The Winton Centre Quick Quiz for Medical Professionals
How well can you navigate the numbers about the potential harms and benefits of different treatments? Can you detect evidence of efficacy from statistical sleight of hand?
Diagnose your skill level with our quick test...
The Winton Centre for Risk and Evidence Communication
On a visit to London, Rudy Giuliani (former Mayor of New York) upset NHS lovers, saying:
‘I had prostate cancer seven years. My chance of survival in the US is 82%; my chance of survival if I was here in England is below 50%.’
He was quoting 5-year survival figures for prostate cancer in the US and UK. From these, Giuliani concluded that treatment for prostate cancer is better in the US than in the UK.
Why is it wrong to draw this conclusion from 5 year survival figures alone?
Your patient Wilfred is trying to decide whether he should get screened for prostate cancer with the specific antigen (PSA) test.
He asks you: ‘Is there reason to think that screening saves lives?’
What evidence do you need to answer Wilfred's question?
The number of cancers detected in screened populations and unscreened populations
The 5-year survival rate for cancer detected because of symptoms, and the the 5-year survival rate for cancer detected by screening.
The mortality rates among screened populations and comparable unscreened populations
The post-test positive value of the PSA screening test to establish how accurate it is at detecting aggressive cancers.
In 1995, there was a major scare over the contraceptive pill: research revealed that it doubled the risk of potentially fatal venous thrombosis.
What statistic is being reported here?
In early 2019, newspapers reported that taking HRT (Hormone Replacement Therapy) during the menopause doubles the risk of blood clots.
Unusually, a lot of the coverage also quoted the researchers pointing out that: ‘For every 10,000 women who take HRT pills, there are just nine extra cases of clots every year.’
This number is helpful. What is it?
Number needed to treat
Percentage point increase
Increase in absolute risk
Odds Ratios are often used to report outcomes in medical trials and epidemiological studies.
Odds ratios are calculated by comparing two simple odds. What are the simple odds of throwing a 6 with a fair dice?
In 2015, the International Agency for Research on Cancer classified bacon and other processed meats as Class 1 carcinogens. This was meant to indicate the strength of the evidence, rather than the size of the effect, but a lot of confused and misleading headlines followed….
The relative risk was widely reported: eating processed meat associated with 18% greater risk of bowel cancer.
In general, we can expect 6 out of every 100 people to get bowel cancer. So out of 100 people who regularly eat bacon, roughly how many would we expect to get bowel cancer?
7 out of 100
17 out of 100
24 out of 100
27 out of 100
Imagine a new drug for osteoporosis has just been approved.
Clinical trials found that 10% of untreated patients with osteoporosis sustained a hip fracture during a 3 year monitoring period, compared with 5% of the patients who took the new drug every day over the same period.
What number of patients need to be treated with the drug to prevent one hip fracture?
Your patient, a 50 year old woman called Mia, had a mammogram as part of a routine screening program rolled out across the country. The result is positive.
Mia is upset, but she knows that screening tests are not infallible.
According to the current NHS screening leaflet, what proportion of women with a positive mammogram do in fact have breast cancer?
We are collecting the results of people’s answers to these quizzes for our research. It would be useful for us to know if you are:
A student health professional
A health professional
None of the above
Click to SUBMIT your answers and get feedback
Want more? Check out our fully accredited & free online courses on 'Communicating potential harms and benefits' for GPs, Surgeons & perioperative specialists here :
You can find out more about the Winton Centre for Risk and Evidence Communication via our website -
- or you can contact us at
This form was created inside of University of Cambridge.