上海浦东新区大肠癌不同初筛模式肠镜检查顺应性

Colonoscopy compliance in high risk population identified by different screening modalities:colorectal cancer screening program in Pudong New Area of Shanghai

目的大肠癌筛查可降低大肠癌发病率和死亡率,但肠镜顺应性低直接影响筛查效果。本研究比较上海浦东新区不同初筛模式肠镜顺应性,以探讨提高肠镜顺应性的策略。方法以社区宣传和居民自愿参加的方式招募大肠癌筛查对象,排除年龄<50岁及>80岁者,最终以上海浦东新区2011-2017年591 769名50～80岁大肠癌筛查参加者为研究对象。所有对象均接受大肠癌危险度评估(问卷)和2次粪便隐血试验(fecal occult blood test,FOBT),问卷阳性或任意1次FOBT阳性者即判定为初筛阳性,建议其自行去医院进行肠镜检查。通过医院报病和随访获得检出结直肠癌病例信息,并通过肿瘤登记系统补充可能的漏诊病例。计算并比较不同筛查模式下阳性结果人群的肠镜顺应性,并采用Logistic回归模型及Spearman等级相关寻找可能的影响因素。结果共检出问卷阳性者60 226人(10.18%),任意1次FOBT阳性76 626人(12.95%),两者均阳性仅11 304人(1.91%)。125 548名初筛阳性者中,仅28 105人(22.39%)接受了肠镜检查。共检出结直肠癌526例,补充可能漏诊病例607例。初筛灵敏度为73.96%,特异度为78.89%,阳性预测值为0.67%。危险度评估结果与FOBT结果一致性较低,Kappa系数仅为0.057 8,χ2=2 354.3,P<0.001。FOBT结果对肠镜顺应性影响最大,与阴性组相比,仅1次阳性者肠镜检查OR为27.92(95%CI:25.64～29.40),2次均阳性者OR为38.27(95%CI:35.77～41.91)。问卷和FOBT串联筛查方法中,"仅问卷阳性"者肠镜顺应性最差(12.08%),"2次FOBT均阳性"者最高(37.68%),χ2=125 922.0,P<0.001。肠镜顺应性与初筛模式灵敏度(r=0.30,P=0.62)无统计学意义相关性,但随特异度(r=0.80,P=0.10)和阳性预测值(r=1.0,P<0.001)升高而上升。结论上海社区大肠癌初筛阳性者肠镜顺应性低,需加强大肠癌防控知识宣教,科学使用问卷评估结果,改善初筛方案特异度及阳性预测值,有效提高肠镜顺应性。

OBJECTIVE Colorectal cancer screening has been suggested to decrease incidence and mortality of the malignancy.The effect of the screening,however,is affected by low compliance to colonoscopy examination.This study was designed to compare colonoscopy compliance in populations under different screening modalities,identify related risk factors for low compliance and thus provide related policy implications.METHODS A colorectal cancer screening program was conducted in Pudong New Area of Shanghai,China,through community mobilization.A total of 591 769 volunteers aged 50-80 years participated in the program during the period of 2011-2017.All subjects had a questionnaire-based risk assessment and two fecal occult blood test(FOBT),then those with a positive result in either method were urged to have a colonoscopy examination by themselves.Colorectal cancer cases were reported by hospitals and identified by follow-up or record linkage with the Shanghai Cancer Registry.Colonoscopy compliance in participants with positive results under different screening modalities were calculated.Logistic regression and Spearman rank correlation were used to identify potential influential factors for colonoscopy compliance.RESULTS A total of 60 226 subjects(10.18%)were positive in risk assessment and 76 626(12.95%)were positive in one or two FOBT,but only 11 304(1.91%)were positive in both methods.Of 125 548 with a positive result(either in risk assessment or in FOBT),28 105(22.39%)had a colonoscopy.Totally526 colorectal cancer cases were detected and additional 607 were identified through the record linkage.The sensitivity,specificity and positive predictive value(PPV)were 73.96%,78.89%and 0.67%respectively for the initial screening modality used in the program.Low consistency was observed between risk assessment and FOBT results,with a Kappa coefficient of 0.057 8(χ2=2 354.3,P<0.001).Colonoscopy compliance was strongly associated with the results of FOBT.Compared with the FOBT negative group,the OR and 95%CI was 27.92(25.64-29.40)for those with one positive result and 38.27(95%CI:35.77-41.91)for those with two positive results.Colonoscopy compliance was observed the lowest in subjects positive only in risk assessment(12.08%),but the highest in those positive in two FOBTs(37.68%)(χ2=125 922.0,P<0.001).Colonoscopy compliance was not correlated with sensitivities(r=0.30,P=0.62)of screening modalities,but increased with increasing specificities(r=0.80,P=0.10)and PPVs(r=1.00,P<0.001).CONCLUSION The results suggest that low colonoscopy compliance,the main challenge for cancer screening program in Shanghai,may be improved by health education,proper use of questionnaire-based risk assessment,increased specificities,and PPVs of screening tests.