System-level intersectoral linkages between the mental health and non-clinical support sectors: a qualitative systematic review

Objectives: Concerns about fragmented mental health service delivery persist, particularly for people with severe and persistent mental illness. The objective was to review evidence regarding outcomes attributed to system-level intersectoral linkages involving mental health services and non-clinical support services, and to identify barriers and facilitators to the intersectoral linkage process.

Methods: A systematic, qualitative review of studies describing attempts to coordinate the activities of multiple service agencies at the policy, program or organisational level was conducted. Electronic databases Medline, PsycINFO and EMBASE were searched via OVID from inception to July 2012.

Results: Of 1593 studies identified, 40 were included in the review - 26 in adult and 14 in vulnerable youth populations. Identified mechanisms to promote positive system-level outcomes included: interagency coordinating committees or intersectoral/interface workers engaged in joint service planning; formalised interagency collaborative agreements; a single care plan in which the responsibilities of all agencies are described; cross-training of staff to ensure staff culture, attitudes, knowledge and skills are complementary; service co-location; and blended funding initiatives to ensure funding aligns with program integration. Identified barriers included: adequacy of funding and technology; ensuring realistic workloads; overcoming 'turf issues' between service providers and disagreements regarding areas of responsibility; ensuring integration strategies are implemented as planned; and maintaining stakeholder enthusiasm.

Conclusions: System-level intersectoral linkages can be achieved in various ways and are associated with positive clinical and non-clinical outcomes for services and clients. Some linkage mechanisms present greater implementation challenges than others (e.g. major technology upgrades or co-location in geographically remote areas). In some instances (e.g. co-location) alternative options may achieve equivalent benefits. Publication bias could not be discounted, and studies using high-quality research designs are scarce. The limited information base applicable to system-level integration argues strongly for the evaluation of the models that evolve in the rollout of the national Partners in Recovery initiative.