Transfer refusals invisible in capacity counts

1. System failure identification

   A unit declines an incoming transfer for reasons of acuity, staffing, or isolation while still reporting available beds. The capacity is real and simultaneously unusable, and the refusal leaves no trace. Coordinators keep routing to a unit that will keep declining.

2. Data pipeline diagnostics

   Bed availability is modelled as a physical property. Whether the unit can accept a given patient depends on staffing ratio and case mix, neither of which the count includes. The count answers whether a bed exists, not whether this patient can occupy it.

3. Predictive computational model

   Record refusals as structured events with a reason code, and report acceptance rate alongside bed count. A unit with beds and a low acceptance rate is a distinct and actionable state. Acceptance rate is the empirical correction to a theoretical capacity figure.

4. Workflow integration

   Surface acceptance rate to regional coordination rather than to the unit. The unit already knows it is refusing, and the coordinators routing patients are the ones operating on bad information. Information is only useful where a routing decision is being made.

Back to the case study database