Measuring individual true change with PROMIS using IRT-based plausible values

Emily H. Ho*, Jay Verkuilen, Felix Fischer

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

Aims: A primary advantage of IRT-based patient-reported outcome measures such as PROMIS short forms and computer-adaptive tests is that each estimate of the latent trait comes with a standard error. Such measurement error needs to be acknowledged, in particular when monitoring individual patients over time. In this study, we use plausible values to account for measurement error and analyze the probability of true within-individual change. Methods: We use a longitudinal, observational study of stable and exacerbated COPD patients (N = 185), providing PROMIS Physical Function and Fatigue T-scores over 3 months. At each measurement, we imputed 1000 plausible values from the scores’ posterior distribution. These were then used to calculate probability of true change using a pre-specified threshold such as minimally important difference supported by the literature, or Δ T- score > 0. We demonstrate assessment of change in individuals and in groups, across different measures (Short Forms and CATs), and at various levels of confidence. Results: Using plausible value imputation and with 95% certainty, 47.5% of participants in the exacerbated group reported less fatigue, compared with 26.5% of participants in the stable group. Comparison of Short Forms and CATs suggests that CATs have better ability to detect change compared to short forms. We also illustrate this method using an individual’s probability of change at different time points. Conclusion: Plausible values offer a flexible way to include measurement error in analysis of individuals and on sample level. Assessment of probability of true change can complement existing distribution-based approaches and facilitates interpretation of improvement or decline.

Original languageEnglish (US)
Pages (from-to)1369-1379
Number of pages11
JournalQuality of Life Research
Volume32
Issue number5
DOIs
StatePublished - May 2023

Keywords

  • COPD
  • Chronic illness
  • Item response theory
  • Meaningful change
  • PROMIS
  • Plausible values

ASJC Scopus subject areas

  • Public Health, Environmental and Occupational Health

Fingerprint

Dive into the research topics of 'Measuring individual true change with PROMIS using IRT-based plausible values'. Together they form a unique fingerprint.

Cite this