Background: Translating scores to a common metric provides advantages that support the applied collaborative care model in the Netherlands. McCall's T-score is frequently used, and the optimal...Show moreBackground: Translating scores to a common metric provides advantages that support the applied collaborative care model in the Netherlands. McCall's T-score is frequently used, and the optimal approach to obtain this score is through Item Response Theory (IRT). However, IRT is complicated, requires dedicated software, and a large dataset. This study aimed to validate an alternative procedure to obtain T-scores from IRT. Methods: This study used data from an existing study, which comprised a population-based sample and a patient sample. We estimated the relation between raw scores and IRT-based T-scores with curve fitting and established conversion formulas to transform raw scores into calculated T-scores. We illustrated the process with raw scores from the Beck Depression Inventory, the Inventory of Depressive Symptomatology - Self Rated, and the Montgomery-Åsberg Depression Rating Scale. We performed a correlational analysis to assess the validity of the calculated T-scores. We also determined cut-off values using ROC analysis. Results: The curve-fitting procedure resulted in third-order polynomial regression equations to use as conversion formulas. The validity of calculated T-scores was supported by their high correlation with theta-based T-scores. ROC analysis provided cut-off values, which were comparable to the previous studies, using raw scale scores. Conclusion: The curve-fitting procedure yielded sufficiently valid calculated T-scores for all instruments in comparison to the theta-based T-scores from IRT. The resulting cut-off values demonstrated that calculated T-scores were able to distinguish patients from the general population. The practical use of the results is discussed.Show less