Wet combing compared with pediculicides for head lice: single blind randomised study

Head lice are a common problem in schoolchildren, but can physical treatments beat chemicals? Martin Dawes looks at a single blinded randomised study that compared the two

This month's paper is Hill N, Moor G, Cameron MM, Butlin A, Preston S, Williamson MS, et al. Single blind, randomised, comparative study of the Bug Buster kit and over the counter pediculicide treatments against head lice in the United Kingdom. BMJ 2005;331:384-6. You can read it by going to studentbmj.com and clicking on the link.

Why do the study?

You have to take your hat off to the head louse. For thousands of years it has been a source of irritation and disgust. Described in ancient Egyptian and Greek medical texts and today, with 699 000 hits on Google, the mostly harmless head louse has developed into an apparently fearsome pest. In the past 2000 years, various treatments have been proposed. Not one has worked sufficiently for it to be regarded as a panacea.

The prevalence of head louse infestation in primary schools in the United Kingdom is 2%, which does not seem high. But 37% of children had had head lice in the previous year (incidence)—so the problem is considerable.

This study investigated the use of the “Bug Buster kit,” a kit comprising four fine toothed combs with instruction to use them with conditioner four times in two weeks. This was compared with treatment with pediculicides. There are several forms of pediculicide treatment, some of which are more than 80 years old (malathion).¹ The recent interest in the non-pharmaceutical approach is because of increasing parental concern about the use of pediculicides in children. So far only minor adverse events have been reported with the use of these agents.² The comparison of effectiveness of comb and pediculicides is certainly not new.³

What is the question—PICTO?

Determining exactly what question the researchers were trying to answer is important. For any therapeutic trial, you want to know five things:

The Patients

The Intervention

Whether there was a Comparison group

The Time between intervention and follow up

The Outcome.

In some cases this is clearly identified usually towards the end of the introduction section of the paper. In this paper the authors wrote, “We compared the effectiveness of the current (1998) Bug Buster kit [intervention] with over the counter pediculicides [comparison] containing malathion or permethrin among representative populations from four counties in England and one county in Scotland.” Their patients were children with head lice aged between 2 and 15 years, and their outcome and time was live head louse found five days after application for insecticides or at 15 days after starting using the Bug Buster kit.

What is the study design?

This is a randomised single blinded study. The blinding refers to the fact that the patients knew what they were using—this could not have been avoided. The nurses evaluating the success of treatment, however, did not know which treatment had been used. The control relates to three aspects of the study design. Firstly, usually the investigators will control the two groups for important factors, such as duration of infestation with head lice or severity of infestation. That is, they would make adjustments to the randomisation process before recruitment to control for this. Secondly, they used a control group, in this case the pediculicide group. And thirdly, the study had some form of regulatory control, which in this case was provided by ethical review.

CDC/DR JURANECK
Effective treatment will stop you feeling lousy

What are the details of the study?

The investigators used general practitioners, school based head lice awareness campaigns, and posters in pharmacies to recruit patients in several regions of the country. This seems a sensible pragmatic approach.

A considerable problem with this study was that the general practitioners recruiting patients in the trial were all given the randomisation list. That is, they could see who was having which treatment before even talking to the patient about joining the study. This lack of concealment of the randomisation list from the recruiting clinician may have resulted in selection bias. If I were a general practitioner, I might have recruited patients with severe infestation only if they were to get pediculicide and less severely affected patients if they were to get the Bug Buster kit. This in turn may lead to as much as a 30% increase in apparent efficacy of bug busting,⁴ compared with a trial in which randomisation was adequately concealed. Concealment of randomisation is extremely important but reported in only 60% of articles.⁵

Patients were randomised to treatment using a list of randomly generated numbers. The simplest way to allocate patients is to use a list of random numbers between zero and nine. If the number is less than five the patient is to have experimental treatment, and if the number is five or greater the patient gets placebo. If you look at one of these lists that are found at the back of most statistic text books you may find the following sequence—1, 4, 2, 3, 1, 3, 1, 2, 6, 6. By chance, the first 8 numbers are less than 5 so that the first 8 patients would all receive Bug Buster kits. Even using the randomisation list, therefore, it is quite possible that some general practitioners only used one form of treatment as the numbers recruited in some regions was low. This additional bias might have been avoided by using block randomisation.

The first table in most papers about treatments allows us to compare some aspects of the two groups and identify the effectiveness of randomisation. The items mentioned differ little clinically. But the duration of the attack, the length, thickness or oiliness of hair, and most importantly the intensity of infestation are not mentioned. So we do not know whether the randomisation process actually achieved its aim.

In real life, blinding of the evaluators is problematic as participants will often say what they were using even though the evaluator has asked them not to. The lack of blinding of the patients may have resulted in a 15% larger difference in the results compared with the results if it had been blinded.⁴

What were the results?

On the face of it this paper shows the effectiveness of wet combing with conditioner over pediculicides. The cure rate of 57% is much higher than the pediculicide treatment cure rate of 13%. The latter rate is surprisingly low when compared with the results in other trials, which are generally 70-80%.^{2 6} You have to ask why the rate of success of the pediculicides in this trial is so low compared, for example, with a recent trial testing phenothrin against dimeticone,⁷ which found cure rates of 75% and 70%, respectively.

The authors of the current study point out that their pediculicide treatment was a single dose and was aqueous based. The British National Formulary states, “Permethrin is active against head lice but the formulation and licensed methods of application of the current products make them unsuitable for the treatment of head lice.” For malathion, the single use is cited as a reason for failure by the authors. The British National Formulary advises a double dose, and double doses of phenophrin and dimeticone were used by Burgess and colleagues.⁷

Was it a good study?

Doing good research is hard. The whole system of medical research is biased because of quickly prepared grant applications with little time and frequently no resources to work out the details. This inevitably leads to research that has not considered important factors that might lead to the bias. Unfortunately, this study has several flaws that weaken its conclusions making them less valid.

In clinical practice, we need several randomised controlled trials collated in a systematic review before we really know what the truth is. Head louse infestation is costing the United States an estimated $350m (£193m; €284m) a year and yet there are few randomised controlled trials.⁸ At the bottom of each paper you will often see the phrase “more research is needed,” which in this case is absolutely true.

What conclusions can you draw?

Possible explanations for the large effect of Bug Buster kits and the smaller effect of pediculicide in the current study are that the results are true, the results are due to chance, or the study was biased against treatment with pediculocide. Despite these reservations, this paper confirms that Bug Buster kits, in the right hands, seem to be effective. Indeed, from previous evidence it looks as though Bug Buster treatment is probably as effective as pediculicide treatment applied twice.

Competing interests: MD has taken part in a pharmaceutical sponsored national committee to develop educational materials covering the management of head louse.

This article is adapted from one first published in the BMJ (2005;331:362-3).

Martin Dawes, chair of family medicineDepartment of Family Medicine, , McGill University, 515 Avenue des Pins, Montreal, Quebec, H2W 1S4, Canada
Email: martin.dawes@mcgill.ca


studentBMJ 2005;13:309-352 September ISSN 0966-6494

1. Parasitical preparations. In: British Medical Association and British Pharmaceutical Society of Great Britain. British national formulary. London: BMA, BPSGB, 2005: 589. (No 49.)
2. Dodd CS. Interventions for treating headlice. Cochrane Database Syst Rev 2001(3):CD001165.
3. Auden GA. The problem of the head louse. Lancet 1921;198:370-2.
4. Juni P, Altman DG, Egger M. Assessing the quality of controlled clinical trials. BMJ 2001;323:42-6.
5. Hewitt C, Hahn S, Torgerson DJ, Watson J, Bland JM. Adequacy and reporting of allocation concealment: review of recent trials published in four general medical journals. BMJ 2005;330:1057-8.
6. Vander Stichele RH, Dezeure EM, Bogaert MG. Systematic review of clinical efficacy of topical treatments for head lice. BMJ 1995;311:604-8.
7. Burgess IF, Brown CM, Lee PN. Treatment of head louse infestation with 4% dimeticone lotion: randomised controlled equivalence trial. BMJ 2005;330:1423.
8. Jones KN, English JC 3rd. Review of common therapeutic options in the United States for the treatment of pediculosis capitis. Clin Infect Dis 2003;36:1355-61.