Alison Krentel 1, Peter Fischer 2, Paul Manoempil 3, Taniawati Supali 4,Gérard Servais 1 and Paul Rückert 1

Transcription

1 Tropical Medicine and International Health doi: /j x volume 11 no 11 pp november 2006 Using knowledge, attitudes and practice (KAP) surveys on lymphatic filariasis to prepare a health promotion campaign for mass drug administration in Alor District, Indonesia Alison Krentel 1, Peter Fischer 2, Paul Manoempil 3, Taniawati Supali 4,Gérard Servais 1 and Paul Rückert 1 1 Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) SISKES Project, Kupang, Indonesia 2 Department of Helminthology, Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany 3 District Health Authority, Alor, Indonesia 4 Department of Parasitology, University of Indonesia, Jakarta, Indonesia Summary We report the results of two surveys of people s knowledge, attitudes and practices (KAP) regarding lymphatic filariasis (LF) in Alor District, eastern Indonesia. The results of the surveys were used to prepare and evaluate the social mobilization component of a pilot mass drug administration (MDA) in five villages. In the study area, the filarial parasites Brugia timori and Wuchereria bancrofti are highly endemic. Frequent and severe adverse reactions after MDA may occur especially in areas endemic for B. timori and therefore, a special communication strategy was designed to inform and to educate communities about LF and its control. The first KAP survey was conducted as a baseline pre-mda with diethylcarbamazine and albendazole and the second as a post-intervention evaluation in order to obtain information on the impact of the communication campaign. Before the information campaign and the subsequent MDA, 54% of the study population had heard of one of the three main terms for LF, whereas after health education and MDA, 89% had heard of at least one of the three terms. Similarly, pre-mda, 21% reported having had previously taken the treatment for filariasis, while post-mda, 88% reported having taken the treatment during the pilot treatment period. The historical fears and traumatic experiences associated with past LF treatment campaigns in Indonesia were averted since both the communication campaign and the MDA were designed appropriately for and together with the community. As a result, compliance was sufficient in the first round to successfully begin the elimination process. keywords Brugia timori, Wuchereria bancrofti, Indonesia, mass drug administration, lymphatic filariasis elimination, community mobilization Introduction Lymphatic filariasis (LF) is one of the important public health and socio-economic problems facing many countries in the developing world and is considered an indicator of poverty (Durrheim et al. 2004). In 1997, LF was targeted for elimination mainly through annual mass drug administration (MDA) to those living in endemic areas (Ottesen et al. 1997; Ottesen 2000). MDA of drugs kill the microfilariae efficiently but have only limited macrofilaricidal efficacy. In order to interrupt parasite transmission, MDA has to be sustained for a period of up to 6 years providing that a significant proportion of the community complies with treatment. As a result, communities must be informed and educated on LF and the importance of sustained compliance over this period of time. Compliance rates can be negatively affected by neglect to inform the community and consequently hinder elimination efforts (Riji 1986; Richards et al. 1995; Babu & Kar 2004). Indonesia is the only country in the world endemic for the three filarial species that cause LF: Wuchereria bancrofti, Brugia malayi, Brugia timori. It is the largest country in South-east Asia endemic for LF and much of its distribution in the country still remains unknown. However about one-third of the world s burden of human Brugia infections and a significant number of W. bancrofti infections are assumed to occur in Indonesia. In 2001, Indonesia joined the Global Programme to Eliminate Lymphatic Filariasis (GPELF) which proposes the administration of a single dose regimen of diethylcarbamazine (DEC) and albendazole in endemic areas. Due to the high prevalence, remote distribution and diversity of filarial infections, LF elimination in Indonesia can be considered one of the greatest challenges of GPELF. The national ª 2006 Blackwell Publishing Ltd 1731

2 programme uses a pilot-oriented approach, focussing on five pilot districts in the provinces of East Kalimantan, Sumatra, south-east Sulawesi, Jambi and East Nusa Tenggara (Alor District). In Alor District, LF is caused by W. bancrofti and B. timori. In some villages, the prevalence of microfilaraemics was reported to be as high as 27% and about 80% had filarial-specific IgG4 antibodies (Supali et al. 2002b). Adverse reactions resulting from treatment studies using multiple low doses of DEC for W. bancrofti or B. timori infections were reported decades ago (Partono et al. 1979, 1981; Partono & Purnomo 1985). Adverse reactions following single dose treatment with DEC alone or combined with albendazole, as recommended by WHO, have been shown to be more severe for those individuals infected with B. timori than for those with W. bancrofti (Supali et al. 2002a). In the past, adverse reactions have hindered the progress of LF elimination in Indonesia and resulted in initial hesitancy to use single dose treatment MDA as the preferred method of elimination for the national programme. The past negative effect of adverse reactions was attributed to a lack of clear social mobilization and education to the community (Oemijati 1999). To prepare a communication strategy, researchers require information on knowledge, behaviour and attitudes specific to the community where the intervention is planned. There are few studies detailing results from knowledge surveys conducted in preparation for LF control activities. Most studies deal with W. bancrofti infection, or are in single projects in India, Africa or Latin America (Eberhard et al. 1996; Ahorlu et al. 1999; Babu & Nayak 2003; el-setouhy & Rio 2003; Suma et al. 2003; Babu et al. 2004). The number of studies from South-east Asia is limited, in particular those dealing with brugian filariasis. A small study from a B. malayi-endemic area on Sumatra, Indonesia, indicated that even after the presence of an LF research team in the village, knowledge about the disease remained limited (Sudomo et al. 1993). In order to build up sustainable national filariasis elimination programmes in South-east Asia, more operational examples are needed which use baseline data in the development of health education materials for LF, especially in areas endemic for brugian filariasis where adverse reactions have been shown to be more severe. This paper reports the results of an investigation into people s knowledge, attitudes and practices regarding LF in Alor District, eastern Indonesia through the administration of two knowledge, attitudes and practice (KAP) surveys and how these results were used to prepare and evaluate the social mobilization component of a pilot MDA in five villages. For the larger Indonesian elimination programme, this study demonstrated the process of developing a simple and effective health promotion campaign to assist LF control activities, to address the possibility of adverse reactions and to stress the need for community support. This model can be applied to other parts of Indonesia as well as to other national programmes where adverse reactions are possible and where communication with the community has been neglected or underfunded. Study area and methods Study area Alor is a small district consisting of 15 islands in Nusa Tenggara Timur Province in eastern Indonesia. The largest islands of the district are Alor and Pantar. Located north of Timor Island, the terrain is rugged and volcanic. There are approximately 14 known languages (with >100 dialects reported) in the district with a population of around ( The five villages which participated in the KAP survey and pilot project included Alila Timur (population 770), Maukuru (437), Kamot (878), Probur (1650) and Pante Deere (642). These villages were selected based on their geographical location (beach, mountain, rice field) and the presence or absence of W. bancrofti and/or B. timori. The District Health Authority in Alor reported microfilaria rates using thick night blood film examinations for Alila Timur of 1.7% in 1991 (W. bancrofti), for Maukuru of 2.9% in 2001 (W. bancrofti and B. timori) and for Kamot of 9.7% in 1996 (W. bancrofti and B. timori). No microfilariae data were available for Probur; however, it is known that there are LF cases caused by W. bancrofti based on the presence of hydrocele cases in the population. Probur is also close to the B. timori-endemic area of western Alor. There were no reported LF cases in Pante Deere. Based on preliminary studies using the ICT antigen detection test for W. bancrofti infection and the Brugia Rapid IgG4 antibody test for detection of brugian filariasis, it is assumed that the actual prevalence of LF in the study villages exceeds the prevalence reported by the District Health Authority (T. Supali, P. Manoempil, P. Fischer, P. Rueckert, unpublished data, Supali et al. 2004). The general population is elementary school educated and most people are subsistence farmers. The study villages are located throughout Alor Island; three on the coastline (Alila Timur, Maukuru and Pante Deere), one in a mountainous area (Probur) and one, Kamot, in a ricegrowing plateau with swamps in and around the village. Some of the village neighbourhoods are located 1 2 h from the village centre by boat or car ª 2006 Blackwell Publishing Ltd

3 Methods A fully scheduled questionnaire was developed to be administered to respondents by trained interviewers. The same interviewers participated in both rounds of interviews. Interviewers were from the district health authority in the district capital, from the provincial capital or privately hired. To control for bias, interviewers were required to wear street clothes as opposed to the Indonesian civil service uniform. Questions were both closed and open-ended, eliciting responses about the following: (1) knowledge of LF in terms of local names, transmission, symptoms, prevention and the information source; (2) attitudes if the disease was considered to be a problem and if so, what kind of problem; and (3) practices with regard to participation in MDA and treatment-seeking behaviour for symptoms. In addition, information was requested on reported compliance with treatment and perceived adverse reactions. Two previous anthropological surveys concerning LF in Alor provided baseline information for the questionnaire (Setyawati et al. 2002a,b). The KAP questionnaire was tested for clarity and comprehension in one village of Alor District, Fanating, where B. timori is responsible for LF cases (Supali et al. 2002b). Changes to the questionnaire were made accordingly. The questionnaire and the educational material were written in Bahasa Indonesia, the national language of Indonesia. Interviews were conducted at the household level. Respondents were given the opportunity to refuse participation and were ensured confidentiality. Only those respondents above the age of 15 years were interviewed to ensure respondent comprehension of the questions. Furthermore, symptoms of LF generally appear during adolescence (Witt & Ottesen 2001). The sample size was calculated from the estimated population above 15 years in the five villages (2629 persons) using 5% confidence interval and the population parameter set at 0.5. The total number of respondents required was 335 persons; however, 40 additional respondents were added to take into account possible refusals and missing information. The samples were divided per village according to the proportion of village population to the total reference population. For example, the population of Kamot is 18% (473 persons) of the total reference population (2629 persons); therefore, 18% (66 persons) of the samples were interviewed in Kamot. An interview team was trained over a 2-day period on questionnaire administration. The villages were divided into clusters. One interviewer was responsible for 15 interviews within each cluster. Following the Expanded Programme on Immunisation (EPI) (Henderson & Sundaresan 1982) sampling method, an interviewer would start by throwing a pen in the centre of the village to show the direction of the first interview. He would continue in the direction of the nearest household and interview one person per household, continuing in the same direction until he had finished 15 interviews. The person interviewed in each household was not necessarily the head of the household, but was chosen randomly controlling for gender. If no one was present in the next house, then the interviewer proceeded to the next house. Each interviewer was responsible for interviewing seven men and seven women and the fifteenth person from either gender to ensure an even gender distribution of interviewees. After the first round of the KAP survey, a comprehensive health promotion campaign using locally specific materials was developed from the baseline information. A variety of materials (film, brochure, song, flip chart and poster) was selected for development in order to provide information for all levels of education, all religions and all levels of literacy. The materials were coordinated so that colours, images and fonts were the same, thus providing uniformity in the campaign. Technical information on LF was checked by parasitologists working in Alor District and simplicity of terms and concepts were maximized. All materials were checked with community leaders and members before production and distribution. Health staff trained community volunteers (drug distributors and village leaders) who were responsible for conducting a village census and for disseminating information about LF, the upcoming treatment and the possibility of adverse reactions. The MDA followed within 10 days of the completion of the information campaign. The health workers in collaboration with the villagers determined the most appropriate way to set up drug delivery in their village house to house or multiple treatment posts. Thereafter, MDA was performed using a single dose DEC (based on 6 mg/kg translated into agebased dosing table for simplicity) combined with albendazole (400 mg) which was previously reported to be effective in Alor District (Supali et al. 2002a; Fischer et al. 2003, 2004; Oqueka et al. 2005). Village drug distributors were given analgesics, anti-pyretic and antihistamines and a simplified dosage table to administer for adverse reactions. In those villages with B. timori, health staff remained in the village on the first night to relieve the apprehensions of the villagers and to assist if more severe adverse reactions appeared. To conduct the post-intervention KAP survey within the shortest recall period possible to get detailed information of self-reported adverse reactions, the KAP survey was re-administered in the pilot villages within 2 weeks of the MDA, leaving minimal time for mopping up those persons who were out of the village during the drug distribution. ª 2006 Blackwell Publishing Ltd 1733

4 Table 1 Demographics of the sample population in five pilot villages on Alor Island, Indonesia, for rounds 1 (February 2002, before MDA) and 2 (June 2002, after MDA) of the filariasis KAP survey Methods for data analysis were identical for both the pre-mda round (round 1) and the post-mda round (round 2). For both sets, data were double entered into Epi Info 6 by two teams of two data entry assistants. Analysis was done using chi-square tests for significance. After data cleaning, analysis was performed on 375 respondents from the first round and 386 respondents from the second round of interviews. Ethical approval for the KAP surveys was given by the District Health Authority of Alor District. Results Demographics of rounds 1 and 2 Round 1 Round 2 Has completed primary 72% (270/375) 71% (275/386) education Christian 84% (315/375) 88% (340/386) Owns a radio 17% (63/375) 16% (62/386) Owns a TV <1% (3/375) 2.3% (9/386) Has a metal toll roof 54% (204/375) 58% (223/386) Has a house with bamboo walls 39% (146/375) 31% (121/386) Round 1 comprised 375 respondents, specifically 190 (51%) males and 185 (49%) females and round 2 had 386 respondents with the same gender distribution. Demographically both rounds showed similar distributions for most variables (Table 1). Age distribution for both rounds was also similar with 19% in round 1 between the 15 and 25 years and 13% in round 2; 34% and 36% between 26 and 35 years; 23% in both rounds between 36 and 45 years and 24% in both rounds between 46 and 90 years. Round 1 Awareness of filariasis and local names. In addition to the more scientific term filaria, two other common Indonesian names for LF exist: kaki gajah which literally translates to elephant leg and boa besar which means large fruit, referring to hydrocele. Most people had heard of kaki gajah compared with the other names (v 2 ¼ 30.66, P < 0.001) (Table 2). In total, 207 respondents (55%) had heard of one of the three names for the disease; 19% of the respondents answered that there was another name for LF other than these three primary names and listed 23 other names for the disease, confirming the linguistic richness of the district. Knowledge. Those respondents who had heard of one of the three names for LF (n ¼ 207) gave the following causes: germs (27%), hard work (8%), worms in the blood (16%), mosquitoes (11%) and mud in the rice field (10%). Other causes listed were women, swamps and genetic. These results echoed the earlier findings of two anthropological studies (Setyawati et al. 2002a,b). There is some degree of interchangeability between cause and transmission as respondents also listed mosquitoes as causing filariasis; however, about one-third also cited them as transmitting filariasis. This indicates a lack of detailed and precise knowledge about the disease itself. When asked to identify symptoms of LF, 50% listed elephantiasis of the leg, 49% listed fever and 31% were able to identify swelling in the genital area as a symptom. Attitudes. Sixty per cent of the respondents perceived LF to be a problem. These respondents (n ¼ 124) further described the problem as economic (defined as inability to work, loss of job, expenditure for medication and/or transport to health facility), medical (defined as pain, fever, in bed, itching, inability to walk) and social (defined as inability to interact with the community, rejected by peers, desertion by family/husband/wife ); 51% perceived Table 2 Results concerning general knowledge about filariasis from rounds 1 (February 2002) and 2 (June 2002) of the KAP survey Round 1 Round 2 v 2, P-value for difference of proportions (d.f. ¼ 1) Has heard of filariasis 20% (76/375) 89% (344/386) , P < Has heard of kaki gajah 38% (143/375) 92% (353/386) , P < Has heard of boa besar 26% (96/375) 81% (312/386) , P < Knows filariasis is caused by worms in the blood 16% (34/207*) 67% (242/359*) , P < Knows filariasis is transmitted by mosquitoes 33% (69/207*) 77% (277/359*) , P < Knows at least three symptoms of filariasis 37% (77/207*) 71% (253/359*) 61.97, P < Knows that infection with filariasis can be prevented 57% (119/207*) 97% (348/359*) , P < *Number of individuals who know at least one name for filariasis ª 2006 Blackwell Publishing Ltd

5 LF to be an economic problem, whilst 24% cited it a medical problem and 19% as a social problem. The remaining 6% did not specify the kind of problem they perceived LF to be. Treatment and prevention. About 49% of respondents agreed that LF could be treated and 23% reported that they had previously been treated for LF; 57% of those who had heard of LF agreed that the disease could be prevented. Most associated LF prevention with avoiding mosquitoes; 31% stated that it can be prevented by cleaning the mosquito breeding ground and 27% stated prevention is by avoiding mosquito bites. A further 30% thought that drinking medication would prevent filariasis. The relatively low percentages for all known methods for prevention indicate a lack of knowledge on prevention of mosquito bites and the treatment of filariasis as methods for addressing the disease. Development and implementation of communication strategy Based on the findings from the first round of interviews, a comprehensive communication strategy was developed to be specific to the people of Indonesia, in particular the eastern region. The following conclusions were drawn from the first KAP survey for the development of the health promotion campaign. (1) Due to the low education level in the villages, simple language is imperative. (2) Include many pictures and graphics in order to grab attention. (3) Add visual and oral forms of media in order to reach those who may not be able to read. (4) Traditional forms of mass media television, radio and newspaper are inaccessible to most people. (5) Basic information about LF must be included, as 45% had not even heard of the disease and do not know how it is transmitted. (6) Basic information is necessary on the prevention of mosquito bites in order to encourage the reduction of morbidity from other mosquito vector-borne diseases. (7) Basic information must be given on what can be done for patients with chronic manifestations (i.e. regular washing of affected area, etc.). (8) As LF is perceived as an economic and social problem rather than a merely medical one, these non-medical issues should be addressed. Campaign materials were developed out of these recommendations. Following discussion with villagers and the health authority, the campaign slogan Berantas Filaria (Eliminate Filaria) was chosen, with Alor Sehat 2010 (Alor Healthy 2010) written underneath. The term filaria was introduced for the campaign, instead of using the scientific term penyakit filaria (the disease filariasis) or the terminology previously used by the national campaign kaki gajah (elephant leg). As both hydrocele and elephantiasis of the leg exist in Indonesia, LF campaigns should address both manifestations. The strategy for the communication campaign was developed by researching past awareness activities used for public health campaigns in Alor, existing health surveys on Alor (health facility and medical anthropology) and discussions with key persons in civil society and government. The planning team learned from the communication and programmatic experiences of an ongoing parasitological survey which treated B. timori in several communities in Alor. The combination of the low capacity of the health facilities, the high potential for adverse reactions, the remote location of many villages, the potential for economic loss during treatment and the endemic nature of the disease in Alor signalled the need for an awareness campaign rooted in the community. The district health staff would not be the only channel to educate the population about LF, administer treatment or treat adverse reactions; the community would take an active role. In order to support the campaign, an 18-min film was produced with images from Alor District and from a World Health Organization film on LF. The film addressed the cause, transmission, symptoms and signs, consequences of the disease, prevention of mosquito bites, the international and national LF campaign, case management of lymphoedema, those eligible for MDA and promotion by key community leaders (local district regent and religious leaders), and the social and economic consequences of the disease. A song was created in collaboration with a local music group. It used a traditional style of music and was used to call people together for a traditional dance associated with this kind of music (poco-poco). Because the main chorus repeated the name filaria over and over again, familiarity with the term was created. The cassette was distributed to local minibus drivers and to health centres. The oral educational materials, as outlined above, were supplemented by a flipchart, poster, brochure and sticker. The language was simple and many pictures were included. The written materials explained that the future of their children is in their hands by stating that children who were born at the start of the mass treatment would not be at risk for LF, so participation of the whole community was required to protect the future health of the children. Following the development of the materials, it was necessary to test them in order to make sure that they were clear and attractive. A series of focus group discussions was conducted in two villages (men and women were tested separately) as well as one focus group discussion with civil society members. In addition, 12 interviews were taken with randomly selected people on the streets in the district ª 2006 Blackwell Publishing Ltd 1735

6 Table 3 Knowledge about lymphatic filariasis by sex as determined by KAP surveys from rounds 1 and 2 Round 1 (February 2002) Round 2 (June 2002) M(n ¼ 190) F (n ¼ 185) v 2, P-value for difference of proportions (d.f. ¼ 1) M (n ¼ 199) F (n ¼ 187) v 2, P-value for difference of proportions (d.f. ¼ 1) Has heard of filariasis 50 (26%) 26 (14%) 8.07, P ¼ (91%) 164 (88%) 0.77, P ¼ Has heard of kaki gajah 81 (43%) 62 (34%) 3.28, P ¼ (94%) 167 (89%) 3.33, P ¼ (elephant leg) Has heard of boa besar (big fruit or ball) 59 (31%) 37 (20%) 6.01, P ¼ (84%) 144 (77%) 2.94, P ¼ capital. Finally parasitologists from the University of Indonesia ensured that the materials were scientifically correct. Some of the comments received during the material testing resulted in changes to the campaign. For example, respondents requested that words such as prevalence, per cent and epidemic be removed and replaced with easier concepts. Many liked the photos and pictures and found them clear. One man who was illiterate explained that although he could not read or write, someone could explain about LF to him using the pictures in the brochure. Others commented on the potentially shocking nature of seeing photographs of genitalia, however during the group discussion, people agreed that as these images were for health, it was permissible to show them. The focus group discussions provided an opportunity for campaign planners to discuss components of the MDA and the awareness strategy with community members and stakeholders. For example, issues about adverse reactions were raised as well as the importance of providing post- MDA medical care in villages to reduce fear. Civil society members were enlisted to support the campaign during these discussions. In terms of timing, strategy and concept development (design, testing and revision) took 10 weeks, preparation of the campaign and drug distribution [training of health staff and community health workers (CHW)] took 3 weeks and the campaign itself took 5 weeks, including mopping up of those persons who missed the initial drug distribution. Round 2 in comparison with the baseline. In order to evaluate the mass media developed to support the MDA, a second round of KAP survey was administered shortly after the campaign s dissemination and the MDA (approximately 15 weeks after round 1). The populations of the two surveys did not show any difference with respect to their socio-demographic variables (Table 1). As seen in Table 2, there are clear improvements in knowledge after the implementation of the communication campaign and the MDA that followed. There was an improvement in overall awareness about LF, in particular the introduced term filaria rather than the two more common local names. Furthermore, respondents reported increased knowledge on cause, transmission and symptoms of disease. There was no difference between men and women in awareness about LF in round 2, as there had been in round 1 (Table 3). This suggests that the communication campaign reached the intended population equally in both genders. Round 2 showed a shift in attitudes, whereas 60% felt filariasis to be a problem in round 1, 89% did in round 2 (v 2 ¼ 66.07, P < 0.001). This implies that respondents internalized the knowledge communicated about LF. Round 2 treatment and adverse reactions. In round 2, 359 of 386 individuals had heard of one of the names for LF and therefore completed the entire questionnaire. About 89% of those interviewed reported taking the required treatment for filariasis. Of the 321 persons reporting compliance, 98% reported taking the treatment in 2002, the same year as the trial in the pilot villages; 90% reported experiencing some form of adverse reaction to the treatment (Table 4). Of the 290, 75% sought relief for their side effects from the CHW, who were trained to administer anti-pyretic drugs and antihistamines, Table 4 Self-reported adverse reactions reported by KAP survey respondents in round 2 (June 2002, N ¼ 290*) following a single dose diethylcarbamazine combined with albendazole to eliminate lymphatic filariasis Round 2 Fever (no thermometer used) 152 (52%) Headache or dizziness 151 (52%) Queasiness 119 (41%) Itching 92 (32%) *290/321 (90%) of those who complied with treatment reported adverse reactions ª 2006 Blackwell Publishing Ltd

7 vs. 21% who sought relief from CHW in the first round (v 2 ¼ 55.39, P < 0.001). Discussion In the past, efforts to eliminate LF in Indonesia have been negatively affected due to lack of communication about adverse reactions to the community. Partono and Purnomo (1985) reported results from one study where low dosage of DEC for 2 weeks was given by the heads of household to community members in West Flores, eastern Indonesia. The authors listed that one of the major factors which may have accounted for the unsatisfactory results of the study included the presence of adverse reactions which led to non-compliance and the ignorance of the community with regard to LF. Based on the results of this study, the authors recommend that adverse reactions could be effectively handled by good communication with the population. In other studies when communities were well informed, adverse reactions did not affect compliance rates. As Partono et al. (1979) reported in another study in West Flores Indonesia, 88% of those treated showed adverse reactions; however, no one dropped out of the treatment programme because the community had been well informed about the possibility of adverse reactions following treatment. The experiences described in Alor District showed a similar result. After the communication campaign, communities were well informed about treatment and possible adverse reactions, and 89% of the examined, eligible population reported compliance with treatment despite understanding the possibility for adverse reactions. This is confirmed by the results of the MDA for the remaining villages on Alor Island where 79% of the total population ( ) complied with treatment. The district health authority prepared a similar communication and drug distribution campaign modelled after the pilot project. If communities are reminded at the beginning of each round of MDA about the nature and onset of adverse reactions then there is less of a collective panic in the village when adverse reactions do occur. In fact, anecdotal evidence received during the pilot project showed that some villagers interpreted the presence of adverse reactions as a sign that the treatment was effective. Communities should be well informed before MDA begins about the possibility of adverse reactions; especially as they are likely to be the most severe in the first treatment round as microfilaria rates are the highest (Supali et al. 2002a). In subsequent rounds of treatment, awareness prevents community members refusing participation because of fear of continuing adverse reactions. McLaughlin et al. (2003) document the adverse reactions reported in a cohort of people in Haiti. If community members were not informed at the onset of the drug administration, then painful adverse reactions or even rumours of severe reactions discouraged some community members from participating in future treatment rounds. A similar result was found by Babu et al. (2006), where the possibility of adverse reactions discouraged people in India from taking the pills. In Alor District where B. timori is present and adverse reactions more severe, it was particularly important that the community understood what might happen before anyone took the treatment and that the presence of adverse reactions signalled that the drug was working. Our study was the first KAP survey in South-east Asia to precede a health promotion campaign and MDA and thereafter. The results of round 1 are in line with results from studies in other areas with a different socio-demographic background. More people knew about the main symptoms of LF such as lymphoedema and hydrocele than about the disease name filariasis (Eberhard et al. 1996). Male individuals appear to be more aware of the symptoms, especially hydrocele than females (Babu et al. 2004). In most communities it is not widely known that mosquitoes transmit the disease agent and very fewer individuals know that worms in the blood cause the disease (Eberhard et al. 1996; Babu et al. 2004; Das et al. 2005). Some studies even do not strictly differentiate between vector and disease agent (Ramaiah et al. 1996; Das et al. 2005). We observed in our study population some interchangeability between transmission and cause of LF. A study from Haiti indicated that knowledge about the symptoms of LF, its transmission and control has a positive impact on the compliance of MDA (Mathieu et al. 2004). This is supported by our results where 89% of the individuals who knew about filariasis and completed the questionnaire also reported participation in MDA. We have established that it is essential to provide the community with basic information about treatment and the possibility of adverse reactions; however, it is also important how the communication campaign is conducted. The KAP survey in round 1 and previous anthropological studies provided the research team with concrete information about the level of knowledge about LF in the community and directed the communication campaign to those gaps in knowledge in the community. Additionally, the baseline information indicated what forms of media would best suit such a rural and under-educated population. As a result, the multimedia communication strategy was appropriate and easy to understand. Complicated information on transmission cycles and clinical manifestations was not necessary, and the campaign used instead simple messages outlining the names of the disease, what kind of treatment was to be ª 2006 Blackwell Publishing Ltd 1737

8 administered, the possibility of adverse reactions and the importance of total community compliance. The effect of the campaign is evidenced by the rise in knowledge levels and in compliance levels between rounds 1 and 2. The campaign also reached men and women equally, righting the previous gender imbalance in knowledge about the disease. The open nature of the communication strategy at the community level meant that men and women were both exposed to the same messages. This may have been the first time some women had seen a hydrocele and connected it with filariasis. The time period between the campaign and its evaluation was short and in larger district or national level campaigns, evaluation is likely to take longer to implement. However, there are advantages to evaluating as soon after MDA as possible in order to maximize people s recall for this year s activities. This is especially noteworthy as governments plan their second, third and fourth rounds of drug distribution and awareness activities. As this campaign was a pilot project for the national Indonesian LF Elimination Programme, the following recommendations for assessing community knowledge, attitudes and practices and for development of locally specific communication strategies were made to the central and provincial levels: 1 Keep health promotion and communication strategies simple and with minimal technical information. Focus on how treatment will affect the individual and the community and the importance of total community compliance. 2 Use both village leaders and health staff in communication and drug distribution activities. 3 Deal with miscommunication and confusion in the community immediately and tackle any rumours. 4 Promote the elimination of intestinal helminths as a positive side effect (Oqueka et al. 2005). 5 Ensure that one member of the health staff remains in those villages with higher microfilaria rates during the first two nights after treatment to reassure the community in the event of adverse reactions. 6 Make certain that the MDA follows swiftly after the communication campaign so that momentum and motivation from the information will not be lost. Subsequent drug administration programmes in Indonesia have been based on these recommendations. Control programmes for LF have been in operation in Indonesia for nearly 30 years using a low dosage of DEC as established by Partono et al. (1979); however, the number of LF infections is still among the highest in South-east Asia. Due to the diversity of filarial parasites and the high microfilaria densities in many areas it appeared to be necessary to evaluate the WHO-recommended single dose treatment strategy using DEC combined with albendazole in Indonesia before following this strategy in the national filariasis control programme and joining GPELF (Supali et al. 2002a). This strategy is more promising for filariasis elimination, but in the light of adverse reactions, the long duration of the MDA and the necessary participation of asymptomatics, health promotion and social mobilisation are crucial for the success and sustainability of the intervention. This study used a simple and reliable method (KAP survey) to design and evaluate a multimedia health promotion campaign. The results are an essential milestone in the Indonesian LF campaign by demonstrating that a communication approach which is adequate, well designed and uses appealing communication materials is possible, feasible and affordable. These results assisted in the policy change by the Indonesian government to accept the use of DEC and albendazole through MDA in endemic areas, in particular where B. timori is present. Although it is well documented that health promotion is essential for the success of MDA for the control of filarial parasites (Riji 1986; Richards et al. 1991, 1995; Ramaiah et al. 1996), it is unfortunately often not prioritized in programme budgets. In order to achieve the elimination goals set for 2020 both within Indonesia and globally, health promotion campaigns based on empirical information should be emphasized in all elimination programmes to ensure community acceptance and compliance over the long term. Acknowledgements We are grateful to the leaders and the inhabitants of the five villages who participated in the survey, and also thank the health staff of the District Health Authority in Alor who participated in all levels of the survey. The study received support from funds of the Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) SISKES Project, Kupang, Indonesia. References Ahorlu CK, Dunyo SK, Koram KA, Nkrumah FK, Aagaard-Hansen J & Simonsen PE (1999) Lymphatic filariasis related perceptions and practices on the coast of Ghana: implications for prevention and control. Acta Tropica 73, Babu BV & Kar SK (2004) Coverage, compliance and some operational issues of mass drug administration during the programme to eliminate lymphatic filariasis in Orissa, India. Tropical Medicine and International Health 9, ª 2006 Blackwell Publishing Ltd

9 Babu BV & Nayak AN (2003) Footcare among lymphoedema patients attending a filariasis clinic in South India: a study of knowledge and practice. Annals of Tropical Medicine and Parasitology 97, Babu BV, Hazra RK, Chhotray GP & Satyanarayana K (2004) Knowledge and beliefs about elephantiasis and hydrocele of lymphatic filariasis and some socio-demographic determinants in an endemic community of Eastern India. Public Health 118, Babu BV, Rath K, Kerketta AS, Swain BK, Mishra S & Kar SK (2006) Adverse reactions following mass drug administration during the Programme to Eliminate Lymphatic Filariasis in Orissa State, India. Transactions of the Royal Society of Tropical Medicine and Hygiene 100, Das D, Kumar S, Dash AP & Babu BV (2005) Knowledge of lymphatic filariasis among the population of an endemic area in rural Madhya Pradesh, India. Annals of Tropical Medicine and Parasitology 99, Durrheim DN, Wynd S, Liese B & Gyapong JO (2004) Editorial: Lymphatic filariasis endemicity an indicator of poverty. Tropical Medicine and International Health 9, Eberhard ML, Walker EM, Addiss DG & Lammie PJ (1996) A survey of knowledge, attitudes, and perceptions (KAPs) of lymphatic filariasis, elephantiasis, and hydrocele among residents in an endemic area in Haiti. American Journal of Tropical Medicine and Hygiene 54, el-setouhy MA & Rio F (2003) Stigma reduction and improved knowledge and attitudes towards filariasis using a comic book for children. Journal of the Egyptian Society of Parasitology 33, Fischer P, Djuardi Y, Ismid IS et al. (2003) Long-lasting reduction of Brugia timori microfilariae following a single dose of diethylcarbamazine combined with albendazole. Transactions of the Royal Society of Tropical Medicine and Hygiene 97, Fischer P, Supali T & Maizels RM (2004) Lymphatic filariasis and Brugia timori: prospects for elimination. Trends in Parasitology 20, Henderson RH & Sundaresan T (1982) Cluster sampling to assess immunization coverage: a review of experience with a simplified sampling method. Bulletin of the World Health Organization 60, Mathieu E, Lammie PJ, Radday J et al. (2004) Factors associated with participation in a campaign of mass treatment against lymphatic filariasis, in Leogane, Haiti. Annals of Tropical Medicine and Parasitology 98, McLaughlin SI, Radday J, Michel MC et al. (2003) Frequency, severity, and costs of adverse reactions following mass treatment for lymphatic filariasis using diethylcarbamazine and albendazole in Leogane, Haiti, American Journal of Tropical Medicine and Hygiene 68, Oemijati S (1999) Current Situation of Filariasis in Indonesia and its Control. World Health Organization, Jakarta. Internal Report. Oqueka T, Supali T, Ismid IS et al. (2005) Impact of two rounds of mass drug administration using diethylcarbamazine combined with albendazole on the prevalence of Brugia timori and of intestinal helminths on Alor Island, Indonesia. Filaria Journal 4, 5. Ottesen EA (2000) The global programme to eliminate lymphatic filariasis. Tropical Medicine and International Health 5, Ottesen EA, Duke BO, Karam M & Behbehani K (1997) Strategies and tools for the control/elimination of lymphatic filariasis. Bulletin of the World Health Organization 75, Partono F & Purnomo (1985) Combined low dosage and short term standard dose treatment with diethylcarbamazine to control Timorian filariasis. Acta Tropica 42, Partono F, Purnomo & Soewarta A (1979) A simple method to control Brugia timori by diethylcarbamazine administration. Transactions of the Royal Society of Tropical Medicine and Hygiene 73, Partono F, Purnomo, Oemijati S & Soewarta A (1981) The long term effects of repeated diethylcarbamazine administration with special reference to microfilaraemia and elephantiasis. Acta Tropica 38, Ramaiah KD, Kumar KN & Ramu K (1996) Knowledge and beliefs about transmission, prevention and control of lymphatic filariasis in rural areas of south India. Tropical Medicine and International Health 1, Richards F, Klein RE, Gonzales-Peralta C et al. (1991) Knowledge, attitudes and perceptions (KAP) of onchocerciasis: a survey among residents in an endemic area in Guatemala targeted for mass chemotherapy with ivermectin. Social Science and Medicine 32, Richards FO, Jr, Klein RE, Gonzales-Peralta C et al. (1995) Knowledge, attitudes and practices during a community-level ivermectin distribution campaign in Guatemala. Health Policy and Planning 10, Riji HM (1986) Comparison of knowledge on filariasis and epidemiologic factors between infected and uninfected respondents in a Malay community. Southeast Asian Journal of Tropical Medicine and Public Health 17, Setyawati I, Fina A, Liklikwatil E & Padu K (2002a) Final Report: Anthropological Study of Lymphatic Filariasis in Alor District: Perceived Causes, Symptoms and Treatments. German Agency for Technical Cooperation (GTZ) SISKES Project, Kupang. Internal Report. Setyawati I, Liklikwatil E & Padu K (2002b) Research Report: Community Perceptions of Health Problems: Perceived Causes and Treatment Pattern East Pantar, Alor. German Agency for Technical Cooperation (GTZ) SISKES Project, Kupang. Internal Report. Sudomo M, Kasnodiharjo & Oemijati S (1993) Social and behavioral aspects of filariasis transmission in Kumpeh, Jambi, Indonesia. Southeast Asian Journal of Tropical Medicine and Public Health 24 (Suppl. 2), Suma TK, Shenoy RK & Kumaraswami V (2003) A qualitative study of the perceptions, practices and socio-psychological suffering related to chronic brugian filariasis in Kerala, southern India. Annals of Tropical Medicine and Parasitology 97, ª 2006 Blackwell Publishing Ltd 1739

10 Supali T, Ismid IS, Ruckert P & Fischer P (2002a) Treatment of Brugia timori and Wuchereria bancrofti infections in Indonesia using DEC or a combination of DEC and albendazole: adverse reactions and short-term effects on microfilariae. Tropical Medicine and International Health 7, Supali T, Wibowo H, Ruckert P et al. (2002b) High prevalence of Brugia timori infection in the highland of Alor Island, Indonesia. American Journal of Tropical Medicine and Hygiene 66, Supali T, Rahmah N, Djuardi Y, Sartono E, Ruckert P & Fischer P (2004) Detection of filaria-specific IgG4 antibodies using Brugia Rapid test in individuals from an area highly endemic for Brugia timori. Acta Tropica 90, Witt C & Ottesen EA (2001) Lymphatic filariasis: an infection of childhood. Tropical Medicine and International Health 6, Corresponding Author Alison Krentel, Department of Public Health and Policy, Public and Environmental Research Unit, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK. Tel.: +44 (207) ; Fax: +44 (207) ; alison.krentel@lshtm.ac.uk Utilisation des connaissances, attitudes et pratiques des surveillances sur la filariose lymphatique pour la préparation d une campagne de promotion de l administration en masse de médicament dans le district de Alor en Indonésie Nous rapportons les résultats de deux études sur les connaissances, les attitudes et les pratiques des gens en rapport avec la filariose lymphatique (FL) dans le district de Alor dans le sud de l Indonésie. Les résultats des ces études ont été utilisés pour préparer et évaluer le volet mobilisation sociale d un projet pilote d administration en masse de médicament dans cinq villages. Dans la région étudiée, les filaires Brugia timori et Wuchereria bancrofti sont fortement endémiques. Des réactions adverses fréquentes et sévères surviennent principalement après une administration en masse de médicament dans les régions endémiques pour B. timori. Dès lors, une stratégie spéciale de communication a été conçue pour informer et éduquer les communautés au sujet de la FL et de son contrôle. La première étude a été menée à partir de données de base avant l administration en masse de diethylcarbamazine et d albendazole. La seconde étude a été une évaluation post-intervention afin d obtenir des informations sur l impacte de la campagne de communication. Avant la campagne d information et les administrations en masse de médicaments subséquentes, 54% de la population étudiée avait déjà entendu parler d au moins un des trois termes principaux sur la FL contre 89% de la population après une éducation de santé et une administration massive de médicament. De la même façon, avant l administration en masse de médicament, 21% de la population rapportait avoir déjà suivi le traitement pour la filariose durant l étude pilote contre 88% après l administration en mass de médicament. Les craintes historiques et les expériences traumatiques associées aux campagnes précédentes pour le traitement de FL en Indonésie ont été rectifiées puisque la campagne de communication tout comme l administration en masse de médicament a été conçue de façon appropriée pour et avec la communauté. Comme résultat, l adhésion était suffisante dans le 1 er tour et le processus d élimination a commencé avec succès. mots clés Brugia timori, Wuchereria bancrofti, Indonésie, administration en masse de médicament, élimination de la filariose lymphatique, mobilisation de la communauté Utilización de encuestas de conocimiento, actitudes y prácticas sobre filariasis linfática, en la preparación de una campaña de promoción de la salud para la administración masiva de medicamentos en el Distrito de Alor, Indonesia Reportamos los resultados de dos encuestas sobre los conocimientos, actitudes y prácticas (CAP) acerca de la filarialisis linfática (FL) en el Distrito de Alor, al este de Indonesia. Los resultados de las encuestas se utilizaron para preparar y evaluar el componente de movilización social de una administración masiva de medicamentos (AMM) en cinco poblados. En el área de estudio los parásitos Brugia timori y Wuchereria bancrofti son altamente endémicos. Después de la AMM pueden ocurrir con frecuencia efectos adversos severos, especialmente en áreas endémicas para B. timori, y por lo tanto se diseñó una estrategia de comunicación especial para informar y educar las comunidades sobre la FL y su control. La primera encuesta de CAP se realizó como una línea de base pre-amm con dietilcarbamazina (DCM) y albendazol y la segunda como una evaluación post-intervención, con el fin de obtener información sobre el impacto de la campaña de comunicación. Antes de la campaña de información y la AMM, un 54% de la población había oído hablar de alguno de los tres principales términos de FL, mientras que después de la campaña educativa y la AMM, el 89% había escuchado al menos uno de los tres términos principales. De forma similar, pre-amm un 21% reportaban haber tomado previamente tratamiento para la filariasis, mientras que post-amm un 88% reportó haber tomado tratamiento durante el periodo de pilotaje. Se logró evitar los miedos históricos y las experiencias traumáticas asociadas a campañas pasadas de tratamiento para la FL en Indonesia, puesto que tanto la campaña de comunicación como la AMM se diseñaron apropiadamente para y con la comunidad. Como resultado, el cumplimiento fue suficiente durante la primera ronda como para iniciar con éxito el proceso de eliminación. palabras clave Brugia timori, Wuchereria bancrofti, Indonesia, administración masiva de medicamentos, eliminación filariasis linfática, movilización comunitaria 1740 ª 2006 Blackwell Publishing Ltd