Assignment Week 5

library(dplyr)
library(tidyr)
library(tidyverse)
library(reshape2)
library(readxl)
library(httr)

Importing the Data

cnames <- read_csv("https://raw.githubusercontent.com/cassie-boylan/DATA-607/main/israeli_vax_data.csv", n_max=0,show_col_types=FALSE) %>% names()

vax_info <- read_csv("https://raw.githubusercontent.com/cassie-boylan/DATA-607/main/israeli_vax_data.csv", col_types = "cnnnnl", na = c(""," ","NA"), skip = 1, )

Introducting the Data

The dataset used in this exercise is a tabular count of the population fully vaccinated against covid-19 and not vaccinated by age group below and above 50 years of age. Severe cases by 100K of the population are also included in this dataset. I will be tidying the data before calculating total population and efficacy of the vaccine by comparing percentage of severe case frequency among vaccinated and non-vaccinated age groups.

Renaming the Columns

cnames <- cnames[c(-3,-5)]
cnames

## [1] "Age"          "Population %" "Severe Cases" "Efficacy"

names(vax_info)[1] <- cnames[1]
names(vax_info)[2] <- "not_vaxed"
names(vax_info)[3] <- "fully_vaxed"
names(vax_info)[6] <- cnames[4]

Replacing NA values

vax_info <- vax_info %>% fill(Age)

Subsetting, Melting & Establishing New Variable Columns

vaxed <- select(vax_info, Age, fully_vaxed, not_vaxed)

vaxed <- melt(vaxed, id="Age", variable.name = "vax_status", value.name = "total")

percent = subset(vaxed, vaxed$total < 100) %>%
  transmute(Age,
            vax_status,
            percent = total/100)

population = subset(vaxed, vaxed$total > 100)

Joining Tidied Subsets Back Together

vaxed <- inner_join(percent,population, by=c("Age","vax_status"))

names(vaxed)[names(vaxed) == "percent"]<- cnames[2]
names(vaxed)[names(vaxed) == "total"]<- "Population"

severe_cases <- select(vax_info,Age,4:5) %>% group_by(Age)

severe_cases <- melt(severe_cases, na.rm = TRUE, id="Age",variable.name = "vax_status", value.name = "per_100K")

severe_cases <- severe_cases %>%
  transmute(Age,
            vax_status = ifelse(
                            grepl("Not",vax_status),
                        "not_vaxed","fully_vaxed"),
          per_100K)

final_df <- inner_join(vaxed,severe_cases,by=c("Age","vax_status"))

Determining Proportion of Population Partially Vaxed

There are 3 population subsets being observed - those who are not vaccinated, those who are fully vaccinated and the small percentage that have only received 1 dose.

partial_vax <- final_df %>% group_by(Age) %>%
  summarize(
    partial_vax_perc = 1 - (sum(`Population %`))
  )

Calculating Total Population

final_df <- final_df %>% group_by(Age,vax_status) %>%
  mutate(
    Total_Population = prettyNum(Population / `Population %`,big.mark = ",")
  )

Calculating Efficacy

I am subsetting to view the severe cases in which hospitalization has occurred. For this analysis, I am creating a wide structure to better separate those cases per the populations of vaccinated and unvaccinated.

final_df_severe <- final_df %>% select(Age, vax_status,per_100K) %>% pivot_wider(names_from=vax_status, values_from=per_100K)

We can observe the efficacy rate is 97.44% among the population younger than 50 years of age.

The efficacy rate is 83.04% among the population greater than 50 years of age.

This indicates that of those who received the vaccine, a 97% and 83% reduction in hospitalization was observed over the population that did not get vaccinated.

final_df_severe %>% mutate(
  Efficacy = round(1 - ((fully_vaxed/100000) / (not_vaxed/10000)),4)
  )

## # A tibble: 2 x 4
## # Groups:   Age [2]
##   Age   fully_vaxed not_vaxed Efficacy
##   <chr>       <dbl>     <dbl>    <dbl>
## 1 <50            11        43    0.974
## 2 >50           290       171    0.830

Comparing Severe Case Rate Among Vaccinated & Unvaccinated

Here we can see that out of the population who have been fully vaccinated, there are only 5 cases per 100k. Out of the population that have not been vaccinated, the severe case rate triples to 16 cases per 100k.

rate_df <- final_df %>%
  select(vax_status, Population, per_100K) %>% 
  group_by(vax_status) %>%
 summarize(severe_total = sum(per_100K),
         pop_total = sum(Population)/100000,
         rate = severe_total/pop_total,
         .groups = "drop")