HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Extract Full Text (PDF) Submit a response Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Freedman, V. A. Articles by Martin, L. G. Search for Related Content PubMed Articles by Freedman, V. A. Articles by Martin, L. G.

Vicki A. Freedman, PhD and Linda G. Martin, PhD

Vicki A. Freedman is with the Polisher Research Institute, Philadelphia Geriatric Center, Jenkintown, Pa. Linda G. Martin is with the Population Council, New York, NY.

Correspondence: Requests for reprints should be sent to Vicki A. Freedman, PhD, Polisher Research Institute, 261 Old York Rd, Suite 427, PO Box 728, Jenkintown, PA 19046 (e-mail: vfreedman{at}pgc.org).

We thank Wang and Badley for their comments, but we believe that the issues they raise as methodological concerns are better interpreted as differences in disciplinary approaches and that the validity of our analysis remains intact.

Their main concern is that we did not translate decomposition results into population attributable risk estimates, a concept long used in epidemiology. Yet demographers have a tradition of using decomposition techniques to distinguish whether a population rate is changing because of shifts in population composition or because of shifts in rates among subgroups. Wang and Badley are correct that the corresponding prevalence rates and coefficients can be used to calculate adjusted population attributable risks, but the fact that we did not do so simply reflects a difference in focus.

Similarly, their suggestion that we should have explored population effects has no substantive implications. Indeed, their points regarding osteoporosis simply mirror what we wrote: "Older Americans were increasingly likely to report having osteoporosis" and "osteoporosis . . . became significantly less debilitating over time" (p 1757). Moreover, the decomposition analysis allowed statistical testing of whether these patterns contributed to improvements in functioning.

Their concerns that comparisons of coefficients across surveys require more caution than we advised (on pp 1756–1757, which included a meticulous accounting of survey changes, as well as caution about omissions), that results should be shown as percentages rather than decimals, that we reported but did not emphasize the poor fit of the model for upper-body limitations, and that P values should be reported at the .01 level rather than higher levels (reflecting their greater concern for type I over type II error) are questions of space limits and taste. More troubling to us is their calculation of much higher P values for changes in 3 conditions reported in Table 2. Without more detail on how they computed their values, it is not possible to reconcile the differences. Substantively, however, we noted (on p 1757) that 2 of the 3 conditions they cite did not increase beyond what would be expected as a result of population shifts, and thus we did not emphasize them in our discussion of the results.

In sum, we believe the substantive conclusion of our paper—that functioning has improved despite increased reports of chronic disease—is not altered by the concerns of Wang and Badley. We look forward to their and others' work on this important topic.

This Article Extract Full Text (PDF) Submit a response Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Freedman, V. A. Articles by Martin, L. G. Search for Related Content PubMed Articles by Freedman, V. A. Articles by Martin, L. G.