The Link-me pragmatic randomised controlled trial: Can a patient-completed decision support tool improve mental health outcomes in primary care?

Problem

Over- and under-treatment of mental health problems in primary care is common. Stepped care aims address this by matching patients to the least intensive intervention effective for their level of need. However, tools to easily match patients to an appropriate intervention in real-time are required. Link-me is a multifaceted, digitally supported, and systematic approach to stepped care. A patient-completed Decision Support Tool (DST) allocates patients according to their predicted depressive and anxiety symptom severity (minimal/mild, moderate, severe) in three months’ time, and provides treatment pathways matched to group allocation. This randomised controlled trial (RCT), examined whether Link-me improved psychological distress at 6 months, relative to usual care plus attention control, for the minimal/mild or severe groups.

Approach

Adult patients (18-75) attending 23 general practices in 3 Australian states were invited to complete the Link-me DST on a purpose-built online platform which included a randomisation function. The comparison arm received usual care plus attention control (information on community-based resources). The intervention arm received feedback on Link-me DST responses, treatment priority-setting, reflection on their motivation to change, and an evidence-based treatment recommendation (low intensity service options for the minimal/mild group and a collaborative care-based, motivational-interviewing informed intervention referred to as care navigation for the severe group). This presentation reports on patient-reported outcome measures completed online at 6 months. The primary outcome was psychological distress (Kessler Psychological Distress Scale [K10]). Secondary outcomes included depression, anxiety, and quality of life.

Findings

Of 15,474 patients screened, 7,985 were eligible; of those, 1,671 were allocated into the minimal/mild (n = 830) or severe symptom groups (n = 841) and randomised. Intention to treat analyses found that overall, Link-me resulted in greater reductions in psychological distress at 6 months than usual care, with a standardised mean difference (effect size) of -0.09 (95 % CI: -0.17 to -0.01). The intervention effect differed for the two symptom severity groups: 0.04 (95% CI: -0.17 to 0.24) for the minimal/mild symptom group and -0.26 (95% CI: -0.43 to -0.09) for the severe symptom group. Supplementary analyses showed increasingly large and more clinically meaningful effect sizes associated with delivery of additional elements of the intervention in the severe group.

Consequences

These preliminary findings suggest that the Link-me approach to matching interventions to patient need is feasible in primary care and can improve mental health outcomes, particularly amongst those with severe symptoms. Implementation of Link-me into routine practice could reduce unnecessary treatment burden and improve allocation of treatment resources. Future analysis will examine the longer-term effect of the intervention.