Performance of Initial Screening Tests for Colorectal Cancer and Subsequent Adherence to Colonoscopy: An Ecological Study

Yihui Yang,Junjie Huang,Weimiao Wu,Hung N. Luu,Foong-Ming Moy,Songsong Tan,Jiongxing Fu,Tao Ying,Mellissa Withers,Dang The Hung,Dandan Mao,Sikun Chen,Martin CS Wong,Wanghong Xu
DOI: https://doi.org/10.2139/ssrn.3537081
2020-01-01
SSRN Electronic Journal
Abstract:Background: Adherence to colonoscopy in colorectal cancer (CRC) screening is at a low level and varies greatly across populations. No study has examined the influencing factors of colonoscopy adherence at a population level, and this remains to be an unaddressed knowledge gap. We therefore aimed to analyze the correlation between performance of initial CRC screening tests and subsequent adherence to colonoscopy base on an ecological evaluation. Methods: The sensitivity, specificity, and positive predictive value (PPV) of initial CRC screening tests and subsequent adherence to colonoscopy in different populations were extracted from relevant studies published in English up to December 31, 2018. The age-standardized incidences (ASR) of CRC across populations in the year of screening were obtained from the Cancer Statistics data. Locally weighted regression smoothing was used to examine the relationship between adherence to colonoscopy and ASR of CRC. Cut-off points of adherence on ASR were identified by Joinpoint Regression and used in Regression Discontinuity as potential thresholds. Spearman’s correlation analysis was used to calculate correlation coefficients. Findings: We identified 192 eligible articles. The median and interquartile range of adherence to colonoscopy was 82·5% (71·8%-90·5%). The adherence was significantly correlated with ASR of CRC in the study populations (r=0·187, p<0·001) and marginally with the PPV of initial screening tests (r=0·105, p=0·077). The change points of the adherence to colonoscopy were identified at the ASR of 32·5 and 142·5/100,000, and the ratio of adherence to colonoscopy at 143/100,000 was 0·900. In countries / areas with a moderate ASR, adherence to colonoscopy was negatively correlated with the sensitivity (r=-0·367, p=0·023) and positively with the PPV (r=0·143, p=0·040), whereas in populations with a high ASR, flexible sigmoidoscopy was more widely used as an initial screening test and achieved a high subsequent adherence to colonoscopy. Interpretation: Adherence to colonoscopy is correlated with the performance of initial screening tests, particularly among populations with moderate CRC incidence. Initial screening tests with high PPV or specificity may help to increase population adherence to colonoscopy, but different strategies should be used in countries / areas with a high or moderate incidence of CRC.Funding Statement: This study was supported by the Local High Level University Construction Project of Shanghai and the Health Commission of the Pudong New Area of Shanghai (PW2019A-5). Declaration of Interests: The authors declare that they have no competing interests.Ethics Approval Statement: Not required.
What problem does this paper attempt to address?