Of brains, biceps and baloney

Guest Post by Geoff Russell. Geoff is a mathematician and computer programmer and is a member of Animal Liberation SA. His recently published book is CSIRO Perfidy.

NASA climate scientist James Hansen’s recent book Storms of my Grandchildren makes accessible the evidence behind the judgement of many climate scientists that we need to get atmospheric carbon dioxide back to 350 ppm (or perhaps 300-325 to be really safe) to avoid dangerous climate tipping points. But he also makes it clear that merely redesigning the global energy infrastructure isn’t enough, other important climate forcings like methane, nitrous oxide and black carbon must also be reduced.

What do we need to do?

Here’s Hansen’s todo list. Stick it on the fridge.

1. Phase out all coal fired power stations by 2030. Of course, you can still use coal if you sequester all the emissions, … good luck with that.
2. Undo 200 years of deforestation. We need to start this now, but it will take over 100 years and contribute a reduction of about 50ppm by 2150.
3. Reduce non-carbon dioxide forcings. Hansen is a little vague here, but the argument implies that pre-industrial levels are required.

Now, if the next sentence doesn’t hit like a shattering ice-shelf, then reread until it does. All three items are mandatory. This isn’t a smorgasbord where you pick what you want and ignore the rest. With countries around the world still building new coal power plants, the first todo is looking shaky. Fortunately the second and third are technically easier. We don’t need any new science or technologies but the politics are diabolical.

You can’t tackle reforestation without a global food system rethink. People who’ve read my previous posts on BNC understand this, but be patient while I race through a little background for new readers.

As with reforestation, steep reductions of methane, black carbon and nitrous oxide forcings also require a rethink of the global food system. This is because 96 megatonnes of the 350 mega tonnes of anthropogenic methane emitted annually are due to livestock. It’s also livestock production which is responsible for the bulk of the annual global conflagrations responsible for preventing plenty of natural reforestation while also contributing rather a lot of black carbon. This is covered in Boverty I. The good news is that 38 megatonnes of methane emissions will go when we stop mining coal and another 73 megatonnes are tied up with oil and gas production and can be relatively easily dealt with when there is a will to do so.

The livestock reforestation impediment

Currently, a major sticking point on reforestation is the attitude to animal product consumption of the UN FAO which is summed up in the just released report on the greenhouse gases associated with the dairy sector: Without concerted action, emissions [from livestock] are unlikely to fall. On the contrary, they are rising, as global demand for meat, milk and eggs continues to grow rapidly. Projected population growth and rising incomes are expected to drive total consumption higher–with meat and milk consumption doubling by 2050 compared to 2000 (FAO, 2006b).

The cognitive dissonance at the UN FAO in understanding that livestock is currently the largest driver of deforestation, but also planning for a doubling of meat and milk consumption by 2050 while trying to reform a frontier cowboy culture is extreme. Any growth in the global livestock industry will make ending deforestation difficult and the required massive global reforestation impossible.

A huge part of the problem is that decades of meat and dairy industry propaganda has left people with the a cult-like certainty that they are some kind of wonder food. This sentiment is echoed in a recent special edition of Science onFood Security in a paper by H. Charles Godfray et al:

… in developing countries, meat represents the most concentrated source of some vitamins and minerals, which is important for some individuals such as young children.

Note that Godfray felt no need to justify his claim. Henry Thoreau wrote about a similar prejudice in 1852:

One farmer says to me, You cannot live on vegetable food solely, for it furnishes nothing to make bones with; and so he religiously devotes a part of his day to supplying his system with the raw material of bones; walking all the while he talks behind his oxen, which, with vegetable-made bones, jerk him and his lumbering plow along in spite of every obstacle.

How can you argue with the likes of Godfray when Science, one of the world’s top peer reviewed science journals allows him to get away with unsubstantiated assertions? There is not even any science to debate if you don’t justify your claims. Thankfully, the 2006 UN Livestock’s Long Shadow (LLS) report provides a hint of science in its justification for pushing livestock products in the developing world:

Children in particular have been shown to benefit greatly in terms of physical and mental health when modest amounts of milk, meat, or eggs are added to their diets, as shown by long-term research carried out in Kenya (Neumann 2003)

The above two quotes go to the heart of the international stranglehold of the livestock industry on the only organisations with enough political clout to have a chance of driving a major global reforestation effort. The players like the UN, the EU and major national Governments.

This post examines the second of these quotes in detail.

But what about all the starving children?

The LLS quote is some serious blackmail. It implies that nobody who cares about starving children could possibly suggest any reductions in global meat production, particularly in the developing world. There have been plenty of calls for a decrease in global meat production via a transfer of meat to the developing world. The most important of these was back in 2007 from an Australian team writing in one of the world’s top medical journals, The Lancet. They proposed the world’s average 100 gram per day meat intake be reduced to 90 grams per day with a hefty redistribution to even out global consumption. High income countries would drop from 200-250 grams per day to 90 grams per day and low income countries would increase their meat from 25-50 grams per day to 90 grams per day. Such a move might halt deforestation, depending on if and when the global population levels off, but it will clearly not be enough to allow the necessary massive reforestation.

The Lancet paper also contains a nice little qualitative table claiming that the proposed increase in meat in the developing world would heavily decrease childhood stunting. Such a claim is in line with the LLS quote, but no reference was given. Again as with Godfray, it seems nobody thought any evidence was needed.

LLS had 6 authors with the lead being the coordinator of the UN FAO’s Livestock Environment and Development Initiative (LEAD) … Henning Steinfeld. Is the quote just a demonstration that LEAD is a livestock industry pawn, or is it simply good science? Let’s look at the Neumann studies.

Show me the data!

The Neumann paper cited by LLS is part of a set that appeared in a 2003 supplement in the Journal of Nutrition.

The papers describe a study involving 554 children in Kenya provided supplemental food on a daily basis for 12 months. This is solid, careful, painstaking clinical research involving a supervised team of over 100 locals taking blood samples, preparing food, measuring biceps, administering IQ tests, freezing and transferring blood samples to the US for processing and dealing with a maze of logistical difficulties. All the meat was shipped into the rural area from Nairobi. Many of the children involved came from families with cattle, but they rarely eat or milk them. A Control group got no extra food at all. Why did they agree to take part? They were given a milking goat at the end of the research. Great PR for the dairy biz.

Randomisation

The randomisation to different extra-food groups was done by school. All the children of the selected age group in one school got more meat, those in another school got more milk and another got just plain food. So 12 schools were allocated to one of 4 groups … 3 schools per group. It’s easy to understand why this randomisation procedure was used, but equally easy to see how something unusual in even a single school might cause problems.

Hang in there while I describe accurately the extra food the children got. This kind of detail is unusual in a blog, but BNC is different and the details matter.

Who got fed what

The children had a median age of 7.4 years old and food intake before the study was highly variable with a quarter of the children being stunted at the start of the study. In addition to a Control group getting no supplementary food, there were 3 types of daily supplement, denoted Meat, Milk and Energy by the study authors. I’ll call them Meat, Milk, and Plants. Calling the Plant food group Energy makes it sound like that’s the only thing Plants can provide … a revealingly silly choice of terminology.

The three supplementary food groups were built around a local stew made of maize, beans and greens and all balanced to contain about 240 Calories. So the children either got stew with meat (60 grams of beef mince), stew with milk (200 ml), stew only, or nothing.

Comparing the LLS quote with the actual research

Now, how does LLS’s description of the results compare with what actually happened? Note that all the key studies are publicly available thanks to the enlightened policies of the Journal of Nutrition. You can read them yourself.

First, there was no egg in any of the supplementary feeding. Oops, strike one for LLS. Second, is 60 grams per day for a 7 year old modest? It’s almost double the Australian National Health and Medical Research Council recommended meat intake for a 7 year old. It’s double the per capita daily production of beef in Kenya. It’s close to double the average red meat intake of Australian 7 year olds (subscription required for this link). After a few months in the study, the meat supplement was increased to 85 grams/day and the milk increased to 250 ml. To describe this intake as modest seems a poor choice of adjective.

The daily food supplement was called a snack by the researchers, and had about the same caloric value as a standard McDonald’s hamburger which has 90 grams of beef mince, similar to the 85 grams in the Meat snack. This is also close to the 90 grams of meat recommended by the Lancet authors (although they recommended more of that meat be chicken or pig meat). The beans and greens however would have made the Meat snack rather more nutritious than a hamburger.

Did the children benefit greatly in physical or mental health as LLS claimed?

The title of the paper describing the physical impacts seems clear: Food Supplements Have a Positive Impact on Weight Gain and the Addition of Animal Source Foods Increases Lean Body Mass of Kenyan School children. But as with everything else about this research, you have to actually read the damn papers, not just the titles and not just the abstracts to find out what actually happened.

All the intervention groups gained an average of 10% (0.4 kilograms) more weight than the Control group but there were essentially no changes in height for age … sorry about the jargon, what does this mean? No change in stunting.

There were no statistically significant differences in height gain, body fat (as measured by skin fold tests) and a few other measures, but the meat group (and not the milk group) got statistically significantly bigger biceps … how much bigger? … after all, this is where the Increases Lean Body Mass of the title comes from. So, are we talking little Kenyan Schwarzeneggers? Not quite. The biceps were bigger in circumference by less than 1 millimeter, the Meat group’s bicep increased by 7.1 mm compared with 6.5 mm in the Plants group. Also, as usual, the paper’s title is misleading because it wasn’t all animal source foods which achieved this mighty sub-millimeter muscular gain, only meat. Milk produced the same gains as Plants. The area of the biceps was also bigger in the meat (but not in the milk group) but whether either of these changes was due to nutrition or the possibility that one or more schools did more physical education wasn’t discussed.

The other aspect of physical health that could be construed as important is micronutrient status. We will deal with that below.

Oh yes, I almost forgot. The researchers were alert to all manner of possible confounding problems. They even measured the food given at home to see if it changed as a result of knowing that the children were getting extra at school. It did. The Control children and the Meat children both got an increase of food at home (about 150 Calories), while the Milk and Plants children received a similar sized decrease in home nutrition. The quantity of extra or reduced home feeding wasn’t uniform over time but the direction of the changes persisted throughout the study.

The researchers went to great lengths to measure physical activity, but didn’t report any details except to claim an improvement. The activity results were published in 2007. I could describe them, but this post will be quite long enough as it is. Suffice to say that, regardless of the supplementation, many of the children would still not have been getting enough food to support high activity. I base this claim on the variance in the reported energy intake of the children and the Australian National Health and Medical Research Council (NHMRC) recommended energy intakes for children, suitably adjusted because Australian children are a little bigger at the same age. The reported extra 150 Calories per day over the 12 months for the meat group didn’t rate a mention in the 2007 paper which duly reported more high activity among the Meat group compared to any others.

That’s strike two for LLS, there were no great physical benefits for the meat children over and above the benefits of extra food.

Is taller better?

It’s worth noting here that while stunting is a definite indicator undernutrition, it doesn’t follow that maximising growth rates in children is good. There is a really good reason not to maximise growth rates. The big “C”. Your height as an adult is a good cancer predictor … greater height equals greater cancer risk. The 2007 World Cancer Research Fund report explains how it works and gives “adult attained height” the rare accolade of having been convincinglyshown to be a cause of bowel and breast cancers with a probable role in other cancers. So if additional animal source foods do maximise growth, then this is evidence against them, not for them.

Great improvements in mental health?

Now for the last of the LLS claims. Recall, LLS also told the world that animal foods, all of them, produce great benefits for mental health. I call this LLS’s meat head claim.

I’ll begin with a quote from the abstract of the relevant study before revealing the actual results. The abstract is beautifully crafted to mislead people who don’t read the entire study while being somewhat defensible in the face of a claim of fraudulent misrepresentation. First, comes the quotable quote, the take home message, the thing that will survive in the annals of meat industry propaganda:

Results suggest that supplementation with animal source food has positive effects on Kenyan children. …

There you have the guts of the LLS claim. But the LLS promoted “positive effects” into “benefitted greatly”. The abstract follows up on this general claim with a semblance of the truth expressed as abstractly and vaguely as possible:

However, these effects are not equivalent across all domains of cognitive functioning, nor did different forms of animal source foods produce the same beneficial effects.

Now let’s see what happened. The researchers measured 3 things: arithmetic, verbal skills and Raven’s Progressive Matrices (RPM). There were virtually no differences on the first two. Even the Control group, living on their normal diet and the promise of a goat made almost identical gains to the children with the burger-equivalent snacks.

But the Raven’s results showed a statistically significant impact. Keep in mind that the bicep increase (of less than 1mm) was also “statistically significant” … which is a rather different concept from “important”. An RPM test consists of a matrix of geometrical patterns with one missing, as in the example from Wikipedia on the left. Usually, the matrix is also accompanied by a set of possible candidate patterns.

The Meat group did statistically significantly better than the Plants and Control groups, while the Milk group statistically significantly worse. The size of the effect was similar. What we are talking about here is a relative change in the slope of the increase in RPM scores as the children aged.

Despite the lower RPM rise rate in the milk group, neither the study authors nor the LLS authors are recommending less milk to prevent a decline in mental health and there was no mention of rethinking the Control group’s free milking goat and perhaps delivering it in sliced and diced form.

By now, you will understand that research cited by LLS doesn’t show what it was supposed to. Not even close. It was funded by the USAID Global Livestock Nutrition Collaborative Research Support Program and was a substantial study carried out by well qualified people with a financial and professional interest in showing that animal foods are a god-send to poor children in developing countries. But apart from the occasional misrepresented and tiny result, they found nothing. This must have disappointed another of their funding sources: The National Cattleman’s Beef Association.

The sloppy, inaccurate and uncritical citation of these non-results by otherwise careful LLS authors just reflects what happens when people have been brainwashed by the tunnel vision of the dominant meat obsessed cuisine challenged culture.

Summing up

Remember that we began with a study cited by LLS which had UCLA Professor Charlotte Neumann as the lead author. Here is its full title.

Animal Source Foods Improve Dietary Quality, Micronutrient Status, Growth and Cognitive Function in Kenyan School Children: Background, Study Design and Baseline Findings

The title makes four claims and we can now summarise their accuracy:

1. Animal source foods increase dietary quality. Vacuously true by Neumann’s definition of quality.
2. Animal source foods increase growth … trivially true, but did it increase growth more than plant source foods? No.
3. Cognitive functions … if the meat RPM increase is considered important, then the milk decrease should be similarly considered important … I’d judge both to be trivial and confounded.
4. Micronutrient status … with the exception of B12, this is false. Again we need to read a subsidiary study. This paper says that none of the supplementary feeding had any impact on any biochemical nutrient measures except B12. Even with B12, the results will surprise some people. The rate of serious B12 deficiency dropped in the Meat and Milk supplemented groups, but the rate of moderate deficiency actually increased in the Meat group.The status paper has various results tables. Let me just cherry pick a few results, not because they prove anything, but just because they will surprise normal meat eaters. Serum zinc levels fell in all the groups, but fell most in the meat group. Oops, not good. Ditto copper. Plasma folate fell more in the Meat and Milk group than in the Plants group. Hemoglobin levels rose more in the Plants group than in the Meat and Milk groups. Serum iron increases in the Plants group were double what they were in the Milk group. The researchers defined anemia as having hemoglobin levels below 115 g/dl. The group which had the biggest fall in anemia rates was … wait for it … the Control group!There’s an old saying that when all you have is a hammer, everything looks like a nail. There are clearly some complex interactions happening between many factors in these children, some of which probably are not on anybody’s radar, let alone researchers who see animal foods as the ultimate hammer.

The issue of B12 is important and came up in the blog responses to Boverty II. The children given Meat or Milk in the Kenyan study didn’t all end up with good B12 blood levels. Judging by the rise in moderate deficiency, some went backwards. How could this happen? The B12 in animal foods is bound to protein and not well absorbed. The B12 in supplements is easier to absorb, doesn’t come with saturated fat, bowel cancer causing heme iron and other carcinogens, and can be supplied to 9 billion people without the deforestation that animal products on the required scale would entail. Older people (over 50) frequently have sub-optimal absorption anyway, which is why the US Institutes of Medicine advises all people over 50 to use B12 supplements … whether or not they eat meat. B12 fortified foods are common in developed countries, they need to become globally ubiquitous … much as iodine is in salt.

So you can stop reading now … if your only concern is the possible deleterious impact that reforestation and a consequent reduction in global livestock could have on global health, particularly to vulnerable children in developing countries. In my previous BNC post, Boverty Blues, I explained the mechanics of the livestock anchor chain depressing reforestation and agricultural productivity in many parts of Africa.

But the rest of the story about this research is fascinating and should be told.

RPM’s have risen in Kenya without animal foods!

Was the size of the RPM improvement as cognitively significant as the mighty increase in bicep circumference? The researchers don’t hazard an opinion. Most of them are also co-authors on another paper involving Kenyan children and RPM scores. This paper shows that there are ways of getting genuinely large increases in RPM scores without adding any animal source foods. The paper reports on a 1984 cohort of Kenyan children of similar age who also underwent RPM testing. The difference between the average RPM scores in the 1984 and 1998 cohorts was 4.5 points. This result held even when the 1984 cohort was carefully filtered to make it closely match the characteristics of the 1998 cohort. As we shall see, this isn’t a one off. RPM scores have been rising globally for decades and the increases are the subject of much research.

What do we know about the possible causes of this particular RPM increase over time? We know with a fair degree of certainty that it wasn’t caused by any increase in animal source foods … because while the 1998 children were better fed than the 1984 group, all of the additional food was plant food. Interestingly, I didn’t find any articles by this group with a title like: Increase in plant foods drives large IQ increase in Kenyan children.

But wait … there’s still more.

In search of the vanishing cohort

The Kenyan research actually involved not one but two different groups of children. My account above only described one. But there is mention of two cohorts in the main Neumann paper. The second cohort had 500 children and was enrolled 12 months after the project start. This cohort was enrolled after a drought and a teacher’s strike caused local food and logistic problems. These extra cohort could be used, they said, either as a replication or to increase the statistical power of the research.

But something happened to Cohort II. The cognitive functions paper just ignores it as does the micronutrient/dietary quality paper and the physical growth response to supplementation paper.

But Cohort II springs to life in a 2007 paper by Neumann et al … but it has shrunk from 500 to 375 without explanation. Cohort II appears in some end of term school test scores where Meat did best, followed by Plants, then Milk with the un-supplemented Control group bringing up the rear. It also appears in a figure describing bicep size changes.

I have emailed Professor Neumann asking what happened to Cohort II, but have so far not had a response.

RPM increases

Last but not least, RPM is a very interesting type of test. We have already noted the Kenyan increase over time without any animal food increments. RPM is a component of most IQ test batteries and children have been getting steep improvements on it (as well as some other IQ tests) for decades, prompting some to speculate that just as improved nutrition is responsible for height increases over the last 80 or so years, it is also responsible for IQ increases. Except that it isn’t.

A recent paper by one of the people (James Flynn) who discovered the effect which now bears his name, The Flynn Effect, demolishes the theory. The paper is called: Requiem for nutrition as the cause of IQ gains: Raven’s gains in Britain 1938-2008. The name nearly says it all, except that it also considers data far beyond Britain. The point that concerns us is that you can train for RPM and improve. This is fine, except that it doesn’t necessarily bringing arithmetic improvements.

The Flynn paper shows that once above some basic threshold level, it isn’t nutrition that drives performance on RPM nor improvements on RPM. It’s easy to dream up simplistic theories about what is driving these increases, but a paper by John Raven (a son of the RPM designer), demolishes more than a few such one-factor theories. Flynn’s own hypothesis about the cause of the increase, presented in his book “What is intelligence?”, is considerably more subtle.

Concluding Remarks

This post began with climate change.

Dealing with climate change requires a global reforestation effort, but that can’t happen without a dietary change and a dietary change won’t happen while people in positions of authority in developed countries sincerely believe that their own meat based diet should be the goal of developing countries. The chain is clear.

The false nutritional beliefs are based on decades of advertising lies and plenty of sloppy reporting of scientific results. This has been a longish post to untangling a tiny part of the tangled web of nutritional misinformation that has to be dispelled as part of our efforts to avoid dangerous climate change.