JMIR Publications

ABSTRACT

Background:

Infant growth is commonly assessed using separate percentile charts for height, weight, and head circumference. An integrated multidimensional approach may provide complementary information.

Objective:

To describe RaschKIDMAP, a lightweight web application that applies Rasch measurement to multidimensional growth assessment in infants aged 0–24 months.

Methods:

RaschKIDMAP integrates height, weight, and head-circumference measurements with sex- and age-matched reference data. For each assessment, the application identifies 50 reference infants matched by sex and month of age, then appends the assessed infant as a 51st record. Measurements are transformed within the matched group and analyzed using a probability-based continuous-response Rasch model, with optional TAM rating-scale and partial-credit models. Outputs include person measures, item difficulties, standard errors, fit statistics, residual Z-scores, and empirical percentiles.

Results:

The application generates an accessible visual report comprising individual and group forest plots, a KIDMAP item-residual display, a Wright map, original comparison data, person and item tables, and conventional percentile plots. It accepts routine anthropometric inputs and returns results immediately without a database backend.

Conclusions:

RaschKIDMAP is a proof-of-concept tool designed to complement conventional single-indicator growth percentiles with an integrated Rasch-based representation of infant growth. Further empirical validation is required before clinical implementation.