Building Long-Term System Partnerships With ICBs and NHS Trusts
Long-term system partnerships are built by what happens repeatedly in day-to-day operations: whether interfaces work, whether escalation is timely, whether disputes are resolved fairly, and whether learning leads to verified improvement. Commissioners notice stability: fewer repeat interface failures, predictable assurance, and transparent handling of shared risk. Providers that succeed long-term make partnership “operational”, not rhetorical. This article sits within working with ICBs in mental health and connects to mental health service models and pathways, focusing on the practical partnership habits that sustain commissioning confidence and safe delivery over multiple contract cycles.
What makes partnerships “long-term” in commissioning reality
In commissioning terms, long-term partnership is not a promise; it is a pattern of evidence across time. ICBs tend to renew confidence when a provider can show:
- Predictable governance: stable rhythms of reporting, sampling, learning and action tracking.
- Interface improvement: recurring problems (referral bounce, delayed escalation) reduce over time due to verified changes.
- Transparent risk ownership: shared risks are logged, escalated and managed without blame-driven culture.
- Credible performance: data reconciles to records and variation is explained with operational evidence.
- Rights-respecting practice: restrictions (where used) are proportionate, reviewed and stepped down.
These factors reduce commissioner burden: fewer crises in contract management, fewer surprises, and clearer assurance.
Partnership “habits” that sustain trust
1) A shared operating rhythm that combines data and case evidence
Effective partnerships agree a rhythm: monthly operational review, quarterly quality and learning review, and periodic deep dives into interface themes. Crucially, the rhythm combines trend data with a small set of case-based assurance traces so the discussion stays grounded and audit-ready.
2) Joint ownership of interface problems and redesign
Interface problems are inevitable. What matters is whether partners redesign processes. Mature providers bring evidence of interface failure patterns (for example, declined referral reasons, delayed escalation steps) and propose practical fixes: minimum datasets, clearer thresholds, revised escalation ladders, and staff briefings. They then verify change through re-audit.
3) Dispute resolution routes that protect safety
Contested thresholds will happen. Long-term partnerships protect safety by having a dispute route that is timely and recorded, paired with interim safety planning while disputes are resolved. This prevents “standoffs” where risk rises during disagreement.
4) Workforce stability and consistent supervision at interfaces
Partnership reliability depends on staff understanding. Providers should maintain supervision coverage, scenario-based training on thresholds and escalation, and manager oversight of high-risk cohorts. Commissioners will often infer partnership maturity from workforce stability and the consistency of staff decision-making.
Operational examples (how partnerships are strengthened over time)
Example 1: Turning recurring referral friction into a stable pathway interface
Context: The first year of delivery involves frequent referral bounce between a provider and a Trust team. Staff frustration grows and contract meetings become dominated by disputes about scope and thresholds.
Support approach: The provider proposes an interface improvement cycle: agree a referral minimum dataset, implement structured decline reasons, and hold a short weekly edge-case call for eight weeks. The provider also creates a simple “referral decision guide” for staff and referrers.
Day-to-day delivery detail: Triage staff use consistent eligibility language. Declined referrals are recorded with structured reasons, and themes are reviewed weekly. The provider shares trend data monthly and proposes targeted actions (referrer briefing, revised form prompts, updated guidance). After changes, the provider runs a small sample audit to verify that acceptance/decline decisions are documented clearly and consistently.
How effectiveness/change is evidenced: Reduced avoidable declines, improved time-to-first-contact, and fewer disputes in contract meetings because evidence is clearer. Evidence includes referral logs, improvement actions and re-audit results.
Example 2: Joint learning from escalation delays improves system response
Context: Escalation delays occur, especially out of hours or across certain localities. Partners disagree about whether information quality or capacity is the driver. The ICB wants assurance that the system is learning rather than repeating the issue.
Support approach: The provider initiates a joint escalation deep dive: review a small set of anonymised escalation timelines, identify where delay occurred, and agree process changes (standard escalation summary, escalation ladder, clearer response expectations). Re-audit is planned after implementation.
Day-to-day delivery detail: Staff use the standard summary; managers record escalation steps and outcomes. Repeat escalations trigger manager review and plan updates. Governance tracks delay themes monthly and escalates unresolved interface risks formally. After agreed changes, a sample review checks whether documentation quality and timeliness improved.
How effectiveness/change is evidenced: Improved escalation timeliness, fewer repeat delays for the same reasons, and clearer shared understanding of responsibilities. Evidence includes escalation logs, deep dive outputs and re-audit findings.
Example 3: Rights and safety strengthened through restrictions review discipline
Context: In response to safeguarding concerns, restrictions persist across handovers without review. The ICB is concerned about rights risk and potential inspection findings.
Support approach: The provider implements a restrictions standard and a review register shared in governance. Restrictions must be time-limited, reviewed, and stepped down where safe, with clear rationale and evidence of least restrictive alternatives considered.
Day-to-day delivery detail: Supervisors require explicit review decisions. During handovers, restrictions and review dates are part of the minimum dataset so the receiving team understands expectations. Governance reviews long-running restrictions monthly and requires action plans where step-down is not happening. Quarterly sampling verifies that the standard is embedded and that restrictions are linked to active safeguarding and positive risk-taking plans.
How effectiveness/change is evidenced: Reduced duration of restrictions, stronger proportionality documentation, and clearer step-down decisions over time. Evidence includes the register, sampled files and governance minutes showing corrective action and verification.
Explicit expectations that must be met
Commissioner expectation
Commissioners expect partnership to reduce uncertainty and contract management burden over time. They will look for stable governance rhythms, clear interface controls, transparent risk management, credible performance evidence, and verified improvement through sampling and re-audit. They also expect disputes to be resolved through agreed routes with interim safety planning.
Regulator / Inspector expectation (e.g. CQC)
CQC expects effective leadership and governance to deliver safe, consistent care and continuous improvement. Inspectors will test whether partnership arrangements support safeguarding, timely escalation and continuity, and whether learning changes practice. They will also look closely at proportionality and review of restrictive practices where present.
Long-term credibility is built on repeatable evidence
The providers who sustain long-term partnerships do not attempt to prove trust with statements; they reduce doubt with predictable evidence. When the system can see how you recognise risk, escalate, learn and verify improvement — and when those patterns hold over time — partnership becomes the default position rather than something you have to sell.