Design and sample size

The design was a cross-sectional study which included quantitative HHSs and qualitative SDIs. The IDP camp and a local community were selected as study sites. Semi-structured in-depth interviews (SDIs) were conducted with nine key informants each in the IPD camp and the local community for a total of 18 SDIs. The intended sample size of the HHSs was calculated using a 5% precision, a 95% confidence interval of the standard value normal distribution and an estimated 20% of household heads who know that mosquitoes transmit dengue virus [21]. The calculated result was a total of 250 households for the HHSs, with individuals sampled equally in the IPD camp and the Laiza town.

Concept definition

People’s health beliefs usually affect their knowledge, treatment-seeking and preventive behaviors of diseases. Health beliefs are the general perceptions of elements related to health and disease, not just special beliefs about dengue, i.e., people’s perception of the effect of religious, socioeconomic and natural elements on their health and their perception of disease causes. More specifically, respondent’s knowledge regarding DF include their knowledge of the clinical symptoms, perceived risk of contracting DF, treatment-seeking, disease vectors, dengue virus transmission and preventive methods [5].

Study site and population

All study participants were of the Kachin tribe. The civil war broke out on 9^th June 2011. Shortly thereafter, fleeing populations resettled in camps by the border areas. About one month (July–August 2011) after the first conflict, three camps which caught about a population of 12,000 were established along the Myanmar–China border. The Hpun Lum Yang IDP camp with about five thousand IDPs is one of the three camps. The local residents mainly live in Laiza town with a population of about six thousand. The two study sites are located in the same valley with an altitude from 230 to 260m, KSR2, Myanmar, and two sites have similar landscape and ecological environments. Hot temperature, adequate precipitation and lush forest fit for the growth and reproduction of mosquitoes and also for the transmission of vector-borne diseases [9, 12, 13]. There is a clinic that cannot do any laboratory diagnosis test for dengue in the IDP camp. The Laiza City Hospital is the only health facility that can detect dengue with RDTs in the area. Both the clinic in IDP camp and the Laiza City Hospital provide free diagnosis and treatment for the IDP and local residents.

Household questionnaire survey

The household was the unit of sampling for this survey. The survey instrument was developed in Chinese for a previous study in Shan Special Region IV (SR4), Myanmar [5]. The family wealth index (FWI) was assessed by the household’s physical assets, such as housing, walls, roofs, and transportation means, and was categorized into five groups (Table 1). The local government of KIO assigns a number to each household in KSR2, so a simple computer randomization was used to select households by using existing household numbers in the two study sites for the survey. Researchers firstly visited every selected household and told the household head about the purpose of this study and related questions that would be asked. When oral informed consent was obtained, the questionnaire was administered to each household head. Investigators who understood both the Kachin ethnical language and Chinese from the Laiza City Hospital asked every question in the Kachin ethnical language and then filled out the questionnaire in Chinese [5, 12, 22].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Principal components for the construction of the family wealth index (FWI).

https://doi.org/10.1371/journal.pntd.0008321.t001

Semi-structured in-depth interviews

The SDI guideline was developed and 18 key informants were interviewed [5], with nine from each of the two study sites. Interviewees were composed of two clergymen, two health workers, three village leaders and two village representatives who were selected by the other villagers at each of the two study sites. The topics that were discussed included villager beliefs including religion and the causes of diseases, local health problems, especially local dengue situations; knowledge, treatment-seeking and prevention behaviors and their potential reasons [5, 23]. Investigators who understood both the Kachin ethnical language and Chinese from the Laiza City Hospital conducted the interviews in Kachin ethnical language and translated answers to Chinese researchers who took notes.

Data analysis

Quantitative data were entered in Excel 2007 and were analyzed in Epi Info 7.2 (Centers for Disease Control and Prevention, USA). The percentage and its 95% confidence interval (CI) were computed for each health belief, knowledge item, and treatment seeking and prevention behavior. The percentage of each health belief, knowledge item and behavior between the IDP camp and the local community was compared by chi-squared test. Microsoft Office Excel 2007 was used to enter data from the SDIs. The records were coded according to the contents of the questions and then entered into cells in Microsoft Office Excel 2007. The same content records were combined by code sequencing. The records of each content were independently analyzed by two researchers to generate themes. The two researchers then compared and discussed their findings to finalize the findings [5, 24].

Ethics

The study was approved by the Department of Health of the KIO, Myanmar, and ethical approval was granted by the Ethics Committee of the Yunnan Institute of Parasitic Diseases in China. Verbal consent was approved by the Ethics Committee as an acceptable form, as the study was interview-based and did not include any human specimens. All study participants were 15-years of age or older. According to the World Medical Association Declaration of Helsinki, the purpose and procedures of the study were explained and disclosed to the participants before obtaining their consent. Participation was entirely voluntary, and participants could pass on a question, take a break or withdraw their consent from the study without providing any explanation at any time. Their continued consent was assumed if they did not refuse to answer questions.

Demographics of the HHS and SDI participants

A total of 137 household heads from the IDP camp and 121 from the local community participated in the HHS, and all completed questionnaires were assessed as valid. The mean age of the respondents was 30-years-old (median: 28.0, range: 15–79). The ages of household heads were younger in the IDP camp than in the local community (P = 0.002), mean age 28.2-years-old (median: 26.0, range: 15–79) in the IDP camp versus 32.1-years-old (median: 30.0, range: 15–69) in the Laiza Town. The gender proportions were similar (P = 0.473), overall proportion of male household heads was less than one fourth in the two sites. About half the respondents reported 4–6 years of formal school education, and the proportion of respondents who reported seven or more years of school education was higher in the IDP camp than in the Laiza town (P<0.0001). Most of these households were poor with a family wealth index (FWI) of 1–3. The IDP respondents were poorer than local respondents who have their own houses (Table 2). The 18 key informants of the SDI included seven males (four from the IDP camp and three from the local community) and eleven females (five from the IDP camp and six from the local community), ranging from 26 to 69-years-old. Results of the SDI found that most of the male young adults were recruited into armed forces, so most of the household heads were young females. The on-going conflict has led to change of demographic characteristics and socioeconomic vulnerability. The protracted displacement resulted in significant impacts on livelihoods at the household level. Male adults who were not recruited into armed forces also had to look for work out of the camp. This has resulted in women and children comprising the majority of the IDP population.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Results of the household questionnaire survey in Kachin Special Region II, Northeastern Myanmar.

https://doi.org/10.1371/journal.pntd.0008321.t002

Themes identified from qualitative study

All participants of the study are Kachin people. The analysis of qualitative data identified five themes for both the IDP and local residents: 1) inconsistency between their religious and health beliefs, 2) people having sound knowledge about dengue because of their received formal school education in childhood and health education in response to the DF outbreaks in 2017 and also their preferring chatting in their spare time in their traditional habit, 3) people only able to seek free treatment from public health sector due to poverty, 4) people complying with instructions of the local government to conduct environment management for dengue vector control, 5) people like fogging and spraying with insecticides in spite of environmental management usually being regarded by health professionals as the most effective intervention for Aedes sp control. The analysis of qualitative data also identified a difference in perception of dengue infection between the IPD and local residents, and differences between health workers and other key informants. All nine key informants from the IDP camp felt that the camp conditions promoted disease transmission and the IDP had a higher risk of dengue transmission. All four key informant health workers did not feel that dengue was a serious disease or one of major health problems because of too many other health problems to be solved in the area. The more detailed results of qualitative study are interspersed with the quantitative data in each section below.

Health beliefs in general

A total of 215 (83.3%) of the HHS respondents believed in Christianity and Catholicism, 147 (68.4%) of the 215 respondents agreed that the God will protect good people, and 141(65.6%) of them agreed that people with evil practices may be punished by diseases. However, the four key informant clergymen were in agreement that the Bible does not say that “Evil will be recompensed with evil”, noting that the Bible says “Everyone is a sinner” and “Jesus Christ had been punished for everyone. We would not be punished as long as we believe in the God and the Jesus.” Comparison between results of the HHSs and the SDIs documented the inconsistency between their beliefs of health and religions.

A total of 251 (97.3%) respondents of the HHSs believed that all of natural factors influenced people’s health. About half of household heads thought that forests, rivers, streams and clear water pools near home were associated with diseases. There were 256 (99.2%) respondents who also believed that people’s diseases were related to poor hygiene. If there was not polluted water near home, people would be more healthy. There were more respondents of the HHSs in the IDP camp than the local residents who believed that rainy weather (P = 0.0001) and water body (P = 0.0051) were related to disease incidence. All nine key informants from the IDP camp felt that the camp conditions promoted diseases. One interviewee noted “The camp is too crowded” while others highlighted the conditions of the camp (“All the grounds are soil” and “Raining makes the camp muddied and dirty”) as contributing to disease transmission (“These are easy to cause diseases.”). Results of the HHSs showed that 147 (57.0%) of the HHS respondents thought that many flowers, grass and trees around houses would benefit health (Table 2). Most of the key informants agreed that their fellow residents in both the IDP camp and the local community did not know that bamboo and tree stump holes, even leaf axils of some plants, could be the breeding habitats of dengue mosquitos, and their fellow residents just thought that flowers, grass and trees could make people happy and healthy.

Dengue knowledge

A total of 255 (98.8%) of the HHS respondents had heard about dengue. They knew that fever was the basic symptom of DF. There was a higher proportion of the household heads who knew fever and headache symptoms of DF in the IDP camp than in the local community (P<0.05). However, less than 50% of the HHS respondents knew the more specific symptoms of DF such as orbital pain, pantalgia and rash. Most of the HHS respondents perceived DF as a serious illness or a deadly disease (Table 2). All 18 key informants confirmed that there were dengue outbreaks in their communities in 2017, and they received information, education and communication (IEC) about dengue from the health sector during response to the outbreaks. Fewer participants of the HHS in the IDP camp thought that they could easily contract dengue than in the local community (P = 0.0048) (Table 2). However, all nine key informants from the IDP camp thought that the crowded camp could easily promote diseases transmission, so they were at a higher risk of dengue infection. One key informant said “People move frequently out and in the Camp. Young adults are soldiers of the armed forces who live in forests, but they often come back to see their wives, children and parents. Some people go out of the camp working at other parts of Kachin State or even the south of Myanmar and Yunnan Province of China. They may bring the diseases into the camp”. All four key informant health workers from both the IDP camp and local community did not feel that dengue was a serious disease or one of the major health problems. The four health workers agreed that there were more pressing health problems to be solved than dengue in the area.

A total of 224 (87.8%) HHS respondents knew that DF was communicable; and 233 (91.4%) of them felt that it was transmittable from person to person directly by physical contact, speaking and breathing, etc. There were more participants who thought dengue was directly transmittable from person to person in the local community (97.5%) than the IDP camp (86.1%). Only 66 (25.6%) knew that DF was a viral disease, and there was no significant difference about what cause the dengue between the IPDs and the local residents (P = 0.8968). A total 206 (79.8%) of 258 HHS respondents confirmed that dengue was transmitted by mosquitoes, further, 203 (84.9%) of 239 HHS respondents who answered the vector question knew that piebald or Aedes were dengue-transmitting mosquitoes; 150 (58.1%) asserted that Aedes bit in the daytime; 246 (95.3%) reported that water containers or small-scale pond water were the principal breeding sites for the Aedes larvae; 255 (99.2%) regarded that turning containers upside down could help to reduce dengue transmitting mosquitoes to prevent dengue transmission. About half of the HHS participants thought that adult mosquito interventions including fogging and spraying with insecticides, use of mosquito coils and of bed nets could prevent dengue infection (Table 2). These results of the HHS demonstrate that Kachin people had a good knowledge regarding dengue vectors and transmission. Key informants noted several reasons for this sound dengue knowledge, including, 1) most of the Kachin people (>80%) had formal school education and could read in Kachin language; 2) the residents received IEC including posters, leaflets and lectures about dengue during the dengue outbreaks in 2017; 3) the Kachin people have a traditional habit that they enjoy chatting in their spare time, and dengue was one of topics that they talked about.

Treatment seeking behaviors

Among the HHS respondents, only one (0.4%) household head said that he took drugs himself at first when he had fever. The 257 (99.6%) others said that they would go the public health facilities for treatment first (Table 2). The same result was obtained from the SDI. A key informant said, “Department of Health of the KIO provides medication free for all people regardless of ethnicities and nationalities”. Another noted that “People are poor. They cannot pay for drugs and services of private drug outlets, clinics and hospitals”. In the IDP camp, people sought free treatment in the public clinic first. When they were not getting well, they sought treatment from the Laiza City Hospital. If they did not become well, a few patients might seek treatment from the private sector or hospitals in China if their family had funds, however, most families (>90%) did not have money to do so. Key informants also noted that that public health sector only had limited drugs, equipment and services for primary health care. It was reported that the Laiza City Hospital did not have enough RDTs for dengue detection, so the number of dengue cases diagnosed by laboratory test was limited.

Prevention behaviors

A total of 210 (84.1%) HHS respondents reported that they turned containers upside down within five days. Around a half of them said that they used mosquito coils, cleaned their houses and surroundings every day. There was no significant difference in these prevention behaviors between the IDPs and local residents. There was a higher proportion of the HHS respondents who reported use of bed nets at the IDP camp than the local community (P = 0.0001) (Table 2). Results of the SDI expanded upon these prevention behaviors. All key informants agreed that behavior change communication (BCC) was carried out in order to respond to the dengue outbreaks in 2017, and people have complied with instructions of the local government about dengue control interventions. One key informant mentioned that “The malaria program delivered a lot of bed nets, but some of the nets are broken and cannot prevent mosquito bite effectively”. It was also noted that residents like fogging and spraying with insecticides. One key informant health worker from the local community said, “Insecticides can kill pests including mosquitoes and flies to prevent diseases. This year (2018), with help from China, we have conducted six rounds of fogging and spraying with insecticides. Now there is not any case of locally infected dengue, only two imported DF patients who came back from Myitkyina.”