Advanced neoplasia detection in the fourth round of colorectal cancer screening appears to be highly dependent on prior round adherence behavior.
Evidence indicates that consistent participation in colorectal cancer (CRC) screening, including fecal immunochemical testing (FIT), is vital to maintaining a proactive preventative strategy. Studies suggest that long-term adherence may also forecast advanced neoplasia (AN) identification in future screenings. The challenge has been to articulate the most effective, simplified measures to predict AN in patterns of longitudinal adherence among screened patients.
Simplified Measures
For a study published in Preventative Medicine, L. de Jonge, PhD candidate, and colleagues examined FIT screening results in a Dutch pilot study, a program in the Piedmont region of Italy, and a program in Reggio Emilia province of Italy to observe the detection rate of AN in the fourth round of a patient’s screening regimen. The five simplified measures applied to this examination were:
- Adherence to previous screening invitations
- Consistency of screenings
- Frequency of screenings
- Frequency plus adherence to previous invitations
- Quantity of invitations covered
As written by de Jonge and colleagues, “The aim of this study was to use real-time data to compare multiple measures of longitudinal adherence and determine the best simplified measure to predict the participation rate, positivity rate, and detection rate of AN in subsequent screening round in FIT-based CRC screening programs.”
Study participants were drawn from closed cohorts, had been invited to four prior screening rounds, and did not have positive FIT results in the prior three rounds. Variations of the examined patterns of participation were developed using all possible configurations of screening compliance. Predictive performance of the variations was determined by applying chi-square deviance.
Identifying Advanced Neoplasia
The identification rate of AN per 1,000 participants was 5.1 in the Dutch study, 3.4 in the Piedmont program, and 3.0 in the Reggio Emilia program (Table). Among participants who attended screening once, in their fourth screening round, AN was identified in the Dutch study between 2.3 and 4.6 per 1,000, between 1.5 and 2.4 in the Piedmont program, and between 0.9 and 1.5 in the Reggio Emilia program. An increase to attending two screenings identified AN in the fourth screening with a variation between 1.6 and 2.1 in the Dutch study, between 1.3 and 1.7 in the Piedmont program, and between 1.0 and 1.5 in the Reggio Emilia program. For those who participated in screenings consistently, the AN identification rate in the fourth screening was 0.015 in the Dutch study, 1.0 in the Piedmont program, and 1.0 in the Emilia Reggio program.
The simplified measure adherence to previous screening invitations estimated an identification rate of AN in the fourth screening round per 1,000 participants of 4.4 in the Dutch study, 2.6 in the Piedmont program, and 2.0 in the Reggio Emilia program if the participant had not acted upon their prior invitation for screening. If the participant had responded to their prior invitation to screening, rates were 1.6 in the Dutch study, 1.1 in the Piedmont program, and 1.1 in the Reggio Emilia program.
Frequency & Adherence
In applying the simplified measure of frequency plus adherence, the estimated identification rate of AN in the fourth screening was between 1.0 and 4.5 in the Dutch study, between 1.3 and 2.0 in the Piedmont program, and between 1.3 and 1.5 in the Reggio Emilia program. The researchers note that the lowest deviance rate for the identification of AN in the fourth round of screening occurred when applying the frequency plus adherence simple measurement.
de Jonge and colleagues wrote, “Our findings suggest that both recent adherence ‘adherence previous invitation’ and adherence to prior screening ‘frequency’ are determinants of participation, positivity. and AN detection in the subsequent screening round… FIT positivity and AN detection in the second round was almost doubled among previous non-attenders compared to attenders. This supports the fact that repeated screening is essential in FIT-based CRC screening programs.”