Lung Cancer Mortality Data-Guided Donation
- Shamini V De Silva
- May 11
- 6 min read
Updated: May 13
Big Picture Goal:
Reduce the mortality rate of Lung Cancer in the U.S.
Project Objective: A Data-Guided Donation
By the end of the 13-weeks, donate to one or more non-profit organizations running evidence-based programs to reduce lung cancer mortality rates, or an intermediate outcome, in key counties within the United States.
Learning Objective:
This project was completed as part of a 13-week training program where the learning objective was to practice the 7 core data skills (querying, entering, cleaning, enhancing, analyzing, visualizing, and humanizing data).
5 Steps
Overview. A team of Community Data Interns at BroadStreet Institute reviewed data on lung cancer mortality rates in the U.S. from September to December 2025 and identified non-profit organizations doing evidence-based strategies.
Step 1. Review the lung cancer data to create a list of 📍key counties by state
Step 2. Teams prioritize 1 📍key county
Step 3. Review of evidence-based interventions in the scientific literature
Step 4. Identify non-profit organizations doing evidence-based solutions in key regions.
Step 5. Donate to non-profit organizations
Inputs:
💰 Money: $235 to donate
⏰ Time: 13 weeks (September to December 2025)
👥 Team: 66 people (15 leaders, 51 team members)
🧰 Inventory:
🧰 Data Tools: Google Sheets, Datawrapper, Flourish
📊 Data Source: CDC WONDER Underlying Cause of Death (CDC WONDER, 2025)
Indicator Definitions:
Lung cancer death defined as ICD-10 Code: C34, "malignant neoplasm (cancer) of the bronchus and lung."
Mortality rate for lung cancer. Crude mortality was calculated from number of overall number of lung cancer deaths for every 100,000 people, 2018-2023.
Methods.
Eight teams, each composed of 4-6 members, used CDC WONDER mortality data (CDC WONDER, 2025) to explore lung cancer mortality rates across various U.S. states. Each team identified a specific county as a 📍key county, searched evidence-based interventions, and then localized non-profit organizations that tackle lung cancer mortality rates, or an intermediate outcome, in the county.
Step 1. Create a list of 📍key counties by state
The objective for Step 1 was for each team to create a short list of counties of interest in assigned states
To begin the process, 29 U.S. states were randomly assigned to members in each team. Some team members were assigned to the same state. Each team member explored county-level mortality rates in their assigned states and selected 1 county to propose. They proposed either the county with the highest overall crude mortality rate (Data source: CDC Wonder, Underlying Cause of Death, 2018 - 2023) for lung cancer within their assigned state, or the highest number of deaths for further analysis.
Step 2. Teams prioritize 1 📍key county
The objective for Step 2 was for each team to propose and prioritize one county from amongst the list of counties selected by team members in Step 1.
Each team refined its county proposals after gaining insight from overall crude mortality estimates for 2018-2023, gender disparities, and population figures. Teams selected one county based on the following criteria: the highest mortality rate (2018-2023), the largest gender mortality disparity, or the largest population.
Step 1 and 2 limitations and caveats:
States and counties may be excluded from the decision process if team members were not assigned to a state or did not propose a key county for their state. In total 29 of the 50 states were represented in this analysis. This means that 21 states were missing from consideration: California, Connecticut, Delaware, Iowa, Kentucky, Louisiana, Maryland, Massachusetts, Minnesota, Missouri, Montana, Nebraska, Oklahoma, Oregon, Rhode Island, South Carolina, Tennessee, Texas, Virginia, Washington, Wisconsin
Duplicate states. Some states were reviewed by more than one team member
Counties with the highest national mortality rates may not be chosen by every team.
Data-guided, value-based decision-making was followed. This means that each team member could propose a county with some flexibility on how the final county was decided.
Step 3. Review of evidence-based interventions in the scientific literature
The objective for Step 3 was to identify evidence-based interventions, including policies and programs, for reducing lung cancer mortality or an intermediate outcome such as lung cancer incidence.
Team members searched PubMed, Cochrane, and the U.S. Preventive Services Task Force (USPSTF) to identify evidence-based interventions that reduce mortality or address intermediate outcomes. They focused on the highest level of evidence, namely systematic reviews and meta-analyses. On USPSTF, recommendations that were ranked as A or B were reviewed.
Step 3 Limitations and caveats:
Evidence-based interventions published in other databases may not be considered.
Step 4. Identify non-profit organizations doing evidence-based solutions in key regions.
The objective for Step 4 was for each team to locate non-profit organizations that were actively running evidence-based programs that would serve the population in their selected county.
Every team prioritized 1 county: Eight teams explored eight counties. The teams identified evidence-based programs serving populations in these counties and submitted their proposals for review. Specifically, teammates were searching for non-profit organizations implementing evidence-based programs or promoting evidence-based policies to reduce lung cancer mortality or an intermediate outcome (e.g. lung cancer incidence).
At least one organization was found in all eight counties. The identified organizations and programs were independently reviewed by BroadStreet leaders to determine whether they aligned with a decision matrix (see table "Decision Matrix") proposed by the BroadStreet Institute Steering Committee.
Limitations and caveats:
Non-profit organizations working on evidence-based interventions may not be considered if team members did locate them on the internet.
Decision Matrix | |
|---|---|
Criteria # | BroadStreet criteria for donation eligibility |
Criteria 1. Initial screening criteria | Organizations are running programs in key places. Note: "Key places" are counties where there are opportunities to reduce mortality (i.e. mortality rates are high, the number of deaths are high, disparities are high, etc.). |
Criteria 2. Initial screening criteria | The program(s) being run by the organization within that key place are evidence-based solutions for reducing lung cancer mortality (or an intermediate outcome). |
Criteria 3. Initial screening criteria | Organizations are non-profits 501(c)3 to which we could donate (i.e. not a government agency). Note: Non-profit are still eligible if they receive government funding. |
Criteria 4. | A donation, if given, could be reasonably assumed to be going to: (1) The key county and (2) The program of interest (i.e. the money would not go to another program or place). |
Criteria 5. optional | Organizations are at least 1 year old and have filed IRS 990s for 501(c)3. |
Criteria 6. optional | Organizations have an annual report in which we could see the number of people served. |
Criteria 7. | Is the organization free from obvious red flags or causes of concern? |
Project Outcomes
Key Counties. The eight counties proposed by teams were:
Adams County, Ohio;
Bennington County, Vermont;
Lincoln County, West Virginia;
Montmorency County, Michigan.
Sharp County, Arkansas;
Somerset County, Maine;
Stone County, Arkansas; and
Sussex County, Delaware.
County selection criteria varied by team. For example: Some counties were chosen that had the highest overall mortality rate compared to other counties. Other counties were identified as having the highest gender disparities among the shortlisted counties.
Evidence-based Programs Serving Key Counties
At the end of the criteria review process, two organizations were identified as suitable for donations. This included CARTI serving residents in Stone County, Arkansas and the Maine Cancer Foundation serving Maine, which we hoped included Somerset County, Maine.
CARTI and the Maine Cancer Foundation passed the decision matrix criteria. Both programs offer lung cancer screening, an evidence-based intervention in accordance with the USPSTF grade B recommendations (U.S. Preventive Services Task Force, 2021). The review team was unsure whether donations to the Maine Cancer Foundation would be allocated to Somerset County in Maine (Criteria 4). However, since Maine ranks third among U.S. states with regard to lung cancer mortality, we believe a donation would benefit the entire state including Somerset County.
We believe that these programs are serving 3 high priority places:
Somerset County, Maine had the highest overall mortality rate compared to other counties considered by the team.
Stone County, Arkansas was identified as having the highest gender disparities among the shortlisted counties.
Sharp County, Arkansas has high mortality rates in both males and females.
Overall, based on our evaluation, we decided to donate $157 to CARTI, which serves two key counties in Arkansas, and $78 to the Maine Cancer Foundation.
Citations
Data Source:
Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS) (2025). Underlying Cause of Death 2018-2023 on CDC WONDER Online Database, released 2025. Data are compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program. Retrieved from https://wonder.cdc.gov/controller/saved/D158/D464F012 in Sept 2025.
References:
U.S. Preventive Services Task Force. (2021). Recommendation: Lung Cancer Screening.https://www.uspreventiveservicestaskforce.org/uspstf/recommendation/lung-cancer-screening
Author List:
A huge thank you to all of the leaders and team members and leaders who made this Data-Guided Donation Possible.
Leaders
Breanna Martin
Diana Wong
Dinika Mirpuri
Isha Baxi
Karan Wadhawan
Kesari Singh
Laila Al Afnan
Liz Meehan
Liz Montoya
Monica Villanueva
Nowrin Nisa
Shamini De Silva
Zaporah Nelson
Team Members
Alex Reifel
Arathy Karakkatt
Beau Braddock
Bilsuma Adema
Brad Lipson
Colleen Brewer
Crystal Green
Diamond Crumby
Emily Berry
Emily DeVane
Frank Dolecki
Gabby DeSisto
Gabriel Rodriguez
Hafsa Akhtar
Ivy Hu
Karina Desai
Katherine Brown
Kevin Lee
Kylie Chow
La'Keitha Patterson
Liz Montoya
Madeline Wallace
Mahbod Kazemarab
Manmeet Kaur
Marlee Graeser
Mekelit Kidane
Micaela Atkinson
Miyalla Tarver
Moses Quintana
Noah Burney
Priscilla Macias
Radhika Agarwal
Rhiannon Stracener
Sarah Roy
Sharon Joanna Azariah
Sydney Case
Sydney Levin
Yahia Moustafa
Zexuan Wang
