Performing the test: dietary restrictions and number of samples

In order to reduce the probability of a false positive result, dietary restrictions are usually recommended when guaiac-based tests are used. Retesting of subjects with a positive test (possibly with dietary restrictions being recommended) represents an alternative option adopted in some programmes to deal with this problem. A review of 5 trials (10 359 participants overall) comparing Guaiac FOBT with and without dietary restriction found a significant difference in compliance in favour of testing without dietary restrictions only in the trial where restrictions were particularly extensive. Authors concluded that advice to restrict the diet and avoid NSAIDs and vitamin C does not substantially reduce completion rate except perhaps when the dietary restrictions are particularly extensive (Pignone et al. 2001). More recent randomised trials (Cole et al. 2003; Federici et al. 2005; van Rossum et al. 2008) have demonstrated that better compliance can be achieved using iFOBT compared to a guaiacbased test. These results are not explained by the nature of the test but by lack of dietary and drug restrictions and easier and more pleasant sampling methods. Indeed, dietary restriction was associated with a significant decrease in participation also among people offered iFOBT test, compared to controls receiving the same test who where not advised to control their diet

Summary of evidence 

 Compliance is affected by dietary restriction and number of stool samples to be collected. Compliance is found to be consistently higher when the test adopted does not require modification of a subject’s diet and sampling is limited to one bowel movement (I).  

Examination of the samples, test interpretation and reporting 

Detailed protocols on handling the stool samples must be available and followed. Identification and tracing of the sample through the entire process should be ensured by adopting appropriate labelling allowing the sample and patient’s ID code to be linked. Automated check protocols should be implemented in order to avoid mismatching of the results. All data, including test results, should have a regular backup system.

An operational definition for an inadequate screening test should be made explicit in the programme protocol, taking into account the characteristics of the test (i.e. the stability and the storage requirements of the tests) as well as the testing procedure adopted

Protocols should be in place to define the appropriate test and the algorithm used to classify a test result (as negative or positive). For quantitative or semi-quantitative iFOBTs, an explicit definition of cut-off levels for haemoglobin concentration should be defined. Protocols or rules for combining results when using multiple samples, the number of samples that are needed to evaluate the test result, etc. must be in place. When using a quantitative test, provision should be made to record the information concerning the actual amount of haemoglobin, both for tests classified as negative and for those classified as positive.

Some people may present with clinical conditions such as inflammatory bowel disease (Crohn’s disease or haemorrhagic recto-colitis), which may explain a positive FOBT result. In such cases, if no cancers were detected, then the screening result should be classified as negative for the purposes of the screening programme. These patients should then be referred for treatment in the appropriate clinical setting.