Facebook
Twitter
LinkedIn
Email
 |
Checking for direct PDF access through Ovid | |
Excerpt
The findings would suggest that the symptom patterns described in Appendix 1 are independent and would result in distinct classifications that might be used for initiating a rehabilitation strategy. The gross statistical analyses might lead one to accept this conclusion. However, one must consider the limitations associated with this study to place these claims in perspective. The patient pool did not represent a balance among all pain patterns. In fact, nearly two thirds of pain patterns among the patients were associated with pattern 1. Pattern 1 represented the most easily identified pattern and thus biased the results in favor of its intertester reliability. Examination of Table 2 suggests that performance among the other patterns between testers was not nearly as consistent. Thus, the results indicate that only one pain pattern (of those including a significantly large number of patients) had good intertester reliability. A patient population group with a different balance of low back disorder distribution might fare very differently in intertester agreement.
The achievements associated with this effort should be viewed in perspective. The claims of the authors only involve intertester reliability; they do not reflect on the ability of the method to help classify patients into clinically meaningful groups. Only large, independent test set, cross-sectional evaluations and prospective evaluations can address these issues.
