A method was developed to investigate the transfer of bacteria from the hands of healthcare workers (HCWs). The method involved standardised hand contact between the HCW and a recipient wearing sterile gloves, followed by sampling of the bare hands of the HCW and the gloved hands of the recipient by the glove juice method. The duration of contact, degree of friction and dryness of the hands could be varied. We investigated the applicability of the method for measuring transfer from hands artificially contaminated with Escherichia coli as well as from naturally contaminated hands following a 30s contact time with moderate friction and dry hands. Only a small proportion of bacteria on donor hands was recovered from the recipient: 0.15% for E. coli and 0.07% for natural hand flora. A smaller proportion of E. coli was recovered from bare skin compared with gloves, suggesting reduced survival of bacteria as a result of contact with natural skin. We suggest that these data are clinically relevant, and may indicate low transfer of bacteria during short contact with dry hands. This method is suitable to investigate the effect of potential risk factors for ineffective hand hygiene and the effect of hand hygiene procedures on contact transmission in clinical studies with large numbers of HCWs.