EFFECT OF DIFFERENT SAMPLING TECHNIQUES ON ODDS RATIO ESTIMATES USINGHOSPITAL-BASED CASES AND CONTROLS

Potential biases introduced by the use of hospital admission records h ave rarely been discussed in the veterinary literature. Veterinary Med ical Teaching Hospital (VMTH) patient records kept at the University o f California, Davis (UCD) School of Veterinary Medicine provide a uniq ue opportunity to perform in-depth analyses on the effect of different control selection (sampling) techniques on odds ratio (OR) estimates for disease risk factors in a retrospective case-control study. Horses with Corynebacterium pseudotuberculosis abscesses (134) and the (seco ndary) study base population (source for controls) were identified, an d a 'gold standard' OR for each category of the factors admission type , age, breed and sex was derived. Example data were used to calculate sampling ratios (SRs), defined as the ratio between any sample proport ion (of an arbitrary risk factor) and the study base proportion for th is risk factor. Sampling ratios different from 1.0 introduced biases i nto the observed OR estimates, when compared with the 'gold standard' OR. Three randomized samples (simple random, stratified random, system atic 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 f our factors were derived. The matched and two different disease sample s ('colic' and 'fractures') had especially wide ranges of observed SRs (and large errors in the OR estimates), whereas simple random and sys tematic sampling had comparably narrow ranges (less biased OR estimate s). For the three randomized sampling techniques under study, repeated sampling was used to derive SR distributions. The SRs were approximat ely normally distributed. Analysis of variance and covariance showed t hat simple random and systematic sampling provided SR distributions wi th means closest to 1.0 (expected value) and small standard deviations . The OR estimates obtained from records selected by these two samplin g techniques therefore were least biased. The findings demonstrate the importance of selecting appropriate sampling techniques in addition t o properly defining the study (base) population. Sampling design intro duces uncertainty into the OR estimates. The direction of the bias, ho wever, depends on the OR between factor and disease in the source popu lation (the 'gold standard'), and on the direction and magnitude of th e SR. When combining the results from single and repeated sampling we conclude that sampling design is most influential on the range of the observed SRs (single samples), on the absolute deviation of the SR fro m 1.0 (expressed as SR Delta Mean) and on the SR standard deviation (S D) (repeated sampling). (C) 1997 Elsevier Science B.V.