---
title: "The Effects of Recruitment status on completion of clinical trials"
author: "Will King"
format: html
editor: source
---


# Setup

```{r}
library(bayesplot)
available_mcmc(pattern = "_nuts_")
library(ggplot2)
library(patchwork)
library(tidyverse)
library(rstan)
library(tidyr)
library(ghibli)
library(xtable)
#Resources: https://github.com/stan-dev/rstan/wiki/RStan-Getting-Started

#save unchanged models instead of recompiling
rstan_options(auto_write = TRUE)
#allow for multithreaded sampling
options(mc.cores = parallel::detectCores())

#test installation, shouldn't get any errors
#example(stan_model, package = "rstan", run.dontrun = TRUE)
```

```{r}
################ Pull data from database ######################
library(RPostgreSQL)

driver <- dbDriver("PostgreSQL")

get_data <- function(driver) {

con <- dbConnect(
    driver,
    user='root',
    password='root',
    dbname='aact_db',
    host='will-office'
    )
on.exit(dbDisconnect(con))

query <- dbSendQuery(
    con,
#    "select * from formatted_data_with_planned_enrollment;"
"
select 
    fdqpe.nct_id
    --,fdqpe.start_date
    --,fdqpe.current_enrollment
    --,fdqpe.enrollment_category
    ,fdqpe.current_status 
    ,fdqpe.earliest_date_observed 
    ,fdqpe.elapsed_duration
    ,fdqpe.n_brands as identical_brands
    ,ntbtu.brand_name_count 
    ,fdqpe.category_id
    ,fdqpe.final_status
    ,fdqpe.h_sdi_val
    --,fdqpe.h_sdi_u95
    --,fdqpe.h_sdi_l95
    ,fdqpe.hm_sdi_val
    --,fdqpe.hm_sdi_u95
    --,fdqpe.hm_sdi_l95
    ,fdqpe.m_sdi_val
    --,fdqpe.m_sdi_u95
    --,fdqpe.m_sdi_l95
    ,fdqpe.lm_sdi_val
    --,fdqpe.lm_sdi_u95
    --,fdqpe.lm_sdi_l95
    ,fdqpe.l_sdi_val
    --,fdqpe.l_sdi_u95
    --,fdqpe.l_sdi_l95
from formatted_data_with_planned_enrollment fdqpe
    join \"Formularies\".nct_to_brands_through_uspdc ntbtu
        on fdqpe.nct_id = ntbtu.nct_id 
order by fdqpe.nct_id, fdqpe.earliest_date_observed 
;
"
    )
df <- fetch(query, n = -1)
df <- na.omit(df)

query2 <-dbSendQuery(con,"select count(*) from \"DiseaseBurden\".icd10_categories ic where \"level\"=1;")
n_categories <- fetch(query2, n = -1)

return(list(data=df,ncat=n_categories))
}


get_counterfact_base <- function(driver) {

con <- dbConnect(
    driver,
    user='root',
    password='root',
    dbname='aact_db',
    host='will-office'
    )
on.exit(dbDisconnect(con))

query <- dbSendQuery(
    con,
    "
    with cte as (
    --get last recruiting state
    select fd.nct_id, max(fd.earliest_date_observed),min(fd2.earliest_date_observed) as tmstmp
    from formatted_data fd 
    	join formatted_data fd2 
    	on fd.nct_id=fd2.nct_id and fd.earliest_date_observed < fd2.earliest_date_observed 
    where fd.current_status = 'Recruiting'
    	and fd2.current_status = 'Active, not recruiting'
    group by fd.nct_id 
    )
    select 
        fdqpe.nct_id
        --,fdqpe.start_date
        --,fdqpe.current_enrollment
        --,fdqpe.enrollment_category
        ,fdqpe.current_status 
        ,fdqpe.earliest_date_observed 
        ,fdqpe.elapsed_duration
        ,fdqpe.n_brands as identical_brands
        ,ntbtu.brand_name_count 
        ,fdqpe.category_id
        ,fdqpe.final_status
        ,fdqpe.h_sdi_val
        --,fdqpe.h_sdi_u95
        --,fdqpe.h_sdi_l95
        ,fdqpe.hm_sdi_val
        --,fdqpe.hm_sdi_u95
        --,fdqpe.hm_sdi_l95
        ,fdqpe.m_sdi_val
        --,fdqpe.m_sdi_u95
        --,fdqpe.m_sdi_l95
        ,fdqpe.lm_sdi_val
        --,fdqpe.lm_sdi_u95
        --,fdqpe.lm_sdi_l95
        ,fdqpe.l_sdi_val
        --,fdqpe.l_sdi_u95
        --,fdqpe.l_sdi_l95
    from formatted_data_with_planned_enrollment fdqpe
        join \"Formularies\".nct_to_brands_through_uspdc ntbtu
            on fdqpe.nct_id = ntbtu.nct_id 
    	join cte 
    		on fdqpe.nct_id = cte.nct_id 
    			and fdqpe.earliest_date_observed = cte.tmstmp
    order by fdqpe.nct_id, fdqpe.earliest_date_observed 
    ;
    "
    )
df <- fetch(query, n = -1)
df <- na.omit(df)

query2 <-dbSendQuery(con,"select count(*) from \"DiseaseBurden\".icd10_categories ic where \"level\"=1;")
n_categories <- fetch(query2, n = -1)

return(list(data=df,ncat=n_categories))
}


d <- get_data(driver)
df <- d$data
n_categories <- d$ncat

cf <- get_counterfact_base(driver)
df_counterfact_base <- cf$data



################ Format Data ###########################

data_formatter <- function(df) {
categories <- df["category_id"]

x <- df["elapsed_duration"]
x["identical_brands"] <- asinh(df$identical_brands)
x["brand_name_counts"] <- asinh(df$brand_name_count)
x["h_sdi_val"] <- asinh(df$h_sdi_val)
x["hm_sdi_val"] <- asinh(df$hm_sdi_val)
x["m_sdi_val"] <- asinh(df$m_sdi_val)
x["lm_sdi_val"] <- asinh(df$lm_sdi_val)
x["l_sdi_val"] <- asinh(df$l_sdi_val)


#Setup fixed effects
x["status_NYR"] <- ifelse(df["current_status"]=="Not yet recruiting",1,0)
x["status_EBI"] <- ifelse(df["current_status"]=="Enrolling by invitation",1,0)
x["status_Rec"] <- ifelse(df["current_status"]=="Recruiting",1,0) 
x["status_ANR"] <- ifelse(df["current_status"]=="Active, not recruiting",1,0)


y <- ifelse(df["final_status"]=="Terminated",1,0)

#get category list


return(list(x=x,y=y))
}

train <- data_formatter(df)
counterfact_base <- data_formatter(df_counterfact_base)

categories <- df$category_id

x <- train$x
y <- train$y

x_cf_base <- counterfact_base$x
y_cf_base <- counterfact_base$y
cf_categories <- df_counterfact_base$category_id
```



# Fit Model




```{r}
################################# FIT MODEL #########################################
inherited_cols <- c(
    "elapsed_duration"
    #,"identical_brands"
    #,"brand_name_counts"
    ,"h_sdi_val"
    ,"hm_sdi_val"
    ,"m_sdi_val"
    ,"lm_sdi_val"
    ,"l_sdi_val"
    ,"status_NYR"
    ,"status_EBI"
    ,"status_Rec"
    ,"status_ANR"
)
```



```{r}
beta_list <- list(
        groups = c(
        `1`="Infections & Parasites",
        `2`="Neoplasms",
        `3`="Blood & Immune system",
        `4`="Endocrine, Nutritional, and Metabolic",
        `5`="Mental & Behavioral",
        `6`="Nervous System",
        `7`="Eye and Adnexa",
        `8`="Ear and Mastoid",
        `9`="Circulatory",
        `10`="Respiratory",
        `11`="Digestive",
        `12`="Skin & Subcutaneaous tissue",
        `13`="Musculoskeletal",
        `14`="Genitourinary",
        `15`="Pregancy, Childbirth, & Puerperium",
        `16`="Perinatal Period",
        `17`="Congential",
        `18`="Symptoms, Signs etc.",
        `19`="Injury etc.",
        `20`="External Causes",
        `21`="Contact with Healthcare",
        `22`="Special Purposes"
    ),
    parameters = c(
        `1`="Elapsed Duration",
        # brands
        `2`="asinh(Generic Brands)",
        `3`="asinh(Competitors USPDC)",
        # population
        `4`="asinh(High SDI)",
        `5`="asinh(High-Medium SDI)",
        `6`="asinh(Medium SDI)",
        `7`="asinh(Low-Medium SDI)",
        `8`="asinh(Low SDI)",
        #Status
        `9`="status_NYR",
        `10`="status_EBI",
        `11`="status_Rec",
        `12`="status_ANR"
    )
)

get_parameters <- function(stem,class_list) {
    #get categories and lengths
    named <- names(class_list)
    lengths <- sapply(named, (function (x) length(class_list[[x]])))
    
    #describe the grid needed
    iter_list <- sapply(named, (function (x) 1:lengths[x]))
    
    #generate the list of parameters
    pardf <- generate_parameter_df(stem, iter_list)
    
    #add columns with appropriate human-readable names
    for (name in named) {
        pardf[paste(name,"_hr",sep="")] <- as.factor(
            sapply(pardf[name], (function (i) class_list[[name]][i]))
        )
    }
     
    return(pardf)   
}

generate_parameter_df <- function(stem, iter_list) {
    grid <- expand.grid(iter_list)
    grid["param_name"] <- grid %>% unite(x,colnames(grid),sep=",")
    grid["param_name"] <- paste(stem,"[",grid$param_name,"]",sep="")
    return(grid)
}

group_mcmc_areas <- function(
        stem,# = "beta"
        class_list,# = beta_list
        stanfit,# = fit
        group_id,# = 2
        rename=TRUE,
        filter=NULL
        ) {
    #get all parameter names
    params <- get_parameters(stem,class_list)
    
    #filter down to parameters of interest
    params <- filter(params,groups == group_id)
    #Get dataframe with only the rows of interest
    filtdata <- as.data.frame(stanfit)[params$param_name]
    #rename columns
    if (rename) dimnames(filtdata)[[2]] <- params$parameters_hr
    #get group name for title
    group_name <- class_list$groups[group_id]
    #create area plot with appropriate title
    p <- mcmc_areas(filtdata,prob = 0.8, prob_outer = 0.95) +
        ggtitle(paste("Parameter distributions for ICD-10 class:",group_name)) +
        geom_vline(xintercept=0,color="grey",alpha=0.75)
    
    d <- pivot_longer(filtdata, everything()) |> 
        group_by(name) |> 
        summarize(
            mean=mean(value)
            ,q025 = quantile(value,probs = 0.025)
            ,q975 = quantile(value,probs = 0.975)
            ,q05 = quantile(value,probs = 0.05)
            ,q95 = quantile(value,probs = 0.95)
            )
    return(list(plot=p,quantiles=d,name=group_name))
}

parameter_mcmc_areas <- function(
        stem,# = "beta"
        class_list,# = beta_list
        stanfit,# = fit
        parameter_id,# = 2
        rename=TRUE
        ) {
    #get all parameter names
    params <- get_parameters(stem,class_list)
    #filter down to parameters of interest
    params <- filter(params,parameters == parameter_id)
    #Get dataframe with only the rows of interest
    filtdata <- as.data.frame(stanfit)[params$param_name]
    #rename columns
    if (rename) dimnames(filtdata)[[2]] <- params$groups_hr
    #get group name for title
    parameter_name <- class_list$parameters[parameter_id]
    #create area plot with appropriate title
    p <- mcmc_areas(filtdata,prob = 0.8, prob_outer = 0.95) +
        ggtitle(parameter_name,"Parameter Distribution")
    
    d <- pivot_longer(filtdata, everything()) |> 
        group_by(name) |> 
        summarize(
            mean=mean(value)
            ,q025 = quantile(value,probs = 0.025)
            ,q975 = quantile(value,probs = 0.975)
            ,q05 = quantile(value,probs = 0.05)
            ,q95 = quantile(value,probs = 0.95)
            )
    return(list(plot=p,quantiles=d,name=parameter_name))
}


```

Plan: select all snapshots that are the first to have closed enrollment (Rec -> ANR)
```{r}
#delay intervention
intervention_enrollment <- x_cf_base[c(inherited_cols,"brand_name_counts", "identical_brands")]
intervention_enrollment["status_ANR"] <- 0
intervention_enrollment["status_Rec"] <- 1
```

```{r}
counterfact_delay <- list(
    D = ncol(x),#
    N = nrow(x),
    L = n_categories$count,
    y = as.vector(y),
    ll = as.vector(categories),
    x = as.matrix(x),
    mu_mean = 0,
    mu_stdev = 0.05,
    sigma_shape = 4,
    sigma_rate = 20,
    Nx = nrow(x_cf_base),
    llx = as.vector(cf_categories),
    counterfact_x_tilde = as.matrix(intervention_enrollment),
    counterfact_x = as.matrix(x_cf_base)
)
```

```{r}
fit <- stan(
    file='Hierarchal_Logistic.stan', 
    data = counterfact_delay,
    chains = 4,
    iter = 5000,
    seed = 11021585
    )
```








## Explore data


```{r}
#get number of trials and snapshots in each category
group_trials_by_category <- as.data.frame(aggregate(category_id ~ nct_id, df, max))
group_trials_by_category <- as.data.frame(group_trials_by_category)

category_count <- group_trials_by_category |> group_by(category_id) |> count()

```





## Fit Results


```{r}
################################# ANALYZE #####################################
print(fit)
```



# Counterfactuals

```{r}
generated_ib <- gqs(
    fit@stanmodel,
    data=counterfact_delay,
    draws=as.matrix(fit),
    seed=11021585
    )
```

```{r}
df_ib_p <- data.frame(
    p_prior=as.vector(extract(generated_ib, pars="p_prior")$p_prior)
    ,p_predicted = as.vector(extract(generated_ib, pars="p_predicted")$p_predicted)
)

df_ib_prior <- data.frame(
    mu_prior = as.vector(extract(generated_ib, pars="mu_prior")$mu_prior)
    ,sigma_prior = as.vector(extract(generated_ib, pars="sigma_prior")$sigma_prior)
)

#p_prior
ggplot(df_ib_p, aes(x=p_prior)) +
    geom_density() + 
    labs(
        title="Implied Prior Distribution P"
        ,subtitle=""
        ,x="Probability Domain 'p'"
        ,y="Probability Density"
    )
ggsave("./EffectsOfEnrollmentDelay/Images/DirectEffects/prior_p.png")

#p_posterior
ggplot(df_ib_p, aes(x=p_predicted)) +
    geom_density() + 
    labs(
        title="Implied Posterior Distribution P"
        ,subtitle=""
        ,x="Probability Domain 'p'"
        ,y="Probability Density"
    )
ggsave("./EffectsOfEnrollmentDelay/Images/DirectEffects/posterior_p.png")

#mu_prior
ggplot(df_ib_prior) +
    geom_density(aes(x=mu_prior)) + 
    labs(
        title="Prior - Mu"
        ,subtitle="same prior for all Mu values"
        ,x="Mu"
        ,y="Probability"
    )
ggsave("./EffectsOfEnrollmentDelay/Images/DirectEffects/prior_mu.png")

#sigma_posterior
ggplot(df_ib_prior) +
    geom_density(aes(x=sigma_prior)) + 
    labs(
        title="Prior - Sigma"
        ,subtitle="same prior for all Sigma values"
        ,x="Sigma"
        ,y="Probability"
    )
ggsave("./EffectsOfEnrollmentDelay/Images/DirectEffects/prior_sigma.png")
```



```{r}
check_hmc_diagnostics(fit)
```





### Intervention: Delay close of enrollment

```{r}
counterfact_predicted_ib <- data.frame(
    p_predicted_default = as.vector(extract(generated_ib, pars="p_predicted_default")$p_predicted_default)
    ,p_predicted_intervention = as.vector(extract(generated_ib, pars="p_predicted_intervention")$p_predicted_intervention)
    ,predicted_difference = as.vector(extract(generated_ib, pars="predicted_difference")$predicted_difference)
)


ggplot(counterfact_predicted_ib, aes(x=p_predicted_default)) +
    geom_density() + 
    labs(
        title="Predicted Distribution of 'p'"
        ,subtitle="Intervention: None"
        ,x="Probability Domain 'p'"
        ,y="Probability Density"
    )
ggsave("./EffectsOfEnrollmentDelay/Images/DirectEffects/default_p_generic_intervention_base.png")

ggplot(counterfact_predicted_ib, aes(x=p_predicted_intervention)) +
    geom_density() + 
    labs(
        title="Predicted Distribution of 'p'"
        ,subtitle="Intervention: Delay close of enrollment"
        ,x="Probability Domain 'p'"
        ,y="Probability Density"
    )
ggsave("./EffectsOfEnrollmentDelay/Images/DirectEffects/default_p_generic_intervention_interv.png")

ggplot(counterfact_predicted_ib, aes(x=predicted_difference)) +
    geom_density() + 
    labs(
        title="Predicted Distribution of differences 'p'"
        ,subtitle="Intervention: Delay close of enrollment"
        ,x="Difference in 'p' under treatment"
        ,y="Probability Density"
    )
ggsave("./EffectsOfEnrollmentDelay/Images/DirectEffects/default_p_generic_intervention_distdiff.png")
```


```{r}


pddf_ib <- data.frame(extract(generated_ib, pars="predicted_difference")$predicted_difference) |>
    pivot_longer(X1:X169)

#TODO: Fix Category names
pddf_ib["entry_idx"] <- as.numeric(gsub("\\D","",pddf_ib$name))
pddf_ib["category"] <-  sapply(pddf_ib$entry_idx, function(i) df$category_id[i])
pddf_ib["category_name"] <- sapply(pddf_ib$category, function(i) beta_list$groups[i])


ggplot(pddf_ib, aes(x=value,)) +
    geom_density(bins=100) +
    labs(
        title = "Distribution of predicted differences"
        ,subtitle = "Intervention: Delay close of enrollment"
        ,x = "Difference in probability due to intervention"
        ,y = "Probability Density"
    ) + 
    geom_vline(aes(xintercept = 0), color = "skyblue", linetype="dashed") 
ggsave("./EffectsOfEnrollmentDelay/Images/DirectEffects/p_generic_intervention_distdiff_styled.png")

ggplot(pddf_ib, aes(x=value,)) +
    geom_density(bins=100) +
    facet_wrap(
        ~factor(
            category_name, 
            levels=beta_list$groups
            )
        , labeller = label_wrap_gen(multi_line = TRUE)
        , ncol=4) +
    labs(
        title = "Distribution of predicted differences | By Group"
        ,subtitle = "Intervention: Delay close of enrollment"
        ,x = "Difference in probability due to intervention"
        ,y = "Probability Density"
    ) + 
    geom_vline(aes(xintercept = 0), color = "skyblue", linetype="dashed") +
    theme(strip.text.x = element_text(size = 8))
ggsave("./EffectsOfEnrollmentDelay/Images/DirectEffects/p_generic_intervention_distdiff_by_group.png")

ggplot(pddf_ib, aes(x=value,)) +
    geom_histogram(bins=100) +
    facet_wrap(
        ~factor(
            category_name, 
            levels=beta_list$groups
            )
        , labeller = label_wrap_gen(multi_line = TRUE)
        , ncol=5) +
    labs(
        title = "Histogram of predicted differences | By Group"
        ,subtitle = "Intervention: Delay close of enrollment"
        ,x = "Difference in probability due to intervention"
        ,y = "Predicted counts"
    ) + 
    #xlim(-0.25,0.1) +
    geom_vline(aes(xintercept = 0), color = "skyblue", linetype="dashed") +
    theme(strip.text.x = element_text(size = 8))
ggsave("./EffectsOfEnrollmentDelay/Images/DirectEffects/p_generic_intervention_histdiff_by_group.png")
```

Get the probability of increase over probability of a decrease

```{r}
mean(counterfact_predicted_ib$predicted_difference)
```
Thus adding a Delay close of enrollment increases the probability of termination by 16.72% on average for
the snapshots investigated.



```{r}
n = length(counterfact_predicted_ib$p_predicted_intervention)
mean(rbinom(n,1,as.vector(counterfact_predicted_ib$p_predicted_intervention)))
mean(rbinom(n,1,as.vector(counterfact_predicted_ib$p_predicted_default)))
```




# Diagnostics

```{r}
#| eval: false
#trace plots
plot(fit, pars=c("mu"), plotfun="trace")


for (i in 1:4) {
    print(
        mcmc_rank_overlay(
        fit, 
        pars=c(
            paste0("mu[",4*i-3,"]"),
            paste0("mu[",4*i-2,"]"),
            paste0("mu[",4*i-1,"]"),
            paste0("mu[",4*i,"]")
            ), 
        n_bins=100
        )+  legend_move("top") +
             scale_colour_ghibli_d("KikiMedium")
    )
}
```

```{r}
#| eval: false
plot(fit, pars=c("sigma"), plotfun="trace")

for (i in 1:4) {
    print(
        mcmc_rank_overlay(
        fit, 
        pars=c(
            paste0("sigma[",4*i-3,"]"),
            paste0("sigma[",4*i-2,"]"),
            paste0("sigma[",4*i-1,"]"),
            paste0("sigma[",4*i,"]")
            ), 
        n_bins=100
        )+  legend_move("top") +
             scale_colour_ghibli_d("KikiMedium")
    )
}
```

```{r}
#| eval: false
#other diagnostics
logpost <- log_posterior(fit)
nuts_prmts <- nuts_params(fit)
posterior <- as.array(fit)

```

```{r}
#| eval: false
color_scheme_set("darkgray")
div_style <- parcoord_style_np(div_color = "green", div_size = 0.05, div_alpha = 0.4)
mcmc_parcoord(posterior, regex_pars = "mu", np=nuts_prmts, np_style = div_style, alpha = 0.05)
```

```{r}
#| eval: false
for (i in 1:4) {
    mus = sapply(3:0, function(j) paste0("mu[",4*i-j ,"]"))
    print(
        mcmc_pairs(
            posterior,
            np = nuts_prmts,
            pars=c(
                mus,
                "lp__"
            ),
            off_diag_args = list(size = 0.75)
        )
    )
}



```

```{r}
#| eval: false
mcmc_parcoord(posterior,regex_pars = "sigma", np=nuts_prmts, alpha=0.05)
```

```{r}
#| eval: false

for (i in 1:4) {
    params = sapply(3:0, function(j) paste0("sigma[",4*i-j ,"]"))
    print(
        mcmc_pairs(
            posterior,
            np = nuts_prmts,
            pars=c(
                params,
                "lp__"
            ),
            off_diag_args = list(size = 0.75)
        )
    )
}
```


```{r}
#| eval: false
for (k in 1:22) {
for (i in 1:4) {
    params = sapply(3:0, function(j) paste0("beta[",k,",",4*i-j ,"]"))
    print(
        mcmc_pairs(
            posterior,
            np = nuts_prmts,
            pars=c(
                params,
                "lp__"
            ),
            off_diag_args = list(size = 0.75)
        )
    )
}}
```






# TODO
- [ ] Double check data flow. (Write summary of this in human readable form)
    - Is it the data we want from the database 
        - Training
        - Counterfactual Evaluation
            - choose a single snapshot per trial.
    - Is the model in STAN well specified.
- [ ] work on LOO validation of model