Olive Oil Can Tap Dance! Read online

  Olive oil can tap dance

  by Zoë Harcombe

  Published by Columbus Publishing Ltd 2013

  Epub edition

  www.columbuspublishing.co.uk

  Edition 201312-01

  Copyright © 2013 Zoë Harcombe

  Zoë Harcombe has asserted her right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988.

  All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying or otherwise without the prior permission of the author. This ebook is licensed for your personal enjoyment only. This ebook may not be re-sold or given away to other people. If you would like to share this book with another person, please purchase an additional copy for each person you share it with. If you are reading this book and did not purchase it, or it was not purchased for your use only, then you should return it and purchase your own copy. Thank you for respecting the hard work of this author.

  Typesetting by Raffaele Bolelli Gallevi

  Brand and product names are trademarks or registered trademarks of their respective owners.

  All links to external sources were correct at the time of writing each post. Some may have changed since

  The information provided in this book should not be construed as personal medical advice or instruction. No action should be taken based solely on the contents of this book.

  Readers should consult appropriate health professionals on any matter relating to their health and well being.

  The information and opinions provided here are believed to be accurate and sound and are based on the best judgments of the author, but readers who fail to consult appropriate health authorities assume the risk of any injuries. Neither the author nor the publisher can be held responsible or liable for any loss or claim arising from the use, or misuse, of the content of this book.

  Table of Contents

  Introduction

  How medical studies are manipulated

  Kellogg’s Coco Pops Advert

  Kellogg’s Special K Mini Breaks

  Olive Oil can Tap Dance!

  The calorie myth & eating 36,000 calories a day

  Weight Watchers works – according to a study funded by Weight Watchers

  Statins with your burger? Better add a pregnancy test too

  “Surgery is better than dieting, says conflicted doctor”

  One in ten adults dangerously obese…

  Weight Watchers ProPoints plan – what’s it all about?

  Cholesterol & heart disease – there is a relationship, but it’s not what you think

  The White Paper on Public Health (Andrew Lansley)

  UK Women are the “World’s Worst Dieters”

  Red meat & cancer & very bad journalism

  Five-a-day – is it enough?!

  Bariatric Surgery: What’s it all about?

  MyPlate – the new American USDA food pyramid

  Who’s teaching our children about ‘nutrition’?

  The Vegetarian Myth – Lierre Keith

  Red meat & diabetes?

  Nutrition – where will a student be taught the truth?

  Introduction

  Probably my best blogging experience was waking up on the morning of March 14th 2012 to find that one of my heroes had commented on one of my blog posts overnight. The comment was "Hi Zoe My simple comment: nice job. Thanks gt." gt stood for Gary Taubes! I had posted a blog the day before entitled "Red meat & mortality & the usual bad science" and I'd had one of those 'viral' moments when it seemed I'd been one of the first to cover a breaking story and my analysis was widely shared.

  That's what makes blogging so rewarding. We all have things that we want to say and we all want to counter bad science when we see it. The internet generally and blogging particularly have given the little person a voice (and I am little!) We don't just have to take things 'put out there' any more - we can answer back and then other people can share what we do, so that we can punch above our weight.

  I started blogging in 2007, but it tended to be brief comments about something in the media, which needed addressing. I started blogging in earnest - or should that be in anger?! - in 2010 and I found myself putting serious time and effort into tracking down conflicts of interest, locating medical journal articles to dissect the original data, finding links that would help other people researching each topic and so on.

  Over the next three years, I came to see the same tricks over and over again - manipulating 'research' to make a desired point. As Einstein famously said - "If you know what you're looking for it isn't research." The tricks played in presenting data in a misleading way are summarised in the opening chapter - this is not on open view anywhere on my sites. I hope that you enjoy it and I hope that it helps you to dissect studies in the way that I enjoy doing.

  This book is a collection of my blog posts on www.zoeharcombe.com from February 2010 to September 2011. There are 20 in total - comprising approximately 40,000 words. They are on open view on my blog, but we have pulled together this book so that you can read them at your convenience off-line. This is the first of a series of books collating my blog posts in this way - look out for the others.

  This first collection of blogs includes a classic post on the true relationship between cholesterol and heart disease. Conditions from diabetes to cancer are covered. We look at bariatric surgery and conflicts of interest. Kellogg's get a bashing and Weight Watchers doesn't fare much better. But then nor does the English health minister. Is the USA dietary advice any better? Does the calorie formula hold? What's the deal with five-a-day? And who's indoctrinating our children - and nutritional students? If you want to win an argument with a vegetarian - it's all in this book.

  Thank you for reading this and for not simply accepting the nonsense that is spouted by our governments and their advisors. I hope one day that they will be sued for what they have done to their fellow humans. Until then - bloggers will keep blogging.

  How medical studies are manipulated

  This article was inspired by an article that appeared in the media on Monday 31st October 2011. (Link) The original journal article was actually accepted in July and published in August, so I’m not sure why it was suddenly ‘hot news’ on Halloween (Link). The Daily Mail article opened with: “A daily dose of drugs designed to lower cholesterol could also slash the risk of breast cancer recurring, say researchers. They found that women who had developed a breast tumour were nearly 30 per cent less likely to suffer a relapse if they took a type of statin called simvastatin.”

  Normally, I can see through ‘research’ in a few seconds and at least know the game that has been played straight away. I must admit that this one baffled me. Every common sense bone in one’s body says – the opposite would be the case. Healthy cells need cholesterol, cancerous cells need even more cholesterol – how can there even be an association between taking statins and cancer reoccurrence? I circulated the original study and the Daily Mail report amongst The International Network of Cholesterol Sceptics (thincs.org) and got the response within an hour or so. It made me think that it would be beneficial to document all the tricks being played on us – not just on Halloween, but every day of the year.

  Before we look at the tricks, let’s just remind ourselves of the two main types of study:

  Types of studies

  There are two main types of studies – a) epidemiological and b) clinical trials.

  a) An epidemiological study is one that looks at populations. This kind of study gathers data on different groups of people and then tries to look for patterns and differences
between those populations. The populations can be small (a few people) or entire countries. Epidemiological studies can only suggest association, but they may throw up interesting observations that will help to focus further research.

  The Seven Countries Study was an epidemiological study. Such studies don’t have to be done on a whole country basis and indeed are often more helpful when done on a section of a population basis. Hence, the Seven Countries Study included people from the countries Finland (2), Greece (2), Holland (1), Italy (3), Japan (2), USA (1) and Yugoslavia (5). There were 16 regions studied within these countries (the number of regions studied in each country is in brackets). We call a population being studied in an epidemiological study a “cohort”. So there were 16 cohorts in the Seven Countries Study.

  Epidemiology is a useful and interesting science, but not an end in itself. It should suggest connections to be investigated further, but not make assertions about causation. You can see that it has the potential to be abused and misinterpreted all the time. As examples, we can observe that approximately 64% of Greek men smoke and 36.7% of UK men smoke. (Link) We could observe in parallel that the national debt of Greece is higher than the national debt of the UK. We would be unlikely to try to connect these two pieces of information (unless we tried to argue that Greece was less productive as a nation because of cigarette breaks!) However, if we observed in comparison data that Greek death rates were substantially higher than those in the UK, we could suspect that smoking has something to do with dying and investigate this further.

  b) Clinical trials are studies where a specific intervention is made to test a hypothesis. Usually there is one group of people called the control group and one group of people called the intervention group. The control group either has nothing done to it, or is given a placebo (dummy medication in effect), so that they don’t know that they are NOT being treated. The intervention group is ‘treated’. In rare studies, there are more than two groups. As an example – one group may have no change in diet, group two may have fish oil added to their diet and group three may have olive oil added to their diet. This would be done to see if either fish oil or olive oil is better than no oil. The more of the following components that a clinical trial has, the better it is considered to be:

  – Large numbers of people. Broadly speaking, the larger the study the better. If we compare two people and discover something, the chance of this being coincidental is quite high. If we observe a pattern in 100,000 people, this is more significant (statistically, as well as logically).

  – Randomly assigned people. Clinical trials should NOT be able to select which people go in which group otherwise bias could be introduced.

  – ‘Blind’ people. This means that people don’t know whether they are in the control group or the intervention group. It requires that (if the trial is about a new drug), both groups take a drug that looks the same, but one group is taking the actual drug and the other is taking a dummy pill. This overcomes the well known placebo effect – that people feel better just because they are being treated. Both groups should have similar placebo effects and then any real effects should be comparable. (You can see that it would be virtually impossible to have a ‘blind’ diet trial).

  – Double blind. This means that both the participants and the researchers don’t know who is taking which drug. A clinical trial sets out to test a hypothesis. Human nature says that we will look for what we are expecting. Hence if we think that a drug will lower blood pressure and we get what looks to be a surprise reading, we may try to dismiss it, exclude it or explain it away if the reading seems contrary to what we are expecting. If we don’t know if the reading for person 123 is someone on the drug or the placebo, we are less able to bias the results.

  – Only one thing should change. This is the most difficult one to achieve. The perfect trial would have group A given a drug and group B given a dummy and absolutely nothing else would change whatsoever – no change in diet, location, job, stress, happiness, marital status, family circumstance etc. Clearly this is impossible to achieve. The size of the study is assumed to help compensate for this. In a large study, with people randomly assigned, it is as likely that there were similar lifestyle changes in both groups, such that any impact of the drug should still stand out above this. However, many, if not most, of the studies that you will see are done on very small numbers.

  As an example, the June 2011 headlines, that low calorie diets could ‘cure’ diabetes came from a study of just 11 people (Link) This study illustrates another way in which a ‘clinical trial’ experiment can be done. An intervention can be done with one group of people and the rest of the population (in this article – the two and a half million other diabetics) can be considered the control group.

  That’s the basis structure of studies being undertaken in universities, hospitals, research centres and Research & Development (R&D) labs world-wide, right now. Here are the ways in which the studies can be manipulated:

  1) The funding of the study.

  This is the starting point and one of the most significant opportunities for bias. Studies are expensive. Drug studies are phenomenally expensive. Surgical studies even more expensive again. Someone has to fund these. The most likely funder is the pharmaceutical company. As a general rule, only those who stand to gain are going to spend a lot of money on research. Drug companies are in fact obliged to do clinical trials, before a drug can be approved and therefore have to fund these. Drug companies are also able to analyse the results and be involved at every stage of the process. As we will see in the following points, there are many ways in which data can be manipulated, exaggerated and/or excluded, which may be unethical, but sadly not illegal.

  Epidemiological studies are also extremely expensive and can be funded by governments, which may have a genuine interest in important discoveries. For example, if governments could prove that eating five portions of fruits and vegetables a day reduces the risk of cancer, they could consider the cost of promoting such a message (or even giving vouchers for produce) vs. the cost of treating cancer to their public health service. However, epidemiological studies are also run by humans and humans are prejudiced (this literally means to pre-judge) and they may be inclined to try to find what they think the study will show.

  When Ancel Keys received government funding for the Seven Countries Study he was already convinced that fat and/or cholesterol were the cause of heart disease. When this is the preconceived view, it influences what one does and what one looks for, and sees, in data.

  As a matter of interest, the EPIC study (the European Investigation into Cancer) has cohorts in Denmark, France, Germany, Greece, Italy, The Netherlands, Norway, Spain, Sweden and the United Kingdom. The UK has cohorts based in Cambridge (Norwich) and Oxford. The Norwich cohort has 30,000 participants and this is funded by an MRC (Medical Research Council) Programme Grant and by additional grant funding from Cancer Research UK, the Stroke Association and the Food Standards Agency. So, it’s an investigation into cancer, partly funded by Cancer Research UK – Cancer Research UK are not going to want to feel that their money has been wasted by the results being “consumption of five fruit and veg a day makes no difference.” And yet, this was what the European study concluded in April 2010 and the UK study concluded in November 2010.

  In November 2010, the UK part of the EPIC study published their findings in the British Journal of Cancer (Link). Professor Tim Key concluded that: “The possibility that fruit and vegetables may help to reduce the risk of cancer has been studied for over 30 years, but no protective effects have been firmly established.”

  If you can get to the source of the funding of any study, you will invariably find conflict of interest. One problem is that the funders are making it increasingly difficult to find that conflict. The article will either conveniently leave out the funding source or declarations of interest will be vague and difficult to find or third party bodies will be used to give implied legitimacy to the study. For example, the Weight W
atchers study in July 2010 (Link) was ostensibly done by the government Medical Research Council. The Sunday Times needed to do very little digging before they discovered that Weight Watchers paid approximately £900,000 to the MRC to do a very specific study: The control group should go to their GP and get weight loss advice and nothing more. The intervention group should attend Weight Watchers for a year and then the study can achieve the desired headline “Weight Watchers works better than going to the GP” (Link).

  If the study itself is not actually funded with conflict, look out for the conflicts of interest of the ‘researchers’. Example 1, on 18 May 2010, Peta Bee reported in the UK Daily Mail “Giving up bread can make you fat: Gluten is good for you.” The expert in this report was Dr. Emma Williams of the British Nutrition Foundation. Excellent job done there, keeping the many food industry sponsors of the BNF happy. (Link).

  Example 2, on 18 January 2010, the headline news was “Ban butter to save thousands of lives says heart surgeon.” Shyam Kolvekar called for a ban on butter and said people should switch to ‘a healthy spread’. Mr Kolvekar's comments were issued by KTB, a public relations company that works for Unilever, the makers of Flora margarine. Felicity Lawrence, writing in The Guardian on 23 January 2010 noted that the surgeon’s hospital, University College London Hospital (UCLH), had received a fee from KTB for filming Kolvekar performing heart surgery, as part of Unilever's campaign to highlight the dangers of eating too much saturated fat (Link).

  2) The subjects of the study.

  Roland Rat, or Mickey Mouse, rather than Fred Bloggs, or Jane Doe, is the subject of many studies. We clearly can’t remove ovaries randomly in women, so we remove them in rats and see what happens. The study usually reports any findings reasonably accurately and notes findings in rats without suggesting that these findings will be replicated in humans. However, to make the headlines, the summary has to be human. One never sees the headline “rats exercise better after eating chocolate”, but the headlines in August 2011 were “How chocolate can help your workout.” (Link) Either the press release of the study, or the media story, will make the quantum leap from rats to humans. What happened to the rats may happen to humans, but we won’t know unless we study humans.