Common Registration Assessment, Part 1
Relative risk and relative risk reduction
Risk, relative risk and what a 95% confidence interval tells you. Three original questions with the working shown.
Published 6 October 2026. Independent revision material, not GPhC questions, not for patient care.
The June 2026 key insights from the GPhC named medical statistics among the Part 1 areas where candidates showed room to improve. Relative risk is the starting point, because the other measures are built from it.
The Pharmaceutical Journal's series on evaluating studies defines risk as the number of events divided by the number of people in the group. Relative risk compares the treatment group with the control group, and relative risk reduction is the proportion by which the risk falls.
The method
- Work out the risk in each group: events divided by the number of people in that group.
- Divide the treatment risk by the control risk to get relative risk (RR).
- Find relative risk reduction with 1 minus RR, or (control rate minus treatment rate) divided by the control rate.
- Check any confidence interval: if it contains 1, the RR is not statistically significant.
Three practice questions
Work each one on paper first, then open the answer. All drugs and patients are fictional.
Question 1
In a trial, 18 of 600 patients taking a new anticoagulant had a stroke, compared with 30 of 500 patients in the control group. What is the relative risk of stroke with the new anticoagulant? These figures are made up for the exercise.
Show the answer
Answer: 0.5
Working: Risk with the drug = 18 ÷ 600 = 0.03. Risk in the control group = 30 ÷ 500 = 0.06. RR = 0.03 ÷ 0.06 = 0.5. The denominators are the group sizes, which are different here, so work out each risk before dividing.
Question 2
An event occurs in 8.0% of patients on a treatment and in 12.5% of patients on control. What is the relative risk reduction, as a percentage? These figures are made up for the exercise.
Show the answer
Answer: 36%
Working: RR = 8.0 ÷ 12.5 = 0.64. RRR = 1 − 0.64 = 0.36, which is 36%. The second method gives the same answer: (12.5 − 8.0) ÷ 12.5 = 0.36.
Question 3
A study reports an RR of 1.30 for bleeding with a new drug compared with control (95% confidence interval 0.95 to 1.78). Is the finding statistically significant, and what does the RR say about the direction of risk? These figures are made up for the exercise.
Show the answer
Answer: Not significant; the risk is higher with the drug
Working: The interval 0.95 to 1.78 includes 1, the value that means no difference between groups, so the finding is not statistically significant. An RR above 1 means the risk is higher in the treatment group, so the point estimate suggests a 30% relative increase in bleeding, but the interval allows for no difference.
What the journal article covers
The article defines risk, relative risk, relative risk reduction, absolute risk reduction, number needed to treat and odds ratio, with worked examples on a migraine trial, apixaban against warfarin and a diabetes drug. Its apixaban example explains why an RR whose confidence interval includes 1 is not significant. The questions on this page use different numbers.
Where marks are lost
- Using the same denominator for both groups when the group sizes differ.
- Putting the control risk on top, which turns an RR below 1 into one above 1.
- Calling a result significant because the point estimate is below 1 when the interval contains 1.
Frequently asked questions
What does an RR of 1 mean?
The risk is the same in the treatment and control groups, so there is no difference in effect.
When is a relative risk not statistically significant?
When its 95% confidence interval includes 1. For a difference between two means or two rates, the equivalent value is 0.
Sources
- The Pharmaceutical Journal, understanding risk and clinical utility, dropouts, data handling and the discussion section when evaluating studies (11 September 2026)
- GPhC, Common Registration Assessment specification and permitted items for 2026
The questions above are original and use fictional drugs. PreRegExamPrep is not affiliated with or endorsed by the General Pharmaceutical Council.
Practise until the method is automatic
Try 15 free questions with worked answers. No sign-up required.
More calculation topics
- Dose by weight calculations for the GPhC assessment
- Volume to give: liquid medicine calculations
- Percentage strength calculations: w/v, w/w and ratios
- Dilution calculations with C1V1 = C2V2
- Infusion rate calculations in mL per hour
- Drip rate calculations: drops per minute
- Body surface area calculations
- Creatinine clearance (Cockcroft-Gault) calculations
- Moles and millimoles calculations
- Displacement volume calculations for reconstitution
- Quantity to supply calculations
- Using a provided formula in calculations
- Dose information from packaging and labels
- Diluting a stock to a lower strength
- Medicine cost and switch calculations
- Rounding rules in calculations
- Enteral feed rate calculations
- Electrolyte content over time
- Infusion dose per kg per hour
- Rounding doses to measurable volumes
- Absolute risk reduction and number needed to treat
- Odds ratio and confidence intervals
- Relative versus absolute risk reduction
- Loading dose and volume of distribution
- Top-up doses from measured drug levels
- Salt correction factors
- Mass and molar drug concentrations
- Half-life and falling drug levels
- Steady-state concentration and infusion rate
See also the formula sheet, the eight sample questions and the approved calculator page.