ABSTRACT
One of the most commonly observational study designs employed in veterinary is the cross-sectional study with binary outcomes. To measure an association with exposure, the use of prevalence ratios (PR) or odds ratios (OR) are possible. In human epidemiology, much has been discussed about the use of the OR exclusively for case-control studies and some authors reported that there is no good justification for fitting logistic regression when the prevalence of the disease is high, in which OR overestimate the PR. Nonetheless, interpretation of OR is difficult since confusing between risk and odds can lead to incorrect quantitative interpretation of data such as "the risk is X times greater," commonly reported in studies that use OR. The aims of this study were (1) to review articles with cross-sectional designs to assess the statistical method used and the appropriateness of the interpretation of the estimated measure of association and (2) to illustrate the use of alternative statistical methods that estimate PR directly. An overview of statistical methods and its interpretation using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines was conducted and included a diverse set of peer-reviewed journals among the veterinary science field using PubMed as the search engine. From each article, the statistical method used and the appropriateness of the interpretation of the estimated measure of association were registered. Additionally, four alternative models for logistic regression that estimate directly PR were tested using our own dataset from a cross-sectional study on bovine viral diarrhea virus. The initial search strategy found 62 articles, in which 6 articles were excluded and therefore 56 studies were used for the overall analysis. The review showed that independent of the level of prevalence reported, 96% of articles employed logistic regression, thus estimating the OR. Results of the multivariate models indicated that logistic regression was the method that most overestimated the PR. The findings of this study indicate that although there are methods that directly estimate PR, many studies in veterinary science do not use these methods and misinterpret the OR estimated by the logistic regression.
ABSTRACT
The purpose of this study was to use mixed logistic regression to evaluate the association between Neospora caninum serostatus and abortion, accounting for the effects of farms. The main objective was to explore the interpretation of this model and discuss the contribution of this analysis to our understanding of the disease's epidemiology. A mixed-effects logistic model using farms as a random effect and the serostatus for N. caninum, age of cattle and farm location as fixed effects was performed. The data from 1256 cows over 15 months of age from 60 farms were used, and the abortion information was obtained from farm records. A significant association between N. caninum serostatus and abortion was found (p<0.0001). Seropositive cows had 6.63 times greater odds of having a history of previous abortion (95% CI: 4.35-13.37). There was remarkable variability across farms in the probability of a cow having a history of an abortion. Including the effects of the farms in the regression, it was possible to estimate an intraclass correlation coefficient (ICC) of 16%. That means that 16% of the variation in abortion occurrence that was not explained by the fixed effects was due to farms. In practical terms, this variation means that while there are farms with several seropositive cows and no/few abortion cases, the opposite is also true, with a high number of abortions in farms with low/medium seroprevalence. The remaining variability (84%) was not explained by the effect of N. caninum, age, region, and the effect of farms, i.e., it is due to unknown factors that are causing abortions. The results of this study reinforce the importance of N. caninum as a cause of abortions and demonstrate the significant heterogeneity in the probability of a cow having a history of an abortion that cannot be explained completely by N. caninum serostatus, age or location. Including the hierarchical structure of the population along with correct interpretation of the model estimates helps us understand the effect of the farms, i.e., the probability of a cow abortion conditional to the farms, which also contributes to our understanding of the epidemiology of abortions caused by neosporosis. The use of hierarchical models and reporting the ICC should be encouraged.
