Five guidelines met our inclusion criteria, developed by: the World Health Organization (WHO), the Public Health Agency of Canada (CA), the Hong Kong Centre for Health Protection (HK), the Public Health England advisory committee on malaria prevention (UK) and the Centre for Disease Control (USA).
We excluded an article providing advice for travellers from New Zealand, as it was not official national guidance, and although our PUBMED search suggested there may be guidelines from Japan, The Netherlands and Spain, these were unavailable in English. No guidance was identified from Australia or Scotland. A brief summary of these guidelines is outlined in Table 1.
Agree II Appraisal of Guidelines
The summary AGREE II scores for each of the five guidelines are shown in Table 2.
Domain 1: Scope and Purpose. All five guidelines covered both drug and non-drug interventions to prevent malaria in travellers, and all aimed to assist clinicians as they advise travellers. The UK and HK guidelines provided the most complete description, clearly outlining the objectives, health issue and target population to which the guidelines apply.
Domain 2: Stakeholder Involvement. All guidelines gave a named list of the individuals who primarily authored the document, although only the UK guidelines stated the roles and responsibilities of these individuals. Three guidelines (HK, USA and WHO) also referred to working groups or external contributors but these were not named. None of the guidelines reported involvement of the target population (travellers).
Domain 3: Rigour of Development. None of the guidelines included a Methods section. None described methods to systematically search for and appraise evidence, or to move from evidence to formulating recommendations. None of the documents described external peer review prior to publication, and only the WHO guideline made reference to a process for future updates.
The Canadian guideline presented the evidence base summarized in GRADE tables. All other guidelines summarized evidence in text form, with variable levels of referencing provided to validate recommendations. USA and WHO guidelines provide no references, HK provide some references and UK guidelines provide a reference for many of the recommendations made.
All guidelines discussed the health benefits and side effects of interventions to some extent, although the level of detail, particularly around safety was highly variable. The Canadian and WHO guidelines gave some risk estimates for the potential benefits and harms.
Domain 4: Clarity of Presentation. All of the guidelines presented some clear recommendations that were unambiguous and specific, and clearly identifiable. However, all guidelines also included discussions of options without clear recommendations, or implied recommendations incorporated into the text and without clear signposting. The Canadian guidelines were presented most clearly, listing all explicit recommendations in a table.
Domain 5: Applicability. None of the guidelines discussed the resource implications of the recommendations, and none described potential barriers or facilitators to implementation. None presented audit criteria. Only the USA guideline presented an outline of the advantages and disadvantages of each drug in a user-friendly table.
Domain 6: Editorial Independence. Only the USA guideline described the sources of funding, and made any reference to potential conflicts of interest among guideline developers. The remaining guidelines said nothing, and therefore scored zero for this domain.
Agreement or Disagreement Between Guidelines
a. Drug Interventions for Malaria Prevention. A summary of the drug recommendations made by each of the guidelines has been presented in Additional file 1 http://www.biomedcentral.com/1471-2458/14/1129/additional. The main areas of conflict and omission are related to information on effectiveness, safety and contraindications:
All guidelines state that atovaquone-proguanil, doxycycline and mefloquine are effective against chloroquine resistant falciparum malaria. Two guidelines quantify this: the Canadian guidelines state that they are 95% effective and the UK guidelines state that they are 90% effective, each referencing different documents.
The Canadian, Hong Kong and USA guidelines state that atovaquone-proguanil has an excellent safety profile, with only the Canadian guidelines quantifying that 8% to 15% of individuals experience nausea, vomiting, abdominal pain or diarrhoea. The guidelines state that the most common side effects are headache and gastrointestinal upsets, but again do not quantify this.
Although the Canadian guideline outlines that atovaquone-proguanil has an excellent safety profile, it outlines that it should be used in caution in patients with HIV and liver disease. This is not discussed in other guidelines;
The US guideline states that atovaquone-proguanil cannot be used in children under 5 kg, whereas all other guidelines state 11 kg as the cut-off;
All guidelines state that doxycycline may be photosensitising and can lead to a rash, however no guidelines quantify this risk;
All guidelines state that doxycycline may lead to gastrointestinal upset and oesophageal ulceration, however no guidelines quantify this risk;
The UK guideline states that doxycycline is contraindicated in children under 12 years, while all other guidelines say 8 years;
All guidelines state that mefloquine can lead to major neuropsychiatric side effects such as seizure, psychosis and suicidal ideation. The Canadian and WHO guidelines quantify this. Canadian guidelines suggest that 1 in 6,000 to 13,000 have seizure or psychosis; and WHO guidelines suggest severe neuropsychiatric disturbance in 1 in 10,000.
The Canadian, Hong Kong and USA guidelines state that mefloquine can lead to minor side effects such as nausea, strange vivid dreams, dizziness, mood changes, insomnia, headache and diarrhoea. The UK and WHO guidelines do not mention this. Only the Canadian guideline quantifies this, suggesting that 95% of people have no or mild, temporary side effects, with 1 in 250 to 500 having mild neuropsychological reactions e.g. anxiety or mood change;
The US guideline states that mefloquine can be given to all ages, while other guidelines state it cannot be used in children below 5 kg;
Stand-by treatments is recommended as being valuable in all guidelines, although the USA guidelines provide very little information regarding this.
b. Non–Drug Interventions for Malaria Prevention. A full outline of the non-drug recommendations made by each of the guidelines has been presented in Additional file 2 http://www.biomedcentral.com/1471-2458/14/1129/additional. There is varying levels of consensus between the guidelines, with a number of significant conflicts.
The main conflicts and omissions are that:
The WHO guidelines states insecticide-treated clothing lasts longer than skin applied insecticide, whereas UK guidelines states the opposite. Other guidelines provide no recommendations regarding this;
Concentrations of DEET recommended in children vary from 10% (CA, HK) to 50% (UK), and in adults vary from 35% (CA, HK) to 50% (UK, USA) in adults. WHO guidelines state that DEET should be used according to manufacturers' recommendations;
The Canadian and WHO guidelines state that DEET can be used at any age, while the UK and USA guidelines state it should be restricted to children over the age of 2 months. Hong Kong guidelines do not provide recommendations on age limits of DEET;
The WHO guidelines state that no other protection is required if a hotel is air-conditioned. The UK and USA guidelines imply that malaria transmission is reduced in air-conditioned rooms, and the Canadian and Hong Kong guideline do not provide recommendations on air-conditioning;
Light covering clothing is recommended by Canada and Hong Kong, while the UK guidelines state this is not effective. USA and WHO guidelines do not provide recommendations on the use of light covered clothing.