You may also like

problem icon

Data Matching

Use your skill and judgement to match the sets of random data.

problem icon

Coin Toss Checker

Can you make a set of random data which will fool the computer?

problem icon

David and Goliath

Does weight confer an advantage to shot putters?

Sleep: the Silent Killer

Age 16 to 18 Challenge Level:
Newpapers often report on scientific announcements, especially if they are health related.

The Daily Mirror proclaimed on its front page on 7th August 2018:


TOO MUCH SLEEP IS DEADLY
More than 8 hours may be sign of ill-health as study shows 'long sleepers' die early

The online version of the article begins with a similar (but more detailed) headline and subheading:

Getting too much sleep could be deadly as study warns more than ten hours a night increases stroke risk
Researchers have urged GPs to check the sleep patterns of patients after shocking study results
 
Based on these headlines alone, what do you think the findings were of the study reported in the article?

Here are two more extracts from the article:

Researchers said underlying diseases could be the cause of regular lie ins – and they urged GPs to check the sleep patterns of patients during visits in a bid to save their lives.

 
“Our study has an important public health impact in that it shows that excessive sleep is a marker of elevated cardiovascular risk. Our findings have important ­implications as clinicians should have greater consideration for exploring sleep duration and quality in consultations."

Do these two extracts match up with what the headlines said?

The actual study is available here: it is a meta-analysis of other studies, which means the researchers have looked at all of the earlier studies they could find and collected all of the relevant results together to give a huge sample size with (one hopes) more reliable results.  They summarise their conclusions (bottom of page 3):
 
Divergence from the recommended 7 to 8 hours of sleep is associated with a higher risk of mortality and cardiovascular events. Longer duration of sleep may be more associated with adverse outcomes compared with shorter sleep durations.

("Higher risk of mortality" means "more likely to die", and "cardiovascular events" means "heart attacks and other nasty things to do with the heart or blood vessels".)
  • What does the word "associated" mean here?
  • How does this conclusion compare to the newspaper headline?
  • How does this conclusion compare to the rest of the newspaper story? 

Clinical implications


The researchers go on to give a clinical implication of their study (page 4):

Our study suggests that abnormal sleep is a marker of elevated cardiovascular risk – and greater attention in consultations with patients to ask, and give advice about optimising the duration and quality of sleep may help reduce the burden of cardiovascular disease.
  • What does this suggest the authors of the scientific paper believe?  Does this fit better with the newspaper headline, with the report conclusion, or neither?
Interpreting data is hard!  It is important to be aware of the limits of what data can tell us, and to be careful when drawing conclusions from it.
  • If the data only shows that there is an association between a longer duration of sleep and cardiovascular disease, but not that a longer duration of sleep causes cardiovascular disease, is advising people who sleep more than 7 to 8 hours to sleep less a good thing, a bad thing or neither?
The online version of the article can be found on the Daily Mirror website.


This resource is part of the collection Statistics - Maths of Real Life