Measurement of Patient-reported Outcomes (PRO) still lacks behind clinical standards. Most established tools are also either too burdensome or too imprecise to be used in clinical practice. Item Response Theory (IRT) methods and Computer Adaptive Tests (CAT) promise to overcome these shortcomings. Simulation studies have shown that individually tailored CATs can provide more precise and less burdensome measurements over a larger measurement range than static tools. Several studies with real CAT application have supported the psychometric superiority of CATs, but results from longitudinal studies are still scarce. IRT item banks also allow scoring different established tools measuring the same construct on one common metric, which could greatly facilitate the harmonization of PRO-assessments.
