49

Be careful of using the risk ratio. An odds ratio is meaningful at any prevalence, but a risk ratio can produce estimates of risk that are greater than 100% in cases where the baseline prevalence is high enough. For example, where the baseline risk is 60% a relative risk of 2 gives a predicted risk of 120%.

Dear Ronán Michael Conroy,

Thanks for you answer, but how should we interpret  odds ratio of 0.529?

A risk ratio is a good measure to use for a meta-analysis if you have data from longitudinal cohorts or clinical trials.  It is generally thought to be easier to interpret than an odds ratio. However, if the outcome is rare (incidence of

Dear Keturah Faurot ,

Thanks for your answer, I understood form your answer that odds ratio is better than risk ratio, isnt it ?

The incidence of tumor  about 18% to 33%

Dear Dr. Al-Moraissi,

It still depends on what sort of data you have available.  At my university,  the risk difference or risk ratio is preferred with an incidence of 18-33% in cohort/RCT data.  However, if the studies in your meta-analysis include case-control data, use the odds ratio instead.

With an incidence that high, the odds ratio diverges from the risk ratio and cannot be interpreted as an estimate of the risk ratio.  However, if you prefer the odds ratio, it is not wrong to use it--you just have to interpret it as an odds ratio.

Hope this helps.

Dear Keturah Faurot,

Really, thank you for your helps

How should we interpret odds ratio of 0.529?

This means that the risk factor is protective with a 47% reduction in the odds of the disease.

How should we interpret odds ratio of 0.529?

When the odds ratio is less than "1" that means the case/group/sample or whatever is less likely to experience the event.

Please go through the link.

Good Luck.

https://www.researchgate.net/post/How_to_interpret_odds_ratios_that_are_smaller_than_1

Odds ratios are used to compare the relative odds of the occurrence of the outcome of interest (e.g. disease or disorder), given exposure to the variable of interest (e.g. health characteristic, aspect of medical history). The odds ratio can also be used to determine whether a particular exposure is a risk factor for a particular outcome, and to compare the magnitude of various risk factors for that outcome.

OR=1 Exposure does not affect odds of outcome

OR>1 Exposure associated with higher odds of outcome

OR

There are researchers who argue that you should not report an Odds Ratio because practioners in particular, but academics just as well, interpret them as Risk Ratio's (see Zhang and Yu, 1998). However, Fleiss & Berlin (2009) argue that statistically, the Odds Ratio is preferable for meta-analysis. For more details on this check our publication. In the paper we also introduce an idea for executing the meta-analysis in (Log) Odds Ratio and then transforming the results back in Risk Ratio/Difference, which may achieve the best of both: good statistics and good interpretation.

Good Luck!

Zhang, J. and Yu, K. F. (1998) ‘What’s the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes.’, JAMA : the journal of the American Medical Association, 280(19), pp. 1690–1. doi: 10.1001/jama.280.19.1690.

Fleiss, J. L., & Berlin, J. A. (2009). Effect sizes for dichotomous data. In H. Cooper, L. V. Hedges & J. C. Valentine (Eds.), The handbook of research synthesis and meta-analysis (Second Ed.) (pp. 237-253). New York, NY: Russell Sage Foundation

Article Meta-Analyze Dichotomous Data: Do the Calculations with Log ...

odds ratio is calculated for case control studies, while relative risk is for cohort studies, so it depends on the study design.

RR is more sensitive than OR!

Use:

RR: Longitudinal cohort study; Clinical trials; If incidence more than 10%;

Comparing before and after treatment.

OR: Cross-sectional study; Case-Control study; Comparing between Main and Control group.

RR and OR: Rare cases in incidence less than 10%.