Methods & data
Methods and data
Recruiting cancer treatment trials open to patients 55 and older, by county, congressional district and state, measured in road-miles from where people live. Built by Neha Agrawal, MD. About · Data API · Key findings · Code and data on GitHub · Report a problem.
Limits and intended use
Not medical advice. A listed trial may not be enrolling at a given site today, and being listed does not mean a patient is eligible. Ask your care team, and check each study's record on ClinicalTrials.gov.
This site is in beta: methods and figures are reviewed weekly and may change. Check figures against Methods & data before citing them, and report problems.
Development and limitations. AI-assisted tools were used in developing this website and its data-processing code. Results depend on the accuracy of source records and the methods used to classify trials, assign locations, and calculate distances. Errors may remain. The map uses approximate highway-network distances; the ZIP finder estimates distance using straight-line distance multiplied by 1.2. These measures describe geographic availability and do not capture all barriers to participation.
This resource is intended for education, exploration, and advocacy. It is not medical advice or a patient-eligibility assessment. Before using a numerical finding in an external publication or policy brief, review its data date, methods, and validation status. Confirm individual trial information with the study team. Please report suspected errors through the issue-reporting page.
Key terms
- Listed trial.
- A recruiting, interventional cancer treatment trial on ClinicalTrials.gov with a US site, accepting some adults aged 55 or older. Listed does not mean a given patient qualifies.
- Road-miles.
- The shortest driving route on the highway network from where people live (each census tract's central point, the Census Bureau's internal point for the tract). Trial sites are placed at the center of their ZIP code or city, so distances to them are approximate. A median road distance is over the residents who have a road route to the target; residents without one (Alaska and all of Hawaii for a broad menu; Alaska and the Hawaiian islands beyond Oahu for an NCI center) are reported as a "no road route" share and counted as beyond every distance threshold.
- Residents short of trials (the map's default layer).
- Each county's circle is sized by the number of residents 55+ with fewer than 20 listed trials within 60 road-miles of home (the published line; the pills on the panel's first figure move it to 5, 10, 50 or 100, and every access figure on the page is then recomputed from the census tracts), summed over the county's census tracts; circles are drawn for counties with at least 500 such residents (99.8% of all such residents). Sizing by people rather than coloring land keeps the map in step with the headline: the counties where most residents are short cover about 62% of the land but hold about 11% of residents 55+. Counties are not shaded on this layer; the “Both” layer (Residents short of trials, over trials in the county) adds each county's own count of listed trials with a recruiting site in it (under 20; 20–99, a limited menu; 100 or more, a broad menu) beneath the circles; the “Metro vs rural” layer colors the circles by the county's USDA rural-urban code (metro, codes 1–3, solid; nonmetro, 4–9, ring), and “Trials in the county” under Color by shows the trial counts on their own. Filters recompute the circles from the census tracts.
- Cancer deaths vs trial access.
- County all-cancer death rates (NCI/CDC State Cancer Profiles, National Vital Statistics System, age-adjusted, five years) set against the share of residents 55+ with fewer than 20 trials within 60 road-miles. "High mortality" is the top third of counties with a published rate; "low access" is at least half of residents short. Counties with fewer than 16 deaths a year on average are classified but hatched; counties the source suppresses are gray. The two measures sit side by side and are never combined into a score. The incidence version of the same view uses the county median instead of the top third.
- A county with 100 or more listed trials, the cut-off ASCO's 2024 county analysis used for a “broad portfolio”.
- A county with 20 to 99 listed trials. This line is the map's own; ASCO's “limited portfolio” was 1 to 99.
- Oncologist (map layer).
- A clinician enrolled in Medicare whose primary specialty is medical oncology, hematology/oncology or hematology, counted in each county where they list a practice location (CMS Care Compare).
- NCI-designated center.
- A cancer center designated by the National Cancer Institute that treats adult patients.
Methods, sources and limits
Loading the methods text…
Data sources and dates
Cite as
Agrawal N. Cancer Clinical Trial Access Map (beta). https://cancertrialaccess.org/map.html (data as of ).
Every view has a link (“Copy link to this view”) that reopens the same place, filters and tab.