Relative risk / Odds ratio
Written by Ronny Gunnarsson and first published on May 5, 2015.
Last revised on June 13, 2020.
There is an updated version of the EPV-calculator found here: https://infovoice.se/epv-calculator/
(This page is still under construction. Sorry for the inconvenience.)
Odds ratio (OR) and relative risk ratio (RR) are typically measures of an association between two binary variables. However, odds ratio can also denote the increased odds for a specified change in an independent variable (read more about this on the page for logistic regression).
In a cohort study an exposure may be associated with a later increased or decreased risk for a specific disease. In this case these associations are often labelled “risk odds ratio (ROR)” and “risk ratio (RR)”.
In cross-sectional studies where the prevalence of one binary variable is associated with the simultaneous prevalence of another binary variable the same measures are often labelled “prevalence odds ratio (POR)” and “prevalence ratio (PR)” (the latter is sometimes also referred to as “prevalence rate ratio”).
OR and ROR are mathematically the same. Similarly, the RR and PR are mathematically the same.
Relative risk is often considered a better estimate of association, at least when the prevalence of any of the variables is above 10% . However, odds ratio is most often uses since most statistical software delivers odds ratios when doing logistic regression.
Useful links
References
{2262766:8UKRBI9Q}
vancouver
default
asc
no
181
%7B%22status%22%3A%22success%22%2C%22updateneeded%22%3Afalse%2C%22instance%22%3A%22zotpress-597f7d2ec3d3ec8baaaf2c5d20ead1ef%22%2C%22meta%22%3A%7B%22request_last%22%3A0%2C%22request_next%22%3A0%2C%22used_cache%22%3Atrue%7D%2C%22data%22%3A%5B%7B%22key%22%3A%228UKRBI9Q%22%2C%22library%22%3A%7B%22id%22%3A2262766%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Martinez%20et%20al.%22%2C%22parsedDate%22%3A%222017-11-10%22%2C%22numChildren%22%3A2%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E1.%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EMartinez%20BAF%2C%20Leotti%20VB%2C%20Silva%20G%20de%20S%20e%2C%20Nunes%20LN%2C%20Machado%20G%2C%20Corbellini%20LG.%20Odds%20Ratio%20or%20Prevalence%20Ratio%3F%20An%20Overview%20of%20Reported%20Statistical%20Methods%20and%20Appropriateness%20of%20Interpretations%20in%20Cross-sectional%20Studies%20with%20Dichotomous%20Outcomes%20in%20Veterinary%20Medicine.%20Front%20Vet%20Sci%20%5BInternet%5D.%202017%20Nov%2010%20%5Bcited%202019%20Jun%2026%5D%3B4.%20Available%20from%3A%20%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.ncbi.nlm.nih.gov%5C%2Fpmc%5C%2Farticles%5C%2FPMC5686058%5C%2F%27%3Ehttps%3A%5C%2F%5C%2Fwww.ncbi.nlm.nih.gov%5C%2Fpmc%5C%2Farticles%5C%2FPMC5686058%5C%2F%3C%5C%2Fa%3E%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Odds%20Ratio%20or%20Prevalence%20Ratio%3F%20An%20Overview%20of%20Reported%20Statistical%20Methods%20and%20Appropriateness%20of%20Interpretations%20in%20Cross-sectional%20Studies%20with%20Dichotomous%20Outcomes%20in%20Veterinary%20Medicine%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Brayan%20Alexander%20Fonseca%22%2C%22lastName%22%3A%22Martinez%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Vanessa%20Bielefeldt%22%2C%22lastName%22%3A%22Leotti%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Gustavo%20de%20Sousa%20e%22%2C%22lastName%22%3A%22Silva%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Luciana%20Neves%22%2C%22lastName%22%3A%22Nunes%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Gustavo%22%2C%22lastName%22%3A%22Machado%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Lu%5Cu00eds%20Gustavo%22%2C%22lastName%22%3A%22Corbellini%22%7D%5D%2C%22abstractNote%22%3A%22One%20of%20the%20most%20commonly%20observational%20study%20designs%20employed%20in%20veterinary%20is%20the%20cross-sectional%20study%20with%20binary%20outcomes.%20To%20measure%20an%20association%20with%20exposure%2C%20the%20use%20of%20prevalence%20ratios%20%28PR%29%20or%20odds%20ratios%20%28OR%29%20are%20possible.%20In%20human%20epidemiology%2C%20much%20has%20been%20discussed%20about%20the%20use%20of%20the%20OR%20exclusively%20for%20case%5Cu2013control%20studies%20and%20some%20authors%20reported%20that%20there%20is%20no%20good%20justification%20for%20fitting%20logistic%20regression%20when%20the%20prevalence%20of%20the%20disease%20is%20high%2C%20in%20which%20OR%20overestimate%20the%20PR.%20Nonetheless%2C%20interpretation%20of%20OR%20is%20difficult%20since%20confusing%20between%20risk%20and%20odds%20can%20lead%20to%20incorrect%20quantitative%20interpretation%20of%20data%20such%20as%20%5Cu201cthe%20risk%20is%20X%20times%20greater%2C%5Cu201d%20commonly%20reported%20in%20studies%20that%20use%20OR.%20The%20aims%20of%20this%20study%20were%20%28%29%20to%20review%20articles%20with%20cross-sectional%20designs%20to%20assess%20the%20statistical%20method%20used%20and%20the%20appropriateness%20of%20the%20interpretation%20of%20the%20estimated%20measure%20of%20association%20and%20%28%29%20to%20illustrate%20the%20use%20of%20alternative%20statistical%20methods%20that%20estimate%20PR%20directly.%20An%20overview%20of%20statistical%20methods%20and%20its%20interpretation%20using%20the%20Preferred%20Reporting%20Items%20for%20Systematic%20Reviews%20and%20Meta-Analyses%20%28PRISMA%29%20guidelines%20was%20conducted%20and%20included%20a%20diverse%20set%20of%20peer-reviewed%20journals%20among%20the%20veterinary%20science%20field%20using%20PubMed%20as%20the%20search%20engine.%20From%20each%20article%2C%20the%20statistical%20method%20used%20and%20the%20appropriateness%20of%20the%20interpretation%20of%20the%20estimated%20measure%20of%20association%20were%20registered.%20Additionally%2C%20four%20alternative%20models%20for%20logistic%20regression%20that%20estimate%20directly%20PR%20were%20tested%20using%20our%20own%20dataset%20from%20a%20cross-sectional%20study%20on%20bovine%20viral%20diarrhea%20virus.%20The%20initial%20search%20strategy%20found%2062%20articles%2C%20in%20which%206%20articles%20were%20excluded%20and%20therefore%2056%20studies%20were%20used%20for%20the%20overall%20analysis.%20The%20review%20showed%20that%20independent%20of%20the%20level%20of%20prevalence%20reported%2C%2096%25%20of%20articles%20employed%20logistic%20regression%2C%20thus%20estimating%20the%20OR.%20Results%20of%20the%20multivariate%20models%20indicated%20that%20logistic%20regression%20was%20the%20method%20that%20most%20overestimated%20the%20PR.%20The%20findings%20of%20this%20study%20indicate%20that%20although%20there%20are%20methods%20that%20directly%20estimate%20PR%2C%20many%20studies%20in%20veterinary%20science%20do%20not%20use%20these%20methods%20and%20misinterpret%20the%20OR%20estimated%20by%20the%20logistic%20regression.%22%2C%22date%22%3A%222017-11-10%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%2210.3389%5C%2Ffvets.2017.00193%22%2C%22ISSN%22%3A%222297-1769%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.ncbi.nlm.nih.gov%5C%2Fpmc%5C%2Farticles%5C%2FPMC5686058%5C%2F%22%2C%22collections%22%3A%5B%5D%2C%22dateModified%22%3A%222019-06-26T11%3A31%3A25Z%22%7D%7D%5D%7D
1.
Martinez BAF, Leotti VB, Silva G de S e, Nunes LN, Machado G, Corbellini LG. Odds Ratio or Prevalence Ratio? An Overview of Reported Statistical Methods and Appropriateness of Interpretations in Cross-sectional Studies with Dichotomous Outcomes in Veterinary Medicine. Front Vet Sci [Internet]. 2017 Nov 10 [cited 2019 Jun 26];4. Available from:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5686058/
There is an updated version of the EPV-calculator found here: https://infovoice.se/epv-calculator/