I am writing in response to an article recently published by Côté et al in Arthritis Care & Research (1). The title, Early aggressive care and delayed recovery from whiplash: isolated finding or reproducible result? would suggest that the authors were able to determine that “early aggressive care,” a term that was not operationally defined, in some way delayed patient recovery in this cohort. This category of “aggressive care” indicated only that the patients were seen by both general practitioners and chiropractors. Since, according to the authors, the group treated with aggressive care had more serious whiplash injuries, this does not seem surprising.
In their conclusion, the authors inform the reader that “combining chiropractic and general practitioner care appears to confer no benefit to patients” (1). This conclusion, however, is based entirely on claim duration, which was defined as the time from the injury to the time of claim closure. No objective measure of “benefits” was actually collected. Moreover, claim closure is an administrative function of the insurer. Owing to the retrospective nature of this study and the artificial reference frame (claim closure) used for a proxy for recovery, the analysis of data and conclusions drawn from it are potentially skewed. This proxy was justified by the authors by referring to a previous study in which they found that claim closure roughly coincided with lower (but not absent) neck pain. In addition to the information bias introduced by misclassifying some symptomatic subjects as “recovered,” it is also likely that in some cases symptom recovery preceded claim closure. Because the present study was retrospective rather than longitudinal, the temporal relationship between clinical recovery and claim closure would have been opaque to the authors and potentially confounds the study with information bias.
Although total number of visits to practitioners is one method of looking at efficacy of treatment, this is only meaningful within the context of total or near total recovery. If recovery is not obtained, the treatment is not efficacious, regardless of the amount provided. Because approximately 30–50% of whiplash victims do not fully recover (2–4) and since the authors made no attempt to determine symptomatic recovery in this cohort, this measure becomes ambiguous. In addition to measures of symptomatic recovery such as the Neck Disability Index and other similar instruments, other meaningful outcome measures and tests of efficacy would include total number of sick leave days, total cost of care, cost of job retraining and other disability costs, and costs of over-the-counter medications. Such assessments were not made.
The authors' comment that “our study augments the evidence that too much health care too early after a soft tissue injury negatively influences the prognosis of whiplash patients” is unsupported by the study design and their findings. The retrospective nature of the study design requires some awkward further questioning, specifically, how did they determine whether treatment was either “too much” or “too early”? More importantly, one simply cannot validate or test a prognosis (a prediction of outcome) if the actual outcome is unknown.