In order to evaluate the accuracy and reproducibility of determination of hepatocyte growth factor (HGF) levels in faeces, the stability of HGF in samples processed in different ways was investigated. An ELISA method was used for determination of HGF concentrations. Faeces samples from healthy controls and patients with infectious diarrhoea were studied. It was found that faeces HGF concentration remained stable irrespective of whether samples were freeze‐thawed several times, kept for 6, 12 or 24 h at room temperature or refrigerated for 6, 12, 24 or 36 h; the levels of HGF did not change significantly when samples were freeze‐dried. Adding protease inhibitor to the faeces samples did not affect the HGF levels. There were no significant differences between HGF levels using phosphate buffered saline (PBS) (pH 7.4) or NaCL as buffer, but it was observed that levels of HGF were significantly lower in the samples that were diluted in distilled water. Although both HGF and albumin through various mechanisms may increase in faeces during infectious diarrhoea, there was no significant correlation between faeces HGF levels and albumin levels, which might indicate local production of HGF in the bowel in response to infection. It is concluded that determination of faeces HGF levels is feasible with a high degree of stability. Increased HGF levels in faeces might represent a local production of HGF during bowel injury and might be of use as a diagnostic and monitoring assay.