Predicting no-shows in paediatric orthopaedic clinics

Area of Science:

• Healthcare Management
• Clinical Informatics
• Pediatric Orthopedics

Background:

• Clinic no-shows (NS) present a significant challenge to healthcare systems, impacting revenue and patient care.
• Previous research has not utilized electronic health records (EHR) to identify high-risk groups for NS in pediatric orthopedics.

Purpose of the Study:

• To identify predictors of non-attendance at pediatric orthopedic outpatient appointments using discrete data from EHR systems.
• To leverage EHR data for proactive management of patient no-shows.

Main Methods:

• Analysis of 131,512 encounters from January 2014 to March 2016.
• Classification and Regression Trees (CART) were employed to identify predictive variables.
• Variables included appointment status, patient demographics, insurance type, and appointment scheduling details.

Main Results:

• The overall no-show rate was 11.8% (15,543 encounters).
• Key predictors for NS identified by CART were: duration between scheduling and appointment, insurance type (private vs. government), and specific orthopedic clinic type.
• Patients with private insurance and wait times ≤38.5 days had the lowest NS rate (7.8%), while those with wait times >38.5 days for fracture or sports clinics had the highest (29.3%).

Conclusions:

• Insurance type and appointment scheduling duration are significant predictors of non-attendance in pediatric orthopedics.
• These findings enable the prediction and intervention for at-risk patient groups.
• Despite predictive capabilities, the inherent complexity of no-shows is highlighted by persistent rates even in optimized groups.