I’m actually dealing with this problem at work right now (a small group of vocal and opinionated subject-matter experts are driving decision-making about how to solve a problem, and possibly not in a good direction either).

From Jorn Barger’s Robot Wisdom Weblog I linked to a great article by Edward Tufte about how the findings of primary studies gradually lose their power and meaning, when they become repackaged and redistributed by secondary organizations (journalists, public relations firms, think tanks, nongovernmental organizations, governmental agencies, and on and on – see the posts from the previous couple of days):

In repackagings, a persistent rage to conclude denies the complexities, ambiguities and uncertainties of the primary evidence.  A substantial selection bias also operates:  news wins out over olds, as recency of evidence decides relevance of evidence.

You must see the original to get the subtext behind the phrase “rage to conclude”, which is a wonderful quote from Flaubert.

According to Tufte, secondary organizations filter and inevitably corrupt the evidence-based findings in a primary report, as they summarize, simplify, synthesize and spin the material to suit their particular needs.  While not mentioned, I’m certain that blogs would also fall into this category.  Readers continually need to ask themselves if what they are getting from any source truly constitutes information.  A definition of information that I have found useful is in James Q. Wilson’s book Bureaucracy:

. . . a full, accurate, and properly nuanced body of knowledge about important matters.

Paraphrasing Wilson, often what we get instead is:

. . . a torrent of incomplete facts, opinions, guesses and self-serving statements about distant events.

Tufte says that consumers of evidence should stay reasonably close to primary sources and to secondary reporters and synthesizers who provide unbiased interpretations (note that in practice this could be a bit difficult, because it requires taking the time to compare original source and secondary compilation, and the original source might be technical and difficult to understand – time and ease of understanding being two reasons for turning to secondary sources).  Signs of an untrustworthy source are denial of access to primary evidence or repackaged reports that always support a party line.

I get the sense that Tufte feels not being scrupulously intellectually honest in how one writes, describes or summarizes another’s work is inherently corrupting.  Using examples of corporate communications and government intelligence agencies, he notes that in distorting and spinning evidence to others, we may start to do something worse – lying to ourselves.

There’s no longer any value in berating our news outlets for the missed opening to genuinely engage people on this environmental health issue. I’ve gone beyond that, and someday I’ll need to update the mission statement for the blog to reflect that. At the risk of repeating myself, I personally don’t have a need to pay much attention to arsenic in apple juice; I already have an informed opinion about arsenic risks and our family doesn’t drink commercial apple juice. Because it’s mentally stimulating and because it’s useful for me to summarize the key points, especially since the science on arsenic evolves and I haven’t written a summary since the Coeur d’Alene RI/FS in 2000, I’m writing a bit about the health risks from arsenic ingestion – if you happen to read it and find it engaging and informative, then that’s more enjoyable for me and a win-win for the both of us.

A little bit of historical perspective: arsenic, whose name comes from the Greek word for potent, has been known to be poisonous from antiquity. However, in spite of its use by notorious poisoners, it hasn’t been uniformly evil and even today there are still some pharmaceutical uses for arsenic compounds. But, arsenic is undeniably toxic, historically causing adverse effects in pesticide workers, copper miners and industrial workers, and most notably with regard to drinking apple juice, villagers in Taiwan, West Bengal and Chile who have used as drinking water groundwater drawn from aquifers that are enriched with arsenic.

In 1968, an epidemiological study was published that presented evidence of the adverse effects of greater than 300 micrograms per liter (300 ug/L) of arsenic in drinking water in villages in southwestern Taiwan. Since the 1920s, this area had been known for the occurrence of blackfoot disease, a disorder of the blood vessels in the legs that could result in gangrene in the toes and feet. Further epidemiological studies of this part of the world revealed a dose-response relationship of arsenic concentrations in groundwater and a host of diseases including cancer of the skin, lung, liver, bladder and kidney, a disfiguring skin disorder (hyperpigmentation and hyperkeratosis), and a range of blood vessel disorders including heart disease. A “dose-response relationship” is the gold standard in toxicology providing the strongest evidence that a substance causes a human health risk. Confirmation of these results had been provided from studies of areas in Argentina and Chile where arsenic was naturally found in groundwater used for drinking water.

In a tragic example of the best intentions going wrong, the United Nations Children Fund in the 1970s responded to epidemics of waterborne infectious diseases in Bangladesh and West Bengal by promoting an effort to switch the population from surface water to drinking water. Millions of tube-wells were drilled into what were later determined to be arsenic-rich aquifers. Groundwater in many of these wells was found to be contaminated with arsenic at concentrations of 500 to 1,000 ug/L or higher. More than 500 million people living within a 500,000 square kilometer area of Bangladesh and India were placed at risk of chronic arsenic poisoning, which the World Health Organization called “the worst mass poisoning in history”. Other studies in West Bengal also indicate that increased risks of reproductive effects and infant mortality, and decrements in intellectual development in children could be associated with elevated arsenic concentrations in groundwater.

Here’s the couple of places that were the sources the preceding synopsis:

2005 news item in Environmental Health Perspectives

“Arsenic Exposure and Toxicology – A Historical Perspective”, prepared for the 50th anniversary of the Society of Toxicology

The preceding synopsis is important in putting arsenic in apple juice into perspective. It forms the basis for a lot of our thinking about tolerable levels of arsenic in food and water; these studies help define the levels at which adverse health effects in humans are pretty much known to occur – the real magic comes into place in using the information to define the levels that will not pose significant health risks. And zero is the wrong answer.

I’m surprised there is any surprise that education and numeracy doesn’t necessarily translate into concern about climate change. A few observations come to mind: the science around assessing climate change hazards is dominated by uncertainties – that’s not a bad thing, it just is when it comes to using the findings for decision making. The research published by Tversky and Kahaneman indicates that we’re generally really bad at making decisions with uncertainty. The research published by Gerd Gigerenzer indicates that even smart people aren’t as numerate about risks as they think. As an aside, it’s probably a fallacy to think that all educated people will converge onto one answer, regardless of other cultural factors.

The simple issue with climate change is that the mitigation and adaptation that might be needed is going to require society-transforming changes and the leadership to compel or persuade a lot of people to take actions that aren’t in their short term interest, for the benefit of future generations. And, we don’t even know if any of this is necessary – but, by the time we’re more certain, it may be too late and we as a species (particularly in Western societies) will be screwed. That seems to be the key issue, and more education about climate science is unlikely to help resolve this dilemma.

One line of inquiry Jim Hrynyshyn might consider is how opinions about climate change correlated with income and education – people who are well-off are the ones who will be asked to make the most significant changes to their lives (i.e. give up stuff they already have) to achieve climate mitigation or adaptation.

Originally, I had thought that before it’s too late and peoples’ attentions are drawn elsewhere, there’s an opportunity to convey some risk information about arsenic. Couldn’t get together in a fashion that’s readable to the folks who would actually benefit from it, in the time allowed (at this posting, Dr. Oz and arsenic-contaminated apple juice have disappeared from the news cycle). After awhile, blog fatigue set in – I already have an informed opinion about arsenic risks; my family and I don’t drink the amber-colored sugar water that passes for commercial apple juice (the youth drink the orangy stuff that passes for orange juice, but that’s a different topic); the burning need to inform readers about arsenic risks is subsiding as short attention spans turn to other matters.

Look, here’s the short answer: arsenic is all around us. It occurs naturally in rocks and soil, rivers and streams and groundwater. Certain human activities such as mining and pesticide use can introduce arsenic into the environment; note: the use of arsenical pesticides was discontinued decades ago, however, concentrations of arsenic in soil resulting from pesticide use can still be found around orchards. Ingestion of arsenic is well known to produce an array of toxic effects, however that fact needs to be tempered with the understanding that the “dose makes the poison” (which is the title of one of the best plain-language guides on toxicology – a bit dated, but still sound information). Arsenic has been associated with adverse effects with human populations in different parts of the world including Taiwan and Bengladesh who were exposed to levels in drinking water exceeding 300 parts per billion (ppb) over a long period of time (much higher levels than found in apple juice). The distinguishing adverse effects associated with chronic ingestion of arsenic are skin lesions (hyperkeratoses and hyperpigmentation) and skin cancer. Other adverse effects from ingestion exposure include cancer of the internal organs (liver, bladder and kidney) and a vascular disease known as “blackfoot disease” (note: blackfoot disease is a condition endemic in an area of Taiwan where there are naturally elevated arsenic concentrations in drinking water). The U.S. Environmental Protection Agency (EPA) is also evaluating other potential health risks such as liver, bladder and kidney cancer as part of its work in developing a new drinking water standard for arsenic. What we know about chronic health risks is from ingestion of high levels in drinking water. Ingestion of low levels in food and water?  Reasoned speculation would be that the contribution to total cancer risks would be small – but that’s not a point you could argue as a fact, and it gets to be difficult to address it scientifically, because it’s a hypothesis that’s not readily falsifiable. Would need to do some calculations to elucidate that better – check back later. . . .

In the meantime, read “The Dose Makes the Poison”, and struggle through some monographs produced by the National Academy of Sciences, at least the executive summaries, published in 1983, 1999 and 2001. And find somewhere else to get informed about arsenic health risks other than Dr. Oz.

Environmental progress seems to be more difficult compared with the 1960s, 1970s or 1980s. The early successes resulted from going after low-hanging fruit (DDT, burning rivers, smog) which didn’t require a lot of individual sacrifice or change. It’s understandable that Western governments are reluctant to confront their citizens with the news that confronting climate change or resource depletion could involve curtailing everyone’s standard of living. The future of environmentalism involves a social revolution over values, and I’m not sure the professional environmentalist class is cut out for that. In the US, the professional environmentalists are still looking for legal/bureaucratic solutions from a system that’s either corrupt (Congress), or has undergone regulatory capture (the agencies), is too conservative for them (the courts) or has thrown them under the bus (the current Administration). Without governmental allies, against the money and media influence exerted by the regulated industries environmentalists are bringing a knife into a gun fight.

There appears to be some growing awareness of this problem. Some would argue that environmentalism needs to get more religious (see Lynn White for the definitive statement on faith and the environment). Someone is advocating a crowdsourcing approach blending today’s social networking tools with the values of the Summer of Love, though I’m with Terry Mann, the writer character played by James Earl Jones in “Field of Dreams”.

What’s an environmentalist to do if money, mainstream media and government aren’t in your corner, and you have difficult truths to convey? Satire and ridicule is a start, perhaps. Recognize that you’re fighting a culture war and start reading about 4GW. Stop being reactive and start preparing to play the long game. Time is on the side of environmentalism.

That needs some context. The group blog Lawyers, Guns and Money, which was a regular reading stop for me until they added as a writer obesity denier Paul Campos, recently posted a blog item about the impact of the meltdown of the Fukashima reactor in the United States. It was the standard cut-and-paste from another blog item which alleges that an apparent increase in infant mortality in the Pacific Northwest, reported on by the Centers for Disease Control, is associated with fallout from the meltdown of the Fukushima reactor. The item gets debunked multiple times in the comments, an example of the self-correcting nature of the Internet, and comments raise the typical issues about the original source (I skim Counterpunch for many things, but not thoughtful commentary). But there are other issues with regard to giving wider distribution to this piece. The original sourcing is understandable – the authors are anti-nuclear activists, though what they’ve written reflects badly on anti-nuclear activists as a whole. The placement is understandable – Counterpunch is a strongly leftist publication, one that I go to for leads or dirt but not thoughtful commentary. What is less understandable is how the item got apparently wider distribution, including Al Jazeera. It’s a statement and not a complimentary one about the judgment and questioning attitude on the part of journalists and Internet commenters.

On the favorable side, there was some good citizen science done at the blog The Capacity Factor, where a guest poster obtained the raw mortality data and conducted an analysis using the professional-grade statistics freeware package R (I’ve recently downloaded and started learning to use it – R is awesome). Events like this restore my faith in blogging and encourage me to get back out there. But the high-traffic blogs such as LGM remain a major disappointment for the ordinary reader such as myself.

As I’ve found recently in Tom Slee’s blog, Dr. Tetlock was able to identify the traits that made for more or less successful punditry. The experts who were more accurate than others tended to be much less confident that they were right. In addition, the less-unsuccessful experts are described as “foxes” (those who know many things) and the more-unsuccessful ones “hedgehogs” (those who know one big thing), after an essay by Isaiah Berlin. The hedgehogs seem to be winning the battle over eyeballs and consistent with the First Great Premise*, winning that battle determines what becomes truth.

*From Charles Pierce’s book, Idiot America: How Stupidity Became a Virtue in the Land of the Free, Doubleday, 2009. In Chapter Two, he articulates Three Great Premises about how we determine the truth of things these days. The First Great Premise: any theory is valid if it sells books, soaks up ratings or otherwise moves units; the Second Great Premise: anything can be true if someone says it loudly enough; the Third Great Premise: fact is that which enough people believe.

However, far be it for me to simply criticize CR’s work without making the attempt to try and communicate health risk issues with heavy metals in protein powders more clearly.  So, I’ll take a run at talking about cadmium, because I kind of ran arsenic into the ground with the last post (Note that an expanded version of this post, providing a more detailed discussion of cadmium risks from protein powders, can be found here).

Cadmium is naturally occurring in soil and water, and can accumulate into the food chain.  For nonsmokers, dietary exposure to cadmium is the most likely route of exposure, with nuts, grains and vegetables providing the largest contributions to dietary exposure.  The toxic effects of cadmium on humans have been extensively studied, and evidence in humans indicates that the kidneys and bones are important target organs for cadmium toxicity.  Scientific studies of adverse effects to the kidney have been used by government agencies to estimate the limits on cadmium ingestion that do not have an appreciable risk of adverse effects; these limits are based on the most sensitive indication of adverse kidney effects, and include a margin of safety.  Authorities responsible for assessing the safety of dietary supplements, such as the U.S. Pharmacopeia (USP) use these very same limits.  The article in CR uses these USP limits to draw attention to cadmium levels in protein powders and drinks.

But is your risk from cadmium really different with protein powders?  In general, the amount of cadmium exposure and risk from protein powders used as meal replacements probably is not appreciably different from meals prepared from ordinary foods.  This is not to say that protein powders (or foods for that matter) are free of risk.  Some recent scientific papers (examples here and here) argue that the margins of safety on cadmium exposure are either very small or inadequate.  However, it is arguable that one’s risk is no different with or without protein powders or drinks in the diet (some of the homework supporting this assertion is here).

The article in CR could lead you to think that in general everyone using protein powders are at risk of kidney disease from cadmium.  There are people who are at increased risk of adverse effects from cadmium exposure: age, general health and nutritional status, whether one is a smoker (smoking makes a substantial contribution to cadmium exposure) and the presence of pre-existing kidney disease, are key risk factors.  I tried to think of an example of someone who might be at an increased risk of cadmium-related adverse effects from using protein powders.  Here’s what I came up with:

A skinny hard-gainer teen-aged kid who, in addition to consuming a lot of protein powder also eats a lot of spinach, sunflower seeds, whole-wheat bread and oatmeal.  He or she would either be employed or have well-off parents in order to afford all of the protein powder being consumed.  He or she also would have kidney problems related to diabetes or some other preexisting condition.  Bonus points for being a smoker.

Keep all of this up through middle-age, and this individual might be at an appreciable risk of kidney toxicity or osteoporosis from cadmium exposure.  The take-away point, for me at least, is that normally healthy people including protein powders and drinks in conjunction with a healthy diet (and not smoking. . .) shouldn’t appreciably be increasing exposure or risk from cadmium.

Ultimately, it’s up to you what you do to maintain and enhance your health.  What people need to do that is information that explains as clearly as possible what the risks from heavy metals might be, and provide the information balancing benefits versus risks of using protein powders.  The sound-bites that were supplied in the article in CR generally were pretty un-informative for users of protein powders about the relative benefits and risks.

However, far be it for me to simply criticize CR’s work without making the attempt to try and communicate health risk issues with heavy metals in protein powders more clearly.  So, I’ll take a run at talking about cadmium, because I kind of ran arsenic into the ground with the last post.

So, what’s the nature of the problem we are trying to address?  According to CR’s testing, a few protein powder products contain cadmium concentrations that are higher than the U.S. Pharmacopeia’s permitted daily exposure (PDE) limits.  We’ll defer for a moment the question of how much information CR’s testing actually provides about potential cadmium exposure from consuming protein powders – simply publicizing that any cadmium has been detected has been sufficient cause to elevate concerns about adverse health effects in the minds of consumers.

The toxic effects of cadmium on humans have been extensively studied.*   Humans consuming cadmium-contaminated rice in Japan have experienced adverse effects in the kidney and bones. Inhalation of cadmium by workers in certain industries has also been associated with kidney toxicity (inhalation by workers is associated with other effects such as lung diseases including lung cancer, but we’re interested in the risks from ingesting cadmium). There is conflicting evidence as to whether or not cadmium exposure produces high blood pressure in humans; cigarette smoking is an important source of cadmium exposure, but by itself may be a confounding factor. Excessive cadmium ingestion exposure in combination with low dietary intake of iron may be associated with anemia.

These pieces of information are useful in understanding the kinds of toxic effects that could be associated with cadmium exposure, based on what human populations have experienced in the past – typically from elevated levels of exposure.  However, these aren’t enough by themselves; there is additional information that needs to be considered to better understand how exposure under a specific circumstance (using protein powders, for example) could pose a health risk.

For example, elevated levels of exposure to cadmium are associated with an increased risk of kidney disease.  Lower levels of exposure do not produce observable kidney disease, but will be associated with an increased prevalence of abnormal biomarkers.  In this case, the abnormal biomarkers would be increased levels of certain low molecular-weight proteins in the urine.  These biomarkers are the earliest and most sensitive indications of changes in kidney functioning.  Some accumulation of cadmium in the kidney without apparent toxic effect is thought to be possible because cadmium in the kidney becomes bound to a protein called metallothionein.  However, when cadmium levels in kidney tissue become high enough, the amount of cadmium not bound to metallothionein becomes high enough to cause kidney toxicity.

Effects to the bone (osteoporosis, increase in bone fractures) were first observed in Japanese women living in areas with high cadmium contamination.  This was a disorder referred to as “itai-itai” (ouch-ouch) disease.  Other factors that may have been important in itai-itai disease included multiple pregnancies, poor nutrition (low calories, reduced calcium, protein, vitamin D, iron and zinc intakes).  The adverse effects in bones occur in the presence of kidney disease from cadmium exposure, and the kidney is thought by some to be a more sensitive target organ for cadmium toxicity.  Other researchers argue that effects to the bones, such as osteoporosis, occur at the same levels that produce kidney toxicity.  However, currently, the limits on cadmium exposure used to assess the risks of dietary intake consider the kidney to be the most sensitive target organ.

The excretion in the urine of biomarkers (low molecular-weight proteins) has been shown to increase due to cadmium-related alteration in kidney function; alteration in these biomarker levels is considered the most sensitive toxic effect from cadmium exposure.

This information has been used to identify a cadmium exposure level in humans associated with minimal health risks.

The Agency for Toxic Substances and Disease Registry (ATSDR) used a 10% increased risk of low molecular-weight proteinuria (proteinuria – excretion of excessive levels of protein in the urine) as an indication of cadmium-associated kidney toxicity.  Selection of the 10% value is based on a typical limit of detection for adverse effects in health effects studies.  ATSDR reviewed numerous scientific studies and estimated the cadmium dose in the body corresponding to the 10 percent increased risk of low molecular-weight proteinuria; that internal dose was estimated in those studies from the levels of cadmium measured in the urine of exposed persons (the researcher’s short-hand for describing this internal dose is the “urinary cadmium dose, 10%” or UCD10).

The next step is to estimate the daily intake rate of cadmium into the body that corresponds to the UCD10.  This uses the results from a “physiologically-based pharmacokinetic model” (the short-hand is “PBPK model”).  The PBPK model provides a mathematical description of the absorption, distribution throughout the body, accumulation in target organs and excretion of ingested cadmium.  The results from the PBPK model can then be used to relate the UCD10 to a daily intake rate over a lifetime.  That daily intake rate is known as the Minimal Risk Level (MRL).  The MRL is defined by ATSDR as: “an estimate of daily human exposure to a substance that is likely to be without an appreciable risk of adverse effects (noncarcinogenic) over a specified duration of exposure.”

This is the approach that U.S. Pharmacopeia (USP) used to develop its Permitted Daily Exposure (PDE) value for cadmium.  As USP described it:

Using data from select environmental studies examining the relationship of urinary cadmium and the prevalence of elevated levels of biomarkers of renal function ATSDR issued the provisional Minimal Risk Level (MRL) for chronic cadmium exposure. The 95% lower confidence limit of urinary cadmium dose corresponding to the probability to exceed in 10% the risk of low molecular weight proteinuria has been estimated as 0.5 ug/g creatinine, assuming accumulation over a 55-year period.  This value corresponds to an intake of 0.33 ug/kg/day in females, for which, applying a safety factor of 3 for human variability. ATSDR has set the MRL to 0.1 ug/kg/day.

The factor of 3 for human variability takes into consideration sensitive individuals (I talk more about sensitive individuals in a moment).  There are also a couple of critical assumptions to be aware of here.  First, the 95% lower confidence limit of the UCD10 is used to calculate the corresponding daily intake rate (i.e. the MRL).  The lower confidence limit provides for a more protective daily intake rate.  Second, the UCD10 is based on constant exposure to cadmium over a 55 year lifetime.  I come back to this latter point when discussing scenarios that may be associated with increased risks from cadmium exposure.

USP assumes that an individual with a 50 kg body weight (about 110 lbs) is exposed to the MRL of 0.1 ug/kg-day to calculate the PDE of 5 ug/day (0.1 x 50 = 5).

So far, we’ve identified the key adverse effects associated with cadmium ingestion, and we’ve quantitatively evaluated the levels that produce adverse health effects.  Recall from the previous post these were two of the three critical pieces of information needed to properly assess the health risks from heavy metals exposure.  The third piece of information is to explore the patterns and magnitude of human exposure to cadmium.

For nonsmokers, dietary exposure to cadmium is the most likely route of exposure.  Cadmium is naturally occurring in soil and water, and can accumulate into the food chain.  Cadmium levels in food can vary depending on the type of food, agricultural practices, the amount of atmospheric deposition (particles in air that deposit onto the ground) and industrial contamination in soil.  In general, leafy vegetables, grains, root vegetables and seeds (nuts and sunflower seeds) have higher levels of cadmium than other foods.  Organ meats such as liver and kidneys, and seafood, concentrate cadmium and therefore have relatively higher levels.  Dairy products, beef and poultry, and fruits have relatively lower levels of cadmium.

The Food and Drug Administration (FDA) performs laboratory analyses of metals in food as part of its Total Diet Study (TDS).  The TDS data might be a good way of comparing with other foods the cadmium concentrations reported in protein powders by CR; that is, if CR had actually reported concentration data (the problems with CR’s analytical results are a topic for another post).  CR’s lack of transparency in reporting metals data for protein powders makes it difficult to compare the risks relative to other foods.

Maybe there’s another way to get at this.  Based on food intake rates and concentrations in foods, a typical daily intake of cadmium for the U.S. population is estimated to be 18.9 micrograms per day or ug/day (note for the more scientific types: that statistic is the geometric mean).  According to CR’s data, cadmium exposures from protein powders and drinks range from 1.6 to 5.6 ug/day (based on the assumption that someone consumes three servings per day).  It could be difficult to argue that the cadmium intake is in addition to the daily intake from other foods; part of the criticism of protein powders is that they are consumed to the exclusion of other foods.   Without answering this question about how cadmium exposure from consumption of protein powders relates to cadmium exposure from all dietary sources, it is difficult to argue either that protein powders significantly increase one’s risk of adverse effects from cadmium OR do not increase one’s risk.

How about some examples?  I created some hypothetical healthy meals that people trying to get lean or stay lean might eat.  Using the TDS cadmium data (note for the nitpicking types, I just used the first quarter 2005 data – FDA samples foods quarterly as part of the TDS – I’m just trying to quickly get a rough indication of exposure here), I estimated the cadmium intake with these meals.  I’ll have to share the calculations in another post, because this is getting pretty long now by blog standards, and there are still a number of points that need to be made.  I present a range of estimated cadmium intakes, because of how the TDS reports non-detected analytical results.**  These examples are as follows:

• Breakfast consisting of two scrambled eggs, 1 cup of oatmeal and 6 oz of grapefruit: 1.07 to 1.21 ug cadmium.
• Snack consisting of 2 oz of peanut butter, two slices of whole-wheat bread and a banana: 2.05 to 2.4 ug cadmium.
• Dinner consisting of an 8 oz chicken breast, one cup of steamed broccoli and a cup of rice: 4.1 to 4.6 ug (the grains and veggies contained the higher concentrations of cadmium).
• A spinach salad with mushrooms, an ounce of shredded cheese, whole wheat croutons and Italian dressing weighs in at 4.8 to 4.9 ug – with most of the exposure coming from the vegetables.

As CR tells us, consuming three servings of protein drinks or protein powders corresponds to 1.6 to 5.6 ug of cadmium.  Eyeballing all of these numbers indicates they all (from the protein powders and example meals) fall within a narrow range – between 1 and 6 micrograms, and the exposures from the protein powders don’t look terribly distinguishable from the meals.  The point here is to not scare you about eating real food.  In a narrow sense, it’s probably correct that other sources of protein such as meat, milk and eggs provide lower exposures to cadmium.  However, if protein powders or drinks are being used as meal replacements, particularly if they are replacing consumption of grains, nuts and vegetables, then the difference in cadmium exposure with or without protein powders in the diet is probably insubstantial.

There are factors that could make an individual more susceptible to cadmium exposure than most, including age, general health and nutritional status, other sources of exposure to cadmium (for example, cigarette smoke), and exposure to other toxic substances – say mercury, for example.  These factors could affect how well cadmium is absorbed in the body, or could reduce the capacity to detoxify or excrete cadmium.  In addition, organ damage from preexisting diseases could affect cadmium toxicity.  Smoking substantially increases cadmium exposure.  Nutritional deficiencies, such as calcium or iron deficiencies, could increase cadmium absorption from the gastrointestinal tract – though this has not been demonstrated consistently in all of the pertinent studies.  Protein deficiencies could affect a person’s ability to detoxify cadmium.  Individuals with decreasing kidney function from unrelated causes – such as diabetes or aging – could be more sensitive to adverse effects from cadmium.

So, what kind of a scenario could we envision for an individual who could be at risk from ingesting cadmium in protein powders, if we overlay CR’s paradigm with some of the known risk factors.  What comes to mind for me is this:  a skinny hard-gainer teen-aged kid who, in addition to consuming a lot of protein powder also eats a lot of spinach, sunflower seeds, whole-wheat bread and oatmeal.  He or she would either be employed or have well-off parents in order to afford all of the protein powder being consumed.  He or she also would have kidney problems related to diabetes or some other preexisting condition.  Bonus points for being a smoker.  Keep all of this up through middle-age, and this individual might be at an appreciable risk of kidney toxicity or osteoporosis from cadmium exposure.

This in no way is being dismissive of the health risks from cadmium, though I hope it’s been a useful counterpoint to the faulty messages in CR’s article.  There are recent scientific papers (here, here and here) highlighting cadmium as a public health concern, noting that the most recent evidence suggests the margins of safety are very small or inadequate between levels that could begin to produce adverse effects and normal intakes from diet and smoking (so don’t smoke, already).  However, it’s important to keep in mind that the MRL (which again is the threshold for the most sensitive effects from a lifetime of cadmium exposure, reduced by an additional factor of three, and is also the number used by USP for assessing cadmium levels in diet supplements) is as protective as feasible, based on the available scientific information.  Would occasional fluctuations in exposure above that level cause harm?  That would depend – how large were the fluctuations, how often did they occur and did they occur to someone who is a sensitive individual, like a diabetic, a smoker or someone already experiencing kidney disease.  The take-away point from this analysis, for me at least, is that in normally healthy people including protein powders and drinks in conjunction with a healthy diet (and not smoking. . .) shouldn’t appreciably be increasing exposure or risk from cadmium.

Ultimately, it’s up to you what you do to maintain and enhance your health.  Some of the experts that CR interviewed did folks a disservice by saying in effect you shouldn’t consume protein powders because the heavy metals in them might increase the risk of adverse health effects; instead, they should have taken the time to explain to people as clearly as possible what those risks might be, and provide the information balancing benefits versus risks of protein powders – information that might be more useful in helping users make their own decisions about using these products.  But that’s a harder thing to do – the topic just doesn’t fit into a sound-bite, which is all that the authors allowed from these experts (I’ve expended about 2,800 words on it so far, and still have only scratched the surface).  The writers of the article in CR just didn’t try very hard, and the sound-bites they provided generally were pretty un-informative for users of protein powders about the relative benefits and risks.

Stay tuned.  I’ve got three more contaminants to go through.  The shorter version of this post minus the extended analysis can be found here.

Notes:

*Much of the discussion of cadmium-related adverse health effects in humans has been drawn from the ATSDR’s toxicological profile for cadmium.

**The TDS will report foods in which no cadmium was detected as zero mg/kg (milligrams of cadmium per kilogram of food).  However, the TDS also will report the “limit of detection” or LOD, which is the lowest concentration that the analytical method used on the food could detect.  So, if the TDS reports for whole milk a cadmium LOD of 0.002 mg/kg (or ug/g, microgram cadmium per gram of food), that means that cadmium intake associated with consuming an 8 oz cup of milk (226 grams) is probably less than 0.45 ug, though how much less might be difficult to say.   So, I reported a range of estimates – one without the LOD (foods reported as not detected are treated as zero) and one with the LOD (foods reported as not detected are assumed to have contamination at the LOD).  Again, this is a topic for a separate post.