Directional Bias in Polyp Detection During Colonoscopy: A Prospective Observational Study

Supplementary material for a prospective observational study on directional bias in polyp detection during colonoscopic withdrawal, conducted at a tertiary center between June 2024 and February 2025. The study analyzed 1,185 lesions from 432 patients, recording the short-axis direction of initial lesion visualization. The data was authored by figshare admin karger and last updated on 2026-04-24.

Use Cases

• Analyze the distribution of lesion detection between upper-left and lower-right directions based on the short-axis directional classification (SADCO).
• Investigate factors associated with detection in the upper-left direction using multivariable logistic regression.
• Compare detection rates across lesion size, bowel segment, and macroscopic type as described in the subgroup analyses.
• Study the detection bias for advanced adenomas and advanced neoplasia as reported in the results.

Strengths

• Dataset is based on a prospective observational study with 1,185 lesions from 432 patients, providing a substantive clinical sample.
• Includes detailed subgroup analyses and multivariable logistic regression, suggesting structured variables for factors like lesion size and endoscopist experience.
• Published under a permissive CC-BY-4.0 license, allowing for reuse and redistribution.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for specific analytical needs.
• The dataset is very small at 73.2 KB, indicating limited scope, likely containing summary results or a data dictionary rather than raw patient-level records.

Provenance

Source

figshare admin karger

Collection Method

Prospective observational study conducted at a tertiary center, using the short-axis directional classification (SADCO).

Time Range

June 2024 to February 2025

Freshness

Last updated 2026-04-24 11:56:00; freshness should be verified.