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

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