Expert-Governed Modality Change¶

A safety-governed protocol — instantiates Plateau Detection and Switching

In safety- or human-affecting settings, requires qualified review before switching intervention modality after a plateau — so a change of approach protects the people it affects rather than merely relabeling failure.

Expert-Governed Modality Change is the archetype applied where the plateaued path is an intervention on people — a treatment, a therapy, an educational or safeguarding approach. In these settings a switch is not a free operation: changing modality can carry risk, and "the current approach has plateaued" can be misused to withdraw care, exclude the hard-to-serve, or dress up abandonment as strategy. This mechanism's distinguishing feature is a governance gate: before any modality change, a qualified reviewer must confirm the plateau is real, weigh the risk of switching versus continuing, and check that the move doesn't abandon or disadvantage the person it affects. The switch trigger is deliberately not automatic — the plateau evidence opens a review, and it is the review, not a threshold, that authorizes the change.

Example¶

A mental-health clinic uses measurement-based care: a patient's depression symptoms are tracked on a standard scale at each visit. After a course on one medication, the scores plateau — improvement has stalled well short of remission across several measurements. Expert-Governed Modality Change is what turns that plateau into a reviewed change rather than a reflexive one. The plateau evidence triggers a clinician review, not an automatic switch: the clinician confirms the response really has stalled (adequate dose, adequate duration, not a measurement blip), weighs the risks of changing versus staying, and, in a stepped-care logic, moves to a different modality — augmentation, a switch of agent, or adding psychotherapy — with the patient's informed involvement. The guardrail is explicit: the change must aim at better outcomes for this patient, and the plateau can't be used to quietly discharge a difficult case. The switch happens because a qualified person judged it right, on the record.

How it works¶

Let plateau evidence open a review, not trigger a switch. A stalled response prompts a mandatory qualified review; the change is never automatic in a human-affecting setting.
Gate the change on qualified judgment. A reviewer with the relevant expertise confirms the plateau, weighs switch-vs-continue risk, and authorizes (or declines) the modality change.
Apply the stakeholder guardrail. The review explicitly checks that the switch serves the affected person and doesn't abandon, exclude, or disadvantage them — with their informed involvement where appropriate.

Tuning parameters¶

Review threshold — how much plateau evidence opens a review. A low bar catches stalled cases early but floods reviewers; a high bar conserves attention but leaves people on a spent approach too long.
Reviewer seniority — the level of expertise the gate requires. Higher seniority improves safety but adds delay and bottlenecks; lower seniority is faster but riskier.
Guardrail strictness — how much protection the affected person's interests are given against expedient switching. Stricter guardrails prevent abandonment; over-strict ones can trap someone on a non-working modality.
Documentation depth — how fully the rationale is recorded. Thorough records enable accountability and later review; heavy documentation slows the response.

When it helps, and when it misleads¶

Its strength is bringing the discipline of stop-and-switch into settings where a careless switch harms people — it prevents both grinding on with a plateaued treatment and the opposite abuse, where "it's plateaued" becomes cover for discharging a hard case. Anchoring the trigger in qualified review is the logic of stepped care: intensify or change approach based on measured response, under clinical governance.^[1]

Its failure modes are institutional. The review can become a rubber stamp that ratifies whatever the clinician already intended, or a bottleneck so slow that people languish on a spent modality waiting for sign-off. And the guardrail can be inverted — the language of "person-centered switching" used to justify a change that actually serves the institution's convenience. The discipline is a genuine, evidence-gated review with the affected person's informed involvement, a rationale documented for accountability, and monitoring that the switch actually helped rather than merely closed a case.

How it implements the components¶

Expert-Governed Modality Change fills the safety-governed-switch subset — a reviewed change that protects the affected person:

stakeholder_guardrail — its core: the explicit check that a modality change serves the affected person and does not abandon, exclude, or disadvantage them.
alternative_strategy — the new intervention modality (a different treatment, therapy, or approach) authorized by the review.
switch_trigger — the review-gated moment: plateau evidence opens a qualified review, and the review — not a threshold — authorizes the change.

It does not compute the plateau signal or run a controlled trial (marginal_response_metric, false_plateau_check → Diminishing Returns Detection, Controlled Experiment After Plateau); it governs whether and how a switch happens in a human-affecting setting.

Instantiates: Plateau Detection and Switching — it is the safety-governed instance of stop-and-switch for human-affecting interventions.
Sibling mechanisms: Strategy Switch Decision Tree · Controlled Experiment After Plateau · Diminishing Returns Detection · Marginal Gain Dashboard · Escalation Stop Workflow · Plateau Review Cadence · Ad Fatigue Switching · Training Plateau Adjustment · Product Growth Plateau Response · Process Redesign After Plateau · Saturation-Aware Resource Allocation

Notes¶

The switch trigger here is intentionally not the automatic threshold used by Escalation Stop Workflow. In human-affecting settings, plateau evidence opens a qualified review rather than firing a change — the governance gate is the whole point, and speed is traded for safety.

References¶

[1] Stepped care — a care model in which treatment intensity or modality is adjusted based on the patient's measured response, under clinical governance, rather than left unchanged or escalated blindly. It is the real-world discipline this protocol generalizes: a measured plateau prompts a reviewed change of approach. ↩