Back to Publications

General comments about Leclaire et al., article and the "Point of View"

Serge A. Gracovetsky, PhD and Nicholas M. Newman, MD.

To the Editor:

We would like to make the following comments about some remarks regarding the Spinex International's performance made in the study by Leclaire et a1.¹ and in the accompanying "Point of View."

Leclaire et al. state in their article that the Spinex International "has greater diagnostic accuracy in detecting simulation than does a single clinical evaluation." We must point out that the Spinex International did not "detect" simulators, but correctly diagnosed simulators' true conditions (normal/abnormal) in spite of simulation.

The capacity to recognize the true medical (normal/abnormal) status of the simulator (clinicians, 30%; Spinex International, 70%) should not be misconstrued as the ability to detect the intent (honest/dishonest) of the subject (clinicians, 25%; Spinex International, 18%)² Leclaire et al. confuse the detection of simulators with the determination of the true clinical status in spite of simulation. The data show that the clinicians and the Spinex International perform poorly at the former, but that the Spinex International performs relatively well at the latter. Therefore the added remark to the effect that the Spinex International did not achieve "a high level of success In detecting simulators," must refer to the intent and not to the clinical status of the simulator.

We agree that the Spinex International data are not intended to represent the sole source of information on which clinicians base their diagnoses; Spinoscopic functional data, like radiologic data are only one of the many possible elements that the clinician must interpret in his or her diagnosis. Leclaire's study data suggest that the Spinex International would be beneficial in cases of contradiction between the physical examination and the reports of the patient, where the accuracy of the clinical examination drops drastically.

To explain the clinicians' poor performance in recognizing the true medical condition of the simulators, it was suggested that the clinical exam in the study "assumes an honest and cooperative patient." This is not so The study protocol specifically advised the clinicians of the existence of simulators among the subjects and required these clinicians to classify each subject as honest or dishonest. Thus, the failure of the clinicians to correctly diagnose the true medical condition of the simulators cannot be attributed to ignorance nor to the assumption that the patient was honest and collaborating.

It was implied that if the medico-legal examination were more complete than the clinical examination performed in the study the clinician's performance would have improved. This interesting suggestion is not supported in the literature. In fact, none of the other techniques such as crosscheck, distraction testing, behavioral assessment, or non-organic physical signs appear to have been validated with a controlled group of simulators.³ Hence, there is no evidence that the addition of any test could have increased the clinicians' score for simulators. Because the clinicians were chosen specifically for their experience in the medico-legal evaluation of low back pain, their clinical examinations must be considered the best possible under the circumstances, and indeed they may well have implicitly used at least some of the additional techniques suggested.

Regarding the 99% score of the clinicians for the honest subjects, a close analysis of the data demonstrated that such a score can be obtained by asking a simple question "Do you have pain" and taking the answer at face value.² Therefore this high score does not represent evidence that the clinical examination is effective, a point noted by many and highlighted in this study by the inclusion of simulators.^4,5

It also was remarked that the authors' conclusions were in agreement those of with previous studies. Previous studies, presumably including that of Newton and Waddell,⁶ did not investigate the Spinex International, nor did they incorporate a group of simulators. Therefore there are no comparable studies in support of this statement.

Nor are there solid grounds for suggesting that, "despite the present findings, the evidence is not sufficient to justify the use of (Spinoscopy) for medico-legal purposes." On the contrary, in medico-legal cases where simulation is an issue, Leclaire's study quantified the shortcomings of the clinical examination and demonstrated the need for additional information beyond what the unaided clinician can provide. Knowing that a patient can or cannot lift 20 kg may be considerably more useful than knowing he is tender at L4-LS.²

References

1. Leclaire R, Esdaile JM, Jequier JC, Hanley JA, Rossignol M, Bourdouxhe M. Diagnostic accuracy. of technologies used in low back pain assessment: Thermography, Triaxial Dynamometry, Spinoscopy and Clinical Examination. Spine 1996;21: 1325-1331

2. Quebec Worker's Compensation Board. Evaluation Clinique experirnentale de Ia validit� d'instruments pour Ia diagnostique de loinbagies: Spinoscopie, dynanornetrie triaxiale (B-200) et thermographie, en regard de l'examen clinique (1995). Original clinical data are available from the undersigned and the Quebec WCB division (IRSST).

3. Waddell C, McCulloch IA, Kummel E, Verner MB. Non-organic physical signs in Iow-back pain. Spine 1980;5: 117-125.

4. Spitzer W0, Leblanc FE, Dupuis M. Scientific approach to the assessment and management of activity-related spinal disorders. Spine 1987;12(Suppl):1-59.

5. Van den Hoogen HMM, Koes BW, van EikJThM, Bouter LM. On the accuracy of history, physical examination, and erythrocyte sedimentation rate in diagnosing low back pain in general practice. Spine 1595;20: 318-27.

6. Newton M, Waddell G. Trunk strength testing with iso-machines. Part 1: Review of a decade of scientific evidence. Spine 1993;18: 801-824.