Developing and Validating a Pediatric Potentially Avoidable Transfer Quality Metric

Area of Science:

• Healthcare Quality Improvement
• Pediatric Patient Safety
• Health Informatics

Background:

• Pediatric potentially avoidable transfers are a significant concern in healthcare quality.
• Accurate identification of these transfers is crucial for targeted interventions.
• Existing methods for identifying such transfers may lack precision.

Purpose of the Study:

• To evaluate and refine a quality metric for identifying pediatric potentially avoidable transfers.
• To establish the operating characteristics of the metric using physician medical record review as the gold standard.
• To validate an optimized metric definition in a separate patient sample.

Main Methods:

• Development of an initial metric definition based on diagnosis and procedure codes.
• Estimation of the initial metric's operating characteristics (c-statistic).
• Iterative refinement of the metric definition to optimize the c-statistic, followed by validation on a separate dataset.

Main Results:

• The initial metric definition yielded a c-statistic of 0.63.
• After 22 revisions, the optimized metric identified transfers discharged within 24 hours without specific specialized diagnoses/procedures.
• The optimized metric demonstrated a sensitivity of 80.6%, specificity of 85.7%, and a c-statistic of 0.83 in the validation sample.

Conclusions:

• The developed quality metric for pediatric potentially avoidable transfers shows satisfactory operating characteristics.
• This metric provides a feasible and validated approach to measure this important healthcare outcome.
• The findings support the use of this metric for quality assessment and improvement in pediatric care.