| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||
LETTER TO THE EDITOR |
Columbia University Medical Center New York, New York
Address correspondence to Huai Yong Cheng, MD, MPH, The Allen Pavilion, Columbia University Medical Center, 5141 Broadway, New York, NY 10034-1159. E-mail: hyc2105{at}columbia.edu
To the Editor:
Lower socioeconomic status (SES) was associated with elevated levels of several inflammatory markers in a study published by Dr. Koster and her colleagues in a recent issue of the Journal (1). They also found that unhealthy behaviors such as smoking and disease status failed to explain completely the association between SES and inflammatory markers (1). I believe that the analytical approach used by Dr. Koster and colleagues is only partially satisfactory, and additional findings could have been obtained by the full use of all the potentials of regression analysis. For example, inclusion of standardized regression coefficient for SES-related variables and confounders in Tables 3–5 could have been used to rank the hierarchical importance of the independent variables in prediction of the inflammatory marker. In addition, a careful examination of the partial R2 attributable to each SES-related variable or confounder may have provided insight on the percentage of variability of the inflammatory markers that could be "explained" by each independent variable or combination of variables. Standardized regression coefficient and R2 are useful and powerful parameters in multiple regression analysis. They should be fully utilized to explore the complicated relationship between inflammatory markers and SES, including the effect of potential confounders.
References
| ||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
