中国成年人排便频率与结直肠癌发病风险的前瞻性关联分析

目的 分析排便频率与结直肠癌发病风险的前瞻性关联。方法 研究对象来自于中国慢性病前瞻性研究，基线调查开展的时间为2004年6月至2008年7月，在剔除基线时自报患有恶性肿瘤的个体后，最终有510 134人纳入分析。本研究使用的随访数据截止到2016年12月31日，采用Cox比例风险回归模型计算排便频率和结直肠癌发病的关联。结果 研究对象平均随访9.9年，随访期间3 056人发生结直肠癌；在分癌种分析时，纳入结肠癌1 548例，直肠癌1 475例。控制潜在混杂因素后，与排便频率基本1次/d者相比，排便频率>1次/d者发生结直肠癌、结肠癌和直肠癌的HR值分别为1.24（95% CI：1.12~1.39）、1.12（95% CI：0.95~1.31）和1.37（95% CI：1.18~1.59）。按随访时间进行亚组分析，随访前5年对应的HR值分别为1.59（95% CI：1.36~1.86）、1.43（95% CI：1.14~1.80）和1.76（95% CI：1.41~2.19）；随访5年后，上述关联均无统计学意义（交互P<0.05）。与排便频率基本1次/d者相比，排便频率更低者的发病风险无统计学意义。结论 排便频率>1次/d者在随访前5年发生结直肠癌的风险升高。异常增加的排便频率作为一个容易识别的症状，应该在自我健康管理和结直肠癌早期筛检的实践工作中加以重视。

Association between the frequency of bowel movements and the risk of colorectal cancer in Chinese adults

Objective To examine the association between the frequencies of bowel movement (BMF) and the risk of colorectal cancer (CRC). Methods In this study, 510 134 participants from the China Kadoorie Biobank (CKB) were included, after excluding those who reported as having been diagnosed with cancer at the baseline survey. The baseline survey was conducted from June 2004 to July 2008. The present study included data from baseline and follow-up until December 31, 2016. We used the Cox proportional hazards regression models to estimate the HR and the 95% CI of incident CRC with BMF. Results During an average follow-up period of 9.9 years, 3 056 participants were documented as having developed colorectal cancer. In the site-specific analysis, 1 548 colon cancer and 1 475 rectal cancer were included. Compared with participants who had bowel movements on the daily base, the multivariable-adjusted HR (95% CI) for those who had more than once of BMF were 1.24 (1.12-1.39) for CRC, 1.12 (0.95-1.31) for colon cancer, and 1.37 (1.18-1.59) for rectal cancer. We further examined the association between BMF and CRC, according to the stages of follow-up, the corresponding HR (95% CI) for CRC, colon and rectal cancer were 1.59 (1.36-1.86), 1.43 (1.14-1.80), and 1.76 (1.41-2.19) for the first five years, while such associations became statistically insignificant in the subsequent follow-up (P for all interactions were <0.05), as time went on. As for CRC, colon or rectal cancers among participants who had lower bowel movements, the risks were not significantly different from those who had bowel movements everyday. Conclusions Participants who had BMF more than once a day, appeared an increased risk of CRC in the subsequent five years. Since abnormal increase of bowel movements is easily recognizable, programs should be set up on health self-management and early screening for CRC.