DaysOfCode.Rmd

---
title: "Days of Code Austin 2018"
author: "Carlos Santillan"
date: "February 10, 2018"
#output:   md_document
output:   html_document
---
# Analysis of Opiod drug overdoses data from the CDC


Carlos Santillan

csantill@gmail.com

https://www.linkedin.com/in/carlos-santillan/
  
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)

```




```{r libraries ,echo=FALSE, Warning=FALSE,message=FALSE}
library(knitr)
library(tidyverse)
library(ggplot2)
library(fiftystater)

data("fifty_states") # this line is optional due to lazy data loading
```

## Load the data
Load CDC data 
Dataset: Underlying Cause of Death, 1999-2016

### Load Monthly Data 

Group By: State, Year, Month, ICD Chapter

ICD-10 Codes: 

* X40 (Accidental poisoning by and exposure to nonopioid analgesics, antipyretics and antirheumatics)
* X41 (Accidental poisoning by and exposure to antiepileptic, sedative-hypnotic, antiparkinsonism and psychotropic drugs, not elsewhere classified)
* X42 (Accidental poisoning by and exposure to narcotics and psychodysleptics [hallucinogens], not elsewhere classified)
* X43 (Accidental poisoning by and exposure to other drugs acting on the autonomic nervous system)
* X44 (Accidental poisoning by and exposure to other and unspecified drugs, medicaments and biological substances) 
* X60 (Intentional self-poisoning by and exposure to nonopioid analgesics, antipyretics and antirheumatics)
* X61 (Intentional self-poisoning by and exposure to antiepileptic, sedative-hypnotic, antiparkinsonism and psychotropic drugs, not elsewhere classified)
* X62 (Intentional self-poisoning by and exposure to narcotics and psychodysleptics [hallucinogens], not elsewhere classified)
* X63 (Intentional self-poisoning by and exposure to other drugs acting on the autonomic nervous system) 
* X64 (Intentional self-poisoning by and exposure to other and unspecified drugs, medicaments and biological substances) 
* Y10 (Poisoning by and exposure to nonopioid analgesics, antipyretics and antirheumatics, undetermined intent) 
* Y11 (Poisoning by and exposure to antiepileptic, sedative-hypnotic, antiparkinsonism and psychotropic drugs, not elsewhere classified, undetermined intent), 
* Y12 (Poisoning by and exposure to narcotics and psychodysleptics [hallucinogens], not elsewhere classified, undetermined intent)
* Y13 (Poisoning by and exposure to other drugs acting on the autonomic nervous system, undetermined intent)
* Y14 (Poisoning by and exposure to other and unspecified drugs, medicaments and biological substances, undetermined intent)

Source: 

Centers for Disease Control and Prevention, National Center for Health Statistics. Underlying Cause of Death 1999-2016 on CDC WONDER Online Database, released December, 2017. Data are from the Multiple Cause of Death Files, 1999-2016, as 
 compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program. Accessed at http://wonder.cdc.gov/ucd-icd10.html on Feb 10, 2018 4:36:12 PM 

* Note: Year, Month data for the state of South Dakota and North Dakota was unavailable due to CDC Suppression, data was inputed by floor (AnnualDeaths/ 12 ) for each year

Use of this data allowed under https://wonder.cdc.gov/DataUse.html#

CDC Data was generated from the following Query https://wonder.cdc.gov/controller/saved/D76/D29F799


```{r loadmonthlydata, echo=FALSE}

cdcMonthlyStateData        <- read_tsv("data/statefull_2018.txt")
colnames(cdcMonthlyStateData) <- make.names(colnames(cdcMonthlyStateData)) # Set standard names (remove spaces, etc)
cdcMonthlyStateData$Notes  <- cdcMonthlyStateData$Notes <- NULL
cdcMonthlyStateData$Deaths <- as.integer(cdcMonthlyStateData$Deaths)
cdcMonthlyStateData <- cdcMonthlyStateData %>% separate(Month.Code,c("yr","Month.Code"))
cdcMonthlyStateData$Year<- as.character(cdcMonthlyStateData$Year)
cdcMonthlyStateData$YearMonth <- paste(cdcMonthlyStateData$Year , cdcMonthlyStateData$Month.Code,sep="-")

cdcMonthlyStateData <- cdcMonthlyStateData %>% select(
  State,YearMonth,Year,Month.Code,Deaths)
#save the RDS
saveRDS(cdcMonthlyStateData,"data/cdcMonthlyStateData.rds")
saveRDS(cdcMonthlyStateData,"shiny/data/cdcMonthlyStateData.rds")


```
### Load Annual Data 

Group By: State, Year, ICD Chapter

ICD-10 Codes: 

* X40 (Accidental poisoning by and exposure to nonopioid analgesics, antipyretics and antirheumatics)
* X41 (Accidental poisoning by and exposure to antiepileptic, sedative-hypnotic, antiparkinsonism and psychotropic drugs, not elsewhere classified)
* X42 (Accidental poisoning by and exposure to narcotics and psychodysleptics [hallucinogens], not elsewhere classified)
* X43 (Accidental poisoning by and exposure to other drugs acting on the autonomic nervous system)
* X44 (Accidental poisoning by and exposure to other and unspecified drugs, medicaments and biological substances) 
* X60 (Intentional self-poisoning by and exposure to nonopioid analgesics, antipyretics and antirheumatics)
* X61 (Intentional self-poisoning by and exposure to antiepileptic, sedative-hypnotic, antiparkinsonism and psychotropic drugs, not elsewhere classified)
* X62 (Intentional self-poisoning by and exposure to narcotics and psychodysleptics [hallucinogens], not elsewhere classified)
* X63 (Intentional self-poisoning by and exposure to other drugs acting on the autonomic nervous system) 
* X64 (Intentional self-poisoning by and exposure to other and unspecified drugs, medicaments and biological substances) 
* Y10 (Poisoning by and exposure to nonopioid analgesics, antipyretics and antirheumatics, undetermined intent) 
* Y11 (Poisoning by and exposure to antiepileptic, sedative-hypnotic, antiparkinsonism and psychotropic drugs, not elsewhere classified, undetermined intent), 
* Y12 (Poisoning by and exposure to narcotics and psychodysleptics [hallucinogens], not elsewhere classified, undetermined intent)
* Y13 (Poisoning by and exposure to other drugs acting on the autonomic nervous system, undetermined intent)
* Y14 (Poisoning by and exposure to other and unspecified drugs, medicaments and biological substances, undetermined intent)

Source: 

 Centers for Disease Control and Prevention, National Center for Health Statistics. Underlying Cause of Death"
"1999-2016 on CDC WONDER Online Database, released December, 2017. Data are from the Multiple Cause of Death Files, 1999-2016, as"
"compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program. Accessed"
"at http://wonder.cdc.gov/ucd-icd10.html on Feb 13, 2018 11:26:09 AM"


Use of this data allowed under https://wonder.cdc.gov/DataUse.html#

CDC Data was generated from the following Query https://wonder.cdc.gov/controller/saved/D76/D29F799
https://wonder.cdc.gov/controller/saved/D76/D30F044

```{r loadAnnualdata, echo=FALSE}

cdcAnnualData        <- read_tsv("data/statefull_annual_2018.txt")
colnames(cdcAnnualData) <- make.names(colnames(cdcAnnualData)) # Set standard names (remove spaces, etc)
cdcAnnualData$Notes  <- cdcAnnualData$Notes <- NULL
cdcAnnualData$Year<- as.character(cdcAnnualData$Year)

#str(cdcAnnualData)

cdcAnnualData <- cdcAnnualData %>% select(State,Year,Deaths,Population,Crude.Rate,Age.Adjusted.Rate)
cdcAnnualData %<>%  arrange(State,desc(Year))  %>% group_by(State) %>%mutate(PercentChange = round(((Deaths - lead(Deaths)) * 100 / lead(Deaths)),digits=2))

# Calculate the US totals
#us_agg_data <- cdcAnnualData %>% group_by(Year)  %>% summarise(Deaths=sum(Deaths), Population = #sum(Population),Crude.Rate = format((sum(Deaths) / sum(Population)) * 100000,  big_mark = "," ,digits = 2))
#us_agg_data ["State"] <- "US Aggregated"
#cdcAnnualData <- cdcAnnualData %>% rbind(us_agg_data)

```
## Load Geo data

Latitudeo and Longitude  Coordinates for each state were downloaded from https://inkplant.com/code/state-latitudes-longitudes


```{r loadGeoData, echo=FALSE}
state.geom <- read.csv("data/state_geom.csv")
state.geom$name = as.character(state.geom$name)
saveRDS(state.geom,"data/state_geom.rds")
saveRDS(state.geom,"shiny/data/state_geom.rds")

# Join geocoordinates with CDC data

cdcAnnualData <- left_join(cdcAnnualData,state.geom,by=c('State'='name'))
cdcAnnualData <- cdcAnnualData %>% select(State,Year,Deaths,Population,PercentChange,Crude.Rate,Age.Adjusted.Rate,Lat=latitude,Long=longitude)
saveRDS(cdcAnnualData,"data/cdcAnnualData.rds")
saveRDS(cdcAnnualData,"shiny/data/cdcAnnualData.rds")

cdcMonthlyStateData <- left_join(cdcMonthlyStateData,state.geom,by=c('State'='name'))
cdcMonthlyStateData <- cdcMonthlyStateData %>% select(State,YearMonth,Year,Month.Code,Deaths,Lat=latitude,Long=longitude)



```



### Exploration 




```{r exploreMonthly, echo=FALSE}
str(cdcMonthlyStateData)
head(cdcMonthlyStateData)

summary(cdcMonthlyStateData)
print("Deaths reported in 2016 ")
print(cdcMonthlyStateData %>% filter(Year=="2016") %>% select(Deaths) %>%sum)
print("Deaths reported in the state of Texas 2016 ")
kable( cdcMonthlyStateData %>% filter(Year=="2016",State=="Texas") %>% select(Deaths) %>%sum)

```

```{r plottexas, echo=FALSE}
grp_state <- cdcMonthlyStateData %>% group_by(State,YearMonth)%>% select(State,YearMonth,Deaths)

tx <- filter(grp_state,State=="Texas")
head(tx)


ggplot(tx,aes(x=YearMonth, y=Deaths))+
  geom_point()  +
  ggtitle("Deaths attributed to opiods for the state of Texas","1999-2016")

ggplot(tx,aes(x=YearMonth, y=Deaths,group=1))+
  geom_line(stat="identity") +
  stat_smooth(method='loess') +
  ggtitle("Deaths attributed to opiods for the state of Texas","1999-2016")


```


```{r plottus, echo=FALSE}
grp_us <- cdcAnnualData %>% group_by(Year)%>% select(Year,Deaths) %>%summarise(Deaths= sum(Deaths))
grp_us2 <- grp_us %>% arrange(desc(Year))  %>% mutate (PercentChange =  (Deaths-lead(Deaths)) * 100 / lead(Deaths))

kable(grp_us2 %>% arrange(desc(Year)),format.args = list(decimal.mark = ".", big.mark = ","),caption="Deaths attributed to opiods in the United State")

ggplot(grp_us,aes(x=Year, y=Deaths))+
  geom_point()  +
  ggtitle("Deaths attributed to opiods in the United States","1999-2016")

ggplot(grp_us,aes(x=Year, y=Deaths,group=1))+
  geom_line() +
  stat_smooth(method="glm") +
  ggtitle("Deaths attributed to opiods in the United States","1999-2016")



```

```{r plotmap, echo=FALSE}

grp_state2 <-cdcAnnualData %>% filter(Year=="2016") %>%group_by(State)%>% summarise(Deaths= sum(Deaths)) %>% arrange(desc(Deaths))

kable(head(grp_state2),format.args = list(decimal.mark = " ", big.mark = ","),caption="States with Highest number of deaths for 2016")
kable(tail(grp_state2),format.args = list(decimal.mark = " ", big.mark = ","),caption="States with Lowest number of deaths for 2016")
grp_state2$State <- tolower(grp_state2$State)
p <- ggplot(grp_state2, aes(map_id = State)) + 
  # map points to the fifty_states shape data
  scale_fill_gradient(low="white", high="red")+
# Diverging color scheme
  geom_map(aes(fill = log10(Deaths)), map = fifty_states) +
  
  expand_limits(x = fifty_states$long, y = fifty_states$lat) +
  coord_map() +
  scale_x_continuous(breaks = NULL) + 
  scale_y_continuous(breaks = NULL) +
  labs(x = "", y = "") +
  theme(legend.position = "bottom", 
        panel.background = element_blank())


p + fifty_states_inset_boxes() 

```



``` {r}
sessionInfo(package = NULL)
```