This study, published in the International Journal of Colorectal Disease, analyzes one of the largest colorectal cancer (CRC) screening datasets to date, involving over 5.6 million participants in Tianjin, China. The researchers identify a significant "compliance gap"—only 27.1% of high-risk individuals followed through with a recommended colonoscopy—and pinpoint specific clinical symptoms and demographic factors that drive both patient adherence and the eventual detection of advanced adenomas.
TL;DR
Despite the availability of free screening, only about a quarter of high-risk individuals in this massive 5.6-million-person cohort complied with colonoscopy recommendations. While visible symptoms like bloody stool are the strongest drivers of patient compliance, objective factors like age and positive fecal tests are the most reliable predictors of finding advanced, pre-cancerous lesions.
What Problem Does This Paper Address?
Colorectal cancer is a leading cause of death globally, yet it is uniquely preventable because it usually develops from identifiable pre-cancerous lesions (polyps and advanced adenomas) over a long period. While the Fecal Immunochemical Test (FIT) is a common non-invasive screening tool, colonoscopy remains the "gold standard" for both diagnosis and immediate intervention.
The challenge is that screening programs often fail at the finish line: patients who test "high-risk" on preliminary screens frequently decline the follow-up colonoscopy. This paper seeks to understand the barriers to compliance and the "diagnostic yield" (the rate of actual disease found) in a real-world, large-scale public health setting.
Methods
The researchers conducted a population-based study in Tianjin, China, between 2012 and 2019. The screening followed a two-stage protocol:
- Preliminary Screening: Participants (aged 40–75) completed a High-Risk Factor Questionnaire (HRFQ) and a FIT.
- Risk Stratification: Individuals were labeled "high-risk" if they had a positive FIT, a family history of CRC, or specific digestive symptoms (e.g., chronic diarrhea or constipation).
- Follow-up: High-risk individuals were advised to undergo a free colonoscopy.
The team used logistic regression models to calculate Odds Ratios (ORs), identifying which factors were most strongly associated with a patient’s decision to get a colonoscopy and the likelihood of finding an Advanced Adenoma (AA) during that procedure.
Datasets
The study utilizes a massive longitudinal dataset from the Tianjin CRC screening programme:
- Initial Recruitment: 5,670,924 participants.
- High-Risk Group: 275,708 individuals (approx. 5.27% of the screened population).
- Colonoscopy Cohort: 74,685 individuals who complied with the follow-up.
- Final Analytical Sample: 150,676 participants (after filtering for missing variables like BMI and education level).
Key Insights and Ideas
- The "Symptom" Driver: Compliance is heavily driven by "alarm symptoms." Participants with a history of mucous bloody stool were 8.2 times more likely to comply with a colonoscopy than those without. Chronic diarrhea also strongly boosted compliance (OR = 5.73). This suggests that patients are much more motivated by visible illness than by "silent" risk factors like a positive lab test.
- The Education Gap: Higher educational attainment was a significant predictor of compliance (OR = 1.87 for high vs. low education). This points to a need for better health literacy and targeted outreach for less-educated populations.
- Anatomical Distribution: A striking majority of lesions were found in the distal colon/rectum (the lower part of the colon). Specifically, CRC cases in the distal colon were nearly four times more frequent than in the proximal colon.
- The AA Detection Trend: Over the seven-year study period, the detection rate for Advanced Adenomas increased from 4.17% to 6.80%. The authors attribute this to improved endoscopic technology and better-trained gastroenterologists.
Key Results
- Compliance Rate: Only 27.1% of high-risk individuals underwent the recommended colonoscopy.
- Diagnostic Yield: Among those who had a colonoscopy, the study found:
- 1,101 cases of CRC (1.47% yield).
- 4,143 cases of Advanced Adenomas (5.55% yield).
- 27,643 cases of Non-Advanced Adenomas (37.01% yield).
- Predictors of Advanced Adenomas (AA): The strongest predictor for finding an AA was age (70–75 years: OR = 3.72) and a positive FIT result (OR = 1.65). Interestingly, while bloody stool drove compliance, it was less predictive of AA than age and FIT results.
Limitations and Future Work
The authors acknowledge several limitations:
- Self-Reporting Bias: The HRFQ relies on participant memory for symptoms and family history, which can be subjective.
- Geographic Specificity: The data is limited to Tianjin; results might differ in regions with different dietary habits or healthcare infrastructures.
- Unmeasured Confounders: The study did not account for health insurance type or the frequency of regular medical check-ups, which likely influence compliance.
Future Directions: The authors suggest that future programs should focus on "solicitude" for participants with low education or those who have experienced major adverse life events (which were found to correlate with higher risk but lower compliance). They also emphasize the need to promote non-invasive tests to bridge the gap for those who refuse colonoscopies.
Summary updated