There is an emerging international literature demonstrating clinical and cost-effectiveness of sub-acute residential mental health services. To date, however, there is limited information on the profile of consumers accessing these models of care. This study aimed to understand the profile of the population served by adult sub-acute residential mental health services in Victoria, Australia (known as Prevention and Recovery Care; PARC) and to compare PARC service consumers with consumers admitted to psychiatric inpatient units within public hospitals.
Using 5 years (2012-2016) of a state-wide database of routinely collected individual level mental health service data, we describe the socio-demographic and clinical profile of PARC service consumers compared to consumers of psychiatric inpatient units including for primary diagnosis and illness severity. Using admissions as the unit of analysis, we identify the characteristics that distinguish PARC service admissions from psychiatric inpatient admissions. We also examine and compare length of stay for the different admission types.
We analysed 78,264 admissions representing 34,906 individuals. The profile of PARC service consumers differed from those admitted to inpatient units including for sex, age, diagnosis and illness severity. The odds of an admission being to a PARC service was associated with several socio-demographic and clinical characteristics. Being male or in the youngest age grouping (< 20 years) significantly reduced the odds of admission to PARC services. The presence of primary diagnoses of schizophrenia and related disorders, mood, anxiety or personality disorders, all significantly increased the odds of admission to PARC services. Predictors of length of stay were consistent across PARC and inpatient admission types.
Our findings suggest PARC services may serve an overlapping but distinguishably different consumer group than inpatient psychiatric units. Future research on sub-acute mental health services should be cognizant of these consumer differences, particularly when assessing the long-term effectiveness of this service option.