Trial based cost-effectiveness evaluation of Link-me: A systematic approach to stepped mental health care in primary care

Problem

Stepped care has been promoted to reduce health care costs without compromising outcomes in the management of mental health. However, there is limited empirical evidence to support the cost-effectiveness of stepped care for mental health treatment in primary care. One challenge to its implementation is the identification of patient need to allow timely matching to appropriate interventions. Link-me is a multifaceted, digitally supported, systematic approach to stepped care beginning with a patient-completed Decision Support Tool (DST) allocating patients into two treatment pathways (low intensity services, care navigation) based on predicted depressive and anxiety symptom severity (minimal/mild, severe) in 3 months. We conducted an economic evaluation within a parallel, stratified individually randomised controlled trial (RCT) to evaluate the costs and outcomes of Link-me compared to usual care for patients in the minimal/mild and severe groups.

Approach

The RCT screened 15,474 adults attending general practices in 3 Australian states. Health sector costs included the DST, care navigation and other health care services used by participants during the trial. Societal costs included health sector costs plus lost productivity. DST and care navigation cost were determined through provider records. Other health care service use (i.e. hospitalisations) and lost productivity were captured through a self-report resource use questionnaire at 6-month follow-up. Standard Australian unit costs were applied (2018/2019 AUD). Outcomes included psychological distress [Kessler Psychological Distress Scale (K10)] and quality adjusted life years [EQ-5D-5L, Australian value set]. Incremental cost-effectiveness ratios were calculated as the difference in average costs between the Link-me and usual care groups, divided by the difference in average outcome with nonparametric bootstrapping to calculate 95% confidence intervals.

Findings

Among 1,671 participants (830 minimal/mild and 841 severe), Link-me resulted in greater reductions in psychological distress at 6 months than usual care (standardised mean difference -0.10 (95% CI: -0.18 to -0.01). Link-me was associated with higher mean health sector costs than usual care of $24 (95% CI $8.3 to $43.8) per person across all participants, and $333 (95% CI $125 to $623) per person for the severe symptom group. Across all participants, we observed an incremental cost per 1-point decrease in K10 score of $1,018 (95% CI $259 to $10,471) from the health sector perspective and $1,282 (95% CI Dominant to $21,964) from the societal perspective. For the severe symptom group the incremental cost per 1-point decrease in K10 score was $896 (95% CI 234 to 3,978) from the health sector perspective and $1,359 (95% CI Dominant to 8,677) from the societal perspective.

Consequences

These preliminary findings suggest the Link-me approach led to improvements in clinical outcomes with higher costs. A planned longer-term follow up evaluation will examine whether improvements in outcomes and trends in cost differences are maintained.