I 3rd-party tested the UK's top vitamin D3 K2 supplements. Here are the results.

One in four UK adults takes a vitamin D supplement. When I emailed the top-selling brands, most of them couldn't show me a third-party test for the batch on the shelf. So I bought seven of them, sent them to an accredited lab with the seals still on, and then sent a second bottle of each to a second, more precise lab. Six passed on dose, one didn't, and one was on the edge. Every number is below, plus what each brand said when I asked, and a Queen Mary professor on whether 4,000 IU a day is safe.

Why vitamin D

One in four adults in the UK takes a vitamin D supplement. It's the one the NHS actually tells everyone to take through autumn and winter. So many of you asked for it in the comments on my last testing video, particularly Nutrition Geeks' D3 K2, which I didn't test then. And when I emailed the top brands on Amazon UK, a lot of them could not provide a third-party test report for their supplements. So here we are.

Seven products, all D3 with K2, all labelled 4,000 IU: Nutrition Geeks, WeightWorld, EVO Nutrition, Howard James, Oxford Vitality, Holland & Barrett, and my own brand, Natural Foundation Supplements.

The brands, and what they said when I asked

Before testing anything, I asked each brand, as a customer, for a third-party test of the batch I'd bought. Here's what came back.

WeightWorld

Started in the UK in 2005, now owned by an Indian corporate group with a portfolio of brands. They replied that they do third-party testing but were out of stock, and would test when the next batch started. I told them I'd placed an order an hour earlier and asked for the report for the batch that was clearly in stock on Amazon. They replied saying the report for my batch was attached. It wasn't. I asked again, and they eventually sent a heavy-metals test from 2024, which tells you nothing about dosage.

Nutrition Geeks

If you haven't seen my investigation into their other products, it's linked. I asked them for a third-party test of their vitamin D, and they sent me their public response to that video. No test result for vitamin D.

EVO Nutrition

I'd never heard of them, but they sell more than 5,000 bottles of D3 K2 a month on Amazon UK. There isn't much information out there beyond the fact that they're GMP certified (good manufacturing practice, a manufacturing-process accreditation, which is a big green flag). I asked for their third-party testing and didn't get a reply.

Howard James

Also new to me, selling around 1,000 units of D3 K2 a month on Amazon UK. There's no website, just an Amazon store, so you can't find out anything about them. On Companies House the registered address is residential, and so is the address on the bottle. As far as I could find, there's also no food business registered for the company, which you're legally supposed to do so that safety, hygiene and marketing standards can be checked. Because there's no website, I couldn't contact them to ask for a test.

Oxford Vitality

A well-established UK brand; you can tell from their Trustpilot how long they've been around. They're currently building their own testing lab in their facility, and they provided a third-party test for active ingredients, which they use to calculate the weights for their tablets.

Holland & Barrett

The UK's largest supplement retailer, which needs no introduction. I asked for their third-party testing and they said they don't give out third-party testing.

Natural Foundation Supplements

My brand. We third-party test every batch, and there's a QR code on the bottle you can scan to see the report for that specific batch.

How they were tested

All seven were bought at retail, the way you'd buy them. Each shipment to the lab went with paperwork explaining what it was for, and a custom note asking the lab to confirm that the manufacturer's seal was still intact when it arrived. That matters, because it's how you know I haven't bought the products and swapped the tablets out.

Each supplement was tested for vitamin D3 content and for heavy metals (arsenic, cadmium, lead, mercury) using validated methods at a UKAS-accredited, ISO 17025-accredited laboratory. A second bottle of each then went to a second accredited lab with a tighter measurement uncertainty, which I'll explain now, because it matters for how you read the numbers.

The 10% problem

Every result here carries a measurement uncertainty. The first lab's established method for vitamin D3 has an uncertainty of about 10%. That's the standard for vitamin D testing, unfortunately: 4,000 IU sounds like a lot, but in weight it's about 100 micrograms, a minuscule fraction of the tablet. Finding and measuring something at that level is genuinely difficult chemistry, and the smaller the amount, the wider the margin for error.

Ten percent was the smallest uncertainty I could find when I ran the first tests in July. In September I found a second lab that could do about 5%, so I sent a fresh bottle of each product there too. Two labs, two batches in most cases, so we can see how that changes the picture. Bear the 10% in mind as you read the first set.

Results: lab one

Vitamin D3 per tablet, against a 4,000 IU label claim, at the first lab:

EVO Nutrition passed. The label claims 4,000 IU and the product contained 4,079. Bang on. Well done to them; heavy metals were well within UK limits too.

WeightWorld are a big Amazon seller, so I was surprised when their D3 K2 came back at 3,508 IU, 12.3% under. Not a huge amount, but not ideal either, and just outside what the 10% uncertainty could explain. Heavy metals were fine.

Oxford Vitality passed: 4,058 IU, basically perfect, and heavy metals miles inside the limits.

Holland & Barrett, the UK's biggest supplement retailer by far, contained 4,553 IU, slightly over. That's good to see, because vitamins in supplements degrade over time and storage conditions affect how much. More on that below.

Howard James contained only 2,822 IU instead of 4,000. That's 29% under, far outside any measurement uncertainty. This is a top seller on Amazon that, as far as I could find, is registered to a residential address with no food business registration.

Nutrition Geeks came back at 3,870 IU. That's 3% under the label, which is inside the method's uncertainty: it could be off by 10% either way, so I treat this as a pass.

Natural Foundation, my brand, tested at 4,647 IU. We add a deliberate overage because vitamins degrade over time, especially in heat, so that at the end of shelf life you're still getting the dose on the label. We also test our D3 for K2 content now; that's on our testing database.

Results: lab two

The second lab has a measurement uncertainty of about 5%, so it's considerably more precise. One caveat: apart from Howard James, the bottles I sent had different batch numbers from the first round, so this is partly a test of batch-to-batch consistency as well as of the labs.

WeightWorld actually performed worse at the second lab: 3,175 IU, about 21% under. Howard James was the same batch at both labs and gave almost exactly the same result, 2,847 IU, still 29% under the 4,000 on the label. When two labs with different methods land within 25 IU of each other, that isn't uncertainty; that's the product.

One mistake on my side: for Oxford Vitality we accidentally sent their 1,000 IU D3 to the second lab rather than the 4,000 IU product, which is why that number is lower. It still passed, with a slight overage. Everyone else came back relatively similar to their first result, and EVO and Holland & Barrett again showed a healthy overage, which I'll come back to.

Heavy metals

All seven products were within UK limits for heavy metals, and by a long way. Every detectable lead and cadmium result was between roughly 1% and 4.5% of its legal limit; the single highest reading in the entire set was under 5% of the UK limit. Arsenic has no UK limit set for supplements, so it's reported but not judged. These limits are set per supplement, not against your total daily intake of heavy metals.

Why some came in over

Holland & Barrett, EVO and my own brand all came in above 4,000 IU. That's not sloppiness; it's deliberate, and here's why.

Vitamins degrade over time, especially in heat, and nobody's entirely sure by how much under real storage conditions. Responsible manufacturers add a small overage so that at the end of the shelf life you're still getting the dose on the label. We're running an experiment on our own batch right now, retesting it at three, six and nine months at room temperature, because we think the loss is 10 to 15% but we want to know for sure.

It's also genuinely hard to hit exactly 4,000 IU. The amount of D3 active that gives you that dose is tiny; the powder it comes in mostly starts life as sheep's wool, fun fact. When that powder goes into the mixer with everything else in the tablet, the mix has to be perfectly even, what's called a uniform mix. Even if a manufacturer weighs the right amount of D3, if it isn't mixed perfectly, one tablet can be under and the next over. We've spent a long time perfecting that process by trial and error, so that we can third-party test every batch, publish the result, and be confident it will pass. (And if you think the "extra" ingredients that make that mixing possible are the devil, I made a whole video on why that's a load of BS.)

Is 4,000 IU safe?

Every product here is dosed at 4,000 IU, or is supposed to be. That's the amount the NHS gives as the upper limit: "Do not take more than 100 micrograms (4,000 IU) of vitamin D a day as it could be harmful." Yet on Amazon UK you can buy 10,000 IU products, and some brands sell 50,000 IU.

To understand whether that's a good idea, I went to Queen Mary University of London to speak to Prof. Adrian Martineau, who has researched vitamin D for over twenty years. The full interview is here; the parts that matter for this test:

"One issue is around safety. People might say that you can take as much as you like and it won't harm you, and that's not right. If you take more than 4,000 units a day as an adult on a regular basis, then you could be at risk of having too much vitamin D, and that can push up your blood calcium, which can have adverse consequences, particularly for your renal function." Prof. Adrian Martineau · Clinical Professor of Respiratory Infection and Immunity, Queen Mary University of London

"The thing about vitamin D that distinguishes it from vitamin C is that it's a fat-soluble vitamin. With vitamin C, if you took a big dose, you would basically wee it out within 24 hours. With vitamin D, it's stored in the body. The lowest level at which toxicity has been reported is around about 220 nanomoles per litre." Prof. Adrian Martineau · on why the ceiling exists

His rules of thumb for an adult with little sun or dietary vitamin D: 400 IU a day gets your blood level to about 25 nmol/L, 1,000 IU to about 50, 2,000 IU to about 75. I hadn't asked him about 4,000 IU specifically, so I followed up: his estimate is that 4,000 IU takes you to around 100 nmol/L, which is under half the lowest level at which toxicity has been reported. A few of the products here measured around 5,000 IU after their overage, which by the same rules lands at roughly 110 nmol/L.

Remember what he said about genes: some of us absorb it two or three times better than others, and heavier people need more for the same level. That's exactly why vitamin D dosing is controversial and why health authorities struggle to give one number for everyone. The NHS 400 IU is a low dose, set purely to prevent deficiency disease like rickets at 25 nmol/L, and to make sure nobody goes too high. Plenty of people think that's too low; in the summer, if you're getting a decent amount of sun, most people don't need a supplement at all, with all the caveats about latitude, time of day, skin exposed and skin colour that he lays out in the interview.

The high-dose marketing game

UK brands do sell D3 K2 at 10,000 IU, over double what the NHS calls the safe limit. There's very strong evidence for the benefits of taking your vitamin D from deficient to somewhere in the normal range. There is not much evidence at all for pushing higher than that, despite what some people on the internet would have you believe.

The supplement industry is saturated. Thousands of brands are competing, so they're constantly looking for a narrative to stand out, and high-dose D3 is a perfect example: most people assume more is better. Is a high dose of D3 going to cure depression or help you lose weight, as some brands claim? No.

What I do

I take 4,000 IU most days from about October through March, and I get blood work done now and again, checking vitamin D along with everything else, to make sure my markers are in line. Vitamin D is very simple. It doesn't need to be overcomplicated.

Whether you buy from my brand or from one of the others that passed here, the best thing you can do is ask for a third-party test of the batch number in your hand. Two of seven brands could show me one. That's the number I'd most like to change.

Documents and sources

1. VideoI 3rd Party Tested The Top Vitamin D3 Supplements - the video this post is written from, with the lab certificates on screen
2. InterviewProf. Adrian Martineau on vitamin D, the full interview - levels, doses, toxicity, sun, genetics, K2
3. GuidanceNHS: Vitamin D - the 10µg (400 IU) advice and the 100µg (4,000 IU) upper limit
4. GuidanceUK Government: vitamin D for vulnerable groups - autumn and winter supplementation advice
5. InvestigationThe Nutrition Geeks investigation - my earlier tests of their other products
6. ExplainerAre supplement fillers toxic? - why the mixing ingredients matter for dose accuracy
7. DatabaseNF Supplements batch-testing database - my own brand's reports, searchable by batch number, including this D3 K2 batch's dose and heavy metals

The lab certificates for every result above (both labs, all seven products) are shown in the video and compiled in the free PDF below; the original PDFs are available on request via the contact page.