top of page

How do we measure infant mortality?

  • Writer: Tracy Flood
    Tracy Flood
  • Mar 11
  • 3 min read

Updated: Apr 8

Is infant mortality high in the U.S.? How do we know? Is there a metric or a way to measure it?


Youtube Short: Is infant mortality high in the U.S.?

If we want to improve infant mortality in the United States (U.S.), we would first need to know: 

  • Q1: What is it?

  • Q2: How do we measure it?


Then we can ask:

  • Q3: Is it high in the U.S. and, specifically, where?


Defining Infant Mortality


Infant mortality is the death of an infant under 1 year of age.


How is Infant Mortality Measured?


Start by measuring infant mortality by counting the number of deaths in a population over a range of time.


Ex. The U.S. has about 20,000 infant deaths in 2022.

In order to know if the number is high or low and if improvement is possible, we would need to compare one region to another.


Ex. Other nations had fewer deaths compared to the U.S. in 2022


Number (#) of infant deaths in 2022 for the 38 member countries in the Organization for Economic Co-operation and Development (OECD). Data Source: Estimates developed by the UN Inter-agency Group for Child Mortality Estimation (UNICEF, WHO, et al.). Retrieved March 2025 from data.worldbank.org 
Number (#) of infant deaths in 2022 for the 38 member countries in the Organization for Economic Co-operation and Development (OECD). Data Source: Estimates developed by the UN Inter-agency Group for Child Mortality Estimation (UNICEF, WHO, et al.). Retrieved March 2025 from data.worldbank.org 

 


However, it may be unfair to compare nations with a different number of births. In order to make comparisons fair, we must divide the total number of deaths by the number of births in that geography. Since that number tends to be small, we can multiply it by 1,000. That will give us the metric: Infant Mortality Rate.


Infant Mortality Rate (IMR) = # of deaths for every 1,000 live births

In the U.S. the rate was about 5 deaths for 1,000 births. The U.S. rate is relatively high compared to other nations in the OECD (The Organization for Economic Co-operation and Development). It looks like there’s room for the U.S. to improve.



Data Source: 2022 estimates for OECD Member nations  by the UN Inter-agency Group for Child Mortality Estimation (UNICEF, WHO, et al.). Retrieved March 2025 from data.worldbank.org 
Data Source: 2022 estimates for OECD Member nations  by the UN Inter-agency Group for Child Mortality Estimation (UNICEF, WHO, et al.). Retrieved March 2025 from data.worldbank.org 



Data Challenges with Infant Mortality Rate (IMR)


Specifically, where do we focus our efforts? If we want to focus in places with opportunities to improve, we must look at smaller geographies.


Thankfully, infant mortality is a rare event. Rare events can create challenges for data quality.


Ex. A county in South Dakota has 200 births a year, last year they had 1 death and the year before they had zero. How does this community compare to the U.S. Benchmark of 5?




Using an average over a few years can help to smooth the year-to-year rates. Once we average over 5-years, we can see that the local rate is low compared to the U.S. Benchmark of 5.

 

Ex. Over 5 years, the average infant mortality rate in this South Dakota County was 3 deaths per 1,000. 


Example county infant mortality rates: 1-year vs. 5-year timespans. Infant Mortality Rate (IMR) is deaths for 1,000 live births. Note that the U.S. Benchmark is an infant mortality rate of 5 (not shown).
Example county infant mortality rates: 1-year vs. 5-year timespans. Infant Mortality Rate (IMR) is deaths for 1,000 live births. Note that the U.S. Benchmark is an infant mortality rate of 5 (not shown).


For the next question, we will ask if we have data for every county in the U.S. and, if so, where can we find it?



Try it Yourself

We used Canva for data visualizations after processing data in this Google Sheet: https://docs.google.com/spreadsheets/d/1XuCYIQMSWqoF_kV4UU5msHPO5A0izHW2eghVfy-XkWw/edit?gid=1577752676#gid=1577752676



References

The main reference used for U.S. numbers and rates is listed below. Please note that rounding and truncation was done for simplicity of demonstrating the calculation technique:

Ely, D.M., Driscoll, A.K. (2024) Infant mortality in the United States, 2022: Data from the period linked birth/infant death file. National Vital Statistics Reports; vol 73 no 5. Hyattsville, MD: National Center for Health Statistics. 2024. DOI: https://dx.doi.org/10.15620/cdc/157006. Retrieved from: www.cdc.gov/nchs/data/nvsr/nvsr73/nvsr73-05.pdf in March 2025.

Data Sources


[1] World Bank Group (2025). Number of Infant Deaths. Retrieved from https://data.worldbank.org/indicator/SH.DTH.IMRT in March 2025.

[2] World Bank Group (2025). Mortality Rate, Infant (per 1,000 births). Retrieved from https://data.worldbank.org/indicator/SP.DYN.IMRT.IN in March 2025.

[3] Open Climate Data (2025). Country Groups, Data, OECD. List of Retrieved from https://github.com/openclimatedata/countrygroups/blob/main/data/oecd.csv in March 2025.



The BroadStreet Institute is a registered 501(c)3. We are a growing community of volunteer data enthusiasts who believe that good data can heal the world. Volunteer teams manage the Community Health Impact Fund. All are welcome to give, volunteer, and to join our community.

We believe that good data can heal the world. Our vision is that those working to improve health have access to the best information possible. Our mission is to empower community health changemakers through training in data-guided decision-making and leadership. 

 

We are proud to have The BroadStreet Data Co-op as a major partner and sponsor. Website made with 🤍 in Milwaukee.

© 2024 by BroadStreet Institute

SITE CONTENT:

We sent information on big events and updates. We respect your inbox and your privacy. This email will only be used to share our Newsletter.

CONNECT WITH US!

  • Instagram
  • LinkedIn
  • TikTok
  • Youtube
  • bluesky
  • Threads
  • X
  • Reddit
  • Facebook
  • GitHub
bottom of page