Behavioural aspects of travellers in their use of malaria presumptive treatment.

Behavioural aspects of travellers in their use of malaria presumptive treatment P. Schlagenhauf,' R. Steffen,2 A. Tschopp,3 P. Van Damme,4 M.-L. Mittelholzer,5 H. Leuenberger,6 & C. Reinke7 The use of stand-by treatment for malaria by travellers depends on their knowledge, attitudes and be- haviour. We examined the behavioural aspects of a cohort of travellers from Switzerland to low-risk malarial areas who, on recruitment, were provided with a kit containing medication for stand-by treat- ment, guidelines on the diagnosis of malaria, and materials for collection of blood samples for later con- firmation of malaria. All subjects were urged to seek medical advice at the first signs of possible malar- ial symptoms. Illness (fever as the main indicator) was reported by 123 of the 1187 participants, often accompa- nied by shivering/chills (36.6%), headache (35.0%), gastrointestinal symptoms (69.9%), and myalgia and/or arthralgia (41.5%). Two-thirds of those ill failed to seek medical attention despite their symptoms and pretravel advice. Only 9 (7.3%) were actually beyond the reach of medical attention. The stand-by treatment was self-administered by 6 travellers, only one of whom had confirmed malaria. Two non- serious adverse events were reported. All users consulted a physician after administering the presump- tive treatment. This stand-by approach is limited by inappropriate behaviour and poor malaria aware- ness among travellers. These negative factors can be mitigated by development of an improved kit con- taining a simple test for self-diagnosis. Introduction More than 10 000 travellers every year fall ill with malaria and the case fatality rate in infections due to Plasmodium falciparum varies between 0.4% and 8.7% (1-3). Both the World Health Organization (4, 5) and the U.S. Centers for Disease Control (6) recommend that in addition to, or sometimes instead of chemoprophylactic drugs, travellers to malaria- endemic areas may carry with them suitable medica- tion for presumptive or stand-by treatment (SBT) for emergency use when malaria is suspected and medi- I Division of Communicable Diseases, Institute for Social and Preventive Medicine, University of Zurich, Sumatrastrasse 30, 8006 Zurich, Switzerland. Requests for reprints should be sent to Dr Schlagenhauf. 2 Professor and Head, Division of Communicable Diseases, Institute for Social and Preventive Medicine, University of Zurich. 3Statistician, Division of Biostatistics, Institute for Social and Preventive Medicine, University of Zurich. 4Research Assistant, Department of Epidemiology and Commu- nity Medicine, University of Antwerp, Antwerp, Belgium. 5 Tropical Medicine Unit, F. Hoffmann La Roche, Basle, Switzer- land. 6 Professor and Head, Institute of Pharmacy, Basle, Switzer- land. 7Institute of Pharmacy, Basle, Switzerland. Reprint No. 5589 cal help is unavailable. This altemative is envisaged primarily for travellers to remote malarious areas where access to diagnostic and therapeutic facilities is limited. The rationale behind this recommendation is that in areas of low transmission the risk of ad- verse events attributed to chemoprophylaxis (sup- pressive therapy) exceeds the benefit of avoided in- fections (7). Presumptive treatment can be life-saving since falciparum malaria in a non-immune traveller requires prompt treatment. Knowledge of, attitudes towards and practices of stand-by treatment are poorly documented. There is some information on the use of presumptive therapy by travellers (8) and airline crews (9, 10). Data on associated problems such as treatment failures (11), inappropriate use for non-malarial fevers, and inci- dence of adverse events are scarce. There is very little information on the behavioural aspects of trav- ellers who are prepared for stand-by treatment, including the geographic location, type and timing of symptoms, chronological sequence of events, avail- ability of medical help, compliance, medical follow- up and final outcome. We therefore conducted a longitudinal study to elucidate the behavioural char- acteristics of travellers with regard to the use of ma- larial stand-by treatment. Volunteers were followed by questionnaire and serologically using blood sam- ples from the fingertip. Bulletin of the World Health Organization, 1995, 73 (2): 215-221 © World Health Organization 1995 215 P. Schlagenhauf et al. Materials and methods Selection of subjects. All travellers, who could nor- mally be recommended to carry stand-by treatment with them according to Swiss guidelines (12), were invited to take part in a cohort study at the Vaccina- tion Centre of the Institute for Social and Preventive Medicine, University of Zurich, between March and November 1992. Their destinations were predomi- nately areas of Central and South America or desti- nations in Asia (Fig. 1). The nature of the study was fully explained to each potential volunteer, who entered the study by giving written informed consent. Excluded from the study were pregnant or nursing women, persons with a history of allergy or severe reaction to components of the trial medica- tion, and those with a history of epilepsy or psychi- atric disorder. Long-term travellers (sojoums of more than six months) were excluded due to follow- up difficulties. Materials. Each volunteer was provided with a kit containing the stand-by treatment which, as recom- mended by Swiss experts in 1992, was the triple combination called Fansimef (mefloquine 250 mg, sulfadoxine 500 mg, and pyrimethamine 25 mg). A single administration of three Fansimef tablets was the recommended dose for persons with a body weight >45 kg in cases of suspected malaria. The front of the kit contained concise information on how to recognize malarial symptoms and what pro- cedure to follow if the condition were suspected. Appropriate use of the medication was defined as a single dose of 3 tablets of Fansimef administered if all of the following were applicable: the traveller was unwell with fever (37 °C on two readings taken at least 2 hours apart) with or without headache and/or myalgia; Fig. 1. Number of travellers and their destinations. More than one destination is possible. 500 400 300 200 0 z 100 0- mesia QcoZ e 0 0d C~~~~~ =bC 0 .9 - d co 0 'u0 EDDE c _ -m 5) .5 fl 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Destinaton - at least 6 days had elapsed since the traveller entered the endemic area; and - medical attention was unavailable within 12 hours of the illness. After this presumptive treat- ment it was stressed as imperative that the travel- ler should visit a physician as soon as possible to confirm the presumptive diagnosis. A thermometer in a protective case was also provided, together with material (cotton swabs and lancets) for sampling fingerprick blood (50 pl) and chromatography paper that was clearly demarcated into three imprinted circles: one each to contain the blood droplet on recruitment, when malarial infec- tion is suspected but before medication is adminis- tered, and after return to Switzerland irrespective of whether malarial treatment had been administered. This last sample was to be collected at least four weeks after the return date to ensure that any malar- ial cases occurring in this time frame were included in the study. Thus, each volunteer provided at least two blood samples, by fingerprick, which were spot- ted directly onto the chromatography paper. Each volunteer also received a questionnaire concerned with the following: - data on the traveller and his/her destination and type of sojourn; - data in cases of illness with fever; - data when presumptive treatment was adminis- tered; and - data on medical consultations, malaria tests and differential diagnoses. A stamped addressed envelope was provided for return of the questionnaire, the chromatography paper blood samples, and unused medication. In ad- dition, a postcard was included in the kit (to be retained by the traveller for six months after return) in case malarial symptoms or delayed adverse events appeared later than one month after return. Written reminders were mailed to those who failed to return the questionnaires and blood samples as instructed. Analysis of blood samples. The blood samples, elu- ted from the chromatography paper, were analysed using enzyme-linked immunosorbent assay (ELISA) for the presence of IgM and IgG antibodies to the synthetic peptide (NANP)50, which represents the immunodominant central repeat region of the major surface protein of P. falciparum sporozoites. Earlier work (13) suggests good sensitivity with this method but to ensure complete detection, the paper samples were later re-tested using the polymerase chain reac- tion (PCR) (14-16). PCR amplification was carried out on a polymorphic region of the major merozoite surface protein 1 (MSPI) of both P. falciparum and WHO Bulletin OMS. Vol 73 1995 2DaZ ,0 216 Use of malaria presumptive treatment by travellers P. vivax (17, 18). Cases of malaria were included only if confirmed by a positive ELISA or a positive PCR, or microscopically from blood smears obtained by the physician consulted when malaria was sus- pected. Persons who were treated presumptively but remained unconfirmed were considered to be non- malarial cases. Statistical analysis. Data were analysed using the SPSSPC software package. The relationship between use of stand-by treatment and other variables was analysed using the chi-squared test and testing Kendall's X coefficient. Significance was defined as P<0.05. Results A total of 1572 volunteers (approximately half of all who received stand-by treatment at the vaccinating centre in the specified period) were recruited, of which 1187 (76%) were evaluated. The remaining volunteers (24%) failed to return their questionnaires despite reminders or were lost to follow-up. The demographic data are shown in Table 1. The dura- tion of the travellers' sojoum varied from less than one week to periods of up to six months. The volun- teers were exposed for 4918 weeks in the low-risk malaria endemic areas. A total of 123 subjects (10.4%) reported that they had been ill with fever during their trip or in the month after retum. Addi- tionally, 14.2% were ill without fever or had some form of minor accident, and 73.5% of all partici- pants regarded themselves as healthy throughout the entire sojourn. The relationship between the travel- ler's sex and incidence of illness was not significant (P=0. 12). Younger travellers (aged <30 years) tended to classify themselves as ill more often than older travellers. This difference was significant (P<0.01). The first group, those reporting illness with fever, were further investigated since fever was one of the prerequisites for presumptive malarial treatment. Symptoms, concomitant with the fever, which could be ascribed to malaria included shivering/chills (36.6%), headache (35.0%), gastrointestinal symp- toms (69.9%), and myalgia and/or arthralgia (41.5%) (Fig. 2). Behaviour in response to illness. In reply to the question: "When did you first suspect malarial infec- tion?", 7 (5.6%) of the 123 persons with fever sus- pected malaria before day six of their trip while 41 (33.3%) contacted a physician within the specified time-frame of 12 hours (including 12 (9.8%) who contacted a doctor but could not obtain a consulta- tion within 12 hours). A total of 82 travellers (66.6%) failed to seek medical attention despite their Table 1: Demographic data on the 1187 participants in the study Characteristic % Characteristic % Age group (years): Duration of stay: <20 1.2 <1 week 5.6 20-29 47.4 2 weeks 17.4 30-39 26.4 3 weeks 29.7 40-49 12.0 4 weeks 27.5 50-59 8.1 -2 months 11.2 .60 3.9 3-6 months 6.0 Unknown 1.0 Unknown 2.5 Sex: Place of staya Male 50.4 Mainly in cities 23.3 Female 48.6 Mainly in towns/ Unknown 1.0 tourist centres 48.2 Villages/countryside 33.1 Type of trip: Tourist 64.5 Adventure 21.9 Business 1.9 Visit family/friends 7.6 Other 2.2 Unknown 1.9 a More than one answer possible. symptoms. This category included 9 persons (7.3%) who did not consult a physician as they were out of reach of medical attention. Of those who received medical attention, the time interval from onset of symptoms to medical consultation was in 95% of cases less than 4 days (range, 4 hours to 30 days; mean, 44 hours), and 59.1% of those who tried did manage to receive medical attention within 12 hours. The behaviour in reaction to perceived illness did not vary significant- ly with age or sex although there was a trend show- ing that women were more compliant with the issued instructions than men. Fig. 2. Percentage distribution of symptoms In 123 patients with fever. A patient could report more than one symptom. 50i 40 20- CD)0L.2 10 .Diarrhoea Chills Headache Myalgia Vomiting Others WHO Bulletin OMS. Vol 73 1995 217 P. Schlagenhauf et al. Influence of education. Those with higher education (university level) were less likely to consider them- selves ill with fever and possible malarial symptoms than the other travellers (P<0.01). Those with a lower educational status were more correct in their behaviour and their reaction to illness (P=0.01). In general, the lower the level of education the greater the likelihood that the traveller would seek medical attention when ill. Users of presumptive treatment. Only six persons, three women and three men, actually used the thera- py for suspected malarial symptoms after the 6th day abroad. This accounts for 0.5% of the complete study population and also for a small proportion (4.6%) of those in the "ill with fever" group. The characteristics of the six users are shown in Table 2. All users of the medication complied with the dosage instructions and administered the three Fansimef tab- lets as a single dose. Two travellers self-adminis- tered their medication back home in Switzerland on the advice of their doctor. The time required for fever clearance ranged from 1 to 3 days. All users consulted a doctor after the administration of the medication to check the presumptive diagnosis; in some cases, medication other than the antimalarial agents was prescribed. Only one of the users actually had a malarial infection (confirmed by blood smear). Adverse events. Subjective adverse events were experienced by two of the six Fansimef users: one case of subjectively severe dizziness resulting in 2 days incapacitation (bed rest required) was reported. The other event was insomnia, lasting for two nights. Fig. 3. Imported malaria cases from Thailand among Swiss travellers, 1983-92, using chemoprophylaxis (till 1988) and stand-by treatment (from 1989). Discussion The participants in this study all travelled to classi- fied low-risk areas where the probability of contac- ting a malarial infection during a limited time period is minimal. The soundness of the antimalarial strategy adopted by the Swiss (i.e., no chemoprophylaxis, but prepared for stand-by treatment) during travel to low-risk areas is demonstrated by the fact that the number of imported malaria cases from areas, such as Thailand, has not increased since this strategy was introduced in the autumn of 1988 (Fig. 3). There are, however, problems inherent in the strategy as shown in our study. Diagnosis of malaria is difficult, even for medical professionals and it is likely that trav- Table 2: Characteristics of six travellers who used stand-by treatment for presumptive malaria Age/sex Area Onset of Physician Fever (years) of use Symptoms symptoms consulted Diagnosis clearance in: User 1 26/F Colombia Fever, headache, Day 17 Out of reach Gastritis 3 days myalgia, diarrhoea, vomiting User 2 38/M Switzerland Fever, headache, Day 24 Yes Malaria 3 days (returned from chills, myalgia (confirmed by Malaysia) microscopy) User 3 28/F Hong Kong Fever, headache, Day 20 Yes Viral infection 2 days (on arrival from chills, myalgia Indonesia) User 4 58/M Switzerland Fever, diarrhoea, >Day 7 Yes Amoebiasis 2 days (on return from vomiting Indonesia) User 5 28/M Sarawak Fever, shivering, Day 17 Yes (twice) Viral fever 1 day (Malaysia) myalgia, headache, (dengue diarrhoea, vomiting fever) User 6 29/F Nepal Fever, diarrhoea, Day 6 Yes Viral infection 1 day headache, vomiting, shivering, myalgia WHO Bulletin OMS. Vol 73 1995218 Use of malaria presumptive treatment by travellers ellers may mistake influenza or some other febrile illness for malaria and use the stand-by treatment. In this study, 123 persons (10.4%) judged their own clinical symptoms, with "malaria" as one possible diagnosis. Six persons used the stand-by treatment (0.5% of the entire cohort or 5% of those with pos- sible malarial symptoms), but only one had con- firmed malaria. On the other hand, some persons with typical malaria symptoms (82 (66.6%) of those ill) often hesitate to use the treatment-and a few of them may really have malaria. Thus, while wrong use of stand-by treatment may lead to serious adverse effects, failure to use it could result in death due to malaria. The efficacy of the medication as a stand-by treatment cannot be assessed from this study, but travellers to areas of multiple-drug resistance (such as the border areas of Thailand) should be aware that the use of emergency malarial treatment in such areas could act as a short-term febrifuge and symp- tom suppressant without a cure, so that prompt medi- cal attention is always imperative (8). The inappropriate use of stand-by treatment may expose persons without malaria to a significant drug risk. With mefloquine, for example, which is often used alone as a stand-by treatment, the incidence of psychiatric events in prophylaxis is rare (approx. 1 in 13 000), but neurotoxicity is approximately sixty times more probable after treatment than with pro- phylactic use (19). Another drug for presumptive therapy, halofantrine, has been associated with elec- trocardiographic changes, namely prolongation of the QTc interval (20-23). In our study, none of the users experienced a serious adverse event; the ad- verse events reported were probably due to the mefloquine component of Fansimef. Such events have been reported elsewhere (24-26). The behaviour of travellers is unpredictable. Tourists despite being made aware of the urgency of malarial treatment in this study, as elsewhere (27), waited for their symptoms to resolve spontaneously. This was contrary to the prescribed advice. The majority of travellers to Asia and South America can consult a physician if they suspect malaria; only 9 (7.3%) of those with fever in our study considered themselves out of reach of medical attention. Stress should thus be placed on seeking medical attention locally, whenever possible in case of illness. The importance of immediate follow-up consultation after stand-by treatment should also be underlined to determine the accuracy of the presumptive diagnosis, the need for alternative antimalarials in cases of non- efficacy of this treatment, and other treatment in cases of erroneous self-diagnosis. Instructions to travellers should be given orally and in writing, including details of the type and severity of possible malarial symptoms and the cir- cumstances in which the use of stand-by treatment can be considered. A possible future development is a kit which will enable the traveller to test a blood droplet for possible malarial infection using a simple diagnostic technique such as the rapid manual test "paraSight-F", an antigen capture test which is simple to perform and provides a definitive answer (88.9% sensitivity) in approximately ten minutes (28).a Such a test, appropriately simplified for use by travellers, would reduce the risk of a wrong diagno- sis of malaria and enhance the strategy of presump- tive treatment. Self-diagnosis of other frequently occurring illnesses among travellers (e.g., diarrhoea) is comparatively easier because the symptomatology can be clearly defined, but in malaria the clinical symptoms are often mild and atypical. Provision of a diagnostic test would increase the cost of the SBT strategy. A small additional charge may be accept- able compared with the overall cost of travel to dis- tant places. Our results showed only one case of confirmed malaria in a population of 1572 travellers, which is similar to previously reported attack rates for low- risk areas (8). The intervention costs incurred with the single malaria case in this study were approxi- mately Sw.fr. 44000 (approximately US$ 32 000) (i.e., 1572 times the cost of the kit (Sw.fr. 23), plus 41 visits to doctors (Sw.fr. 169 each) and six addi- tional follow-up visits). The visits are at Swiss prices, which in developing countries would cost less. The benefits of this intervention strategy are the prevention of malarial complications. The approxi- mate costs incurred in various outcomes of malarial infections, including death, have- previously been estimated by Dinkel et al. (29). The addition of a simple diagnostic kit to the stand-by treatment would increase the intervention cost but would re- duce the potential for misdiagnosis and inappropriate treatment. Travel to areas with greater malaria attack rates and the possibility of carrying the kit on more than one trip (prolonged expiry dates) would enhance the cost-effectiveness of this development. In summary, the advantages of the stand-by treatment approach are the avoidance of side-effects associated with chemoprophylaxis and the life- saving potential of having presumptive therapy for falciparum infections. One constraint is the wide range of clinical presentations of malaria which makes it difficult to define simple guidelines with a Peyron F et al. Assessment of a rapid manual test for the diagnosis of Plasmodium falciparum malaria (Poster). In: Pro- ceedings of the 3rd Conference on International Travel Medi- cine, 26-29 April 1993, Paris. WHO Bulletin OMS. Vol 73 1995 219 P. Schlagenhauf et al. accurate indications for the safe and appropriate use of antimalarial drugs by travellers. As the diag- nosis of malaria is unlikely to be correct, stand-by treatment is more often overused or underused than used correctly. Incorrect use may actually be danger- ous (30). Travellers should be encouraged to use adequate anti-mosquito measures and to visit a local physician if symptoms occur. Education of the travel- ling public and increasing their understanding of malaria prevention and cure are top priorities. In addition, there is an urgent need for development of a kit containing a test for rapid self-diagnosis by evaluation of a blood droplet. The aim therefore should be refinement of the strategy with the ad- dition of a test for self-diagnosis. Acknowledgements We are grateful for the assistance given by Dr Howard Etlinger (ELISA analysis), Mr Werner Huber (PCR analy- sis), and Mr Hanspeter Jauss (graphics). Resume Aspects comportementaux de l'utilisation du traitement pr6somptif du paludisme par les voyageurs Les connaissances, les attitudes et les pratiques des voyageurs en ce qui concerne le traitement de r6serve (SBT) du paludisme sont mal docu- mentees. Cette 6tude examine le comportement d'une cohorte de voyageurs en provenance de Suisse et se rendant dans des r6gions d'end6mie palustre a faible risque, qui avaient adopt6 les mesures antimoustiques comme strat6gie de pr6- vention du paludisme et qui avaient emport6 un traitement de r6serve a utiliser en cas de besoin. Les sujets ont 616 recrut6s entre mars et novembre 1992; il leur a ete distribu6 un n6ces- saire contenant un traitement m6dicamenteux de r6serve (m6floquine, sulfadoxine et pyrimethami- ne), un thermom6tre m6dical, des informations sur le diagnostic du paludisme, et un mat6riel de pre- I6vement de sang capillaire au bout du doigt en vue de la confirmation ult6rieure de l'infection. Tous ces sujets ont 6t6 instamment invit6s a con- sulter un medecin d6s les premiers sympt6mes pouvant evoquer un paludisme. Sur les 1187 parti- cipants, 123 ont signal6 une maladie (la fi6vre 6tant le principal indicateur), avec frissons (36,6%), c6phal6es (35%), troubles digestifs (69,9%), myal- gies etlou arthralgies (41,5%). Les deux tiers des sujets tomb6s malades n'ont pas consult6 de medecin, malgre leurs sympt6mes et les conseils regus; seuls 9 d'entre eux (7,3%) etaient reel- lement hors de port6e de tout traitement m6di- cal. Six voyageurs ont pris leur traitement de reserve, mais un seul etait atteint de paludisme confirm6. Tous les utilisateurs ont consulte un medecin apres la prise du traitement pr6ventif, et deux r6actions indesirables sans gravit6 ont ete signalees. La strategie du traitement antipaludique de reserve permet d'eviter les effets ind6sirables associ6s a la chimioprophylaxie et a l'avantage d'utiliser des medicaments capables de sauver le malade en cas d'infection a falciparum. 11 est tou- tefois peu probable que le diagnostic de paludis- me soit correct, et le traitement de reserve est plus souvent sur- ou sous-utilise que correctement utilis6. Cette strategie est donc limitee par le com- portement des voyageurs, le manque de connais- sances concernant le paludisme et la difficult6 de savoir s'il faut prendre ou non le traitement. Une solution possible pourrait etre de mettre au point un n6cessaire qui permettrait au voyageur de faire lui-meme, au moyen d'une technique simple, le diagnostic de paludisme sur une goutte de sang pr6lev6e au bout du doigt. References 1. Pohn HP et al. [Infectious diseases: epidemiological situation 1989 in the Federal Republic of Germany]. Bundesgesundheitsblatt, 1991, No. 5: 199-202 (in German). 2. Ohtomo H et al. Clinical evaluation of antimalarial regimens in Japan. Zentralblatt fOr Bakteriologie, Mikrobiologie und Hygiene, Abteilung 1, Originale A, 1987, 264: 513-520. 3. Phillips-Howard P et al. Malaria in Britain: 1977-86. British medical journal, 1988, 296: 245- 248. 4. International travel and health. Vaccination require- ments and health advice. Geneva, World Health Organization, 1994. 5. Development of recommendations for the protection of short-term travellers to malaria endemic areas: Memorandum from two WHO meetings. Bulletin of the World Health Organization, 1988, 66: 177-196. 6. Centers for Disease Control. Health information for international travel. Atlanta, US Department of Health and Human Services, 1992. 7. Steffen R, Behrens RH. Travellers' malaria. Parasi- tology today, 1992, 8: 61-66. 8. Schlagenhauf P, Steffen R. Stand-by treatment of malaria in travellers: a review. Journal of tropical medicine and hygiene, 1994, 97(3): 151-160. 9. Steffen R et al. Malaria prophylaxis and self-thera- py in airline crews. Aviation space and environmen- tal medicine, 1990, Xl: 942-945. 10. Holdener F, Wyss R. [From malaria chemopro- 220 WHO Bulletin OMS. Vol 73 1995 Use of malaria presumptive treatment by travellers phylaxis to emergency self-treatment (1975-1989)]. Schweiz. Rundschau fur Medizin, 1991, 80: 53-56 (in German). 11. Centers for Disease Control. Malaria in travellers returning from Kenya: failure of self-treatment with pyrimethamine/sulfadoxine. Morbidity and mortality weekly report, 1989, 38: 363-364. 12. Raeber PA et al. Prophylaxie du paludisme 1992: cinq remarques d'actualit6. M6decine et hygiene, 1991, 49: 3214-3216. 13. Wijesundera M de S et al. Antibodies to P. falcip- arum sporozoites following a malarial outbreak in a non-endemic area of Sri Lanka. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1990, 84: 35-39. 14. Warhurst DC et al. Simplified preparation of malar- ial blood samples for polymerase chain reaction. Lancet, 1991, 337: 303-304. 15. Kain KC, Lanar DE. Determination of genetic varia- tion within Plasmodium falciparum by using enzy- matically amplified DNA from filter-paper disks impregnated with whole blood. Journal of clinical microbiology, 1991, 29: 1171-1174. 16. Ranford-Cartwright LC et al. Genetic hybrids of Plasmodium falciparum identified by amplification of genomic DNA frim single oocysts. Molecular and biochemical parasitology, 1991, 49: 239-244. 17. Foley M et al. Rapid and simple method for isolat- ing malaria DNA from fingerprick sample of blood. Molecular and biochemical parasitology, 1992, 49: 239-244. 18. Porto M et al. Second form in a segment of the merozoite surface protein 1 gene of Plasmodium vivax among isolates from Rondonia (Brazil). Molec- ular and biochemical parasitology, 1992, 54: 121-124. 19. Weinke T et al. Neuropsychiatric side-effects after the use of mefloquine. American journal of tropical medicine and hygiene, 1991, 45: 86-91. 20. Ter Kuile FO et al. Halofantrine versus mefloquine in treatment of multidrug-resistant falciparum malar- ia. Lancet, 1993, 341: 1044-1049. 21. Nosten F et al. Cardiac effects of antimalarial treat- ment with halofantrine. Lancet, 1993, 341: 1054-1056. 22. Castot A et al. Prolonged QT interval with halofan- trine (Letter). Lancet, 1993, 341:1541. 23. Drug alert: halofantrine. Change in recommenda- tions for use. Weekly epidemiological record, 1993, 68 (37): 269-270. 24. Steffen R et al. Mefloquine compared with other malaria chemoprophylactic regimens in tourists visit- ing East Africa. Lancet, 1993, 341: 1299-1303. 25. Prophylactic and therapeutic use of mefloquine. Weekly epidemiological record, 1989, 64(32): 247-248. 26. Lobel HO et al. Long-term malaria prophylaxis with weekly mefloquine. Lancet, 1993, 341: 848-851. 27. Behrens RH, Phillips-Howard PA. What do travel- lers know about malaria? Lancet, 1989, 2: 1395-1396. 28. Shiff CJ et al. The rapid manual ParaSight-F test. A new diagnostic tool for Plasmodium falciparum infection. Transactions of the Royal Society of Tropi- cal Medicine and Hygiene, 1993, 87: 646-648. 29. Dinkel R et al. [Cost-benefit analysis of malaria pro- phylaxis with mefloquine among travellers to Kenya]. Bern, Verlag Peter Lang, 1990 (in German). 30. Phillips-Howard PA et al. Stevens-Johnson syn- drome due to pyrimethamine/sulfadoxine during pre- sumptive self-therapy of malaria. Lancet, 1989, 2: 803-804. WHO Bulletin OMS. Vol 73 1995 221