| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Letters to the Editor |
Texas Woman’s University Denton, Texas 76204
Address correspondence to: Sydney Lou Bonnick, M.D., FACP, Texas Woman’s University, Denton, Texas 76204.
To the editor:
Dr. Cummings and the editorial staff of JCEM are to be congratulated for encouraging and continuing this very important dialogue. It is also appropriate in discussing the care of the individual patient, as Dr. Cummings has now done, to move from issues of regression to the mean, which have no relevance, to the issues of precision and least significant change (LSC). Dr. Cummings is not correct in suggesting that it was argued in my editorial that only "large" changes in BMD are meaningful (1). If the precision of testing is 1%, then a change from baseline of only 2.77% would be considered significant at the most stringent confidence level of 95%. This magnitude of change would hardly be considered "large." It is difficult to comment on the analysis described in Dr. Cummings’ letter. The magnitude of the LSC is determined by the precision of the technique at a particular skeletal site and the desired level of statistical confidence. In vivo precision values were not given in the original publication by Cummings et al. (2). If the precision at some skeletal site is 2% and the desired level of statistical confidence is 95%, the LSC for a single measurement made at baseline and follow-up would, in fact, be a change of 5.54% from baseline, not 4% as suggested by Dr. Cummings. Therefore, this analysis would include individuals who experienced no significant change in bone density at the end of 1 yr. If the physician anticipates a decline in bone mineral density (BMD) in the absence of therapy, no significant change in BMD may be considered efficacious. The physician would correctly conclude that no change in therapy is necessarily warranted. This is predicated, of course, on understanding the concepts of precision and the LSC. The reality is, however, if the precision at a particular site were a paltry 2%, one would not expect to meet or exceed the LSC in only 1 yr with most of the antiresorptive agents currently available. A testing interval of 2 yr is more appropriate when 95% confidence is desired and only one measurement is made at baseline and follow-up. Although Dr. Cummings does not state the magnitude of the increase in bone density in the second year in his analysis, it is quite likely that many of those individuals would have a change in BMD over the 2-yr interval that would be considered insignificant at the 95% confidence level. It would then remain up to the judgment of the physician as to whether changes in therapy are warranted. Finally, the probability of success with a given therapeutic agent as demonstrated in clinical trials is an excellent reason for initially choosing that agent to treat a particular disease. In the care of the individual patient, it would seem to be a poor reason to abandon reasonable efforts to assess efficacy.
Received April 25, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  R. S. Vasan, J. C. Evans, M. G. Larson, P. W.F. Wilson, J. B. Meigs, N. Rifai, E. J. Benjamin, and D. Levy Serum Aldosterone and the Incidence of Hypertension in Nonhypertensive Persons N. Engl. J. Med., July 1, 2004; 351(1): 33 - 41. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
