Knowledge and practice regarding dengue and chikungunya: a cross-sectional study among Healthcare workers and community in Northern Tanzania

Transcription

1 Tropical Medicine and International Health doi: /tmi volume 22 no 5 pp may 2017 Knowledge and practice regarding dengue and chikungunya: a cross-sectional study among Healthcare workers and community in Northern Tanzania Debora C. Kajeguka 1, Rachelle E. Desrochers 2, Rose Mwangi 1, Maseke R. Mgabo 1,3, Michael Alifrangis 4, Reginald A. Kavishe 1, Franklin W. Mosha 1 and Manisha A. Kulkarni 5 1 Faculty of Medicine, Kilimanjaro Christian Medical University College, Moshi, Tanzania 2 HealthBridge, Ottawa, ON, Canada 3 Institute of Rural Development Planning, Dodoma, Tanzania 4 Centre for Medical Parasitology, University of Copenhagen, Copenhagen, Denmark 5 School of Epidemiology, Public Health & Preventive Medicine, University of Ottawa, Ottawa, ON, Canada Abstract objective To investigate knowledge and prevention practices regarding dengue and chikungunya amongst community members, as well as knowledge, treatment and diagnostic practices among healthcare workers. method We conducted a cross-sectional survey with 125 community members and 125 healthcare workers from 13 health facilities in six villages in the Hai district of Tanzania. A knowledge score was generated based on participant responses to a structured questionnaire, with a score of 40 or higher (of 80 and 50 total scores for community members and healthcare workers, respectively) indicating good knowledge. We conducted qualitative survey (n = 40) to further assess knowledge and practice regarding dengue and chikungunya fever. results 15.2% (n = 19) of community members had good knowledge regarding dengue, whereas 53.6%, (n = 67) of healthcare workers did. 20.3% (n = 16) of participants from lowland areas and 6.5% (n = 3) from highland areas had good knowledge of dengue (v 2 = 4.25, P = 0.03). Only 2.4% (n = 3) of all participants had a good knowledge score for chikungunya. In the qualitative study, community members expressed uncertainty about dengue and chikungunya. Some healthcare workers thought that they were new diseases. conclusion There is insufficient knowledge regarding dengue and chikungunya fever among community members and healthcare workers. Health promotion activities on these diseases based on Ecological Health Mode components to increase knowledge and improve preventive practices should be developed. keywords dengue, chikungunya, knowledge, preventive, diagnostic, practices Introduction Dengue and chikungunya viruses are a major public health problem in many tropical regions of the world, including Tanzania, where Aedes aegypti is the primary vector [1, 2]. However, as dengue and chikungunya present with an undifferentiated febrile illness, it is likely that clinicians usually overlook the possibility of infections other than malaria in patients presenting with fever. In Tanzania overdiagnosis of malaria [3 6] involves both the prescription of antimalarials to patients without evidence of malaria parasitaemia and the frequent absence of treatment for alternative causes of disease [3]. Dengue was documented for the first time in Tanzania in 2010 [7], and again recently in 2013 and 2014, with two outbreaks occurring in Dar Es Salaam city [8, 9]. Chikungunya was first isolated in Tanzania in 1952 [10], yet few studies have been conducted to document active chikungunya cases in northern Tanzania [5,11,12] and no information is available on knowledge of chikungunya among healthcare workers (HCWs) and community members. Current but not acute dengue and chikungunya exposure has been reported in Hai district, Kilimanjaro [12]. Measured Immunoglobulin M (IgM) towards dengue and chikungunya specific antigens using commercial kits was reported to be 1.1% and 13.8%, respectively [12]. Hertz 2017 John Wiley & Sons Ltd 583

2 et al. documented the presence of A. aegypti adult mosquitoes in two areas in Hai district (Boma Ng ombe and Machame), thus suggesting that A. aegypti is the primary local vector of dengue and chikungunya in the area [1]. Successful measures of disease prevention and control depend on community knowledge and health seeking behaviour [13] as well as HCWs practices. Therefore, information on knowledge and practices regarding dengue and chikungunya prevention, diagnosis and treatment seeking behaviours are needed. WHO recommends that sustainable prevention and treatment of disease requires evaluation of social, cultural and community practices leading to spread of disease [14]. Arbovirus infections have been explained by socio-ecological models (SEM) [13, 15 19]. In this regard, some interventions have incorporated the tools of SEM into understanding and designing programmes aimed at preventing and controlling these diseases [20]. To date, numerous achievements in theoretical SEM have influenced the development and implementation of dengue and chikungunya control programmes [15, 21] and SEM control interventions in dengue have shown to be lowcost, feasible and sustainable [21]. As ascertained by social mobilisation and promotion, staff should design and implement behavioural and social interventions that will prepare communities for potential public health measures and promote risk reduction [22]. This can be achieved if SEM contextual studies are done in dengue and chikungunya infection-prone areas. From this insight, the current study employed a SEM approach in understanding the knowledge and practices regarding dengue and chikungunya among HCWs and community members in northern Tanzania. The information gathered from this study highlights the areas that need to be addressed through health promotion as well as knowledge and skills that need to be imparted for enhancing the fight against dengue and chikungunya in Tanzania. Materials and methods Study area and design Hai has both urban and rural settings, classified as tropical savannah, although the climate varies considerably due to the influence of Mount Kilimanjaro, situated in the north-eastern area of the District. On average, the District receives 700 mm of rainfall in the lowlands, 1250 mm in the mid zone and 1750 mm in the upper zone [23] during the long rains (March June) and the short rains (November December). In May 2015, we conducted a cross-sectional study in Hai district, Tanzania, to evaluate knowledge and practices regarding dengue and chikungunya among HCWs and community members. Both quantitative and qualitative data were collected. Data collection Selection criteria. We randomly selected six villages (three from each elevation zone): Nshara, Kware and Machame-uroki from the highland; Rundugai, Magadini and Boma-ngombe from the lowland. Within these villages, 13 health facilities were randomly selected to be included in the study. Community members were informed about the study through local leaders and were voluntarily assembled at the selected heath facilities, and participants (n = 125) were randomly selected. Heads of health facilities were informed about the study, after which HCWs (n = 125) were asked to voluntarily participate in the study. Community members (n = 32) and HCWs (n = 8) were recruited to participate in focus group discussion (FGDs) and in-depth interviews (IDIs) in consultation with the local leaders from their villages and head of heath facility, respectively. Quantitative survey. Face-to-face interviews were conducted using structured questionnaires with questions specifically designed for HCWs and community members. The study site was divided into two zones (high and low elevation) to control for possible variation in knowledge of dengue and chikungunya. The two zones were ecologically different in terms of climate and vegetation, which have known effects on malaria transmission [24,25] which may also influence dengue and chikungunya transmission. The questions aimed at ascertaining the community s understanding of the disease process (symptoms, transmission, aetiology and vector), risk factors (season, time of day, location) and standard preventive strategies (mosquito nets, use of mosquito repellants). Information regarding knowledge and practise of dengue and chikungunya, type and location of facility, and demographic characteristics of the respondents were gathered. Knowledge assessment among community members and HCWs. Knowledge of dengue and chikungunya was quantified using a knowledge score as described by Itrat et al., (2008) and Al-zurfi et al., (2015) with few modifications. In assessing the knowledge of the community members from the six selected villages (three from each elevation zone), good knowledge was assessed as participants answering correctly questions pertaining to signs, symptoms, mode of transmission and preventive measures John Wiley & Sons Ltd

3 for dengue or chikungunya. For HCWs, good knowledge was assessed as participants answering correctly questions pertaining to signs, symptoms and diagnostic practices for dengue or chikungunya. Correct answers for each knowledge item were coded as 10 while incorrect answers were coded as 0. The total knowledge scores ranged from 0 to 80 for community members, with scores of 40 or higher being considered good and 30 or lower being considered poor, while total knowledge scores ranged from 0 to 50 for HCWs, with scores of 40 or higher being considered good and 30 or lower being considered poor. Qualitative survey. We conducted a qualitative survey based on IDIs among HCWs and FGDs among community members. People were selected on the basis of gender and residence. Eight (8) IDIs were conducted, three and five from highland and lowland areas, respectively. Six (6) FGDs were conducted, three from the highlands and three from lowlands. FGDs lasted for a maximum of 90 min, and were conducted in areas selected by the local leaders. FGDs comprised six to eight participants, who were segmented by gender and neighbourhood, and were led by an experienced medical social scientist fluent in the Swahili language and accompanied by a note taker to record observations. Discussions were based on a semistructured topic guide that focused on knowledge and prevention practices regarding dengue and chikungunya. The following were major themes: Knowledge regarding dengue and chikungunya Conducive environment for breeding and survival of mosquitoes Preventive practices regarding dengue and chikungunya IDIs lasted for a maximum of 60 min, primarily targeting government and private health facilities. An IDI inclusion criterion was having worked with the facility for at least six months prior to study beginning. Interviews were conducted in English and Swahili language, in a private office within the facility by the study authors (MRM & DCK) with two research assistants making up a team of four. One researcher led the interview while the other recorded observations and any non-verbal communication. Interview guides containing open-ended questions were used to allow flexibility for probing to gain more insight on knowledge about dengue and chikungunya as well as diagnostic practices. Specific themes included were: knowledge on dengue and chikungunya, fevers diagnosis, RDT perceptions; influences on testing and treatment decisions. Ethical approval and consent to participate. The Tanzanian National Institute for Medical Research approved the protocol of this study with approval number NIMR/ HQ/R.8a.Vol. IX/1898. Additional ethical approval was obtained from the Ottawa Health Sciences Network Research Ethics Board (OHSN-REB) protocol number H. The objectives of the study were explained to the respondents and a signed informed consent was obtained from all participants. Data analysis All completed questionnaires were double-checked and verified on the same day for completeness and consistency. The outcome variables were knowledge and practices while the independent variables were age, gender, level of education/medical role, income (for community members), work experience and attended trainings (for HCWs). Cross-tabulations of categorical variables were calculated using chi-square tests (v 2 ). Analysis of Variance (ANOVA) was used to study the differences in mean knowledge score among participants from lowland and highland areas. All data analyses were carried out using STATA v.13.1 (StataCorp, College Station, Texas, USA). Qualitative data were translated from Kiswahili to English, transcribed and then categorised according to the themes. After identification of themes, inter-consistency checks were done before starting the analysis. Thematic analysis was used to analyse qualitative data gathered from the IDIs and FGDs. Results Characteristics of the study population A total of 290 participants were included in this study. In the quantitative survey, 250 participants were randomly selected and interviewed, comprised of 125 community members and 125 HCWs. In the qualitative survey, 40 participants were included in the qualitative study, comprised of 32 community members and eight HCWs. Table 1 summarises demographic characteristics of the of the quantitative survey. For the qualitative survey, females (65%, n = 26) were more represented than men (35%, n = 14). Results of the qualitative survey are summarised in Table 2. Demographic characteristics of community members Among community members, 78.4% (n = 98) were female. The majority of the participants had primary 2017 John Wiley & Sons Ltd 585

4 Table 1 Mean dengue knowledge scores for community members and HCWs in the lowland and highland of Hai District Lowland Highland Total Variables %(n) Mean (SD) % (n) Mean (SD) % (n) Mean (SD) All community members 63.2 (79) 16.9 (17.1) 36.8 (46) 9.1 (13.7) 100 (125) 14.0 (16.4) Age categories years 92.6 (25) 22.0 (19.1) 7.4 (2) 25.0 (21.2) 100 (27) 22.2 (18.8) Years 75.8 (25) 12.0 (15.8) 24.2 (8) 5.0 (10.6) 100 (33) 10.3 (14.8) Years 45.0 (18) 18.3 (17.2) 55.0 (22) 13.1 (15.8) 100 (40) 15.5 (16.4) Years 44.0 (11) 14.5 (13.6) 56.0 (14) 2.8 (6.1) 100 (25) 8.0 (11.5) P = 0.20 P = 0.03 P < 0.01 Gender Male 55.6 (15) 29.3 (19.8) 44.4 (12) 2.5 (4.5) 100 (27) 17.4 (20.1) Female 65.3 (64) 14.0 (15.2) 34.7 (34) 11.4 (15.2) 100 (98) 13.1(15.2) P = P = 0.05 P = 0.23 Level of education No formal 37.5 (6) 3.3 (5.1) 62.5 (10) 3.0 (4.8) 100 (16) 3.1 (4.7) Primary 65.6 (63) 17.6 (17.9) 34.4 (33) 9.7 (14.6) 100 (96) 14.9 (17.2) Secondary 76.9 (10) 21.0 (13.7) 23.1 (3) 23.3 (15.2) 100 (13) 21.5 (13.4) P = 0.10 P =.0.07 P < 0.01 Income per month (TZs) < (58) 17.1 (16.9) 37.6 (35) 7.4 (12.9) 100 (93) 13.4 (16.1) (13) 12.3 (15.8) 40.9 (9) 15.5 (17.4) 100 (22) 13.6 (16.1) > (8) 24.2 (22.2) 20.0 (2) 10.0 (0.0) 100 (10) 21.1 (20.2) P = 0.33 P = 0.29 P = 0.41 Occupational Farmer 54.4 (43) 20.0 (19.1) 45.6 (36) 7.5 (13.6) 100 (79) 14.3 (17.8) Businessman/Employed 71.0 (22) 15.0 (17.1) 29.0 (9) 12.2 (10.9) 100 (31) 14.1 (15.4) Unemployed 93.3 (14) 10.7 (6.1) 6.7 (1) 40.0 (-) 100 (15) 12.6 (9.6) P = 0.17 P = 0.04 P = 0.93 All HCWs 68.8 (86) 26.7 (7.1) 31.2 (39) 19.7 (6.7) 100 (125) 24.6 (7.7) Gender Male 76.2 (16) 28.8 (6.2) 23.8 (5) 24.0 (8.9) 100 (21) 27.6 (7.0) Female 67.3 (70) 26.4 (7.2) 32.7 (34) 19.1 (6.2) 100 (104) 24.0 (7.7) P = 0.24 P = 0.13 P = 0.05 Age category <25 years 90.0 (9) 27.8 (6.7) 10.0 (1) 20.0 (-) 100 (10) 27.0 (6.7) 25 to 40 years 75.0 (45) 28.0 (6.3) 25.0 (15) 20.6 (7.0) 100 (60) 26.1 (7.2) >40 years 58.2 (32) 25.0 (8.0) 41.8 (23) 19.1 (6.7) 100 (55) 22.5 (8.0) P = 0.17 P = 0.80 P = 0.02 Medical Role Medical Doctor 90.9 (10) 28.0 (6.3) 9.1 (1) 10.0 (-) 100 (11) 26.3 (8.1) Clinical officer 20.0 (1) 20.0 (-) 80.0 (4) 27.5 (5.0) 100 (5) 26.0 (5.5) Ass. Med Officer 42.9 (3) 30.0 (0.0) 57.1 (4) 20.0 (11.5) 100 (7) 24.3 (9.8) Nursing Officer 75.3 (61) 26.7 (7.7) 24.7 (20) 18.0 (6.2) 100 (81) 24.6 (8.2) Health Attendant 56.2 (9) 27.8 (4.4) 43.8 (7) 21.4 (3.8) 100 (16) 25.0 (5.2) Lab Technician 40.0 (2) 20.0 (0.0) 60.0 (3) 20.0 (0.0) 100 (5) 20.0 (0.0) P = 0.58 P = 0.08 P = 0.77 Experience < 5 years 85.3 (29) 26.9 (6.0) 14.7 (5) 20.0 (0.0) 100 (34) 25.9 (6.1) 6 to 10 years 76.7 (33) 29.7 (4.7) 23.3 (10) 21.0 (8.8) 100 (43) 27.7 (6.8) 11 to 15 years 14.3 (2) 15.0 (7.1) 85.7 (12) 19.2 (6.7) 100 (14) 18.6 (6.6) > 15 years 64.7 (22) 23.6 (9.0) 35.3 (12) 19.2 (6.7) 100 (34) 22.1 (8.4) p < 0.01 P = 0.91 P < 0.01 TZs, Tanzanian shillings John Wiley & Sons Ltd

5 Table 2 Results of the qualitative survey in Hai District Knowledge regarding dengue and chikungunya among community members Preventive practices regarding dengue and chikungunya among community members Knowledge regarding dengue and chikungunya among HCWs Fever diagnosis among HCWs Dengue! Yes! I recently heard about dengue eruption in Dar Es Salaam. Said an old man aged 79 years from FGD1). This is going to be tragedy! We have not been able to fight malaria, again this Dengue has come. We are going to perish Said a woman aged 47 years from Rundugai village (FGD2). It is my first time to hear from you that mosquito can transmit Chikungunya. Said participant No4 a man aged 75 years from (FGD1). My Lord Jesus! I don t know where this world is heading! a new disease Chikungunya has come again! Where shall we go. Said a lady ages 52 years (FGD1). The government is spraying pesticides to prevent mosquitos to be carried by buses to other regions in the country, currently nothing seems to be done in our community. Said an old man aged 79 years (FGD1). This is going to be tragedy! We have not been able to fight malaria, again this Dengue has come. We are going to perish. Said a woman aged 47 years from Rundugai village (FGD2). It is of recent when we were called by District Medical officer and informed about outbreak of Dengue. But I and some of my colleagues we were not aware of the presence of this disease before. Imagine! We thought it is a new disease (said a Male Clinical Officer from facility No1 who has a working experience of over 15 years). I thought I was the only medical practitioner being unaware of chikungunya disease. I have realized that majority of my colleagues and workmates are not aware of this disease. This was revealed when I shared it in our whatsapp group, majority seemed to be astonished by presence of chikungunya as being transmitted by mosquito! Remarked a Medical Assistant from facility No3 who has been employed by the government for over 15 years. It is my first time to hear about this disease [Chikungunya]!. I wonder how come I didn t hear about it even when I was in medical college! Remarked a Clinical Officer from health facility No4 who has worked for over 10 years. I googled to learn more about Dengue, this was after the dengue outbreak in Dar. But I have just heard of chikungunya today from you! Said a male Laboratory technician from health facility No1 who has 15 years working experience. I have encountered many cases where:- mrdt reads negative while a patient has fever and headache. I always prescribe antimalarials and antibiotics. This has proven to give patients immediately relief and eventually get healed completely Response from Clinical Officer with 8 years working experience- from health facility No 2. To my understanding none of the health facilities in Moshi has rapid test for dengue and chikungunya. I hope this is an opportunity to get these kits Response from a Medical Assistant from facility No3 who has been employed by the government for over 15 years. education (76.8%, n = 96), followed by those with no formal education (12.8%, n = 16). Most participants were from the age group 45 to 59 years (32.0%, n = 40) and most of the study participants reported having low monthly income (75.0%, n = 93). The majority of the community members who participated were farmers (63.2%, n = 79) followed by businessmen/women (16.8%, n = 21), unemployed (12.0%, n = 15) and other employed (8.0%, n = 10). Employment was defined as working and being paid and not being self-employed John Wiley & Sons Ltd 587

6 Knowledge and preventive practices regarding dengue among community members Of all community members, 61.6% (n = 77) had heard of dengue and 44.9% (n = 61) had heard of it through radio (Table 3). Most participants could recognise fever as a symptom of dengue (27.0%, n = 24). Fewer participants recognised headache (6.7%, n = 6), joint pain (7.9%, n = 7) and nausea/vomit (2.2%, n = 2) as signs of dengue infection. We found that 38.0% (n = 30) knew that mosquitoes were the transmitting vectors, while only 26.3% (n = 21) of these knew that the mosquitoes that transmit dengue are daytime biters. Most participants (64.6%, n = 51) did not recognise outdoor containers (e.g. tires) as breeding sites for Aedes mosquitoes, but 11.4%, (n = 9) recognised that flower jars or clean water were potential breeding sites (Table 3). Many respondents stated that they do not use any preventive measures against dengue (47.7%, n = 41). Only one respondent reported using window screens to keep mosquitoes of the house, one reported using mosquito repellant and two reported using mosquito nets; while 9.3% (n = 8) stated that they use other preventive measures like removing possible breeding grounds, covering water containers and clearing vegetation nearby their house (Table 3). On the other hand, this was reported in the discussion with the community; Dengue! Yes! I recently heard about dengue eruption in Dar Es Salaam. The government is spraying pesticides to prevent mosquitos to be carried by buses to other regions in the country, currently nothing seems to be done in our community Said an old man aged 79 years, (FGD1) This is going to be tragedy!. We have not been able to fight malaria, again this Dengue has come. We are going to perish. (FGD2). Table 3 Knowledge and prevention practices regarding dengue and chikungunya among community members Variable (N = 125) Dengue %(n) Chikungunya %(n) Where respondent heard about [infection] TV 6.6 (9) 1.6 (2) Radio 44.9 (61) 0 (0) Magazine 0.7 (1) 1.6 (2) Healthcare workers 2.9 (4) 0 (0) Neighbours 7.4 (10) 0 (0) Others 1.5 (2) 0 (0) Never heard 36.0 (49) 96.8 (121) Signs and symptoms recognised by respondent Fever 27.0 (24) 3.2 (4) Headache 6.7 (6) 0 (0) Joint pain 7.9 (7) 0 (0) Nausea/vomit 2.2 (2) 0 (0) Bleeding 3.4 (3) 0 (0) Do not know 52.8 (47) 96.8 (121) Mode of Transmission Ticks 2.5 (2) 0 (0) Mosquito 38.0 (30) 2.4 (3) Do not know 59.5 (47) 97.6 (122) Mosquito biting behaviour Day time 26.3 (21) 0 (0) Evening 1.3 (1) 0 (0) Morning 1.3 (1) 0 (0) Night 6.3 (5) 1.6 (2) Do not know 65.0 (52) 98.4 (123) Breeding sites Used car tire 0 (0) 0 (0) Clean water/flower jars 11.4 (9) 0 (0) Dirty water 21.5 (17) 0 (0) Waste 2.5 (2) 2.4 (3) Do not know 80.0 (100) 97.6 (122) Prevention practices Use bed net 2.3 (2) 2 (1.6) Mosquito repellent 1.2 (1) 0 (0) Window screen 1.2 (1) 0 (0) Do not use anything 47.7 (41) 1 (0.8) Others 9.3 (8) 0 (0) Do not know way of prevention 38.4 (33) 97.6 (122) Knowledge and preventive practices regarding chikungunya among community members Of all community members, 3.2% (n = 4) had heard of chikungunya, either through television (1.6%, n = 2) or radio (1.6%, n = 2). All participants who had heard about chikungunya reported fever as a symptom of chikungunya (3.2%, n = 4), while 2.4% (n = 3) knew that mosquitoes were the transmitting vectors. Only 1.6% (n = 2) reported that Aedes mosquitoes bite during the night, while 1.6% (n = 2) reported that mosquito nets can be used as a preventive measure against chikungunya (Table 3). The community members added this during the FGDs: It is my first time to hear from you that mosquito can transmit Chikungunya. Said participant No. 4, a man aged 75 years. (FGD1) A woman aged 52 years added; my Lord Jesus! I don t know where this world is heading to! a new disease Chikungunya has come again! Where shall we go. (FGD1) John Wiley & Sons Ltd

7 Demographic characteristics of HCWs Among HCWs, the majority were female (83.2%, n = 104). Most of the study participants were nurses (64.8%, n = 81), followed by health attendants (12.8%, n = 16), medical doctors (8.8%, n = 11), clinical officers (4.0%, n = 5) and laboratory technicians (4.0%, n = 5). Most HCWs had 6-10 years of experience (34.4%, n = 43), followed by those who had been in practices for less than 5 years (27.2%, n = 34) and more than 15 years (27.2%, n = 34) and lastly those worked 11 to 15 years (11.2%, n = 14). The majority of HCWs were aged between 25 to 40 years (48.0%, n = 60), Table 1. Knowledge regarding dengue among HCWs Almost all HCWs (96.8%, n = 121) knew about dengue, through the media. The majority reported that the risk of dengue was low in this study area (44.8%, n = 56). Only 2.4% (n = 3) reported that they had heard of a dengue case in their community. Only % (n = 84) had been involved in health promotion for dengue and 32.8% (n = 41) were not involved in any health promotion about dengue (Table 4). However, findings from the indepth interviews indicated that some awareness campaigns were in progress as indicated by a male clinical officer who had a working experience of over 15 years: Recently, we were called by District Medical officer and informed about outbreak of Dengue. But my colleagues and I, were not aware of the presence of this disease before. Imagine! We thought it is a new disease. (Clinical officer, facility No1) Knowledge regarding chikungunya among HCWs The majority of HCWs (87.2%, n = 109) had not heard of chikungunya. Of those who had heard, 7.2% received the knowledge through media and 5.6% (n = 7) through books. Most HCWs (89.6%, n = 112) did not know whether chikungunya was a problem in the community, only 4.0% (n = 5) reported that the chikungunya risk is high and 7.2% (n = 9) reported that they had heard of a chikungunya case in their community. None of the HCWs had been involved in chikungunya health promotion (Table 4). Participants in the in-depth interviews shared similar sentiments. A medical assistant who had been employed by the government for over 15 years had this to say; I thought I was the only medical practitioner being unaware of chikungunya disease. I have realized Table 4 Knowledge regarding dengue and chikungunya among HCWs in Hai District Variable that majority of my colleagues and workmates are not aware of this disease. This was revealed when I shared it in our whatsapp group, majority seemed to be astonished by presence of chikungunya as being transmitted by the mosquito! (Medical Assistant, facility No3) The lack of awareness regarding chikungunya was not limited to the lower cadre of HCWs only, as indicated by the two HCWs who had worked for over 10 and 15 years respectively, It is my first time to hear about this disease [Chikungunya]! I wonder how come I didn t hear about it even when I was in medical college! (Clinical Officer health facility No 4) I googled to learn more about Dengue, this was after the dengue outbreak in Dar. But I have just heard of chikungunya today from you! (Laboratory technician health facility No 1) Fever diagnosis Dengue (N = 125)% (n) Chikungunya (N = 125)% (n) Have you heard about [infection]? Yes 96.8 (121) 12.8 (16) No 3.2 (4) 87.2 (109) Where did you heard about [infection]? Media (TV or Radio) 96.8 (121) 7.2 (9) Books 3.2 (4) 5.6 (7) Personal communication 0 (0) 0 (0) Others 0 (0) 0 (0) How big is the problem? High 0 (0) 4.0 (5) Moderate 7.2 (9) 0.8 (1) Low 44.8 (56) 5.6 (7) Donot know 48.0 (60) 89.6 (112) Have you heard of any cases in this community/area? Yes 2.4 (3) 7.2 (9) No 97.6 (122) 92.8 (116) Have you participated in heath promotion about [infection]? Public lecture 32.0 (40) 0 (0) Source reduction 7.2 (9) 0 (0) Individual advice 4.0 (5) 0 (0) Small group discussion 4.0 (5) 0 (0) Demonstration 20.0 (25) 0 (0) Not involved 32.8 (41) 100 (125) HCWs were asked about their daily practices regarding use of malaria rapid diagnostic tests (mrdt) for malaria or fever diagnosis. Among HCWs, 45.6% (n = 57) and 2017 John Wiley & Sons Ltd 589

8 34.4% (n = 43), respectively, reported always or often using mrdt. 20.0% (n = 25) reported never using mrdt. When asked what prompts them to use a mrdt, only 24.0% (n = 24) stated for fever diagnosis, 74.0% (n = 74) stated diagnosis of patients with symptoms other than fever (joint pain, headache, nausea and vomit), while only 2.0% (n = 2) stated that they use the test during the rainy season. A small portion of HCWs (9.6%, n = 12) reported that mrdt-negative patients always request for antimalarials. Most HCWs reported that whenever they receive mrdt negative results for patients presenting with malaria-like symptoms they prescribe antibiotics (48.5%, n = 47) or diagnose another disease (40.2%, n = 39). If mrdt test kits are of stock, 33.6% (n = 42) diagnose fever clinically while 46.4% (n = 58) request for microscopy test. Among all HCWs interviewed only 4.0% (n = 5) reported that they have a dengue diagnostic tool, that is dengue rapid test (there is no chikungunya diagnostic tool). Only 7.2% (n = 9) and 0.8% (n = 1) had been trained for dengue and chikungunya diagnosis, respectively (Table 5). Similarly, clinical officers commented: I have encountered many cases where: mrdt reads negative while a patient has fever and headache. I always prescribe antimalarials and antibiotics. This has proven to give patients immediate relief and eventually get healed completely. (Clinical Officer with 8 years working experience- health facility No 2) A medical assistant who has been employed by the government for over 15 years stated: To my understanding none of the health facilities in Moshi has rapid test for dengue and chikungunya. I hope this is an opportunity to get these kits. (Medical Assistant facility No3) Knowledge score difference regarding dengue and chikungunya Table 1, summarises mean dengue knowledge scores. Among community members, few participants (15.2%, n = 19) had good knowledge of dengue (knowledge score of 40 or above) and this was significantly higher in lowland compared to highland areas (20.3% vs. 6.5% P = 0.03). The mean dengue knowledge score was higher in lowland areas (mean SD) compared to highland areas (mean SD). In general, only age and education level were statistically associated with higher mean knowledge scores (P < 0.01 for both categories). In the highland area, age categories, gender and Table 5 Diagnostic practices regarding dengue, chikungunya and malaria among HCWs in Hai District Variable (N = 125 if no other indication) % (n) Have you been trained in using mrdts? Yes 83.2 (104) No 16.8 (21) How often do you use mrdts? Always 45.6 (57) Often 34.4 (43) Never 20.0 (25) What prompt you to use a mrdt? (n = 100) Patient with fever 24.0 (24) Malaria season (Rainy season) 2.0 (2) Patient with other symptoms 74.0 (74) If mrdt results are negative, do patients request antimalarials prescription? (n = 100) Always 9.6 (12) Often 34.4 (43) No 36.0 (45) Do you think patients trust mrdt negative results? (n = 102) Yes 42.4 (53) No 39.2 (49) What do you do if mrdt test is negative? (n = 97) Prescribe antimalarials 10.3 (10) Prescribe antibiotics 48.5 (47) Prescribe both antibiotic and antimalarials 1.0 (1) Diagnose other diseases 40.2 (39) If mrdt test kits are out of stock, how do you diagnose fever? (n = 100) Microscopy 46.4 (58) Clinical diagnosis 33.6 (42) Is there a diagnostic tool for dengue? Yes 4.0 (5) No 96.0 (120) Is there a diagnostic tool for chikungunya? Yes 0 (0) No 100 (125) Have you ever been trained for dengue diagnosis? Yes 7.2 (9) No 92.8 (116) Have you ever been trained for chikungunya diagnosis? Yes 0.8 (1) No 99.2 (124) mrdt, Malaria rapid diagnostic test. occupation showed statistically significant difference of mean knowledge score within groups, while in the lowland, a statistically significant difference was observed for the gender category only. Among HCWs, 53.6% (n = 67) had a good knowledge score for dengue and this was significantly higher in lowland compared to highland areas (68.6% vs 20.5%, P < 0.001). In general, for both highland and lowland areas, the mean knowledge scores between categories were significantly different regarding gender, age and John Wiley & Sons Ltd

9 working experience (P = 0.05, P = 0.02 and P < 0.01, respectively). However, within highland and lowland areas, observed mean knowledge scores did not significantly differ by gender, age categories or medical role and experience (Table 1). Only 2.4% (n = 3) of HCWs had a good overall knowledge of chikungunya; due to these low numbers the mean knowledge score of chikungunya was not calculated. Discussion This study employed a SEM approach in understanding the knowledge and practices regarding dengue and chikungunya among HCWs and community members in northern Tanzania. By assessing knowledge about mosquito vectors, disease transmission and preventive practices, we were able to identify knowledge gaps that could be targeted to improve individual and community level action against arboviruses and increase healthcare workers ability to detect and manage disease. Community members Our study in the Kilimanjaro Region of Tanzania found that more than half of the community members surveyed had heard about dengue, while only 3.2% had heard of chikungunya, despite evidence for local circulation of chikungunya virus [5, 11, 12]. Respondents from lowland areas were more aware of dengue and chikungunya than respondents from highland areas. Our findings are similar to a study conducted in Kilosa, Tanzania, which found minimal awareness of community members regarding non-malaria febrile illnesses including chikungunya fever [26]. This could reflect either a lack of consideration of infections other than malaria among community members, or the fact that chikungunya and dengue have been recognised recently only as an ongoing health concern in Tanzania, so the community may not have had the opportunity to hear about these arboviral diseases. Good knowledge of mosquito vectors, and the signs and symptoms of dengue fever, is essential in identifying the disease and in seeking early and appropriate medical treatment to save lives [16], while poor knowledge of a disease in a community can increase the burden of communicable disease. This study found low knowledge regarding dengue amongst community members (only 15.2% had good knowledge scores). Studies from other countries outside of sub-saharan Africa have shown considerable variation in knowledge at the community level, with good knowledge among 12% of respondents in a study in Nepal [27], compared to 54% in Jamaica [17], 61.45% in Philippines [16] and 63.2% in Malaysia [28]; in all studies, knowledge was acquired through media (TV and radio) and from health professionals. In our study, more than half of the study population had heard of dengue but only a small proportion identified fever as a recognisable symptom, and most of the respondents were not able to correctly state other symptoms of dengue. This is similar to studies conducted in Malaysia [29] and Pakistan [30], where respondents were aware of dengue but their knowledge on dengue symptoms and transmission remained insufficient. It is likely that the low level of knowledge regarding dengue in our study is due to lack of regular health awareness programs on this infection, despite the two recent experiences of outbreaks in the country. The A. aegypti mosquito is known to bite mostly during the day [31]. Only 6.3% (n = 21) of community members were aware of the day-biting behaviour of the vector. This is similar to a study that was conducted in India, where most participants were unaware that dengue vector mosquitoes bite during the day [13]. Most participants did not use any preventive measures, although a small proportion reported using window screening, bed nets and repellants while others reported that they drain stagnant water and clear bushes around their houses. Community understanding of the ecology of arboviruses, including mosquito breeding and control, is vital at this juncture. To fill this gap in knowledge it is essential to design health programs that will educate residents on personal protection against mosquitoes and control of breeding sites. In this context, a SEM is suggested to be applied in health promotion. HCWs: Diagnosis and Treatment Almost all HCWs (96.8%) knew about dengue, while in contrast, the majority (87.2%) had not heard about chikungunya. HCWs had good dengue knowledge in general, with those from lowland areas having significantly higher knowledge scores than those from the highlands. Dengue, chikungunya, malaria and other fever-causing pathogens have similar symptoms that are difficult to differentiate clinically [32]. In Kilimanjaro region, malaria prevalence is less than 1% [33]; it is therefore likely that most cases diagnosed as malaria by HCWs are in fact viral and bacterial infections. This has been reported in a study by Crump et al., (2013) in Kilimanjaro, which found that bacteria and arboviral diseases were prevalent but unrecognised as cause of febrile illness. In the absence of diagnostic capacity, most fever cases are treated as malaria cases. Our study found that even if a mrdt was negative, 10.3% of HCWs reported prescribing an antimalarials or an antibiotic. When malaria rapid tests are 2017 John Wiley & Sons Ltd 591

10 not available, most HCWs (33.6%) reported diagnosing fever based on clinical signs and symptoms without laboratory confirmation. Therefore, patients are being overtreated with either antimalarials or antibiotics without evidence of the pathogens. Over-treatment has contributed much on the global health challenge which lead to the emergence of drug resistance, unnecessary adverse drug effects and increased treatment costs [34 36]. Most of the HCWs had heard of and had basic knowledge of dengue, but few reported having been trained on dengue diagnosis. This is similar to a research from Karachi, Pakistan, where it was also found that physicians had basic knowledge of dengue, but needed training in clinical diagnosis [37, 38]. Effects of lack of knowledge among healthcare professionals and improper diagnosis of diseases can worsen patient conditions, including increased likelihood of mortality. Mortality rates of 6.2% and 10.8% among patients who were malaria smear negative have been reported in two large hospitals in the area, Kilimanjaro Christian Medical Centre and Mawenzi Regional Hospital [39]. This highlights an urgent need for training on diagnosis of dengue and other fever-causing pathogens in the area. This will also help to improve disease management that will likely reduce over-prescription of antimalarials and antibiotic drugs. Only one health professional in our study reported being trained on chikungunya diagnosis. Chikungunya has been prevalent in Tanzania for many years [10], but most of the HCWs were unaware of the disease. This could be due to the fact that no cases have been reported since the 1952 outbreak in Newala [10]. In this regard, it is important to include diagnosis of dengue and chikunguya (and arboviruses in general) in medical college and university curricula to raise awareness among students. However, the limitations of this study are acknowledged including the small sample size used for the qualitative and quantitative surveys. In this regard, there is a need for a geographical-wide representative survey to assess the knowledge and practices regarding dengue in the general population. Conclusion There is insufficient knowledge regarding dengue and chikungunya fever among community members and HCWs. Low levels of knowledge regarding dengue and chikungunya amongst the study population signifies that this disease may easily be confused with other common causes of fever at the health facility and community levels. Given the emerging arbovirus outbreaks worldwide, arboviruses should be included in national campaigns against mosquito-borne infections to raise public awareness. A SEM approach could help in devising a plan for increasing awareness among community members about the symptoms, mode of transmission and improve preventive practices. Acknowledgements We thank all volunteers who participated in the study in Hai district, and acknowledge the support of Grand Challenges Canada and DANIDA in the Building strong Universities (BSU) project. We thank Kate Stechyshyn for assisting in the development of survey tools. References 1. Hertz JT, Lyaruu LJ, Ooi EE et al. Distribution of Aedes mosquitoes in the Kilimanjaro Region of northern Tanzania. Pathog Glob Health 2016: 110(3): Mboera LEG, Mweya CN, Rumisha SF et al. The risk of dengue virus transmission in Dar es Salaam, Tanzania during an epidemic period of PLoS Negl Trop Dis 2016: 10(1): Reyburn H, Mbatia R, Drakeley C et al. Overdiagnosis of malaria in patients with severe febrile illness in Tanzania: a prospective study. BMJ 2004: 329(7476): Chandler CIR, Jones C, Boniface G, Juma K, Reyburn H, Whitty CJM. Guidelines and mindlines: why do clinical staff over-diagnose malaria in Tanzania? A qualitative study. Malar J 2008: 7(53): Crump J, Morrissey A, Nicholson W et al. Etiology of severe non-malaria febrile illness in Northern Tanzania: a prospective cohort study. PLoS Negl Trop Dis 2013: 7(7): D Acremont V, Kilowoko M, Kyungu E et al. Beyond malaria causes of fever in outpatient Tanzanian children. N Engl J Med 2014: 370(9): Moi M, Tomohiko T, Kotaki A et al. Importation of dengue virus type 3 to Japan from Tanzania and Cote d Ivoire. Emerg Infect Dis 2010: 16(11): MoHSW. Dengue outbreak in Dar Es Salaam, Tanzania WHO. Dengue outbreak in the United Republic of Tanzania (Situation as of 30 May 2014). WHO Library Lumsden WH. An epidemic of virus disease in Southern Province, Tanganyika Territory, in II. General description and epidemiology. Trans R Soc Trop Med Hyg 1955: 49 (1): Hertz JT, Munishi OM, Ooi EE et al. Chikungunya and dengue fever among hospitalized febrile patients in Northern Tanzania. Am J Trop Med Hyg 2012: 86(1): Kajeguka DC, Kaaya R, Mwakalinga S et al. Prevalence of dengue and chikungunya virus infections in north-eastern Tanzania: a cross sectional study among participants presenting with malaria-like symptoms. BMC Infect Dis 2016: 16(183): Malhotra G, Yadav A, Dudeja P. Knowledge, awareness and practices regarding dengue among rural and slum John Wiley & Sons Ltd

11 communities in North Indian city, India. Int J Med Sci Public Heal. 2014: 3(3): WHO. Dengue guidelines for diagnosis, treatment, prevention and control [Internet]. WHO Library (Available from: ngue-diagnosis.pdf) [5 Apr 2014] 15. Quintero J, Brochero H, Manrique-Saide P et al. Ecological, biological and social dimensions of dengue vector breeding in five urban settings of Latin America: a multi-country study. BMC Infect Dis 2014: 14(1): Yboa BC, Labrague LJ. Dengue knowledge and preventive practices among rural residents in Samar Province, Philippines. Am J Public Health Res 2013: 1(2): Shuaib F, Todd D, Campbell-stennett D, Ehiri J, Jolly E. Knowlege.attitudes and practices regarding dengue infection in Westmoreland, Jamaica. West Indian Med J 2010: 59(2): Espino F, Marco J, Salazar NP, Salazar F, Mendoza Y, Velazco A. Community-based dengue vector control: experiences in behavior change in Metropolitan Manila, Philippines. Pathog Glob Health 2016: 2012(106): Mukherji S. Dengue: a runaway epidemic and a bewildered public health worker. Med J Armed Forces India 2015: 71 (1): Stewart-Ibarra AM A Socio-Ecological Analysis of Vulnerability to Dengue Fever in Southern Coastal Ecuador. State University of New York. Available at: works.umi.com/35/53/ html. 21. ADB/WHO, Managing regional public goods for health: community-based dengue vector control. Asian Development Bank and World Health Organization, p. 70. Available at: Community_based_dengue_vector_control.pdf?ua=1 [Accessed 12 July 2015]. 22. WHO. Global strategy for dengue prevention and control World Heal Organ. 2012; (Available from: [Accessed 20 January 2015] 23. HDC. Hai district council [Internet] (Available from: PROFILE.doc) [18 Apr 2016] 24. Drakeley CJ, Carneiro I, Reyburn H et al. Altitude-dependent and -independent Variations in plasmodium falciparum prevalence in Northeastern Tanzania. J Infect Dis 2005: 191: Kulkarni M a, Rowland M, Alifrangis M et al. Occurrence of the leucine-to-phenylalanine knockdown resistance (kdr) mutation in Anopheles arabiensis populations in Tanzania, detected by a simplified high-throughput SSOP-ELISA method. Malar J 2006: 5: Chipwaza B, Mugasa JP, Mayumana I, Amuri M, Makungu C, Gwakisa PS. Community knowledge and attitudes and health workers practices regarding non-malaria febrile Illnesses in Eastern Tanzania. PLoS Negl Trop Dis 2014: 8(5): Dhimal M, Aryal KK, Dhimal ML et al. Knowledge, attitude and practice regarding dengue fever among the healthy population of highland and lowland communities in Central Nepal. PLoS ONE 2014: 9(7): Al-zurfi BMN, Fuad MDF, Abdelqader MA et al. Knowledge, attitude and practice of dengue fever and heath education programme among students of Alam Shah Science. Malaysian J Public Health Med 2015: 15(2): Naing C, Ren WY, Man CY et al. Awareness of dengue and practice of dengue control among the semi-urban community: a cross sectional survey. J Community Health 2011: 36 (6): Itrat A, Khan A, Javaid S et al. Knowledge, awareness and practices regarding dengue fever among the adult population of dengue hit cosmopolitan. PLoS ONE 2008: 3(7): e Gubler DJ. Dengue and Dengue Hemorrhagic fever. Clin Microbiol Rev 1998: 11(3): Mardekian SK, Roberts AL. Diagnostic Options and Challenges for Dengue and Chikungunya Viruses. BioMed Research International 2015: 2015: UNICEF. Tanzania HIV/AIDS and Malaria Indicator Survey Wongsrichanalai C, Barcus MJ, Muth S, Sutamihardja A, Wernsdorfer WH. A review of malaria diagnostic tools: microscopy and rapid diagnostic test (RDT). Am J Trop Med Hyg 2007: 77(2): Osei-Kwakye K, Asante KP, Mahama E et al. The Benefits or otherwise of managing malaria cases with or without laboratory diagnosis: the experience in a district hospital in Ghana. PLoS ONE 2013: 8(3): Llor C, Bjerrum L. Antimicrobial resistance: risk associated with antibiotic overuse and initiatives to reduce the problem. Ther Adv Drug Saf 2014: 5(6): Rafique I, Arif M, Saqib N, Siddiqui S, Munir MA, Malik IA. Dengue knowledge and its management practices among physicians of major cities of Pakistan. J Pak Med Assoc 2015: 65: Thaver AM, Sobani ZA, Qazi F, Khan M, Zafar A, Beg MA. Assessing the need for training: general practitioners knowledge, attitude and practice concerning dengue and malaria in Karachi, Pakistan. Int Health 2011: 3(2): Moon AM, Biggs HM, Rubach MP et al. Evaluation of in-hospital management for febrile illness in northern Tanzania before and after 2010 World Health Organization guidelines for the treatment of malaria. PLoS ONE 2014: 9(2): 1 7. Corresponding Author Debora C. Kajeguka, Faculty of Medicine, Kilimanjaro Christian Medical University College, P.O. Box 2240, Moshi, Tanzania. dkajeguka@gmail.com 2017 John Wiley & Sons Ltd 593