Wednesday, 3 December 2014

Misreporting in the scientific literature.

When I was reading the American Heart Association Scientific Statement (AHA Statement) published in 2009, in which Dr Lustig also participated and which I mentioned in my previous article, I have come across interesting reporting of findings of two studies. 

In the section titled Dietary Sugars and Obesity on page 1014 I found this passage: 


The highlighted references were these: 


Something did not quite seem right to me so I checked the two studies myself. Unfortunately I no longer have a free access to the full content of the studies which were not made available in full, so I rely on their abstracts now. Nonetheless, the abstracts seemed to be a sufficient source of information for this purpose, therefore hear what I found so interesting that made me write this article. 

According to the AHA Statement, the reference 69, Rolls et al. (2006) suggests that when people had offered a larger sugar sweetened beverage (SSB), they also ate more solid food as a consequence of that and therefore the SSBs make people eating more so they gain weight. However, the aim of the study was: 
"We tested the effect on energy intake of increasing the portion size of all foods and beverages served over 2 consecutive days."
Therefore they did not examine how increased portion size of soft drink increased food intake at all. 
Their findings: 
"There was a significant effect of portion size on energy intake in both men and women (P<0.0001). Increasing portions by 50% increased daily energy intake by 16% (women: 335 kcal/day; men: 504 kcal/day), and increasing portions by 100% increased intake by 26% (women: 530 kcal/day; men: 812 kcal/day). Energy intake did not differ between the 2 days of each week. Daily ratings of fullness were lowest in the 100% portion condition (P=0.0004), but did not differ significantly in the 150% and 200% conditions."
So, what the study reported was that people ate more when they were offered more, whether it was a drink or food and that extra energy in drink consumed did not lead to naturally decreased portion of the food eaten, therefore the body does not recognize the excess of consumed calories. What's more, people did not compensate on the second day for the overeating during the first day. They received palatable food, they were given larger portions and they ate more of it. That is it. Nothing was reported about the stimulating effect of SSB on food intake and the study did not aim to examine this effect either. There was a difference, however. As you can see, the increased portion by 50% or 100% did not lead to increased energy intake by 50% or 100%, it was less. I assume people were limited by the size of their stomachs, but should they continue eating like this, they would manage to eat it all over time as obese people easily shovel in larger quantities of food and drink than slim people. There is also a psychological mechanism in us: we tend to eat more from a bigger portion than from a smaller portion, and the scientists confirmed this by the study.  

Their conclusion was: 
"Increasing the portion size of all foods resulted in a significant increase in energy intake that was sustained over 2 days. These data support suggestions that large portions are associated with excess energy intake that could contribute to increased body weight."
See? Nothing about the effect of drink on the consumed size of meal. What were they talking about in the Scientific Statement of the American Heart Association? 

Regarding the reference 70 corresponding to Flood et al. 2006, the Statement gives the same impression: the increased size of SSB led to an increase in consumed portion of food. 

What was the aim of the study? 
"This study examined the impact of increasing beverage portion size on beverage and food intake."
Well, this could suggest what was claimed in the AHA Statement. However, it could as well mean that they also studied whether the participants consume more of the drink when they are offered more, and whether this increased portion makes them to compensate with lower food intake, as I have mentioned in my comment to the first study above.
What were the actual findings of this second study? 
"Increasing beverage portion size significantly increased the weight of beverage consumed, regardless of the type of beverage served (P<0.05). As a consequence, for the caloric beverage, energy intake from the beverage increased by 10% for women and 26% for men when there was a 50% increase in the portion served (P<0.01). Food intake did not differ between conditions, so when the energy from the caloric beverage was added to the energy from food, total energy intake at lunch was increased significantly (P<0.001) compared with noncaloric beverages."
I am sure you understand what was written in there, but let me just clarify: 

  • The study reports that when they increased a portion of the drink offered to the participants, they consumed more of it. 
  • Diet cola and water did not make difference in the consumed volume of drink when compared with sugary drink. It is still a fluid and the sweetness, whether from the sugar or artificial sweetener did not make difference from pure water. 
  • Obviously, when you offer a caloric drink at the lunch and this drink is consumed, the energy value of the ingested content (food and drink together), will increase. Would you expect something different when compared to the drink of the same volume but with little or no calories?
  • Food intake did not differ - yet the Statement from 2009 suggests that this study reported increased food intake after consuming SSB. I have no words for commenting this. 
But let's check the conclusion: 
"Serving a larger portion of beverage resulted in increased beverage consumption, and increased energy intake from the beverage when a caloric beverage was served. Serving a caloric beverage resulted in an overall increase in total energy consumed at lunch. Therefore, replacing caloric beverages with low-calorie or noncaloric beverages can be an effective strategy for decreasing energy intake."
Now compare the underlined parts of the test with the text in the AHA 'Scientific' Statement: 
"When the size of a regular cola was increased ... energy intake from food increased..." 
Have you noticed the misreporting of the information? The study did not suggest the increased consumption of food at all! I am speechless. Who wrote that stupid paragraph? 

All what the study reported was that people again did not compensate for the sugar in the drink, or even to the volume of the drink with reduced food intake to balance the calories. The fact is that in both studies there was nothing reported about the increased food intake as a direct consequence of consuming SSB. 

Just check the AHA Statement paragraph again, so that you do not have to scroll up and down: 


And that is not all, I have found some more of such 'scientific' writing. And that was only from the materials I have checked or with which I was familiar from my own research. I did not inspect every single reference they used. 

In one of my following articles will make you see the cherry picking reporting in this particular statement. 

No comments:

Post a Comment