Working With ICBs in Community Mental Health: What Providers Need to Get Right
Working well with an Integrated Care Board (ICB) is not primarily about relationship management; it is about predictable operational practice that commissioners can trust. ICBs need clarity on how your service fits within the local system, how risk is shared and escalated, and how performance is measured without creating avoidable reporting burden. This article sits within working with ICBs in mental health and links to mental health service models and pathways, focusing on what providers must get right in day-to-day governance, assurance, and evidence so contract monitoring and mobilisation checks are straightforward rather than adversarial.
How ICBs typically assess providers in practice
ICBs will often talk about “partnership working”, but contract assurance tends to concentrate on a small set of practical tests:
- System fit: can you describe your service boundaries, referral criteria, interfaces and escalation routes in a way that matches the local pathway reality?
- Control of risk: can you evidence timely safeguarding action, proportionate risk management, and least restrictive practice where restrictions are relevant?
- Performance credibility: do your reported metrics reconcile to case files and operational logs, and can you explain variation across cohorts/localities?
- Governance maturity: do incidents lead to learning and verified improvement, and can you show action tracking and re-audit?
The biggest avoidable failure is presenting a “model” that sounds coherent but cannot be evidenced when commissioners sample cases, escalation logs and governance minutes.
Operational foundations for strong ICB working
1) Define boundaries and interfaces so escalation is unambiguous
Providers should be able to evidence, in plain operational terms, what the service does and does not do, and how people move between teams (including crisis interfaces, step-up/step-down, discharge transitions, and multi-agency safeguarding). This is not just a pathway document; it must be reflected in care plans, staff decision-making and recorded escalation actions.
2) Align your assurance pack to the ICB’s contract questions
A good assurance pack is stable and repeatable. It typically includes: service activity and responsiveness; safeguarding and serious incident themes; escalation patterns; complaints and learning; audit results on non-negotiables; workforce stability and supervision coverage; and a small set of impact measures that can be traced back to routine records. The goal is “auditable confidence”, not volume.
3) Use shared risk language and clear escalation routes
ICBs are often managing system pressure across Trusts, VCSE providers, local authority interfaces and primary care. Providers build trust by being explicit about shared risk: what you can control directly, what depends on partner response, and how interface risks are raised and tracked. This is where mature providers differ: they document interface issues and show follow-through, rather than informally “chasing” and hoping for the best.
Operational examples (what good looks like on the ground)
Example 1: Mobilisation and early contract assurance for a new locality
Context: A provider mobilises a community mental health support service across two localities. The ICB is concerned about uneven access, referral drift, and whether staff understand escalation thresholds from day one.
Support approach: The provider creates a mobilisation control plan with three operational deliverables: (1) a referral and eligibility decision guide aligned to the local pathway; (2) an escalation and safeguarding decision framework; (3) a simple quality audit tool for the first 12 weeks focused on non-negotiables (baseline, goals, risk plan, escalation routes, review decisions).
Day-to-day delivery detail: For the first month, team leads hold daily short huddles reviewing new referrals, rejected referrals (with reasons), and any cases escalating. Each week, the manager samples a small set of new starts to confirm that care plans reflect pathway boundaries and escalation logic. Where drift is seen (for example, referrals outside scope), the provider records the pattern and agrees a corrective action with the ICB (refresher briefing for referrers, updated referral form prompts, or clearer triage messaging).
How effectiveness/change is evidenced: The ICB sees stable acceptance criteria, reduced inappropriate referrals over time, and consistent early evidence trails in case files. Evidence includes mobilisation logs, sampling results, and governance minutes showing how issues were detected and resolved.
Example 2: Managing shared risk at the crisis interface
Context: A cohort experiences repeated crisis escalation. Staff report that crisis response thresholds and access routes vary by time and locality, creating inconsistency and frustration. The ICB wants assurance that the provider is not “handing off” risk or delaying escalation.
Support approach: The provider implements a crisis-interface protocol that includes: early warning indicators recorded at each contact for high-risk cases; step-up thresholds and actions; a standard escalation summary format (what changed, what actions have been tried, current risks, and what is being requested); and a manager review step for repeat escalations.
Day-to-day delivery detail: Staff record early warning indicators and actions taken (not just discussion). When escalation is required, the standard summary is used so partners receive consistent information quickly. Managers review all repeat escalations weekly and identify patterns: delays, unclear thresholds, or partner access issues. Where access issues persist, the provider escalates formally through agreed routes and logs outcomes, so shared risks are visible and trackable.
How effectiveness/change is evidenced: Fewer late-stage crises, improved timeliness of step-up actions, and clearer documentation of escalation decisions. Evidence is triangulated from escalation logs, file sampling and recorded interface escalations with outcomes.
Example 3: Safeguarding coordination across system partners
Context: Exploitation and self-neglect concerns increase. The ICB is concerned about safeguarding timeliness, variable protection planning, and whether restrictive responses are becoming default.
Support approach: The provider introduces a safeguarding operating rhythm: weekly safeguarding huddles for active cases, a decision tree for thresholds and “what to do today” actions, and a least restrictive check where restrictions are considered. A monthly audit tests referral timeliness, action completion and review cadence.
Day-to-day delivery detail: Each safeguarding case has named actions with deadlines. Team leads check completion and record multi-agency contact and outcomes. Supervision uses scenarios to test threshold understanding and to ensure staff can explain proportionality decisions. Where partner response is slow, escalation routes are used and logged, ensuring shared risk is visible to the ICB without becoming blame-driven.
How effectiveness/change is evidenced: Faster referrals, higher completion of safeguarding actions, fewer repeat concerns without learning, and reduced long-running restrictions. Evidence includes huddle records, audit outputs, and sampled case evidence trails.
Explicit expectations that must be met
Commissioner expectation
ICBs expect auditable assurance, not narrative assurance. Providers should be able to reconcile reported performance to case files and operational logs, explain variation, and evidence action tracking and verification (re-audit/sampling) when issues are identified. They also expect interface risks to be raised and managed through agreed escalation routes.
Regulator / Inspector expectation (e.g. CQC)
CQC expects safe, person-centred practice delivered consistently, including through system interfaces. Inspectors will triangulate staff understanding, records, safeguarding responsiveness and learning from incidents. They will test proportionality, least restrictive practice where relevant, and whether governance translates into day-to-day control rather than paperwork.
Making ICB working easier over time
The most reliable way to improve commissioner relationships is to reduce uncertainty: stable standards, consistent evidence trails, and quick correction when drift appears. If you can show how your service model operates in reality — including interfaces, escalation, learning and verification — contract management becomes a structured conversation about improvement rather than a debate about credibility.