Background: Executing safe transitions of care for hospitalized patients is important because of the vulnerability of patients during this time period.  While hospital-based providers frequently schedule appointments on behalf of their patients prior to discharge, cancellations and non-attendance (“no-shows”) for post-discharge follow-up appointments are anecdotally common and diminish the quality  of patient care.  Little is known about the incidence of and reasons for non-attendance at these appointments.

Methods: This retrospective single-site cohort study examined all outpatient follow-up appointments scheduled prior to hospital discharge by providers at our academic medical center over a one-year time period.  Patients included in the analysis were hospitalized on general medicine or medical subspecialty services; appointments were included if they were scheduled during dates that a patient was hospitalized.  Appointments booked with providers unaffiliated with our health system were not examined, and appointments where the patient expired prior to the appointment date were excluded.  Appointment outcomes were subsequently categorized as attended, cancelled, or no-showed.  The reason for each cancellation, as entered by outpatient clinic staff, was secondarily characterized as medically-related (e.g., prior hospital readmission), patient-related (e.g., patient choosing to reschedule), or systems-related (e.g., clinic having to reschedule the appointment).  Statistical calculations were performed using Chi-squared analyses with a pre-determined α = 0.05.

Results: 10,644 follow-up appointments (representing 6505 unique hospital encounters) were scheduled prior to hospital discharge and included in our analysis.  49% of appointments were attended, 32% were cancelled, and 19% were no-shows.  Statistically significant risk factors (p < 0.001 in each case) for non-attendance included: age under 40 (RR 1.12), Medicaid insurance (RR 1.25), driving distance within 15 miles (RR 1.10), and appointments scheduled beyond 1 week after discharge (RR 1.23, p < 0.001).  75% of cancellations occurred within a week of the appointment, with 51% happening within three days and 28% happening on the day of the appointment.  Reasons for cancellation were predominantly patient-related (59%), while 32% were systems-related and 7% were medically-related.

Conclusions: Over half of post-discharge follow-up appointments scheduled by providers on behalf of their patients at our urban academic medical center were not attended as scheduled.  This finding was particularly evident for patients who were younger, had Medicaid insurance, lived closer to our hospital, or had appointments scheduled for dates further from discharge. Based on these findings, our institution is in the process of actively exploring potential interventions to better engage patients and their families in the appointment-scheduling process.