University of Illinois at Chicago
Browse

Minimum Clinically Important Difference in Medical Studies

Download (420.86 kB)
journal contribution
posted on 2016-08-01, 00:00 authored by Hedayat AS, Wang J, Xu T
In clinical trials, minimum clinically important difference (MCID) has attracted increasing interest as an important supportive clinical and statistical inference tool. Many estimation methods have been developed based on various intuitions, while little theoretical justification has been established. This paper proposes a new estimation framework of the MCID using both diagnostic measurements and patient-reported outcomes (PRO’s). The framework first formulates the population-based MCID as a large margin classification problem, and then extends to the personalized MCID to allow individualized thresholding value for patients whose clinical profiles may affect their PRO responses. More importantly, the proposed estimation framework is showed to be asymptotically consistent, and a finite-sample upper bound is established for its prediction accuracy compared against the ideal MCID. The advantage of our proposed method is also demonstrated in a variety of simulated experiments as well as two phase-3 clinical trials.

Funding

Research is supported in part by The U.S. National Science Foundation Grants DMS- 0904125 (Hedayat) and DMS-1306394 (Hedayat).

History

Publisher Statement

This is the pre-peer reviewed version of the following article: Hedayat, A. S., Wang, J. and Xu, T. Minimum clinically important difference in medical studies. Biometrics. 2015. 71(1): 33-41. 10.1111/biom.12251.

Publisher

Wiley

Language

  • en_US

issn

0006-341X

Issue date

2015-03-01

Usage metrics

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC