Aims The aims of this study were (1) to determine the measurement accuracy of a widely used guaiac faecal occult blood test (gFOBT) compared with an immunochemical faecal occult blood test (iFOBT) during in vitro studies, including their analytical stability over time at ambient temperature and at 4°C; and (2) to compare analytical imprecision and other characteristics between two commercially available iFOBT methods.

Methods Faecal specimens from healthy volunteers negative for occult blood were spiked with red cell lysate to give concentrations of 0 (A), 2.5 (B) and 4.5 (C) mg Hb/g faeces respectively. Samples from each pool were then tested by nine blinded assessors in order to determine the measurement accuracy. Sample stability for the gFOBT at ambient temperature (18°C) was determined by repeating the gFOBT analysis on faecal samples (A, B and C) consecutively for 7 days. Stability for the iFOBT was tested on spiked faecal samples stored at 4°C and also at ambient temperature (18°C). Testing for other analytical performance characteristics including sample carryover and imprecision was performed for both iFOBT methods.

Results Nine blinded assessors using gFOBT achieved correct readings for eight of nine samples from Pool A, five of nine from Pool B and seven of nine from Pool C. Overall, the sensitivity and specificity of gFOBT were 67% and 89% respectively. No discrepant results were detected with iFOBT. Faecal samples applied to gFOBT cards immediately postcollection gave positive results for the next 7 days only for pool C. In contrast, the results of iFOBT remained stable up to 14 days at 18°C. The within-run imprecision and sample carryover showed robust results with both iFOBT methods.

Conclusions The superior analytical stability and measurement accuracy demonstrated by the iFOBT in this in vitro study confers advantages over traditional qualitative gFOBTs and supports their suitability for more widespread use in population-based colorectal-cancer screening programmes.