The Odds Ratio is “portable” across baseline risk but not the Relative Risk: Time to do away with the log link in binomial regression (original) (raw)
Related papers
Annals of Epidemiology, 2002
In cohort studies of common outcomes, odds ratios (ORs) may seriously overestimate the true effect of an exposure on the outcome of interest (as measured by the risk ratio [RR]). Since few study designs require ORs (most frequently, case-control studies), their popularity is due to the widespread use of logistic regression. Because ORs are used to approximate RRs so frequently, methods have been published in the general medical literature describing how to convert ORs to RRs; however, these methods may produce inaccurate confidence intervals (CIs). The authors explore the use of binomial regression as an alternative technique to directly estimate RRs and associated CIs in cohort studies of common outcomes. METHODS: Using actual study data, the authors describe how to perform binomial regression using the SAS System for Windows, a statistical analysis program widely used by US health researchers. RESULTS: In a sample data set, the OR for the exposure of interest overestimated the RR more than twofold. The 95% CIs for the OR and converted RR were wider than for the directly estimated RR. CONCLUSIONS: The authors argue that for cohort studies, the use of logistic regression should be sharply curtailed, and that instead, binomial regression be used to directly estimate RRs and associated CIs.
Expression and interpretation of relative risk and odds ratio in biomedical research studies
Indian Journal of Community Health
Relative risk and odds ratio are commonly used in the biomedical research studies; however, expression and interpretation must be done very carefully. A risk ratio and an odds ratio are used in cohort studies but only odds ratio is used in case control studies. However, relative risk or risk ratio is found to be frequently used in the interventional biomedical research studies. The relative risk and odds ratio provide important information regarding the effect of a risk factor on the outcome of interest. The relative risk and odds ratio of 1 suggests that there is no difference between two groups. A value >1 suggests increase risk, while a value <1 suggest reduction of risk. If the confidence interval meets or includes value 1.00 (line of no difference) indicates there is no difference between the groups.
Prevalence odds ratio versus prevalence ratio: choice comes with consequences
Statistics in medicine, 2016
Odds ratio, risk ratio, and prevalence ratio are some of the measures of association which are often reported in research studies quantifying the relationship between an independent variable and the outcome of interest. There has been much debate on the issue of which measure is appropriate to report depending on the study design. However, the literature on selecting a particular category of the outcome to be modeled and/or change in reference group for categorical independent variables and the effect on statistical significance, although known, is scantly discussed nor published with examples. In this article, we provide an example of a cross-sectional study wherein prevalence ratio was chosen over (Prevalence) odds ratio and demonstrate the analytic implications of the choice of category to be modeled and choice of reference level for independent variables. Copyright © 2016 John Wiley & Sons, Ltd.
International Journal of Environmental Research and Public Health
Evidence-based medicine is predicated on the integration of best available research evidence with clinical expertise and patient values to inform care. In medical research, several distinct measures are commonly used to describe the associations between variables, and a sound understanding of these pervasive measures is foundational in the clinician’s ability to interpret, synthesize, and apply available evidence from the medical literature. Accordingly, this article aims to provide an educational tutorial/topic primer on some of the most ubiquitous measures of association and risk quantification in medical research, including odds ratios, relative risk, absolute risk, and number needed to treat, using several real-world examples from the medical literature.
An odd measure of risk: use and misuse of the odds ratio
Obstetrics & Gynecology, 2001
To determine how often the odds ratio, as used in clinical research of obstetrics and gynecology, differs substantially from the risk ratio estimate and to assess whether the difference in these measures leads to misinterpretation of research results. METHODS: Articles from 1998 through 1999 in Obstetrics & Gynecology and the American Journal of Obstetrics and Gynecology were searched for the term "odds ratio." The key odds ratio in each article was identified, and, when possible, an estimated risk ratio was calculated. The odds ratios and the estimated risk ratios were compared quantitatively and graphically. RESULTS: Of 151 studies using odds ratios, 107 were suitable to estimate a risk ratio. The difference between the odds ratio and the estimated risk ratio was greater than 20% in 47 (44%) of these articles. An odds ratio appears to magnify an effect compared with a risk ratio. In 39 (26%) articles the odds ratio was interpreted as a risk ratio without explicit justification. CONCLUSION: The odds ratio is frequently used, and often misinterpreted, in the current literature of obstetrics and gynecology.
Measures of effect: relative risks, odds ratios, risk difference, and ‘number needed to treat’
2007
Epidemiological studies aim at assessing the relationship between exposures and outcomes. Clinicians are interested in knowing not only whether a link between a given exposure (e.g. smoking) and a certain outcome (e.g. myocardial infarction) is statistically significant, but also the magnitude of this relationship. The 'measures of effect' are indexes that summarize the strength of the link between exposures and outcomes and can help the clinician in taking decisions in every day clinical practice. In epidemiological studies, the effect of exposure can be measured both in relative and absolute terms. The risk ratio, the incidence rate ratio, and the odds ratio are relative measures of effect. Risk difference is an absolute measure of effect and it is calculated by subtracting the risk of the outcome in exposed individuals from that of unexposed.
Journal of dairy science, 2012
The objectives were (1) to explain why the risk ratio (RR) is an appropriate measure of association when the outcome of interest is dichotomous (e.g., displaced abomasum or no displaced abomasum) in both cohort studies and randomized trials; and (2) to outline an applied method for estimating the RR using currently available software. Interest in the association between multiple risk factors and a yes or no outcome is very common in the dairy industry; historically, logistic regression, which reports odds ratios (OR), was the method available in common statistical packages to evaluate this kind of association. However, the OR can overestimate the magnitude of the response in cohort studies and randomized trials when the outcome frequency is large. In addition, the interpretation of odds is not intuitive; fortunately, recent advances in statistical software have allowed the estimation of the RR. Because SAS software (SAS Institute Inc., Cary, NC) is commonly used to analyze data, thi...