Association measures in cross-sectional studies with complex sampling: odds ratio and prevalence ratio

Revista Brasileira de Epidemiologia

Francisco, Priscila Maria S. Bergamo³  Goldbaum, Moisés¹  Donalisio, Maria Rita³  Barros, Marilisa Berti de Azevedo³  Cesar, Chester Luis Galvão¹  Carandina, Luana² 

[1] USP;UNESP;UNICAMP

关键词: Cross-sectional studies;    Odds ratio;    Prevalence ratios;   

DOI  :  10.1590/S1415-790X2008000300002

学科分类：过敏症与临床免疫学

来源: SciELO

【 摘 要 】

The objective forthis paper was to present and discuss the use of odds ratios and prevalenceratios using real data with a complex sampling design. We carried out a cross-sectionalstudy using data obtained from a two-stage stratified cluster sample from astudy conducted in 2001-2002 (n = 1,958). Odds ratios and prevalence ratioswere obtained by unconditional logistic regression and Poisson regression, respectively,for later comparison using the Stata statistical package (v. 7.0). Confidenceintervals and design effects were considered in the evaluation of the precisionof estimates. Two outcomes of a cross-sectional study with different prevalenceswere evaluated: vaccination against influenza (66.1%) and self-referred lungdisease (6.9%). In the high-prevalence scenario, using prevalence ratios theestimates were more conservative and we found narrower confidence intervals.In the low-prevalence scenario, we found no important numeric differences betweenthe estimates and standard errors obtained using the two techniques. A designeffect greater than one indicates that the sample design has increased the varianceof the estimate. However, it is the researcher's task to choose which techniqueand measure to use for each data set, since this choice must remain within thescope of epidemiology.

【 授权许可】

CC BY