 6 and dropped. If MDI 1 and 6 are different both are kept. Enddate is moved to visit 6 date.
+# 5.   Defining the visit 6 date as same as enddate if visit delay is <7.
+
+# ds <- ls_nas$mdi
+# ref <- subjects
+
+
+time_reg_correction <- function(ds, ref) {
+  
+  # Splitting by rnumb, to treat each subj
+  ls <- split(ds, ds$rnumb)
+  
+  # ref: https://r-coder.com/progress-bar-r/
+  # Setting up simple progress bar
+  pb <- txtProgressBar(min = 0, max = length(ls), style = 3)
+  
+  
+  # Getting the rnumbs for subsetting in order of the list
+  nms <- names(ls)
+  # Applying correction to each element in list
+  ls_c <- lapply(seq_along(ls), function(i) {
+    cat(paste(i,"af",length(ls)))
+    # Updates the current state
+    setTxtProgressBar(pb, i)
+    
+    # Current rnumb
+    nm <- nms[i]
+    # print(i)
+    df <- ls[[i]]
+    # Only run if both instance 2 and 4 are present
+    if (all(c(2, 4) %in% select(df, matches("instance", ignore.case = TRUE))[[1]])) {
+      
+      # Define common variables
+      inc_time <- as.numeric(ref$inc_time[ref$rnumb == nm])
+      end_date <- as.Date(ref$enddate[ref$rnumb == nm])
+      rdate <- as.Date(ref$rdate[ref$rnumb == nm])
+      inst <- grep("instance", tolower(colnames(df)))
+      
+      # Define relevant data columns to substitute/modify
+      cols <- (inst + 1):ncol(df)
+      
+      # Define date column index
+      date_col_index <-
+        match(colnames(select(df, ends_with("00"))), colnames(df))
+      
+      # Step 1
+      
+      ## Correction if inc_time is performed
+      if (inc_time <= 38 & all(is.na(df[df[inst] == 2, cols]))) {
+        # Substitutions to transfer instance 4 meassures and leave as NA
+        df[df[inst] == 2, cols] <- df[df[inst] == 4, cols]
+        df[df[inst] == 4, -c(inst,date_col_index)] <- NA
+      }
+      
+      # Step 2
+      
+      if (all(is.na(df[date_col_index]))) {
+        # If both visit dates are missing, instance 2 date is substituted with enddate
+        df[date_col_index][df[inst] == 2] <-
+          as.character(end_date)
+      }
+      
+      # Step HELPA
+      ## Extra help step to impute enddate as visit 4 if instance 4 is present but no date.
+      if (is.na(df[date_col_index][df[inst] == 4])){
+        df[date_col_index][df[inst] == 4] <- as.character(end_date)
+      }
+      
+      ## Last steps are only performed if date values are available for both instance 2 and 4
+      if (!any(is.na(df[date_col_index]))){
+      
+      # Step 3
+      end_delay <- as.numeric(difftime(df[date_col_index][df[inst] == 4],
+                                       end_date, 
+                                       units = "days"))
+      
+      # If enddate is before last visit, the last visit data is dropped.
+      if (end_delay > 2) {
+        df[df[inst] == 4,-c(inst,date_col_index)] <- NA
+      }
+      
+      # Step 4
+      visit_delay <-
+        as.numeric(difftime(df[date_col_index][df[inst] == 4], df[date_col_index][df[inst] ==
+                                                                                    2], units = "days"))
+      
+      if (purrr::is_empty(visit_delay)){
+        visit_delay <- NA
+      }
+      
+        
+      ## Test if data has manually been inserted at visit 2, though should should be visit 4 time wise
+      second2fourth <-
+        if_else((visit_delay <= 7 |
+                   all(is.na(df[df[inst] == 2, cols]))) &
+                  inc_time > 38,
+                TRUE,
+                FALSE,
+                missing = FALSE)
+      
+      ## Move data from 2. visit to 4. 
+      if (second2fourth) {
+        # df[date_col_index][df[inst] == 2]
+        
+        ## If all missing at last visit, 1. visit is copied
+        if (all(is.na(df[df[inst] == 4, cols]))) {
+          df[df[inst] == 4, cols] <- df[df[inst] == 2, cols]
+        }
+        
+        ## If entries at visit 2 and 4 are identical, vist 2 is deleted
+        if (identical(df[df[inst] == 4, cols], df[df[inst] == 2, cols])) {
+          df[df[inst] == 2, -c(inst,date_col_index)] <- NA
+        }
+        
+        ## If data entries are NA at visit 2, the date is also deleted
+        if (all(is.na(df[df[inst] == 2, cols]))) {
+          df[,date_col_index][df[inst] == 2] <- NA
+        }
+        
+        ## The registered enddate is copied to the visit 4 date
+        df[,date_col_index][df[inst] == 4] <-
+          as.character(end_date)
+      }
+      
+      # Step 5
+      ## Ensure enddate is visit 4 if visit delay is <7
+      if (visit_delay < 7) {
+        df[,date_col_index][df[inst] == 4] <- as.character(end_date)
+      }
+      }
+    }
+    df
+  })
+  
+  ls_c |> bind_rows()
+  
+}
+
+longlist2wide <-
+  function(list,
+           id.name = "rnumb",
+           instance = "instance",
+           inst.glue = "{.value}_{instance}") {
+    
+    # ref: https://r-coder.com/progress-bar-r/
+    # Setting up simple progress bar
+    pb <- txtProgressBar(min = 0, max = length(list), style = 3)
+    
+    l <- lapply(seq_along(list), function(i) {
+      
+      # Updates the current state
+      setTxtProgressBar(pb, i)
+      
+      lst <- list[[i]] |> data.frame()
+      
+      rep_inst <-
+        length(levels(factor(lst[, colnames(lst) == instance]))) > 1
+      
+      
+      k <- lapply(split(lst, f = lst[[id.name]]), function(j) {
+        cname <- colnames(j)
+        vals <-
+          cname[!cname %in% c(id.name,
+                              instance)]
+        s <- tidyr::pivot_wider(
+          j,
+          names_from = instance,
+          values_from = all_of(vals),
+          names_glue = inst.glue
+        )
+        s[!colnames(s) %in% instance]
+      })
+      
+      k |> dplyr::bind_rows()
+      
+      
+    })
+    
+  }
+
+na_recode <- function(ds,vars_na,new_na="no"){
+  for (i in seq_along(vars_na)){
+    ds[i][is.na(ds[i])] <- new_na
+  }
+  ds
+}
+
+
+

2 Longterm/generalised odds ratio.R

@@ -0,0 +1,9 @@
+## Generaloised odds ratio
+## 
+## Tournament based approach
+## 
+
+
+library(genodds)
+
+

2 Longterm/grotta_bars.R

@@ -0,0 +1,26 @@
+library(rankinPlot)
+
+?rankinPlot::grottaBar
+
+dta<-read.csv("/Volumes/Data/exercise/source/background.csv",colClasses = "character", na.strings = c("NA","","unknown"))[,c("rtreat","mrs_0","mrs_1","mrs_6","hypertension","diabetes","civil")]
+
+df <- dta |> select(c("rtreat","mrs_0","mrs_1","mrs_6")) |> pivot_longer(cols = -rtreat)
+
+x<-table(mRS=df$value,
+      Group=df$rtreat,
+      Time = df$name)
+
+grottaBar(x,groupName="Group",
+          scoreName = "mRS",
+          strataName="Time",
+          colourScheme ="custom"
+) +
+  scale_fill_viridis_d(direction=-1)
+
+dta |> select(-c("mrs_0","mrs_1")) |> generic_stroke(group = "rtreat", score = "mrs_6", variables = c("hypertension","diabetes","civil"))
+
+library(stRoke)
+cc<-dta[complete.cases(dta),]
+talos <- cc[sample(1:nrow(cc),200),] |> select(-mrs_0)
+
+save(talos,file="talos.rda")

2 Longterm/hr coef plot.R

@@ -0,0 +1,173 @@
+source(here::here("1 PA Decline/dst import.R"))
+
+file <- project.aid::docx2list("/Users/au301842/Library/CloudStorage/OneDrive-Personal/Research/PhD/2 TALOS opfølgning/Manuskript/Arkiv/Longterm risk_v1_0.docx", data.type = "table cell")
+
+ds <- file |>
+  purrr::pluck(3) |>
+  (\(.x){
+    setNames(.x, letters[seq_len(ncol(.x))])
+  })() |>
+  dplyr::filter(dplyr::row_number() <= dplyr::n() - 1) |>
+  setNames(c("var", purrr::map(1:3, \(.x)paste0(c("hr", "ci"), .x)) |> purrr::list_c())) |>
+  dplyr::mutate(dplyr::across(dplyr::everything(), ~ gsub("—", "", .x))) |>
+  (\(.x){
+    split.default(.x, factor(project.aid::str_extract(names(.x), "\\d$"), labels = c("Univariable", "Multivariable", "Imputed"))) |>
+      purrr::imap(\(.y, .i){
+        dplyr::bind_cols(var = .x[1], .y) |>
+          tidyr::separate_wider_delim(
+            col = dplyr::starts_with("ci"),
+            delim = ", ",
+            # names_sep = "_",
+            too_few = "align_start",
+            names = c("low", "high")
+          ) |>
+          (\(.z){
+            names(.z)[2] <- "hr"
+            .z
+            # setNames(c("var","hr","low","high"))
+          })() |>
+          dplyr::mutate(model = .i)
+      })
+  })() |>
+  dplyr::bind_rows()
+
+create_log_tics <- function(data) {
+  sort(round(unique(c(1 / data, data)), 2))
+}
+
+forest_plot <- function(data,
+                        group.colors = viridisLite::viridis(3,option = "D"),
+                        x.tics = create_log_tics(c(1, 1.5, 3, 6)),
+                        point.shape = rep(23, 3),
+                        legend.title = "",
+                        dodge.width = .8,
+                        wrap.col) {
+  
+  data |>
+    ggplot2::ggplot(ggplot2::aes(x = log(hr), y = labels, color = model, fill = model)) +
+    ggplot2::geom_vline(ggplot2::aes(xintercept = 0), linewidth = .5, linetype = "dashed") +
+    ggplot2::geom_point(ggplot2::aes(shape = model),
+      # position = ggplot2::position_dodge(width = dodge.width),
+      size = 7
+    ) +
+    ggplot2::geom_errorbarh(ggplot2::aes(xmax = log(high), xmin = log(low)),
+      # position = ggplot2::position_dodge(width = dodge.width),
+      size = .5,
+      height = .2,
+      color = "gray50"
+    ) +
+    # ggplot2::position_dodge(width = 2, preserve = "total")+
+    ggplot2::scale_x_continuous(
+      breaks = log(x.tics),
+      labels = x.tics,
+      limits = log(range(x.tics))
+    ) +
+    ggplot2::scale_color_manual(values = group.colors) +
+    ggplot2::scale_fill_manual(values = group.colors) +
+    ggplot2::scale_shape_manual(values = point.shape) +
+    ggplot2::theme_bw() +
+    ggplot2::theme(
+      panel.grid.minor = ggplot2::element_blank(),
+      # legend.title = ggplot2::element_text(""),
+      legend.position = "bottom"
+    ) +
+    ggplot2::ylab("") +
+    ggplot2::xlab("Hazards ratio (log)") +
+    ggplot2::labs(
+      shape = legend.title,
+      color = legend.title,
+      fill = legend.title
+    ) +
+    ggplot2::facet_wrap(facets = ggplot2::vars(model), ncol = wrap.col)
+}
+
+# LETTERS[1:8] |> purrr::map(\(.x){
+#   viridisLite::viridis(3,option = .x)|> 
+#     project.aid::color_plot(ncol = 3)
+#   }) |> patchwork::wrap_plots(ncol=1)
+  
+headers <- c("Clinical data","Lifestyle factors","Socioeconomic factors","Assessments at follow-up")
+
+ds_new <- ds |>
+  dplyr::mutate(dplyr::across(c("hr", "low", "high"), ~ as.numeric(.x)),
+    var = factor(var, levels = rev(unique(var))),
+    model = factor(model, levels = unique(model))
+  ) |> 
+    dplyr::mutate(level=apply(is.na(dplyr::pick(hr,low,high))|dplyr::pick(hr,low,high) == "",1,sum),
+                  labels=dplyr::case_when(level==3&var%in%headers~marquee::marquee_glue("**{var}**"),
+                                          level==3&var!="Body mass index"~marquee::marquee_glue("*{var}*"),
+                                          .default=marquee::marquee_glue("{var}")),
+                  labels = factor(labels, levels = rev(unique(labels))))
+  
+(p1 <- ds_new |> (\(.x){
+    forest_plot(.x, wrap.col = length(levels(.x$model)))
+  })()+
+    ggplot2::theme(axis.text.y =marquee::element_marquee(vjust = .77,
+                                                         # hjust = -1,
+                                                         size=14)))
+
+
+ggplot2::ggsave(
+  filename = here::here("2 Longterm/hr_plot_facets.png"),
+  plot = p1+
+    ggplot2::theme(
+      strip.background = ggplot2::element_blank(),
+      strip.text.x = ggplot2::element_blank(),
+      legend.text = ggplot2::element_text(size=14),
+      axis.text.x=ggplot2::element_text(size=10),
+      axis.ticks.length.y = ggplot2::unit(3,"mm"),
+      axis.ticks.y = ggplot2::element_line(color = "white")
+    ),
+  units = "mm",
+  width = 250,
+  height = 300,
+  # pointsize = 10,
+  dpi = 300,
+)
+
+
+headers <- c("Clinical data","Lifestyle factors","Socioeconomic factors","Assessments at follow-up")
+
+(t <- ds  |> 
+  dplyr::mutate(dplyr::across(dplyr::everything(), ~ gsub("—", "", .x))) |>
+  dplyr::mutate(dplyr::across(c("hr", "low", "high"), ~ as.numeric(.x)),
+                var = factor(var, levels = rev(unique(var))),
+                model = factor(model, levels = unique(model))
+  ) |> 
+    dplyr::mutate(level=apply(is.na(dplyr::pick(hr,low,high))|dplyr::pick(hr,low,high) == "",1,sum),
+                  labels=dplyr::case_when(level==3&var%in%headers~glue::glue("**{var}**"),
+                                          level==3~glue::glue("*{var}*"),
+                                          .default=glue::glue("{var}"))) |> 
+  dplyr::filter(model=="Univariable")  |> 
+  ggplot2::ggplot(
+    # ggplot2::aes(x = x, y = var)
+    ) +
+  ggtext::geom_richtext(ggplot2::aes(x = 0, y = var,label = labels),fill = NA,
+                        label.colour = NA,
+                        hjust="right") +
+  ggplot2::scale_x_continuous(
+    limits = c(-.1,0)
+  )+
+  ggplot2::theme_void())
+
+(w <- patchwork::wrap_plots(list(t,
+                                 # patchwork::plot_spacer(),
+                           p1+
+                             ggplot2::theme(
+                               strip.background = ggplot2::element_blank(),
+                               strip.text.x = ggplot2::element_blank(),
+                               axis.text.y = ggplot2::element_blank(),
+                               plot.margin = ggplot2::unit(c(1,1,1,0), 'cm'),
+                               # axis.ticks.y.length = ggplot2::unit(0, "pt"),
+                               # axis.ticks.y = ggplot2::element_blank(),
+                               # panel.border=ggplot2::element_blank(), 
+                               panel.spacing = ggplot2::unit(0, "cm")
+                             )),ncol=2,widths = c(1.2,3),axes = "collect"))
+
+ggplot2::ggsave(
+  filename = here::here("2 Longterm/hr_plot_facets_labels.png"),plot = w,
+  units = "mm",
+  width = 200,
+  height = 220,
+  dpi = 300,
+)

2 Longterm/kamila-clustering.R

@@ -0,0 +1,84 @@
+## Not run: 
+# import and format a mixed-type data set
+library(kamila)
+data(Byar, package='clustMD')
+
+ds <- readr::read_csv("2 Longterm/assigndata.csv",na = c("","NA")) |> na.omit()
+
+ds <- ds |> dplyr::mutate(mrs_1 = mrs_1>2)
+
+cat_i <- lapply(ds,is.character) |> purrr::list_c()
+con_i <- lapply(ds,is.double) |> purrr::list_c()
+
+xor(cat_i,con_i)
+
+clin_clust=2
+# Byar$logSpap <- log(Byar$Serum.prostatic.acid.phosphatase)
+
+# conInd <- c(5,6,8:10,16)
+# conVars <- Byar[,conInd]
+# conVars <- data.frame(scale(conVars))
+
+conVars <- ds[,con_i]
+conVars <- data.frame(scale(conVars))
+
+# catVarsFac <- Byar[,-c(1:2,conInd,11,14,15)]
+# catVarsFac[] <- lapply(catVarsFac, factor)
+# catVarsDum <- dummyCodeFactorDf(catVarsFac)
+
+catVarsFac <- ds[,cat_i]
+catVarsFac <- lapply(catVarsFac, factor) |> dplyr::bind_cols() |> as.data.frame()
+catVarsDum <- dummyCodeFactorDf(catVarsFac)
+
+# Modha-Spangler clustering with kmeans default Hartigan-Wong algorithm
+gmsResHw <- gmsClust(conVars, catVarsDum, nclust = clin_clust)
+
+# Modha-Spangler clustering with kmeans Forgy-Lloyd algorithm
+# NOTE searchDensity should be >= 10 for optimal performance:
+# this is just a syntax demo
+gmsResLloyd <- gmsClust(conVars, catVarsDum, nclust = clin_clust,
+                        algorithm = "Lloyd", searchDensity = 15)
+
+# KAMILA clustering
+kamRes <- kamila(conVars, catVarsFac, numClust=2:7, numInit=10, calcNumClust="ps")
+
+# Plot results
+# ternarySurvival <- factor(Byar$SurvStat)
+# levels(ternarySurvival) <- c('Alive','DeadProst','DeadOther')[c(1,2,rep(3,8))]
+plottingData <- cbind(
+  conVars,
+  catVarsFac,
+  KamilaCluster = factor(kamRes$finalMemb))
+# plottingData$Bone.metastases <- ifelse(
+#   plottingData$Bone.metastases == '1', yes='Yes',no='No')
+# 
+# # Plot Modha-Spangler/Hartigan-Wong results
+# msPlot <- ggplot(
+#   plottingData,
+#   aes(
+#     x=logSpap,
+#     y=Index.of.tumour.stage.and.histolic.grade,
+#     color=ternarySurvival,
+#     shape=MSCluster))
+# plotOpts <- function(pl) (pl + geom_point() + 
+#                             scale_shape_manual(values=c(2,3,7)) + geom_jitter())
+# plotOpts(msPlot)
+
+# Plot KAMILA results
+kamPlot <- ggplot(
+  plottingData,
+  aes(
+    x=pase_0,
+    y=pase_6,
+    color=KamilaCluster,
+    shape=KamilaCluster))
+plotOpts(kamPlot)
+
+
+plotting_ls <- tibble(KamilaCluster = factor(kamRes$finalMemb),
+                      MSCluster = factor(gmsResHw$results$cluster)) |> 
+  purrr::map(\(x) cbind(x, ds))
+
+plotting_ls |> purrr::map(\(y) {
+  y |> gtsummary::tbl_summary(by=x) |> gtsummary::add_p() |> gtsummary::add_overall()}) |> 
+  gtsummary::tbl_merge()

2 Longterm/kmeans-clustering.R

@@ -0,0 +1,114 @@
+ds <- readr::read_csv("2 Longterm/assigndata.csv", na = c("", "NA")) |> na.omit()
+
+ds <- ds |> dplyr::mutate(mrs_1 = mrs_1 > 2)
+
+
+library(tidymodels)
+library(tidyverse)
+
+ds |>
+  na.omit() |>
+  kmeans(centers = 3)
+
+
+ds_num <- ds |> mutate(
+  across(where(is.double), ~ scale(.x)),
+  across(is.character, ~ as.numeric(factor(.x)))
+)
+
+ds_num |> kmeans(centers = 3)
+
+set.seed(321)
+kclusts <-
+  tibble(k = 1:9) %>%
+  mutate(
+    kclust = map(k, ~ kmeans(ds_num, .x)),
+    tidied = map(kclust, tidy),
+    glanced = map(kclust, glance),
+    augmented = map(kclust, augment, ds_num)
+  )
+
+kclusts
+
+clusters <-
+  kclusts %>%
+  unnest(cols = c(tidied))
+
+assignments <-
+  kclusts %>%
+  unnest(cols = c(augmented))
+
+clusterings <-
+  kclusts %>%
+  unnest(cols = c(glanced))
+
+ggplot(assignments, aes(x = pase_0, y = age)) +
+  geom_point(aes(color = .cluster), alpha = 0.8) +
+  facet_wrap(~k)
+
+ggplot(clusterings, aes(k, tot.withinss)) +
+  geom_line() +
+  geom_point()
+
+PCA
+
+pairs(assignments[-c(2:4)],
+  gap = 0,
+  bg = c("red", "yellow", "blue")[assignments$k],
+  pch = 21
+)
+
+
+pc.out <- prcomp(ds_num, center = TRUE, scale = TRUE)
+
+pc.sum <- summary(pc.out)
+
+pscr <- tibble(
+  x = 1:dim(pc.sum$importance)[2],
+  Proportion = pc.sum$importance[2, ],
+  Cumulative = pc.sum$importance[3, ]
+) %>%
+  pivot_longer(cols = -x) %>%
+  ggplot(aes(x = x, y = value, color = name)) +
+  geom_line() +
+  geom_point() +
+  ylim(0, 1) +
+  labs(
+    title = "Scree plot",
+    color = "Variance"
+  ) +
+  ylab("Variance") +
+  xlab("Principal components")
+
+
+###
+###
+
+install.packages("VarSelLCM")
+library(VarSelLCM)
+
+# Please indicate the number of cores you wan to use for parallelization
+nb.CPU <- 4
+# clustering without variable selection (about than 10/20 sec on 4 CPU)
+res_without <- ds |> select(!mrs_1) |> 
+  mutate(across(is.character, ~ factor(.x))) |> as.data.frame() |> 
+  VarSelCluster(
+    gvals = 1:5,
+    crit.varsel = "BIC",
+    vbleSelec = FALSE,
+    nbcores = nb.CPU
+  )
+
+summary(res_without)
+
+plot(res_without)
+
+plot(x=res_without, y="mdi_1")
+
+plot(x=res_without, y="sex")
+
+print(res_without)
+
+coef(res_without)
+
+VarSelShiny(res_without)

2 Longterm/sankey events.R

@@ -0,0 +1,191 @@
+
+# source("1 PA Decline/data_format.R")
+
+# NEW QUARTILES
+
+
+# Visuals - sankey
+# https://stackoverflow.com/questions/50395027/beautifying-sankey-alluvial-visualization-using-r
+
+
+## Painting
+
+df_raw <- readr::read_csv(here::here("2 Longterm/DDV 241031/event_sankey_data.csv")) 
+
+df <- df_raw|> 
+  dplyr::rename(pase_0_cut=pase_0_quartile,
+                pase_6_cut=pase_4_quartile) |> 
+  dplyr::mutate(change=dplyr::case_when(
+    pase_0_cut==1 & pase_6_cut==1 ~ "ll",
+    pase_0_cut %in% 2:4 & pase_6_cut %in% 2:4 ~ "hh",
+    pase_0_cut<pase_6_cut ~ "hop",
+    pase_0_cut>pase_6_cut ~ "drop"
+  ),
+  dplyr::across(c(pase_0_cut,pase_6_cut,change),as.factor))
+
+
+df |> sankey_ready()
+
+sankey_ready <- function(data,change.var="change"){
+  df <- data 
+  # |> 
+  #   dplyr::count(dplyr::across(dplyr::all_of(c("pase_0_cut", "pase_6_cut",change.var)))) |> 
+  #   dplyr::mutate(dplyr::across(dplyr::starts_with("pase_"),\(.x) factor(.x))) |> 
+  #   setNames(c("pase_0_cut", "pase_6_cut","change","n"))
+  
+  lbs0 <-
+    c(
+      paste0("1st \n(n=", sum(df$n[df$pase_0_cut == "1"]), ")"),
+      paste0("2nd \n(n=", sum(df$n[df$pase_0_cut == "2"]), ")"),
+      paste0("3rd \n(n=", sum(df$n[df$pase_0_cut == "3"]), ")"),
+      paste0("4th \n(n=", sum(df$n[df$pase_0_cut == "4"]), ")")
+    )
+  
+  
+  lbs6 <-
+    c(
+      paste0("1st \n(n=", sum(df$n[df$pase_6_cut == "1"]), ")"),
+      paste0("2nd \n(n=", sum(df$n[df$pase_6_cut == "2"]), ")"),
+      paste0("3rd \n(n=", sum(df$n[df$pase_6_cut == "3"]), ")"),
+      paste0("4th \n(n=", sum(df$n[df$pase_6_cut == "4"]), ")")
+    )
+  
+  
+  levels(df$pase_0_cut) <- lbs0[1:length(levels(df$pase_0_cut))]
+  levels(df$pase_6_cut) <- lbs6[1:length(levels(df$pase_6_cut))]
+  
+  df$pase_0_cut <- factor(df$pase_0_cut, levels = rev(levels(df$pase_0_cut)))
+  df$pase_6_cut <- factor(df$pase_6_cut, levels = rev(levels(df$pase_6_cut)))
+  
+  df$change <- factor(df$change, levels = c("hh","hop", "drop", "ll"))
+  
+  if (change.var=="change"){
+    df |> dplyr::mutate(first_grp=ifelse(substr(pase_0_cut,1,1)==1,"low","higher"))
+  } else if (change.var=="change_any"){
+    df |> dplyr::mutate(first_grp=dplyr::case_when(
+      substr(pase_0_cut,1,1)==1 ~ "low",
+      substr(pase_0_cut,1,1) %in% 2:3 ~ "mid",
+      substr(pase_0_cut,1,1)==4 ~ "high"))
+  }
+}
+
+# hops <- "#66c1a3" # grey
+# # drops <- "#990033"  # Midtrød
+# drops <- "#CE0045" # Lighter Midtrød
+# nos <- "grey80" # Light grey
+# 
+# # border <- "#00596B"
+# # box <- "#008099"
+# 
+# border <- "#EA571D"
+# box <- "#1E4B66"
+# 
+# higher <- "yellow"
+# low <- "purple"
+
+library(ggalluvial)
+
+library(ggplot2)
+
+# project.aid::color_plot(viridisLite::turbo(4))
+
+plot_sankey <- function(data,
+                        # palette=viridisLite::turbo(4),
+                        hops = "#1AE4B6FF",
+                        drops = "#FABA39FF",
+                        hh = "#30123BFF",
+                        ll = "#7A0403FF",
+                        border = "#EA571D",
+                        box = "#1E4B66",
+                        higher = "#1E4B66",
+                        mid = "#1E4B66",
+                        low = "#1E4B66",
+                        alpha = 0.8,
+                        a1=pase_0_cut,
+                        a2=pase_6_cut,
+                        a1.grp=first_grp,
+                        text.size = 4
+){
+  
+  if (length(unique(data[[ncol(data)]]))>2) {
+    fills <- c(higher,low,mid)
+  } else {
+    fills <- c(higher,low) 
+  }
+  
+  cls <- c(hh, hops, drops, ll)
+  # stratum.grp <- c(df[["first_grp"]],df[["last_grp"]])
+  
+  # cls <- palette
+  # browser()
+  ggplot(data, aes(y = n, axis1 = {{a1}}, axis2 = {{a2}})) +
+    geom_alluvium(
+      aes(fill = change, color = change),
+      width = 1 / 16,
+      alpha = alpha,
+      knot.pos = 0.4,
+      curve_type ="sigmoid"
+    ) +
+    geom_stratum(aes(fill={{a1.grp}}),
+                 # geom_stratum(aes(fill=stratum_grp),
+                 size = 2, 
+                 width = 1 / 3.4,
+                 # fill = box,
+                 color = border
+    ) +
+    geom_text(stat = "stratum", 
+              aes(label = after_stat(stratum)), 
+              colour = "white", 
+              size = text.size,
+              lineheight = 1) +
+    scale_x_continuous(
+      breaks = 1:2,
+      labels = c("Pre-stroke\nPASE quartile", "Six months\nPASE quartile")
+    ) +
+    scale_fill_manual(values = c(cls,fills),na.value = box) +
+    scale_color_manual(values = cls) +
+    ggtitle("PA level changes from \npre-stroke to post-stroke")
+}
+
+# c("change","change_any") |> purrr::map(\(.x){
+#   df_raw |>
+#     sankey_ready(change.var = .x)
+# }) |> 
+#   purrr::map(\(.x){
+#     .x |> plot_sankey(text.size=4.5)
+#   }) |> 
+#   patchwork::wrap_plots()
+
+pal <- viridisLite::turbo(12)
+
+project.aid::color_plot(pal)
+
+df |>
+  sankey_ready() |> 
+  plot_sankey(text.size=4.5)
+
+p_delta <- df |>
+  sankey_ready() |> 
+  plot_sankey(text.size=4.5)
+
+ggplot2::ggsave(filename = here::here("2 Longterm/sankey_event_stroke.png"),
+                p_delta +
+                  theme_void() +
+                  theme(
+                    legend.position = "none",
+                    # panel.grid.major = element_blank(),
+                    # panel.grid.minor = element_blank(),
+                    # axis.text.y = element_blank(),
+                    # axis.title.y = element_blank(),
+                    axis.text.x = element_text(),
+                    # text = element_text(size = 5),
+                    plot.title = element_blank(),
+                    # panel.background = element_rect(fill = "white"),
+                    plot.background = element_rect(fill="white"),
+                    panel.border = element_blank()
+                  ),
+                units = "mm",
+                width = 84,
+                height = 70,
+                # pointsize = 30,
+                dpi = 600)

2 Longterm/sdmt correction.R

@@ -0,0 +1,24 @@
+# During study time, the SDMT was performed in a non-standardised way until around 2015-02-18, only giving subjects 60 seconds.
+# Some filled assessments were marked and were all manually evaluated for correction.
+# A decision was reached to multiply old results by a factor of 1.5 assuming a proportional increase in completed fields.
+
+
+
+sdmt_time<-openxlsx::read.xlsx("/Volumes/Data/source/tid.sdmt.xlsx") |> 
+  tidyr::pivot_longer(cols = c(tid.1md, tid.6md)) |> mutate(deltager=as.character(deltager))
+sdmt_time$name <- as.double(as.character(factor(sdmt_time$name,labels = c("2","4"))))
+
+# Setting cut date
+sdmt_cut<-as.Date("2015-02-18")
+
+# Joining datasets to have date of visit
+sdmt_corr <- left_join(ls_nas$sdmt,sdmt_time |> mutate(name=as.character(name)), by = c("rnumb"="deltager","instance"="name"))
+
+# Modyfying old correction table to be complete
+sdmt_corr$value <- if_else(sdmt_corr$talos_sdmt00<sdmt_cut|sdmt_corr$value==60,60,90,missing = 90)
+
+# Write for database upload and easier future handling
+# write.csv(select(sdmt_corr, rnumb, instance, value),"2 Longterm/sdmt_time_correction.csv")
+
+# Multiplying meassure by correction valued turned weight and rounded
+sdmt_corr$talos_sdmt01a <- round(as.numeric(sdmt_corr$talos_sdmt01a)*(90/sdmt_corr$value),0)

2 Longterm/survival.R

@@ -0,0 +1,242 @@
+#survival
+
+# survfit experiments
+# 
+
+
+library(ggsurvfit)
+
+
+## This is what I wanted (!)
+(p <- survfit2(Surv(time, status) ~ surg, data = df_colon) |>
+  ggsurvfit(linewidth = 1,) +
+  add_confidence_interval() +
+  add_risktable() +
+  add_quantile(y_value = 0.6, color = "gray50", linewidth = 0.75) +
+  # limit plot to show 8 years and less
+  coord_cartesian(xlim = c(0, 8)) +
+  # update figure labels/titles
+  labs(
+    y = "Percentage Survival",
+    title = "Recurrence by Time From Surgery to Randomization",
+  ) +
+  # reduce padding on edges of figure and format axes
+  scale_y_continuous(label = scales::percent, 
+                     breaks = seq(0, 1, by = 0.2),
+                     expand = c(0.015, 0)) +
+  scale_x_continuous(breaks = 0:10, 
+                     expand = c(0.02, 0)))
+
+
+
+head(df_colon)
+
+library("survival")
+library("survminer")
+
+variables <- c("sex", "age", "adhere", "extent", "surg")
+
+cox_model <- coxph(as.formula(paste("Surv(time, status) ~",paste(variables,collapse="+"))), data = df_colon)
+
+summary(cox_model)
+
+ggsurvplot(survfit(cox_model), data=df_colon, palette = "#2E9FDF",
+           ggtheme = theme_minimal())
+
+# Checks
+
+(ph_check <- survival::cox.zph(cox_model))
+
+survminer::ggcoxzph(ph_check, var=c("sex", "age", "adhere", "surg"),
+                    font.main = 10, 
+                    font.x = 10, 
+                    font.y = 10)
+
+
+## Below is the nest best solution
+## Try at smoothing the survival curve
+
+## Working code to get satisfying object
+df <- survfit2(Surv(time, status) ~ surg, data = df_colon) |> 
+  tidy_survfit(type = "survival")
+
+df_split <- split(df,df$strata)
+
+df_smoothed <- purrr::reduce(lapply(c("estimate","conf.low", "conf.high"), function(j) {
+  do.call(rbind,
+          lapply(seq_along(df_split), function(i) {
+            nms <- names(df_split)[i]
+            y <-
+              predict(mgcv::gam(as.formula(paste0(
+                j[[1]], " ~ s(time, bs = 'cs')"
+              )), data = df_split[[i]]))
+            df <- data.frame(df_split[[i]]$time, y, nms)
+            names(df) <- c("time", paste0(j[[1]], ".smooth"), "strata")
+            df
+          }))
+}),dplyr::full_join) |> full_join(df)
+
+ggplot(data=df_smoothed) +
+  geom_line(aes(x=time, y=estimate.smooth, color = strata))+
+  geom_ribbon(aes(x=time, ymin = conf.low.smooth, ymax = conf.high.smooth, fill = strata), alpha = 0.50) +
+  # geom_smooth(aes(x=time, y=estimate, color = strata), method = "gam", formula = y ~ s(x, bs = "cs")) +
+  # reduce padding on edges of figure and format axes
+  scale_y_continuous(label = scales::percent, 
+                     breaks = seq(0, 1, by = 0.2),
+                     expand = c(0.015, 0), limits = c(0,1)) +
+  scale_x_continuous(breaks = 0:10, 
+                     expand = c(0.02, 0))+
+  labs(
+    y = "Percentage Survival",
+    title = "Recurrence by Time From Surgery to Randomization",
+  ) +
+  # limit plot to show 8 years and less
+  coord_cartesian(xlim = c(0, 8))
+
+
+
+## Dendrogram
+
+df_colon[do.call(c, lapply(seq_len(ncol(df_colon)), function(i) {
+  is.double(df_colon[[i]])
+}))][-1] |> scale() |> dist() |> hclust(method="average") |> ggdendro::ggdendrogram()
+
+
+## Better example??
+## 
+
+library(tidyverse)
+library(survival)
+library(purrr)
+library(ggsurvfit)
+library(cobs)
+
+## Data
+plot.type <- "survival"
+
+x <- survfit2(Surv(time, status) ~ surg, data = df_colon)
+df <-
+  tidy_survfit(x, type = plot.type) %>% dplyr::mutate(survfit = c(list(x),
+                                                                  rep_len(list(), dplyr::n() - 1L)))
+method <- "gam"
+
+df_split <- split(df,df$strata)
+
+df_smoothed <- purrr::reduce(lapply(c("estimate","conf.low", "conf.high"), function(j) {
+  do.call(rbind,
+          lapply(seq_along(df_split), function(i) {
+            nms <- names(df_split)[i]
+            x = df_split[[i]]$time
+            if (method=="loess"){
+              y <-
+                predict(loess(as.formula(paste0(
+                  j[[1]], " ~ time"
+                )), data = df_split[[i]]))
+            } else if (method=="gam"){
+              y <-
+                predict(mgcv::gam(as.formula(paste0(
+                  j[[1]], " ~ s(time, bs = 'cs')"
+                )), data = df_split[[i]]))
+            } else if (method=="cobs") {
+              
+              if (plot.type=="survival"){
+                ## This will make the plot start in (0,1)
+                con <- rbind(c( 0,min(x),1))
+                ## This ensures a monotonic decreasing slope 
+                ## for the estimate, not the CIs
+                if (j[[1]]=="estimate"){
+                  direction="decrease"
+                } else {direction="none"}
+                
+              } else if (plot.type=="risk"){
+                con <- rbind(c( 0,min(x),0))
+                if (j[[1]]=="estimate"){
+                  direction="increase"
+                } else {direction="none"}
+              }
+              
+              m <- cobs(x,df_split[[i]][[j]], 
+                        constraint=direction, 
+                        nknots = 4,
+                        pointwise= con,
+                        degree = 2,)
+              y <- predict(m, x)[, 'fit']
+            }
+            
+            df <- data.frame(x, y, nms)
+            names(df) <- c("time", paste0(j[[1]], ".smooth"), "strata")
+            df
+          }))
+}),dplyr::full_join) |> full_join(df)
+
+## Plotting
+ggplot(data=df_smoothed) + 
+  geom_line(aes(x=time, y=estimate.smooth, color = strata))+
+  geom_ribbon(aes(x=time, ymin = conf.low.smooth, ymax = conf.high.smooth, fill = strata), alpha = 0.50)
+
+## Weighted GAM approach
+## https://stackoverflow.com/a/66705556/21019325
+
+## It does not work. Gonna stop here due to lack of time.
+## Apparantly 
+
+dat_orig <- df_split[[1]][,c("time","estimate")]
+x1=0
+y1=1
+# set.seed(123)
+# N = 100
+# x <- sort(runif(N) * 4 - 1)
+# f <- exp(4*x)/(1+exp(4*x))
+# y <- f + rnorm(N) * 0.1
+# x = c(-1, x)
+# y = c(-0.1, y)
+# dat = data.frame(x = x, y= y)
+x <- do.call(c,c(x1,dat_orig[1]))
+y <- do.call(c,c(y1,dat_orig[,2]))
+dat <- data.frame(x=x,y=y)
+k <- 13
+
+library(mgcv)
+
+fit0 <- gam(y ~ s(x, k = k, bs = "cr"),data=dat)
+# predict from unconstrained GAM fit
+
+newdata <- data.frame(x = x)
+newdata$y_pred_fit0 <- predict(fit0, newdata = newdata)
+
+# Show regular spline fit (and save fitted object)
+# f.ug <- gam(y~s(x,k=k,bs="cr"))
+
+# explicitly construct smooth term's design matrix
+sm <- smoothCon(s(x,k=k,bs="cr"),dat,knots=NULL)[[1]]
+# find linear constraints sufficient for monotonicity of a cubic regression spline
+# it assumes "cr" is the basis and its knots are provided as input
+f.mono <- mono.con(sm$xp,up = FALSE)
+
+G <- list(
+  X=sm$X,
+  C=matrix(0,0,0),  # [0 x 0] matrix (no equality constraints)
+  sp=fit0$sp,       # smoothing parameter estimates (taken from unconstrained model)
+  p=sm$xp,          # array of feasible initial parameter estimates
+  y=dat[,2],           
+  w= c(1e8, rep(1,nrow(dat_orig))),  # weights for data
+  Ain=f.mono$A, # matrix for the inequality constraints
+  bin=f.mono$b, # vector for the inequality constraints
+  S=sm$S,  # list of penalty matrices; The first parameter it penalizes is given by off[i]+1
+  off=0   # Offset values locating the elements of M$S in the correct location within each penalty coefficient matrix.  (Zero offset implies starting in first location)
+)
+
+p <- pcls(G)       # fit spline (using smoothing parameter estimates from unconstrained fit)
+
+# predict 
+newdata$y_pred_fit2 <- Predict.matrix(sm, data.frame(x = newdata$x)) %*% p
+# plot
+ggplot(data=newdata) + 
+  geom_line(aes(x=x, y=y_pred_fit2))
+  # geom_ribbon(aes(x=time, ymin = conf.low.smooth, ymax = conf.high.smooth, fill = strata), alpha = 0.50)
+# 
+plot(y ~ x, data = dat)
+lines(y_pred_fit0 ~ x, data = newdata, col = 2, lwd = 2)
+lines(y_pred_fit2 ~ x, data = newdata, col = 4, lwd = 2)
+abline(v = -1)
+abline(h = -0.1)

2 Longterm/survival_two_groups.R

@@ -0,0 +1,132 @@
+# =============================================================================
+# Retrospective Sample Size Evaluation — Event-Free Survival (2 Groups)
+# Freedman Method | works directly from hazard ratio and observed events
+# Appropriate for: unequal group sizes, single censored-at-event designs
+# =============================================================================
+
+# --- 0. Install & Load Packages ----------------------------------------------
+
+# Uncomment to install if needed:
+# install.packages("gsDesign")
+
+library(gsDesign)
+
+
+# --- 1. Observed Data --------------------------------------------------------
+
+# n           : number of participants per group
+# events      : number of observed events per group
+# person_time : total person-years of follow-up per group
+#               = sum of each individual's follow-up duration
+#               naturally accounts for censoring and variable follow-up
+# max_follow_up : maximum follow-up time (years)
+
+n_A           <- 323;   events_A <- 80;   person_time_A <- 2089
+n_B           <- 56;    events_B <- 32;   person_time_B <- 294
+max_follow_up <- 9      # years
+
+# Derived quantities
+hazard_A <- events_A / person_time_A   # events per person-year
+hazard_B <- events_B / person_time_B
+hr       <- hazard_B / hazard_A        # observed hazard ratio (B vs A)
+
+overall_event_rate <- (events_A + events_B) / (n_A + n_B)
+pi_A     <- n_A / (n_A + n_B)
+pi_B     <- n_B / (n_A + n_B)
+
+cat("=== Observed Data Summary ===\n")
+cat(sprintf("Group A : N=%-4d  Events=%-3d  Person-years=%6.0f  Hazard=%.4f\n",
+            n_A, events_A, person_time_A, hazard_A))
+cat(sprintf("Group B : N=%-4d  Events=%-3d  Person-years=%6.0f  Hazard=%.4f\n",
+            n_B, events_B, person_time_B, hazard_B))
+cat(sprintf("Hazard ratio (B vs A)  : %.3f\n", hr))
+cat(sprintf("Allocation (A / B)     : %.1f%% / %.1f%%\n", pi_A*100, pi_B*100))
+cat(sprintf("Overall event rate     : %.3f\n", overall_event_rate))
+
+
+# --- 2. Required Events — Freedman Method ------------------------------------
+# Freedman (1982): works directly from the hazard ratio.
+# More stable than Schoenfeld when group sizes are unequal, because it does
+# not depend on a variance term V that collapses under unequal allocation.
+#
+# Formula: E = (z_alpha/2 + z_beta)^2 * (1 + hr)^2 / (hr - 1)^2
+#
+# Note: hr must not equal 1 (no difference). If hr < 1, invert it so
+# the formula always uses the ratio > 1.
+
+required_events_freedman <- function(alpha = 0.05, power = 0.80, hr) {
+  if (hr == 1) stop("Hazard ratio must not equal 1 (no detectable difference).")
+  hr      <- ifelse(hr < 1, 1/hr, hr)   # ensure hr > 1
+  z_alpha <- qnorm(1 - alpha / 2)
+  z_beta  <- qnorm(power)
+  E       <- (z_alpha + z_beta)^2 * (1 + hr)^2 / (hr - 1)^2
+  return(ceiling(E))
+}
+
+E_required <- required_events_freedman(alpha = 0.05, power = 0.80, hr = hr)
+N_required <- ceiling(E_required / overall_event_rate)
+
+cat("\n=== Freedman Method (alpha=0.05, power=0.80) ===\n")
+cat("Required total events :", E_required, "\n")
+cat("Observed total events :", events_A + events_B, "\n")
+cat("Required total N      :", N_required, "\n")
+cat("Observed total N      :", n_A + n_B, "\n")
+cat("Difference (req-obs)  :", N_required - (n_A + n_B), "\n")
+
+
+# --- 3. Validation via gsDesign::nSurv ---------------------------------------
+
+cat("\n=== gsDesign Validation ===\n")
+tryCatch({
+  gs <- nSurv(
+    lambdaC = hazard_A,
+    hr      = hr,
+    sided   = 2,
+    alpha   = 0.05,
+    beta    = 0.20,
+    T       = max_follow_up,
+    minfup  = 0
+  )
+  print(gs)
+}, error = function(e) {
+  cat("gsDesign error:", conditionMessage(e), "\n")
+})
+
+
+# --- 4. Sensitivity Analysis Across Power Levels -----------------------------
+
+power_levels <- c(0.70, 0.75, 0.80, 0.85, 0.90)
+
+sensitivity <- data.frame(
+  power      = power_levels,
+  events_req = sapply(power_levels, function(pw)
+    required_events_freedman(0.05, pw, hr))
+)
+sensitivity$n_req    <- ceiling(sensitivity$events_req / overall_event_rate)
+sensitivity$adequate <- ifelse(sensitivity$n_req <= (n_A + n_B), "Yes", "No")
+
+cat("\n=== Sensitivity Analysis by Power Level ===\n")
+print(sensitivity)
+
+
+# --- 5. Interpretation -------------------------------------------------------
+
+cat("\n=== Interpretation ===\n")
+cat(sprintf("Hazard ratio: %.3f — Group B events occur %.1fx faster than Group A\n",
+            hr, hr))
+
+if (E_required <= (events_A + events_B)) {
+  cat("Event count : ADEQUATE — observed events meet Freedman requirement.\n")
+} else {
+  cat(sprintf("Event count : INSUFFICIENT — %d observed, %d required.\n",
+              events_A + events_B, E_required))
+}
+
+if (N_required <= (n_A + n_B)) {
+  cat("Sample size : ADEQUATELY POWERED at 80%.\n")
+} else {
+  cat(sprintf(paste0("Sample size : UNDERPOWERED at 80%% — ",
+                     "N=%d observed, N=%d required (shortfall: %d).\n"),
+              n_A + n_B, N_required, N_required - (n_A + n_B)))
+  cat("Type II error risk is elevated — interpret null results with caution.\n")
+}

2 Longterm/varsellcm-clustering.R

@@ -0,0 +1,33 @@
+ds <- readr::read_csv("2 Longterm/assigndata.csv", na = c("", "NA"))
+
+library(VarSelLCM)
+
+# Please indicate the number of cores you wan to use for parallelization
+nb.CPU <- 4
+# clustering without variable selection (about than 10/20 sec on 4 CPU)
+clusters <- ds |> select(!mrs_1) |> na.omit() |> 
+  mutate(across(is.character, ~ factor(.x))) |> as.data.frame() |> 
+  VarSelCluster(
+    gvals = 1:5,
+    crit.varsel = "BIC",
+    vbleSelec = FALSE,
+    nbcores = nb.CPU
+  )
+
+summary(clusters)
+
+plot(clusters)
+
+plot(x=clusters, y="mdi_1")
+
+plot(x=clusters, y="sex")
+
+print(clusters)
+
+coef(clusters)
+
+VarSelShiny(clusters)
+
+clusters@partitions@zMAP
+
+gtsummary::tbl_summary(tibble(ds |> na.omit(), class=clusters@partitions@zMAP),by=class) |> gtsummary::add_p()

2 Longterm/wsc-sankey.R

@@ -0,0 +1,58 @@
+
+grps <- rbind(c("low0","low6","1"),
+      c("high0","low6","2"),
+      c("low0","high6","3"),
+      c("high0","high6","4"))
+
+tmes <- c(48,62,55,323)
+
+tbl <- data.frame(grps,tmes)
+
+colnames(tbl) <- c("t1","t2","grp","n")
+
+library(ggalluvial)
+
+cls <- viridisLite::turbo(4,direction=-1)
+
+cls
+stRoke::color_plot(viridisLite::turbo(4,direction=-1))
+
+p <- ggplot(tbl,aes(y = n, axis1 = t1, axis2 = t2)) +
+  geom_alluvium(
+    aes(fill = grp, color = grp),
+    width = 1 / 16,
+    alpha = .6,
+    knot.pos = 0.4
+  ) +
+  # geom_stratum(aes(size=10),width = 1 / 4,
+  #              fill = box,
+  #              color = border) +
+  # geom_text(stat = "stratum", aes(label = after_stat(stratum)), colour = "white", size = 20) +
+  scale_x_continuous(breaks = 1:2,
+                     labels = c("Pre-stroke\nquartile", "Six months\nquartile"))     +
+  scale_y_continuous(breaks=c(0,103,488))+
+  scale_fill_manual(values  = cls) +
+  scale_color_manual(values = cls) +
+  # labs(title=NULL, legend=NULL)+
+  theme_minimal()
+
+png(
+  filename = "sankey_change_WSC.png",
+  units = "mm",
+  width = 500,
+  height = 500,
+  pointsize = 15,
+  res = 300
+); p+
+  theme_minimal() +
+  theme(
+    legend.position = "none",
+    panel.grid.major = element_blank(),
+    panel.grid.minor = element_blank(),
+    axis.text.y = element_blank(),
+    axis.title.y = element_blank(),
+    axis.text.x = element_blank(),
+    plot.title = element_blank(),
+    panel.background = element_rect(fill='transparent'), 
+    plot.background = element_rect(fill='transparent', color=NA)
+  ); dev.off()