Performance of longitudinal item response theory models in shortened or partial assessments
- PDF / 540,259 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 80 Downloads / 159 Views
ORIGINAL PAPER
Performance of longitudinal item response theory models in shortened or partial assessments Leticia Arrington1,2
•
Sebastian Ueckert1 • Malidi Ahamadi2 • Sreeraj Macha2 • Mats O. Karlsson1
Received: 5 March 2020 / Accepted: 18 June 2020 Ó The Author(s) 2020
Abstract This work evaluates the performance of longitudinal item response (IR) theory models in shortened assessments using an existing model for part II and III of the MDS-UPDRS score. Based on the item information content, the assessment was reduced by removal of items in multiple increments and the models’ ability to recover the item characteristics of the remaining items at each level was evaluated. This evaluation was done for both simulated and real data. The metric of comparison in both cases was the item information function. For real data, the impact of shortening on the estimated disease progression and drug effect was also studied. In the simulated data setting, the item characteristics did not differ between the full and the shortened assessments down to the lowest level of information remaining; indicating a considerable independence between items. In contrast when reducing the assessment in a real data setting, a substantial change in item information was observed for some of the items. Disease progression and drug effect estimates also decreased in the reduced assessments. These changes indicate a shift in the measured construct of the shortened assessment and warrant caution when comparing results from a partial assessment with results from the full assessment. Keywords Item response theory Composite score Pharmacometrics Item information
Introduction Composite scores from clinical assessments which use rating scales are a common method for the evaluation of ability or disability in a wide range of therapeutic areas. Generally, they aim at capturing different aspects of a disease by combining a variety of symptoms into a single composite score. These scores are used in clinical practice for diagnosis and to guide treatment of patients, but are also common endpoints in clinical trials. The statistical data analysis of clinical trials with composite assessments as endpoints is traditionally based on the total composite score only. Similarly, pharmacometric models of disease Electronic supplementary material The online version of this article (https://doi.org/10.1007/s10928-020-09697-x) contains supplementary material, which is available to authorized users. & Mats O. Karlsson [email protected] 1
Department of Pharmaceutical Biosciences, Uppsala University, P.O. Box 591, 751 24 Uppsala, Sweden
2
Merck & Co. Inc, Kenilworth, NJ, USA
progression historically described the evolution of disease with total score [1–3]. In recent years, however, item response (IR) theory has gained increased interest in application within pharmacometric frameworks [4]. IR originated in the field of psychometrics where it is a common methodology for the evaluation of achievement test outcomes. IR analysis is a statistical methodolo
Data Loading...
