What Predicts a Completed Reschedule? A Cohort Study of Appointment Friction

Research question, population, and method

A reschedule is often counted as a simple calendar edit, but the customer experience is a sequence of choices. This cohort study asks which observable features of a reschedule request are associated with completion: the time of the request, available alternatives, channel, appointment type, reason category, or need for staff review. The study does not assume that a completed move is always a good outcome; a customer may accept an unsuitable time because the path is difficult. Each cohort begins at a cancellation or change request and ends at a completed replacement, a documented decision not to continue, or a defined unresolved cutoff. Requests that remain open are retained as unresolved rather than silently classified as failures. That distinction is essential for appointment scheduling operations because an empty slot, an unanswered message, and a completed replacement represent different states of demand.

What the event record can show

Build the cohort from event history, not from the final calendar. Capture when the request was received, when options were offered, how many usable options were shown, whether a response was received, and whether a second touch was needed. Code the reason only when it is explicitly stated; do not infer dissatisfaction from delay. Separate customer-selected timing constraints from internal availability constraints. A reschedule can fail because no suitable slot exists, because the customer did not receive the offer, because the request requires an approval, or because the follow-up was not owned. Those causes imply different remedies. A scheduling operator can present approved alternatives, confirm the selected time, preserve the cancellation reason, and route exceptions. The operator cannot turn an unavailable resource into an available one or make a policy exception without authorization.

How to analyze the scheduling workflow

Compare cohorts by appointment type and original lead time. A request for a short recurring visit may have many substitutes, while a specialist appointment may have only a narrow set. A headline completion rate that combines them can hide the access problem that matters most. Report time to first option, number of options offered, time to decision, eventual outcome, and the share that needed more than one contact. Where possible, inspect the records in each outcome group. A customer who never replied after receiving three clear options differs from a request that sat unassigned for three days. The evidence should make that distinction visible. If a workflow change is tested, keep the offer language, channel, ownership, and eligibility rules stable enough to interpret the result, and record any concurrent seasonal or staffing change.

Role boundaries and interpretation risks

The central analytical risk is confusing customer preference with operational friction. A longer reschedule time may reflect a narrow preference, but it may also reflect poor option quality. Record whether the offered alternatives met the original service, resource, location, and time-window requirements. If the system cannot express those constraints, preserve them in a structured note or a review queue. Do not use a “next available” label as evidence that the option was suitable. For schedulingAppointment readers, the practical decision is often whether to improve option presentation, add an escalation trigger, or protect capacity for a high-constraint appointment class. The cohort design helps because it connects a workflow touch to an outcome without claiming that the touch caused the outcome.

Evidence-led conclusion and limitations

This study cannot establish causality from routine records. Reschedule reasons may be incomplete, channels may be measured differently, and unresolved requests may be under-recorded. External reminder research concerns attendance more directly than rescheduling, so it provides context rather than a direct benchmark. The evidence-led conclusion is that reschedule completion should be analyzed as a constrained matching problem with an ownership path, not as a binary calendar edit. A defensible local report keeps unresolved requests in the denominator, distinguishes suitable from merely open slots, and states the observation window. That gives an appointment team a grounded basis for deciding where customer choice ends, where calendar capacity begins, and where a human escalation is necessary.

Research methodology

Methodology: This is a structured evidence review and proposed local cohort audit about reschedule friction. The cited sources were compared by population, intervention or workflow, outcome definition, and evidence scope; their estimates were not pooled. A local operator should preserve event history, define the denominator before measurement, retain unresolved records, and compare a fixed observation window. The approach supports a bounded operating baseline, not a universal target. Limitations include heterogeneous appointment types, channels, policies, and incomplete routine records.

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. AHRQ, Improving Primary Care: AHRQ guidance on measuring and improving primary-care workflows.
2. Cochrane, mobile phone messaging reminders: Systematic review of text and phone reminders for appointments.
3. Dantas et al., no-shows in appointment scheduling: Systematic review describing variation in no-show research and definitions.
4. BLS, Receptionists: Occupational description of front-desk scheduling work.

Related content

• Research library
• Appointment scheduling service
• Scheduling support

Questions for scheduling operators