作者: Lawrence F. Paszat , Linda Rabeneck , Nancy N. Baxter , Meredith A. Goldwasser , Refik Saskin
DOI:
关键词: Internal medicine 、 Odds ratio 、 Colonoscopy 、 Oncology 、 Conditional logistic regression 、 Colorectal cancer 、 Case-control study 、 Medicine 、 Population based 、 Administrative claims 、 Index date
摘要: Measurements: Administrative claims data were used to detect exposure any colonoscopy and complete (to the cecum) from January 1992 an index date 6 months before diagnosis in each case patient same assigned matched controls. Exposures patients controls compared by using conditional logistic regression control for comorbid conditions. Secondary analyses done see whether associations differed site of primary CRC, age, or sex. Results: 10 292 51 460 identified; 719 (7.0%) 5031 (9.8%) had undergone colonoscopy. Compared with controls, less likely have attempted (adjusted odds ratio [OR], 0.69 [95% CI, 0.63 0.74; P 0.001]) OR, [CI, 0.57 0.69; 0.001]). Complete was strongly associated fewer deaths left-sided CRC 0.33 0.28 0.39]) but not right-sided 0.99 0.86 1.14]).