If you haven’t already read my post about Osmosis Skincare’s Drinkable Sunscreen, you might want to do that before you read this one. A brief overview is that Osmosis Skincare claims drinking their “harmonized water” prevents sunburn. I complained to the Advertising Standards Authority that this claim wasn’t backed up by evidence and the complaint was upheld, now those products have been removed from their New Zealand website as a result.

In their response to this complaint (which you can read via the ASA’s website), Osmosis Skincare said that (emphasis mine):

This is a new type of technology being used in this way and Head office can reference the internal research they did showing the product to be effective, but their independent clinical trial isn’t until the 28th of June, whereby they will put 30 people outside for one hour in San Diego, CA at noon supervised by a plastic surgeon.

Osmosis Skincare NZ – Response to ASA complaint 14/287

When reading this, I noticed that the study design seemed to be lacking a control group (which prohibits randomisation and blinding) and that the sample size seemed very small. These are all properties of low-quality science, but with only this brief summary to go by I couldn’t draw in any firm conclusions. In any case, as they were making therapeutic claims without any evidence to back them up, the complaint was upheld.

---

A few days ago, the head office of Osmosis Skincare issued a press release regarding this “independent clinical trial”, which has now been completed. I have to say I wasn’t particularly surprised to read that the trial was actually not independent. The press release introduces it by saying:

Osmosis Pür Medical Skincare executed the [Harmonized Water UV Neutralizer] line’s first clinical trial on June 28, 2014.

Osmosis Skincare – First Drinkable Sunscreen Releases Clinical Trial Results

The press release also claimed that the trial was randomised (emphasis mine):

The randomized clinical trial was designed to evaluate a new technology…

Osmosis Skincare – First Drinkable Sunscreen Releases Clinical Trial Results

When a clinical trial is randomised, that means that participants are allocated into different groups in a way that is determined randomly. These different groups typically consist of an experimental group, which receives the treatment being tested, and a control group, which receives either a placebo or sham version of the treatment or the standard of care against which the experimental treatment is being compared. Randomly allocating participants into these groups helps avoid any systematic differences between the groups that could be a source of bias in the results.

Of course, this necessitates multiple groups for participants to be included in. With a sample size of 30, this means each group would only contain around 15 people. However, the press release isn’t done with its surprises yet:

24 patients ranging from 18 to 60 with various ethnic backgrounds and skin types were exposed to one hour of sun to one side of the body between noon and 1pm after ingesting 3ml Osmosis Harmonized Water UV Neutralizer

Osmosis Skincare – First Drinkable Sunscreen Releases Clinical Trial Results

The press release doesn’t mention 6 other participants in a control group, so it’s not clear yet if there was another (smaller) group that just isn’t mentioned here or if, for some reason, they went with 24 participants instead of the 30 they’d planned on using earlier.

The “Summary of Results” says that:

All 24 patients were evaluated before, and immediately after the exposure as well as the following morning. There was no evidence of sunburn on 16 patients, 5 had minor or partial sunburns and 3 had notable sunburns in the study.

…

This proves UV Neutralizer effectively limited the sun damage for a majority of the users that consumed it.

Hang on a minute, “proves”? I’m not sure how on Earth they’d expect to be able to honestly evaluate the effectiveness of their product without first establishing some baseline to use as a comparison. It reads as though they’ve assumed that, if their product didn’t work, then all participants would have received notable sunburns. If they want to gather evidence regarding the effectiveness of their product in an intellectually honest way, they’d need to either compare it to a placebo or to the standard treatment, so they can actually see if it cause different results than nothing or if it’s as good as the real deal.

Instead, though, we get quotes like this:

The definitive results from this trial prove that the scalar wave technology in Harmonized Water works.

Dr. Ben Johnson, Founder and CEO of Osmosis Pür Medical Skin Care – First Drinkable Sunscreen Releases Clinical Trial Results

The press release then provides links to the clinical trial and photos of participants.

If you click on those links, you’ll probably notice a couple of things right away. First, instead of being published in a peer-reviewed scientific journal, the link to the clinical trial takes you to a folder on “box.com”, a cloud storage website. Second, the PDF containing photos of participants only contains 16 participants, not the 24 we’re told participated in the trial. I’ve no idea why that is the case, or how they determined which 8 participants to exclude.

The paper, entitled “Evaluation of a Novel Form of Sun Protection”, seems laid out pretty much as one would expect from a real clinical trial, published in a peer-reviewed scientific journal. Its first author is Paul Ver Hoeve, the doctor that supervised the experiment, and the second author is Ben Johnson, the founder and CEO of Osmosis Skincare (although that conflict of interest isn’t stated in the paper itself). I’m really wondering why this was ever described as “independent”. Hopefully this was just a piece of honest confusion on the part of whoever was liaising with the ASA on behalf of Osmosis Skincare NZ.

Unsurprisingly, although the authors included 8 distinct references (out of 11 in total) for the claim that topical sunscreens can cause inflammation, this claim went unreferenced:

Upon ingestion of 2-3 ml of the [harmonized] water, the scalar waves reportedly work their way through the molecules of water in the body until they reach the water in the dermis. This process has been shown to take an hour on an empty stomach, 90 minutes if any food is present in the stomach.

Paul Ver Hoeve, Ben Johnson – Evaluation of a Novel Form of Sun Protection

The “Subjects and Methods” section starts with a very concerning sentence. As far as I can tell, this is the entire basis for describing the study as “randomized”:

In this study, 24 patients were randomly selected as test subjects with no consideration for their natural skin tone.

Paul Ver Hoeve, Ben Johnson – Evaluation of a Novel Form of Sun Protection

As I mentioned earlier, when a clinical trial is described as “randomised” that means participants are randomly allocated to different treatment groups. It absolutely does not mean “patients were randomly selected as test subjects”. I’m not sure what that even means. It sounds like it’s referring to their recruitment method, but no more detail is given so I can’t really tell what this is supposed to say about the study design.

In my opinion, using this as justification for describing this experiment as a “randomized clinical trial” is very misleading.

The report goes on to state that:

The decision was made to not do a double-blind test for this application because of the ethical implications of knowingly causing a sunburn in many people.

Paul Ver Hoeve, Ben Johnson – Evaluation of a Novel Form of Sun Protection

Despite this apparent ethical concern, I can see no indication of the trial being approved by an Institutional Review Board (IRB). I believe IRB approval is required of all human subject research in the USA that is publicly funded, but privately funded research like this doesn’t have the same requirement. I’m not sure how it would affect a trial’s chances of being published in a peer-reviewed journal, perhaps someone who knows more could weigh in via the comments.

Performing “a double-blind test” in this case would require giving some participants water then leaving them to burn in the sun for an hour, and they’re right to say that’s pretty obviously unethical. I think the fact that they realise this but it’s also what they did in their trial should have been a pretty big red flag that they could use some ethical oversight.

They also don’t mention what seems to me is obviously a more ethical and rigorous way to conduct the trial. Even if they didn’t manage blind the researchers, it would have been better for everyone if the experimental group were compared to a control group that had applied topical sunscreen instead. This would allow them to have a “randomized clinical trial” that is actually randomised, as well as giving them a baseline to compare their results to so they might have a chance to draw some useful conclusions.

The actual results of the experiment were that:

There was no evidence of a sunburn on 16 patients, 5 had minor or partial sunburns and 3 had significant sunburns in the study.

Paul Ver Hoeve, Ben Johnson – Evaluation of a Novel Form of Sun Protection

Now I don’t know about you, but 1/3 of participants getting sunburned doesn’t exactly sound like a rousing success to me. They went on to try to justify these failures by saying:

All of the patients who burned said they would not normally lay [sic] out in the sun for one hour. Many of them said they burn with the use of other sunscreens as well.

Paul Ver Hoeve, Ben Johnson – Evaluation of a Novel Form of Sun Protection

The report gives no indication that the other 16 participants were asked the same questions, so there’s no way of telling if this could have contributed to the results, let alone accounted for them. As far as I know, this hasn’t prompted Osmosis Skincare to put a warning label on their products that it’s not effective for people that don’t usually expose themselves to the sun very much. It also doesn’t stop the authors from putting the results down to these answers entirely:

While the results were not 100%, the authors believe this was due solely to the excessive amount of sun they received to their relatively virgin skin and their overall health.

Paul Ver Hoeve, Ben Johnson – Evaluation of a Novel Form of Sun Protection

---

This trial had a tiny sample size, and was uncontrolled (therefore also non-randomised and unblinded), as well as being industry-funded and co-authored by the founder of the company that makes and sells the product. And on top of all that it didn’t even seem to have had particularly positive results.

If there’s any conclusion that can be drawn from this study, it’s that Osmosis Skincare is willing to do bad science and use its mediocre results to promote their products. Considering they were already making the same therapeutic claims before this experiment, I can’t say I find that surprising.

If Osmosis Skincare NZ stands by this research, and considers it rigorous enough to justify the sort of claims they were previously making about these products without substantiation, then they should appeal the ASA’s decision to uphold a complaint against them. I’d certainly like to see them try.

If you see this product being promoted or sold in New Zealand, please contact the Society for Science Based Healthcare with the details.