adrace.rmd

---
title: "Untitled"
author: "Smriti karki"
date: "9/2/2020"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```


##library loading
```{r}
library(haven)
library(dplyr)
library(tidyverse)
library(Hmisc)
library(expss)
```
```{r}
adrace <- read_sas("E:/Numeric mind/adam/New_Study-master/New_Study-master/adam/adrace.sas7bdat",NULL)
dm <- read_sas("E:/Numeric mind/adam/New_Study-master/New_Study-master/sdtm/dm.sas7bdat",NULL)
adsl <- read_sas("E:/Numeric mind/adam/New_Study-master/New_Study-master/adam/adsl.sas7bdat",NULL) 
suppdm <- read_sas("E:/Numeric mind/adam/New_Study-master/New_Study-master/sdtm/suppdm.sas7bdat",NULL)
```
```{r}
adrace1 <- dm %>% select(STUDYID,USUBJID)
temp <- adsl %>% select(USUBJID,SITEID,LTSUBJID,AGE,AGEU,SEX,RACE,RACEN,COUNTRY,PPROTFL,SCRNFL,ITTEFL,SAFFL,ENRLFL,LTFUFL,TRT01P,TRT01PN,TRT01A,TRT01AN,TRTSDT,TRTEDT,MNTEDT,ARACE)
adrace1 <- adrace1 %>% full_join(temp)
```

#SUPPDM.QVAL WHERE QNAM=RACEOR, RACEORX WHERE X =1,2,.... WHEN QVAL="AFRICAN AMERICAN/AFRICAN HERITAGE" REPORT AS "BLACK OR AFRICAN AMERICAN", WHEN QVAL="AMERICAN INDIAN OR ALASKAN NATIVE" REPORT AS "AMERICAN INDIAN OR ALASKA NATIVE".
```{r}
adrace11 <- suppdm %>% select(USUBJID,QNAM,QVAL) %>% 
  filter(QNAM =="RACEOR" | QNAM =="RACEOR1" | QNAM =="RACEOR2") %>% 
  rename(RACEOR = QVAL)
adrace1 <- adrace1 %>% left_join(adrace11)
adrace1$RACEOR[adrace1$RACEOR  == "AFRICAN AMERICAN/AFRICAN HERITAGE"]  <-  "BLACK OR AFRICAN AMERICAN"
adrace1$RACEOR[adrace1$RACEOR  == "AMERICAN INDIAN OR ALASKAN NATIVE"]  <-  "AMERICAN INDIAN OR ALASKA NATIVE"

label(adrace1$RACEOR) <- "Original Race"
# rm(adrace11)
```
#TRTP,TRTPN,TRTA,TRTAN
```{r}
adrace1 <- adrace1 %>% mutate(TRTP = TRT01P)
adrace1 <- adrace1 %>% mutate(TRTPN = TRT01PN)
adrace1 <- adrace1 %>% mutate(TRTA = TRT01A)
adrace1 <- adrace1 %>% mutate(TRTAN = TRT01AN)
adrace1 <- apply_labels(adrace1,TRTP = "Planned Treatment")
adrace1 <- apply_labels(adrace1,TRTPN = "Planned Treatment (N)")
adrace1 <- apply_labels(adrace1,TRTA = "Actual Treatment")
adrace1 <- apply_labels(adrace1,TRTAN = "Actual Treatment (N)")
```
#ASEQ
```{r}
adrace1 <- adrace1 %>% 
  group_by(USUBJID) %>% 
  mutate(ASEQ = row_number())
label(adrace1$ASEQ)<- "Analysis Sequence Number"
```
#FOR ASIAN RACEOR VALUES, ALL TEXT AFTER THE DASH IS STRIPPED PRIOR TO CONCATENATION OR ASSIGNMENT. IF DM.RACE='ASIAN' ASSIGNED AS FOLLOWS (USING THE VALUES OF DM.RACEOR): SUBJECTS WHO SELECT ONE OR MORE OF JAPANESE HERITAGE, EAST ASIAN HERITAGE, AND SOUTH EAST ASIAN HERITAGE, AND NO OTHER ASIAN SUBCATEGORY, ARE ASSIGNED 'JAPANESE HERITAGE/EAST ASIAN HERITAGE/SOUTH EAST ASIAN HERITAGE'; SUBJECTS WHO SELECT CENTRAL/SOUTH ASIAN HERITAGE AND ANY OTHER ASIAN SUBCATEGORY ARE ASSIGNED 'MIXED ASIAN RACE'; SUBJECTS WHO SELECT ONLY CENTRAL/SOUTH ASIAN HERITAGE ARE ASSIGNED 'CENTRAL/SOUTH ASIAN HERITAGE'. IF DM.RACE='WHITE' AND MORE THAN ONE SUBCATEGORY IS PRESENT ASSIGN AS 'MIXED WHITE RACE';. IF DM.RACE='MULTIPLE' THEN RACECOMB IS THE CONCATENATION OF ALL DM.RACEX IN ALPHABETICAL ORDER.
```{r}
test <- adrace1 %>% 
  select(ASEQ,RACE,ARACE,ASEQ,RACEOR) %>% 
  mutate(raceor_pre = sub("-.*","", RACEOR)) %>% 
  mutate(raceor_sup = sub(".*-","",RACEOR))
test1 <- test %>% filter(RACE == "ASIAN") %>% 
  mutate(RACECOMP = case_when(
    (raceor_sup == "JAPANESE HERITAGE" | raceor_sup == "EAST ASIAN HERITAGE" | raceor_sup ==  "SOUTH EAST ASIAN HERITAGE")~ "JAPANESE HERITAGE/EAST ASIAN HERITAGE/SOUTH EAST ASIAN HERITAGE",
    raceor_sup == "CENTRAL/SOUTH ASIAN HERITAGE" ~ "CENTRAL/SOUTH ASIAN HERITAGE",
    TRUE ~ "MIXED ASIAN RACE")) %>% 
  select(USUBJID,ASEQ,RACECOMP)
test2 <- test %>% 
  filter(ARACE == "MULTIPLE") %>% 
  arrange(raceor_pre) %>% 
  group_by(USUBJID) %>% 
  mutate(RACECOMP= paste0(raceor_pre,collapse = " & ")) %>% 
  select(USUBJID,ASEQ,RACECOMP)
test1 <- test %>% full_join(test2) 
test1$RACECOMP[is.na(test1$RACECOMP)]<- ""
test1$RACECOMP <- as.character(test1$RACECOMP)
adrace1$RACECOMP <- as.character(adrace1$RACECOMP)

adrace1 <- full_join(test1,adrace1) 
adrace1 <- adrace1 %>% select(-raceor_pre,-raceor_sup)

```
#FOR DM.RACE='WHITE' OR 'ASIAN', IF SUBJECT HAS MORE THAN ONE DM.RACEOR VALUE SET TO 'MIXED WHITE RACE' OR 'MIXED ASIAN RACE', RESPECTIVELY; FOR DM.RACE='MULTIPLE' ASSIGNED AS THE CONCATENATION OF ALL ADRACE.RACEOR VALUES FOR THAT SUBJECT, SEPARATING EACH VALUE WITH " & ".

```{r}
test <- adrace1 %>% 
  select(USUBJID,ASEQ,RACE,RACEOR,ARACE) %>% 
  mutate(raceor_pre = sub("-.*","", RACEOR)) %>% 
  mutate(raceor_sup = sub(".*-","",RACEOR))
test1 <- test %>% 
    filter(RACE== "WHITE" | RACE == "ASIAN") %>% 
    mutate(RACECDET = raceor_sup)
test1$RACECDET <- as.character(test1$RACECDET)

test2 <- test %>% 
    filter(ARACE == "MULTIPLE") %>% 
    arrange(raceor_pre) %>% 
    group_by(USUBJID) %>% 
    mutate(RACECDET= paste0(raceor_pre,collapse = " & ")) %>% 
    select(USUBJID,ASEQ,RACECDET)
test1 <- test1 %>% full_join(test2) %>% select(USUBJID,ASEQ,RACECDET)
adrace1 <- full_join(test1,adrace1) %>% distinct()
adrace1 <- adrace1 %>% select(-QNAM)
rm(adrace11,temp,test,test1,test2)
adrace1$RACECDET[is.na(adrace1$RACECDET)] <- " "
```
```{r}
adrace1$RACECOMP[is.na(adrace1$RACECOMP)] <- " "
```
```{r}
adrace1 <- adrace1 %>% select(STUDYID,SITEID,USUBJID,LTSUBJID,AGE,AGEU,SEX,RACE,RACEN,COUNTRY,PPROTFL,SCRNFL,ITTEFL,SAFFL,ENRLFL,LTFUFL,TRT01P,TRT01PN,TRT01A,TRT01AN,TRTP,TRTPN,TRTA,TRTAN,TRTSDT,TRTEDT,MNTEDT,ASEQ,ARACE,RACEOR,RACECOMP,RACECDET)
adrace1<- apply_labels(adrace1,RACECOMP = "Race and Racial Combinations")
adrace1<- apply_labels(adrace1, RACECDET= "Race and Racial Combination Details")

```