In a previous study (Felce, de Kock and Repp, 1986), appropriate client activity was found to be considerably higher in two community homes for severely and profoundly mentally handicapped adults than in a number of institutions. Client activity was also shown to be related to the extent of staff interaction. The method of this study was replicated in a two-year follow-up to investigate whether these desirable levels of staff interaction and client activity were maintained. Client engagement in appropriate activity remained similar for five subjects but had fallen slightly for a further five. The duration of instruction given to the majority of subjects by staff had declined but so too, for a variety of reasons, had staff–client ratios. Overall, the follow-up data give a general impression of the maintenance of client activity with some subjects engaging with greater independence. However, interpretation is dependent on the role that staff–client ratio may play in determining staff performance and in particular their level of interaction with clients.