Johannes Blom, MD, PhD, is an associate professor at Karolinska Institutet,

Johannes Blom, MD, PhD

A new analysis suggests the mortality benefit of population-based colorectal cancer screening may be greater than previously estimated.

In a previous prospective cohort study of 379448 individuals in the Stockholm-Gotland region of Sweden, early invitation to fecal occult blood test screening was associated with a 14% reduction in colorectal cancer mortality after up to 14 years of follow-up.

The new analysis re-evaluated the screening program in a cohort of 376511 individuals, accounting for potential sources of bias that may have diluted the earlier estimate, including screening exposure among individuals in the control group and cancers diagnosed after screening invitations had ended. After adjustment, invitation to screening was associated with a 26% reduction in colorectal cancer mortality. With further adjustment for nonadherence, participation in screening was associated with a 43% reduction.

 14% mortality reduction with invitation to screening

 ~2/3 of the "control" group were eventually invited to screening too, diluting the comparison

 Some screening-group deaths were from cancers diagnosed years after invitations had ended, when they were no longer "exposed"

 26% mortality reduction with invitation to screening

 Accounts for people invited who never completed screening

 43% mortality reduction with actual participation in screening

In an interview with HCPLive, Johannes Blom, MD, PhD, associate professor at Karolinska Institutet, discussed why the original estimate may have underestimated the effectiveness of colorectal cancer screening and what the new findings suggest about the impact of participation in organized screening programs.

Q&A: 43% Lower Mortality Seen in Colorectal Screening Participants, With Johannes Blom, MD, PhD

HCPLive: Can you start by explaining what this study looked at, and why you felt the original estimate of how much colorectal cancer screening reduces deaths needed a second look?

Our previous evaluation of the Stockholm-Gotland colorectal cancer screening program showed a 14% reduction in colorectal cancer mortality among individuals invited to screening. However, we suspected that this estimate underestimated the true effectiveness of the program.

A major reason is that many individuals in the control group were eventually invited to screening, meaning they were not truly unexposed. At the same time, during follow-up, some people in the screening group died of colorectal cancer that were diagnosed years after screening invitations had stopped. At that point, they were no longer exposed to screening. Both factors tend to dilute the apparent effectiveness of screening.

The aim of the current study was therefore to re-evaluate the effectiveness of the program using a new statistical approach that accounts for these sources of bias. The method also made it possible to estimate the association between actual participation in screening and colorectal cancer mortality.

HCPLive: When you compare people who were screened to people who weren't, what can make that comparison misleading, and why?

Comparing people who were screened (participating) with people who were not is actually not feasible. Individuals who choose to participate in screening tend to differ from those who do not. They are often more engaged in preventive health care and may have different lifestyle and health characteristics. Therefore, some of the difference in colorectal cancer mortality between participants and nonparticipants may reflect differences between the groups themselves rather than the effect of screening alone. This highlights the importance of using a method that accounts for this self-selection bias.

HCPLive: Some people counted as "screened" had actually gone years without another screening, and some people counted as "unscreened" had eventually been invited to screen too. You developed a new way to correct for that mismatch. In simple terms, what does that correction actually do?

Blom: When the screening program was implemented, birth cohorts were randomized to different years for their first invitation. Our first analysis compared individuals invited to screening early with those invited later or not invited at all. As a result, approximately two-thirds of the control group were eventually invited to screening at least once.

Moreover, some individuals in the screening group died of colorectal cancer that were diagnosed years after screening invitations had stopped, when they were no longer exposed to screening.

The method adjusts for these situations, providing an estimate that more closely reflects the true effectiveness of screening.

HCPLive: Once you made those corrections, what did you find, both for people who were simply invited to get screened and for people who actually went through with it?

Blom: After adjusting for contamination between the groups, invitation to screening was associated with a 26% reduction in colorectal cancer mortality.

When we further accounted for nonadherence and estimated the association with actual participation in screening, the mortality reduction increased to 43%.

These findings suggest that the benefit of organized fecal occult blood test screening is substantially greater than earlier estimates indicated.

HCPLive: How do your findings line up with what's been seen in other countries that run similar colorectal cancer screening programs?

Blom: There are relatively few studies evaluating the effectiveness of ongoing screening programs on colorectal cancer mortality because it is difficult to identify a valid control group.

However, studies from Italy and Taiwan have also reported substantial reductions in colorectal cancer mortality among individuals who participate in screening. Although screening strategies, participation rates, and analytical methods differ between countries, the overall message is consistent. Participation in organized stool-based colorectal cancer screening is associated with a substantial reduction in the risk of dying from colorectal cancer.

HCPLive: What should this tell patients and doctors about how well colorectal cancer screening actually works?

Blom: Our findings suggest that the real-world effectiveness of colorectal cancer screening has been substantially underestimated because of the methodological challenges inherent in evaluating ongoing screening programs. The main message is that colorectal cancer screening works, and it may work better than many people realize.

Our findings suggest that inviting people to participate in organized screening programs could prevent more than one quarter of colorectal cancer deaths, while participation in screening was associated with more than 40% lower colorectal cancer mortality.

For clinicians, these findings reinforce the importance of encouraging eligible individuals to participate in screening. For patients, they provide further evidence that participation in an organized stool-based screening program is associated with substantially lower colorectal cancer mortality. At the population level, increasing participation remains one of the greatest opportunities to further reduce deaths from colorectal cancer.

Editor’s Note: Blom reported no relevant disclosures. 

Blom J, Nyström L, Jonsson H. Fecal occult blood screening outcomes adjusted for contamination bias and nonadherence. JAMA Netw Open. 2026;9(8):e2629856. doi:10.1001/jamanetworkopen.2026.29856

Blom J, Saraste D, Törnberg S, Jonsson H. Routine fecal occult blood screening and colorectal cancer mortality in Sweden. JAMA Netw Open. 2024;7(2):e240516. doi:10.1001/jamanetworkopen.2024.0516

Articles

Karolinska Institute's Johannes Blom, MD, PhD, explains why colorectal cancer screening's real-world mortality benefit has been underestimated.