NTL Study Group Programming Task #1 - Health Data analysis
Data
Data can be obtained from this link
Data is downloaded and read from local harddisk
We start by loading needed libraries and reading the data
library(dplyr)
library(tidyr)
library(ggplot2)
library(knitr)
NHCS_data <- read.csv("NCHS.csv")
dim(NHCS_data)## [1] 15028 6NHCS data file has 6 columns & 15028 rows
We start by checking how our data is structured
str(NHCS_data)## 'data.frame': 15028 obs. of 6 variables:
## $ Year : int 1999 1999 1999 1999 1999 1999 1999 1999 1999 1999 ...
## $ X113.Cause.Name : Factor w/ 17 levels "Accidents (unintentional injuries) (V01-X59,Y85-Y86)",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ Cause.Name : Factor w/ 17 levels "All Causes","Alzheimer's disease",..: 17 17 17 17 17 17 17 17 17 17 ...
## $ State : Factor w/ 52 levels "Alabama","Alaska",..: 1 2 3 4 5 6 7 8 9 10 ...
## $ Deaths : int 2313 294 2214 1287 9198 1519 1034 267 161 5961 ...
## $ Age.adjusted.Death.Rate: num 52.2 55.9 44.8 47.6 28.7 39 29.3 35.3 28.4 35.7 ...summary(NHCS_data)## Year
## Min. :1999
## 1st Qu.:2003
## Median :2007
## Mean :2007
## 3rd Qu.:2011
## Max. :2015
##
## X113.Cause.Name
## Accidents (unintentional injuries) (V01-X59,Y85-Y86): 884
## All Causes : 884
## Alzheimer's disease (G30) : 884
## Assault (homicide) (*U01-*U02,X85-Y09,Y87.1) : 884
## Cerebrovascular diseases (I60-I69) : 884
## Chronic liver disease and cirrhosis (K70,K73-K74) : 884
## (Other) :9724
## Cause.Name State
## All Causes : 884 Alabama : 289
## Alzheimer's disease : 884 Alaska : 289
## Cancer : 884 Arizona : 289
## Chronic liver disease and cirrhosis: 884 Arkansas : 289
## CLRD : 884 California: 289
## Diabetes : 884 Colorado : 289
## (Other) :9724 (Other) :13294
## Deaths Age.adjusted.Death.Rate
## Min. : 10 Min. : 1.30
## 1st Qu.: 294 1st Qu.: 8.30
## Median : 838 Median : 18.90
## Mean : 10233 Mean : 86.53
## 3rd Qu.: 2737 3rd Qu.: 46.30
## Max. :2712630 Max. :1087.30
## NA's :15 NA's :111Objectives
After checking structure of our data, we will use NHCS dataser to answer the few questions we have in hand ..
1. What were the causes of death in this dataset?
unique(NHCS_data$Cause.Name)## [1] Unintentional Injuries
## [2] All Causes
## [3] Alzheimer's disease
## [4] Homicide
## [5] Stroke
## [6] Chronic liver disease and cirrhosis
## [7] CLRD
## [8] Diabetes
## [9] Diseases of Heart
## [10] Essential hypertension and hypertensive renal disease
## [11] Influenza and pneumonia
## [12] Cancer
## [13] Suicide
## [14] Kidney Disease
## [15] Parkinson's disease
## [16] Pneumonitis due to solids and liquids
## [17] Septicemia
## 17 Levels: All Causes Alzheimer's disease ... Unintentional Injuries2. What was the total number of deaths in the united states from 1999 to 2015?
# we notice there is a cause named 'All Causes' .. hence we count in 2 ways:
# get total death by filtering on 'All Causes' ,
# get total death for all records except 'All Causes'
sum(filter(NHCS_data, NHCS_data$Cause.Name == "All Causes")$Deaths, na.rm = TRUE) ## [1] 84341636sum(filter(NHCS_data, NHCS_data$Cause.Name != "All Causes")$Deaths, na.rm = TRUE)## [1] 69280514# check if states coulmn has all states
unique(NHCS_data$State)## [1] Alabama Alaska Arizona
## [4] Arkansas California Colorado
## [7] Connecticut Delaware District of Columbia
## [10] Florida Georgia Hawaii
## [13] Idaho Illinois Indiana
## [16] Iowa Kansas Kentucky
## [19] Louisiana Maine Maryland
## [22] Massachusetts Michigan Minnesota
## [25] Mississippi Missouri Montana
## [28] Nebraska Nevada New Hampshire
## [31] New Jersey New Mexico New York
## [34] North Carolina North Dakota Ohio
## [37] Oklahoma Oregon Pennsylvania
## [40] Rhode Island South Carolina South Dakota
## [43] Tennessee Texas United States
## [46] Utah Vermont Virginia
## [49] Washington West Virginia Wisconsin
## [52] Wyoming
## 52 Levels: Alabama Alaska Arizona Arkansas California ... Wyoming# # we notice we have 'United States' listed among states, hence we count death in all 4 cases:
# get total death by filtering on 'All Causes' & 'United States',
# get total death by filtering on 'All Causes' & individual States,
# get total death by filtering on individual Causes' & 'United States',
# get total death by filtering on individual Causes & individual States
death_count <- c(
sum(filter(NHCS_data, NHCS_data$Cause.Name == "All Causes" , NHCS_data$State =="United States")$Deaths, na.rm = TRUE),
sum(filter(NHCS_data, NHCS_data$Cause.Name == "All Causes" , NHCS_data$State !="United States")$Deaths, na.rm = TRUE),
sum(filter(NHCS_data, NHCS_data$Cause.Name != "All Causes" , NHCS_data$State =="United States")$Deaths, na.rm = TRUE),
sum(filter(NHCS_data, NHCS_data$Cause.Name != "All Causes" , NHCS_data$State !="United States")$Deaths, na.rm = TRUE)
)
names(death_count) <- c('All Causes & United States','All Causes & States','Causes & United States','Causes & States')
death_count## All Causes & United States All Causes & States
## 42170818 42170818
## Causes & United States Causes & States
## 34640315 346401993. What is the number of deaths per year from 1999 to 2015?
# Sum Death per Year from 1999 to 2015
death_per_year <- filter(NHCS_data, NHCS_data$Cause.Name == "All Causes" , NHCS_data$State =="United States")
death_per_year <- aggregate(death_per_year['Deaths'], by=death_per_year['Year'], sum,na.rm=TRUE, na.action=NULL)
# Display Sum of Deaths per year from 1999 to 2015
ggplot(death_per_year, aes(Year, Deaths)) + geom_line()+labs(title = "Sum of Deaths per Year in the USA from 1999 to 2015")
4. Which states had the highest number of deaths overall?
# Sum Death per Year from 1999 to 2015
death_per_states <- filter(NHCS_data, NHCS_data$Cause.Name == "All Causes" , NHCS_data$State !="United States")
death_per_states <- aggregate(death_per_states['Deaths'], by=death_per_states['State'], sum)
# Display top 10 states with highest deaths
ggplot(death_per_states[
order( -death_per_states[,2]),
][1:10,], aes(State, Deaths)) + geom_bar(stat = "identity", fill="blue", alpha=0.5)+labs(title = "Top 10 Deaths per State in the USA from 1999 to 2015") + theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0))
5. What were top causes of death in the USA in this period?
# Sum Death per Cause from 1999 to 2015
death_per_cause <- filter(NHCS_data, NHCS_data$Cause.Name != "All Causes" , NHCS_data$State =="United States")
death_per_cause <- aggregate(death_per_cause['Deaths'], by=death_per_cause['Cause.Name'], sum,na.rm=TRUE, na.action=NULL)
death_per_cause <- death_per_cause[order( -death_per_cause[,2]),][1:10,]
# Display top 10 states with highest deaths
ggplot(death_per_cause, aes(Cause.Name, Deaths)) + geom_bar(stat = "identity", fill="blue", alpha=0.5)+labs(title = "Top 10 Deaths per Cause in the USA from 1999 to 2015") + theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0))