I am editing a biomedical review article for publication and the authors would like to display the percentage value with the corresponding raw values from the data. How do I format this?

Example: "According to X report, 5 of 400 (1.25%) patients recieving CMV prophylaxis developed encephalitis."

Is this format correct?

I have also seen: "5 (1.25%) of 400 patients" "5 of 400 patients (1.25%)"

Any advice is appreciated!

  • 2
    Is there a style guide of your target publication or of the institution you are writing for? – Helmar Nov 13 '16 at 18:11
  • 1
    "5 of 400 patients, i.e. 1.25% receiving ..." – Graffito Nov 13 '16 at 20:06
  • I think that if the % immediately precedes "patients," it needs to be "% of patients." And that's awkward. So use "Reportedly, 5 of 400 patients receiving CMV prophylaxis developed encephalitis (1.25%)." But this is a matter of style. – jejorda2 Nov 14 '16 at 13:46

I think if you use your own critical judgement on this you will reach the best solution. The first of the alternatives you mention seems best to me:

"According to X report, 5 of 400 (1.25%) patients recieving CMV prophylaxis developed encephalitis."

Why? Because you can't say (x%) until you have written a of b, and any later in the sentence it is dissociated from the values.

However, perhaps a better solution would be:

"According to X report 1.25% (5 of 400) patients recieving CMV prophylaxis developed encephalitis."

Why? I assume that the primary interest is the overall percentage. The sample size is important, but it's more a qualifier, so it seems more natural to me to place it in parentheses.

Or course, context is important. If X report is being compared to Y report, then the focus might be on different sample sizes, in which case you would put the emphasis there, with something like:

"According to X study of 400 patients, 1.25% recieving CMV prophylaxis developed encephalitis."

So the important thing is not to follow some convention, but to use a sentence structure that conveys the information most clearly (and honestly).


For some reason I'm still unaware of, I receive the British medical journal The Lancet. In perusing their review articles, they used both your schemes. However, what is consistent, is a statement preceding the numerical values, giving their presentation a purpose. A savvy audience will probably want to know something additional to a percentage such as an odds ratio or a confidence interval. Once you start adding numbers in a professional journal, the readers will want to know how trustworthy your data is.

For example,

Collins, et. al., examined the rate of encephalitis in recently CMV-vaccinated patients ages 8-17. They found 5 of 400 patients (1.3%, odds ratio 2.5/1, 95% CI 0.74-0.81) became symptomatic using the BRAIN method.

Also, be careful with the number of significant digits. It's unlikely the study is adequately powered for 3 significant digits; you're better off sticking with 2 throughout, hence 1.3% rather than 1.25%.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.