⚡ Quick Navigation
You’ve run your analysis and out popped an odds ratio of 1.79. So what does that actually tell you? In short, it means the odds of the outcome in the exposed group are 79% higher than in the unexposed group (since 1.79 - 1 = 0.79). But that simple translation hides a ton of nuance. I’ve spent the last decade teaching statistics to medical residents, and I can tell you that 1.79 is often the most misinterpreted number in a paper. So let’s break it down properly.
In this guide, I’ll explain exactly what an odds ratio of 1.79 means, walk through a real-world example, show you why it’s not the same as relative risk, and point out the errors that even experienced researchers make. I’ll also give you the exact wording you can use in your own write-up.
What Does 1.79 Mean in Plain English?
An odds ratio (OR) compares the odds of an event happening in one group to the odds of it happening in another. If the OR is 1.0, the odds are identical. If it’s greater than 1, the exposure is associated with higher odds. So an OR of 1.79 means the odds in the exposed group are 1.79 times the odds in the unexposed group. That's a 79% increase in odds, not risk.
Many people conflate odds and probability. Let me untangle that. Probability is the chance of an event among all possible outcomes. Odds is the ratio of the event happening to it not happening. For example, if 20 out of 100 people get a disease, the probability is 0.20, but the odds is 20/80 = 0.25. So when we say 'odds are 79% higher', we're talking about that ratio, not the overall chance.
Here's where it gets tricky: For rare events (say, less than 10% occurrence), odds and probability are close, and an OR of 1.79 approximates a 79% higher risk. But if the event is common, the OR can exaggerate the risk. I'll show you that with numbers later.
How to Interpret 1.79: A Step-by-Step Example
Let’s use a hypothetical clinical trial. I’ve seen people jump straight to the result without checking the data structure, so let’s do this clearly.
Scenario: A new drug is tested to reduce cardiovascular events. The table below shows the number of patients who experienced an event (heart attack) and those who didn't.
| Group | Event | No Event | Odds |
|---|---|---|---|
| Drug | 179 | 1000 | 179/1000 = 0.179 |
| Placebo | 100 | 1000 | 100/1000 = 0.100 |
Step 1: Calculate the odds in each group.
Step 2: Divide the drug group odds by the placebo group odds: 0.179 / 0.100 = 1.79.
So the drug group has 1.79 times the odds of having a cardiovascular event compared to the placebo group. In other words, the odds are 79% higher. But look at the actual event rates: 15.2% in the drug group (179/1179) and 9.1% in the placebo group (100/1100). The relative risk (RR) is 15.2/9.1 = 1.67. So the OR (1.79) is higher than the RR (1.67) because the event is not super rare. This gap is exactly why you need to be careful about what you claim.
In a published paper, you might write: 'The odds of a cardiovascular event were higher in the drug group than in the placebo group (OR = 1.79; 95% CI ...).' But you should not say 'the risk increased by 79%' unless you're reporting an RR from a randomized trial where you can directly estimate risk.
Odds Ratio vs Relative Risk: Why 1.79 Might Not Be What You Think
The difference between OR and RR is not just academic. It changes the message. RR compares probabilities directly; OR compares odds. For rare events, they’re nearly equal. For common events, OR diverges and can make an effect look bigger than it is.
Let me show you with a simple table. Suppose the event rate in the unexposed group is 10%. To get an OR of 1.79, the exposed group event rate would be about 16.3% (check: odds = 16.3/83.7 = 0.195, 0.195/0.111 = 1.75, close). The RR would be 1.63. If the baseline rate is 50%, then an OR of 1.79 corresponds to an odds of 1.79*1 = 1.79, so probability = 1.79/(1+1.79) = 64.2%. The RR is 64.2/50 = 1.28. So a 79% increase in odds is only a 28% increase in risk. That’s a huge difference.
This is why the American Statistical Association and major medical journals (like the BMJ) insist that authors clearly state whether they're reporting odds or risk. In cohort studies with common outcomes, reporting ORs alone can be misleading. I always advise clinical researchers to also calculate RR or absolute risk reduction when the event rate >10%.
When Is an OR of 1.79 Statistically Significant?
An odds ratio of 1.79 is just a point estimate. The real question is whether the confidence interval (CI) includes 1.0. If the 95% CI for your OR includes 1.0, then you cannot confidently say there is an association. For instance, if the CI is 0.98 to 3.27, the result is not statistically significant, even though the point estimate is 1.79. Many people overlook this and get excited too early.
On the other hand, if the CI is 1.12 to 2.86, then the association is significant at the 0.05 level. The width of the CI also tells you about precision. A wide interval like 1.01 to 3.17 suggests a weak and imprecise estimate. I remember a resident once told me: 'But the OR is 1.79, isn't that big?' I had to explain that 'big' depends on the confidence interval and the context. In epidemiology, an OR of 1.79 might be considered a small to moderate effect, depending on the exposure and outcome.
Briefly, statistical significance is not about the size of the OR but whether the CI crosses 1. So always report and interpret the CI alongside the point estimate.
Common Mistakes When Interpreting an OR of 1.79
In my years reviewing manuscripts, I've spotted the same errors over and over. Here are the ones to avoid:
Mistake 1: Treating OR as RR. This is the most common. If you say 'the risk is 79% higher' when you only have an OR from a case-control study, you're overstating the effect. The correct phrasing is 'the odds are 79% higher.'
Mistake 2: Ignoring the baseline risk. An OR of 1.79 means something different if the baseline risk is 0.1% vs 40%. Always supply the absolute numbers or event rates.
Mistake 3: Skipping the confidence interval. I once had a student submit a paper with 'OR = 1.79' and no CI. That’s incomplete. The CI matters as much as the estimate.
Mistake 4: Claiming causation from an observational OR. A couple of years ago, I consulted for a company that found an OR of 1.79 between a certain gene and a disease in a cross-sectional study. They wanted to launch a diagnostic test based on that. I had to stop them. An OR from an observational study does not prove the exposure causes the outcome. It could be due to confounding or reverse causation.
FAQ: Odds Ratio 1.79
My thesis committee asks me to report an OR of 1.79 as a percentage increase. How do I phrase it?
Say 'The odds of the outcome were 79% higher in the exposed group compared to the unexposed group (OR 1.79).' Avoid the word 'risk' unless you are reporting a relative risk from a prospective cohort or a randomized trial. You can also add 'as indicated by an odds ratio of 1.79' to be precise.
If the 95% CI for my OR of 1.79 ranges from 0.95 to 3.37, can I still talk about a meaningful trend?
No. When the CI crosses 1.0, the result is not statistically significant at the conventional 0.05 level. You cannot claim a meaningful association. You can mention it as a 'non-significant trend' but be cautious. Reviewers might criticize this as warping the conclusion. Better to report the CI and let the data speak.
Why does my OR of 1.79 become non-significant after adjusting for confounders?
This usually means that part of the apparent effect was due to confounding variables. In your adjusted model, the OR may shrink toward 1.0 or the CI may widen because you've used up degrees of freedom and the confounders explain some of the variability. It’s not a mistake; it's evidence of confounding. Report both crude and adjusted ORs and discuss what factors drove the change.
So an odds ratio of 1.79 is a meaningful but context-dependent number. It tells you that the odds are 79% higher, but that doesn't automatically translate to a 79% higher risk, nor does it guarantee significance. Always pair the OR with its confidence interval, present absolute risks when possible, and choose your words carefully. If you follow these guidelines, you'll avoid the pitfalls that trip up even experienced researchers.
