Minerals in balance: zinc, copper, selenium and why single high doses can tip the system
Minerals are not switches. They are faders on a mixing desk. Pull one of them far up and you change the whole sound. Here is where that happens in the body and what you can do with it.
A mineral is not a switch. It is a fader on a mixing desk. You cannot pull a single fader far up without changing the overall sound. Including in places you never listened to.
You are standing in front of the shelf. Zinc 50 milligrams, high dose, for skin, hair and immune system. Next to it selenium, 200 micrograms, for the thyroid. Next to that calcium, because your mother had osteoporosis. And on the far right iron, because you are often tired.
Four tubs. Four understandable reasons. And nobody at the shelf tells you that these four talk to each other.
I am not writing this text to talk you out of supplements. I prescribe minerals myself. In my consulting room I experience that quite a lot can be moved with them. That is my clinical observation and not a statement from a trial, and exactly that separation matters to me in this text. I am writing it because with minerals of all things, precision matters most and happens least.
The short version up front: minerals share transport routes, binding sites and control loops. Anyone who takes one of them at a high dose over months moves the others along. Usually quietly. Sometimes to a point where it becomes clinically visible.
What you will find here
- Why minerals run in control loops and not in separate lanes
- Typical zinc deficiency symptoms and why the blood value often misleads
- The zinc copper antagonism, the best documented example there is
- How a side effect became a therapy: zinc in Wilson disease
- Selenium and its narrow window: the U curve and the diabetes signal
- Iron, zinc and calcium: the crowd in the gut and what timing achieves
- Magnesium, calcium, potassium and the question of ratios
- Three levers for tomorrow, plus the safety limits that really count
Why a mineral is never on the road alone
Picture a mixing desk. Twelve faders, one song. You decide the bass is too quiet and pull it up. What happens? The bass gets louder. And the voice gets quieter, although you never touched it. Not because the desk is broken, but because everything runs through the same output stage.
Your mineral household behaves in exactly the same way. There is no channel that carries only zinc and nothing else. There are transporter families with overlapping preferences, there are binding proteins that accept several metals, and there are control loops that react to supply.
This is not an esoteric claim. This is absorption physiology, and it is well studied.
Two researchers at the Technical University of Berlin gathered what is known about zinc uptake in the human small intestine. Their central point: uptake in the gut is one of the most important levers with which the body regulates its overall zinc stores.
In concrete terms that means: when supply is short, the share your gut pulls out of food rises. When supply is plentiful, it falls. On top of that, components of the meal, phytate from grains and legumes above all, influence how much is available in the first place.
What this means for you: the number on the tub is not the number that arrives inside you. Between the two sits a set of rules that keeps readjusting.
For context: this review leans heavily on cell models of the gut. It describes well how the regulation can work, and less well which number comes out in you.
Maares M, Haase H. Nutrients. 2020. DOI: 10.3390/nu12030762And this is exactly where the thinking error sits that we all make at some point. We treat nutrients like refuelling. Empty, so fill it up. More is more.
The body does not work like a tank, though. It works like a thermostat. A thermostat reacts to supply by counter steering. That is why a high single dose can do two things at once: raise the mineral you wanted and shift the regulation of another one.
The question is not: how much zinc do I take? The question is: what does that amount change in the network while it raises the zinc?
That sounds more complicated than it is. In practice it only means: think about what else is running along, and look again after a few months instead of simply carrying on.
And now you know why I never talk about a single substance with minerals without mentioning the neighbours.
Zinc: the symptoms everyone googles, and the blood value that often misleads
You type in "zinc deficiency symptoms". And you read: hair loss. Brittle nails. Constant infections. Wounds that take a long time. Dry, cracked skin. Loss of appetite. Altered sense of taste.
And you think: that is me. Three out of seven at least.
I understand that well. I have looked at this list myself and recognised myself in it. Only that is exactly the trap.
None of these signs belongs to zinc alone. Hair loss can come from the thyroid, from iron deficiency, from a diet, from stress. Susceptibility to infection can be lack of sleep. Cracked skin can be winter. The symptom list is a reason to look closer. It is not the diagnosis.
Ananda Prasad helped describe human zinc deficiency in the 1960s and drew up a balance sheet decades later. He places zinc deficiency as a globally relevant problem which by his estimate affects almost two billion people in developing countries.
As consequences he names delayed growth, disturbances of gonadal function, a weakened immune response and cognitive impairment. As the main cause in many regions he describes the high phytate content of grain based diets, which can lower zinc availability.
He also describes zinc as a signalling molecule in certain immune cells and as a factor that can be linked with oxidative stress and inflammatory messengers in older people.
Prasad AS. J Am Coll Nutr. 2009. DOI: 10.1080/07315724.2009.10719780So if zinc is that important, why do we not simply measure it and know where we stand?
Because serum zinc is one of the shakiest markers in the lab. A large share of the zinc in blood hangs on albumin, the most important transport protein. When albumin drops, the zinc value drops with it. And albumin drops with inflammation, with liver disease, with malnutrition, after operations.
That leads to a pattern which comes up constantly in practice and is rarely named.
A Scottish working group retrospectively reviewed 70 case records of people who had been prescribed zinc. They looked at why the zinc had been prescribed and what happened with the lab values.
Result one: in 48 percent, the low plasma zinc value was probably a consequence of low albumin or a systemic inflammatory response and not an expression of a true deficiency. The diagnosis of zinc deficiency was therefore often a misreading of the lab value.
Result two: 62 percent received zinc doses at a level that can trigger copper deficiency. Copper was measured at all in two out of 70 people.
Duncan A et al. J Clin Pathol. 2015. DOI: 10.1136/jclinpath-2014-202837A low serum zinc value is not proof of zinc deficiency. It is a hint that has to be placed in context.
Without CRP and albumin beside it, it is hard to read. That is not a reproach aimed at laboratory medicine. It is a property of this marker that the whole field works with.
The other direction holds as well: when a deficiency really is there, it cannot always be topped up quickly.
In a randomised double blind trial, 31 nursing home residents with low serum zinc received either 30 milligrams of zinc a day or a very low comparison dose over three months.
The zinc value rose about 16 percent more in the zinc group than in the comparison group. Those who started with particularly low baseline values, however, did not reach the target range even after three months. In parallel, the number of T cells in the blood rose.
What this means for you: a deficiency is not a hole you shovel shut in two weeks. And the check after some months is not a luxury, it is the point at which you see whether anything happened at all.
Barnett JB et al. Am J Clin Nutr. 2016. DOI: 10.3945/ajcn.115.115188And now you know why with zinc I would rather measure once more than guess once more.
Zinc and copper: two that share the same door
Now comes the example that convinced me most that minerals are not a toy. It is well documented, it is understood mechanistically, and it happens quietly.
Zinc and copper compete in the gut. Not only for space, but by a detour that is almost elegant.
When a lot of zinc arrives in the gut, it prompts the intestinal cells to build more of a protein called metallothionein. Metallothionein is a metal catcher. It binds zinc, but it binds copper even more tightly. The bound copper can then stay sitting in the intestinal cell and be excreted along with it instead of reaching the blood.
Picture it like a doorman you hired yourself. You wanted more zinc to come in. The doorman accepts the zinc, but at the same time he holds on to the copper and does not let it through.
An Italian and international working group describes this route in detail, because it is the basis of an established treatment. In Wilson disease, an inherited copper storage disorder, high dose zinc is used to trigger exactly this effect.
The zinc prompts the formation of metallothionein in the intestinal cells, copper uptake from the gut is slowed as a result, and the excess of unbound copper in the body can fall. For people with this disease that is a recognised therapy.
The same principle that is wanted there can become a problem for everyone else when zinc is dosed high over months and copper is never checked.
Squitti R et al. Biomolecules. 2020. DOI: 10.3390/biom10081164Here two worlds part company that stand side by side on the shelf and are still different.
Covering a requirement means: filling a gap. Low doses, freely available, without a therapeutic claim. No medicinal effect is to be expected from them, and that is not their purpose either.
Orthomolecular or therapeutic high dosing means: a nutrient is used deliberately as a tool, in amounts far above the requirement, with a clear goal, limited in time and under laboratory monitoring. Zinc in Wilson disease is exactly that. It is a medically led treatment with follow up checks, not a shelf product.
Anyone who takes the dose from the second world and the supervision from the first gets the risk without the framework. That is the point at which supplements can tip.
How fast does that happen? Faster than most people assume. And already at doses that are freely available.
A Canadian working group gave healthy men 25 milligrams of zinc twice a day or a dummy preparation over six weeks. Plasma copper and a copper dependent blood protein did not change measurably. Another marker did: the activity of copper dependent superoxide dismutase in the red blood cells fell in the zinc group, and the difference became statistically notable after six weeks.
An American group studied adult women over ten weeks on 50 milligrams of zinc a day. Here too the same copper dependent enzyme fell. In addition, ferritin and haematocrit dropped. When iron was given alongside, the iron part could be cushioned, the copper part could not.
Both studies are old, small and without modern blinding standards. The study in women also had no placebo group, comparisons there were made against the participants' own baseline values. They show no disease picture, only a shift in markers. That is exactly why they are interesting: they show that the shift begins early, long before anyone has symptoms.
Fischer PW et al. Am J Clin Nutr. 1984. DOI: 10.1093/ajcn/40.4.743 · Yadrick MK et al. Am J Clin Nutr. 1989. DOI: 10.1093/ajcn/49.1.145And when it runs long enough, a shift in markers can turn into a disease picture. That is the part I consider the most important in this whole article.
A neurologist at the Mayo Clinic worked through the neurological consequences of acquired copper deficiency in humans. What was known before were mainly the blood findings: anaemia and a shortage of certain white blood cells.
Beyond that he describes damage to the spinal cord with a spastic gait and a pronounced disturbance of position sense. As known causes he names stomach operations, excessive zinc intake and absorption disorders. The picture resembles the spinal cord damage seen with vitamin B12 deficiency so closely that it can be mistaken for it.
The decisive sentence for us: giving copper can bring anaemia and the shortage of blood cells back promptly and may possibly prevent further neurological deterioration. Neurological damage that has already occurred, by contrast, often recedes only partly.
Kumar N. Mayo Clin Proc. 2006. DOI: 10.4065/81.10.1371That this is not a theoretical worry is shown by a case report from Canada. It describes a young adult with anaemia and a shortage of defence cells, triggered by excessive zinc intake from freely available products. The blood changes receded after stopping. The authors call it an underestimated cause of changes in the blood count.
Irving JA et al. CMAJ. 2003. PMID: 12874162
The tub of zinc on the shelf and the zinc preparation used in Wilson disease contain the same substance. What separates them is dose, intention and monitoring. Only one of those is printed on the box.
| Situation | What is documented | What follows from it |
|---|---|---|
| 2 x 25 mg zinc over 6 weeks, healthy men | Copper dependent enzyme in erythrocytes falls, plasma copper unchanged | The effect begins before standard values look abnormal |
| 50 mg zinc over 10 weeks, healthy women | The same enzyme falls, ferritin and haematocrit drop in addition | Iron status can be pulled along too |
| Zinc prescriptions in a clinical review | 62 percent at a copper critical dose, copper measured in 2 of 70 | The risk is usually not on anyone's radar |
| Acquired copper deficiency over a longer period | Anaemia, shortage of blood cells, spinal cord damage possible | Blood findings usually recede, nerve damage often only partly |
| Wilson disease, high dose zinc | Deliberate braking of copper uptake via metallothionein | The same mechanism, here used as a recognised therapy |
Higher zinc doses over a longer period belong under medical supervision. With longer use it makes sense to keep an eye on copper, the blood count, CRP and albumin alongside zinc.
If unexplained anaemia, a drop in certain white blood cells, unsteadiness in walking, odd sensations in feet or hands or unsteady standing in the dark appear during longer running zinc use, that is a reason to seek medical advice promptly. Not for panic, but for clarification.
And now you know why with zinc I always ask how long already and at what dose, before I talk about the next capsule.
Selenium: a window, not a funnel
Among the trace elements, selenium holds a special position. Not because it would be particularly important, but because the distance between too little and too much is so small.
Most nutrients behave like a funnel. Too little at the bottom, then a broad range in which everything is fine, and somewhere far up eventually too much. Selenium behaves more like a window. The good range is narrow, and both edges are uncomfortable.
Why do we need it at all? Selenium is a building block of a protein family of its own, the selenoproteins. These include enzymes that buffer oxidative stress, and the so called deiodinases, which can convert the thyroid hormone T4 into the active form T3.
A Romanian review worked through the role of the selenoproteins for the thyroid. It describes two levels: protection of thyroid tissue against oxidative stress and the conversion of thyroid hormones via selenium dependent enzymes.
The authors place the evidence carefully. An association between low selenium status and autoimmune thyroid disease is considered well supported. On the question of whether giving selenium can favourably influence the course, the results are inconsistent, with the exception of mild endocrine orbitopathy.
Their conclusion is remarkably sober: routine selenium supplementation is not recommended. Measuring the baseline value appears sensible in order to see who could benefit at all.
Valea A, Georgescu CE. Hormones (Athens). 2018. DOI: 10.1007/s42000-018-0033-5That brings us to the core. Selenium does not top up endlessly. The selenium dependent enzymes have a saturation point.
An American nutrition scientist compared selenium status markers across graded intake amounts between several animal species, among them rats, mice, lambs, chicks and turkeys.
The pattern was similar across the species. The central enzyme markers rose with increasing selenium intake and then reached a clearly delimited plateau. Above this plateau, additional selenium intake changed nothing further in the markers. The author puts it roughly like this: there was no marker that could meaningfully depict a selenium status above the requirement.
Placing it in context: these are animal data. They prove nothing for humans. They do, however, make it plausible why in humans more selenium above a certain intake could not show any additional benefit.
Sunde RA et al. Adv Nutr. 2016. DOI: 10.3945/an.116.012872When nothing more is added at the top but the substance still ends up in the body, the question of what else it does there becomes interesting. And here the data get uncomfortable.
In a randomised, placebo controlled trial in selenium poor regions of the eastern United States, 1,202 people without diabetes received 200 micrograms of selenium a day or a dummy preparation. They were observed for a mean of 7.7 years.
Recruitment took place in dermatology clinics, and the participants were mostly older with a history of skin cancer. The result is therefore not one to one transferable to everyone.
In the selenium group 58 people developed type 2 diabetes, in the placebo group 39. That corresponds to 12.6 versus 8.4 cases per 1,000 person years, with a hazard ratio of 1.55. The signal was clearest in those who already had the highest selenium levels at the start.
Important for context: diabetes was a secondary endpoint in this trial, and the diagnoses rested largely on self report. The result is a warning signal to be taken seriously, not proof of harm.
Stranges S et al. Ann Intern Med. 2007. DOI: 10.7326/0003-4819-147-4-200708210-00175And one sentence matters to me at this point. If selenium was prescribed for you by a doctor, for example for a thyroid condition, please do not stop it on your own because of this section. Take it instead as a reason to ask at your next appointment about your baseline value and the planned duration.
You could close the case there. You should not, though, because there is a counter test, and it turned out differently.
In Denmark, a country with a rather low selenium supply, 491 people aged 60 to 74 were allocated to 100, 200 or 300 micrograms of selenium a day or a dummy preparation and followed for two years.
After two years the long term blood sugar value did not differ meaningfully between the groups. The authors read this together with the older data as a U shaped relationship: at a low baseline status selenium does not appear to burden sugar metabolism, while at an already good supply additional selenium could be unfavourable.
That is the reason why I never recommend selenium across the board or reject it across the board. It depends on where you start.
Stranges S et al. Diabetes Obes Metab. 2019. DOI: 10.1111/dom.13549Margaret Rayman summed up these contradictions in the Lancet in one sentence that I consider among the wisest in the whole micronutrient literature. In essence: the relationship between selenium and health is inescapably U shaped. Those who have little can benefit from more. Those who have enough to plenty could be disadvantaged by more and should not take selenium products.
And what about the hope selenium grew up with, cancer prevention?
In the SELECT trial, 35,533 healthy men were allocated over years to selenium, vitamin E, both or dummy preparations, and 34,887 of them were analysed. The selenium dose was 200 micrograms a day as selenomethionine.
Neither selenium nor the combination lowered the risk of prostate cancer. In the vitamin E group the number of prostate carcinomas even rose, with a hazard ratio of 1.17 compared with placebo. For selenium alone the hazard ratio was 1.09 and was not statistically secured.
What this means for you: a well meant high dose over years is not a harmless bet. It can also swing the other way. Incidentally, this trial is where the figure comes from on which the European upper limit for selenium was later built.
Klein EA et al. JAMA. 2011. DOI: 10.1001/jama.2011.1437And the upper edge of the window? It is not theoretical. It exists on record, through a manufacturing error.
The American health authorities investigated an outbreak of acute selenium poisoning. The cause was a liquid food supplement that contained 200 times the declared amount of selenium.
Affected were 201 people in ten states. The estimated mean daily dose was around 41,750 micrograms, against a recommended intake of 55 micrograms. Common complaints were diarrhoea, exhaustion, hair loss, joint pain as well as discolouration and brittleness of the nails.
The course is remarkable: even after 90 days, 52 percent still reported nail changes, 35 percent exhaustion and 29 percent hair loss. The case shows two things. First, what a selenium overdose looks like. Second, that quality control for supplements does not match that for medicines.
MacFarquhar JK et al. Arch Intern Med. 2010. DOI: 10.1001/archinternmed.2009.495The European Food Safety Authority set a tolerable upper intake level of 255 micrograms of selenium per day for adults in 2023, including pregnant and breastfeeding women. Hair loss was chosen as the critical endpoint, derived from an observed adverse effect threshold of 330 micrograms a day in the SELECT trial.
The authority names two groups that can realistically exceed this limit: regular users of high dose selenium products and regular eaters of Brazil nuts. Brazil nuts are not a niche topic. A few nuts can already sit in the range of the daily intake, and their selenium content varies considerably depending on the soil they grew in.
Important for understanding this number: an upper limit marks where harm becomes more likely, not where benefit begins. The diabetes signal from the American trial appeared at 200 micrograms, below this limit. Staying under the upper limit therefore does not automatically mean that an intake is sensible or harmless for you.
And now you know why with selenium I ask about the baseline value before I think about the dose.
Iron, zinc and calcium: the crowd in the gut
Up to here it was about two pairs. Now comes the traffic. Because in the gut far more than two minerals meet, and they do it at the same bottlenecks.
Picture a revolving door that three people want to enter at once. Nobody has bad intentions. It simply does not fit.
A Chilean working group measured on human intestinal cells in culture what happens when iron, zinc and copper are offered at the same time.
Copper and zinc inhibited iron uptake. Iron inhibited copper uptake. Zinc did not inhibit copper uptake in this model. When all three metals were offered at the same time at a ratio of one to one to one, the uptake of iron or copper respectively fell by about 40 percent.
Placing it in context: this is cell culture, not a meal. Concentrations are easier to control there and higher than in a real gut. The finding explains a mechanism, it does not put a number on your everyday life.
Arredondo M et al. Biol Res. 2006. DOI: 10.4067/s0716-97602006000100011What is interesting is that this crowding does not only begin at the transport channels. It begins one layer earlier.
A Berlin working group gathered what role the mucus layer of the gut plays for the uptake of zinc, iron, copper and manganese. This layer coats the entire digestive tract, and its properties come above all from the mucins it contains.
Their assessment: the mucins are hard to do without for the uptake of these trace elements, because they keep the metals soluble in the slightly basic environment of the small intestine. And: the competition between the trace elements apparently begins at this mucus layer already, not only at the transporter.
At the same time they name the gap. How strongly this shifts actual availability in humans has not yet been sufficiently studied.
Einhorn V, Haase H, Maares M. J Trace Elem Med Biol. 2024. DOI: 10.1016/j.jtemb.2024.127459Now to calcium, because that is the most common combination in everyday life. Milk with the muesli, a cheese sandwich with an iron tablet, a calcium tablet in the evening together with everything else.
A Swedish working group measured this in 126 people. Between 40 and 600 milligrams of calcium were given with test meals.
The inhibiting effect on iron uptake was clearly dose dependent up to about 300 milligrams. At 300 to 600 milligrams of calcium, iron absorption fell by 50 to 60 percent. Even 165 milligrams of calcium as milk, cheese or a calcium salt lowered uptake in this order of magnitude, and it did so for plant based as well as animal iron.
Hallberg L et al. Am J Clin Nutr. 1991. DOI: 10.1093/ajcn/53.1.112That sounds dramatic. And here I have to brake immediately, because an honest account matters more at this point than a strong sentence.
A Danish nutrition scientist summarised the interactions between micronutrients for practice. Her key statements are nuanced and therefore useful.
First: at the amounts that occur in foods, most micronutrients use their own uptake routes and barely disturb each other. Competition arises above all in watery solution and at higher intake amounts, meaning exactly the situation a supplement creates.
Second: negative effects of iron products on markers of zinc and copper status are described, as are effects of zinc products on iron and copper status.
Third, and this is the important qualification: the inhibiting effect of calcium on iron uptake could not be confirmed in longer supplementation studies. The body appears to adapt over weeks.
Sandström B. Br J Nutr. 2001. PMID: 11509108A meal effect is not automatically a status effect. One measures an hour, the other measures half a year.
For you that means: you do not have to start organising your diet by timetable. But if you have a concrete goal, for example topping up a documented iron deficiency, then spacing things out from calcium can be a simple measure. Just make sure the overall calcium intake does not drop in the process. The goal is shifting it, not leaving it out.
How much spacing? There is research on that too. A Swedish group developed a calculation from absorption data that estimates how much iron is taken up from a meal. Factored in are phytate, polyphenols from coffee and tea, ascorbic acid, meat and fish, calcium, egg, soy protein and alcohol. When that many factors sit in one formula, it says one thing above all: iron uptake from a meal is not a fixed number, it is the result of many simultaneous influences.
Iron and coffee or tea
Polyphenols can markedly lower the uptake of plant based iron. In calculation models for iron absorption this factor is built in permanently.
In practice: put coffee and black tea between meals rather than with them when iron is being topped up.
Iron and calcium
In single meals, iron uptake fell by 50 to 60 percent at 300 to 600 milligrams of calcium. In long term studies a lasting disadvantage could not be confirmed.
In practice: separate them in time when topping up deliberately, otherwise stay relaxed.
Iron and zinc in one capsule
Both compete for the same routes in watery solution and at higher doses. In cell experiments, iron uptake fell markedly when both were offered at the same time.
In practice: take them separately at different times of day instead of using a combination product.
Iron and vitamin C
Ascorbic acid is regarded as a strong promoter of the uptake of plant based iron and is included in absorption models as a favouring factor.
In practice: peppers, lemon or parsley with the plant based iron source is the simplest version of this.
Zinc and phytate
The high phytate content of grain based diets is considered the main reason for poorer zinc availability in parts of the world.
In practice: soaking, sprouting and sourdough fermentation can lower the phytate content. That is kitchen craft, not a supplement.
Zinc and copper in a combination product
Many combination products contain copper at a ratio of roughly 10 to 1 up to 15 to 1 relative to zinc. That is a pragmatic convention, not a proven target.
In practice: with longer zinc use, what counts is less the ratio on the label than whether anyone checks the copper.
And now you know why with minerals the question "when do I take what" is sometimes more important than the question "how much".
Magnesium, calcium and potassium: the second row that often gets forgotten
Zinc, copper and selenium are the trace elements, the ones we need in milligrams or micrograms. Now to the second group, the bulk minerals. There it is about grams, and there the ratio between them is at least as interesting as the single number.
Let us start with calcium and magnesium, because both stand on almost every shelf and are rarely thought about together.
An American working group gathered what is known about the interplay of magnesium, calcium and vitamin D. Their observation: in the United States, calcium intakes rose two to two and a half times faster than magnesium intakes between 1977 and 2012. The ratio of calcium to magnesium in the diet was most recently above 3 to 1.
They discuss indications that ratios below about 1.7 and above about 2.8 could be unfavourable and that a range around 2 is possibly more favourable. As an example they cite that American studies with a high background ratio showed a benefit of more magnesium for the cardiovascular system, while Chinese studies with a low background ratio pointed the other way.
Placing it in context: this is a hypothesis from observational data, not a proven target. The authors themselves name the fact that a reliable marker for whole body magnesium status is still missing.
Rosanoff A, Dai Q, Shapses SA. Adv Nutr. 2016. DOI: 10.3945/an.115.008631If you want to remember one single thought about the bulk minerals, then this one: the absolute amount does not decide alone, the distance to the others in the same control loop matters too.
That is why a high calcium dose alongside scarce magnesium is something different from the same calcium dose alongside a good magnesium supply. The substance is the same. The context is not.
And with that to potassium. Here the tone gets more serious, and for good reason.
Potassium is first of all the opposite of a problem substance. It sits inside the cells, it keeps the excitability of nerves and muscles in balance, and a potassium rich diet is discussed explicitly in positive terms in the specialist literature.
A Dutch and Taiwanese working group worked through the physiology of potassium for kidney medicine. Their starting observation: in nephrology potassium has a bad reputation, because people with chronic kidney disease tend towards high potassium values.
At the same time, observational data show that higher potassium excretion in urine, meaning indirectly a higher potassium intake, goes together with lower blood pressure, lower cardiovascular risk and better kidney courses. The authors describe a mechanism in the kidney tubule through which potassium can help steer sodium excretion.
Their conclusion: given our sodium rich and potassium poor diet, more potassium from foods could be similarly important as less salt. That is a statement about vegetables, fruit, legumes and potatoes. It is not a statement about potassium capsules.
Wei KY et al. Clin Kidney J. 2020. DOI: 10.1093/ckj/sfaa157Why is this distinction so important? Because potassium from foods arrives slowly and in company, while potassium from a capsule arrives as a bolus. And because many people take medication that can raise the potassium level anyway.
A review on hyperkalaemia describes it as a common abnormality in chronic kidney disease, linked with an increased risk of malignant cardiac arrhythmias. Severe hyperkalaemia is considered a medical emergency.
As the first measure in prevention, that same paper explicitly names avoiding triggering factors, among them a high potassium intake through supplements and medication that can raise the potassium level. Most often named are inhibitors of the renin angiotensin aldosterone system.
In concrete terms that means: if you take ACE inhibitors, sartans, aldosterone antagonists or potassium sparing diuretics, or if your kidney function is reduced, potassium products do not belong in self medication. Potassium containing salt substitutes should also be discussed in this situation. Please clarify this medically, with monitoring of potassium and kidney values.
There is also the other direction, incidentally, and it belongs here for completeness. Certain water tablets can flush out more potassium and magnesium. That is one of the reasons why medication plays such a large role in assessing the mineral household. Not as a reproach aimed at the medication, but as part of the whole picture.
Potassium from vegetables and potassium from a capsule are chemically the same and medically not the same.
And now you know why with the bulk minerals I ask about the medication list first and only then about the product.
The four lenses: why minerals show up in so many places at once
In clinical psychoneuroimmunology we look at a symptom through four lenses. With minerals that is not a theoretical detour. It simply explains why the same elements turn up in guides on skin, thyroid, mood and heart.
Nervous system
Copper is contained in enzymes needed for the insulating layer of the nerve pathways and for the formation of certain messengers. An acquired copper deficiency can therefore show up as spinal cord damage with an unsteady gait and a disturbed position sense. A picture that looks so similar to vitamin B12 deficiency that it can be mistaken for it.
Immune system
Zinc takes part in signalling processes in immune cells. In a randomised trial in nursing home residents, the number of T cells in the blood rose under zinc supplementation. Selenoproteins are also among the enzymes that can buffer oxidative stress, and oxidative stress is a running theme of every immune response.
Metabolism
Copper and zinc sit together in the active centre of an enzyme that catches free radicals. It was exactly this enzyme that fell first in the zinc studies. With selenium, in turn, the data show a U: too little is unfavourable, and with an already good supply additional selenium is linked with a signal in the direction of sugar metabolism.
Hormone system
The conversion of the thyroid hormone T4 into the active form T3 runs via selenium dependent enzymes. That is why selenium is discussed so often in thyroid matters. The specialist literature is reserved here: the association with low selenium status is well supported, the benefit of routine supplementation is inconsistent.
These four lenses explain why a single mineral has so many appearances. Not because it would be an all rounder. Rather because a cofactor that runs along in many enzymes can make itself noticed in many places when it becomes scarce or when it is there in excess.
This is not about a lab value on a printout. It is about whether you get through the infection season in autumn without being flat on your back three times. Whether your thyroid can work calmly. Whether your gait is still steady at seventy.
Minerals are not a cosmetic topic. They are infrastructure. And infrastructure is something you only notice when it is missing.
Real food first, and why topping up is still more often justified today
At this point the impression could arise that I am advising against supplements. The opposite is the case, only with an order to it.
The first choice is always the real food from good origin and good husbandry. Not out of romance, but out of biochemistry. An oyster delivers zinc not as an isolated ion but embedded in a matrix of protein, fat and accompanying substances that co determine uptake. A Brazil nut delivers selenium together with everything else that sits inside it. A capsule cannot rebuild that matrix in the same way.
And one more sentence I mean seriously: no supplement replaces sleep, movement, sun, real food and relationships that carry you. If one of these five construction sites is open, a capsule is an answer to the wrong question.
So why is targeted topping up nevertheless more often sensible today than fifty years ago? There are four reasons for that, and I want to present them with their limits and not sell them as a scandal.
Reason 1: soils, varieties, harvest times
- A review of American food composition tables from 1950 to 1999 found lower contents for 6 of 13 nutrients examined across 43 garden crops. The median declines ranged from 6 percent for protein to 38 percent for riboflavin.Davis DR et al. J Am Coll Nutr. 2004. DOI: 10.1080/07315724.2004.10719409
- The methodological caveat belongs with it: analytical procedures have changed, and so have varieties and growing conditions. Part of the effect is presumably explained by high yield varieties in which the same amount of nutrient is spread over more mass.
- With selenium the soil influence is particularly tangible. The selenium content of a Brazil nut varies considerably depending on the growing region, which is one reason why authorities name regular Brazil nut eaters as a risk group for exceeding the limit.
Reason 2: heavily processed foods
- Ultra processed products deliver energy reliably. On micronutrient density they usually come off worse in analyses of dietary surveys.
- At the same time the composition of the diet can shift availability. The high phytate share of grain based fare is considered the main reason for poorer zinc availability in large parts of the world.
- This is not a moral argument against convenience products. It is a calculation: if a large part of the energy comes from nutrient poor sources, the rest has to be all the denser.
Reason 3: consumption through burden and stress
- Protective systems against oxidative stress work with zinc, copper, selenium and magnesium as building blocks. Where more has to be buffered, more consumption is plausible.
- A meta-analysis of ten randomised trials found lower values of a marker of oxidative cell stress under zinc supplementation and higher values for total antioxidant capacity and for glutathione. A marker for nitric oxide did not change.Mousavi SM et al. Pharmacol Res. 2020. DOI: 10.1016/j.phrs.2020.105166
- Important for context: these are laboratory parameters, not disease courses. The connection is biologically plausible and must not be read as proof of clinical benefit.
Reason 4: medicines that can move the mineral household along with them
- Certain water tablets can increase the excretion of potassium and magnesium.
- Stomach acid blockers can impair the uptake of vitamin B12 and magnesium, because stomach acid is needed to release some nutrients.
- For metformin, a prescription only medicine firmly anchored in the guidelines for type 2 diabetes, the association with lower vitamin B12 levels is well described. In a randomised placebo controlled trial over 4.3 years, the vitamin B12 level under metformin was on average 19 percent lower than under placebo. That is a reason to keep an eye on B12, and not a reason to question the medicine. For hormonal contraception, shifts in B6, B12, folate and zinc are discussed, although the data on this are inconsistent.de Jager J et al. BMJ. 2010. DOI: 10.1136/bmj.c2181 · Palmery M et al. Eur Rev Med Pharmacol Sci. 2013. PMID: 23852908
- This is not an argument against these medicines. It is an argument for looking at the mineral household together with the medication list.
- And the most important sentence in this section: please do not stop any of them on your own. Metformin, acid blockers and water tablets are on your plan for good reasons, and stopping them without medical agreement can do considerably more harm than a nutrient gap. The right way round is the opposite one: keep taking the medicine and put the affected values on the lab form.
Targeted topping up is more often justified today than it used to be, not because supplements are fashionable, but because the conditions have changed.
That is something completely different from the statement that everyone needs a combination product. It is the statement that the probability of real gaps has risen and that it is therefore worth looking instead of guessing.
And now you know why I say with the same conviction: eat properly first. And after that: look at what is still missing.
Three levers you can put into practice today
No protocol, no brands, no dosage suggestions. Just three things that lie entirely in your hands.
Lever 1: add up what you are actually taking
- Put all the tubs on the table, including the multivitamin, the protein powder, the electrolyte drink and the fortified muesli.
- Write out the milligrams or micrograms for zinc and selenium and add them up. For many people the sum is a small surprise, because the same minerals sit in several products.
- Hold the zinc sum up against the European upper limit of 25 milligrams per day for adults. That number comes from the European scientific safety assessment for zinc and means everything together, foods and supplements.
- Compare the selenium sum with the European upper limit of 255 micrograms per day for adults. Count regular Brazil nuts in.
Lever 2: set an end date before you start
- Formulate in one sentence what is supposed to change and how you would recognise it.
- Fix when the check happens. A baseline lab panel, an interim check after some weeks to months, then a deliberate decision about carrying on.
- With longer zinc use, copper belongs on the lab form as well, along with the blood count, CRP and albumin. That is the difference between supervised use and an open ended subscription.
- And please do not take from this text the conclusion that you should now buy copper on top. Copper has a narrow window of its own, too much of it can burden the liver, and with a copper storage disease it would be exactly the wrong move. Getting it measured is the step, not restocking.
Lever 3: separate instead of stacking
- Do not put iron and calcium into the same meal when you are topping up iron deliberately. Coffee and black tea rather between meals.
- Iron and zinc at different times of day instead of in one capsule, because they share the same routes at higher doses.
- Vitamin C with the plant based iron source can improve uptake and is the simplest lever on this list. It costs nothing.
- One caveat belongs with it: please only top up iron when a deficiency has actually been measured. Some people absorb and store iron too well, for example with inherited haemochromatosis. In them, extra iron can burden the liver, the heart and the pancreas.
- And one practical point: iron products belong out of the reach of children. Even small amounts can cause severe poisoning in small children. If you suspect a child has swallowed some, that is an emergency, so call the emergency number, 112 in Germany, and do not wait.
With reduced kidney function the excretion of minerals changes fundamentally. Potassium and magnesium can then accumulate. Potassium products and potassium containing salt substitutes do not belong in self medication in this situation, especially not together with ACE inhibitors, sartans, aldosterone antagonists or potassium sparing diuretics.
With liver disease, in pregnancy and while breastfeeding, in children, under blood thinning therapy and with existing long term medication, every higher mineral dose belongs discussed beforehand. Zinc can also impair the uptake of some prescription only medicines. This is described above all for tetracyclines and fluoroquinolones, two antibiotic groups, as well as for levothyroxine and bisphosphonates. It works the other way round too. In these cases spacing them out in time makes sense, and the timing belongs discussed in the consultation.
Iron adds a constellation of its own. Iron only belongs topped up when a deficiency has been measured. Some people store iron too well, for example with inherited haemochromatosis or after frequent transfusions. An unrecognised iron overload can burden the liver, the heart and the pancreas. Iron products also belong stored safely away from children, because even small amounts can cause severe poisoning in small children.
Anyone considering zinc or selenium in the therapeutic range needs a baseline lab panel, a follow up check and an exit plan. This text replaces neither a medical examination nor individual advice.
And now you know why the most honest answer to "which mineral should I take" starts with a question in return: what are you taking right now, since when, and who is looking at it?
Common questions about zinc, copper and selenium
What are typical symptoms of zinc deficiency?
What gets described above all is a higher susceptibility to infections, slower wound healing, skin changes, hair loss, brittle nails, loss of appetite and changes in taste and smell. In children, delayed growth and delayed puberty are added to the list.
The problem with this list: not a single one of these signs belongs to zinc alone. The same symptoms can appear with iron deficiency, thyroid disorders, chronic inflammation or simply too little sleep.
A pattern of symptoms is a reason to look closer, not already a diagnosis. What makes sense is the combination of history, eating habits, medication list and a lab panel with inflammation markers alongside.
How do I know whether my blood zinc value is really low?
Serum zinc is a shaky marker. A large share of the zinc in blood is bound to albumin. When albumin is low or inflammation is running, the measured zinc value drops without the body stores having to be short.
In a British review of 70 case records, the low plasma value in 48 percent of patients was probably explained by low albumin or a systemic inflammatory response and not by a true deficiency.
That is why CRP and albumin belong in the lab panel, together with questions about diet, medication and gut health. Without that context a single zinc value is hard to read.
How much zinc is too much, and when does copper start to tip?
There is no single number at which it tips, but there are study data.
In a controlled investigation in healthy men, a copper dependent enzyme in the red blood cells dropped after only six weeks on twice 25 milligrams of zinc a day. In a ten week study in women on 50 milligrams of zinc a day the same enzyme dropped as well, and ferritin fell in addition.
Both studies are old and small, and one of them had no placebo group. They show no disease picture, but an early shift in laboratory values. That is exactly why they are a reason to look, and not a table of limits.
In a clinical review, 62 percent of those treated received zinc doses at a level that can trigger copper deficiency. Copper was almost never measured.
As a rough orientation, the European safety assessment names a tolerable upper intake level of 25 milligrams of zinc per day for adults, foods and supplements counted together.
Higher zinc doses over months therefore belong under medical supervision, with an eye on copper and the blood count.
Which zinc to copper ratio is considered sensible?
There is no fixed ideal ratio written into guidelines.
In practice people often work with combination products that contain zinc and copper at a ratio of roughly 10 to 1 up to 15 to 1, because that is approximately the relation of the usual intake recommendations. That is a pragmatic convention, not a proven target.
More useful than doing arithmetic with ratios is the question of whether anyone looks at copper at all during longer zinc use. In everyday care that often does not happen, as the clinical review of 70 prescriptions showed.
Which selenium dose is still considered safe?
The European Food Safety Authority set a tolerable upper intake level of 255 micrograms of selenium per day for adults in 2023, including pregnant and breastfeeding women.
Hair loss was chosen as the critical endpoint. The limit was derived from an observed adverse effect threshold of 330 micrograms a day in a large randomised trial, with a safety factor applied.
The authority points out explicitly that regular users of high dose selenium products and regular eaters of Brazil nuts can exceed this limit. Selenium from ordinary foods, by contrast, is considered unproblematic.
Important for understanding: an upper limit marks where harm becomes more likely, not where benefit begins. Staying under this number does not automatically mean that an intake is sensible for you.
Can selenium affect blood sugar?
There is a signal, and I consider it important enough to name openly.
In a randomised trial with 1,202 people and a mean of 7.7 years of observation, more people on 200 micrograms of selenium a day developed type 2 diabetes than on placebo, with a hazard ratio of 1.55. The association was clearest in the top third of baseline selenium levels.
Important for context: diabetes was an additionally analysed endpoint in this trial, and the diagnoses rested largely on self report. That is a signal to be taken seriously, not proof of harm.
A later Danish dose study with 491 participants found no difference in long term blood sugar after two years and read the data as a U shaped relationship.
Anyone living on a selenium rich diet probably has nothing to gain from extra selenium, but possibly something to lose. With a low baseline status the calculation looks different.
And if selenium was prescribed for you by a doctor, please do not stop it on your own because of these figures, but talk about your baseline value and the planned duration at your next appointment.
Can I take iron and zinc together?
In watery solution and at higher doses, iron, zinc and copper compete for the same uptake routes.
In cell experiments on human intestinal cells, iron uptake was inhibited by copper and zinc, and by about 40 percent when all three metals were offered at the same time at a ratio of one to one to one. That is cell culture and not a meal, so the finding explains a mechanism and does not put a number on your everyday life. Reviews also describe that iron products can shift markers of zinc and copper status unfavourably and the other way round.
In practice that means spacing them out in time instead of a permanent combination in one capsule. With a documented deficiency it is worth agreeing the order and the times of day with the doctor who knows your values.
Why should calcium not go into the same meal as iron?
Because calcium can measurably push down iron uptake within the same meal.
In a classic investigation in 126 people, 300 to 600 milligrams of calcium lowered iron absorption from a test meal by 50 to 60 percent, and it did so for plant based as well as animal iron. Even 165 milligrams of calcium as milk or cheese showed this effect.
The honest reading matters here: this is a meal effect. In longer supplementation studies a lasting disadvantage for iron status could not be reliably confirmed, because the body apparently adapts.
For people with a documented iron deficiency, spacing them out in time can still be a simple measure. What matters alongside it: the overall calcium intake should not drop. This is about shifting, not about leaving out.
Are potassium products dangerous if I take blood pressure medication?
Particular caution is called for here.
ACE inhibitors, sartans, aldosterone antagonists and potassium sparing diuretics can raise the potassium level in blood. If reduced kidney function is added, the risk of hyperkalaemia, meaning a potassium value that is too high, goes up.
Severe hyperkalaemia is a medical emergency because it can trigger malignant cardiac arrhythmias. Specialist reviews name leaving out potassium containing supplements as the explicit first step in prevention.
Potassium products and potassium containing salt substitutes therefore do not belong in self medication but in medical hands, with monitoring of potassium and kidney values. Potassium rich foods are looked at separately and are judged largely positively in the specialist literature.
How long should I take minerals?
As a basic stance: as long as needed, with a goal and a review appointment, not as an open ended subscription.
Topping up is a time limited intervention. What makes sense is a baseline lab panel, an interim check after some weeks to months and a deliberate decision about whether to carry on. With zinc in particular this is not a formality, because the copper effect builds slowly.
There are well founded exceptions with permanent substitution, for example with chronic absorption disorders or after certain operations on the digestive tract. Those exceptions are exactly that: well founded, documented and medically supervised.
What I see most often in practice is the opposite: a tub that was started three years ago for a good reason, and since then nobody has looked again.
How this topic connects with the others
Minerals rarely stand alone. If you want to read on somewhere, these paths lead further.
Supplements
The guide this article belongs to
You are hereWhen supplements make sense
Control loops, homeostasis and why more is not better
Orthomolecular or covering a requirement
Why the same substance can be two completely different things
Micronutrient analysis
What different measurement methods can show and where their limits lie
Thyroid and trace elements
Selenium, zinc and iron in interplay with T4 and T3
Functional iron deficiency
When ferritin looks normal and something is still missing
Magnesium forms compared
The map instead of the ranking
Iron guide
The mineral where timing counts most
Sources
- Maares M, Haase H. A Guide to Human Zinc Absorption: General Overview and Recent Advances of In Vitro Intestinal Models. Nutrients. 2020;12(3):762. DOI: 10.3390/nu12030762 [Review]
- Prasad AS. Impact of the discovery of human zinc deficiency on health. J Am Coll Nutr. 2009;28(3):257-265. DOI: 10.1080/07315724.2009.10719780 [Review]
- Duncan A, Yacoubian C, Watson N, Morrison I. The risk of copper deficiency in patients prescribed zinc supplements. J Clin Pathol. 2015;68(9):723-725. DOI: 10.1136/jclinpath-2014-202837 [Cohort, n=70]
- Barnett JB, Dao MC, Hamer DH et al. Effect of zinc supplementation on serum zinc concentration and T cell proliferation in nursing home elderly: a randomized, double-blind, placebo-controlled trial. Am J Clin Nutr. 2016;103(3):942-951. DOI: 10.3945/ajcn.115.115188 [RCT, n=31]
- Fischer PW, Giroux A, L'Abbé MR. Effect of zinc supplementation on copper status in adult man. Am J Clin Nutr. 1984;40(4):743-746. DOI: 10.1093/ajcn/40.4.743 [RCT, 6 weeks]
- Yadrick MK, Kenney MA, Winterfeldt EA. Iron, copper, and zinc status: response to supplementation with zinc or zinc and iron in adult females. Am J Clin Nutr. 1989;49(1):145-150. DOI: 10.1093/ajcn/49.1.145 [Human study, 10 weeks, no placebo group]
- Squitti R, Pal A, Picozza M et al. Zinc Therapy in Early Alzheimer's Disease: Safety and Potential Therapeutic Efficacy. Biomolecules. 2020;10(8):1164. DOI: 10.3390/biom10081164 [Review]
- Kumar N. Copper deficiency myelopathy (human swayback). Mayo Clin Proc. 2006;81(10):1371-1384. DOI: 10.4065/81.10.1371 [Review]
- Irving JA, Mattman A, Lockitch G, Farrell K, Wadsworth LD. Element of caution: a case of reversible cytopenias associated with excessive zinc supplementation. CMAJ. 2003;169(2):129-131. PMID: 12874162 [Case]
- Rayman MP. Selenium and human health. Lancet. 2012;379(9822):1256-1268. DOI: 10.1016/S0140-6736(11)61452-9 [Review]
- Stranges S, Marshall JR, Natarajan R et al. Effects of long-term selenium supplementation on the incidence of type 2 diabetes: a randomized trial. Ann Intern Med. 2007;147(4):217-223. DOI: 10.7326/0003-4819-147-4-200708210-00175 [RCT, n=1,202]
- Stranges S, Rayman MP, Winther KH et al. Effect of selenium supplementation on changes in HbA1c: Results from a multiple-dose, randomized controlled trial. Diabetes Obes Metab. 2019;21(3):541-549. DOI: 10.1111/dom.13549 [RCT, n=491]
- Klein EA, Thompson IM, Tangen CM et al. Vitamin E and the risk of prostate cancer: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA. 2011;306(14):1549-1556. DOI: 10.1001/jama.2011.1437 [RCT, n=34,887]
- MacFarquhar JK, Broussard DL, Melstrom P et al. Acute selenium toxicity associated with a dietary supplement. Arch Intern Med. 2010;170(3):256-261. DOI: 10.1001/archinternmed.2009.495 [Case, outbreak investigation, n=201]
- EFSA Panel on Nutrition, Novel Foods and Food Allergens. Scientific opinion on the tolerable upper intake level for selenium. EFSA J. 2023;21(1):e07704. DOI: 10.2903/j.efsa.2023.7704 [Authority Document]
- Valea A, Georgescu CE. Selenoproteins in human body: focus on thyroid pathophysiology. Hormones (Athens). 2018;17(2):183-196. DOI: 10.1007/s42000-018-0033-5 [Review]
- Sunde RA, Li JL, Taylor RM. Insights for Setting of Nutrient Requirements, Gleaned by Comparison of Selenium Status Biomarkers in Turkeys and Chickens versus Rats, Mice, and Lambs. Adv Nutr. 2016;7(6):1129-1138. DOI: 10.3945/an.116.012872 [Review, animal data]
- Sandström B. Micronutrient interactions: effects on absorption and bioavailability. Br J Nutr. 2001;85 Suppl 2:S181-S185. PMID: 11509108 [Review]
- Arredondo M, Martínez R, Núñez MT, Ruz M, Olivares M. Inhibition of iron and copper uptake by iron, copper and zinc. Biol Res. 2006;39(1):95-102. DOI: 10.4067/s0716-97602006000100011 [In vitro, human cells]
- Einhorn V, Haase H, Maares M. Interaction and competition for intestinal absorption by zinc, iron, copper, and manganese at the intestinal mucus layer. J Trace Elem Med Biol. 2024;84:127459. DOI: 10.1016/j.jtemb.2024.127459 [Mechanism Review]
- Hallberg L, Brune M, Erlandsson M, Sandberg AS, Rossander-Hultén L. Calcium: effect of different amounts on nonheme- and heme-iron absorption in humans. Am J Clin Nutr. 1991;53(1):112-119. DOI: 10.1093/ajcn/53.1.112 [Cohort, n=126]
- Hallberg L, Hulthén L. Prediction of dietary iron absorption: an algorithm for calculating absorption and bioavailability of dietary iron. Am J Clin Nutr. 2000;71(5):1147-1160. DOI: 10.1093/ajcn/71.5.1147 [Review]
- Rosanoff A, Dai Q, Shapses SA. Essential Nutrient Interactions: Does Low or Suboptimal Magnesium Status Interact with Vitamin D and/or Calcium Status? Adv Nutr. 2016;7(1):25-43. DOI: 10.3945/an.115.008631 [Review]
- Wei KY, Gritter M, Vogt L, de Borst MH, Rotmans JI, Hoorn EJ. Dietary potassium and the kidney: lifesaving physiology. Clin Kidney J. 2020;13(6):952-968. DOI: 10.1093/ckj/sfaa157 [Review]
- Kovesdy CP. Updates in hyperkalemia: Outcomes and therapeutic strategies. Rev Endocr Metab Disord. 2017;18(1):41-47. DOI: 10.1007/s11154-016-9384-x [Review]
- Mousavi SM, Hajishafiee M, Clark CCT et al. Clinical effectiveness of zinc supplementation on the biomarkers of oxidative stress: A systematic review and meta-analysis of randomized controlled trials. Pharmacol Res. 2020;161:105166. DOI: 10.1016/j.phrs.2020.105166 [Meta-analysis, k=10]
- Davis DR, Epp MD, Riordan HD. Changes in USDA food composition data for 43 garden crops, 1950 to 1999. J Am Coll Nutr. 2004;23(6):669-682. DOI: 10.1080/07315724.2004.10719409 [Comparative Study, database analysis]
- de Jager J, Kooy A, Lehert P et al. Long term treatment with metformin in patients with type 2 diabetes and risk of vitamin B-12 deficiency: randomised placebo controlled trial. BMJ. 2010;340:c2181. DOI: 10.1136/bmj.c2181 [RCT, n=390]
- Palmery M, Saraceno A, Vaiarelli A, Carlomagno G. Oral contraceptives and changes in nutritional requirements. Eur Rev Med Pharmacol Sci. 2013;17(13):1804-1813. PMID: 23852908 [Review]