Effect of different sampling techniques on odds ...
|Title||Effect of different sampling techniques on odds ratio estimates using hospital-based cases and controls|
|Author(s)||M. G. Doherr, T. E. Carpenter, W. D. Wilson, I. A. Gardner|
|Journal||Preventive Veterinary Medicine|
|Abstract||The effects of different control selection (sampling) techniques on odds ratio (OR) estimates for disease risk factors were analysed in a retrospective case-control study using patient records kept at the School of Veterinary Medicine, University of California, USA. 134 horses with Corynebacterium pseudotuberculosis abscesses and control horses were identified. Example data were used to calculate sampling ratios (SRs), defined as the ratio between any sample proportion (of an arbitrary risk factor) and the study base proportion for this risk factor. Sampling ratios different from 1.0 introduced biases into the observed OR estimates compared with the standard OR. Three randomized samples (simple random, stratified random, systematic sampling), one matched (on date of admission) and three different diagnosis samples (colic, cuts and lacerations, fractures) were selected from the study base, and the SRs for all categories of the 4 factors were derived. The matched and 2 different disease samples (colic and fractures) had especially wide ranges of observed SRs (and large errors in the OR estimates), whereas simple random and systematic sampling had comparably narrow ranges (less biased OR estimates). For the 3 randomized sampling techniques, repeated sampling was used to derive SR distributions. The SRs were approximately normally distributed. Analysis of variance and covariance showed that simple random and systemic sampling provided SR distributions closest to 1.0 and small standard deviations. It is concluded that sampling design is most influential on the range of the observed SRs, the absolute deviation of the SR from 1.0 and on the SR standard deviation.|
Using APA 6th Edition citation style.
Times viewed: 198