Elsevier

The Spine Journal

Volume 12, Issue 11, November 2012, Pages 1035-1039
The Spine Journal

Technical Report
Clinimetric properties of the EuroQol-5D in patients with chronic low back pain

https://doi.org/10.1016/j.spinee.2012.10.030Get rights and content

Abstract

Background context

Clinimetric properties of the EuroQol-5D (EQ-5D) in patients with nonspecific chronic low back pain (CLBP) are largely unknown.

Purpose

To study the criterion validity, responsiveness, and minimal clinically important change (MCIC) of EQ-5D in patients with CLBP.

Study design

Prospective study design carried out in a multispecialist Spine Center in The Netherlands.

Patient sample

One hundred fifty-one patients with CLBP.

Outcome measures

Quality of life (QOL) was measured with EQ-5D, consisting of two scales: one scale measuring QOL with five categorical questions and the other measuring health state on a visual analog scale (0–100). Criterion measures were disability, measured with the Pain Disability Index (PDI) and the Roland Morris Disability Questionnaire (RMDQ), and pain intensity, measured with a numeric rating scale (NRS).

Methods

Pearson correlation coefficients between the EQ-5D and RMDQ, PDI, and NRS were calculated to test the criterion validity. Correlations were interpreted based on predefined criteria. Responsiveness of the EQ-5D was calculated with area under the receiver operating characteristics (ROC) curve. Minimal clinically important change was calculated with the optimal cutoff point under the ROC curve, and sensitivity and specificity were also calculated.

Results

Correlations between EQ-5D and criterion measures ranged between 0.39 and 0.59 and were considered moderate to good. Areas under the ROC curve ranged from 0.59 to 0.72 depending on the external criterion and EQ-5D subscale. The MCIC was 0.03 points for the categorical scales of the EQ-5D and 10.5 points for the EQ-5D visual analog scale.

Conclusions

The EQ-5D is a valid and responsive QOL scale in patients with CLBP.

Introduction

The EuroQol-5D (EQ-5D) [1] is a widely used instrument to assess health-related quality of life (QOL). It is used in cost-effectiveness research based on quality-adjusted life year (QALY) and is recommended by the Dutch Healthcare Insurance Board [2]. Pain, disability, and functional status may affect QOL differently for various diseases. Consequently, the clinimetric properties of EQ-5D may vary according to the disease studied. Interventions for patients with nonspecific chronic low back pain (CLBP) aim to reduce disability and improve functioning and participation, which in turn should be related to QOL. Consequently, EQ-5D should be related to (changes in) functioning and disability to be a valid measure of QOL in patients with CLBP. Clinimetric properties for these specific purposes of EQ-5D have been studied scarcely in the field of CLBP.

It is known that lower QOL is associated with efficiency loss and absenteeism in patients with low back pain [3]. Another study addressed the assumption of relatedness between QOL and functioning and found differences in the responsiveness of EQ-5D in studies among different diseases and countries. Hence, the primary outcome of the therapy does not always relate to QOL in the same manner. Consequently, specific regression coefficients (tarrifs) were constructed for different countries [4], [5]. Responsiveness of the EQ-5D was found to be moderate in a study in which exercise treatment was provided [6] and was sufficient for patients undergoing low-back surgery (area under the receiver operating characteristics [ROC] curve (AUC) of 0.77) [7]. One study reported the differences between pain, disability, and QOL after spinal surgery [8]. In general, responsiveness of the EQ-5D appeared higher in acute back care than in CLBP [6], [7], [9]. To make valid decisions concerning QALYs, however, the primary outcome measure between interventions should be similar, namely QOL. The relation of QOL to reported disability, pain, or functioning should therefore be similar between different diseases. It is, however, unknown if EQ-5D correlates sufficiently with these primary outcomes. Additionally, there is insufficient data concerning responsiveness and minimal clinically important change (MCIC) in a broad perspective of CLBP interventions. The objectives of this study were to test criterion validity, responsiveness, and MCIC of the EQ-5D in patients with CLBP.

Section snippets

Procedures

The Groningen Spine Center is a secondary and tertiary care multidisciplinary center for patients with spinal disorders, that is, part of the University Medical Center Groningen, in the Netherlands. Disciplines involved were one or more of the following: rehabilitation medicine, neurology, and anesthesiology. Interventions were chosen based on medical status and patient preference and included outpatient multidisciplinary rehabilitation with psychology, physiotherapy, and occupational therapy

Results

Data of 151 patients were analyzed. Fifty-three percent received a minimal intervention, 27% rehabilitation, 19% anesthesiology (injections or medication), and 2% of the patients received a combination of treatments. Baseline characteristics are presented in Table 1. At baseline, no patient reached the ceiling or floor of the EQ-5D. At T1, a ceiling score was reached by six patients (4%). At baseline, 15% had one or more missing items on the PDI and 13% on the RMDQ and at T1, 21% had one or

Discussion

The main purpose of this study was to determine the clinimetric properties of the EQ-5D in patients with nonspecific CLBP. From the data of this study, the criterion validity of EQ-5D was confirmed. Responsiveness of EQ-5D is borderline sufficient (AUC=0.71). For the EQ-5D, a small OCP of 0.03 points was found, meaning that a change of 0.03 points in QOL is related to a change on the GPE with a sensitivity of 0.73 and specificity of 0.66. In clinical practice, this finding means that any

References (20)

There are more references available in the full text version of this article.

Cited by (113)

View all citing articles on Scopus

FDA device/drug status: Not applicable.

Author disclosures: RS: Research Support (Investigator Salary): University of Groningen (D); Research Support (Staff/Materials): University of Groningen (D). MFR: Nothing to disclose. BLGNS: Nothing to disclose. MHC: Nothing to disclose. PCAJV: Royalties: Publisher de Tijdstroom (A); Consulting: Boehringer (A); Speaking/Teaching Arrangements: Boehringer (A).

The disclosure key can be found on the Table of Contents and at www.TheSpineJournalOnline.com.

View full text