biostatistics_project.r

biostatistics_project.r

Dataset <- read.csv("Dataset.csv", stringsAsFactors=TRUE, header=TRUE)

Dataset <- Dataset[Dataset$Assignment != "",]

View(Dataset)

drugs <- data.frame(
    Name = c("A", "B", "C", "D", "E", "F", "G", "H", "I", "J", "K", "L", "M", "N", "O", "P", "Q", "R", "S", "T", "U", "V", "W", "X", "Y", "Z", "AA", "AB", "AC", "AD"),
    Color = c("Red", "Red", "Green", "Yellow", "Red", "Yellow", "Red", "Red", "Red", "Red", "Green", "Red", "Red", "Red", "Green", "Red", "Red", "Green", "Yellow", "Red", "Red", "Yellow", "Yellow", "Red", "Yellow", "Red", "Yellow", "Red", "Red", "Green")
)

red_cols <- drugs$Name[drugs$Color == "Red"]
yellow_cols <- drugs$Name[drugs$Color == "Yellow"]
green_cols <- drugs$Name[drugs$Color == "Green"]

Dataset$red = rowSums(Dataset[, red_cols])
Dataset$yellow = rowSums(Dataset[, yellow_cols])
Dataset$green = rowSums(Dataset[, green_cols])

# Move "red" and "yellow" drugs to "green" for EG patients in both experimental and control groups
EG_patients <- Dataset$Therapeutic.Guidances == "EG"
Dataset[EG_patients, "green"] <- rowSums(Dataset[EG_patients, c("red", "yellow", "green")])
Dataset[EG_patients, "red"] <- 0
Dataset[EG_patients, "yellow"] <- 0

# Move "yellow" drugs to "green" for CT patients in the experimental group
# (we have CT only in experimental group)
CT_patients <- Dataset$Therapeutic.Guidances == "CT"
Dataset[CT_patients, "green"] <- rowSums(Dataset[CT_patients, c("yellow", "green")])
Dataset[CT_patients, "yellow"] <- 0

linear_model <- lm(LOS ~ GENDER + AGE + RACE.ETHNICITY + MD + Assignment + Therapeutic.Guidances + red + yellow + green, data=Dataset)
summary(linear_model)