Final Project 607
Andreina Arias
2024-12-10
Introduction
In 2020 Covid-19 seemed to be one of the leading causes of death and it wanted to see if there was a difference in the leading causes of death for NYC and the United States of America (USA), to see if the concerning diseases in NYC were also main concerns to the whole USA. Is there a difference in death rate for NYC and the USA, since is NYC part of the USA. I used data from NYC open data on the leading causes of death in NY since 2007 “https://data.cityofnewyork.us/Health/New-York-City-Leading-Causes-of-Death/jb7j-dtam/data_preview”, which I loaded into GitHub to obtain the raw data of the CSV to us in R. For the data on the leading cause of death in the USA I’ll use data from World Health Organization (WHO) “https://www.who.int/data/gho/data/themes/mortality-and-global-health-estimates/ghe-leading-causes-of-death”, which I will using API (using jsonlite) to extract the data from the data link”https://xmart-api-public.who.int/DEX_CMS/GHE_FULL?$filter=FLAG_RANKABLE%20eq%201%20and%20DIM_COUNTRY_CODE%20eq%20%27USA%27%20and%20DIM_SEX_CODE%20eq%20%27MALE%27%20and%20DIM_AGEGROUP_CODE%20eq%20%27TOTAL%27%20and%20DIM_YEAR_CODE%20eq%202020&$orderBy=VAL_DTHS_RATE100K_NUMERIC%20desc&$select=DIM_GHECAUSE_CODE,DIM_GHECAUSE_TITLE,VAL_DALY_RATE100K_NUMERIC,VAL_DTHS_RATE100K_NUMERIC&$top=10“. This observational study can be helpful seeing which interventions NYC would need, such as in 2020 NYC was black sheep state that everyone wanted to avoid.
NYC leading cause of death data from NYC open data was uploaded into github to load the csv file into R for analysis.
DF1<-read.csv('https://raw.githubusercontent.com/Andreina-A/Project2_Data607/refs/heads/main/New_York_City_Leading_Causes_of_Death_20241013.csv', na.strings=" ")# na.string will convert all empty data into '.'
#Rename to reflect NYC data when joined with USA data and select the data needed
DF1<-DF1 %>% rename("Leading_cause"=Leading.Cause, "Death_rate_NYC"=Death.Rate)%>%
filter(Year == 2020)%>% #Only looking at data from 2020
select(Year, Leading_cause, Death_rate_NYC)
#Renamed Variables in leading cause of death to Match leading cause in USA data, also renamed diseases with long text
DF1 <- DF1 %>%
mutate(Leading_cause = recode(Leading_cause,
"Diseases of Heart (I00-I09, I11, I13, I20-I51)"="All heart diseases", "Alzheimer's Disease (G30)"="Alzheimer disease and other dementias", "Cerebrovascular Disease (Stroke: I60-I69)"="Stroke","Covid-19"= "COVID-19", "Chronic Lower Respiratory Diseases (J40-J47)"="Chronic obstructive pulmonary disease","Malignant Neoplasms (Cancer: C00-C97)"="Trachea, bronchus, lung cancers",
"Accidents Except Drug Posioning (V01-X39, X43, X45-X59, Y85-Y86)"="Accident", "Mental and Behavioral Disorders due to Accidental Poisoning and Other Psychoactive Substance Use (F11-F16, F18-F19, X40-X42, X44)"="Drug use disorders", "Intentional Self-Harm (Suicide: U03, X60-X84, Y87.0)"="Self-harm"))
DF1$Death_rate_NYC<-as.numeric(DF1$Death_rate_NYC) #made values in death numerical for further analysis
DF1 <- na.omit(DF1) #removed rows with missing values.
#view data
head(DF1)## Year Leading_cause Death_rate_NYC
## 1 2020 COVID-19 334.1
## 2 2020 All heart diseases 160.2
## 3 2020 Trachea, bronchus, lung cancers 88.7
## 4 2020 Drug use disorders 45.6
## 5 2020 Diabetes Mellitus (E10-E14) 22.6
## 6 2020 Influenza (Flu) and Pneumonia (J09-J18) 22.5Used aggregate function to group all the leading causes of death in NYC and sum their death rate, also created a column for source to differentiate data from the USA and NYC to use in facet wrap (visual).
DF1 <- aggregate(Death_rate_NYC ~ Leading_cause, data = DF1, sum)%>%
mutate(Source = "NYC") #Source created to use in facet wrap function
head(DF1)## Leading_cause
## 1 Accidents Except Drug Poisoning (V01-X39, X43, X45-X59, Y85-Y86)
## 2 All heart diseases
## 3 All Other Causes
## 4 Alzheimer disease and other dementias
## 5 Assault (Homicide: U01-U02, Y87.1, X85-Y09)
## 6 Chronic Liver Disease and Cirrhosis (K70, K73-K74)
## Death_rate_NYC Source
## 1 79.6 NYC
## 2 1848.1 NYC
## 3 1237.8 NYC
## 4 66.5 NYC
## 5 32.3 NYC
## 6 16.2 NYCLoaded data on the USA from WHO by connecting to API URL, to obtain json file.
# Load API URL
url <- "https://xmart-api-public.who.int/DEX_CMS/GHE_FULL?$filter=FLAG_RANKABLE%20eq%201%20and%20DIM_COUNTRY_CODE%20eq%20%27USA%27%20and%20DIM_SEX_CODE%20eq%20%27MALE%27%20and%20DIM_AGEGROUP_CODE%20eq%20%27TOTAL%27%20and%20DIM_YEAR_CODE%20eq%202020&$orderBy=VAL_DTHS_RATE100K_NUMERIC%20desc&$select=DIM_GHECAUSE_CODE,DIM_GHECAUSE_TITLE,VAL_DALY_RATE100K_NUMERIC,VAL_DTHS_RATE100K_NUMERIC&$top=10"
# Use GET function to request to the API
response <- GET(url)
# Check if the request was successful (HTTP status 200)
if (status_code(response) == 200) {
# obtain the JSON content
data <- fromJSON(content(response, "text"))
# Checked structure to confirm it is a data frame
str(data)
# Extracted and viewed the relevant part of the data (e.g., death statistics)
DF2<- data$value # Assuming 'value' holds the desired data
head(DF2)
} else {
print("Request failed with status code: ")
print(status_code(response))
}## List of 2
## $ @odata.context: chr "https://xmart-api-public.who.int/DEX_CMS/$metadata#GHE_FULL"
## $ value :'data.frame': 10 obs. of 4 variables:
## ..$ DIM_GHECAUSE_CODE : int [1:10] 1130 395 950 1180 680 1140 870 1270 1120 1610
## ..$ DIM_GHECAUSE_TITLE : chr [1:10] "Ischaemic heart disease" "COVID-19" "Alzheimer disease and other dementias" "Chronic obstructive pulmonary disease" ...
## ..$ VAL_DALY_RATE100K_NUMERIC: num [1:10] 3529 2462 888 1272 1001 ...
## ..$ VAL_DTHS_RATE100K_NUMERIC: num [1:10] 167.2 115.6 57.9 50.1 44.3 ...## DIM_GHECAUSE_CODE DIM_GHECAUSE_TITLE
## 1 1130 Ischaemic heart disease
## 2 395 COVID-19
## 3 950 Alzheimer disease and other dementias
## 4 1180 Chronic obstructive pulmonary disease
## 5 680 Trachea, bronchus, lung cancers
## 6 1140 Stroke
## VAL_DALY_RATE100K_NUMERIC VAL_DTHS_RATE100K_NUMERIC
## 1 3528.65 167.20
## 2 2461.82 115.56
## 3 887.57 57.90
## 4 1272.04 50.11
## 5 1001.19 44.26
## 6 1058.35 43.09Cleaned the WHO data by renaming the colums to merge with the NYC data, created a source column for use in the facet wrap function.
# Renamed and Extracted relevant columns from WHO data
DF2 <- DF2 %>% rename("Leading_cause"=DIM_GHECAUSE_TITLE, "Death_rate_USA"=VAL_DTHS_RATE100K_NUMERIC)%>%
select(Leading_cause, Death_rate_USA)%>%
mutate(Source = "USA") #Source created to use in facet wrap function
#View USA data
head(DF2)## Leading_cause Death_rate_USA Source
## 1 Ischaemic heart disease 167.20 USA
## 2 COVID-19 115.56 USA
## 3 Alzheimer disease and other dementias 57.90 USA
## 4 Chronic obstructive pulmonary disease 50.11 USA
## 5 Trachea, bronchus, lung cancers 44.26 USA
## 6 Stroke 43.09 USAMerged NYC leading cause of death data with USA leading cause
of death. Since the USA data only had 10 diseases, there will be NA
values wherever USA didn’t have a disease listed compared to NYC which
more than 10 diseases listed, I removed NA values to only compare the
leading causes.
Merged_data <- left_join(DF1, DF2, by = "Leading_cause")# Merged WHO and NYC data
Merged_data <- bind_rows(DF1, DF2) #merge by rows to cause a NA for th
# View merged data
Merged_data## Leading_cause
## 1 Accidents Except Drug Poisoning (V01-X39, X43, X45-X59, Y85-Y86)
## 2 All heart diseases
## 3 All Other Causes
## 4 Alzheimer disease and other dementias
## 5 Assault (Homicide: U01-U02, Y87.1, X85-Y09)
## 6 Chronic Liver Disease and Cirrhosis (K70, K73-K74)
## 7 Chronic obstructive pulmonary disease
## 8 COVID-19
## 9 Diabetes Mellitus (E10-E14)
## 10 Drug use disorders
## 11 Essential Hypertension and Renal Diseases (I10, I12, I15)
## 12 Influenza (Flu) and Pneumonia (J09-J18)
## 13 Self-harm
## 14 Stroke
## 15 Trachea, bronchus, lung cancers
## 16 Ischaemic heart disease
## 17 COVID-19
## 18 Alzheimer disease and other dementias
## 19 Chronic obstructive pulmonary disease
## 20 Trachea, bronchus, lung cancers
## 21 Stroke
## 22 Drug use disorders
## 23 Kidney diseases
## 24 Hypertensive heart disease
## 25 Self-harm
## Death_rate_NYC Source Death_rate_USA
## 1 79.6 NYC NA
## 2 1848.1 NYC NA
## 3 1237.8 NYC NA
## 4 66.5 NYC NA
## 5 32.3 NYC NA
## 6 16.2 NYC NA
## 7 125.8 NYC NA
## 8 1889.3 NYC NA
## 9 203.4 NYC NA
## 10 167.4 NYC NA
## 11 130.9 NYC NA
## 12 182.5 NYC NA
## 13 8.8 NYC NA
## 14 196.0 NYC NA
## 15 1031.6 NYC NA
## 16 NA USA 167.20
## 17 NA USA 115.56
## 18 NA USA 57.90
## 19 NA USA 50.11
## 20 NA USA 44.26
## 21 NA USA 43.09
## 22 NA USA 39.59
## 23 NA USA 31.10
## 24 NA USA 27.86
## 25 NA USA 23.59Comparison
Used select function to view the comparsion on the leading cause of death for the USA and NYC
# View of the comparison
Merged_data %>%
select(Leading_cause, Death_rate_NYC, Death_rate_USA, Source) %>%
arrange(desc(Death_rate_USA),desc(Death_rate_NYC)) #arranged descending order for the USA and NYC## Leading_cause
## 1 Ischaemic heart disease
## 2 COVID-19
## 3 Alzheimer disease and other dementias
## 4 Chronic obstructive pulmonary disease
## 5 Trachea, bronchus, lung cancers
## 6 Stroke
## 7 Drug use disorders
## 8 Kidney diseases
## 9 Hypertensive heart disease
## 10 Self-harm
## 11 COVID-19
## 12 All heart diseases
## 13 All Other Causes
## 14 Trachea, bronchus, lung cancers
## 15 Diabetes Mellitus (E10-E14)
## 16 Stroke
## 17 Influenza (Flu) and Pneumonia (J09-J18)
## 18 Drug use disorders
## 19 Essential Hypertension and Renal Diseases (I10, I12, I15)
## 20 Chronic obstructive pulmonary disease
## 21 Accidents Except Drug Poisoning (V01-X39, X43, X45-X59, Y85-Y86)
## 22 Alzheimer disease and other dementias
## 23 Assault (Homicide: U01-U02, Y87.1, X85-Y09)
## 24 Chronic Liver Disease and Cirrhosis (K70, K73-K74)
## 25 Self-harm
## Death_rate_NYC Death_rate_USA Source
## 1 NA 167.20 USA
## 2 NA 115.56 USA
## 3 NA 57.90 USA
## 4 NA 50.11 USA
## 5 NA 44.26 USA
## 6 NA 43.09 USA
## 7 NA 39.59 USA
## 8 NA 31.10 USA
## 9 NA 27.86 USA
## 10 NA 23.59 USA
## 11 1889.3 NA NYC
## 12 1848.1 NA NYC
## 13 1237.8 NA NYC
## 14 1031.6 NA NYC
## 15 203.4 NA NYC
## 16 196.0 NA NYC
## 17 182.5 NA NYC
## 18 167.4 NA NYC
## 19 130.9 NA NYC
## 20 125.8 NA NYC
## 21 79.6 NA NYC
## 22 66.5 NA NYC
## 23 32.3 NA NYC
## 24 16.2 NA NYC
## 25 8.8 NA NYCVisual
Used facet wrap function to make a visual comparsion of the leading causes of death of NYC and the USA.
ggplot(Merged_data, aes(x = reorder(Leading_cause, -Death_rate_NYC),
y = ifelse(Source == "NYC", Death_rate_NYC, Death_rate_USA),
fill = Source)) +
geom_bar(stat = "identity", position = "dodge") +
facet_wrap(~ Source, scales = "free_y") + #free_y to have height that will be proportional to the length of the y scale
labs(title = "Comparison of Leading Causes of Death: NYC vs USA (2020)",
x = "Cause of Death",
y = "Death Rate (per 100,000)",
fill = "Location") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1)) + #rotated labels on x-axis
scale_fill_manual(values = c("NYC" = "skyblue", "USA" = "orange"))
t_test_result <- t.test(Merged_data$Death_rate_NYC, Merged_data$Death_Rate_USA, alternative = "two.sided")
t_test_result##
## One Sample t-test
##
## data: Merged_data$Death_rate_NYC
## t = 2.7784, df = 14, p-value = 0.0148
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 109.7122 852.4478
## sample estimates:
## mean of x
## 481.08Null Hypothesis: No significance difference in the death rate for NYC compared to the USA Alternative Hypothesis: There is a difference in the death rates for NYC compared to the USA. Based on the t- test the P value is 0.01 which is less than .05 meaning there is a signicant difference in the death rate for NYC and the USA.
Conclusion
The comparison on the leading cause in 2020 within NYC and the USA were visually demonstrated, using the facet wrap function. The primary leading cause of death in the USA was heart disease and Covid-19 came in second, while in NYC the primary leading cause of death was Covid-19 and heart disease came in second. I wasn’t surprised to see that NYC’s primary cause of death being Covid-19 in 2020 because during the pandemic everyone was afraid to live in NYC and many moved to different states within the USA, Rhode island didn’t even want visitors from NYC. Using the p value I was able to come with the conclusion that there is a significant difference in the death rate for the USA and NYC. Although I noticed there was a difference in death rate value for NYC and the USA, NYC seemed to have high values this could be because of the different denominator being used to calculate the death rate; NYC could have been calculated from the city-level while the USA was calculated based on the entire United Stated including NYC which would use a different way to calculate the death rate compare to NYC. Overall I would say it’s important to see that main leading cause of death is heart disease in NYC and the USA and I feel it’s something we study consider to find presentation. If I were to do further assessment I would compare the leading cause of death of the USA to the world and see if there is a country with a lower death rate for heart disease and learn about their lifestyle to possibly decrease the death in the USA by heart disease.
Reference
Data source for NYC- New York City Department of Health and Mental Hygiene. (n.d.). New York City leading causes of death. NYC Open Data. Retrieved December 10, 2024, from https://data.cityofnewyork.us/Health/New-York-City-Leading-Causes-of-Death/jb7j-dtam/data_preview
Data source for USA- World Health Organization. (n.d.). Leading causes of death. Global Health Estimates. World Health Organization. Retrieved December 10, 2024, from https://www.who.int/data/gho/data/themes/mortality-and-global-health-estimates/ghe-leading-causes-of-death.