Popular science: Five clues to common overstatements

Lazy interpretation creeps in when science is popularised.

Here are some clues to help to detect overstatements:

1. "Because" 

any article that uses this word is veering from the truth. We are living through a Crisis of Attribution; that is, we have no secular theory to explain why a follows b, only that they are associated more often that expected by chance. Our theories are not built to answer the question: "Why?" or "How?" only to make predictions and test for associations.
Suggested alternative: avoid altogether

2. "Scientific" 

Scientists are aware of what we don't know which enables them to ask a discriminant question: ie one that can be tested in our present state of knowledge. The adjective 'scientific' conveys nothing more than: 'I want you to believe this.'
Suggested alternative: "doubtful"

3. "Facts"

In the Words of Friedrich Nietsche: "Es gibt keine Tatsachen, nur Interpretationen" 
'There are no facts, only interpretations.' 
Human perception is notoriously susceptible to error, blue sky is grey to some people and my 'blue' is almost certainly completely different to yours.

Suggested alternative: avoid altogether

4. "Proved" and "Validated" when combined with the 'S' word.

Suggested alternative: "suggested"

5. "Brain chemicals" and "Neuroscientists" and "raises the level of"

In science, theories are contested, try finding two 'neuroscientists' from varying fields who agree with each other; the neuro-anatomists barely talk to the neuro-chemists and as for the neuro-electrophysiologists and the brain scanners, they rarely even see each other.

It's a tough life being a populiser of science, though I can't fault Brian Cox, particularly as I don't understand astrophysics.

UK Cancer Survival rates lag behind: info-graphic

Today's Guardian reports that UK cancer survival lags ten years behind other European countries: 

Which cancers?

UK Cancer survival rates have been improving in the last thirty years. :

source

Cancer is a highly variable disease. Cancer of the prostate, testis and melanoma have shown particular improvement, lung and pancreas less so.

source: Daily Mail

These are the latest figures for common cancers in the first year after diagnosis.And after Five years:

Lung and stomach cancers show less change.Health funding differs widely between regions. 

What about regional differences? 

The picture is complicated, regions in the south and Central belt of Scotland appeart to do best. Source: Daily Mail

Details of the study of almost 3m cases of eleven different cancers diagnosed in 23 different countries between 1995-1999 is here: 

Which countries perform best?

Some countries which spend less on health perform have better survival rates than the UK. Mikre Richards, UK 'Cancer Tsar' puts this down to late diagnosis.

Take cervical cancer, for instance: 

This is internatiuonal data from OECD, showing marginal improvement in the UK compared to countries like Norway and New Zealand.

I can't find the reference for this bar chart which suggests that US healthcare is equivalent to more efficient European countries.

Cotswolds: road death capital of Glos

Which Glos roads are the most dangerous?

What's special about Glos Road deaths?

UK Government statistics show an impressive fall in annual road deaths from 3,000 a year to under 2,000.

But Country roads remain dangerous; more dangerous than motorways and A roads. If you live in a rural area you'll know the reason: speed.  According to Brake one in three drivers admits to driving too fast on country roads.

Which Glos District do you think has the highest number of accidents? Cotswold or Gloucester City?

Here's the stats available from Road Safety Gloucestershire 

32 Glos road casualties in 2013 include one cyclist.

Cotswold roads are among the most dangeropus in the whole of the UK.  How does this compare with previous years?

Here's the data from the UK Govt site

Thirty deaths a year have been average in Gloucestershire, essentially unchanged. Of the West of England rural counties shown here, Dorset has achieved the greatest fall in road deaths while deaths in the other counties remain static.

Brake campaigns in favour of 20 mph speed limits in towns but what can be done in rural areas?

Percentage Medicine, scourge of our times.

Random controlled trials [RCT's] are used to detect benefit from giving treatments to prevent stroke.  But randomised trials depend on weeding out atypical cases and expecting everyone in the study to take their treatment as prescribed.  It ignores the narrative of the patient's life and herds all patients with a condition together, assuming that they are identical.  Normally compliance with treatment is checked regularly and people who drop out are accounted for in the analysis.  

A common measure derived from the studies is 'number needed to treat' [NNT] in order to prevent death or recurrence. How many people need to take common treatments to prevent a further stroke?  What would your estimate be? 5, 10, 15?

Opinions and estimates vary but it is not unusual for doctors to prescribe treatments that have a NNT of over 100, that is 99 people may take the treatment and experience side effects for one person to survive who otherwise would have died. You could call this 'percentage medicine' which follows from the bias of RCT's away from lifestyle changes and towards drug treatments. 

Lifestyle changes such as increasing exercise and weight loss are more difficult to standardise and outcomes are less clear.  Multiple prescribing is the result: anti-platelet drugs [eg aspirin], ACE inhibitors [lisinopril] and statins to lower cholesterol levels.  These may all be continued into advanced age when changes in memory, sight, hearing and interaction between all the treatments prescribed are largely unknown.  Compliance with treatment is almost certainly lower in 'free range' people when compared to well motivated research subjects.

 These treatments work but only in a proportion of cases, sometimes a very small proportion of patients can be shown to benefit from this approach, like 1%.

What else could we do? Drugs will never be the answer, though big pharma would like us to think so. The biggest advances in health have come from public health interventions: immunisation, clean water, improving diets.  We could apply what we already know.  Educate, but you can't force people to exercise more and lose weight. We could be preventing stroke a lot better if we had individualised exercise regimens with incentives for reaching major goals.  Sounds like socialised medicine. The uncontrolled marketing of fatty and sugary foods and slimming remedies has kept the economy ticking over but growth in these areas is impacting on the nation's health. Increasingly sedentary jobs are putting more people at risk.

  Cycling needs to be made safer and why can't cars offer more exercise while you're driving?

"Local author" Promote and Market yourself

   With Thanks to Colleen Laybourne-Smith
 Colleen biographyWho is published by Book Guild through their partnership scheme.  

Last year 3,000 people applied to become partners at the Book Guild and Colleen was one of 85 whose picture book for 5-8 year olds was accepted. In partnership, the author pays the cost of the first print run and receives 30% of the total price of each book sold

The initial Waterstones book launch.  Choose the most local branch .

• Phone local papers and magazines in advance

• Invite friends, relatives, contacts, fellow authors and inform the branch of pre-orders, bring your own copies to sell and use commercial book holders, pretty soft toys and a stand to make an eye-catching display.  

• Use a banner with picture of the book: TITLE, GENRE, AUTHOR, BOOK SIGNING TODAY, PRICE, AGE RANGE  Have matching cards and book marks with links to a website.

• Position your table near related books where you can talk to parents looking for a book: personalise your dedication for a christening or birthday.

Send out review copies, update your author page on Amazon, try to interest libraries, Google your ISBN, title and name: you may find people mention the book in remote places.

Subsequent Waterstones launches: "Local author" is a moveable feast: where you were born, where you went to primary school, secondary school, any previous towns you lived in, where your grandparents live or your present home. Let them know how many books were sold at the initial launch.

• Set clear goals: number of books to sell or sales ranking at Amazon or Waterstones sales rankings also number of blog site hits.

• Further events: Involve independent booksellers, mention them in press articles and ask if they will stock your book in return.  

UK Cycle Death Conspiracy

Rising Road Deaths in the UK-February 2017

There had been a welcome improvement in the annual casualties of car drivers and passengers up to 2010, from over 3,000 a year to less than 2,000 a year.  Deaths of cyclists, though much less common, had been relatively stable over this period.

But cycle deaths in adults are static or rising: Here are the latest trends in quarterly figures showing a rise in numbers killed or seriously injured in all road users.  

* Serious injury: An injury for which a person is detained in hsoptial as an in-patient: fracture, concussion, internal injuries, crushings, burns, excluding friction burns, severe cuts, severe general shock requiring medical treatment and injuries causing death 30 or more days after the accident.

Here are the figures:

The latest report cites the CRASH system which may have resulted in police recording SI more frequently but even so more people are dying on the roads:

A 'non-significant rise' despite these cautious bullet points, it's hard to ignore the rising numbers of road deaths. The UK will likely soon return to 2,000 avoidable deaths per year.

Hidden among the figures and tables is a 22% rise in child deaths on the road, highlighted in an RAC news report: 

International comparison are of limited value but the reduction of Dutch mortality among cyclists to 162 in 2010, despite much higher cycle use than in the UK is striking.

The UK Government has prioritised road safety for car drivers and relatively neglected the needs of cyclists.

References

https://www.gov.uk/government/statistical-data-sets/ras45-quarterly-statistics#table-ras45007
https://www.gov.uk/government/publications/reported-road-casualties-great-britain-main-results-2011

http://www.rospa.com/roadsafety/adviceandinformation/cycling/facts-figures.aspx

http://en.wikipedia.org/wiki/Reported_Road_Casualties_Great_Britain

http://www.thetimes.co.uk/tto/public/cyclesafety/article3313260.ece

http://assets.dft.gov.uk/statistics/releases/reported-road-casualties-gb-main-results-2011/reported-road-casualties-in-great-britain-main-results-2011.pdf

http://www.aviewfromthecyclepath.com/2011/04/netherlands-traffic-deaths-down-again.html