How Does One Appointment Delay Propagate? A Calendar Recovery Study

One delay can create several questions

A late appointment is easy to observe when planned and actual start times are both recorded. What follows is harder. The next appointment may start late, move, be cancelled, or remain unchanged. A customer may receive a notice, not receive it, or have no notice record at all. If a report calls all of these propagation, it makes an assumption about causality. The study asks how to follow the sequence without losing those distinctions. Flow literature helps frame demand, supply, and bottlenecks, but it cannot tell a local operator how much one delay affected a particular calendar. A scheduling record should preserve planned time, observed time, downstream event, communication event, and recovery state. Support can document approved actions and route exceptions; it cannot promise that a delay will or will not affect another appointment.

Trace the calendar state by state

A reliable delay record links events without overwriting the original plan. Keep planned start and duration, actual start and end when known, buffer state, downstream change, notice time, and final disposition. Mark unknown timestamps honestly. A late start followed by an on-time next appointment is different from a late start followed by a reschedule, even if both began with the same variance. The cause may be service complexity, a protected buffer, a cancellation, or a calendar decision made outside the record. These should not be guessed. The scheduler’s role is to preserve the event trail and communicate approved information. The owner decides whether buffers, appointment lengths, escalation, or notice rules should change. This separation prevents a descriptive metric from becoming an unauthorized operational rule.

How to conduct a recovery cohort

Select a fixed period and include appointments with a defined planned start. Group records by whether actual timing is known, then trace the next scheduled event only where the linkage is reliable. Report delay distributions, downstream changes, notice events, and recovery outcomes with separate denominators. Read cases with large delay, no downstream effect, multiple changes, and unknown communication. Compare appointment types because duration and workflow can differ. Note holidays, coverage changes, protected blocks, and cancellations that may affect the same calendar. If an owner tests a notice or recovery process, preserve the prior definitions and document concurrent changes. The study should establish what happened in the observed window. It should not forecast the next day’s effect or claim that one delay explains every later change.

Avoid false chains of causation

The first mistake is linking every later delay to the first late start. The second is treating a calendar edit as proof that the customer experienced a delay. The third is treating a sent message as received or a reschedule offer as accepted. Teams may also use average propagation time without retaining no-effect and unknown cases. A calendar’s protected time can absorb a variance, while a cancellation can create a later opening; both change the observed sequence. Scheduling support should follow approved notice and reschedule procedures, capture stated responses, and escalate decisions about policy or service delivery. It should not diagnose the cause of a delay, alter a buffer, or promise a recovery slot. Measurement is most useful when the role boundary remains explicit.

Conclusion and limitations

The evidence supports measuring delay propagation as a sequence of observed events rather than applying a fixed multiplier to late starts. Preserve planned and actual timing, downstream calendar state, notice events, and recovery outcomes, with unknowns visible. Flow and access sources provide useful concepts but no universal propagation rate or buffer rule. Limitations include missing timestamps, heterogeneous appointment types, protected time, manual calendar edits, and incomplete communication records. The next decision should be narrow: define one downstream relationship, reconcile its exceptions, and test one approved recovery control against a stable baseline. This yields a more honest picture of calendar resilience and customer access. It also avoids blaming a single appointment or promising that a process change will produce a guaranteed result.

Research methodology

Methodology: Combine public flow and access evidence with a proposed event-level audit. Follow appointments with planned start, actual start or unknown, downstream slot effects, notices, and final disposition in a fixed window. Compare sequences descriptively and inspect exceptions; do not infer causality from correlation. Limitations include incomplete timestamps, different service types, protected buffers, manual adjustments, and unrecorded communication.

Data sources and methodology

This brief reports published findings as stated by each source. It does not combine study populations into a new benchmark; local operators should treat the figures as context and measure their own workflow.

1. Institute for Healthcare Improvement, flow: Discusses demand, capacity, and flow relationships.
2. Murray and Berwick, advanced access: Addresses access and appointment supply.
3. AHRQ, care coordination measures atlas: Provides timeliness and coordination measure concepts.

Related content

• Research library
• Appointment scheduling service
• Scheduling support

Questions about propagated delay