Autoencoded sparse Bayesian in-IRT factorization, calibration, and amortized inference for the Work Disability Functional Assessment Battery

The Work Disability Functional Assessment Battery (WD-FAB) is a multidimensional item response theory (IRT) instrument designed for assessing work-related mental and physical function based on responses to an item bank. In prior iterations it was developed using traditional means -- linear factorization, followed by statistical testing for item selection, and finally, calibration of disjoint unidimensional IRT models. As a result, the WD-FAB, like many other IRT instruments, is a posthoc model. In this manuscript, we derive an interpretable probabilistic autoencoder architecture that embeds as the decoder a Bayesian hierarchical model for self-consistently performing the following simultaneous tasks: scale factorization, item selection, parameter identification, and response scoring. This method obviates the linear factorization and null hypothesis statistical tests that are usually required for developing multidimensional IRT models, so that partitioning is consistent with the ultimate nonlinear factor model. We use the method on WD-FAB item responses and compare the resulting item discriminations to those obtained using the traditional method.