Annual visits to GPs by elderly patients
CMAJ 1998;158:299
Re: "The health of Canada's elderly population: current status and future implications" (CMAJ 1997;157[8]:1025-32 [full text / résumé]), by Drs. Mark W. Rosenberg and Eric G. Moore
In response to: M.M. Finkelstein; P.P. Wang & E.M. Badley
We acknowledge that there was a problem in Table 5, which we missed during proofreading — odds ratios are indeed always positive.
As Drs. Wang and Badley point out, the values in Table 5 are regression coefficients, not odds ratios. They were derived according to a procedure similar to the one Dr. Finkelstein outlines, not logistic regression. The heading for column 2 of Table 5 is mislabelled and should read instead "Added number of visits." In our model the dependent variable was the number of added visits to a GP and the independent variables were the medical conditions and the socioeconomic variables identified in Table 5. When all other independent variables are controlled for, the analysis yields the regression coefficients shown in Table 5. The interpretation of the numbers is consistent with the text on page 1030. For example, having Alzheimer's disease increases the number of annual visits to a GP by 1.87, when all other conditions and socioeconomic circumstances are controlled for. The crucial point is that this correction does not change the general argument that specific conditions increase the likelihood that elderly people will make more visits to their GP. We might also add that even though Dr. Finkelstein was working with different data, his results complement ours.
We apologize to CMAJ readers for any confusion caused by Table 5 and the corresponding discussion but would remind them that this unfortunate error in no way negates either the general argument we make about health and socioeconomic conditions of the elderly population and their meaning for the utilization of physician services or any of the other aspects of our article.
Mark W. Rosenberg, PhD
Eric G. Moore, PhD
Professors
Department of Geography
Queen's University
Kingston, Ont.
 |
Send a letter to the editor responding to this letter
Envoyez une lettre à la rédaction au sujet de cette lettre |
Letters