Blog News

Does coconut oil help damaged hair?

We are not convinced!

Coconut oil has recently received lots of attention from the media, and from beauty bloggers in particular. They have claimed that it is a healthier alternative to the more common seed oils for cooking as well as being a miraculous remedy for skin and hair.

Common praises of coconut oil found on the internet are:

– Research has proven that it is the best oil

– Protects against heat damage

– It is the only oil that helps retain protein in hair

– It binds hair proteins and holds them together, resulting in stronger, shinier, and healthier hair

– It repairs split ends faster

– Remains inside the hair and retains moisture instead of evaporating

– It constantly repairs damage and breakage

But does coconut oil help damaged hair for real? We decided to find out!

Most of the above mentioned claims come from the misinterpretation of a single article,“Effect of mineral oil, sunflower oil, and coconut oil on prevention of hair damage”. Bloggers should be very careful when interpreting scientific articles, as it is very easy to draw the wrong conclusions or to generalise findings (do you want to know why? Read this post). We decided to not do the same mistakes and ask a group of scientists to comment on the coconut article. In bold is shown what can be learned after reading the article, followed by the comments from the scientists on the matter.

The effect of mineral, sunflower, and coconut oils on hair was analysed.

“This is a peculiar choice of oils to compare. Instead of analysing the trendier jojoba, argan, olive, almond, or castor oil, the article assesses one oil that is commonly used in cosmetics (coconut) and two that are less frequently used in DIY remedies. It’s like studying whether children would prefer to eat ice cream or cabbage. The conclusion is obvious even before conducting the experiment!”

The oils were tested on cut hair that was A) untreated; B) boiled for 120 minutes; C) bleached for 120 minutes; or D) exposed to simulated sunlight at a temperature of 50°C for 300 hours.

“The procedure for treating hair during this study was very different from the situation that hair is exposed to in the real world.  Nobody would boil their hair (for 2 hours or otherwise!), and the usual time for bleaching is between 20 and 50 minutes, not 120.”

“Why would they boil hair? They claim to simulate taking hot baths, but every biochemist knows that there is a very big difference between exposing proteins to hot water as opposed to boiling it!”

“Were they trying to simulate post-apocalyptic hair scenarios? They should have gone all the way and stuck some hair into a nuclear reactor :)”

Hair samples were treated as described above. The oils were applied to the hair samples either before the treatment, after the treatment, or not applied at all*. Finally, the hair protein loss was analysed in all hair samples.

*Hair samples that were not treated with oil are called controls and are needed to show that the effect we see in the experiment comes from the experimental conditions (oil application) and not because of some other factor.

“Reading the text, it becomes clear that the authors weren’t measuring the hair’s overall proteins loss. Instead they monitored only the specific quantity of protein lost when the hair was combed at the end of the process.

This can be compared to carrying out the following experiment: During autumn, point super-powered leaf blowers directly towards some trees for 2 hours. Then coat the bark of some of the trees with coconut oil and slightly shake the trunk. Now count the number of leaves that flutter down while the tree is being shaken and compare this number across the different trees. The harshest treatment was the exposure to the leaf blower, but the leaves lost during this step were ignored completely!”

“I believe that another method of assessing protein loss should have been used.”

This graph shows protein loss in hair, depending on whether oils were applied before or after the treatment, or not applied at all. The second and fifth couple of bars show protein loss in hair treated with mineral oil and their respective controls.

“Why is there such great variation in protein loss across the hair to which no oil was applied (all the dark bars)? It varies between 93 µg/g and 227 µg/g for the control hair, which is very close to the difference between hair that hasn’t been coated in coconut oil and hair that has (between 70 µg/g and 227 µg/g)! This shows that other factors affect hair protein loss more than oils.”

“Some of the untreated (not boiled, bleached, etc.) hair samples show a protein loss of 92 µg/g, and others show almost the same loss (93 µg/g) after 2 hours of hair bleaching! Are the authors really claiming that after 2 hours in hydrogen peroxide and ammonia solution there is no increase in protein loss? I believe that the method of assessing protein loss is at fault here. Going back to the tree analogy: If a super powerful leaf blower was pointed towards a tree for 2 hours, by the end of that time the tree would have lost most of its leaves. Therefore, if you only start counting leaf loss while shaking the trunk, you will see almost no leaves falling. But this is because there are hardly any leaves left on the tree! In the case of this experiment, if almost no protein is present on the hair, protein loss will be minimal.”

What conclusions can be drawn?

“If you are planning to boil your hair for 2 hours, putting some coconut oil on it will probably reduce protein loss :)”

“Mineral and sunflower oil do not seem to help prevent hair scales (cuticle) from being lost during brushing.”

“There are some hints that coconut oil may help prevent cuticle being lost from hair, and that it may be best to apply it before treating the hair rather than after. However, there is a too wide variation between the untreated hair (control) samples to be able to draw any real conclusions.”

“If you plan to treat your hair with coconut oil, but you don’t go ahead with it, you will have more protein loss than if you were planning to apply mineral or sunflower oil. Therefore, if you think about applying coconut oil, you better go ahead –  if you don’t then the thought of coconut oil will ruin your hair!”

In short, our scientists have the following concerns:

  • They question the choice of oils to be compared in this experiment. Although coconut oil is frequently used in home beauty treatments, mineral and sunflower oils aren’t.
  • Next our experts point out that the significant range of protein loss (from 70 to 227 µg/g) shown by hair that was not treated with oil – in other words, hair that should have been a control sample – make the experiment essentially useless. After all, if regular hair can vary so much, how can we draw any conclusions from a similar variation from treated hair?
  • Thirdly, the methodology for measuring protein loss is questionable. If only the protein lost at the final stage is measured then the stripping of vast amounts of protein before measurement begins – as might happen, for example, if hair is boiled for two hours or soaked in a bleaching solution – will significantly skew the results.
  • Finally, the assignment of the controls is bizarre. How does hair that hasn’t been treated with coconut oil differ from hair that hasn’t been treated by mineral oil, for example? Surely it’s just hair! These huge differences in results between them make all the results invalid.

Our Conclusions:

As you can see, the scientists could not take this study very seriously, and there are certainly many problems with it, ranging from a discombobulated graph to the oddness with the “control”.

Even if the experiments and statistical analysis in the article had been conducted perfectly (which doesn’t often happen; even scientists are human!), the best possible conclusion that the authors and bloggers could have come to is that coconut oil is better than mineral or sunflower oil at preventing some cuticle scales from being lost. Nevertheless, as pointed out by our experts, even this statement might be an exaggeration.

Of course, we’re not claiming that coconut oil has no beneficial effect on hair. What we’re saying is that people need to be critical when reading or referencing scientific articles, and that there’s still a lot of research to be carried out on coconut oil … and its “psychic” powers ; )

I hope this in-depth look at the truth behind this beauty science article has given you something to think about next time you consider treating your hair. If you have a specific scientific article that you’d like our experts to take a look at then let us know!