Final Project

Introduction

This data set includes tables on persons living with HIV/AIDS, newly diagnosed HIV cases, and all-cause deaths in HIV/AIDS cases by gender, age, race/ethnicity, and transmission category. A data frame with 6005 rows and 18 variables. It contains both quantitative and categorical values. The data set here records the reported cases of AIDS diagnosed from 2011 and until 2015. I choose to explore this dataset because I want to have a better insight into HIV/AIDS.

Metadata List

• Year: The year of the diagnosis • Borough: Cities • UHF: the United States hospital fund neighborhood • Gender: Female, male, transgender • Age: the age group of the people the column is describing • Race: Black, White, Latino/Hispanic, Asian/Pacific Island, Other/Unknown • HIV diagnoses: Number of people infected by HIV • HIV diagnosis rate: the HIV diagnosis rate out of 100,000 • Concurrent diagnoses • Percent linked to care within 3 months: • AIDS diagnoses: Number of people infected by AIDS • AIDS diagnosis rate: the rate of AIDS diagnoses out of 100,000 • PLWDHI prevalence: the prevalence of People Living with Diagnosed HIV Infection • Percent viral suppression of the: Number of people dies by HIV or AIDS • Death rate: the rate of deaths out of 100,000 • HIV-related death rate: HIV-related to deaths out of 100,000 • Non-HIV related death rate, the rate of non-HIV related death out of 100,000.

Import the HIV_AIDS_NY Dataset

library(readr)
HIV_AIDS_NY <- read_csv("C:/Users/Mitcheyla$/Desktop/DATA110 -VISUALISATION/HIV_AIDS_NY.csv")

## Rows: 6005 Columns: 18
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr  (5): Borough, UHF, Gender, Age, Race
## dbl (13): Year, HIV diagnoses, HIV diagnosis rate, Concurrent diagnoses, % l...
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

View(HIV_AIDS_NY)

Load Library

library(tidyverse)

## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
## ✔ ggplot2 3.4.0      ✔ dplyr   1.0.10
## ✔ tibble  3.1.8      ✔ stringr 1.5.0 
## ✔ tidyr   1.2.1      ✔ forcats 0.5.2 
## ✔ purrr   0.3.5

## Warning: package 'dplyr' was built under R version 4.2.2

## Warning: package 'stringr' was built under R version 4.2.2

## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()

Checking the dataset

str(HIV_AIDS_NY)

## spc_tbl_ [6,005 × 18] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
##  $ Year                            : num [1:6005] 2011 2011 2011 2011 2011 ...
##  $ Borough                         : chr [1:6005] "All" "All" "All" "All" ...
##  $ UHF                             : chr [1:6005] "All" "All" "All" "All" ...
##  $ Gender                          : chr [1:6005] "All" "Male" "Female" "Transgender" ...
##  $ Age                             : chr [1:6005] "All" "All" "All" "All" ...
##  $ Race                            : chr [1:6005] "All" "All" "All" "All" ...
##  $ HIV diagnoses                   : num [1:6005] 3379 2595 733 51 47 ...
##  $ HIV diagnosis rate              : num [1:6005] 48.3 79.1 21.1 99999 13.6 ...
##  $ Concurrent diagnoses            : num [1:6005] 640 480 153 7 4 20 31 50 32 23 ...
##  $ % linked to care within 3 months: num [1:6005] 66 66 66 63 64 67 66 62 72 68 ...
##  $ AIDS diagnoses                  : num [1:6005] 2366 1712 622 32 22 ...
##  $ AIDS diagnosis rate             : num [1:6005] 33.8 52.2 17.6 99999 6.4 ...
##  $ PLWDHI prevalence               : num [1:6005] 1.1 1.7 0.6 99999 0.1 ...
##  $ % viral suppression             : num [1:6005] 71 72 68 55 57 48 61 66 73 81 ...
##  $ Deaths                          : num [1:6005] 2040 1423 605 12 1 ...
##  $ Death rate                      : num [1:6005] 13.6 13.4 14 11.1 1.4 7.2 9.4 15.9 24.1 33.5 ...
##  $ HIV-related death rate          : num [1:6005] 5.8 5.7 6 5.7 1.4 3.2 5.7 7.8 11.5 10.6 ...
##  $ Non-HIV-related death rate      : num [1:6005] 7.8 7.7 8 5.4 0 4 3.7 8.1 12.6 22.9 ...
##  - attr(*, "spec")=
##   .. cols(
##   ..   Year = col_double(),
##   ..   Borough = col_character(),
##   ..   UHF = col_character(),
##   ..   Gender = col_character(),
##   ..   Age = col_character(),
##   ..   Race = col_character(),
##   ..   `HIV diagnoses` = col_double(),
##   ..   `HIV diagnosis rate` = col_double(),
##   ..   `Concurrent diagnoses` = col_double(),
##   ..   `% linked to care within 3 months` = col_double(),
##   ..   `AIDS diagnoses` = col_double(),
##   ..   `AIDS diagnosis rate` = col_double(),
##   ..   `PLWDHI prevalence` = col_double(),
##   ..   `% viral suppression` = col_double(),
##   ..   Deaths = col_double(),
##   ..   `Death rate` = col_double(),
##   ..   `HIV-related death rate` = col_double(),
##   ..   `Non-HIV-related death rate` = col_double()
##   .. )
##  - attr(*, "problems")=<externalptr>

Look at the dimension of the dataset

dim(HIV_AIDS_NY)

## [1] 6005   18

Summarize the dataset

summary(HIV_AIDS_NY)

##       Year        Borough              UHF               Gender         
##  Min.   :2011   Length:6005        Length:6005        Length:6005       
##  1st Qu.:2012   Class :character   Class :character   Class :character  
##  Median :2013   Mode  :character   Mode  :character   Mode  :character  
##  Mean   :2013                                                           
##  3rd Qu.:2014                                                           
##  Max.   :2015                                                           
##      Age                Race           HIV diagnoses    HIV diagnosis rate
##  Length:6005        Length:6005        Min.   :   0.0   Min.   :    0.0   
##  Class :character   Class :character   1st Qu.:   0.0   1st Qu.:    0.0   
##  Mode  :character   Mode  :character   Median :   3.0   Median :   18.5   
##                                        Mean   :  26.5   Mean   :  119.5   
##                                        3rd Qu.:  13.0   3rd Qu.:   49.4   
##                                        Max.   :3379.0   Max.   :99999.0   
##  Concurrent diagnoses % linked to care within 3 months AIDS diagnoses   
##  Min.   :  0.000      Min.   :    0                    Min.   :    0.0  
##  1st Qu.:  0.000      1st Qu.:   67                    1st Qu.:    0.0  
##  Median :  1.000      Median :   83                    Median :    2.0  
##  Mean   :  5.095      Mean   :25399                    Mean   :   33.3  
##  3rd Qu.:  3.000      3rd Qu.:99999                    3rd Qu.:    8.0  
##  Max.   :640.000      Max.   :99999                    Max.   :99999.0  
##  AIDS diagnosis rate PLWDHI prevalence % viral suppression     Deaths        
##  Min.   :    0.0     Min.   :    0.0   Min.   :    0       Min.   :    0.00  
##  1st Qu.:    0.0     1st Qu.:    0.2   1st Qu.:   71       1st Qu.:    0.00  
##  Median :   10.4     Median :    0.6   Median :   79       Median :    1.00  
##  Mean   :  122.8     Mean   :  317.5   Mean   : 2656       Mean   :   49.45  
##  3rd Qu.:   30.6     3rd Qu.:    1.5   3rd Qu.:   87       3rd Qu.:    8.00  
##  Max.   :99999.0     Max.   :99999.0   Max.   :99999       Max.   :99999.00  
##    Death rate     HIV-related death rate Non-HIV-related death rate
##  Min.   :  0.00   Min.   :    0.0        Min.   :    0.0           
##  1st Qu.:  0.00   1st Qu.:    0.0        1st Qu.:    0.0           
##  Median :  6.00   Median :    3.0        Median :    5.5           
##  Mean   : 10.34   Mean   :20003.2        Mean   :20005.1           
##  3rd Qu.: 14.10   3rd Qu.:   14.4        3rd Qu.:   22.1           
##  Max.   :263.20   Max.   :99999.0        Max.   :99999.0

Verify if we have any missing data in the dataset

sum(is.na(HIV_AIDS_NY))

## [1] 0

See the first and last 10 rows of the dataset

head(HIV_AIDS_NY,10)

## # A tibble: 10 × 18
##     Year Borough UHF   Gender      Age     Race  HIV d…¹ HIV d…² Concu…³ % lin…⁴
##    <dbl> <chr>   <chr> <chr>       <chr>   <chr>   <dbl>   <dbl>   <dbl>   <dbl>
##  1  2011 All     All   All         All     All      3379    48.3     640      66
##  2  2011 All     All   Male        All     All      2595    79.1     480      66
##  3  2011 All     All   Female      All     All       733    21.1     153      66
##  4  2011 All     All   Transgender All     All        51 99999         7      63
##  5  2011 All     All   Female      13 - 19 All        47    13.6       4      64
##  6  2011 All     All   Female      20 - 29 All       178    24.7      20      67
##  7  2011 All     All   Female      30 - 39 All       176    26.9      31      66
##  8  2011 All     All   Female      40 - 49 All       195    33        50      62
##  9  2011 All     All   Female      50 - 59 All       130    23.5      32      72
## 10  2011 All     All   Female      60+     All        57     6.7      23      68
## # … with 8 more variables: `AIDS diagnoses` <dbl>, `AIDS diagnosis rate` <dbl>,
## #   `PLWDHI prevalence` <dbl>, `% viral suppression` <dbl>, Deaths <dbl>,
## #   `Death rate` <dbl>, `HIV-related death rate` <dbl>,
## #   `Non-HIV-related death rate` <dbl>, and abbreviated variable names
## #   ¹​`HIV diagnoses`, ²​`HIV diagnosis rate`, ³​`Concurrent diagnoses`,
## #   ⁴​`% linked to care within 3 months`

tail(HIV_AIDS_NY, 10)

## # A tibble: 10 × 18
##     Year Borough       UHF    Gender Age   Race  HIV d…¹ HIV d…² Concu…³ % lin…⁴
##    <dbl> <chr>         <chr>  <chr>  <chr> <chr>   <dbl>   <dbl>   <dbl>   <dbl>
##  1  2015 Staten Island Willo… Male   30 -… All         0     0         0   99999
##  2  2015 Staten Island Willo… Male   40 -… All         1    18.8       0     100
##  3  2015 Staten Island Willo… Male   50 -… All         0     0         0   99999
##  4  2015 Staten Island Willo… Male   60+   All         0     0         0   99999
##  5  2015 Staten Island Willo… Male   All   All         4    11.2       0      75
##  6  2015 Staten Island Willo… Male   All   Asia…       0     0         0   99999
##  7  2015 Staten Island Willo… Male   All   Black       1    72.4       0     100
##  8  2015 Staten Island Willo… Male   All   Lati…       2    43.9       0     100
##  9  2015 Staten Island Willo… Male   All   Othe…       1   219.        0       0
## 10  2015 Staten Island Willo… Male   All   White       0     0         0   99999
## # … with 8 more variables: `AIDS diagnoses` <dbl>, `AIDS diagnosis rate` <dbl>,
## #   `PLWDHI prevalence` <dbl>, `% viral suppression` <dbl>, Deaths <dbl>,
## #   `Death rate` <dbl>, `HIV-related death rate` <dbl>,
## #   `Non-HIV-related death rate` <dbl>, and abbreviated variable names
## #   ¹​`HIV diagnoses`, ²​`HIV diagnosis rate`, ³​`Concurrent diagnoses`,
## #   ⁴​`% linked to care within 3 months`

Clean up the dataset set

Load library tidyr

library(tidyr)

#make lowercase, remove space, replace % to percent, and rename variables
names(HIV_AIDS_NY) <- tolower(names(HIV_AIDS_NY))
names(HIV_AIDS_NY) <- gsub(" ","",names(HIV_AIDS_NY))
names(HIV_AIDS_NY) <- gsub("-","",names(HIV_AIDS_NY))
names(HIV_AIDS_NY) <- gsub("%", "percent", names(HIV_AIDS_NY))
names(HIV_AIDS_NY[names(HIV_AIDS_NY) == "hiv-relateddeathrate"] <- "hivrelateddeathrate")

## NULL

names(HIV_AIDS_NY)[names(HIV_AIDS_NY) == "non-hiv-relateddeathrate"] <- "nonhivrelateddeathrate"
str(HIV_AIDS_NY)

## spc_tbl_ [6,005 × 18] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
##  $ year                            : num [1:6005] 2011 2011 2011 2011 2011 ...
##  $ borough                         : chr [1:6005] "All" "All" "All" "All" ...
##  $ uhf                             : chr [1:6005] "All" "All" "All" "All" ...
##  $ gender                          : chr [1:6005] "All" "Male" "Female" "Transgender" ...
##  $ age                             : chr [1:6005] "All" "All" "All" "All" ...
##  $ race                            : chr [1:6005] "All" "All" "All" "All" ...
##  $ hivdiagnoses                    : num [1:6005] 3379 2595 733 51 47 ...
##  $ hivdiagnosisrate                : num [1:6005] 48.3 79.1 21.1 99999 13.6 ...
##  $ concurrentdiagnoses             : num [1:6005] 640 480 153 7 4 20 31 50 32 23 ...
##  $ percentlinkedtocarewithin3months: num [1:6005] 66 66 66 63 64 67 66 62 72 68 ...
##  $ aidsdiagnoses                   : num [1:6005] 2366 1712 622 32 22 ...
##  $ aidsdiagnosisrate               : num [1:6005] 33.8 52.2 17.6 99999 6.4 ...
##  $ plwdhiprevalence                : num [1:6005] 1.1 1.7 0.6 99999 0.1 ...
##  $ percentviralsuppression         : num [1:6005] 71 72 68 55 57 48 61 66 73 81 ...
##  $ deaths                          : num [1:6005] 2040 1423 605 12 1 ...
##  $ deathrate                       : num [1:6005] 13.6 13.4 14 11.1 1.4 7.2 9.4 15.9 24.1 33.5 ...
##  $ hivrelateddeathrate             : num [1:6005] 5.8 5.7 6 5.7 1.4 3.2 5.7 7.8 11.5 10.6 ...
##  $ nonhivrelateddeathrate          : num [1:6005] 7.8 7.7 8 5.4 0 4 3.7 8.1 12.6 22.9 ...
##  - attr(*, "spec")=
##   .. cols(
##   ..   Year = col_double(),
##   ..   Borough = col_character(),
##   ..   UHF = col_character(),
##   ..   Gender = col_character(),
##   ..   Age = col_character(),
##   ..   Race = col_character(),
##   ..   `HIV diagnoses` = col_double(),
##   ..   `HIV diagnosis rate` = col_double(),
##   ..   `Concurrent diagnoses` = col_double(),
##   ..   `% linked to care within 3 months` = col_double(),
##   ..   `AIDS diagnoses` = col_double(),
##   ..   `AIDS diagnosis rate` = col_double(),
##   ..   `PLWDHI prevalence` = col_double(),
##   ..   `% viral suppression` = col_double(),
##   ..   Deaths = col_double(),
##   ..   `Death rate` = col_double(),
##   ..   `HIV-related death rate` = col_double(),
##   ..   `Non-HIV-related death rate` = col_double()
##   .. )
##  - attr(*, "problems")=<externalptr>

List the name of the Variables

ls(HIV_AIDS_NY)

##  [1] "age"                              "aidsdiagnoses"                   
##  [3] "aidsdiagnosisrate"                "borough"                         
##  [5] "concurrentdiagnoses"              "deathrate"                       
##  [7] "deaths"                           "gender"                          
##  [9] "hivdiagnoses"                     "hivdiagnosisrate"                
## [11] "hivrelateddeathrate"              "nonhivrelateddeathrate"          
## [13] "percentlinkedtocarewithin3months" "percentviralsuppression"         
## [15] "plwdhiprevalence"                 "race"                            
## [17] "uhf"                              "year"

#get rid of those weird 99999's
HIV_AIDS_NY <- HIV_AIDS_NY %>%
  filter(hivdiagnoses != 99999.0, 
         hivdiagnosisrate  != 99999.0, 
         percentlinkedtocarewithin3months != 99999.0, 
         aidsdiagnoses!= 99999.0, 
         aidsdiagnosisrate != 99999.0, 
         plwdhiprevalence != 99999.0, 
         deaths != 99999.0,
         deathrate != 99999.0
          )
str(HIV_AIDS_NY)

## spc_tbl_ [4,478 × 18] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
##  $ year                            : num [1:4478] 2011 2011 2011 2011 2011 ...
##  $ borough                         : chr [1:4478] "All" "All" "All" "All" ...
##  $ uhf                             : chr [1:4478] "All" "All" "All" "All" ...
##  $ gender                          : chr [1:4478] "All" "Male" "Female" "Female" ...
##  $ age                             : chr [1:4478] "All" "All" "All" "13 - 19" ...
##  $ race                            : chr [1:4478] "All" "All" "All" "All" ...
##  $ hivdiagnoses                    : num [1:4478] 3379 2595 733 47 178 ...
##  $ hivdiagnosisrate                : num [1:4478] 48.3 79.1 21.1 13.6 24.7 26.9 33 23.5 6.7 2.2 ...
##  $ concurrentdiagnoses             : num [1:4478] 640 480 153 4 20 31 50 32 23 2 ...
##  $ percentlinkedtocarewithin3months: num [1:4478] 66 66 66 64 67 66 62 72 68 91 ...
##  $ aidsdiagnoses                   : num [1:4478] 2366 1712 622 22 96 ...
##  $ aidsdiagnosisrate               : num [1:4478] 33.8 52.2 17.6 6.4 13.3 20.3 35.5 24 7.1 1.6 ...
##  $ plwdhiprevalence                : num [1:4478] 1.1 1.7 0.6 0.1 0.3 0.6 1.4 1.3 0.3 0.1 ...
##  $ percentviralsuppression         : num [1:4478] 71 72 68 57 48 61 66 73 81 77 ...
##  $ deaths                          : num [1:4478] 2040 1423 605 1 19 ...
##  $ deathrate                       : num [1:4478] 13.6 13.4 14 1.4 7.2 9.4 15.9 24.1 33.5 13.1 ...
##  $ hivrelateddeathrate             : num [1:4478] 5.8 5.7 6 1.4 3.2 5.7 7.8 11.5 10.6 2.6 ...
##  $ nonhivrelateddeathrate          : num [1:4478] 7.8 7.7 8 0 4 3.7 8.1 12.6 22.9 10.6 ...
##  - attr(*, "spec")=
##   .. cols(
##   ..   Year = col_double(),
##   ..   Borough = col_character(),
##   ..   UHF = col_character(),
##   ..   Gender = col_character(),
##   ..   Age = col_character(),
##   ..   Race = col_character(),
##   ..   `HIV diagnoses` = col_double(),
##   ..   `HIV diagnosis rate` = col_double(),
##   ..   `Concurrent diagnoses` = col_double(),
##   ..   `% linked to care within 3 months` = col_double(),
##   ..   `AIDS diagnoses` = col_double(),
##   ..   `AIDS diagnosis rate` = col_double(),
##   ..   `PLWDHI prevalence` = col_double(),
##   ..   `% viral suppression` = col_double(),
##   ..   Deaths = col_double(),
##   ..   `Death rate` = col_double(),
##   ..   `HIV-related death rate` = col_double(),
##   ..   `Non-HIV-related death rate` = col_double()
##   .. )
##  - attr(*, "problems")=<externalptr>

Questions to Investigate

1- Are the number of HIV/AIDS and deaths decrease over time? 2- What race has higher deaths rate? 3- Are the number of HIV and AIDS rate among the cities are the same?(Anova test) 4- Is there any correlations between percent viral suppression and HIV related death rate? 5- What race have the highest Death rate, HIV diagnosis rate, and AIDS diagnosis rate?

Look up the Number of People die from HIV/AIDS per year and Create a Visualization

To answer the first question about if the number of diagnosis people with HIV/AIDS decrease over time, I will create a new dataset to only include the relevant variables. I will use select(), group_by(, tidyr::gather(),and summarize functions on the HIV_AIDS_NY dataset to gather HIV diagnoses, AIDS diagnoses, and HIV related deaths into “Id” and their frequencies into “frequency”.

hiv_new_variable <- HIV_AIDS_NY %>%
  select(year,deaths, aidsdiagnoses, hivdiagnoses) %>%
  tidyr::gather("id", "frequency", 2:4) %>%
  group_by(year,id) %>%
  summarize(frequency = sum(frequency))

## `summarise()` has grouped output by 'year'. You can override using the
## `.groups` argument.

View the New Variable

hiv_new_variable

## # A tibble: 15 × 3
## # Groups:   year [5]
##     year id            frequency
##    <dbl> <chr>             <dbl>
##  1  2011 aidsdiagnoses     25797
##  2  2011 deaths            21431
##  3  2011 hivdiagnoses      36657
##  4  2012 aidsdiagnoses     22681
##  5  2012 deaths            19770
##  6  2012 hivdiagnoses      33604
##  7  2013 aidsdiagnoses     20742
##  8  2013 deaths            19109
##  9  2013 hivdiagnoses      30959
## 10  2014 aidsdiagnoses     16110
## 11  2014 deaths            18323
## 12  2014 hivdiagnoses      29980
## 13  2015 aidsdiagnoses     14161
## 14  2015 deaths            17466
## 15  2015 hivdiagnoses      27670

We can see in 2011 more people were die from HIV/AIDS. With time , less people die from those diseases compare to the precedent years.For example, in 2015 the number of deaths were 17466.

Visualisation of HIV diagnoses, AIDS diagnoses, and HIV diagnoses, AIDS diagnoses

library(ggalluvial)

p1 <- hiv_new_variable %>%
  ggplot(aes(x=year, y=frequency, alluvium = id)) +
  geom_alluvium(aes(fill = id, ymin=0, ymax=100000), alpha = .9,
           position="stack", stat="identity", curve_type="sigmoid") +
  xlab("Year") +
  ylab("Frequency") +
  ggtitle("HIV Related Deaths, AIDS Diagnoses, 
          and HIV Diagnoses from 2011-2015") +
  scale_fill_brewer(palette = "Set2") + 
  scale_y_continuous() + 
  theme(axis.text.x = element_text(angle = 0))
p1

As we can see with this plot,in 2011 there were more people with AIDS/HIV and the number of deaths decrease over time.

Boxplot to compare the Death Rate among Race

What race has higher deaths rate?

ggplot(data = HIV_AIDS_NY, aes(x = race, y = deathrate, color = race)) +
  ylab("Frequency)") +
  theme_minimal(base_size = 12) +
  ggtitle("Number of Deaths per Cities") +
  geom_boxplot() +
  scale_color_brewer(palette = 'Set1')

In this visualization the death rate for black is a little higher compare to the other races. Also, all of them have some outliers.

Conduct Two Anova Tests to Compare Variances across the means (or average) of different boroughs to See if the Number of HIV and AIDS infection are the same among them.

Anova Test 1

Are the number of HIV and AIDS rate among the cities are the same?(Anova test)

• Null hypothesis (H0): μa = μbx = μbr = μm = μq =μs • μa : Mean for ALL • μbx: Bronx • μbr: Brooklyn • μm: Manhattan • μq: Queens • μs: Staten Island • Confidence Level 95 % and alpha (0.05)

results <- aov(HIV_AIDS_NY$hivdiagnosisrate ~ HIV_AIDS_NY$borough)
summary(results)

##                       Df   Sum Sq Mean Sq F value Pr(>F)    
## HIV_AIDS_NY$borough    5  1125030  225006   82.68 <2e-16 ***
## Residuals           4472 12170155    2721                   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

P-Value : 2e-16 = 0.00000000000000022 < 0.05 Conclusion: Reject the Null Hypothesis H0. The number of HIV diagnoses are not the same in the cities.

Anova Test 2

Null hypothesis (H0): μa = μbx = μbr = μm = μq =μs μa : Mean for ALL μbx: Bronx μbr: Brooklyn μm: Manhattan μq: Queens μs: Staten Island

Confidence Level 95 % and alpha (0.05)

results1 <- aov(HIV_AIDS_NY$aidsdiagnosisrate ~ HIV_AIDS_NY$borough)
summary(results1)

##                       Df  Sum Sq Mean Sq F value Pr(>F)    
## HIV_AIDS_NY$borough    5  479480   95896   73.46 <2e-16 ***
## Residuals           4472 5837973    1305                   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

P-Value : 2e-16 = 0.00000000000000022 < 0.05 Conclusion: Reject the Null Hypothesis H0. The number of AIDS diagnoses are not the same in the cities.

Visualisation to See the Death Rate Related to HIV among NY Borough

boxplot(data = HIV_AIDS_NY, hivrelateddeathrate ~ borough, outline = F, col = "red")

In this visualization, we can see that Staten Island has the highest death rate compared to other cities. The median for All, Bronx is the same. The median for Brooklyn is slightly different from Bronx and All. Queens is the city that has a lower HIV death rate than the other cities.

Look to See if there is a correlation Between Percent Viral Suppresion and HIV Related Death Rate

cor(HIV_AIDS_NY$percentviralsuppression, HIV_AIDS_NY$hivrelateddeathrate)

## [1] -0.0136484

Knowing that correlation is not causation, I want to know if people with HIV decide to take medicine to treat HIV as soon as possible can reduce the amount of HIV in the blood to a very low level which keeps the immune system working and prevent illness. Doing this correlation among the two variables, we can assume that people with HIV/AIDS might may have a possibility to not die from the HIV/AIDS because the two variables are negatively correlated.

Create a New Variable to Focus on hivrelateddeathrate, hivdiagnosisrate, aidsdiagnosisrate

What race have the highest Death rate, HIV diagnosis rate, and AIDS diagnosis rate?

HIVVariable <- HIV_AIDS_NY%>%
  select(race, 
         hivrelateddeathrate, hivdiagnosisrate, aidsdiagnosisrate) %>%
  tidyr::gather("id", "rate", 2:4) %>%
  group_by(race,id) %>%
  summarize(rate=mean(rate)) %>%
  filter(race !="All")

## `summarise()` has grouped output by 'race'. You can override using the
## `.groups` argument.

Look up the Rate of some variables

for (i in 1:nrow(HIVVariable)) {
  if (HIVVariable[i,2] == "percentlinkedtocarewithin3months") {
    HIVVariable[i,2] <- "Linked to Care"
  }
  if (HIVVariable[i,2] == "hivrelateddeathrate") {
    HIVVariable[i,2] <- "Death from HIV"
  }
  if (HIVVariable[i,2] == "hivdiagnosisrate") {
    HIVVariable[i,2] <- "HIV Diagnosis"
  }
  if (HIVVariable[i,2] == "aidsdiagnosisrate") {
    HIVVariable[i,2] <- "AIDS Diagnosis"
  }
}
HIVVariable

## # A tibble: 15 × 3
## # Groups:   race [5]
##    race                   id                rate
##    <chr>                  <chr>            <dbl>
##  1 Asian/Pacific Islander AIDS Diagnosis    11.6
##  2 Asian/Pacific Islander HIV Diagnosis     29.1
##  3 Asian/Pacific Islander Death from HIV 19234. 
##  4 Black                  AIDS Diagnosis    66.3
##  5 Black                  HIV Diagnosis     93.4
##  6 Black                  Death from HIV 19108. 
##  7 Latino/Hispanic        AIDS Diagnosis    30.1
##  8 Latino/Hispanic        HIV Diagnosis     54.2
##  9 Latino/Hispanic        Death from HIV 19672. 
## 10 Other/Unknown          AIDS Diagnosis    20.9
## 11 Other/Unknown          HIV Diagnosis     85.0
## 12 Other/Unknown          Death from HIV 17744. 
## 13 White                  AIDS Diagnosis    19.1
## 14 White                  HIV Diagnosis     40.4
## 15 White                  Death from HIV 18983.

Visualization of race, Death rate, HIV diagnosis rate, and AIDS diagnosis rate

raceplot <- HIVVariable %>%
  ggplot() +
  geom_histogram(aes(x=id, y=rate, fill = race),
      position = "dodge", stat = "identity") +
  ggtitle("AIDS diagnosis Rate, HIV Related Death Rate, 
          and HIV Rate  For Each Race") +
  ylab("Rate out of 100") +
  scale_y_continuous(trans = 'log10') +
  theme(axis.title.x=element_blank())

## Warning in geom_histogram(aes(x = id, y = rate, fill = race), position =
## "dodge", : Ignoring unknown parameters: `binwidth`, `bins`, and `pad`

raceplot

In this histogram, The numbers of HIV Diagnosis for each race are greater than the AIDS diagnosis.The number of Deaths from HIV are almost the same for almost the same.

Create Another Variable to Include Gender, Hiv-related-death-rate, Hiv-diagnosis-rate, AIDS-diagnosi-srate

What gender have the highest Death rate, HIV diagnosis rate, and AIDS diagnosis rate?

HIV3 <- HIV_AIDS_NY %>%
  select(gender, hivrelateddeathrate, hivdiagnosisrate, aidsdiagnosisrate) %>%
  tidyr::gather("id", "rate", 2:4) #%>%
  #filter(race !="All")
  
for (i in 1:nrow(HIV3)) {
  if (HIV3[i,2] == "percentlinkedtocarewithin3months") {
    HIV3[i,2] <- "Care"
  }
  if (HIV3[i,2] == "hivrelateddeathrate") {
    HIV3[i,2] <- "HIV Death"
  }
  if (HIV3[i,2] == "hivdiagnosisrate") {
    HIV3[i,2] <- "HIV"
  }
  if (HIV3[i,2] == "aidsdiagnosisrate") {
    HIV3[i,2] <- "AIDS"
  }
}  

Final Visualization

p3 <- HIV3 %>% 
  ggplot(., aes(id, rate))+
  geom_boxplot()+
  aes(color = gender)+
  facet_wrap(~gender)+
  ylab("Rate of") +
  xlab("") +
  scale_y_continuous(trans = 'log10')
  ggtitle("AIDS Diagnosis Rate, HIV Related Death Rate, 
          and HIV Diagnosis Rate For Gender") +
  theme(axis.text.x = element_text(angle = -45))

## NULL

p3

## Warning: Transformation introduced infinite values in continuous y-axis

## Warning: Removed 2101 rows containing non-finite values (`stat_boxplot()`).

For this plot, I use the facet wrap. This visualization helps us to have an idea about gender-related to AIDS, HIV, and HIV deaths. We can see the median of HIV for females is higher than for males and a little slightly different for All(genders). However, the HIV death for males is higher. The median AIDS for males and all are almost the same but the AIDS for females is higher compared to the other gender.

Essay

HIV is caused by a virus. It can spread through sexual contact, illicit injection drug use or sharing needles, contact with infected blood, or from mother to child during pregnancy, childbirth, or breastfeeding according to the article entitled HIV/AIDS by the Mayo Clinic. Those viruses are life-threatening conditions caused by the human immunodeficiency virus and by damaging people’s immune systems. For those viruses, there is no cure, but medications can control the infection and prevent the progression of the disease. HIV typically turns into AIDS in about 8 to 10 years. However, Access to better antiviral treatments has dramatically decreased deaths from AIDS and most people today don’t develop AIDS.

While working with this dataset, I was a little concerned about the lack of transparency in the data. For example, in the column gender, we have female, male, transgender, and all. All is not a gender by itself. It can include males, females, and transgender also. For borough, race, and age variables, we also have all. Black, white, and Asian can be also included in all. It remains unclear and cannot help to get a real insight into HIV and AIDS. Additionally, there are not any variables that include which way someone can get the virus, we cannot make any regression linear because of that. For example, if HIV spreads through sexual contact, illicit injection drug, sharing needles, contact with infected blood, from mother to child during pregnancy, childbirth, or breastfeeding. This dataset is not sufficient to have a better understanding of the death rate related to HIV/AIDS. Additionally, I was asking myself a lot of questions such as whether is any consent needed to access people’s personal information, and who stored, owns, or controls this dataset although this dataset is from the New York government website. Finally, I wanted to create a multiple regression analysis to see which variable might be correlated, but I was not able to do that because of the dataset. However, I was surprised to see the fourth plot that the death rate among race is almost the same.

Reference

https://www.nyc.gov/site/doh/data/data-sets/hiv-aids-surveillance-and-epidemiology-reports.page

https://www.mayoclinic.org/diseases-conditions/hiv-aids/symptoms-causes/syc-20373524#:~:text=HIV%20is%20caused%20by%20a,helping%20your%20body%20fight%20disease.