Stress-test an add-on case spike by proving each add-on lane has a named owner, a measured age-out, and a reverse path before draft add-ons start governing every OR claim.
Add-on case spikes fail when "added" means a status without a decision artifact, when teams dual-count a draft add-on as both approved and live, and when high-utilization service lines still pin to one thin scheduler because the lane inventory is incomplete. The review needs one written inventory of add-on lanes, one owner who can deny a risky shortcut, and a documented consequence when a failed add-on ages past the deadline.
Map the add-on path
List every add-on class with its response window, waive rule, override behavior, owners, and notification channels. Mark paths that cannot reverse without a manual amendment. Attach the last three add-on spike incidents with raw timelines and any waivers. Include the source of truth for open add-on counts during the observation window.
Define the phases for detect, hold, divert, abort, and restore. Each phase needs an owner and an exit condition. Write the point after which a stuck add-on would require a different procedure, then review whether that action is still permitted. Capture maximum acceptable dwell in measurable units, including which specialties are excluded from the add-on path and why.
Include the calendar of known events for the next two quarters: block freezes, staffing gaps, and equipment outages that shrink the usable response window. A spike budget that ignores those dates will look calm until the week they land.
Loads that expose soft add-ons
- A minority high-volume specialty keeps a side-channel while the aggregate add-on dashboard stays green.
- A restore has already left a room without a trusted add-on state.
- The primary add-on export lags beyond the planned observation window.
- An operator skips an add-on gate because a day-of cutoff is near.
- Automated and human divert messages collide under the new add-on rule.
- Abort authority is unclear at week end and the update lands on the wrong amendment.
For each load, identify detection time, room impact, containment, and the authority to force an add-on rollback. Require commands and export links in the runbook.
Keep ownership after the spike
Run the package in Pingpong with OR ops, nursing, anesthesia, and clinic seats. Ask OR ops which specialty decision becomes unsafe first if add-ons still stick after the claimed window. Ask nursing whether turnover macros can absorb a forced override. Ask owners to show the exact add-on version or export used as the exit condition.
Related reviews include the OR ops lead seat, a block time release rule, a turnover target cut, and an on-call schedule rewrite. Browse the war-game decisions hub for adjacent controls.
Authorize the published add-on path only after a timed drill restores usable add-on hygiene inside the documented budget without an undocumented manual step.