Back to presentation overview

Title: Responsiveness and Minimal Clinically Important Difference for Pain and Disability Instruments in Low Back Pain Patients.

Authors and affiliation: Henrik H. Lauridsen, DC, MSc1, Jan Hartvigsen, DC, PhD1,2, Claus Manniche, MD, DMSc3, Lars Korsholm, PhD4, Niels Grunnet-Nilsson DC, MD, PhD1

1 Clinical Locomotion Science, Institute of Sports Science and Clinical Biomechanics, University of Southern Denmark.
2 Nordic Institute of Chiropractic and Clinical Biomechanics, part of Clinical Locomotion Science.
3 Backcenter Funen, Ringe, part of Clinical Locomotion Science.
4 Department of Statistics, University of Southern Denmark, Odense.

Email address: hlauridsen@health.sdu.dk

Introduction: The proliferation of similar and well-validated back-specific questionnaires has made the choice of a proper instrument for a given situation challenging. However, concurrent comparisons of psychometric properties of these instruments in subpopulations of low back pain (LBP) patients are rare. Many authors advocate that the property of sensitivity to change (responsiveness) should be central in the choice of an evaluative instrument, and commonly reported indices are standardised response mean (SRM), area under the receiver operating characteristic curve (ROCauc), and minimal clinically important difference (MCID). Furthermore, the MCID has been found to depend on baseline score. Whether other target attributes such as patient entry point into the health care system or pain location affect the meaning of change is unknown.

Objectives: 1) To conduct a head-to-head comparison of responsiveness and to calculate the MCID for pain and disability measures in subpopulations of LBP patients. 2) To determine the dependence of the MCID on baseline entry score, patient entry point and pain location.

Methods: The Danish versions of the Oswestry Disability Index (ODI), the 23-item Roland Morris Disability Questionnaire (RMQ), the physical function and bodily pain subscales of the SF36, the pain and disability subscales of the Low Back Pain Rating Scale (LBPRS) and a numerical rating scale for pain (0-10) were completed by 93 patients from the primary sector (PrS) of the Danish health care system and 97 patients from the secondary sector (SeS). Clinical change was estimated using a 7-point transition question and a numeric rating scale for importance (0-10). Responsiveness was operationalised using SRM, ROCauc, and cut-point analysis for optimal sensitivity and specificity. Subpopulation analysis of SRM and MCID was carried out on PrS and SeS patients, and on patients with LBP only and LBP and/or leg pain.

Results: The most responsive outcome measure proved to be the RMQ in PrS and LBP only patients (SRM = 2.1-3.2; ROCauc = 0.82-0.84) whereas the ODI was marginally more responsive in SeS and LBP and/or leg pain patients (SRM = 0.3-1.4; ROCauc = 0.78-0.85). The disability subscale of the LBPRS showed poor responsiveness in 3 out of 4 subpopulations. All pain measures demonstrated similar responsiveness in the subpopulations. The MCID increased with increasing baseline score in PrS and LBP only patients but was only slightly affected by patient entry point and pain location. Most instruments were dependent on baseline entry score and/or pain location when classifying patients as having received an “important improvement” or not.

Conclusion: The RMQ is suitable for measuring change in PrS and LBP only patients whereas the ODI and RMQ are suitable for SeS and LBP and/or leg pain patients. The MCID is baseline score dependent but only in certain subpopulations, and optimal classification of improved patients depends on baseline score and/or pain location. Baseline and subpopulationspecific MCIDs will be presented.