Minimum Clinical Important Differences Of Health Outcomes In A Chronic Pain Population: Are They Predictive Of Poor Outcomes?

Date

2008-04-22T02:41:28Z

Authors

Journal Title

Journal ISSN

Volume Title

Publisher

Psychology

Abstract

Psychometric validation of health outcomes measures ensures that the methods utilized to evaluate treatment effects, and aid in individual patient diagnosis are reliable, valid, and meaningful. A relatively new concept within the psychometric process of validation is the assessment of responsiveness, or the ability of an instrument to detect clinically meaningful change. Clinically meaningful change may be defined through subjective, self-reports of change, physician-based assessment, or through objective outcome criteria. The purpose of the current study was to evaluate clinically meaningful changes in chronic pain health outcome measures, as defined by objective outcome criteria. The average percent change in the Oswestry Disability Index, Million Visual Analog Scale, Short-form 36, Pain Disability Questionnaire, Pain Intensity Scale, and Beck Depression Inventory, were calculated for patients categorized as having Poor, Fair, and Good 1-year socioeconomic outcomes. The predictive ability of the percent change scores were evaluated through logistic regression analysis. Percent difference in BDI and MVAS were predictive of outcome status when combined with pre-treatment scores and age. No other percent difference variables were predictive of outcomes at the individual patient level, negating the application of an MCID for use in a clinical setting for the ODI, PDQ, PI, and BDI. However, a variety of additional pre and post measures were predictive of outcome status. The PDQ was able to predict poor outcomes better than any other scale, and the Pain Intensity, MVAS, and BDI were superior at detecting good outcomes. By combining scales such as MVAS and PI (better sensitivity), which are better at classifying good outcomes, and scales such as the PDQ and ODI, which are better at discriminating among patients that have poor outcomes (better specificity), superior identification of patients at greatest risk for poor outcomes may be realized. The prevalence of pain, critical role of health outcome measures in the field of medicine, and evaluation of MCIDs as a critical aspect in the validation process of measurement scales, highlights the importance of the current project.

Description

Keywords

Citation