Department of Gastroenterology, Kaiser Permanente San Francisco, San Francisco, California, USA.
Division of Gastroenterology, Northwestern University, Chicago, Illinois, USA.
Gastrointest Endosc. 2020 Aug;92(2):355-364.e5. doi: 10.1016/j.gie.2020.02.016. Epub 2020 Feb 21.
Although colonoscopy reduces colorectal cancer (CRC) risk, interval CRCs (iCRCs) still occur. We aimed to determine iCRC incidence, assess the relationship between adenoma detection rates (ADRs) and iCRC rates, and evaluate iCRC rates over time concomitant with initiation of an institutional colonoscopy quality improvement (QI) program.
We performed a retrospective cohort study of patients who underwent colonoscopy at an academic medical center (January 2003 to December 2015). We identified iCRCs through our data warehouse and reviewed charts to confirm appropriateness for study inclusion. iCRC was defined as a cancer diagnosed 6 to 60 months and early iCRC as a cancer diagnosed 6 to 36 months after index colonoscopy. We measured the relationship between provider ADRs and iCRC rates and assessed iCRC rates over time with initiation of a QI program that started in 2010.
A total of 193,939 colonoscopies were performed over the study period. We identified 186 patients with iCRC. The overall iCRC rate was .12% and the early iCRC rate .06%. Average-risk patients undergoing colonoscopy by endoscopists in the highest ADR quartile (34%-52%) had a 4-fold lower iCRC risk (relative risk, .23; 95% confidence interval, .11-.48) than those undergoing colonoscopy by endoscopists in the lowest quartile (12%-21%). After QI program initiation, overall iCRC rates improved from .15% to .08% (P < .001) and early iCRC rates improved from .07% to .04% (P = .004).
We confirmed that iCRC rate is inversely correlated with provider ADR. ADRs increased and iCRC rates decreased over time, concomitant with a QI program focused on split-dose bowel preparation, quality metric measurement, provider education, and feedback. iCRC rate measurement should be considered a feasible, outcomes-driven institutional metric of colonoscopy quality.
虽然结肠镜检查可降低结直肠癌（CRC）的风险，但仍会发生结直肠癌。我们旨在确定 CRC 的发生率，评估腺瘤检出率（ADR）与 CRC 发生率之间的关系，并评估随着机构结肠镜质量改进（QI）计划的启动，CRC 的发生率随时间的变化情况。
我们对在学术医疗中心接受结肠镜检查的患者进行了回顾性队列研究（2003 年 1 月至 2015 年 12 月）。我们通过我们的数据仓库确定了 CRC，并通过审查图表来确认研究纳入的适当性。CRC 定义为在指数结肠镜检查后 6 至 60 个月诊断出的癌症，早期 CRC 定义为在指数结肠镜检查后 6 至 36 个月诊断出的癌症。我们测量了提供者 ADR 与 CRC 发生率之间的关系，并在 2010 年开始的 QI 计划启动后评估了 CRC 发生率随时间的变化情况。
在研究期间共进行了 193939 例结肠镜检查。我们确定了 186 例 CRC 患者。总体 CRC 发生率为 0.12%，早期 CRC 发生率为 0.06%。在最高 ADR 四分位（34%-52%）中接受结肠镜检查的平均风险患者的 CRC 风险降低了 4 倍（相对风险，0.23；95%置信区间，0.11-0.48），而在最低四分位（12%-21%）中接受结肠镜检查的患者风险降低了 4 倍（相对风险，0.23；95%置信区间，0.11-0.48）。QI 计划启动后，总体 CRC 率从 0.15%降至 0.08%（P<0.001），早期 CRC 率从 0.07%降至 0.04%（P=0.004）。
我们证实 CRC 发生率与提供者 ADR 呈负相关。随着 QI 计划侧重于分剂量肠道准备、质量指标测量、提供者教育和反馈，ADR 增加，CRC 发生率随时间下降。CRC 发生率的测量应被视为结肠镜检查质量的可行、以结果为导向的机构指标。
