index.Rmd

---
title: "HIIT vs. MICT"
output:
  html_document:
    fig_caption: yes
    theme: cerulean
    toc: yes
    toc_depth: 3
    toc_float: yes
  pdf_document:
    toc: yes
    toc_depth: '3'
  word_document:
    toc: yes
    toc_depth: '3'
---
This is designed to track study progress! Below is a weekly update.
  
- **Things to keep track of:**

    - Are green/blue participants on track with workouts?
    - Have participants at the end...
        - completed EP surveys and enough workouts?
        - willing to crossover?
    - Are new participants willing to do biopsy?
   
```{r setup, include=FALSE}
# Load packages
knitr::opts_chunk$set(echo = TRUE)
library(tidyverse)
library(readxl)
library(ggplot2)
library(table1)
library(lubridate)
library(knitr)
library(kableExtra)
library(ggpubr)
library(stringr)
library(zoo)
```


```{r demog, message=FALSE, warning=FALSE, echo=FALSE}
# Read in the data generated by REDCapAPI.Rmd
demog<-read.csv("/Users/aubreykr/Google Drive/Shared drives/HIIT and Endurance Study/Data/data/Redcap/Processed_Data/64208_Demographics_Processed.csv")

#----Data Cleaning, features----------
# Pull in data to see if EP surveys are done
epsurveys<-read.csv("/Users/aubreykr/Google Drive/Shared drives/HIIT and Endurance Study/Data/data/Redcap/Raw_Data/64208_WorkoutDates_Raw.csv")
epsurveys<-epsurveys %>% 
  dplyr::select(record_id, redcap_event_name, perceived_stress_scale_pss10_complete)
epsurveys<-epsurveys %>% 
  filter(redcap_event_name == "week_12_ep_surveys_arm_1")
epsurveys<-epsurveys %>% 
  dplyr::rename(EP_surveys = perceived_stress_scale_pss10_complete) %>% 
  dplyr::select(record_id, EP_surveys)
epsurveys<-epsurveys %>% 
  mutate(EP_surveys = ifelse(EP_surveys == "Complete", "YES", ""))

# Create variable for crossover ppts
demog<-demog %>% 
  mutate(crossover = ifelse(record_id >=1000, "yes", "no"))

# Calculate expected num of workouts based on week they are on
demog$expected_num <- gsub("Week ", "", demog$current_week)
demog$expected_num <- (as.numeric(demog$expected_num) - 1)*3

# Flag people who need to be extended
demog$EXTEND <- ifelse(demog$expected_num - demog$total_workouts_in_REDCap >=4, "YES", " ")
demog$EXTEND[demog$Group == "Control"]<- " "

# EP vo2 complete?
demog$EP_VO2[is.na(demog$ep_vo2_max)==FALSE] <- "YES"
demog$EP_VO2[is.na(demog$ep_vo2_max)==TRUE] <- " "

# if missing, they have enrolled and update this
demog$enrollment_status[is.na(demog$enrollment_status) == TRUE] <- "Enrolled (not screened yet)"

# Specify order of weeks for plots
demog$current_week <- factor(demog$current_week, levels=c("Not Training Yet", "Week 1", "Week 2", "Week 3", "Week 4", "Week 5", "Week 6", "Week 7", "Week 8", "Week 9", "Week 10", "Week 11", "Week 12", "Week 13", "Week 14"))

# new column for target HR zone depending on group
demog<-demog %>% 
  mutate(HR_zone = case_when(Group == "MICT" ~ bl_heart_rate_mict,
                             Group == "HIIT" ~ bl_heart_rate_hiit))

# manually assign HR zone for crossover records
demog$HR_zone[demog$record_id == "1070"] <- "122-134"
demog$HR_zone[demog$record_id == "1076"] <- "114-126"
demog$HR_zone[demog$record_id == "1103"] <- "133-149"

demog<-full_join(demog, epsurveys, by=c("record_id"))
```

## Where are participants this week?

This was last updated on `r format(Sys.Date(), format="%B %d, %Y")`.

```{r firstcode, message=FALSE, warning=FALSE, echo=FALSE}

#---------Snapshot of participant progress------------
                                                     
ongoing<-demog %>% filter(enrollment_status == "On-going Randomized"  & is.na(ep_vo2_max)==TRUE | enrollment_status=="Enrolled (not screened yet)" | record_id == 127)
current_num<-length(unique(ongoing$record_id))

demog %>%
  filter(enrollment_status == "On-going Randomized"  & is.na(ep_vo2_max)==TRUE  | enrollment_status=="Enrolled (not screened yet)" | record_id ==127) %>% 
  ggplot(aes(x = current_week, fill=Group, group = factor(record_id))) + geom_bar() +
  geom_text(stat = "count", aes(label = paste0(record_id, sep="\n(", total_workouts_in_REDCap, sep=")"), y = ..count..),
            position = position_stack(vjust = 0.5), color = "black", size = 3, alpha=1) +
  scale_x_discrete(drop = FALSE) + theme_bw() +
  theme(axis.text.x = element_text(angle = 55, vjust = 1, hjust = 1)) +
  xlab(paste0("Plot Last Updated:", sep = " ", Sys.Date())) +
  ggtitle(paste0("Current Participants: n=", sep="", current_num)) + labs(fill = "Record ID") + theme(plot.title = element_text(hjust = 0.5)) +
  scale_y_continuous(breaks = seq(0, 30, 2)) 
```


## Workout Adherence

### On-going participants
```{r workouts0, message=FALSE, warning=FALSE, echo=FALSE}

counts<-demog %>%
   filter(enrollment_status == "On-going Randomized"  & current_week !="Not Training Yet" & is.na(ep_vo2_max)==TRUE)
active_num <- length(unique(counts$record_id))
# Filter for participants who are on-going
demog %>%
   filter(enrollment_status == "On-going Randomized"  & current_week !="Not Training Yet" & is.na(ep_vo2_max)==TRUE) %>%
  dplyr::select(record_id, current_week, Group, total_workouts_in_REDCap, expected_num, quality_fraction, name_of_participant_buddy, HR_zone, EXTEND, EP_surveys) %>% 
  arrange(current_week, record_id)  %>% 
  kbl(col.names = c("Record ID","Current Week","Group","Total Wkts", "Expected Num", "Quality (%)", "Buddy", "HR Target", "EXTEND", "EP_surveys"), align="c", escape=FALSE, caption=paste0("Active Participants: n=", sep="", active_num)) %>%
  kable_styling(full_width = F, position = "center", font_size = 14, latex_options = "striped", bootstrap_options = c("striped", "hover"))



# EP survey check
demog %>%
   filter(enrollment_status == "On-going Randomized" & (EP_VO2 != " " | current_week == "Week 12") & !record_id %in% c(76, 1076, 91, 92, 94, 98, 103, 105, 107, 110, 112, 1091, 117, 119, 123, 115))%>%
  dplyr::select(record_id, current_week, Group, total_workouts_in_REDCap, expected_num, quality_fraction, name_of_participant_buddy, HR_zone, EP_VO2, EP_surveys) %>% 
  arrange(current_week, record_id)  %>% 
  kbl(col.names = c("Record ID","Current Week","Group","Total Wkts", "Expected Num", "Quality (%)", "Buddy", "HR Target", "EP_Vo2?", "EP_surveys"), align="c", escape=FALSE, caption=paste0("EP survey check (randomized and EP VO2 done / at Week 12)")) %>%
  kable_styling(full_width = F, position = "center", font_size = 14, latex_options = "striped", bootstrap_options = c("striped", "hover"))



#Active controls
# demog %>%
#   filter(enrollment_status == "On-going Randomized" & Group == "Control" & current_week !="Not Training Yet"  & !record_id %in% c(103, 76, 98, 91) & record_id <= 1000) %>%
#   dplyr::select(record_id, current_week, Group, name_of_participant_buddy, control_crossover, EP_VO2) %>%
#   arrange(current_week, name_of_participant_buddy) %>%
#   kbl(col.names = c("Record ID","Current Week","Group","Buddy", "Crossover?", "EP VO2?"), align="c", escape=FALSE, caption="Active Controls") %>%
#   kable_styling()
demog$bl_VO2 <- ifelse(is.na(demog$bl_vo2_max), "No", "Yes")

# upcoming ppts
demog %>% 
  filter(enrollment_status %in% c( "On-going Randomized", "Enrolled (not screened yet)") & current_week =="Not Training Yet" ) %>% 
  dplyr::select(record_id, current_week, Group, name_of_participant_buddy, HR_zone, bl_VO2) %>% 
  kbl(col.names = c("Record ID","Current Week","Group","Buddy", "HR zone", "BL VO2 done?"), align="c", escape=FALSE, caption="Upcoming ppts") %>%
  kable_styling()

# not marked as completed, but did VO2
# demog %>%
#   filter(enrollment_status != "Completed" & is.na(ep_vo2_max)==FALSE) %>%
#   dplyr::select(record_id, Group, name_of_participant_buddy, EP_surveys) %>%
#   arrange(desc(record_id)) %>%
#   kbl(col.names = c("Record ID","Group","Buddy", "EP surveys?"), align="c", escape=FALSE, caption="Check on completed ppt EP surveys:") %>%
#   kable_styling()


#Mitra processing
# mitra<-demog %>%
#   select(record_id, enrollment_status, Group)
# 
# write.csv(mitra,"/Users/aubreykr/Desktop/mitradates.csv")
# 
# View(mitra %>%
#   filter(!record_id %in% c(3,4,5,6,7,8,10,12,14,18,19,20,21,26,29,30,36,37,38,40,45,46,50,51,56,60,64,71,72,73,74,75,76,77,80,81,87,88,89,90,95)))
# 
# mitras<-mitra %>%
#   filter(enrollment_status %in% c("Screen failed", "Withdrawn (before training)", "Withdrawn (after training)", "Lost to Follow-Up (before training)", "Lost to Follow-Up (after training)") | record_id >=74 & record_id <1000)
# #df<-read.csv("/Users/aubreykr/Desktop/Mitras/Mitra_ScreenFail_Withdraw.csv")

#mitras
#df<-df %>%
  #rename(record_id = Record.ID)

#mitras<-left_join(mitras, df)

#mitras %>%
  #filter(Date.of.Stress.Test.Visit != "" | Date.of.Endpoint.Stress.Test.Visit !="")
```

### Crossover Participant Tracking
```{r crossovers, message=FALSE, warning=FALSE, echo=FALSE}

demog$crossover_rand_group[demog$crossover_rand_group == "Endurance"]<- "MICT"

demog %>% 
  filter(crossover == "yes") %>% 
  dplyr::select(record_id,  Group, EP_VO2) %>% 
  kbl(col.names = c("Record ID","Rand Group", "EP VO2 done?"), align="c", escape=FALSE, caption="Crossover Participants") %>%
  kable_styling() 

```




```{r workouts,message=FALSE, warning=FALSE, echo=FALSE}
### Completed MICT participants

#Filter for participants who are completed and MICT
# completed_mict<-demog %>%
#   filter(enrollment_status == "Completed" & Group == "MICT") %>%
#   dplyr::select(record_id, quality_workouts_in_REDCap, percent_complete_of_36_wkts, number_polar_used, number_fitbit_used, total_workouts_in_REDCap, sex, agegroup, HR_zone)
# 
# completed_mict %>%
#     arrange(as.numeric(record_id)) %>%
#   kable(col.names = c("Record ID", "Quality Wkts", "Percent Complete of 36 Wkts", "Polar Worn", "Fitbit Worn", "Total Wkts", "Sex", "Age", "HR Target"), align="c") %>%
#    kable_styling(full_width = F, position = "center", font_size = 14, latex_options = "striped", bootstrap_options = c("striped", "hover"))
```


```{r workouts2,message=FALSE, warning=FALSE, echo=FALSE}
### Completed HIIT participants

# Filter for participants who are completed and HIIT
# completed_hiit<-demog %>%
#   filter(enrollment_status == "Completed" & Group == "HIIT") %>%
#   dplyr::select(record_id, quality_workouts_in_REDCap, percent_complete_of_36_wkts, number_polar_used,number_fitbit_used, total_workouts_in_REDCap, sex, agegroup, HR_zone)
# 
# completed_hiit %>%
#   arrange(as.numeric(record_id)) %>%
#    kable(col.names = c("Record ID", "Quality Wkts", "Percent Complete of 36 Wkts", "Polar Worn", "Fitbit Worn", "Total Wkts", "Sex", "Age", "HR Target"), align="c") %>%
#    kable_styling(full_width = F, position = "center", font_size = 14, latex_options = "striped", bootstrap_options = c("striped", "hover"))
```


```{r workouts3,message=FALSE, warning=FALSE, echo=FALSE}
### HIIT vs. MICT
#demog %>% 
#  filter(enrollment_status == "Completed" & Group %in% c("HIIT", "MICT")) %>% 
 # ggplot(aes(y=percent_complete_of_36_wkts, x=Group)) + geom_boxplot() + geom_point() + ylab("Percent Complete of 36 Workouts") +
 # ggtitle("Percent Complete of 36 Workouts by Randomization Group") + theme_bw()+ theme(plot.title = element_text(hjust = 0.5)) + ylim(c(0,100))

# 
# demog %>% 
#   filter(enrollment_status == "Completed" & Group %in% c("HIIT", "MICT")) %>% 
#   ggplot(aes(y=total_workouts_in_REDCap, x=Group)) + geom_boxplot() +geom_point() + geom_hline(yintercept = 32, linetype="dashed") + ylab("Total Workouts in REDCap") +
#   ggtitle("Completed Participants (n=30): Total Workouts") + theme_bw()+ theme(plot.title = element_text(hjust = 0.5)) + ylim(c(0,40))
# 
# demog %>% 
#   filter(enrollment_status == "Completed" & Group %in% c("HIIT", "MICT")) %>% 
#   ggplot(aes(y=quality_workouts_in_REDCap, x=Group)) + geom_boxplot() +geom_jitter(aes(color=factor(record_id))) + geom_hline(yintercept = 32, linetype="dashed") + ylab("Total Workouts in REDCap") +
#   ggtitle("Completed Participants (n=30): Good Workouts") + theme_bw()+ theme(plot.title = element_text(hjust = 0.5)) + ylim(c(0,40)) + labs(color="Participant Num")

```

### NUMBER COMPLETED
```{r progressstrata2, message=FALSE, warning=FALSE, echo=FALSE}
cohort<-demog %>% 
  filter(is.na(ep_vo2_max)==FALSE)

completed_num<-cohort %>% 
  filter(is.na(ep_vo2_max)==FALSE & record_id <= 1000)

completed_num<-length(completed_num$record_id)

cohort %>% 
  filter(is.na(ep_vo2_max)==FALSE & record_id < 1000) %>% 
  ggplot(aes(x=Group, fill=Group, alpha=0.8))+geom_bar(stat="count")+ theme_bw()+
  theme(plot.title = element_text(hjust = 0.5))+xlab("Group")+ylab("Number")+scale_fill_discrete(name = "Group")+
  geom_text(aes(label=..count..),stat='count',vjust=1.5) + 
  ggtitle(paste0(completed_num, sep=" ", "Completed with VO2 max pre/post")) + guides(alpha = "none")



# cohort %>% 
#   filter(is.na(ep_vo2_max)==FALSE & record_id > 1000) %>% 
#   ggplot(aes(x=Group, fill=Group, alpha=0.8))+geom_bar(stat="count")+
#   theme(plot.title = element_text(hjust = 0.5))+xlab("Group")+ylab("Number")+scale_fill_discrete(name = "Group")+
#   geom_text(aes(label=..count..),stat='count',vjust=1.5) + 
#   ggtitle(paste0("Num Crossovers Completed")) + guides(alpha = "none")

```

### Progress toward filling 4 strata
```{r progressstrata, message=FALSE, warning=FALSE, echo=FALSE}

# Create sex + agebin variable
demog$strata[demog$sex == "Female" & demog$agegroup == "18-34 yrs"]<-"F, 18-35"
demog$strata[demog$sex == "Female" & demog$agegroup == "35-65 yrs"]<-"F, 35-65"
demog$strata[demog$sex == "Male" & demog$agegroup == "18-34 yrs"]<-"M, 18-34"
demog$strata[demog$sex == "Male" & demog$agegroup == "35-65 yrs"]<-"M, 35-64"


cohort<-demog %>% 
  filter(enrollment_status == "On-going Randomized" & record_id <1000 | enrollment_status == "Completed" & record_id <1000)

cohort %>% 
  filter(is.na(ep_vo2_max)==FALSE & record_id < 1000) %>% 
  ggplot(aes(x=Group, fill=Group, alpha=0.8))+geom_bar(stat="count")+theme_bw()+
  theme(plot.title = element_text(hjust = 0.5))+xlab("Group")+ylab("Number")+scale_fill_discrete(name = "Group")+
  geom_text(aes(label=..count..),stat='count',vjust=1.5) + 
  ggtitle(paste0(completed_num, sep=" ", "Completed with CPX pre/post")) + guides(alpha = "none") + facet_wrap(~strata)


# Progress Toward Filling 4 Strata (M, F, 18-35 yrs, 35-55 yrs)

cohort %>% 
  ggplot(aes(x=Group, fill=Group, alpha=0.8))+geom_bar(stat="count")+theme_bw()+
  theme(plot.title = element_text(hjust = 0.5))+xlab("Group")+ylab("Number")+scale_fill_discrete(name = "Group")+facet_wrap(~strata)+
  geom_text(aes(label=..count..),stat='count',vjust=1.5) + 
  ggtitle(paste("On-going Randomized or Completed Study: n=", sep="", length(unique(cohort$record_id)))) + guides(alpha = "none")

cohort<-cohort %>% 
  filter(record_id != "81" & record_id !="69" & record_id !="1070" & record_id !="1076" & record_id !="1091")

cohort%>% 
  ggplot(aes(x=Group, fill=Group, alpha=0.8))+geom_bar(stat="count")+ theme_bw()+
  theme(plot.title = element_text(hjust = 0.5))+xlab("Group")+ylab("Number")+scale_fill_discrete(name = "Group")+
  geom_text(aes(label=..count..),stat='count',vjust=1.5) + 
  ggtitle(paste("Sum of on-going + completed: n=", sep="", length(unique(cohort$record_id)))) + guides(alpha = "none")

```


## Study Progress
```{r newppl, message=FALSE, warning=FALSE, echo=FALSE}
# Order levels of enrollment status
demog<-demog %>% 
  mutate(enrollment_status = ifelse(is.na(ep_vo2_max)==FALSE & record_id<1000, "Completed", ifelse(
    is.na(ep_vo2_max)==FALSE & record_id>1000, "Completed (crossover)", enrollment_status
  )))

demog$enrollment_status <- factor(demog$enrollment_status, levels = c("Enrolled (not screened yet)", "Lost to Follow-Up (before training)", "Screen failed", "Withdrawn (before training)",  "On-going Randomized", "Withdrawn (after training)", "Lost to Follow-Up (after training)", "Completed", "Completed (crossover)"))


ggplot(aes(x=enrollment_status, fill = enrollment_status, alpha=0.8), data=demog)+geom_bar(stat="count")+theme_bw()+
  ggtitle(paste("Total Participants: n=", sep="", length(unique(demog$record_id))))+ theme(plot.title = element_text(hjust = 0.5))+xlab("Participant Status")+ylab("Number")+scale_fill_discrete(name = "Group")+ theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust=1))+theme(legend.position = "none")+ geom_text(aes(label=..count..),stat='count',vjust=1.5)

```

### NUMBER CONSENTED
```{r consented, echo=FALSE, message=FALSE, warning=FALSE}
# PLOT CONSENTING
demog$date_consent<-as.yearmon(demog$date_consent)
demog$date_consent <- as.factor(demog$date_consent)
demog$date_consent<-factor(demog$date_consent, levels=c("Feb 2023", "Mar 2023", "Apr 2023", "May 2023", "Jun 2023", "Jul 2023", "Aug 2023", "Sep 2023", "Oct 2023", "Nov 2023", "Dec 2023", "Jan 2024", "Feb 2024", "Mar 2024", "Apr 2024", "May 2024", "Jun 2024", "Jul 2024", "Aug 2024", "Sep 2024", "Oct 2024", "Nov 2024", "Dec 2024", "Jan 2025", "Feb 2025"))

consent<-demog %>% 
  filter(is.na(date_consent)==FALSE)
consented_num<-length(consent$record_id)

demog %>% 
  filter(is.na(date_consent)==FALSE) %>% 
  ggplot(aes(x=date_consent)) + geom_bar(color="purple")+stat_count(geom = "text", color="white", aes(label = ..count..), vjust = 2)+theme_bw()+
  theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust=1)) + ggtitle(paste0("Number consented:", sep=" ", consented_num)) + theme(plot.title = element_text(hjust=0.5))+
  scale_y_continuous(breaks = seq(0,15,by=2)) + xlab("Month Consented")

```

## Month-by-Month Projections
Note: average counts exclude the first month and Dec 2023 for winter closure which had lower rates.
```{r formatdates, echo=FALSE, message=FALSE, warning=FALSE}

# Filter for relevant variables: start and end month, rand group, record_id
projected <- demog  %>% 
  filter(enrollment_status %in% c("Completed", "On-going Randomized")) %>% 
  dplyr::select(record_id, Group, start_month, end_month)

# Specify order of months
projected$start_month<-as.factor(projected$start_month)
projected$end_month<-as.factor(projected$end_month)

projected$start_month<-factor(projected$start_month, levels=c("Apr 2023", "May 2023", "Jun 2023", "Jul 2023", "Aug 2023", "Sep 2023", "Oct 2023", "Nov 2023", "Dec 2023", "Jan 2024", "Feb 2024", "Mar 2024", "Apr 2024", "May 2024", "Jun 2024", "Jul 2024", "Aug 2024", "Sep 2024", "Oct 2024", "Nov 2024", "Dec 2024", "Jan 2025", "Feb 2025"))

projected$end_month<-factor(projected$end_month, levels=c("Apr 2023", "May 2023", "Jun 2023", "Jul 2023", "Aug 2023", "Sep 2023", "Oct 2023", "Nov 2023", "Dec 2023", "Jan 2024", "Feb 2024", "Mar 2024", "Apr 2024", "May 2024", "Jun 2024", "Jul 2024", "Aug 2024", "Sep 2024", "Oct 2024", "Nov 2024", "Dec 2024", "Jan 2025", "Feb 2025"))

projected$start_month[is.na(projected$start_month)]<-"Feb 2024"
projected$end_month[is.na(projected$end_month)]<-"May 2024"
```



```{r plotprojected, echo=FALSE,  message=FALSE, warning=FALSE}

# Plots
current_date <- Sys.Date()
current_date <- as.yearmon(current_date)
current_date <- as.factor(current_date)

counts_start<-projected %>% 
  group_by(start_month) %>% 
  count()
counts_start<-counts_start %>% 
  filter(!start_month %in% c("Apr 2023", "Jan 2024"))
avg_start <- round(mean(counts_start$n),2)

counts_end<-projected %>% 
  group_by(end_month) %>% 
  count()
counts_end<-counts_end %>% 
  filter(!end_month %in% c("Aug 2023", "Jan 2024"))
avg_end <- round(mean(counts_end$n),2)


projected %>% 
  ggplot(aes(x=start_month)) + geom_bar(fill="#F8766D") + stat_count(geom = "text", color="white", aes(label = ..count..), vjust = 2) +theme_bw()+
  ggtitle(paste0("Start Months, average count/month:", sep=" ", avg_start)) + theme(plot.title = element_text(hjust = 0.5)) + geom_vline(xintercept = current_date, linetype="dotted", color="black")+ theme(axis.text.x = element_text(angle = 55, vjust = 1, hjust = 1)) 

projected %>% 
  ggplot(aes(x=end_month)) + geom_bar(fill="#00BFC4") + stat_count(geom = "text", color="white", aes(label = ..count..), vjust = 2) +theme_bw()+
  ggtitle(paste0("End Months, average count/month:", sep=" ", avg_end)) + theme(plot.title = element_text(hjust = 0.5))+ geom_vline(xintercept = current_date, linetype="dotted", color="black") + theme(axis.text.x = element_text(angle = 55, vjust = 1, hjust = 1)) 

# Combined start + end plot

num_projected <- length(unique(projected$record_id))

# Gather and color by timepoint
projected <- gather(projected, key = "Timepoint", value = "Month", c("start_month", "end_month"))
projected$Timepoint <- factor(projected$Timepoint, levels=c("start_month", "end_month"))
projected$Month<-factor(projected$Month, levels=c("Apr 2023", "May 2023", "Jun 2023", "Jul 2023", "Aug 2023", "Sep 2023", "Oct 2023", "Nov 2023", "Dec 2023", "Jan 2024", "Feb 2024", "Mar 2024", "Apr 2024", "May 2024", "Jun 2024", "Jul 2024", "Aug 2024", "Sep 2024", "Oct 2024", "Nov 2024", "Dec 2024", "Jan 2025", "Feb 2025", "Mar 2025"))

ggplot(aes(x=Month, fill=Timepoint), data=projected) + geom_bar() + theme_bw()+
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
  ggtitle(paste0(num_projected, sep=" ", "completing the study by the last date plotted")) + theme(plot.title = element_text(hjust = 0.5))+ geom_vline(xintercept = current_date, linetype="dotted", color="black") + stat_count(geom = "text", color="white", aes(label = ..count..), vjust = 2) + scale_y_continuous(breaks = seq(0,20,by=2))  +theme(axis.text.x = element_text(angle = 55, vjust = 1, hjust = 1)) 

```


## Biopsy Tracker

```{r biopsy, message=FALSE, warning=FALSE, echo=FALSE}

demog$baseline_visit_biopsy_complete <- ifelse(demog$baseline_visit_biopsy_complete=="Complete", 1, 0)
demog$endpoint_visit_biopsy_complete <- ifelse(demog$endpoint_visit_biopsy_complete=="Complete", 1, 0)

biopsy<-demog %>% 
  filter(baseline_visit_biopsy_complete == 1 | endpoint_visit_biopsy_complete == 1) %>% 
  gather(key = "Timepoint", value="biopsy_completed", c("baseline_visit_biopsy_complete", "endpoint_visit_biopsy_complete"))

biopsy<-biopsy %>% 
  dplyr::select(record_id, enrollment_status, sex, agegroup, Group, Timepoint, biopsy_completed) %>% 
  mutate(Timepoint = ifelse(Timepoint=="baseline_visit_biopsy_complete", "Baseline", "Endpoint"))

biopsy$Group[biopsy$record_id == 1076] <- "MICT"

biopsy %>% 
  filter(biopsy_completed==1) %>% 
  group_by(Group, Timepoint) %>% 
  count() %>% 
  kable() %>% 
  kable_styling(full_width = FALSE)

biopsy %>% 
  filter(biopsy_completed==1) %>% 
  arrange(Group, record_id) %>% 
  dplyr::select(record_id, Group, Timepoint) %>% 
  kable() %>% 
  kable_styling(full_width = FALSE)

# biopsy %>% 
#   filter(biopsy_completed==1) %>% 
#   ggplot(aes(x=Group, fill=Timepoint))+geom_bar(position="dodge") + theme_bw() + ylim(c(0,5)) + ggtitle("Number of biopsies complete") + theme(plot.title = element_text(hjust = 0.5))

```

```{r demographics}












```







## Workout Tracking: Wearables vs. REDCap
```{r wearables, message=FALSE, warning=FALSE, echo=FALSE}

## PROCESSING DATA FROM GOOGLE DRIVE ##
# Read in wearable data from Google Drive
data<-read.csv("/Users/aubreykr/Library/CloudStorage/GoogleDrive-aubreykr@stanford.edu/Shared drives/HIIT and Endurance Study/Data/data/workout/workout_allevents.csv")

# Rename to record_id and place at front of dataset
data<-data %>% 
  dplyr::rename(record_id = ppt_id) %>% 
  dplyr::select(record_id, everything())

# Create date 
data<-data %>% 
  mutate(date = as.Date(data$X_realtime))

# Unique participants in the dataset
#length(unique(data$date))
#length(unique(data$record_id))
#(unique(data$sourcetype))
#paste0("Participants in this dataset:", sep=" ", (list(unique(data$record_id))))

# COUNT POLAR WORKOUTS
# Count number of polar workouts
data_polar<-data %>% 
  filter(sourcetype == "WearablePolar") %>% 
  group_by(record_id, date) %>% 
  count()

# Create num minutes per workout column
data_polar$polar_minutes = data_polar$n/60

# COUNT FITBIT WORKOUTS
# Count number of fitbit workouts
data_fitbit<-data %>% 
  filter(sourcetype == "WearableFitbit") %>% 
  group_by(record_id, date) %>% 
  count()

# Create num minutes per workout column
data_fitbit$fitbit_minutes = data_fitbit$n/60

# remove n column
data_polar<-data_polar %>% 
 dplyr::select(-c(n))
data_fitbit<-data_fitbit %>% 
  dplyr::select(-c(n))

# MERGE WEARABLE DATA

# Merge polar and fitbit dataframes
summary<-full_join(data_polar, data_fitbit, by=c("record_id", "date"))

# Total number workouts for Polar, Fitbit
num_polar_workouts <- summary %>% 
 filter(is.na(polar_minutes)!=TRUE) %>% 
 group_by(record_id) %>% 
 summarize(wearable_polar_workouts = length(unique(date)))

num_fitbit_workouts <- summary %>% 
 filter(is.na(fitbit_minutes)!=TRUE) %>% 
 group_by(record_id) %>% 
 summarize(wearable_fitbit_workouts = length(unique(date)))

# SUMMARIZE NUMBER WORKOUTS
#num_polar_workouts
#num_fitbit_workouts

wearables<-full_join(num_polar_workouts, num_fitbit_workouts, by="record_id")
```

```{r wearable merging, messages=FALSE, warning=FALSE, echo=FALSE}

## JOIN MYPHD/POLAR DATA WITH DEMOG FRAME ##

demog<-full_join(demog, wearables, by="record_id")

# calculate num missing
demog$num_missing_polar <- demog$wearable_polar_workouts - demog$number_polar_used
demog$num_missing_fitbit <- demog$wearable_fitbit_workouts - demog$number_fitbit_used

# set up data for table
completed<-demog %>% 
  filter(enrollment_status == "Completed" & Group != "Control") %>% 
  dplyr::select(record_id, number_polar_used, wearable_polar_workouts, num_missing_polar, number_fitbit_used, wearable_fitbit_workouts, num_missing_fitbit)

# create table
completed %>% 
    arrange(as.numeric(record_id)) %>% 
  kable(col.names = c("Record ID", "REDCap polar", "Wearable polar", "Num missing polar", "REDCap fitbit", "Wearable fitbit", "Num missing fitbit"), align="c") %>%
   kable_styling(full_width = F, position = "center", font_size = 14, latex_options = "striped", bootstrap_options = c("striped", "hover"))

```

*Thanks everybody for your hard work!*

```{r vo2 analysis,  echo=FALSE, include=FALSE, message=FALSE}
# vo2 <- gather(demog, key="Timepoint", value="VO2_Max", c(bl_vo2_max, ep_vo2_max))
# vo2$Timepoint[vo2$Timepoint == "bl_vo2_max"] <- "Week 0"
# vo2$Timepoint[vo2$Timepoint == "ep_vo2_max"] <- "Week 12"
# 
# vo2 %>% 
#   filter(Group %in% c("HIIT", "MICT") & enrollment_status == "Completed") %>% 
#   ggplot(aes(x=Timepoint, y=VO2_Max)) + geom_boxplot() + geom_point()+
#   labs(title = "VO2: Baseline vs. Endpoint", x = "Timepoint", y = "VO2 Peak (mL/kg)") + facet_wrap(~Group)+
#   theme(plot.title=element_text(hjust=0.5)) + stat_compare_means(method = "t.test",
#                                                                  label = "p.format",
#                                                                  paired = TRUE)
# # percent change
# demog$vo2change <- (demog$ep_vo2_max - demog$bl_vo2_max) / demog$bl_vo2_max * 100
# demog$bmi_change <- demog$ep_bmi - demog$bl_bmi
# demog$bodyfatchange <- demog$ep_percent_body_fat - demog$bl_percent_body_fat 
# 
# demog %>% 
#   filter(enrollment_status == "Completed" & record_id != 4) %>% 
#   group_by(Group) %>% 
#   summarize(mean = mean(vo2change),
#             sd = sd(vo2change))
# 
# 
# demog %>% 
#   filter(enrollment_status == "Completed" & record_id !="4") %>% 
#   ggplot(aes(y=vo2change, x=Group, color=sex, group=Group)) + geom_boxplot() + geom_point()+ theme_bw()+
#   labs(title = "VO2 Max: Baseline vs. Endpoint", x = "Group", y = "% change in VO2 Max") + 
#   theme(plot.title=element_text(hjust=0.5)) + stat_compare_means(method = "t.test",
#                                                                  label = "p.format",
#                                                                  paired = TRUE) + 
#   geom_hline(yintercept = 0, linetype="dashed") + labs(color="Sex")
# colnames(demog)

```


```{r testdata, echo=FALSE, include=FALSE, message=FALSE, warning=FALSE}

# save<-demog %>% 
#  filter(enrollment_status == "Completed") 
# 
# label(save$Group) <- "Randomization Group"
# label(save$sex)  <- "Sex"
# label(save$agegroup)  <- "Age Group"
# label(save$race) <- "Race"
# label(save$vo2change) <- "VO2 max"
# label(save$bmi_change) <- "BMI "
# label(save$bodyfatchange) <- "Body Fat"
# label(save$quality_workouts_in_REDCap) <- "Workouts Completed"
# units(save$vo2change)       <- "%"
# units(save$bmi_change) <- "kg/m2"
# units(save$bodyfatchange) <- "%"
# units(save$quality_workouts_in_REDCap) <- "%"
# 
# my.render.cont <- function(x) {
#    with(stats.default(x), 
#        sprintf("%0.2f (%0.1f)", MEAN, SD))
# }
#   
# #table1(~sex + agegroup + vo2change + bmi_change + bodyfatchange + I(quality_workouts_in_REDCap/32 * 100) | Group, overall=F,
#      #    render.continuous=my.render.cont, render.missing=NULL, caption="Table 1. Participant Demographics and Preliminary Findings", data=save)
# 
# 
# # another table 1 version
# save<-demog %>% 
#   filter(enrollment_status == "Completed")
# 
# label(save$Group) <- "Randomization Group"
# label(save$sex)  <- "Sex"
# label(save$agegroup)  <- "Age Group"
# label(save$race) <- "Race"
# label(save$vo2change) <- "VO2 max"
# label(save$bmi_change) <- "BMI "
# label(save$bodyfatchange) <- "Body Fat"
# label(save$quality_workouts_in_REDCap) <- "Workouts Completed"
# units(save$vo2change)       <- "%"
# units(save$bmi_change) <- "kg/m2"
# units(save$bodyfatchange) <- "%"
# units(save$quality_workouts_in_REDCap) <- "%"
# 
# save$race[save$race == "Other Asian - please specify below (Hmong, Laotian, Thai, Pakistani, Cambodian, etc)"] <- "Other"
# table1(~sex + agegroup + race + Group,
#          render.continuous=my.render.cont, render.missing=NULL, caption="Table 1. Participant Demographics", data=save)
# 



```





```{r dxa,  echo=FALSE, include=FALSE}

#dxa <- read_xlsx("/Users/aubreykr/Desktop/dxa_feb2.xlsx")

#completed <- demog %>% 
# filter(enrollment_status == "Completed") %>% 
#  select(record_id, Group)

# my.render.cont <- function(x) {
#     with(stats.default(x), 
#         sprintf("%0.2f (%0.1f)", MEAN, SD))
# }
# 
# label(completed$Group) <- "Randomization Group"
# label(completed$sex)  <- "Sex"
# label(completed$agegroup)  <- "Age Group"
# label(completed$race) <- "Race"
# label(completed$bl_vo2_max) <- "VO2 max"
# label(completed$bl_bmi) <- "BMI "
# label(completed$bl_percent_body_fat) <- "Body Fat"
# 
# units(completed$bl_vo2_max)       <- "mL/kg/min"
# units(completed$bl_bmi) <- "kg/m2"
# units(completed$bl_percent_body_fat) <- "%"
# 
# table1(~Group + sex + agegroup + bl_vo2_max +  bl_bmi + bl_percent_body_fat,  data=completed, render.continuous=my.render.cont, caption="Table 1. Baseline Demographics")
# 

# Convert DXA to record ids we can merge
# dxa$study_id <- str_sub(dxa$study_id, -2)
# dxa$study_id <- as.numeric(dxa$study_id)
# dxa<-dxa %>% 
#   rename(record_id = study_id)
# 
# # merge dataframes
# dxa<-dxa %>% 
#   select(record_id, timepoint, contains(c("strength", "endur", "VFAT", "TRUNK", "MUSCLE")))
# dxa<-left_join(completed, dxa, by="record_id")
# # convert to kg
# dxa[, 12:26] <- dxa[, 12:26] / 1000
# 
# 
# # Exploratory plots
# for (col_name in colnames(dxa)[4:26]) {
#   plot_data <- data.frame(value = dxa[[col_name]], timepoint = as.factor(dxa$timepoint), color=dxa$Group)
#   print(ggplot(plot_data, aes(x = timepoint, y = value)) + facet_wrap(~color)+
#     geom_boxplot() + geom_point() +
#     stat_summary(fun = "median", geom = "text", aes(label = round(..y.., 1)), vjust = -0.5, color = "red")+
#     ggtitle(col_name) + theme_bw() + theme(plot.title = element_text(hjust=0.5)))
# }
#   #stat_compare_means(comparisons = list(c("1", "2")), method = "t.test",label = "p.format", paired = TRUE)+
# 
#  my.render.cont <- function(x) {
#     with(stats.default(x), 
#          sprintf("%0.2f (%0.1f)", MEAN, SD))
# }
# table1(~.| Group+timepoint, overall=F,  render.continuous=my.render.cont, data=dxa)
# 
# colnames(demog)
# 
# extra <- demog %>% 
#   select(record_id, sex, agegroup, bl_vo2_max, ep_vo2_max, bl_percent_body_fat, ep_percent_body_fat, bl_bmi, ep_bmi, percent_complete_of_36_wkts, total_workouts_in_REDCap)
# 
# dxa2<-merge(dxa, extra, by="record_id")
# 
# table1(~. | Group + timepoint, overall=F, render.continuous=my.render.cont, data=dxa2)
```




```{r vo2checking, message=FALSE, warning=FALSE, echo=FALSE, include=FALSE}
## Workout adherence and VO2 max

# #vo2<-read_xlsx("/Users/aubreykr/Desktop/VO2_prelim_analysis/Yair_condensed_VO2_1_6_24.xlsx")
# 
# # remove missing rows
# #vo2<-vo2 %>% 
#   #filter(is.na(record_id)==FALSE)
# 
# # change to baseline, endpoint labels
# vo2<-vo2 %>% 
#   mutate(Timepoint = case_when(Timepoint == "V1" ~ "Baseline",
#                                Timepoint == "V2" ~ "Endpoint"))
# 
# # rearrange timepoint to front of column
# vo2<-vo2 %>%
#   relocate(Timepoint, .after=record_id) %>% 
#   arrange(record_id, Timepoint)
# 
# # Calculated variables for VO2, VT1, VT2 
# 
# # % change in VO2 max
# for (i in 1:nrow(vo2)) {
#   if( i %% 2 == 1) { vo2$Peak_VO2_change[i] = 0 }
#   if (i %% 2 == 0) {  vo2$Peak_VO2_change[i] = (vo2$VO2_Peak_mLkg[i] - vo2$VO2_Peak_mLkg[i-1]) /  vo2$VO2_Peak_mLkg[i-1] * 100}
# } 
# 
# 
# # VT1
# for (i in 1:nrow(vo2)) {
#   if( i %% 2 == 1) { vo2$VT1_change[i] = 0 }
#   if (i %% 2 == 0) { vo2$VT1_change[i] = vo2$VT1_vo2[i] - vo2$VT1_vo2[i-1]}
# } 
# 
# # VT2
# for (i in 1:nrow(vo2)) {
#   if( i %% 2 == 1) { vo2$VT2_change[i] = 0 }
#   if (i %% 2 == 0) { vo2$VT2_change[i] = vo2$VT2_vo2[i] - vo2$VT2_vo2[i-1]}
# } 
# 
# # HR 
# for (i in 1:nrow(vo2)) {
#   if( i %% 2 == 1) { vo2$Peak_HR_change[i] = 0 }
#   if (i %% 2 == 0) { vo2$Peak_HR_change[i] = vo2$Peak_HR[i] - vo2$Peak_HR[i-1]}
# } 
# 
# # HR at VT1
# for (i in 1:nrow(vo2)) {
#   if( i %% 2 == 1) { vo2$VT1_HR_change[i] = 0 }
#   if (i %% 2 == 0) { vo2$VT1_HR_change[i] = vo2$VT1_HR[i] - vo2$VT1_HR[i-1]}
# } 
# 
# # HR at VT2
# for (i in 1:nrow(vo2)) {
#   if( i %% 2 == 1) { vo2$VT2_HR_change[i] = 0 }
#   if (i %% 2 == 0) { vo2$VT2_HR_change[i] = vo2$VT2_HR[i] - vo2$VT2_HR[i-1]}
# } 
# 
# # plot VO2 max at BL vs. EP
# #ggplot(aes(x=Timepoint, y=VO2_Peak_mLkg, color=factor(record_id), group=record_id), data=vo2) + geom_point() +
#  # geom_line(aes(group = record_id), linetype = "dashed") +
#  # labs(title = "VO2 Peak: Baseline vs. Endpoint", x = "Timepoint", y = "VO2 Peak (mL/kg)") + facet_wrap(~group) + theme(plot.title=element_text(hjust=0.5))
# 
# # length from first to last vo2 test
# length<-vo2[vo2$Timepoint == "Endpoint",]$date - vo2[vo2$Timepoint == "Baseline",]$date
# record_id<-unique(vo2$record_id)
# dates<-data.frame(cbind(record_id, length))
# vo2<-merge(vo2, dates)
# 
# # change to length in weeks not days
# vo2$length_weeks<-vo2$length / 7
# 
# # plot % change in VO2 max
# vo2 %>% 
#   filter(Timepoint=="Endpoint") %>% 
#   ggplot(aes(x=group, y=Peak_VO2_change)) + geom_boxplot() + geom_hline(yintercept=0, linetype="dotted") + geom_point() + 
#    stat_summary(
#     fun = median,
#     geom = "text",
#     vjust = -0.6,
#     size = 3.2,
#     aes(group = group, label = sprintf("%.2f", ..y..))
#   ) + ylab("Percent change in VO2 max") + xlab("Randomization Group") + ggtitle("Percent change in VO2 max after 12 Weeks") + theme(plot.title=element_text(hjust=0.5))
# 
# # add adherence data and dxa data to vo2 dataset
# subset<-demog %>% 
#   filter(enrollment_status == "Completed") %>% 
#   dplyr::select(record_id, sex, agegroup, total_workouts_in_REDCap, quality_workouts_in_REDCap, percent_complete_of_36_wkts)
# 
# vo2<-merge(vo2, subset, by="record_id")
# 
# # plot correlation between workout adherence and change in VO2
# 
# library(ggpubr)
# 
# vo2 %>% 
#   filter(Timepoint=="Endpoint" & group!="Con" & record_id!="4") %>% 
#   ggplot(aes(x=quality_workouts_in_REDCap, y=Peak_VO2_change, color=group)) + geom_point() + 
#     geom_smooth(method = "lm", se = FALSE, size=0.5) +  # Add linear regression line without confidence interval
#   labs(title = "Number Successful Workouts vs. % Change in VO2 max",
#        x = "Number Successful Workouts",
#        y = "Percent Change in VO2") + theme(plot.title=element_text(hjust=0.5)) + geom_hline(yintercept=0, linetype="dotted", size=0.8)+
#   stat_cor(method = "pearson")
# 
# vo2 %>% 
#   filter(Timepoint=="Endpoint" & group!="Con") %>% 
#   ggplot(aes(x=total_workouts_in_REDCap, y=Peak_VO2_change, color=group)) + geom_point() + 
#   labs(title = "Number Total Workouts vs. % Change in VO2 max",
#        x = "Number Total Workouts",
#        y = "Percent Change in VO2") + theme(plot.title=element_text(hjust=0.5)) + geom_hline(yintercept=0, linetype="dotted", size=0.8)
# 
# 
# # look at intervention length
# 
# vo2 %>% 
#   filter(Timepoint == "Endpoint" & group!="Con") %>% 
#   ggplot(aes(x=length_weeks, y=Peak_VO2_change, group=group, color=group)) + geom_point() +  geom_smooth(method = "lm", se = FALSE, size=0.5) + labs(title = "Number weeks between VO2 max visits vs. % Change in VO2 max",
#        x = "Number weeks between VO2 max visits",
#        y = "Percent Change in VO2") + theme(plot.title=element_text(hjust=0.5)) + geom_hline(yintercept=0, linetype="dotted", size=0.8)+
#    stat_cor(method = "pearson")
# 
# vo2 %>% 
#   filter(Timepoint == "Endpoint" & group!="Con") %>% 
#   ggplot(aes(x=length_weeks, y=quality_workouts_in_REDCap, group=group, color=group)) + geom_point() +  geom_smooth(method = "lm", se = FALSE, size=0.5) + labs(title = "Number weeks between VO2 max visits vs. num quality workouts",
#        x = "Number weeks between VO2 max visits",
#        y = "Number quality workouts in redcap") + theme(plot.title=element_text(hjust=0.5)) +
#    stat_cor(method = "pearson")
# 
# vo2<-vo2 %>% 
#   mutate(wkts_per_week = quality_workouts_in_REDCap / length_weeks)
# 
# # average number of workouts per week
# vo2 %>% 
#   filter(Timepoint == "Endpoint" & group!="Con" & record_id!="4") %>% 
#   ggplot(aes(x=wkts_per_week, y=Peak_VO2_change, group=group, color=group)) + geom_point() +  geom_smooth(method = "lm", se = FALSE, size=0.5) + labs(title = "Average num workouts per week vs. VO2 change",
#        x = "Average num workouts per week",
#        y = "Percent Change in VO2") + theme(plot.title=element_text(hjust=0.5)) +
#    stat_cor(method = "pearson") + geom_hline(yintercept=0, linetype="dotted", size=0.8)
# 
# 
# #model<-lm(percent_change_vo2 ~ sex + agegroup + weight_kg + quality_workouts_in_REDCap + length_weeks + group, data = test)
# #summary(model)
# 
# 
# vo2 %>% 
#   filter(Timepoint == "Endpoint" & group!="Con") %>% 
#   ggplot(aes(x=quality_workouts_in_REDCap, y=Peak_VO2_change, color=group)) + geom_point() + 
#   labs(title = "Changes by Sex, Age",
#        x = "Number Successful Workouts",
#        y = "Percent Change in VO2") + theme(plot.title=element_text(hjust=0.5)) + geom_hline(yintercept=0, linetype="dotted", size=0.8) + facet_wrap(~sex + agegroup)
# 
# 
# ```
# 
# ```{r vt1, message=FALSE, warning=FALSE, echo=FALSE}
# 
# ggplot(aes(x=Timepoint, y=VO2_Peak_mLkg), data=vo2) + geom_boxplot() + geom_point()+
#  labs(title = "VO2: Baseline vs. Endpoint", x = "Timepoint", y = "VO2 Peak (mL/kg)") + facet_wrap(~group) +
#   theme(plot.title=element_text(hjust=0.5)) + stat_compare_means(
#   method = "t.test", # You can also use "wilcox.test" for non-parametric test
#   label = "p.format",
#   paired = TRUE
# )
# 
# ggplot(aes(x=Timepoint, y=VT1_vo2), data=vo2) + geom_boxplot() + geom_point()+
#  labs(title = "VT1: Baseline vs. Endpoint", x = "Timepoint", y = "VO2 Peak (mL/kg)") + facet_wrap(~group) +
#   theme(plot.title=element_text(hjust=0.5)) + stat_compare_means(
#   method = "t.test", # You can also use "wilcox.test" for non-parametric test
#   label = "p.format",
#   paired = TRUE
# )
# 
# ggplot(aes(x=Timepoint, y=VT2_vo2), data=vo2) + geom_boxplot() + geom_point()+
#  labs(title = "VT2: Baseline vs. Endpoint", x = "Timepoint", y = "VO2 Peak (mL/kg)") + facet_wrap(~group) +
#   theme(plot.title=element_text(hjust=0.5)) + stat_compare_means(
#   method = "t.test", # You can also use "wilcox.test" for non-parametric test
#   label = "p.format",
#   paired = TRUE
# )
# 
# 
# ggplot(aes(x=Timepoint, y=Peak_HR), data=vo2) + geom_boxplot() + geom_point()+
#  labs(title = "VO2 HR: Baseline vs. Endpoint", x = "Timepoint", y = "VO2 Peak (mL/kg)") + facet_wrap(~group) +
#   theme(plot.title=element_text(hjust=0.5)) + stat_compare_means(
#   method = "t.test", # You can also use "wilcox.test" for non-parametric test
#   label = "p.format",
#   paired = TRUE
# )
# 
# ggplot(aes(x=Timepoint, y=VT1_HR), data=vo2) + geom_boxplot() + geom_point()+
#  labs(title = "VT1 HR: Baseline vs. Endpoint", x = "Timepoint", y = "VO2 Peak (mL/kg)") + facet_wrap(~group) +
#   theme(plot.title=element_text(hjust=0.5)) + stat_compare_means(
#   method = "t.test", # You can also use "wilcox.test" for non-parametric test
#   label = "p.format",
#   paired = TRUE
# )
# 
# ggplot(aes(x=Timepoint, y=VT2_HR), data=vo2) + geom_boxplot() + geom_point()+
#  labs(title = "VT2 HR: Baseline vs. Endpoint", x = "Timepoint", y = "VO2 Peak (mL/kg)") + facet_wrap(~group) +
#   theme(plot.title=element_text(hjust=0.5)) + stat_compare_means(
#   method = "t.test", # You can also use "wilcox.test" for non-parametric test
#   label = "p.format",
#   paired = TRUE
# )

```

```{r scoringefficacy, message=FALSE, warning=FALSE, echo=FALSE, include=FALSE}

# motiv<-read.csv("/Users/aubreykr/Desktop/exselfefficacy.csv")
# 
# motiv<-motiv %>% 
#   rename(Timepoint = redcap_event_name) %>% 
#   dplyr::select(-c(exeffiency_date_enrolled, exercise_selfefficacy_complete)) %>% 
#   mutate(Timepoint = case_when(Timepoint == "scr_bl_ep_arm_1" ~ "Baseline",
#                                Timepoint == "week_12_ep_surveys_arm_1" ~ "Endpoint",
#                                Timepoint == "month_6_arm_1" ~ "Month 6"))
# 
# # self-efficacy score ranges from 9-28; sum the 9 questions.
# motiv$efficacy_score <- rowSums(motiv[, 3:11])
# 
# # combine efficacy with vo2 data
# motiv<-motiv %>% 
#   dplyr::select(record_id, Timepoint, efficacy_score)
#  
# motiv$record_id <- as.numeric(motiv$record_id)
# 
# vo2<-left_join(vo2, motiv, by=c("record_id", "Timepoint"))
# 
# vo2 %>% 
#   filter(record_id != "8" & record_id != "10" & record_id != "29") %>% 
#   ggplot(aes(x=Timepoint, y=efficacy_score)) + geom_boxplot() + geom_point()+
#   labs(title = "Exercise Self-Efficacy: Baseline vs. Endpoint", x = "Timepoint", y = "Exercise Self-Efficacy (Score 9-36)") + facet_wrap(~group) +
#   theme(plot.title=element_text(hjust=0.5)) + stat_compare_means(
#   method = "t.test", # You can also use "wilcox.test" for non-parametric test
#   label = "p.format",
#   paired = TRUE
# )
# 
# vo2 %>% 
#   filter(record_id != "8" & record_id != "10" & record_id != "29") %>% 
#   ggplot(aes(x=Timepoint, y=efficacy_score, group=group, color=factor(record_id))) + geom_point()  + geom_line(aes(group = record_id), linetype = "dashed") + facet_wrap(~group)
# 
# # plot VO2 max at BL vs. EP
# #ggplot(aes(x=Timepoint, y=VO2_Peak_mLkg, color=factor(record_id), group=record_id), data=vo2) + geom_point() +
#  # geom_line(aes(group = record_id), linetype = "dashed") +
#  # labs(title = "VO2 Peak: Baseline vs. Endpoint", x = "Timepoint", y = "VO2 Peak (mL/kg)") + facet_wrap(~group) + theme(plot.title=element_text(hjust=0.5))
# 
# # make demographics tables
# demog<-vo2 %>% 
#   filter(Timepoint == "Baseline")
# 
# label(demog$sex) <- "Sex"
# label(demog$agegroup) <- "Age Group"
# table1(~sex + agegroup | group, data=demog, overall=F, caption="Demographics for Completed Participants")
```



```{r test, echo=FALSE}
# Run a shiny app
# library(ggplot2)
# library(shiny)
# library(DT)
# 
# # Server
# server <- function(input, output) {
#   # Filter data based on selections
#   output$table <- DT::renderDataTable({
#     data <- completed_mict
#     datatable(data, options = list(rownames = FALSE))
#   })
# }
# 
# # UI
# ui <- fluidPage(
#   titlePanel("Basic DataTable"),
#   # Create a new row for the table.
#   DT::dataTableOutput("table")
# )
# 
# shinyApp(ui, server)

```