Volume 28, Issue 8 , Page 637, October 2005
The Bournemouth Questionnaire: Can it be Used to Monitor and Predict Treatment Outcome in Chiropractic Patients With Persistent Low Back Pain?
Article Outline
To the Editor:
I am writing to dispute the conclusion that the Bournemouth Questionnaire (BQ) is not a useful instrument1 to monitor progress based on data that in the main, relate to properties of the instrument in predicting (and not monitoring) outcomes. This instrument was never designed to be used, nor as far as I am aware has ever been proffered, as either a diagnostic or a predictive instrument. Thus, to test it in this way, and then to make the leap and imply that it is also not useful as an outcome measure, seems to me to be unfair, biased, and unjustified.
Specifically, the BQ, when tested against the revised Oswestry questionnaire in this study, performed as well as the Oswestry in both responsiveness and predicting outcomes. A simple calculation of effect sizes2 from the data presented in table 3 illustrates comparable responsiveness and, by the authors' own admission, no difference in predictive value (although by modifying the BQ, it was possible to predict “fairly well” all 4 of the outcomes tested).
The authors also investigated concurrent validity of the questionnaires. My interpretation was that the questionnaires agreed very well, with very small mean score differences at all time points. The authors stated that, “At all points in time, there was a small mean difference between the sum scores…” However, as they correctly observed, there was a wider scatter in the differences as the average increased (an observation not uncommon in these types of analyses). This later appeared as, “There is considerable disagreement between the …. at all times meaning that one or both of the questionnaires produce unreliable data.” Finally, by assuming that the Oswestry is the criterion standard (a biased assumption because there is no way of knowing the “truth”), the authors imply that the BQ is unreliable. This is not supported by the data. Only well-designed replication studies can judge the reliability of an instrument. A final point of concern is that the authors failed to reference any rigorous validity studies of the translated versions of the questionnaires, thus calling into question the accuracy of the data collected by these instruments in either this or in any other studies.
I do agree with the authors, however, when they state that patients' responses should be categorized using cutoff scores to enable calculation of numbers needed to treat. The authors may be interested to know that this has already been done for the BQ.3, 4
The authors state that it is their opinion (and one I concur with) that the scientific and clinical world should not be further burdened with additional instruments unless they add to it. The BQ adds to it by virtue of being an outcome measure not only with robust psychometric properties comparable with other outcome measures but also one that is shorter and arguably more practical and one that can be used in the confidence that it measures a wide spectrum of outcomes. This view is reflected in the number of chiropractors, at least in the UK, who do find the BQ useful and use it routinely to monitor their patients' progress. I trust that in spite of the contestable conclusions of this paper, they will continue to do so.
References
- . The Bournemouth Questionnaire: can it be used to monitor and predict treatment outcome in chiropractic patients with persistent low back pain?. J Manipulative Physiol Ther. 2005;28:219–227
- . Effect sizes for interpreting changes in health status. Med Care. 1989;27:S178–S189
- . Assessing the clinical significance of change scores recorded on subjective outcome measures. J Manipulative Physiol Ther. 2004;27:26–35
- . Sensitivity and specificity of outcome measures in patients with neck pain: detecting clinically significant improvement. Spine. 2004;29:2410–2417
PII: S0161-4754(05)00229-0
doi:10.1016/j.jmpt.2005.08.007
© 2005 National University of Health. Published by Elsevier Inc. All rights reserved.
Refers to article:
- The Bournemouth Questionnaire: Can it Be Used to Monitor and Predict Treatment Outcome in Chiropractic Patients With Persistent Low Back Pain?
Volume 28, Issue 8 , Page 637, October 2005
