Although they make up only 10% of the AKT examination, statistical questions tend to be more discriminating. It is these discriminating questions that will help decide a pass mark for that cohort. A question everyone gets right is not a good discriminator and vice versa.
Preparing for this aspect of the exam will require a good understanding of commonly used statistical terms within evidence-based medicine.
You may already have this understanding, depending on your background. However you may find the principles challenging so having a basic understanding of the more commonly used terms should help you to achieve success.
This will include terms such as:
- Positive predictive values
- Negative predictive values
- P values
- Pre-test probability
- Post-test probability
- Numbers needed to treat (NNT)
- Numbers needed to harm (NNH)
- Odds ratios
- Absolute risk reduction
- Relative risk reduction
Also be prepared to interpret data on various graphs such as:
- Funnel plots
- Forrest plots
- Box and whisker plot
The above lists are not exhaustive, but are common examples used in AKT questions.
Ensure you familiarise yourself with the RCGP curriculum statements around critical appraisal and evidence-based practice.
This will not be an in-depth test of your evidence interpretation skills but will likely test basic application of your understanding of these terms. Preparation for this can be done in a number of ways. First make sure you are familiar with the terms if you have forgotten their meaning.
The following textbooks may help.
- How to Read a Paper
- Pocket Guide to Critical Appraisal
You may also wish to review any old notes from medical school if still in your possession.
It is important to then practice the statistical questions on one or more of the websites you choose to use for your preparation, for example onexamination.com, pastest.com or passmedicine.com. You may also wish to use practice questions on the RCGP website and review websites such as Bradford VTS and Pennine GP Training.
After you have attempted the questions, there will be explanations for each of the answers. This will also cover some of the more common terms used in evidence interpretation.
If you identify a common area that you are not performing so well at, read around that area.
If you have grasped the terms well or already have a good basic understanding of statistical terms, then you may wish to help other colleagues whom you know who may not understand certain statistical principles well. You could do this one to one, in a small group or even on a much larger scale within your training programme.
- Dr Singh is a GP trainer in Northumberland
Read the other articles in this series
- RCGP information on the AKT, including practice questions
- Resources to help GP trainees use the curriculum including self assessment ratings