Abstract
Background: Studies of assertive community treatment (ACT) have shown various benefits, including reduced hospital bed use. In the UK, this finding was not replicated by randomised controlled trials (RCTs), which lacked fidelity to the model. Conversely, observational studies, while limited by their inherent weakness in implying causality, have shown lower bed use. Against this background many ACT teams are being disestablished in the UK.
Aims: To observe the long-term effect of ACT on bed use, incorporating methods of analysis which mitigate against some weaknesses of observational design.
Methods: Bed use was compared for equal periods of time either side of starting support from an ACT team.
Results: Ninety-three people were followed for up to 10.5 years after starting ACT. Hospital bed use was compared for each person, showing a reduction from a mean of 72 d per year prior to ACT to 44 d per year during ACT (p = 0.0018).
Conclusions: The results demonstrate that ACT is associated with reduced bed use in the UK and that it is possible to use an observational design with enhanced analysis techniques to increase evidence for causality. These techniques may have value in other service evaluations.