I would like to make a paired barchart with month as the fill. Month, however, is a selectinput option and I'm unsure of how to incorporate that into the ggplot commands. Using fill = input$Month or fill= "Reconciled.Month" is leaving the graphics blank. The code worked with only one month, but I would like to be able to compare performance changes over time to see if there has been any change in causal behaviors and, if so, if they are for the better or worse. data sample :
Provider Reconciled Month Per Visit Bill Rate- w/Mal Per Visit Bill Rate Location
Andrews 1 27.68 30.57 Illinois
Davids 2 25.08 29.08 New York
Stein 3 31.39 31.39 New York
Goodman 2 26.00 30.00 Wyoming
Roberts 3 24.34 25.74 Utah
Buckley 1 29.54 33.54 Ohio
Castro 3 0.00 0.00 Ohio
Code:
library(shiny)
library(shinydashboard)
library(ggplot2)
library(rsconnect)
library(lubridate)
library(scales)
library(plyr)
library(dplyr)
library(RColorBrewer)
AdjustedData<- read.csv("Adjusted Data.csv")
Target <- data.frame(yintercept= 40.00, Target= factor(40.00))
AdjustedData$Percentage <- round(AdjustedData$Total.Procedure/AdjustedData$Reconciled.Visits*100, 2)
AdjustedData$Percentage[is.na(AdjustedData$Percentage)] <- 0
opt <- read.csv("optimization.csv")
opt$OptPer <- opt$Optimization*100
opt$Location <- as.character(opt$Location)
opt$Location <- ifelse(opt$Location == "New Jersey" | opt$Location == "Indiana" | opt$Location == "Pennsylvania", "New Jersey/Indiana/Pennsylvania", opt$Location)
opt$Location <- as.factor(opt$Location)
mu <- mean(opt$OptPer)
res <- as.vector(opt$OptPer-mu)
opt <- cbind(opt, res)
ui <- dashboardPage(
dashboardHeader(title = "Performance Metrics"),
dashboardSidebar(
sidebarMenu(
menuItem("Dashboard", tabName = "dashboard", icon = icon("dashboard")),
menuItem("Bill Rate information", tabName = "Bill Rate Information", icon = icon("dollar-sign")),
menuItem("Optimization Rates", tabName = "Optimization Rates", icon = icon("whmcs"))
)
),
dashboardBody(
tabItem( tabName = "dashboard"),
tabItem( tabName = "Bill Rate Information",
fluidRow(
box(title = "Options",
selectInput(inputId = ".Month",
label = "Month:",
choices = c("January" = "1", "February" = "2", "March" = "3"), multiple = TRUE),
selectInput(inputId = "Location",
label = "Location:",
choices = c("Illinois", "Wound MD", "Ohio", "New York")), width = 1000, height = 200), multiple = TRUE,
box(plotOutput("barplot", height = 500, width= 800)),
box(plotOutput("barplot2", height= 500, width = 800))),
box(plotOutput("barplot3", height = 500, width = 800))
)))
server <- function(input, output) {
data <- reactive({
dataset <- subset(AdjustedData, Location == input$Location & Reconciled.Month == input$Month)
})
data2 <- reactive({
dataset2 <- subset(opt, Location == input$Location & Reconciled.Month == input$Month)
})
output$barplot <-renderPlot({
ggplot(data= data(), aes_string(x="Provider",y= "Per.Visit.Bill.Rate")) +
geom_bar(stat="identity", position = "dodge")+
labs(caption = "Data sourced from Provider Compensation Analysis and Monthly Summary \nof Billed Physician Encounters Reports. Rates have been adjusted to exclude \nthe deduction per visit for malpractice insurance where applicable.")+
theme_light()+
theme(plot.caption = element_text(size= 8, hjust = .5))+
theme(axis.text.x=element_text(angle= 45, vjust=.5)) +
ggtitle("Per Visit Bill Rate", input$Location)+
theme(plot.title = element_text(size = 22, hjust = .5, family = "serif"))+
theme(plot.subtitle = element_text(size = 18, hjust = .5, family = "serif"))+
geom_text(aes(label=sprintf("$%.2f",Per.Visit.Bill.Rate)), position=position_dodge(width=0.9),hjust= .5, vjust=0,angle= 0) +
scale_y_continuous(name = "Per Visit Bill Rate ($)", limits = c(0,100),breaks =seq(0,100,10), labels = dollar)+
scale_color_continuous()+
geom_hline(aes(yintercept = yintercept , linetype = Target), data = Target, color = "red", size = 1.0)+
guides(fill=guide_legend(title = "Per Visit Bill Rate ($)", reverse = TRUE))+
theme(plot.title = element_text(hjust = 0.5))
})
output$barplot2 <-renderPlot({
ggplot(data= data(), aes_string(x="Provider",y= "Percentage", fill= "Reconciled.Month")) +
geom_bar(stat = "identity")+
geom_text(aes(x=Provider, y= Percentage, label=sprintf("%.1f%%", Percentage)), position=position_dodge(width=0.9),hjust= .5, vjust= 0 ,angle= 0)+
labs(caption = "Data sourced from Provider Compensation Analysis and \nMonthly Summary of Billed Physician Encounters Reports, March 2018 \nRates have been calculated using the sum of all procedure codes.")+
scale_y_continuous(name = "Percentage (%)", limits = c(0,100),breaks =seq(0,100,10))+
scale_color_continuous()+
theme(plot.caption = element_text(size= 8, hjust = .5))+
theme(axis.text.x=element_text(angle= 0, vjust=.5))+
ggtitle("Procedure Percentages", input$Location)+
theme(plot.title = element_text(size = 22, hjust = .5, family = "serif"))+
theme(plot.subtitle = element_text(size = 18, hjust = .5, family = "serif"))+
guides(fill=guide_legend(title = "Month", reverse =FALSE))
})
output$barplot3 <- renderPlot({
ggplot(data= data2(), aes_string(x = "Provider", y= "OptPer", fill= "OptPer"))+
geom_bar(stat = "identity") +
geom_text(aes(x= Provider, y= OptPer, label=sprintf("%.1f%%", OptPer)), position = position_dodge(width=.6), hjust= .5, vjust=0)+
theme_light()+
theme(axis.text.x=element_text(angle= 45, vjust=.5))+
scale_color_continuous()+
ggtitle("Optimization Percentages", input$Location)+
labs(caption = "Data sourced from Provider Compensation Analysis Reports, March 2018")+
theme(plot.caption = element_text(size= 8, hjust = .5))+
ylab("Optimization Percentage")+
theme(plot.title = element_text(size = 22, hjust = .5, family = "serif"))+
theme(plot.subtitle = element_text(size = 18, hjust = .5, family = "serif"))+
scale_y_continuous(limits= c(0,100), labels = function(x) paste0(x, "%")) +
guides(fill=guide_legend(title = "Optimization Percentage (%)", reverse = TRUE))
})
}
shinyApp(ui = ui, server = server)
