Kevin Fischer (and William Ryan), 2nd place

Abstract

The human body is a complex system with many interacting parts that are not fully characterized. Minerals we consume do not directly translate to mineral levels in the body. Multiple minerals compete for the same transport mechanisms, meaning that ratios can be more important than absolute levels. Minerals from dietary intake exist in multiple forms, which are metabolized and absorbed differently by the body, and they are always consumed in the presence of other nutrients. Supplementation usually involves isolated forms of individual minerals, and even multi-mineral supplements are not guaranteed to have the right ratios for optimal nutrition.

Humans have been eating whole foods and demonstrating good health since before the technology existed to create isolated nutrients, and this represents the null hypothesis of nutrient intake. Supplementation is the alternative hypothesis, and requires evidence of greater efficacy to be recommended. Studies show that dietary intake and certain ranges of blood levels of minerals are correlated with optimal health. Controlled trials of mineral supplements fail to show benefits (and sometimes show harm) outside of the context of established mineral deficiencies. Given this evidence, we need to be optimizing dietary intake of minerals in order to provide optimal health. 4 ounces of grass fed beef liver per week goes a long way towards optimizing dietary mineral intake.

We recommend that patients begin with a comprehensive mineral testing panel, to assess overall mineral status and identify any deficiencies, and do follow up testing after any dietary or supplement-based interventions. From our survey of the literature, we recommend a diet that empirically maximizes nutrient intake and absorption. We recommend that patients keep a food diary to measure compliance to the dietary guidelines and identify any nutritional shortfalls. We only recommend supplementation if they do not comply with or respond to the dietary guidelines.

Recommendations

Our primary recommendation is to consume 4 oz of grass-fed beef liver at most twice per week but at least once per week. Beyond that, increasing intake of other offal, seafood, meat, and leafy green vegetables is also very beneficial. These foods should be added at the expense of nutrient-poor foods such as refined sugar and flour, high-fructose corn syrup, and industrial seed oils.

We also recommend decreasing dietary intake of phytic acid substantially, which is most effectively accomplished by reducing nuts, seeds, grains and legumes in the diet. If necessary, try to consume sprouted or fermented products of these foods. Otherwise, dehulled versions of these foods (e.g. white rice and white flour) also contain low phytic acid levels, though they are nutrient-poor. Replace these foods with the ones above, plus fruits, root vegetables, all other vegetables and dairy (if tolerated).

Other dietary recommendations include remineralizing drinking water with trace elements (particularly magnesium), avoiding coffee, tea and cocoa during meal times, and avoiding boiling as a method of cooking.

We recommend trace mineral nutritional testing as a before and after check on patients implementing this diet to give a metric of nutritional status improvement. We also recommend regular diagnostic mineral testing anytime supplements are indicated, such as in the case of patients who cannot comply with this diet or otherwise don’t respond to it. Note that comprehensive nutritional testing (including fatty acids, organic acids, and coenzyme q10 non-mineral nutrients) represents a better consumer value than some of the more sensitive mineral markers discussed and is our true recommendation but is beyond the scope of this document.

Report Body

Nutrition science has inherent methodological problems that make it difficult to do empirical research. Randomized controlled trials are prohibitively expensive, and there are significant moral considerations involved. Animal models are only valid insofar as the relevant features are similar to humans. In vitro experimentation is useful for piecing out individual causal mechanisms, but there are almost always multiple factors involved. Most human studies are observational and as such can establish correlations to investigate further but cannot establish causal relationships.

The human body is a complex system. There are tens of trillions of cells in the human body, each of which express different portions of their genome to different degrees, signaling to each other over long distances using hormones and short distances using cytokines. Our understanding of this system is constantly growing, but we don’t yet know how much we don’t know. Chemical reactions are occurring all throughout the body, either spontaneous or catalyzed by enzymes, and we are not yet able to model these in detail. This complexity makes it difficult to determine the exact effects of anything we put in our body.

“Eat food high in nutrients” represents something like the null hypothesis on nutrition - human beings were eating food for millions of years before extracting individual constituents was even possible. “Take supplements” is the alternative hypothesis. A comprehensive examination of the literature on supplementation reveals widespread critical flaws, such that we can only recommend supplementation in specific cases of verified deficiencies. Even then, we believe supplementing with specific whole foods to be superior to individual mineral supplementation.

Conventional wisdom is that taking supplemental minerals is the healthful choice. The evidence is actually quite weak for supplemental minerals reducing all cause mortality. Sometimes mineral supplementation shows increased risk of mortality[1]. The problem is that most people take supplements in a vacuum. Supplemental minerals do represent a benefit when fixing deficiencies, but represent a severely unnatural load on the body when supplementing in cases of mineral sufficiency. Personal nutritional testing, one of the best ways to diagnose nutrient deficiencies, is rare and not encouraged by most physicians. Scientific understanding of minerals as an integrated system is woefully inadequate. We do know that some minerals affect the absorption of other minerals: zinc supplementation interferes with the absorption of magnesium[2].

The only combinations of minerals that are well understood to be properly absorbed are the combinations found in whole food. To really, actually understand how vitamins and minerals interact with regards to preventing cancer and promoting health, a systemic approach is necessary[3].

There have been surprising results in the literature, where a vitamin or mineral believed to have a beneficial effect from dietary intake, actually has the opposite effect when consumed as a supplement. One study showed a 39% increase in incidences of lung cancer when Vitamin A and beta carotene were given to smokers, former smokers, and people exposed to asbestos, which was a very surprising and disappointing result[4], given that increased

dietary Vitamin A clearly decreases lung cancer incidence[5]. A more recent study also shows that supplemental multivitamins do not decrease lung cancer risk[6], along with this study[7] showing no decreased heart disease risk. But increased fruit consumption does seem to decrease cancer risk[8]. Another study weakly showed chronic zinc oversupplementation to increase advanced prostate cancer risk by more than 2x, though the observational study did not allow for confounding factors to be ruled out[9].

One problem with supplementation is that vitamins and minerals exist in many different isolated forms. When consumed from food, people get a whole spectrum of bioavailable forms of nutrients. Supplements often only contain one form of a particular nutrient. The Nutritional Prevention of Cancer Trial showed a striking, 63% decrease in prostate cancer when supplemental high-selenium nutritional yeast was given. The nutritional yeast used had a selenium mix consisting of L-selenomethionine, selenocysteine, Se- methylselenocysteine, selenoethionine, selenoglutathione, selenodiglutathione, and selenite[10]. When the study was replicated, scientists voted to supplement with only L- selenomethionine and found no decrease in prostate cancer and a statistically insignificant increase in type 2 diabetes[11]. A discussion of these results proposes that it could be the high selenium nutritional yeast itself that is the active anti-cancer agent[12]. Studies such as this lead us to conclude that vitamin formulators sometimes make very bad decisions when working with the inadequate information available.

Compared to the hypothesis of supplemental minerals improving health outcomes, the evidence is much stronger for non-nutrient deficient diets reducing all or specific cause mortality. Dietary magnesium intake correlates with decreased risk of coronary heart disease[13]. Another study shows increased dietary calcium may correlate with decreased colorectal cancer risk in Swedish men and may correlate with decreased all-cause mortality[14]. Dietary calcium intake directly correlates with decreased stroke risk in a large Japanese study group[15].Higher fruit and vegetable consumption seemed to correlate with decreased heart disease mortality, though the likelihood of the causal connection is disputed[16].

We believe that the evidence indicates strongly that minerals should be supplemented only in the case of marked deficiencies - that dietary minerals are not always shown to be good, but they seem to practically never cause gross harm to the patient. We think that supplementing with whole foods strictly trumps isolated minerals. The evidence on supplementation shows beneficial effects when given to populations likely to have subjects with marked deficiencies, like American elementary schoolchildren from a lower income school. Some of the study subjects had improved verbal IQ, with the investigators hypothesizing that those that responded best to supplements were those deficient in important vitamins and minerals[17].The Linxian trials also showed positive results when supplementing vitamins and minerals in a nutrient deficient population[18].

There is also evidence that for a number of minerals, the ratios of minerals consumed are more important for health than the absolute amounts, an important factor only partially considered in the US government guidelines on nutrition. Nancy Cook’s group concludes that the ratio of sodium to potassium matters greatly for cardiovascular disease, with

approximately 2:1 potassium:sodium consumption having the lowest rate of cardiovascular events. They are also one of very few studies cited here to use a Bayesian Information Criterion for model fitting! Of course, as yet another observational, not double-blind study, cause and effect was not proven[19]. A follow-up analysis of a separate cohort shows similar results[20].

The zinc:copper ratio hypothesis, that imbalances between zinc and copper lead to coronary heart disease, was proposed in 197521. Since then, the zinc:copper ratio has gone on to be a common nutrition metric used in studies looking at many different kinds of disease pathology[22][23][24]. Zinc:Iron is another mineral interaction pair with a useful ratio metric[25][26][27].

That some ratios are accepted as very important but most ratios are ignored is one of the most striking inadequacies of nutritional research. However, moving forward with identifying meaningful mineral ratios quickly becomes intractable because of the combinatorial explosion involved in looking at all minerals and vitamins in relation to each other. Our bodies are nothing if not complex systems. This nutrient ratio combinatorial explosion hypothesis is one of the reasons we feel so strongly about whole foods as opposed to supplements.

There are two ways to improve nutrient status from changing food intake: increase the mineral content of the diet, or increase the mineral availability of the diet.

Increasing mineral content of the diet involves replacing nutrient-poor foods with nutrient- dense foods. The USDA publishes a nutrient database that has measured values of most dietary minerals for almost 8000 foods[28], which we are using for our recommendations.

The strongest finding is that seafood and meat, particularly organ meats, are capable of supplying almost every major essential nutrient: potassium, iron, zinc, copper, manganese, selenium, molybdenum, chromium and iodine. Leafy green vegetables are also very nutrient dense, supplying calcium, magnesium, potassium and iron, plus many of the trace elements. Root vegetables, other vegetables and fruit all contain some of these minerals. Nuts, grains and legumes also have high measured levels of many minerals, which we discuss further below.

Our strongest recommendation of this paper is consumption of 4 oz/week of grass-fed liver, no more than twice/week. Liver is much more readily available than other organ meats (though ideally a mix of organ meats is recommended), and has the greatest quantity and variety of minerals. Grass-fed beef has a higher proportion of desirable minerals than grain fed beef, notably calcium, magnesium, and potassium[29]. Grass fed beef had 37.0 ug/g of zinc compared to 28.9 ug/g in grain fed cows. Calcium was 96.0 ug/g and 84.8 ug/g, respectively. Potassium 3721.5 ug/g to 2242.7ug/g[30]. Direct studies on the effect of liver consumption are rare - funding for research into whole foods generally is largely supplied by industry groups, which are seen as a biased source of funding, whereas pharmaceutical companies are required by the FDA to undergo extensive testing before approval, leading to much more research on drugs than whole foods. Most studies on organs are correlating

multiple aspects of diet with various health outcomes, such as a study where poor Jamaican men eat both high organ meat intake as well as fast food (and see no increased prostate cancer risk), not a particularly useful result[31] One study finds organ meat intake is not associated with pancreatic cancer in China[32]. We found one epidemiological study that did find a correlation between consuming liver more than twice/week and stomach cancer[33], though it is notable that they studied 32 individual foods plus combinations thereof and only discovered three correlations an enormous number of combinations to only find three correlations. One of their was vegetable intake correlating with stomach cancer in women, suspicious on priors. Note that we do not recommend consuming liver more than twice/week due to excessive vitamin A – one recommendation in the literature is 100 g/week (slightly less than 4 oz), and 50 g/week if you are a pregnant woman[34]. One common concern with eating liver is that it accumulates toxins, though the dioxin and PCB content of the most common beef, calf, pig and lamb liver is below tolerable daily intake when eaten twice/week or less and grass-fed liver should be even lower in toxins[35]. Note that fish liver in particular has stronger evidence of health benefits - a study investigating toxin exposure in fish liver inadvertently found consumption was protective against total cancer risk (relative risk = 0.92), despite 37% of the non-consumption group taking cod liver oil supplements. This effect, however, is more likely due to omega-3 intake or vitamin D levels[36] and is thus highly beneficial but outside the scope of this paper.

Foods with low nutrient density occur when an individual component is extracted from a whole food, for example sugar, high-fructose corn syrup, white flour, or refined seed oils. These foods contain calories with few nutrients, which means either calorie intake must increase and/or nutrient intake must decrease when these foods are primarily eaten. In order to include the above nutrient-dense foods, these necessarily need to be reduced to maintain equal caloric intake.

Note that food is not the only source of dietary minerals. Before water purification technology became widely available, significant amounts of magnesium, calcium and other minerals were usually consumed through water. A series of studies have investigated the link between water hardness (high magnesium/calcium levels) and cardiovascular disease and mortality rates. Some find important roles for both[37], others find only magnesium matters[38], others find only calcium matters[39], yet more find that the correlation still holds and isn’t explained by either calcium or magnesium[40], and finally some studies find no correlation whatsoever[41]. This puzzle can easily be explained by the fact that hard water composition is different across regions. Not all of them may have the ideal protective nutrient ratio. Two recent review studies have surveyed the literature, and found that magnesium does seem to have a significant impact while the evidence for calcium is mixed[42][43]. One of the studies estimates that the odds ratio of cardiovascular death is 0.75 in areas with 8.3-19.4 mg/l versus 2.5-8.2 mg/l magnesium content in drinking water. While the research has primarily focused on these two minerals, given our understanding of functional nutrition ratios it seems likely that trace elements are a critical part of the puzzle - for instance, up to a quarter of copper intake may come from drinking water[44]. In a case of confirmed magnesium deficiency, particularly in areas with demineralized water, trace mineral extracts (such as commercially available ConcenTrace Supplement, consisting of minerals evaporated from the Great Salt Lake with excess sodium removed) can be added

to drinking water to replace this source of nutrients.

The other important factor in nutrient status is the bioavailability of minerals in the diet. The main factors we discuss are cooking methods and minimizing various dietary components which actively bind to minerals and chelate them from the body.

Cooking is essentially doing organic chemistry, and has a wide range of effects on different nutrients - here we focus on the mineral content and availability. Unlike molecules, elements cannot be easily destroyed by heat, light, oxidation or pH. Instead, minerals can be physically removed from food, and potassium appears to be the mineral most vulnerable to leaching. 50% of potassium was lost from boiled potato cubes and 75% from boiled shredded potatoes - plus some losses of phosphorus, magnesium, sulfur, zinc, manganese, and iron - though this effect was not present with baking, roasting or microwaving[45]. Most methods of cooking meat also cause moisture loss, which reduces the potassium content of the meat below required intake on high-meat diets[46]. We recommend any cooking method other than boiling for vegetables and preferring stews and soups when most food intake is coming from meat.

The biggest factor in mineral availability is the presence of molecules which bind strongly to and form complexes with minerals, thus rendering them unavailable for use in the body. There is the most evidence around anti-nutrient effects of polyphenols and phytate.

Polyphenols are structural organic molecules found abundantly in plants, the most prevalent of which are condensed tannins (forming up to 50% of the dry weight of leaves). Most of the research on polyphenols and tannins looks at the effects on iron absorption. Dietary iron is either considered heme or non-heme. Heme iron comprises half of iron intake from animal sources, everything else is non-heme iron. Bioavailability of iron varies widely, from 15-35% of heme iron to 2-20% of non-heme iron[47] which is largely determined by polyphenol and vitamin C intake. In a controlled trial looking at tea consumption, 20-50 mg or 100-400 mg polyphenols reduced iron absorption by 50-70% and 60-90% respectively[48], though it is noteworthy that the intervention group had sweetened beverages while the “control” group had unsweetened water. To put this in perspective, average US intake of polyphenols is on the order of 1 gram per day[49]. Tannic acid and gallic acid appear to be very effective, with 5 mg, 25 mg and 100 mg inhibiting absorption by 20%, 67% and 88% respectively[50]. Fortunately, vitamin C can prevent the iron from forming complexes, and also reduces ferric iron to ferrous iron which increases uptake[51]. This effect occurs even at small doses: 50 mg of vitamin C (the amount in a small orange) reduced polyphenol inhibition by 25%, and 100 mg by 50%[52]. This once again hammers home our thesis - nutrients are not meant to be consumed in isolation. While research on other minerals is very sparse, the effects appear quite strong: tannins from walnuts complexed almost 100% of copper in solution, while almond tannins complexed 84% of zinc[53]. Our recommendation with respect to iron is to obtain iron from animal sources if possible, and this may also be true of animal forms of other minerals such as zinc[54]. Due to more general concerns about mineral uptake, we recommend that patients only consume coffee, tea or cocoa between meal times.

The other major anti-nutrient with respect to mineral availability is phytate. Phytate (also

called phytic acid) is a molecule used by plants as their primary form of phosphorous storage. This is most prevalent in the hull (outer shell) of nuts, seeds, grains and legumes. For example, wheat contains 0.72-0.93% phytate by dry weight, corn 0.85-1.02%, peas 1.02%, and soybeans 1.43%[55]. Nuts are even higher, with almonds at 2.11%, cashews 1.22%, and pistachios 2.84% for example[56]. In comparison, one study found rates of 0.11- 0.27% in 8 species of potatoes[57], another looked at 15 species of root vegetable and found the most at 0.17% in taro and no detectable levels of phytate in sweet potatoes[58]. Dietary intake of phytate in the United States varies significantly, with 631 mg for omnivorous women, 746 mg for omnivorous men, 1293 mg college students, and even higher in specific vegetarian populations - compared to just 180 mg in Swedish diets and 219 mg in Italian diets[59].

Phytic acid binds readily to most minerals, albeit with varying strength and stability. Zinc and copper are both very prone to form tightly bound complexes[60]. Reducing phytic acid content from 500 micromol to 110 micromol by leavening bread doubled zinc absorption rates[61]. Magnesium absorption was reduced from 32% to 24% and 13% by adding 0.75 mmol and 1.49 mmol of phytic acid respectively[62]. There is a remarkably strong correlation of -0.8 between iron absorption and phytate levels of infant cereal foods[63]. In a study (funded by Nestle) where all phytic acid was destroyed (which usefully represents an absolute upper bound on effect size), iron absorption from rice was increased by 308%, corn by 496%, oats by 836% and wheat by 1,160%[64]. Phytate in nuts appears to work similarly, cutting iron absorption by more than two-thirds when added to bread (though 50 mg vitamin C prevents this in some cases)[65]. Vitamin A and beta-carotene also appear to have a synergistic effect on iron absorption, increasing rates two to three times from rice, corn and wheat[66], providing more evidence that nutrients should not be consumed in isolation. Traditionally the ratio of phytate to minerals has been used to predict bioavailability, however total phytate content alone is an equal or better predictor[67][68]. This suggests that merely increasing mineral intake is not sufficient to produce adequate dietary intake, instead limiting phytate in the diet seems important for optimal nutrient status.

There are two options to reduce phytate intake: process the food such that phytate is destroyed, or avoid foods which contain phytate. The enzyme phytase is used by various bacteria and animals to digest phytate, though humans only produce 1/30 the amount as rats, who are evolved to eat grains[69]. Some native gut bacteria can produce phytase[70], which suggests that proper maintenance of gut bacteria could theoretically improve mineral digestion, but currently this is speculative. Traditional cultures that rely heavily on grain and legume intake have a variety of preparation methods: milling (to remove the bran), pounding, soaking, sprouting, fermenting, and cooking. In most grains (except for corn) the bran contains most of the phytic acid, so white rice and white flour for example have levels comparable to root vegetables[71]. Unfortunately, the bran also contains most of the nutrients in grain, which makes it little more than a source of starch calories. Cooking alone only appears to destroy about 40% of phytic acid in grains[72], while yeast leavening and cooking destroyed 22-58% of phytic acid in whole wheat bread[73] Soaking, sprouting and fermenting are unlikely to be done at home by any of our patients, however various sprouted and fermented products are available commercially, especially sourdough bread. Particularly when pounded and/or processed at higher temperatures, any of the above methods can reduce phytic acid content by up to 50-75% and may be much more effective in combination[74][75][76][77]. Given the difficulty involved in significantly reducing phytic acid content of foods, our recommendation is to avoid nuts, seeds, grains and legumes insofar as the patient will comply. If such foods are necessary, we recommend obtaining sprouted/fermented products, or using dehulled foods like white rice or flour.

We recommend that patients working to optimize nutrient intake keep a food diary. Superior compliance has been shown with Food Frequency Questionnaires. See this US FFQ[78] and two studies validating the use of questionnaires[79][80]. However we have not found a food frequency questionnaire sensitive to the particular guidelines of our recommended diet. As food diaries have been shown to be quite reliable when followed[81], we recommend patients keep a food diary until we can develop a custom food frequency questionnaire. The food diary is very important Bayesian evidence when used as a tool to combine with nutritional testing.

There is no single specimen that provides a complete and accurate picture of whole body levels of essential minerals. To get a complete picture, it is required to test multiple specimens. The most accepted specimens are whole blood, red blood cells (erythrocytes, RBC), urine, and hair, though other markers exist for testing for specific compounds. Whole blood and serum are the most common baseline nutritional diagnostic test. RBC testing reflects longer term nutritional status, over a window around four months long. Urine testing is very short term, reflecting nutrient processing over a single 24 hour period or even less[82].

It stands to reason that hair testing should be an extremely useful testing indicator, and Laboratory Evaluations for Integrative and Functional Medicine does recommend hair testing as a useful indicator. However, the evidence on hair testing is very mixed and has revealed testing labs rife with outright fraud and incorrect interpretative guides. In 1985 a study found hair testing to be exceptionally unreliable and results could not be replicated[83]. In 2001 this result was followed up on with similar results[84]. An editorial response points out some of the mechanisms for why results were so different, which is most likely the inconsistent digestion methods used to prepare samples and reduce environmental contamination. Also, hair mineral content varies along the length of the shaft, so if the samples of hair were submitted to the lab as powder but not strands, it could explain some of the variation[85][86]. Another study in Germany again found widely different results for the same samples sent to different labs[87]. At its best, from a particularly reputable, independently verified lab, hair testing should be repeatable, but interpretation is very tricky. It seems suspiciously like reading tea leaves, allowing a practitioner to make whatever confirmation they want based on the results, because paradoxical results are common in hair testing. High hair calcium is a sign of calcium deficiency, except when it is actually a sign of too much calcium in the body. Such decisions are left to practitioners and interpretation guides. Certainly hair testing can be useful Bayesian evidence, but because the medical establishment is so rationally skeptical of it I recommend that practitioners don’t offer hair testing, except as a possible confirmation of toxic metal exposure. Environmental contamination can often render hair testing useless (some shampoos are high in selenium, calcium), but this effect is actually a feature with regards to testing for toxic metals.

Tissue samples obtained via biopsy could be one of the most valuable specimens for testing intracellular mineral concentrations but except in the case of particularly pernicious chronic illness or clinical trials where exceptionally accurate testing is need, biopsies are almost certainly not worth the patient discomfort involved[88]. Nuclear Magnetic Resonance Spectroscopy also provides a relatively non-invasive, sensitive means of testing for magnesium[89]. Despite this being known in the literature for decades, NMR magnesium testing is exceptionally rare in the clinic and seems to only be done in research studies.

In choosing which tests to use for a patient, tradeoffs have to be made between price, convenience, and ease of interpretation. By testing all possible specimens for many different functional indicators, a very complete portrait of nutrient levels can be realized.

However, such a complete profile is often unnecessary when paired with patient food diaries. We can use the nutrients a patient regularly consumes as strong Bayesian evidence for confirming partial lab testing results. Instead of quintupling up on multiple mineral indicators, it represents a better value to the patient to choose a more comprehensive but shallower nutritional test, such as a test of amino acids, fatty acids, antioxidants, and coenzyme q10 in addition to minerals. Such a test is beyond the scope of this document, focused mostly on minerals.

Urine testing is accurate for measuring short term absorption of nutrients. A 24 hour collection of urine is commonly done, which means having the patient collect 4 or more urine samples over the course of a 24 hour period. This is done because urinary levels of minerals vary based on very short term food consumption. However urinary mineral analysis does indicate when minerals are being properly absorbed and it can be especially accurate in patients with particularly consistent dietary patterns.

When we started this research, we hoped that we would be able to find that there existed very specific targeted ranges for where one wants to be on nutritional diagnostic testing. The nutrition literature in no way supports this hypothesis. There are various studies making various conclusions about different blood level indicators, such as the prospective Atherosclerosis Risk in Communities showing a 40% reduced risk of sudden cardiac death in patients with the highest quartile magnesium compared to the lowest[90]. However, analysis from the Framingham Framingham Heart Study offspring cohort did not support the hypothesis that low serum magnesium is a risk factor for hypertension or heart disease[91]. Such disparate conclusions make us conclude that the evidence mostly does not let us make confident recommendations on specific ideal nutrient ranges in blood. An exception is made for copper, where top quintile blood levels do seem to be consistently associated with negative health outcomes[92], though based on the ratio hypothesis moderately high copper levels are likely acceptable if sufficient quantities of zinc and iron are consumed. So we do not have micro-optimized ranges for nutritional testing. We instead have a U shaped dose response curves, where both very high and low levels of mineral consumption and blood levels of minerals are associated with negative health outcomes[93] Laboratory reference ranges, always bundled with nutritional tests, offer a rough guideline for the clinician to interpret, but plenty of caveats apply. We do recommend looking at nutritional tests based

on quintiles, rather than a pass fail 95% reference range. 95% reference ranges were developed based on the average American, but this fails in cases where large segments of the population are deficient, such as in magnesium[94]. The 95% reference range for magnesium would give a “pass” to two standard deviations worth of individuals actually below consensus standard for magnesium consumption[95]. Mostly, in interpreting nutritional testing, what a clinician is looking for are cases of marked deficiency or clear overdose where decisive dietary changes or supplemental changes can be recommended, rather than trying to get someone testing in the third quintile for magnesium to move up to the fifth quintile.

Typically iron deficiency is tested for by anemia testing. This was traditionally tested for with a blood test for hemacrit, the percentage of blood volume occupied by red blood cells. Today, whole blood hemoglobin is typically used as the more sensitive test for anemia[96]. Once anemia has been diagnosed, more specific measures of iron can be used. Serum iron is such a test but values can vary greatly over the course of a 24 hour period and can also be rendered inaccurate by infection and inflammatory stress. The ratio of serum iron to iron- bind capacity (serum transferrin) can also be tested for but is subject to the same flaws[97]. In the case of iron deficiency, supplements are an option but are often difficult to absorb. Dietary interventions such as increased beef muscle and organ meat consumption will increase iron while being less likely to have the systemic nutrient absorption issues common to iron supplements.

Magnesium deficiency is difficult to simply test for. Serum magnesium is the most common test specimen for magnesium measurement[98]. Low serum magnesium usually suggests a deficiency, but higher serum magnesium concentration correlates poorly with total-body magnesium stores[99]. Red Blood Cell magnesium may be a better indicator of magnesium deficiency than serum magnesium. Specifically RBC magnesium has been shown to be a better indicator than serum magnesium in diabetic children[100]. RBC magnesium correlates better with bone magnesium than it does with muscle magnesium[101]. Given normal renal function, low magnesium in urine indicates low magnesium intake, but reflects recent dietary magnesium intake rather than long term intake. A magnesium load retention test can provide a more useful indicator of magnesium from a urine sample, as originally described by Seelig[102]. Collect 24 hour sample of urinary magnesium. Infuse 2.4 mg magnesium per kilogram of lean body weight. Collect 24 hour urinary sample post infusion. Then compare the pre and post injection samples. Retention of more than 20-25% of infused magnesium indicates magnesium deficiency. Other studies have maintained that the magnesium load test is a useful indicator[103] However, it requires careful observation of a patient and is therefore a very expensive test and impractical for most clinics. Laboratory Evaluations suggests a modification of the test with oral magnesium rather than IV magnesium. Different rates of oral absorption of magnesium would present a confounding factor with this test, but additional study could result in a cheaper, easier version of the magnesium load urine test. There are also a number of functional magnesium biomarkers that can be useful tools in assessing magnesium status, such as C-reactive protein, thromboxane B2, endothelin-1, and Na/K ATPase. None, however, provide a single sensitive marker of magnesium status[104]. For now, we suggest testing for RBC magnesium and serum magnesium. Along with a food diary, this represents enough information to be able to meaningfully suggest

magnesium supplements when necessary. When a patient is believed to be low in magnesium, either via food intake and/or diagnostic testing, we suggest magnesium heavy trace mineral supplements over isolated magnesium.

The human body typically contains almost a kilogram of calcium, greater than any other element, mostly in bones. Because of this, the goal of calcium status assessment is to see if a patient is in a time-averaged positive balance for calcium so that bone loss doesn’t occur. Proper testing of body calcium balance requires measurement of multiple markers because biochemical tests don’t directly measure the amount of calcium in bone. This can however be measured with DXA scans, which is a test recommended by the US Preventative Services Task Force for patients at high risk of osteoporosis, women over 65[105]. Calcium in the blood consists of about equal amounts of ionized and calcium and albumin-bound calcium. Albumin levels in the blood are a useful additional indicator of functional calcium levels.

Serum calcium status has been shown to rise with calcium supplementation in calcium deficient populations[106], but calcium serum only consistently indicates late-stage calcium deficiency. RBC calcium testing is a useful though incomplete indicator for high RBC calcium, but not meaningful for establishing calcium deficiency. A high RBC calcium test is often followed with RBC fatty acid testing[107]. Hypercalcemia is a disorder found in 5-10% of the population where excess calcium is eliminated through urine[108]. Because of the prevalence of this disorder, urinary calcium measurement functions more as a test for hypercalcemia than as a reflection of dietary calcium intake. We recommend serum calcium and albumin level testing, and RBC testing in patients taking calcium supplements and at risk for too much calcium intake. DXA scans are definitely recommended for all women over 65, and should be used more widely when clinicians have a compelling reason to get a full picture of calcium intake, or to most accurately measure the effects of calcium supplementation or dietary interventions on bone density.

Red blood cells are a relatively sensitive specimen for indicating potassium status compared to serum potassium[109][110]. Red blood cell potassium is a reliable enough indicator to recommend for clinical testing.

Sodium blood testing is not a typical clinical nutrition test, though red blood cell testing for sodium works well[111]. Such tests are typically only done in research studies, and not as a general tool in nutritional clinics, as sodium is well absorbed and food diaries and lifestyle questions provide a sensitive enough indicator of sodium status.

Phosphorus deficiency is quite rare and testing is not usually recommended. However, phosphorus is straightforward enough to test for via serum or whole blood and testing is recommended for patients taking either calcium or phosphorus supplements as calcium and phosphorus negatively impact the absorption of each other[112]. If supplements are taken in isolation without corresponding testing, it’s possible to end up with too much body calcium and a dangerous deficiency of phosphorus, or the other way around.

Zinc probably takes the prize for the most clearly essential element that is also the hardest

to test for. A zinc testing meta-study concludes that plasma, urinary, and hair zinc are reliable markers though as before urinary zinc is only an indicator of short term status and hair testing is extremely controversial[113]. They declared that RBC zinc is not reliable, though Laboratory Evaluations states that RBC zinc is a useful indicator. Erythrocyte metallothionein has been used to test for the narrow range between mild zinc deficiency and too much zinc in the body. King initially proposed plasma + MT zinc testing in 1990[114]. There are many more zinc functional markers that can be tested for, such as mononuclear blood cells, carbonic anydrase levels, alkaline phosphatase, nucleoside phosphorylase, ribonuclease, and 5-nucleotidase[115]. However it is much more information than is useful to test for all of these zinc markers when dietary surveys go a long way to telling you about someone’s underlying zinc deficiency or sufficiency. It seems very likely that within this set of functional markers exists a complete Bayesian test for zinc deficiency and overdose. Developing such a test represents cutting-edge applied nutrition and biochemistry and interpretation and is beyond the scope of this paper. For now, we recommend testing plasma and RBC zinc, with erythrocyte metallothionein as an extra indicator when the clinician thinks it is necessary. In the case of suspected zinc deficiency, we recommend supplementing with one dozen raw oysters per week.

Copper status is most frequently tested for with serum copper or serum ceruloplasmin. Past the point of adequacies, these biomarkers don’t tend to increase much, so copper toxicity can be more difficult to establish. High serum copper concentration along with an elevated copper to ceruloplasmin ratio indicates copper excess. Urinary copper can be used to assess bodily copper status but only reflects short term intake[116]. We recommend testing for serum copper along with serum ceruloplasmin and interpreting together.

Manganese can be tested for with whole blood, RBC, urine, and other methods but as a not very popular mineral, manganese testing methods are not well studied or particularly reliable. Manganese is most frequently measured via RBC or whole blood[117].

Iodine is most frequently tested for with 24 hour urinary collection. Profiling of plasma or serum iodine is more a measure of circulating thyroid hormone than bodily iodine status[118].

Selenium has a narrow therapeutic index as a supplement, so nutritional testing is especially important when supplementation is being considered. There are many forms of selenium in the blood: 60% is found as selenoprotein P, 20% as erythrocyte glutathione peroxidase, and 20% is found bound to albumin. Among these forms, erythrocyte glutathione peroxidase normalizes more rapidly than selenoprotein, leading to the conclusion that selenoprotein P is a more reliable marker for whole body selenium status[119]. RBCs are an indicator, but may not indicate whole body status. A patient with normal RBC selenium might still have insufficient selenium to meet other selenoprotein demands[120]. Whole blood is the most widely used clinical selenium assessment[121]. We recommend whole blood selenium testing as a baseline diagnostic test, with follow-up selenoprotein P testing in those where a clinician recommends supplementation or dietary testing. Given the results discussed earlier of high selenium nutritional yeast versus isolated selenium supplementation, we strongly recommend only supplementing with approximately 200 mcg of selenium yeast and not isolated selenium when dietary interventions are not

applicable.

Molybdenum is found in the body in particular minute amounts, making direct measurement difficult. Direct blood and urine measurements of molybdenum reflect short term dietary intake of molybdenum rather than total body stores. More evaluation is needed before these metrics are considered overtly reliable[122]. Frank molybdenum deficiency is extremely rare. There is only one case of molybdenum deficiency reported in the literature; a patient with Crohn’s Disease receiving long term IV food and nutrients[123]. Because of the exceptional rareness of molybdenum deficiency and the perceived unreliability of the testing indicators, we don’t particularly recommend testing for molybdenum, though the results are worth taking a look at if bundled on a commercially available mineral diagnostic test. Be suspicious of low results though, because if eating a food-based diet, testing inaccuracy is more likely than molybdenum deficiency.

References

1 J. Mursu et al., “Dietary Supplements and Mortality Rate in Older Women: The Iowa Women’s Health Study,” Archives of Internal Medicine 171, no. 18 (October 10, 2011): 1625-­‐1633.

2 H Spencer, C Norris, and D Williams, “Inhibitory effects of zinc on magnesium balance and magnesium absorption in man,” Journal of the American College of Nutrition 13, no. 5 (October 1994): 479-­‐484.

3 Peter Greenwald et al., “Clinical trials of vitamin and mineral supplements for cancer prevention,” The American Journal of Clinical Nutrition 85, no. 1 (January 2007): 314S-­‐317S.

4 G S Omenn et al., “Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease,” The New England Journal of Medicine 334, no. 18 (May 2, 1996): 1150-­‐1155.

5 RichardB. Shekelle et al., “DIETARY VITAMIN A AND RISK OF CANCER IN THE WESTERN ELECTRIC STUDY,” The Lancet 318, no. 8257 (November 1981): 1185-­‐1190.

6 C. G. Slatore et al., “Long-­‐Term Use of Supplemental Multivitamins, Vitamin C, Vitamin E, and Folate Does Not Reduce the Risk of Lung Cancer,” American Journal of Respiratory and Critical Care Medicine 177, no. 5 (November 20, 2007): 524-­‐530.

7 Han-­‐Yao Huang et al., “Multivitamin/Mineral supplements and prevention of chronic disease: executive summary,” The American Journal of Clinical Nutrition 85, no. 1 (January 2007): 265S-­‐268S.

8 Stephanie A. Smith-­‐Warner et al., “Fruits, vegetables and lung cancer: A pooled analysis of cohort studies,” International Journal of Cancer 107, no. 6 (December 20, 2003): 1001-­‐1011.

9 Michael F Leitzmann et al., “Zinc supplement use and risk of prostate cancer,” Journal of the National Cancer Institute 95, no. 13 (July 2, 2003): 1004-­‐1007.

10 L C Clark et al., “Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin. A randomized controlled trial. Nutritional Prevention of Cancer Study Group,” JAMA: The Journal of the American Medical Association 276, no. 24 (December 25, 1996): 1957-­‐1963.

11 S. M. Lippman et al., “Effect of Selenium and Vitamin E on Risk of Prostate Cancer and Other Cancers: The Selenium and Vitamin E Cancer Prevention Trial (SELECT),” JAMA: The Journal of the American Medical Association 301, no. 1 (January 7, 2009): 39-­‐51.

12 Maria E Goossens, Frank Buntinx, and Maurice P Zeegers, “Re: Selenium and vitamin E: interesting biology and dashed hope,” Journal of the National Cancer Institute 101, no. 19 (October 7, 2009): 1363-­‐1364; author reply 1364.

13 Robert D Abbott et al., “Dietary magnesium intake and the future risk of coronary heart disease (the Honolulu Heart Program),” The American Journal of Cardiology 92, no. 6 (September 15, 2003): 665-­‐669.

14 Susanna C Larsson et al., “Calcium and dairy food intakes are inversely associated with colorectal cancer risk in the Cohort of Swedish Men,” The American Journal of Clinical Nutrition 83, no. 3 (March 2006): 667-­‐673; quiz 728-­‐ 729.

15 Mitsumasa Umesawa et al., “Dietary intake of calcium in relation to mortality from cardiovascular disease: the JACC Study,” Stroke; a Journal of Cerebral Circulation 37, no. 1 (January 2006): 20-­‐26.

16 F. L. Crowe et al., “Fruit and vegetable intake and mortality from ischaemic heart disease: results from the European Prospective Investigation into Cancer and Nutrition (EPIC)-­‐Heart study,” European Heart Journal 32, no. 10 (January 18, 2011): 1235-­‐1243.

17 S J Schoenthaler et al., “The effect of vitamin-­‐mineral supplementation on the intelligence of American schoolchildren: a randomized, double-­‐blind placebo-­‐controlled trial,” Journal of Alternative and Complementary Medicine (New York, N.Y.) 6, no. 1 (February 2000): 19-­‐29.

18 W J Blot et al., “The Linxian trials: mortality rates by vitamin-­‐mineral intervention group,” The American Journal of Clinical Nutrition 62, no. 6 Suppl (December 1995): 1424S-­‐1426S.

19 Nancy R Cook et al., “Joint effects of sodium and potassium intake on subsequent cardiovascular disease: the Trials of Hypertension Prevention follow-­‐up study,” Archives of Internal Medicine 169, no. 1 (January 12, 2009): 32-­‐ 40.



20 Q. Yang et al., “Sodium and Potassium Intake and Mortality Among US Adults: Prospective Data From the Third National Health and Nutrition Examination Survey,” Archives of Internal Medicine 171, no. 13 (July 11, 2011): 1183-­‐ 1191.

21 L M Klevay, “Coronary heart disease: the zinc/copper hypothesis,” The American Journal of Clinical Nutrition 28, no. 7 (July 1975): 764-­‐774.

22 J Kadrabová et al., “Plasma zinc, copper and copper/zinc ratio in intrinsic asthma,” Journal of Trace Elements in Medicine and Biology: Organ of the Society for Minerals and Trace Elements (GMS) 10, no. 1 (April 1996): 50-­‐53.

23 H Lai et al., “Plasma zinc, copper, copper:zinc ratio, and survival in a cohort of HIV-­‐1-­‐infected homosexual men,” Journal of Acquired Immune Deficiency Syndromes (1999) 27, no. 1 (May 1, 2001): 56-­‐62.

24 Idris Yücel et al., “Serum copper and zinc levels and copper/zinc ratio in patients with breast cancer,” Biological Trace Element Research 40, no. 1 (January 1994): 31-­‐38.

25 P Whittaker, “Iron and zinc interactions in humans,” The American Journal of Clinical Nutrition 68, no. 2 Suppl (August 1998): 442S-­‐446S.

26 N W Solomons, “Competitive interaction of iron and zinc in the diet: consequences for human nutrition,” The Journal of Nutrition 116, no. 6 (June 1986): 927-­‐935.

27 Solomons and RA Jacob Studies on the bioavailability of zinc in humans: effects of heme and nonheme iron on the absorption of zinc Am J Clin Nutr 1981 34: 4 475-­‐482, n.d.

28 “USDA Nutrient Database”, n.d., http://ndb.nal.usda.gov/.

29 J. M. Leheska et al., “Effects of conventional and grass-­‐feeding systems on the nutrient composition of beef,” Journal of Animal Science 86, no. 12 (July 18, 2008): 3575-­‐3585.

30 “Ruminant Nutrition: Effect of Production Systems on Performance, Body Compostion and Lipid and Mineral Profiles of Soft Tissue in Cattle J. E. Williams, D. G. Wagner, L. E. Walters, G. W. Horn, G. R. Waller, P. L. Sims, and J. J. Guenther J ANIM SCI October 1983 57:1020-­‐1028” (n.d.).

31 Maria Jackson et al., “Are food patterns associated with prostate cancer in Jamaican men: a preliminary report,” Infectious Agents and Cancer 4, no. Suppl 1 (2009): S5.

32 B T Ji et al., “Dietary factors and the risk of pancreatic cancer: a case-­‐control study in Shanghai China,” Cancer Epidemiology, Biomarkers & Prevention: A Publication of the American Association for Cancer Research, Cosponsored by the American Society of Preventive Oncology 4, no. 8 (December 1995): 885-­‐893.

33 M L McCullough et al., “A prospective study of diet and stomach cancer mortality in United States men and women,” Cancer Epidemiology, Biomarkers & Prevention: A Publication of the American Association for Cancer Research, Cosponsored by the American Society of Preventive Oncology 10, no. 11 (November 2001): 1201-­‐1205.

34 M Nelson, “Vitamin A, liver consumption, and risk of birth defects.,” BMJ 301, no. 6762 (November 24, 1990): 1176-­‐1176.

35 Alwyn Fernandes et al., “Dioxins (PCDD/Fs) and PCBs in offal: occurrence and dietary exposure,” Chemosphere 81, no. 4 (September 2010): 536-­‐540.

36 M Brustad et al., “Vitamin D status in a rural population of northern Norway with high fish liver consumption,” Public Health Nutrition 7, no. 6 (September 2004): 783-­‐789.

37 T Crawford and M D Crawford, “Prevalence and pathological changes of ischaemic heart-­‐disease in a hard-­‐water and in a soft-­‐water area,” Lancet 1, no. 7484 (February 4, 1967): 229-­‐232.

38 T W Anderson et al., “Letter: Ischemic heart disease, water hardness and myocardial magnesium,” Canadian Medical Association Journal 113, no. 3 (August 9, 1975): 199-­‐203.

39 Chun-­‐Yuh Yang et al., “Calcium and magnesium in drinking water and risk of death from acute myocardial infarction in Taiwan,” Environmental Research 101, no. 3 (July 2006): 407-­‐411.

40 T W Anderson, “Serum electrolytes and skeletal mineralization in hard-­‐ and soft-­‐water areas,” Canadian Medical Association Journal 107, no. 1 (July 8, 1972): 34-­‐37.

41 Lina J. Leurs et al., “Relationship between Tap Water Hardness, Magnesium, and Calcium Concentration and Mortality due to Ischemic Heart Disease or Stroke 
in the Netherlands,” Environmental Health Perspectives 118, no. 3 (October 27, 2009): 414-­‐420.

42 Silvano Monarca et al., “Review of epidemiological studies on drinking water hardness and cardiovascular diseases,” European Journal of Cardiovascular Prevention and Rehabilitation: Official Journal of the European

Society of Cardiology, Working Groups on Epidemiology & Prevention and Cardiac Rehabilitation and Exercise Physiology 13, no. 4 (August 2006): 495-­‐506.

43 Louise A Catling et al., “A systematic review of analytical observational studies investigating the association between cardiovascular disease and drinking water hardness,” Journal of Water and Health 6, no. 4 (December 2008): 433-­‐442.

44 Bonnie Stern, Risk Assessment for Environmental Health. U-­‐shaped dose-­‐response curve for risk assessment of essential trace elements: Copper as a case study. (San Francisco: John Wiley and Sons, n.d.).

45 P C Bethke and S H Jansky, “The effects of boiling and leaching on the content of potassium and other minerals in potatoes,” Journal of Food Science 73, no. 5 (June 2008): H80-­‐85.

46 Stephen D Phinney, “Ketogenic diets and physical performance,” Nutrition & Metabolism 1, no. 1 (17 2004): 2. 47 E R Monsen, “Iron nutrition and absorption: dietary factors which impact iron bioavailability,” Journal of the American Dietetic Association 88, no. 7 (July 1988): 786-­‐790.

48 R F Hurrell, M Reddy, and J D Cook, “Inhibition of non-­‐haem iron absorption in man by polyphenolic-­‐containing beverages,” The British Journal of Nutrition 81, no. 4 (April 1999): 289-­‐295.

49 J Kühnau, “The flavonoids. A class of semi-­‐essential food components: their role in human nutrition,” World Review of Nutrition and Dietetics 24 (1976): 117-­‐191.

50 M Brune, L Rossander, and L Hallberg, “Iron absorption and phenolic compounds: importance of different phenolic structures,” European Journal of Clinical Nutrition 43, no. 8 (August 1989): 547-­‐557.

51 L Hallberg, M Brune, and L Rossander, “The role of vitamin C in iron absorption,” International Journal for Vitamin and Nutrition Research. Supplement = Internationale Zeitschrift Für Vitamin-­‐ Und Ernährungsforschung. Supplement 30 (1989): 103-­‐108.

52 Malulee Tuntawiroon, “Dose-­‐dependent inhibitory effect of phenolic compounds in foods on nonheme-­‐iron absorption in men,” American Journal of Clinical Nutrition 53 (February 1991): 554-­‐557.

53 Magdalena Karamać, “Chelation of Cu(II), Zn(II), and Fe(II) by Tannin Constituents of Selected Edible Nuts,” International Journal of Molecular Sciences 10, no. 12 (December 22, 2009): 5485-­‐5497.

54 N R Meyer, M A Stuart, and C M Weaver, “Bioavailability of zinc from defatted soy flour, soy hulls and whole eggs as determined by intrinsic and extrinsic labeling techniques,” The Journal of Nutrition 113, no. 6 (June 1983): 1255-­‐1264.

55 “HIDVEGI M. ‘PHYTIC ACID CONTENT OF CEREALS AND LEGUMES AND INTERACTION WITH PROTEINS.’ PERIODICA POLYTECHNICA SER. CHEM. ENG. VOL. 46, NO. 1–2, PP. 59–64 (2002)” (n.d.).

56 Barbara F. Harland, Suzette Smikle-­‐Williams, and Donald Oberleas, “High performance liquid chromatography analysis of phytate (IP6) in selected foods,” Journal of Food Composition and Analysis 17, no. 2 (April 2004): 227-­‐ 233.

57 Ibid.

58 Brian Q Phillippy, John M Bland, and Terence J Evens, “Ion chromatography of phytate in roots and tubers,” Journal of Agricultural and Food Chemistry 51, no. 2 (January 15, 2003): 350-­‐353.

59 Ready NR. “Chapter 3. Occurrence, Distribution, Content, and Dietary Intake of Phytate” Food Phytates. CRC Press, 2001, n.d.

60 P Vohra, G A Gray, and F H Kratzer, “Phytic acid-­‐metal complexes,” Proceedings of the Society for Experimental Biology and Medicine. Society for Experimental Biology and Medicine (New York, N.Y.) 120, no. 2 (November 1965): 447-­‐449.

61 B Nävert, B Sandström, and A Cederblad, “Reduction of the phytate content of bran by leavening in bread and its effect on zinc absorption in man,” The British Journal of Nutrition 53, no. 1 (January 1985): 47-­‐53.

62 Bohn T et al. “Phytic acid added to white-­‐wheat bread inhibits fractional apparent magnesium absorption in humans” American Journal of Clinical Nutrition. 2004 79:418 –23., n.d.

63 J D Cook et al., “The influence of different cereal grains on iron absorption from infant cereal foods,” The American Journal of Clinical Nutrition 65, no. 4 (April 1997): 964-­‐969.

64 Hurrell RF, et al. "Degradation of phytic acid in cereal porridges improves iron absorption by human subjects, n.d. 65 B J Macfarlane et al., “Inhibitory effect of nuts on iron absorption,” The American Journal of Clinical Nutrition 47, no. 2 (February 1988): 270-­‐274.

66 M Layrisse et al., “New property of vitamin A and beta-­‐carotene on human iron absorption: effect on phytate and polyphenols as inhibitors of iron absorption,” Archivos Latinoamericanos De Nutrición 50, no. 3 (September 2000): 243-­‐248.

67 Puspa R. Saha, Connie M. Weaver, and April C. Mason, “Mineral Bioavailability in Rats from Intrinsically Labeled Whole Wheat Flour of Various Phytate Levels,” Journal of Agricultural and Food Chemistry 42, no. 11 (November 1994): 2531-­‐2535.

68 Weaver, C.M.,et al. Calcium absorption from foods. In: Nutritional Aspects of Osteoporosis, Burckhardt, P. and Heaney, R.P., Ed., Serono Symposia Publications from Raven Press 85:133–137. 1991.]., n.d.

69 T H Iqbal, K O Lewis, and B T Cooper, “Phytase activity in the human and rat small intestine.,” Gut 35, no. 9 (September 1, 1994): 1233-­‐1236.

70 Giuseppe Famularo et al., “Probiotic lactobacilli: an innovative tool to correct the malabsorption syndrome of vegetarians?,” Medical Hypotheses 65, no. 6 (2005): 1132-­‐1135.

71 Ready NR. “Chapter 3. Occurrence, Distribution, Content, and Dietary Intake of Phytate” Food Phytates. CRC Press, 2001, 3.

72 P Gahlawat, “Antinutritional content of developed weaning foods as affected by domestic processing,” Food Chemistry 47, no. 4 (1993): 333-­‐336.

73 Joan M. McKenzie-­‐Parnell and Neill T. Davies, “Destruction of phytic acid during home breadmaking,” Food Chemistry 22, no. 3 (January 1986): 181-­‐192.

74 Isabelle Lestienne et al., “Relative contribution of phytates, fibers, and tannins to low iron and zinc in vitro solubility in pearl millet (Pennisetum glaucum) flour and grain fractions,” Journal of Agricultural and Food Chemistry 53, no. 21 (October 19, 2005): 8342-­‐8348.

75 C Hotz, R S Gibson, and L Temple, “A home-­‐based method to reduce phytate content and increase zinc bioavailability in maize-­‐based complementary diets,” International Journal of Food Sciences and Nutrition 52, no. 2 (March 2001): 133-­‐142.

76 Saray Servi, Hazım Özkaya, and Abdullah S. Colakoglu, “Dephytinization of wheat bran by fermentation with bakers’ yeast, incubation with barley malt flour and autoclaving at different pH levels,” Journal of Cereal Science 48, no. 2 (September 2008): 471-­‐476.

77 Halina Kozlowska et al., “Natural Fermentation of Lentils: Influence of Time, Concentration and Temperature on the Kinetics of Hydrolysis of Inositol Phosphates,” Journal of the Science of Food and Agriculture 71, no. 3 (July 1996): 367-­‐375.

78 http://riskfactor.cancer.gov/diet/FFQ.English.June0304.pdf, n.d.

79 E B Rimm et al., “Reproducibility and validity of an expanded self-­‐administered semiquantitative food frequency questionnaire among male health professionals,” American Journal of Epidemiology 135, no. 10 (May 15, 1992): 1114-­‐1126; discussion 1127-­‐1136.

80 W C Willett et al., “Reproducibility and validity of a semiquantitative food frequency questionnaire,” American Journal of Epidemiology 122, no. 1 (July 1985): 51-­‐65.

81 R L Karvetti and L R Knuts, “Validity of the estimated food diary: comparison of 2-­‐day recorded and observed food and nutrient intakes,” Journal of the American Dietetic Association 92, no. 5 (May 1992): 580-­‐584.

82 Richard Lord and J. Alexander Bralley, Laboratory Evaluations for Integrative and Functional Medicine, n.d.

83 S Barrett, “Commercial hair analysis. Science or scam?,” JAMA: The Journal of the American Medical Association 254, no. 8 (August 23, 1985): 1041-­‐1045.

84 S Seidel et al., “Assessment of commercial laboratories performing hair mineral analysis,” JAMA: The Journal of the American Medical Association 285, no. 1 (January 3, 2001): 67-­‐72.

85 S J Steindel and P J Howanitz, “The uncertainty of hair analysis for trace metals,” JAMA: The Journal of the American Medical Association 285, no. 1 (January 3, 2001): 83-­‐85.

86 M Yukawa, M Suzuki-­‐Yasumoto, and S Tanaka, “The variation of trace element concentration in human hair: the trace element profile in human long hair by sectional analysis using neutron activation analysis,” The Science of the Total Environment 38 (September 1984): 41-­‐54.

87 G Drasch and G Roider, “Assessment of hair mineral analysis commercially offered in Germany,” Journal of Trace Elements in Medicine and Biology 16, no. 1 (2002): 27-­‐31.

88 Disilvestro Robert, Handbook of minerals as nutritional supplements (CRC Press, 2005).

89 R E London, “Methods for Measurement of Intracellular Magnesium:NMR and Fluorescence,” Annual Review of Physiology 53, no. 1 (October 1991): 241-­‐258.

90 James M Peacock et al., “Serum magnesium and risk of sudden cardiac death in the Atherosclerosis Risk in Communities (ARIC) Study,” American Heart Journal 160, no. 3 (September 2010): 464-­‐470.

91 Abigail May Khan et al., “Lack of association between serum magnesium and the risks of hypertension and cardiovascular disease,” American Heart Journal 160, no. 4 (October 2010): 715-­‐720.

92 Mursu et al., “Dietary Supplements and Mortality Rate in Older Women.”

93 Catherine A Mulholland and Diane J Benford, “What is known about the safety of multivitamin-­‐multimineral supplements for the generally healthy population? Theoretical basis for harm,” The American Journal of Clinical Nutrition 85, no. 1 (January 2007): 318S-­‐322S.

94 S Johnson, “The multifaceted and widespread pathology of magnesium deficiency,” Medical Hypotheses 56, no. 2 (February 2001): 163-­‐170.

95 Lord and Bralley, Laboratory Evaluations.

96 Elizabeth M. Ross, “Evaluation and Treatment of Iron Deficiency in Adults,” Nutrition in Clinical Care 5, no. 5 (September 2002): 220-­‐224.

97 Wardlaw, G.M., Hampl, J.S., and DiSilvestro, R.A., Perspectives in Nutrition, 6th ed., McGraw-­‐Hill Education, Madison, 2004, chap. 12., n.d.

98 Michael E Maguire and James A Cowan, “Magnesium chemistry and biochemistry,” Biometals: An International Journal on the Role of Metal Ions in Biology, Biochemistry, and Medicine 15, no. 3 (September 2002): 203-­‐210.

99 Lord and Bralley, Laboratory Evaluations.

100 Enver Simşek, Meltem Karabay, and Kenan Kocabay, “Assessment of magnesium status in newly diagnosed diabetic children: measurement of erythrocyte magnesium level and magnesium tolerance testing,” The Turkish Journal of Pediatrics 47, no. 2 (June 2005): 132-­‐137.

101 A C Alfrey, N L Miller, and D Butkus, “Evaluation of body magnesium stores,” The Journal of Laboratory and Clinical Medicine 84, no. 2 (August 1974): 153-­‐162.

102 Seelig MS. Magnesium Deficiency in the Pathogensis of Disease. New York: Plenum Medical Books; 1980., n.d. 103 L Gullestad et al., “The magnesium loading test: reference values in healthy subjects,” Scandinavian Journal of Clinical and Laboratory Investigation 54, no. 1 (February 1994): 23-­‐31.

104 Lord and Bralley, Laboratory Evaluations.

105 http://www.uspreventiveservicestaskforce.org/uspstf/uspsoste.htm, n.d.

106 J M Pettifor et al., “The effect of dietary calcium supplementation on serum calcium, phosphorus, and alkaline phosphatase concentrations in a rural black population,” The American Journal of Clinical Nutrition 34, no. 10 (October 1981): 2187-­‐2191.

107 Lord and Bralley, Laboratory Evaluations.

108 Giuseppe Vezzoli, Laura Soldati, and Giovanni Gambaro, “Update on primary hypercalciuria from a genetic perspective,” The Journal of Urology 179, no. 5 (May 2008): 1676-­‐1682.

109 M Bahemuka and H M Hodkinson, “Red-­‐blood-­‐cell potassium as a practical index of potassium status in elderly patients,” Age and Ageing 5, no. 1 (February 1976): 24-­‐30.

110 H S LANS, I F STEIN Jr, and K A MEYER, “The relation of serum potassium to erythrocyte potassium in normal subjects and patients with potassium deficiency,” The American Journal of the Medical Sciences 223, no. 1 (January 1952): 65-­‐74.

111 D. C. Tosteson, Ethel Shea, and R. C. Darling, “POTASSIUM AND SODIUM OF RED BLOOD CELLS IN SICKLE CELL ANEMIA 1,” Journal of Clinical Investigation 31, no. 4 (April 1, 1952): 406-­‐411.

112 Robert P Heaney and B E C Nordin, “Calcium effects on phosphorus absorption: implications for the prevention and co-­‐therapy of osteoporosis,” Journal of the American College of Nutrition 21, no. 3 (June 2002): 239-­‐244.

113 Nicola M Lowe, Katalin Fekete, and Tamás Decsi, “Methods of assessment of zinc status in humans: a systematic review,” The American Journal of Clinical Nutrition 89, no. 6 (June 2009): 2040S-­‐2051S.

114 J C King, “Assessment of zinc status,” The Journal of Nutrition 120 Suppl 11 (November 1990): 1474-­‐1479.

115 Lord and Bralley, Laboratory Evaluations.

116 Ibid.

117 Ibid.

118 Ibid.

119 Glenn A. Jacobson et al., “Selenoprotein P Analysis in Human Plasma: A Discrepancy Between HPLC Fractionation of Human Plasma with Heparin-­‐Affinity Chromatography and SDS-­‐PAGE with Immunoblot Analysis,” Biological Trace Element Research 107, no. 3 (2005): 213-­‐220.

120 K M Brown and J R Arthur, “Selenium, selenoproteins and human health: a review,” Public Health Nutrition 4, no. 2B (April 2001): 593-­‐599.

121 Lord and Bralley, Laboratory Evaluations.

122 Ibid.

123 N N Abumrad et al., “Amino acid intolerance during prolonged total parenteral nutrition reversed by molybdate therapy,” The American Journal of Clinical Nutrition 34, no. 11 (November 1981): 2551-­‐2559.