Community Over Clinic: The RECOVER Trial Puts a Number on What We Already Suspected
- RCT, n=891 (506 meeting SMI/SPMI criteria), Hamburg: the RECOVER model — Assertive Community Treatment (ACT) + Crisis Resolution Teams (CRT) — reduced total societal costs by €6,681 per patient compared to treatment as usual (p<0.01) over 12 months.
- Outpatient costs increased by €1,593, but were more than offset by inpatient savings of €6,382 — the economic logic of community psychiatry in one number.
- The ICER showed RECOVER to be *dominant*: more effective and less costly simultaneously; cost-effectiveness probability exceeded 99% across all willingness-to-pay thresholds tested.
- Psychosocial functioning improved in the RECOVER group relative to TAU — better outcomes at lower cost from a societal perspective.
The debate about inpatient versus community-based psychiatric care is decades old. What has been lacking is an adequately powered RCT that quantifies the cost differential across a societal perspective — not just direct healthcare costs, but productivity losses, social care, and the full economic footprint of severe mental illness. The RECOVER trial from Hamburg provides exactly that data, and the finding is unambiguous: community-based assertive care costs less and works better.
What the Economics Show
The RECOVER model combines two interventions that are evidence-based separately but rarely studied together at scale: Assertive Community Treatment (ACT), which provides intensive outreach and case management for the most severely ill, and Crisis Resolution Teams (CRT), which provide rapid home-based crisis support as an alternative to emergency hospitalization. In the RECOVER trial, 506 patients meeting criteria for severe and persistent mental illness (SMI/SPMI) were randomized. Both arms received care; the RECOVER arm received these additional structured community services.
The headline figure — €6,681 lower societal cost per patient per year — comes from a simple arithmetic that mental health policymakers have been slow to act on: preventing one inpatient admission (average savings €6,382) more than pays for the increased outpatient investment (€1,593). The ICER analysis goes further: RECOVER doesn't just break even on cost — it dominates on both dimensions simultaneously. A 99%+ probability of cost-effectiveness means this is not a borderline finding.
Psychosocial functioning improvement adds the clinical outcome component. RECOVER patients showed significantly better functional outcomes than TAU at 12-month follow-up. This matters because SMI/SPMI patients are often excluded from functional outcome studies on the assumption that their prognosis is uniformly poor. RECOVER challenges that assumption directly.
For Your Practice
For clinicians in systems that still rely heavily on inpatient care for SMI patients, this trial provides economic ammunition for service restructuring. The data directly supports the argument that outpatient investment in ACT and CRT pays for itself through avoided hospitalizations — not eventually, but within a 12-month window.
The practical challenge is implementation. ACT requires fidelity to a demanding model: small caseloads (typically 1:10 ratio), 24/7 availability, multidisciplinary team structure, and active outreach rather than passive office-based care. CRT requires rapid response capacity that most outpatient teams lack. Neither is cheap to set up. But the RECOVER data makes the business case that these upfront structural costs are recovered through reduced inpatient utilization within the fiscal year.
Preventing one psychiatric hospitalization saves more than an entire year of intensive community outreach — RECOVER proves the math holds at the system level.
Hamburg healthcare system may not generalize to settings with different hospitalization costs or social care structures. The 12-month follow-up window may not capture all cost dynamics. TAU varied across the trial period.