Editor’s note: Thanks to Joe for Biofortified’s first post of 2012! Here’s to a great year of science blogging!
I like to think of myself as a skeptical blogger. I like to engage in critical thinking about scientific issues because this is an important aspect of my job as a graduate assistant. When I move into the workforce, I’ll still need some basic skills to parse evidence because this is my job as a scientist. Mythbusting is a great opportunity to do this, and I enjoy discussing things which may help people who read my posts whenever I can. Being an entomologist gives me some rather interesting opportunities to do this, which is leading me to discuss head lice of all things.
In my last post, Do OTC Head Louse Treatments Work? Part 1: Mechanisms, I explained how the most commonly used FDA-approved treatments worked. In addition to those science-based products, there are many products that have no evidence of efficacy behind their claims, and that rely on fear to make a sale. What I’ve seen deeply concerns me not only as a scientist trying to make the world a better place, but as a parent trying to raise my daughter the best that I can. In this post, I’ve taken a few commonly sold products and listed some ways in which I think they play fast and loose with their claims.
A very brief review of how the nervous system works and how pesticides work in general can be found in the video below:
How do we know if treatments work?
Before getting down to the business of mythbusting, I think it’s appropriate to discuss how we know various products work. Treatments are assessed through clinical trails where infested volunteers subject themselves to putative treatments which are known as in vivo trials. In some cases, the lice are removed from the volunteers and exposed to the treatments in petri dishes which are known as in vitro treatments. In vitro treatments must be performed with the louse’s biology in mind because removing the louse from the host means that the louse is no longer in its natural environment. If the louse is not in it’s natural environment, the results gained from such a test may not be applicable to a real infestation. In vitro tests can disprove that a product works under ideal conditions, but proof of efficacy ultimately requires that the product be tested in real world conditions.
Clinical trials must have large numbers of people (and large numbers of lice) and untreated control groups. After all, insects are surprisingly fragile critters and even water or non-insecticidal shampoos may result in a small amount of mortality which is insignificant to treatment. Water or noninsecticidal shampoos can also temporarily clog the insect’s spiracles, resulting in immobile lice which could be interpreted as dead by a careless counter. Removal can physically injure the lice, which could cloud trial results if results are drawn from collected lice.
Another important aspect of clinical trials is blinding and randomization which make sure the person who is counting the lice isn’t aware of the treatment the person received. The human mind is a surprisingly bad tool for science because we tend to see patterns where none exist, and we may unintentionally superimpose patterns that don’t exist. Since everything in nature has some amount of variability (Anastasia is about six inches shorter than I am, Bug Girl is about a foot shorter than I am, and my boss is about a foot taller than I am for a quick example) we use statistics to tell us what the probability is that our results are due to random chance, eventually ending up with something known as a P-value. Followup observations are also required to show that the patient remained louse free, that is that there weren’t any hidden adults or unhatched eggs because the unhatched eggs can restart infestations.
Last week, I discussed some common OTC head louse treatments. While effective, there are some problems with resistance for some OTC treatments which results in failure of some treatments. This is a product of evolution where some lice are able to survive treatment because they have some random mutations which just so happen to be beneficial in a pesticide filled environment. The mechanisms of this resistance are actually similar to agricultural pests which have been treated with the same product.
One thing astute readers may have noticed is that I didn’t shy away from the use of the word ‘pesticide’ when discussing these treatments. One of my very first posts on Biofortified revolved around the definition of the word ‘pest’ which is completely anthropocentric. A pest is any critter which annoys us in the slightest, and a pesticide is a compound which kills a pest. Insecticides are used to kill insect pests and head louse treatments are referred to as ‘pediculicides’ because they kill lice. All pediculicides are insecticides (because they kill lice, which are insects), and many of the less toxic insecticides used in agriculture have been repurposed as pediculicides. Often times with head louse treatments, you hear companies claim with great pride that their products are pesticide free, are great at killing lice and that no resistance has evolved to their treatment.
Well… how do these claims stack up?
The use of agricultural insecticides to treat head lice is somewhat of an uncomfortable truth, and many companies have taken advantage of this to market head louse treatments. Despite what any label you read may say, any product which claims to kill lice is an insecticide by definition. It doesn’t matter if these are plant extracts, because pyrethrum falls straight into this category and it is classified as an insecticide. In fact, I would even go so far as to argue that a product is engaging in false advertising if it claims to kill headlice while being pesticide free. This, of course, doesn’t mean that all products must directly interfere with the inner workings of lice to be potential treatments.
Some compounds like mineral oil are used as insecticides in agriculture to kill aphids by suffocating them. The product marketed as ‘Lice MD‘ in the picture above claims to kill lice through a similar mechanism. The fact that these chemicals do not interfere with the neurological systems of insects does not mean that the product isn’t an insecticide. If the product claims to kill lice, as Lice MD does, it is claiming to be an insecticide.
Uneasiness about insecticides has also given rise to many products which are derived from natural sources; these are popular because of a general assumption that natural products are safer than synthetic insecticides. The advantage of this from a company’s point of view is that these products don’t have to go through safety or efficacy testing, depending on how they’re marketed. The Dietary Supplement Health and Education Act of 1994 allows many products to go straight to market without testing under the guise of ‘supplements’ which allows them to make sometimes outlandish health-related claims. Homeopathic products are similarly exempt from safety and efficacy testing, which gives companies a great loophole to sell products which make medicinal claims.
This is a successful tactic because it plays on the unease parents have about treating their children with insecticides to kill lice. Unfortunately for these uneasy parents, the assumption that natural products are less harmful than synthetic products doesn’t always hold true. The LD50s for many synthetic pyrethroids are higher than their natural counterparts. Ricin and amantin are both incredibly powerful poisons derived from plants and fungi respectively. Eucalyptus oil, if used improperly as a head louse treatment, can have dire consequences including seizures and death. Many natural components can have chronic effects, too. Cyclopamine, derived from Vetratum californicum, causes some rather disturbing birth defects by inhibiting developmental pathways. Rotenone, a pesticide once used widely in organic agriculture, has been linked to Parkinson’s disease in workers exposed to sublethal doses of the toxin over the course of a very long time. Aflatoxins are powerful carcinogens produced by fungi which threaten food supplies all over the world. Even something as seemingly innocuous as a cockroach sex pheromone can be carcinogenic.
To be safe, it doesn’t matter if a chemical is derived from natural sources. Instead, safety depends on how the chemical interacts with the molecular machinery that keeps us alive. The safest way to make a new product is to construct it with chemicals where we know what everything does, as opposed to treating with soups of unknown composition. Unfortunately, this isn’t always possible because purifying and testing a compound for effects is extremely expensive and can take years of effort.
One of the components in the Quit Nits formula is a plant called Delphinium, a plant genus which is famed for its toxic alkaloids that poison cattle and make this plant genus a pest of cattle pastures. I’ll discuss this claim further in another paragraph, but the plant is in the formula at a concentration that is most likely too low to harm either lice or people. Other claims on this product are technically honest, but misleading. Some components of this product have been used in agriculture as insecticides. Lice consume a blood diet, and would probably have to eat these pesticides to see any effect. Components of the shampoo are toxic, but the concentrations these components are used in are harmless to people and are probably harmless to lice as well. As far as I can tell, most of these components haven’t been tested against lice in literature available to researchers. The statements made in the Quit Nits product comparison chart are misleading on multiple levels.
Many products give misleading statistics that aim to trick parents into believing the product works. For example, let’s take a look at a product called Lice Sheild. They give a description of an experiment on their website that seems like a good test on the surface but is missing any information that actually allows you to draw any conclusions. For instance, they give P-values in their experimental setup, but do not give any information needed to verify their results. The P-values given imply statistical significance, but without any information on repetitions, sample sizes, means, or deviations the information is basically useless. There’s no way to check their math to see if the statistics were correctly performed. There are also no details on how many lice they used for the experiment. If they used two repetitions of five lice (the minimum required for 80% repellency with four lice moving between hair strands and away from the treatment), this would be a negligible result. If they used ten repetitions of 500 lice, the results would be a bit stronger. There simply isn’t enough information here to determine if the statistics were correctly performed.
Also lacking is a description of the experimental arena which is important because there are many ways in which you can test repellency that would potentially interfere with louse movement. If they placed the louse in a container in a patch of hair, one wouldn’t expect lice to move away from the hair at an appreciable rate. Many products are tested by placing the lice on a piece of filter paper and looking at the percentage of lice which move away from the product. While a test like this may appear to show some repellent activity, it’s not actually a very good measure of how good your product repels lice in real world conditions. Using in vitro tests means you’re trapping the lice in a place with the repellent and giving them essentially unlimited time to make their choice. This is a big problem because they’re using a timeframe that’s irrelevant to head louse transmission. Lice are mainly transmitted through hair to hair contact, and it’s rare for two people to be in hair to hair contact for this amount of time.
In short, showing 80% repellency is very different than saying that you have a 80% reduction in your chances of getting head lice. The company which makes Lice Shield uses questionable statistics to claim their product repels 80% of lice under conditions that don’t reflect the conditions where lice are transmitted, then turns around and claims in their FAQ this means that there is an 80% reduction in head louse transmission without any evidence for this claim. These two claims are quite different, because repellency doesn’t necessarily translate to a reduction of infestation.
Homeopathic Medicines are Marketed Differently than FDA Approved Drugs
Homeopathy is a system of beliefs which claim that serially diluting a ‘medicinal’ substance makes it stronger. These dilutions are pretty specific, for example the letter X denotes a 1:10 dilution. If a product is diluted 6X, it’s diluted to 10^-6 which is about one millionth of the original concentration. The idea that diluting a potential louse treatment makes it stronger is ridiculous because if the chemicals in any treatment interfere with insect biochemisty, they must be within a certain range to have an effect. Too much, the person gets poisoned (but the lice still die). Too little, and the lice survive and resistance can build up after the more susceptible individuals are culled from the population.
Homeopathy defies essentially every principle in science from biology to physics. Homeopaths claim that water has memory but, to paraphrase Tim Minchin, this ‘memory’ of water seems infinite when paired with some substances but water seems to have a case for amnesia when it comes to more harmful substances. Homeopathy has no a plausible mode of action.
The next paragraph of my post may get me into hot water with some of my skeptic friends. Many skeptical bloggers have taken on homeopathy. Let me restate: homeopathy has no plausible mode of action. I want to put this in writing to avoid the inevitable criticism from other skeptical bloggers. I also want to avoid the quote-mining from naturopaths who may want to say that I support homeopathy. I am not claiming the efficacy of homeopathy because there is no evidence that it works, and there is no plausible mechanism by which this practice could possibly work.
I am considering some homeopathic products as potentially effective. Why? Homeopathic formulas are exempt from safety and efficacy testing by the FDA which gives many products a free pass when it comes to clinical trials. Many products aren’t actually homeopathic because they contain ingredients in concentrations that could potentially have an effect. These products are often classified as homeopathic so they can make medicinal claims. A cold remedy product marketed under the name Zicam is a good example of this. Zicam was a solution of zinc which was marketed to treat the common cold after it was shown that zinc ions could interfere with viral replication in in vitro tests. The product was eventually recalled by the FDA because it was found to destroy the sense of smell. Another example of a product sold as a homeopathic remedy with potentially active components is sold under the name of Quit Nits.
Unlikely Modes of Introduction
Quit Nits bills itself as a homeopathic remedy and contains a bunch of plant extracts from several different species. As a result of intense selection by insect herbivory, all plants have some sort of anti-herbivore defense. Many plants have toxic components as a result of being under selective pressure to develop such components over the course of millions of years. Plants represent a wonderful treasure trove of different types of novel pesticide chemistries. After all, this is how we got pyrethrum. Despite the fact plant extracts are potentially plausible pesticides in and of themselves, we shouldn’t assume that any plant can kill any insect.
The first thing that raises a red flag for me in the Quit Nits formula is the mode of introduction of this pesticide. Some pesticides (see this RNAi post, for example) must be eaten to be toxic, and these are referred to as stomach poisons. Others can be absorbed, and are referred to as contact poisons. While this product does have toxic components, the fact these plants have natural toxins doesn’t automatically mean that they’ll be absorbed by the lice. Because the lice feed by inserting their mouthparts into the host, it seems very unlikely to me that they’d actually be able to pick up any pesticide by eating it unless the pesticide was in the blood of the host in appreciable amounts. Thus, any active ingredient would have to be absorbed through the exoskeleton.
A second thing that I am concerned about is the formulation. Spraying plant extracts on crops and lathering the same stuff into hair and then washing it off are very different modes of introduction. Pesticidal activity may not be preserved by the shampoo, even if the substance is downright toxic to bugs when dissolved in water. This stuff needs to be tested on lice in the formulation offered for sale before it can be said to have insecticidal activity. The mode of introduction and dosage play vital roles in the insecticidal activity. It’s possible the active ingredients wouldn’t retain their insecticidal activity in shampoo or that they wouldn’t be in contact with the lice long enough to be toxic. No pesticide kills every insect with equal efficacy in every situation. This is why the ultimate test of any pesticide is to test it on the pest in the situation you’re going to use it in, in the formulation in which it will be used.
Third, the mode of action of the two active ingredients means that they are unlikely to affect lice. Extract of the plant Delphinium and extract of a plant called Sabadilla (Schoenocaulon sp.) are listed as active ingredients at one part per million. There appears to be little work evaluating Delphinium for insecticidal activity, but Sabadilla and Delphinium both contain veratridine which acts as a stomach poison in insects. Because lice feed by inserting their mouthparts into the skin of the host and sucking blood from capillaries under the skin, I have a tough time believing they’d actually pick the insecticide up in appreciable amounts unless the toxins were absorbed directly into the bloodstream. Since the active ingredient is toxic to humans, there would probably be some major issues with the product if it made it’s way into the bloodstream. The mode of action here renders me skeptical that the lice would pick up a toxic dose of the pesticide in the first place.
Pesticides Used in Doses Unlikely to be Effective
The biggest problem with Quit Nits is the concentrations of the active ingredients. It’s the dose that makes the poison and if you look at the label of the product in question, there are two ingredients that are listed as parts per million and one component that’s in there at a 1:100 dilution.
Purified components of Sabadilla have been used as pesticides for high value orchards like oranges and mangoes. The lowest concentration for semi-purified Sabadilla alkaloids is about .1 g/l, or about one part in 10,000 if we’re going by weight. The extract of the plant seeds, of which the alkaloids are only a small part, is about a hundred times lower than this in the quit nits shampoo. The seeds of Schoenocaulon contain 2-4% insecticidal alkaloids by weight, which means the alkaloids from Sabadilla are present at one part in 25,000,000 in the shampoo.
Delphinium contains veratridine in appreciable amounts as well and has a large amount of other toxic alkaloids in addition to veratridine. It’s difficult to know what concentrations the insecticidal alkaloids are present in Delphinium because there are simply many potentially insecticidal alkaloids in these plants. However, we can make some educated guesses because researchers have purified alkaloids from Delphinium. The individual components are present in milligram amounts with all the alkaloids being present at about 6 grams per kilogram of plant tissue. If we assume the insecticidal alkaloids are present at a concentration of five grams per kilogram of plant material to make our math easy, this means that the insecticidal components comprise about one part in two hundred per unit weight. The concentration of plant in the shampoo is about two parts per million, which means the alkaloids are present at one two hundredth (1/200) this concentration. Given generous assumptions of grams per kilogram amounts, the active ingredients would be present part per hundred million concentrations if we assumed all of the alkaloids in the plants had insecticidal activity.
These plants combined are in about one part in 500,000 in the shampoo. This means that the concentrations of the insecticidal alkaloids is about one part in 4-6*10^-8 parts depending on the alkaloid concentrations of the plants used. Because the lowest concentration of this pesticide used in agriculture is one part in ten thousand, this comes out to a ballpark figure of somewhere around 1,000 times lower than the lowest dose used in agriculture. The dose the lice will be exposed to in the shampoo won’t be great, as there will only be a couple grams of the shampoo used on the entire scalp. To give you an idea of what the pesticidal concentrations are in other louse products, pyrethrum is generally in antilouse shampoos at one part per hundred (one percent). Malathion is generally present at one part in two hundred parts, or one-half percent. This means that the crude alkaloids from the plant extracts would be present at one one millionth the concentration of the active ingredients that have known insecticidal activity. The improbable mode of action combined with the low amounts of active ingredients in the plant means that I would assume these ingredients are essentially inert without proof that they kill lice at these concentrations.
These ingredients aren’t the main stuff in the Quit Nits treatment, though. The plant extracts listed above are in parts per million, but Quassia amara extract is present at a 1:100 dilution…about 10,000 times higher than Sabadilla and Delphinium. Furthermore, it’s in the ballpark of the Pyrethrum extract. So what about Quassia?
Ingredients with No Proof of Efficacy
Quassia amara is an interesting plant because it contains one of the most bitter substances in the world. These substances are called quassinoids, and have been examined for insecticidal and antifeedant activities against a wide range of pests. In many cases extracts and purified components from Quassia have been shown to have insecticidal and antifeedant activity, but it wasn’t always clear to me whether the antifeedant activity was so strong that it led to mortality. In other words, it was difficult to tell if the substance made the food taste so bad to the bug that they’d rather starve than eat. Either way, we’re interested in it’s activity against lice in particular.
There have only been two papers which have examined Quassia extracts against lice. One appears in a Dutch journal in 1978 and another in a Spanish journal in 1991. Since these papers are in rather obscure low impact journals, I was not able to access them directly through my library and instead had to rely on their descriptions in review articles. The review articles weren’t exactly favorable towards Quassia as a louse treatment. The Dutch paper claimed high efficacy, but the experiment was apparently ran as an un-controlled, un-randomized, un-blinded experiment and counts as nothing as far as proof goes. The Spanish paper claimed high efficacy, but the review states that the Spanish paper concluded that Quassia would only have repellent effects but didn’t mention whether the extract had a clinically relevant success rate. There have been no well performed tests of Quassia as a head louse treatment, and the few tests that have been performed have yielded conflicting results. There’s simply no proof that the “active” ingredients in Quit Nits work.
Quit Nits also sells a repellent spray that has undergone independent testing. One paper compared it’s repellent activity using a filter paper repellency test, incubating the lice with filter paper treated with repellent on one side and water on the other. The objective was to measure what percentage of the lice moved away from the treatment. At the earliest time point measured (two hours), Quit Nits performed about as well as water. At later time points, there was some non-significant repellent activity. Another paper looked at the repellency of Quit Nits under real world (or close to real world) conditions and looked at whether lice would transfer to hair under approximated hair-hair contact conditions, if the lice would move on the hair treated with Quit Nits repellent spray, or if the lice would feed on the forearm of one of the authors who performed the study. For the hair tests, KY jelly was used to simulate greasy hair and Quit Nits fared no better than this. For the skin tests, bare skin not receiving any treatment was used and the lice exposed to Quit Nits treated skin fed just as well as those on bare skin. Quit Nits repellent spray simply doesn’t repel lice, as far as the current experiments show.
Some Products Have no Plausible Mode of Action
The first example of a product with no mode of action is a product called X-pel. In fairness, I’ve only seen this at a few small grocery stores in Iowa, but the fact something like this is sold at all really worries me. The product is a shampoo which consists of ground up honeybees, phenol, and an uncommon species of Rhododendron at femptogram concentrations (15X), or one part in one quadrillion. On their website, they give a couple of vague descriptions of various tests. The tests contain very little methodology and give no statistical information about their results. They claim a few ‘major universities’ were involved in the testing of the product, but neglect to give any sort of contact information or any publications generated as I described in the Quit Nits treatment. They have a video on their website, below, where they show an in-vitro test that consists of them drowning a louse in the shampoo. Because lice can be inactive for a long time following immersion in water, there is no evidence given that the lice in this video were actually killed. They also show an uncontrolled, un-randomized, un-blinded test of a single subject without any apparent followup as proof that their product works. Phenol here is the most likely ingredient for insecticidal activity, as the concentration of the Rhododendron is far too low to do anything at all. Quite frankly, I’m not sure how honeybees are supposed to kill head lice unless we assume the venom glands were somehow involved, but I find this unlikely. The ingredients used in this product are only used in vanishingly small concentrations, and there’s really no way to justify using ground up honeybees to treat head lice.
Another product marketed under the name Licefreee! is little more than a concentrated sodium chloride solution. While it’s plausible the product could suffocate the lice, the data for suffocants in head lice treatment isn’t exactly convincing. Because lice are coated in a waxy layer that prevents dehydration, I find the claim that a 10% salt solution will kill lice suspect. As far as I can tell, there’s no evidence of this product works either because I’ve only seen this mentioned in passing under ‘folk treatment’ sections of review articles. I’ve seen no primary literature articles dealing with concentrated salt solutions as lice-killers.
Many of these companies use a variety of tactics to sell their products that have nothing to do with efficacy. Many use highly questionable advertising methods, like capitalizing on patient fears of synthetic medicines and pretending to identify with their customers to sell them products of uncertain effectiveness. Some of these products even go as far as to claim to be pesticide free while still claiming to kill lice. Many of these products claim to have been invented by parents, but as a parent myself I cannot imagine marketing a questionable head louse treatment and this is a big part of why I’ve written this post.
The science-based products currently on the market that I mentioned in Part 1 have been thoroughly studied and activity proven with the obvious exception for strains of lice that are resistant to some treatments. Even though there is a risk to any product you’re bound to use, the risks of these products have been investigated and have been taken into consideration when formulating treatment regimens. I can certainly understand anxiety about exposing kids to pesticides, but when looking at alternative treatments one needs to ask whether they’re safe and effective. Extraordinary claims require extraordinary evidence. If a product you spend money on makes any sort of claim, you should consider the claim extraordinary and ask for evidence behind the claim.
Burgess, I. (2009). Current treatments for pediculosis capitis Current Opinion in Infectious Diseases, 22 (2), 131-136 DOI: 10.1097/QCO.0b013e328322a019
Leone, P. (2007). Scabies and Pediculosis Pubis: An Update of Treatment Regimens and General Review Clinical Infectious Diseases, 44 (Supplement 3) DOI: 10.1086/511428
Canyon, D., & Speare, R. (2007). Do head lice spread in swimming pools? International Journal of Dermatology, 46 (11), 1211-1213 DOI: 10.1111/j.1365-4632.2007.03011.x
Heukelbach, J., Canyon, D., Olivera, F., Muller, R., & Speare, R. (2008). Efficacy of over-the-counter botanical pediculicides against the head louse based on a stringent standard for mortality assessment. Medical and Veterinary Entomology, 22 (3), 264-272 DOI: 10.1111/j.1365-2915.2008.00738.x
Lebwohl, M., Clark, L., & Levitt, J. (2007). Therapy for Head Lice Based on Life Cycle, Resistance, and Safety Considerations PEDIATRICS, 119 (5), 965-974 DOI: 10.1542/peds.2006-3087
Leone, P. (2007). Scabies and Pediculosis Pubis: An Update of Treatment Regimens and General Review Clinical Infectious Diseases, 44 (Supplement 3) DOI: 10.1086/511428
Greive KA, & Barnes TM (2011). In vitro comparison of four treatments which discourage infestation by head lice. Parasitology research PMID: 22030833
Canyon, D., & Speare, R. (2007). A comparison of botanical and synthetic substances commonly used to prevent head lice (Pediculus humanus var. capitis) infestation International Journal of Dermatology, 46 (4), 422-426 DOI: 10.1111/j.1365-4632.2007.03132.x
Rossini, C., Castillo, L., & González, A. (2007). Plant extracts and their components as potential control agents against human head lice Phytochemistry Reviews, 7 (1), 51-63 DOI: 10.1007/s11101-006-9026-0