- Four recent studies examined cancer screening and early detection across primary care, colorectal screening, longitudinal follow-up, and early-onset colorectal cancer research.
- Together, they show that earlier detection depends on more than access to a test. It also depends on where screening enters care, how recommendations change what is found, whether patients return, and how earlier risk may be recognized in younger adults.
- Taken together, the findings suggest that early detection is best understood as a care pathway, not a single screening event.
Earlier cancer detection can change the course of care. It means finding precancerous lesions before they progress, identifying disease at a more treatable stage, and creating more time for intervention.
These recent studies using Truveta Data offer a real-world view of the path to earlier detection, from screening and follow-through to research on even earlier risk identification.
1. Earlier detection starts in primary care
A study from University of Michigan exploring cervical cancer screening found that family medicine and internal medicine clinicians now provide most cervical cancer screening in the US. Overall, these clinicians delivered 61.9% of cervical cancer screenings, rising to 68.4% among adults ages 50–65. The same study found that only 70.5% of eligible individuals received at least one cervical cancer screening during the study period.
The finding is straightforward but important: much of screening begins in primary care, not specialty care. That shifts where researchers may need to look for missed opportunities, especially as cervical cancer screening moves toward primary HPV testing and self-collection.
2. Policy changes can shift detection earlier
A colorectal cancer screening study from Truveta Research examined adults ages 40–64 before and after the 2021 US Preventive Services Task Force (USPSTF) recommendation to begin average-risk screening at ages 45–49. After the recommendation changed, the share of screened patients ages 45–49 rose from 9% to 19%. In that same group, adenomatous polyp findings increased from 45.7% to 53.4%, while adenocarcinoma findings fell from 2.1% to 1.0%.
While the authors note this does not prove causality on its own, these findings highlight an important real-world signal that expanded colorectal cancer screening in this age group was accompanied by more precancerous findings and fewer adenocarcinomas.
3. The first test is not enough
Stool-based colorectal cancer screening only works as intended when patients repeat testing at recommended intervals, with FIT and multi-target stool DNA tests designed to be repeated every 1–3 years. However, in a real-world study of 372,522 average-risk patients who received at least one stool-based test, 52% had only one test over four years of follow-up. Less than half received more than one test, indicating low adherence to repeat testing.
The same study found lower repeat adherence among women, Black individuals, Native American or Pacific Islander individuals, and unmarried individuals. Region and provider specialty were also strong predictors of whether patients received more than one test.
These findings highlight the practical implication that a screening program only works when patients stay on schedule, and one of the biggest weaknesses in screening may come after the first test.
4. The next opportunity is earlier risk identification
The next opportunity in early detection may be earlier risk identification, especially for younger adults who fall outside routine screening windows.
Researchers from Truveta explored early-onset colorectal cancer prediction using machine learning and large language models to analyze conditions, lab results, and observations from up to six months before diagnosis to identify patients at elevated risk. The fine-tuned GPT-4o model identified 73% of patients who went on to be diagnosed with early-onset colorectal cancer and correctly ruled out risk in 91% of patients who did not.
This is early-stage research, not screening guidance, but it suggests that earlier risk may leave detectable clinical signals in the record before diagnosis, creating a new area for research in younger adults outside routine screening windows.
Why it matters
These studies highlight how early detection succeeds or fails across the care pathway, not at a single visit. Real-world data can help bring that pathway into view by revealing where screening begins, how clinical care changes after recommendations shift, and where opportunities remain for earlier identification.
Read the full studies
- Characterizing Cervical Cancer Screening in the US: Preparing for the Era of Self-Collection, JABFM
- Colorectal cancer screening and detection following USPSTF recommendations for ages 45-49: Insights from a large EHR cohort, Journal of Clinical Oncology
- Real-World Adherence to Repeat Testing for Stool-Based Non-Invasive Colorectal Cancer Screening Tests Among Individuals With Average Risk, The American Journal of Gastroenterology
- Predicting Early-Onset Colorectal Cancer with Large Language Models, AMIA Annual Symposium Proceedings
