Commentary on Piumatti et al. 2018

Commentary on Piumatti et al. 2018 Journal of Public Health | Vol. 40, No. 2, p. 312 Correspondence Commentary on Piumatti et al. 2018 a car as benefit, and reverse movement as harm, this is a claim that putting on the brakes is producing harm. This is a well-designed study with interesting results. Even if one did accept the bizarre argument that ‘harm’ Unfortunately, these results provide no justification for the starts when the gradient starts increasing, there are two rea- headline conclusions drawn by the authors, which are in dir- sons why there is no foundation for any claim that the data ect contradiction to their own analysis. identifies a threshold of 10 g/day. First, the authors claim in the Results section that � Daily alcohol consumption is plotted on a log-scale ‘Cognitive performance declined as alcohol consumption (although the fact that the model is fitted on a log-scale increased beyond 10 g/day (Fig. 1)’. This is not true. provides no reason for this). If the graph were plotted on Looking at Fig. 1 clearly reveals that reaction times continue a standard linear scale, the curve would become convex, to improve beyond 10 g/day until reaching a minimum at with no straight lines, and so there would be no point of around 16 g/day, the current UK guidelines. inflexion at all, and the positive change in gradient would Second, in the discussion they claim that ‘Cognitive per- start from the lowest possible alcohol level. Therefore, formance improved as alcohol consumption increased up to the whole idea of a point of inflexion is a product of their 10 g/day and then deteriorated as alcohol consumption arbitrary choice of a logarithmic scale. increased beyond 10 g/day’. Again, this is in direct contra- � The claimed point-of-inflexion of 10 g/day was chosen as diction to their Fig. 1. such by the authors, as it is a pre-selected ‘knot’ in the They then go on to say ‘Our findings suggest that to pre- restricted cubic spline on log-consumption, where a serve cognitive performance 10 g/day is a more appropriate straight line is guaranteed to turn into a curve. upper limit.’ And yet their Fig. 1 clearly shows reaction times at this threshold are slower than at the current government It is, to put it very mildly indeed, deeply regrettable that such guidelines of 16 g/day. clearly unjustified conclusions have become part of the scientific These statements are repeated in the Abstract: ‘Cognitive literature, and have already been quoted uncritically by journalists. function declined as alcohol use increased beyond 10 g/ day….Consuming more than one UK standard unit of alco- David Spiegelhalter hol per day is detrimental to cognitive performance.’, none Centre for Mathematical Sciences, University of of which can be justified from their own analysis. Cambridge, Cambridge, UK I understand that the authors might argue that ‘harm’ starts at 10 g/day, since at this point the straight line of Fig. Address correspondence to David Spiegelhalter, 1 turns into a curve—technically, the gradient with respect E-mail: d.spiegelhalter@statslab.cam.ac.uk to the logarithm of daily consumption starts increasing. This doi: 10.1093/pubmed/fdy038 is an extraordinary idea. If we identify forward movement of Advance Access Publication March 1, 2018 312 © The Author(s) 2018. Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com Downloaded from https://academic.oup.com/jpubhealth/article-abstract/40/2/312/4915977 by Ed 'DeepDyve' Gillespie user on 24 July 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Public Health Oxford University Press

Commentary on Piumatti et al. 2018

Journal of Public Health , Volume Advance Article (2) – Mar 1, 2018
Free
1 page

Loading next page...
1 Page
 
/lp/ou_press/commentary-on-piumatti-et-al-2018-mk6ekv5OXn
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1741-3842
eISSN
1741-3850
D.O.I.
10.1093/pubmed/fdy038
Publisher site
See Article on Publisher Site

Abstract

Journal of Public Health | Vol. 40, No. 2, p. 312 Correspondence Commentary on Piumatti et al. 2018 a car as benefit, and reverse movement as harm, this is a claim that putting on the brakes is producing harm. This is a well-designed study with interesting results. Even if one did accept the bizarre argument that ‘harm’ Unfortunately, these results provide no justification for the starts when the gradient starts increasing, there are two rea- headline conclusions drawn by the authors, which are in dir- sons why there is no foundation for any claim that the data ect contradiction to their own analysis. identifies a threshold of 10 g/day. First, the authors claim in the Results section that � Daily alcohol consumption is plotted on a log-scale ‘Cognitive performance declined as alcohol consumption (although the fact that the model is fitted on a log-scale increased beyond 10 g/day (Fig. 1)’. This is not true. provides no reason for this). If the graph were plotted on Looking at Fig. 1 clearly reveals that reaction times continue a standard linear scale, the curve would become convex, to improve beyond 10 g/day until reaching a minimum at with no straight lines, and so there would be no point of around 16 g/day, the current UK guidelines. inflexion at all, and the positive change in gradient would Second, in the discussion they claim that ‘Cognitive per- start from the lowest possible alcohol level. Therefore, formance improved as alcohol consumption increased up to the whole idea of a point of inflexion is a product of their 10 g/day and then deteriorated as alcohol consumption arbitrary choice of a logarithmic scale. increased beyond 10 g/day’. Again, this is in direct contra- � The claimed point-of-inflexion of 10 g/day was chosen as diction to their Fig. 1. such by the authors, as it is a pre-selected ‘knot’ in the They then go on to say ‘Our findings suggest that to pre- restricted cubic spline on log-consumption, where a serve cognitive performance 10 g/day is a more appropriate straight line is guaranteed to turn into a curve. upper limit.’ And yet their Fig. 1 clearly shows reaction times at this threshold are slower than at the current government It is, to put it very mildly indeed, deeply regrettable that such guidelines of 16 g/day. clearly unjustified conclusions have become part of the scientific These statements are repeated in the Abstract: ‘Cognitive literature, and have already been quoted uncritically by journalists. function declined as alcohol use increased beyond 10 g/ day….Consuming more than one UK standard unit of alco- David Spiegelhalter hol per day is detrimental to cognitive performance.’, none Centre for Mathematical Sciences, University of of which can be justified from their own analysis. Cambridge, Cambridge, UK I understand that the authors might argue that ‘harm’ starts at 10 g/day, since at this point the straight line of Fig. Address correspondence to David Spiegelhalter, 1 turns into a curve—technically, the gradient with respect E-mail: d.spiegelhalter@statslab.cam.ac.uk to the logarithm of daily consumption starts increasing. This doi: 10.1093/pubmed/fdy038 is an extraordinary idea. If we identify forward movement of Advance Access Publication March 1, 2018 312 © The Author(s) 2018. Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com Downloaded from https://academic.oup.com/jpubhealth/article-abstract/40/2/312/4915977 by Ed 'DeepDyve' Gillespie user on 24 July 2018

Journal

Journal of Public HealthOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off