A study of laboratory based faecal occult blood testing in Melbourne, Australia

Faecal occult blood tests (FOBT) are widely used in clinical practice and are under increasing scrutiny as a tool for colorectal cancer screening. However, there is little information regarding the quality of testing performed in pathology laboratories. Therefore, we asked 13 pathology laboratories in Melbourne, Australia, to test coded contrived faecal samples prepared from a composite stool specimen which had been spiked to various concentrations of haemoglobin. The samples were provided to the laboratories in two forms: (i) on/in the sample collection device appropriate for the faecal occult blood test they normally used; and (ii) as a moist faecal sample. Some variation in threshold analytical sensitivity between laboratories for the same FOBT was observed for Hemoccult® SENSA®, ColoRectal®, Hematest®, MonoHaem® and Hemolex® suggesting that, at least for those tests, technician training could be improved. Two tests, Hematest® and an in-house FOBT did not perform as well as the other FOBT. When samples were sent in moist form, Hemoccult® SENSA® (P= 0.0002), ColoRectal (P= 0.02) and MonoHaem® (P= 0.04) had significantly lower overall positivity rates; for Hemolex® the decrease was not significant (P = 0.3). The lower positivity rate with moist samples is important, given that 11 of the 13 laboratories in the study stated that they receive at least some samples in moist form. Thus, technician training and laboratory procedure need to be reviewed to maximize the benefits of faecal occult blood testing in clinical practice, especially with its expanding role in colorectal cancer screening.