Data Exploration on Disparities in Life Expectancy

Posted on Feb 9, 2017
The skills the author demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.

Life Expectancy and Public Policy

Among the policy proposals that the new administration and Congress are likely to consider are changes affecting Social Security and Medicare. It is often argued that Americans are living longer and raising the eligibility age for Social Security and Medicare is a logical, necessary, and fair. But does the data show U.S. today have a consistent, higher life expectancy, regardless of income, geography, or gender?

This question is worth examining not only because of the possible changes to retirement benefits, but also because it can help to uncover social differences worthy of additional inquiry. Data compiled by the Health Inequality Project and available online provides an ideal starting point for this inquiry.

Plotting female and male life expectancy by household income reveals that life expectancy rises sharply with income, then tapers off. The gap between female and male life expectancies is also clear and persistent, and rises as income levels grow.

As with other life expectancy figures discussed here, the figures in this graph are race-adjusted, with differences in life expectancy associated with race or ethnicity, which warrant additional research, having been removed.

Data Exploration on Disparities in Life Expectancy

U.S. Life Expectancy Estimates by Household Income and Gender


With incomes concentrated near the low end of the graph, there is sparse data available above about $250,000 in household income. A logarithmic scale for income may be more revealing.

The next graph updates the previous one by transforming household income on a log scale.

Data Exploration on Disparities in Life Expectancy

U.S. Life Expectancy Estimates by Household Income (Log Scale) and Gender

For the bulk of observations, which are found between roughly $20,000 and $200,000, it appears that a linear model could approximate the relationship well. However, the slope would be steeper for men than for women. Note that the slopes are lower at the lowest incomes and especially at the highest incomes, compared with the middle income range. It appears that statements about a rising U.S. life expectancy require greater nuance given the wide range of life expectancies and their relationship with income levels and gender.

Data on Geographic patterns

If life expectancy varies by household income level and gender at the national level, are these patterns consistent from state to state, or are they more pronounced in some states or regions? One way to evaluate this is a graph that compares the life expectancy of people in the bottom quartile of household income with those in the top quartile of income in the same state. To isolate the influence of gender, two graphs can be used to illustrate the patterns for women and men separately, and they can then be compared.

In the graph of women’s life expectancy by state below, the points for every state would fall on the diagonal line if women in the lowest income quartile had the same life expectancy (horizontal axis) as women in the highest income quartile (vertical axis). In no state is this close to reality.

Hawaii serves as an example below and is indicated by an annotation. As the note on the graph explains, the vertical distance from the diagonal line to the point above it represents 4.3 years of life expectancy that women in the lowest income quartile lose compared with their high-income peers in the state.

Data Exploration on Disparities in Life Expectancy

State-Level Female Life Expectancy by Household Income Quartile


There are considerable differences among states in the life expectancies of women of both income groups and some variation in the number of years lost by women in the lower income group.

Regional differences, though not clear-cut, may also be present. States in the Northeast, shown in red, are clustered mostly on the right due to higher life expectancies for lower income women than most other states. In addition, states in the Midwest, indicated in blue, appear to be a bit farther from the diagonal line than other states, meaning that the disparity in female life expectancy by income seems to be greater there. Further research would be needed to tell whether other groupings of states, such as according to shared policies or leading industries, rather than geography, would be more meaningful.

Male life expectancy by the state for first and fourth quartiles of household income reveals an even larger gap. The dashed line that connects the point representing Indiana to the diagonal line means that lower income men in that state have a life expectancy that is 9.7 years less than their high-income counterparts.

State-Level Male Life Expectancy by Household Income Quartile

State-Level Male Life Expectancy by Household Income Quartile

There still appear to be some regional patterns, such as higher life expectancies for men with first quartile household income in most Northeastern states, but they seem to be weaker than for women.

The number of years of life expectancy that women lose out on if they are in the first quartile of household income instead of the fourth can also be seen on a map. The closer states are to blue in this graph, the smaller the gap in life expectancy between women of the first and fourth income quartiles. Clearly, states such as California and New York have significantly smaller disparities than others, such as Kansas.

LostΒ Years of Female Life Expectancy (Race-Adjusted) For Quartile 1 Vs. Quartile 4 Household Income

Similar patterns can be seen for men. However, note that the scale is different because the disparity is much higher overall.

Lost Years of Male Life Expectancy (Race-Adjusted) For Quartile 1 Vs. Quartile 4 Household Income

Lost Years of Male Life Expectancy (Race-Adjusted) For Quartile 1 Vs. Quartile 4 Household Income

Again, California and New York appear to have smaller gaps, but their gaps for men are comparable to the largest gaps of any state for women.

For states such as Wyoming and Indiana, lower income men lose almost a decade of life compared with higher income men. Β 

Changes over time

Over the period from 2001 to 2014, female life expectancy has been rising. This is the case at both the 25th and 75th percentile of household income. However, there remains a large gap in life expectancy between the lower and upper income groups, depicted as the shaded area between the lines, that is persistent, if not growing.

Female Life Expectancy Estimates by Income Percentile

Female Life Expectancy Estimates by Income Percentile

Some additional research into the reversal that appears in 2004 would be appropriate. Life expectancy figures around 86 years are found at percentiles just above and just below 75th percentile. It is not clear whether the anomaly is an erroneous entry or has another explanation.

For male life expectancy, the trend is similar, with increases for both first and fourth income quartiles, and a large gap remaining. For men, the gap appears to be widening due to the modest rise in life expectancy for lower income men.

Male Life Expectancy Estimates by Income Percentile

Male Life Expectancy Estimates by Income Percentile

How do the trends for men and women compare?

Overlaying the two previous graphs, we can see that, while the overall life expectancy is rising, large disparities remain based on income and gender. The life expectancy of higher income men appears comparable to that of lower income women. At the same time, there is a wide and growing gap between the life expectancies of lower income men and the other groups (higher income men as well as women of either income group).

Life Expectancy Estimates by Gender and Income Percentile

Life Expectancy Estimates by Gender and Income Percentile


Topics for future research

It is clear that a story that states that life expectancy has now reached a high level for all Americans is far too simple to serve as a basis for public policy. Differences in life expectancy by income level, gender, and state or region are pronounced. Further analysis is needed to inform the policy-making discourse.

Other questions are also worth exploring. These include:

  • How does life expectancy compare in rural vs. urban areas and how are they changing?
  • How do differences in public policy, such as provision of subsidized health insurance or other health-related benefits, affect life expectancy?
  • Does the degree of income inequality of a state affect disparities in life expectancy?
  • What factors explain the comparatively low life expectancies of lower income men, which appear to lag in growth behind other groups?

About Author

Evan Frisch

Evan Frisch has more than a decade and a half of experience using technology and data to achieve results for organizations in the private, public, and non-profit sectors. Evan received his undergraduate degree with honors from Yale University,...
View all posts by Evan Frisch >

Related Articles

Leave a Comment

No comments found.

View Posts by Categories

Our Recent Popular Posts

View Posts by Tags

#python #trainwithnycdsa 2019 2020 Revenue 3-points agriculture air quality airbnb airline alcohol Alex Baransky algorithm alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus ames dataset ames housing dataset apartment rent API Application artist aws bank loans beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep boston safety Bundles cake recipe California Cancer Research capstone car price Career Career Day citibike classic cars classpass clustering Coding Course Demo Course Report covid 19 credit credit card crime frequency crops D3.js data data analysis Data Analyst data analytics data for tripadvisor reviews data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization database Deep Learning Demo Day Discount disney dplyr drug data e-commerce economy employee employee burnout employer networking environment feature engineering Finance Financial Data Science fitness studio Flask flight delay gbm Get Hired ggplot2 googleVis H20 Hadoop hallmark holiday movie happiness healthcare frauds higgs boson Hiring hiring partner events Hiring Partners hotels housing housing data housing predictions housing price hy-vee Income Industry Experts Injuries Instructor Blog Instructor Interview insurance italki Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter las vegas airport lasso regression Lead Data Scienctist Lead Data Scientist leaflet league linear regression Logistic Regression machine learning Maps market matplotlib Medical Research Meet the team meetup methal health miami beach movie music Napoli NBA netflix Networking neural network Neural networks New Courses NHL nlp NYC NYC Data Science nyc data science academy NYC Open Data nyc property NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time performance phoenix pollutants Portfolio Development precision measurement prediction Prework Programming public safety PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R Data Analysis R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn seafood type Selenium sentiment analysis sentiment classification Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau teachers team team performance TensorFlow Testimonial tf-idf Top Data Science Bootcamp Top manufacturing companies Transfers tweets twitter videos visualization wallstreet wallstreetbets web scraping Weekend Course What to expect whiskey whiskeyadvocate wildfire word cloud word2vec XGBoost yelp youtube trending ZORI