Psychometric Considerations in Developing PROMIS® Measures for Early Childhood

Area of Science:

• Pediatric Health Measurement
• Psychometrics
• Child Development

Background:

• Lifespan health and disease states are often detectable in early childhood.
• The Patient-Reported Outcome Measurement Information System (PROMIS®) previously covered ages 5 and older.
• There was a need for developmentally appropriate measures for younger children.

Purpose of the Study:

• To extend the PROMIS® system to early childhood (ages 1-5 years).
• To develop and psychometrically evaluate parent-report measures for young children.
• To create lifespan-coherent health outcome measures for a broader age range.

Main Methods:

• Collected 2 waves of parent-report data from children aged 1-5.
• Utilized item response theory (IRT) methods for item pool development and testing.
• Evaluated unidimensionality, model fit, and item parameters for 12 distinct item pools.

Main Results:

• Successfully developed 12 PROMIS® item pools for early childhood, including Global Health, Physical Activity, Sleep, and psychosocial domains.
• Confirmed unidimensionality for several key domains in initial analyses.
• Combined related concepts (e.g., Sleep Disturbance/Impairment) and refined others (e.g., Self-Regulation) for improved measurement.

Conclusions:

• PROMIS® has been successfully extended to cover children aged 1-5 years using rigorous mixed-methods.
• Developmentally appropriate, lifespan-coherent parent-report measures are now available.
• These new measures are publicly accessible in English and Spanish.