Tanvetyanon is correct in noting that there are differences in patient characteristics in the groups with and without a chiropractic benefit that may have contributed to some of the effects seen in the study.1 These limitations are in fact noted in our article. To be sure, these differences had a high level of statistical significance, as they would have to be given the extremely large sample size. However, to characterize the different rates of comorbid conditions as “much more prevalent” is not accurate. In fact, the magnitudes of these differences are rather modest, and while there may be alternative methods to control for these differences, the methods used in our study were robust and our conclusions are reasonable.
Legorreta AP. Medical Costs Associated With Chiropractic Care—Reply. Arch Intern Med. 2005;165(11):1314. doi:10.1001/archinte.165.11.1314-a