BACKGROUND This study compared long-term outcome of serious mental illness in two states using a matched design to assess psychiatric rehabilitation programmes; Vermont subjects participated in a model psychiatric rehabilitation programme, while the Maine group received more traditional care.
METHOD Maine and Vermont subjects (n = 269) were matched by age, sex, diagnosis, and chronicity. Demographic, illness, and life history information were abstracted from hospital records by clinicians blind to outcome. DSM-III criteria were applied retrospectively. Outcome was assessed by clinicians blind to history.
RESULTS Vermont subjects alive at follow-up (n = 180) were more productive (P < 0.0009), had fewer symptoms (P < 0.002), better community adjustment (P < 0.001) and global functioning (P < 0.001) than Maine subjects (n = 119).
CONCLUSIONS Outcome differences may be due to Vermont's model programme and a policy of allowing an earlier opportunity for community life.