Abstract
To the Editor:
With inefficient patient access, a health system will not achieve the Triple Aim of improving the patient experience and the health of populations, and reducing the per capita cost of health care. Substance abuse, homelessness, and unemployment impede access. 1 Some interventions to improve access include reducing patient wait time 2 and patient reminders.3,4 McMullen and Netland found a direct correlation between a patient failing to appear for an appointment (no-show) and the scheduled time to appointment. 5
We predicted that patients are more likely to keep a subspecialty appointment if there is a short wait time between scheduling an appointment and the actual appointment date. Therefore, as part of a performance improvement project, we examined appointment data from the Division of Cardiology, a division of Einstein Physicians Philadelphia, over a 2-week period. The Albert Einstein Medical Center Philadelphia is an urban tertiary care facility, staffed by 28 cardiologists.
We analyzed the relationship between the interval from the date on which the appointment was scheduled to the actual appointment date (wait time), and we compared this to the appointment status (show vs no-show or late cancellation). A no-show was defined as a patient who did not keep his/her appointment and did not call at least 24 hours prior to the scheduled appointment. An appointment was considered a late cancellation if the patient canceled the appointment within 24 hours of the scheduled appointment time. Fisher exact test was used to determine statistical significance.
We identified 228 appointments for unique patients that were requested during the period from June 15 to 30, 2015. Of these patients, 72.8% were African American, and 38.2% had Medicaid/Medicare. The mean age of the cohort was 63 years. No-show rates for appointments made within 2 weeks appeared to differ compared to no-show rates among patients with appointments made more than 2 weeks in advance. Therefore, we performed a binary analysis comparing no-show rates between patients with wait time less than (group 1) and more than (group 2) 2 weeks. The average no-show rate in group 1 was 5.8% compared with 29.1% in group 2 (P = .004). The late cancelation rate in group 1 was 0% versus 11.3% in group 2 (P = .52).
In summary, we demonstrated an inverse correlation between the wait time for cardiology appointments and the no-show rate. A wait time of more than 2 weeks significantly increased the probability that a patient would not show. Patients also may be less likely to cancel their appointments if the wait time is less than 2 weeks, but the sample may have been too small to achieve statistical significance.
To validate our findings and their broader applicability, the same analysis would need to be conducted in other types of practices and with populations with a higher rate of commercial insurance. Finally, we did not collect information on patient factors that could affect the likelihood of keeping an appointment, such as comorbid illnesses, the source of the referral to the subspecialty practice, and the diagnosis for which the patient was to be seen.