Methods for shortening patient-reported outcome measures

Daphna Harel, Murray Baron

Research output: Contribution to journalArticle

Abstract

Patient-reported outcome measures are widely used to assess patient experiences, well-being, and treatment response in clinical trials and cohort-based observational studies. However, patients may be asked to respond to many different measures in order to provide researchers and clinicians with a wide array of information regarding their experiences. Collecting such long and cumbersome patient-reported outcome measures may burden patients, increase research costs, and potentially reduce the quality of the data collected. Nonetheless, little research has been conducted on replicable, and reproducible methods to shorten these instruments that result in shortened forms of minimal length. This manuscript proposes the use of mixed integer programming through Optimal Test Assembly as a method to shorten patient-reported outcome measures. This method is compared to the existing standard in the field, which is selecting items based on having high discrimination parameters from an item response theory model. The method is then illustrated in an application to a fatigue scale for patients with Systemic Sclerosis.

Original languageEnglish (US)
JournalStatistical Methods in Medical Research
DOIs
StateAccepted/In press - Jan 1 2018

Fingerprint

Systemic Sclerosis
Optimal Test
Observational Study
Systemic Scleroderma
Mixed Integer Programming
Model Theory
Research
Clinical Trials
Fatigue
Discrimination
Observational Studies
Research Personnel
Costs and Cost Analysis
Patient Reported Outcome Measures
Costs
Experience
Therapeutics
Standards
Form
Data Accuracy

Keywords

  • generalized partial credit model
  • Item response theory
  • optimal test assembly
  • patient-reported outcome measure
  • shortened form

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability
  • Health Information Management

Cite this

Methods for shortening patient-reported outcome measures. / Harel, Daphna; Baron, Murray.

In: Statistical Methods in Medical Research, 01.01.2018.

Research output: Contribution to journalArticle

@article{0b2373ef9ad945169366b4134f2f637d,
title = "Methods for shortening patient-reported outcome measures",
abstract = "Patient-reported outcome measures are widely used to assess patient experiences, well-being, and treatment response in clinical trials and cohort-based observational studies. However, patients may be asked to respond to many different measures in order to provide researchers and clinicians with a wide array of information regarding their experiences. Collecting such long and cumbersome patient-reported outcome measures may burden patients, increase research costs, and potentially reduce the quality of the data collected. Nonetheless, little research has been conducted on replicable, and reproducible methods to shorten these instruments that result in shortened forms of minimal length. This manuscript proposes the use of mixed integer programming through Optimal Test Assembly as a method to shorten patient-reported outcome measures. This method is compared to the existing standard in the field, which is selecting items based on having high discrimination parameters from an item response theory model. The method is then illustrated in an application to a fatigue scale for patients with Systemic Sclerosis.",
keywords = "generalized partial credit model, Item response theory, optimal test assembly, patient-reported outcome measure, shortened form",
author = "Daphna Harel and Murray Baron",
year = "2018",
month = "1",
day = "1",
doi = "10.1177/0962280218795187",
language = "English (US)",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",

}

TY - JOUR

T1 - Methods for shortening patient-reported outcome measures

AU - Harel, Daphna

AU - Baron, Murray

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Patient-reported outcome measures are widely used to assess patient experiences, well-being, and treatment response in clinical trials and cohort-based observational studies. However, patients may be asked to respond to many different measures in order to provide researchers and clinicians with a wide array of information regarding their experiences. Collecting such long and cumbersome patient-reported outcome measures may burden patients, increase research costs, and potentially reduce the quality of the data collected. Nonetheless, little research has been conducted on replicable, and reproducible methods to shorten these instruments that result in shortened forms of minimal length. This manuscript proposes the use of mixed integer programming through Optimal Test Assembly as a method to shorten patient-reported outcome measures. This method is compared to the existing standard in the field, which is selecting items based on having high discrimination parameters from an item response theory model. The method is then illustrated in an application to a fatigue scale for patients with Systemic Sclerosis.

AB - Patient-reported outcome measures are widely used to assess patient experiences, well-being, and treatment response in clinical trials and cohort-based observational studies. However, patients may be asked to respond to many different measures in order to provide researchers and clinicians with a wide array of information regarding their experiences. Collecting such long and cumbersome patient-reported outcome measures may burden patients, increase research costs, and potentially reduce the quality of the data collected. Nonetheless, little research has been conducted on replicable, and reproducible methods to shorten these instruments that result in shortened forms of minimal length. This manuscript proposes the use of mixed integer programming through Optimal Test Assembly as a method to shorten patient-reported outcome measures. This method is compared to the existing standard in the field, which is selecting items based on having high discrimination parameters from an item response theory model. The method is then illustrated in an application to a fatigue scale for patients with Systemic Sclerosis.

KW - generalized partial credit model

KW - Item response theory

KW - optimal test assembly

KW - patient-reported outcome measure

KW - shortened form

UR - http://www.scopus.com/inward/record.url?scp=85053288386&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053288386&partnerID=8YFLogxK

U2 - 10.1177/0962280218795187

DO - 10.1177/0962280218795187

M3 - Article

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

ER -