1 of 33

LIFE EXPECTANCY (WHO)

��Sai Manoj Chatrathi, Lokeshwar Reddy Gowkanapalli, Trivikram Gummaraj Shivakumar Sridevi, Anjani Sowmya Bollapragada

2024-11-29

2 of 33

CONTENTS

  • Public Health Question
  • Literature Review Highlights
  • Data Description
    • Variable Description
  • Dataset Issues and Handling
  • Exploratory Data Analysis
  • Regression Models
  • Data Analysis
  • Shiny Application
  • Conclusion
  • Future Enhancements
  • References

3 of 33

PUBLIC HEALTH QUESTION

What are the key factors influencing life expectancy globally, and how do these factors differ between developed and developing countries over time?

Importance

  • Life expectancy is a crucial measure of a population’s overall health and well-being.
  • Understanding the various factors that contribute to life expectancy is vital for developing targeted public health policies and interventions.
  • This study is important as it identifies actionable areas for improving life expectancy, especially in developing nations, where healthcare resources may be limited and disparities are high.

Data Source: The dataset is obtained from Kaggle, sourced from the World Health Organization (WHO) and the United Nations (UN).

4 of 33

LITERATURE REVIEW HIGHLIGHTS

  • Previous Research: Studies have examined economic (income, GDP) & demographic factors (healthcare spending) impacting life expectancy
  • Identified Gaps: Limited focus on the impact of immunization coverage and predominance of cross-sectional studies; mixed-effects models for longitudinal analysis are rare.
  • Recent Advances: Notable healthcare improvements in developing countries over the past 15 years, influencing life expectancy trends.
  • Relevance of This Study: Integrates immunization and health-related factors and utilizes mixed-effects models for robust, longitudinal analysis of global data.

5 of 33

DATA DESCRIPTION

​

​

​

Variable

Description

Data Type

Country

Name of the country

String

Year

Year of the record

Integer

Status

Developed or Developing status of the Country

String

Life.expectancy

Life Expectancy in age

Float

6 of 33

DATA DESCRIPTION

​

​

​

Variable

Description

Data Type

Adult.Mortality

Adult mortality rates (probability of dying between 15 and 60 years per 1000 population)

Float

infant.deaths

Number of infant deaths per 1000 population

Integer

Alcohol

Alcohol consumption per capita

Float

percentage.expenditure

Expenditure on health as a percentage of GDP per capita

Float

7 of 33

DATA DESCRIPTION

​

​

​

Variable

Description

Data Type

Hepatitis.B

Hepatitis B immunization coverage among 1-year-olds (%)

Float

Measles

Number of reported cases per 1000 population

Integer

BMI

Average Body Mass Index of the entire population

Float

under.five.deaths

Number of under-five deaths per 1000 population

Integer

8 of 33

DATA DESCRIPTION

​

​

​

Variable

Description

Data Type

Polio

Polio immunization coverage among 1-year-olds (%)

Float

Total.expenditure

General government expenditure on health as a percentage of total government expenditure (%)

Float

Diphtheria

DTP3 immunization coverage among 1-year-olds (%)

Float

HIV/AIDS

Deaths per 1000 live births due to HIV/AIDS (0-4 years)

Integer

9 of 33

DATA DESCRIPTION

​

​

​

Variable

Description

Data Type

GDP

Gross Domestic Product per capita (USD)

Float

Population

Population of the Country

Integer

thinness..1.19.years

Prevalence of thinness among children and adolescents (ages 10-19) (%)

Float

thinness..5.9.years

Prevalence of thinness among children (ages 5-9) (%)

Float

10 of 33

DATA DESCRIPTION

​

​

​

Variable

Description

Data Type

Income.composition.of.resources

Human Development Index in terms of income composition of resources (0 to 1 index)

Float

Schooling

Number of years of schooling

Integer

11 of 33

VARIABLE DESCRIPTION

​

  • Primary Response Variable:

​

​

​

  • Predictor Variables:

​

Life Expectancy

The average number of years a person is expected to live, based on age-specific mortality rates

Economic Factors

Health Factors

Social Factors

Status, Income composition of resources, Total Expenditure, GDP

Adult Mortality, Infant Deaths, Alcohol Consumption, Measles Cases, BMI, Under Five Deaths, Polio Coverage, Diphtheria Coverage, HIV Prevalence, Thinness Factors

​

Schooling

12 of 33

VARIABLE DESCRIPTION

​

    • Rationale :

​

Economic variables

Health-related variables

Social Factors

Economic variables are critical as they reflect a country’s financial capability to invest in healthcare.

Health-related variables, including immunization and mortality rates, directly affect population health.

Social factors like schooling impact health literacy and access to care, influencing life expectancy.

13 of 33

VARIABLE DESCRIPTION

​

    • Confounding Variables and Their Impact :

Handling Confounders

The analysis accounted for potential confounders by employing mixed-effects models, which allow for random effects to capture unobserved variability among countries.

Confounding Variables

Factors like population size, political stability, and healthcare infrastructure could confound the relationships between predictors and life expectancy.

14 of 33

DATASET ISSUES AND HANDLING:

15 of 33

DATASET ISSUES AND HANDLING:

16 of 33

EXPLORATORY DATA ANALYSIS (EDA)

Conducted a summary of the data, including checking data types, identifying missing values, and ensuring consistency in variable formats

17 of 33

EXPLORATORY DATA ANALYSIS (EDA)

18 of 33

EXPLORATORY DATA ANALYSIS (EDA)

19 of 33

LOGISTIC REGRESSION

20 of 33

RIDGE REGRESSION

21 of 33

LASSO REGRESSION

22 of 33

ROC CURVE COMPARISON

23 of 33

FACTORS AFFECTING LIFE EXPECTANCY

24 of 33

LIFE EXPECTANCY VS ECONOMIC STATUS

25 of 33

LIFE EXPECTANCY OVER TIME

26 of 33

LIFE EXPECTANCY VS SCHOOLING

27 of 33

GEOSPATIAL VISUALIZATION

28 of 33

R SHINY APPLICATION

29 of 33

R SHINY APPLICATION

30 of 33

CONCLUSION

  • Key Determinants of Life Expectancy:
    • Income Composition of Resources, HIV Prevalence and Infant Deaths
  • Model Accuracy:
    • Logistic and Lasso Regression have performed better than Ridge Regression.
  • Disparities Between Countries:
    • Life Expectancy is better in Developed Countries than in Developing Countries.
  • Temporal Trends:
    • Average Life Expectancy is increasing over the years.
  • Role of Education:
    • Schooling shows a significant improvement in LIfe Expectancy.

31 of 33

FUTURE ENHANCEMENTS

  • Leverage machine learning algorithms, such as Gradient Boosting or Neural Networks, to improve life expectancy predictions.
  • Explore region-specific trends to identify areas with significant improvements or declines.
  • Add variables such as income inequality (Gini index) and employment rates to assess their impact on life expectancy.
  • Create a "What-If" scenario analysis to evaluate the potential outcomes of policy changes, such as increased vaccination rates or reduced infant mortality.
  • Explore disparities in life expectancy between urban and rural areas or among different socio-economic groups within countries.

32 of 33

REFERENCES

​

​

33 of 33

THANK YOU!