Skip to content

Window-Closure Review

Closure review — instantiates Critical-Window Intervention Timing

Judges whether the receptive window has closed or is about to, and applies a stop rule that halts window-dependent escalation and hands off to protected alternatives rather than pushing harder past closure.

The most dangerous moment in a window-timed intervention is the one after the window has quietly closed, when the team keeps escalating an approach that can no longer work. Window-Closure Review governs the far edge: it judges whether receptivity has ended — or forecasts that it soon will — and issues a stop rule that ends window-dependent effort and redirects to protected later routes. Its defining move is the stop-and-switch: rather than intensifying a fading intervention on the hope the window is still open, it declares closure against a preset criterion and converts that judgment into an explicit halt. It is the mirror of opening detection: where readiness assessment decides begin, closure review decides stop, because pushing further now only costs.

Example

A stroke team is treating an ischemic stroke, where clot-dissolving thrombolysis is effective only within a narrow window from symptom onset — roughly four and a half hours in standard guidance.[1] As the clock runs, the review tracks the closure criterion: time-from-onset against the guideline limit, plus imaging signs that the salvageable tissue is gone. When the criterion is met — the window has closed for this patient — the stop rule fires: no further escalation of the time-critical drug, because past the window its expected harm now outweighs its benefit.

Critically, the stop is a switch, not an abandonment. Firing the rule routes the patient to the protected alternatives that remain appropriate after closure — supportive management and rehabilitation planning — rather than leaving them with nothing. The review's contribution is precisely this disciplined handoff: it prevents both the harm of pushing a closed-window treatment and the harm of treating "window closed" as "care over."

How it works

The review holds a window-closure criterion — the observable condition that marks receptivity as ended (elapsed time past a limit, loss of a readiness signal, an imaging or developmental marker) — and a closure forecast and stop rule that both projects when closure will arrive and specifies what halts when it does. It runs as a recurring check against that criterion; when closure is met or imminently forecast, it fires the stop rule, ending window-dependent escalation and triggering the handoff to protected later routes. Its output is a decision — continue, closing, prepare to switch, or closed, stop and hand off — not a new estimate of the window.

Tuning parameters

  • Closure threshold — how much evidence of closure is required before stopping. A tight threshold stops promptly and avoids wasted or harmful late effort; a loose one keeps trying longer at rising cost.
  • Forecast lead — how far ahead closure is projected so the switch can be prepared. More lead readies the alternative in time but may switch a still-open case early.
  • Stop firmness — whether the stop is a hard halt or a review trigger. Hard stops prevent escalation creep; soft ones allow case-by-case override but invite it.
  • Handoff target — which protected alternative the stop routes to. Choosing this in advance is what keeps closure from defaulting to "nothing further offered."

When it helps, and when it misleads

It is most valuable when continuing a window-dependent intervention past closure is not merely futile but harmful or costly, and when a real alternative exists to switch to. Its failure mode is a miscalibrated criterion: too eager and it stops while the window is still open, denying a real chance; too reluctant and it escalates a dead approach. The gravest misuse is running it backwards as a denial instrument — declaring the window "closed" to justify withdrawing support that is inconvenient to provide, converting an uncertain group-level cutoff into a deterministic verdict against an individual. The discipline is to base closure on this system's evidence rather than a population average, to pair every stop with a real protected handoff rather than a void, and to keep the criterion honest by separating "the window is closed" from "we would prefer to stop."

How it implements the components

  • window_closure_criterion — it defines and evaluates the observable condition under which receptivity is judged to have ended.
  • closure_forecast_and_stop_rule — it projects closure and holds the rule that halts window-dependent escalation and initiates the switch.

It does not decide the opening (readiness_signal, window_opening_criterion — see Window-Opening Readiness Assessment), does not re-estimate the window's model from new data (critical_window_model, window_uncertainty_band — see Adaptive Window Re-estimation), and does not build the post-closure remediation route it hands off to (missed_window_fallback_path — see Missed-Window Remediation Plan).

Notes

A stop from this review is not a claim that nothing can ever help again. Whether a closed window can be reopened is a separate question answered by Reconsolidation or Reopening Protocol; what protected route to offer after closure belongs to Missed-Window Remediation Plan. The review's job ends at the halt and the handoff.

References

[1] Intravenous thrombolysis for ischemic stroke carries a well-established time limit from symptom onset (about 4.5 hours in standard guidance), past which expected harm outweighs benefit. It is a real, defined closure criterion — used here only to illustrate a stop rule, not as a clinical recommendation — and its existence is why the review pairs a time-based closure test with a switch to protected care. ↩