1. The DATA

The Behavioral Risk Factor Surveillance System (BRFSS) objective is to collect uniform, state-specific data on preventive health practices and risk behaviors that are linked to chronic diseases, injuries, and preventable infectious diseases that affect the adult population (i.e 18 years and older) residing in the USA.The data are collected each year, in this project we will work with the data collected in 2013. Factors assessed by the BRFSS in 2013 include (https://www.cdc.gov/brfss/about/index.htm)

  1. Tobacco use,
  2. HIV/AIDS knowledge and prevention,
  3. exercise,
  4. immunization,
  5. health status,
  6. healthy days — health-related quality of life,
  7. health care access,
  8. inadequate sleep,
  9. hypertension awareness,
  10. cholesterol awareness,
  11. chronic health conditions,
  12. alcohol consumption,
  13. fruits and vegetables consumption,
  14. arthritis burden, and s
  15. seatbelt use.
  16. and others variables

Since 2011, BRFSS conducts both landline telephone- and cellular telephone-based surveys. In conducting the landline telephone survey, interviewers collect data from a randomly selected adult in a household using disproportionate stratified sampling, which is more efficient that simple random sampling. In conducting the cellular telephone version of the BRFSS questionnaire, interviewers collect data from randomly selected respondents with an equal probability of selection.

Could the data be generalizable? As random sampling was used for both data collection methods, the data for the sample is generalizable to the USA population.

This is an observational study, the researches not introduce any type of intervention on this study to modify the data. Because this study is not an experimental study, we cannot establish causality with any probable results, and correlation does not prove causation.

2. The Research Questions

Research Question 1

Does obesity(high Boy Mass Index (BMI)) aggravate the risk for heart attack? Wha is the relation between BMI and high cholesterol levels?
Reasoning: This questions tries to answer the possible impact of obesity in aggravating the health risks of heart attack and the possible relation between BMI and high cholesterol levels.

Cardiac arrest, one of the most commonly occuring medical condition affecting people across all backgrounds, is the abrupt loss of heart function in a person who may or may not have been diagnosed with heart disease. It can come suddenly, or in the wake of other symptoms. Cardiac arrest is often fatal, if appropriate steps aren’t taken immediately.

This research question will try to look for an association between high cholesterol levels and cardiac risks with increase in BMI.

Research Question 2

Is an increase of risk to develop asthma in overweight and obese populations?.
Reasoning: Obesity is a worldwide epidemic with a prevalence that has tripled in the last two decades. Obesity has been suggested to be a risk factor for the development of more difficult-to-control asthma. Although the mechanisms underlying the asthma–obesity relationship are not fully understood, several possible explanations have been put forward. This research question will try to find possible association between BMI levels and population diagnosed with atshma in this population.

Research Question 3

Is there any association between income and health care coverage?
Reasoning: It is clearly supposed that income and health care coverage are related, we can see if that is demonstrable using this data set.

3. Exploratory Data Analysis for each research question presented above

Packages to be use

library(ggplot2)
options( warn = -1)
library(ggplot2)
suppressMessages(library(dplyr));library(dplyr)
suppressMessages(library(statsr));library(statsr)
suppressMessages(library(corrplot));library(corrplot)
setwd("C:/Users/Olinto/Desktop/Statistics With R/Project Probability and Data")
load("brfss2013.Rdata")

Research Question 1

Does obesity(high BMI) aggravate the risk for heart attack? What is the relation between BMI and high cholesterol levels? Reasoning: This questions tries to answer the possible impact of obesity in aggravating the health risks of heart attack and the possible relation between BMI and high cholesterol levels.

Variables in consideration for this research question are:

bmi5cat: Computed Body Mass Index Categories
toldhi2: Ever Told Blood Cholesterol High
cvdinfr4: Ever Diagnosed With Heart Attack

Q1<-brfss2013%>% select(X_bmi5,X_bmi5cat,toldhi2,cvdinfr4) %>%
          filter(!is.na(X_bmi5),!is.na(X_bmi5cat),!is.na(toldhi2),!is.na(cvdinfr4))

plot(Q1$cvdinfr4,main="Diagnosed With Heart Attack ", xlab = "Ever Diagnosed with Heart attack",ylab = "Count")

Barplots of proportions

Barplot1 = barplot(CV_On_weight_yes, main="BMI vs ever diagnosed with Heart attact", ylab = "Percent",ylim=c(0,15), xlab="BMI", names.arg = c("Underweight", "Normal Weight", "Overweight", "Obese"),
                   col = c("grey90","grey80","grey50", "red"))

text(Barplot1, round(CV_On_weight_yes,2)*1.3, labels=round((CV_On_weight_yes),1))

FIGURE 6: USA population diagnosed with Heart Attack versus Body Mass Index in percent. Populations with high percent of Heart Attack are populations with underweight and Obese (7.4 and 8.2 respectively), compared with a 5.2 % of population with Normal Weight. (BRFSS2013 data set)

Barplot2 = barplot(BP_On_weight_yes,main="BMI vs Ever Told Blood Cholesterol High", ylab = "Percent",ylim=c(0,65), xlab="BMI",names.arg = c("Underweight", "Normal Weight", "Overweight","Obese"),col = c("grey90","grey80","grey50", "red"))

text(Barplot2, round(BP_On_weight_yes,2)*1.08, labels=round((BP_On_weight_yes),1))

FIGURE 7:Blood Cholesterol(Ever Told Blood Cholesterol High) as function of Body Mass Index. More than 50% of the Obese population has High Blood Cholesterol compared with only 30.5 % for the Underweight population.(BRFSS2013 data)

Conclusion/discussion for “Research question 1”

  1. Only a small percent of population appears in the underweight classification, FIGURE 2

  2. Very high percent of this population appears with high blood cholesterol, FIGURE 2

  3. Figure 6 shows a clear correlation between “ever diagnosed with Heart Attack” variable with an increase in weight between Normal-weight and obese. High percent value was found for those classified within underweight. (very Low BMI)

  4. Clear evidence of relation between weight and blood cholesterol was found, FIGURE 7,These results were expected because the weight has a strong correlation with Blood pressure.

Research Question 2

Is an increase of risk to develop asthma in overweight and obese populations?.
Reasoning: Obesity is a worldwide epidemic with a prevalence that has tripled in the last two decades. Obesity has been suggested to be a risk factor for the development of more difficult-to-control asthma. Although the mechanisms underlying the asthma–obesity relationship are not fully understood, several possible explanations have been put forward.

This research question will try to find possible association between BMI levels and population diagnosed with atshma in this population.

** Variables under consideration are: **

** _bmi5cat: Computed Body Mass Index Categories asthma3: Ever Told Had Asthma**

Q2<-brfss2013%>% select(X_bmi5,X_bmi5cat,asthma3) %>%
  filter(!is.na(X_bmi5),!is.na(X_bmi5cat),!is.na(asthma3))


plot(Q2$asthma3,main="Ever Told Had Asthma ", xlab = "Ever Told Had Asthma",ylab = "Count")

FIGURE 8:

plot(Q2$X_bmi5cat,main="BMI", xlab = "BMI categorization",ylab = "Count")

FIGURE 9 Population distribution as a function of BMI. Only small percents of the population appears as Underweight. The majority of the USA population can be classify between Normal weight, Over Weight or Obese.

Analysis of proportions. This analysis will be focused only in the population that was ever told that have asthma to see what BMI subpopulation is most affected in percent view

xUnderweightT<- count(Q2%>% select(X_bmi5,X_bmi5cat,asthma3) %>%  filter(X_bmi5cat == "Underweight"  ))
xNormalWeightT<- count(Q2%>% select(X_bmi5,X_bmi5cat,asthma3) %>% filter(X_bmi5cat == "Normal weight"  ))
xObeseT<- count(Q2%>% select(X_bmi5,X_bmi5cat,asthma3) %>%filter(X_bmi5cat == "Obese"  ))
xOverweightT<- count(Q2%>% select(X_bmi5,X_bmi5cat,asthma3) %>% filter(X_bmi5cat == "Overweight" ))

BMI<-as.numeric(c(xUnderweightT,xNormalWeightT,xOverweightT,xObeseT))
UnderweightTasthma<- count(Q2%>% select(X_bmi5,X_bmi5cat,asthma3) %>%  filter(X_bmi5cat == "Underweight" & asthma3 == "Yes"  ))
NormalWeightTasthma<- count(Q2%>% select(X_bmi5,X_bmi5cat,asthma3) %>% filter(X_bmi5cat == "Normal weight" & asthma3 == "Yes" ))
ObeseTasthma<- count(Q2%>% select(X_bmi5,X_bmi5cat,asthma3) %>%filter(X_bmi5cat == "Obese"& asthma3 == "Yes"  ))
OverweightTasthma<- count(Q2%>% select(X_bmi5,X_bmi5cat,asthma3) %>% filter(X_bmi5cat == "Overweight" & asthma3 == "Yes"))

Asthma_underweight<- (UnderweightTasthma/xUnderweightT)*100
Asthma_Normalweight<- (NormalWeightTasthma/xNormalWeightT)*100
Asthma_Obese<- (ObeseTasthma/xObeseT)*100
Asthma_Overweight<- (OverweightTasthma/xOverweightT)*100

Asth_BMI<-as.numeric(c(Asthma_underweight,Asthma_Normalweight,Asthma_Overweight,Asthma_Obese))


Barplotx1 = barplot(Asth_BMI, main=" Asthma versus BMI", ylab = "Percent",ylim=c(0,25), xlab="BMI", names.arg = c("Underweight", "Normal Weight",  "Overweight","Obese"),
                   col = c("grey90","grey80","grey50", "red"))

text(Barplotx1, round(Asth_BMI,2)*1.1, labels=round((Asth_BMI),1))

FIGURE 10: Distribution of population that have asthma related to BMI in percent. Clearly, we can see an increase in asthma related to BMI between Normal weight to Obese. High percents were found in the population classified as underweight (14.1%)

Conclusion/discussion. Is an increase of risk to develop asthma in overweight and obese populations. That can be seen on FIGURE 10. Clearly, we can see an increase of risk to develop asthma as related to an increase in weight. FIGURE 10 shows an 18.3% of asthma on that population classified within obese compared with only a 11.5% on the population with Normal weight. A high percent of asthma appears in the population classified within underweight.

Research Question 3

Is there any association between income and health care coverage? Reasoning: It is clearly suposed that income and helth care coverage are related, we can see if that is demostrable using this data set.

plot(brfss2013$income2, brfss2013$hlthpln1, xlab = 'Income Level', ylab = 'Health Care Coverage', main =
'Income Level versus Health Care Coverage')

FIGURE 11: Relation between income and health care coverage in this population.

Conlusion/Discussion. These results were expected. A strong associoation exists between income levels and healthcare coverage. Curiosity, these results show a percent of the population with high income level that does not have healthcare coverage. FIGURE 11