Ownership, utilization, and motivations for use or non-use of bed nets among community members at risk of malaria along the Thai-Myanmar border


 Background

With the goal for malaria elimination in Thailand set for 2024, increased coverage and utilization of bed net, especially insecticide-treated net (ITN) or long-lasting insecticidal net (LLIN) is a key strategy. This study aims to provide the necessary information about bed net ownership and utilization among the population at risk of malaria living along the Thai-Myanmar border in Tak province.
Methods

A cross-sectional study was conducted using a mixed-method approach in 331 households from 5 hamlets in the villages of the Thai-Myanmar border. The research tools included a questionnaire, bed net inspection, and semi-structured interviews. Log-binomial regression was used to explore the sociodemographic factors associated with bed net utilization. The qualitative analysis employed a thematic analysis approach.
Results

This survey found that 98.5% of households had at least one bed net per household, and 74.3% had at least one ITN/LLIN. However, only 30.8% of households reached the standard policy set by the Minister of Public Health of one ITN/LLINs per two persons. Most residents used bed net (92.1% used in the previous night and 80.9% used every day). For those using bed nets, however, 61.9% used ITNs or LLINs the night before and 53.1% used them every day. Nonetheless, the usage rates of bed nets (any type) in the previous night among children and pregnant women were high, reaching 95.3% and 90.0%, respectively. Three explanatory variables including “not stay overnight in the forest”, “preference of free bed nets”, and “sufficient numbers of bed nets to cover all sleeping spaces”, showed statistically significant association with bed net use every day. The major reasons for the regular use of bed nets in both household and the forest were to prevent mosquito biting. The reasons for not using bednets in the household were discomfort feelings from heat, perception of unnecessity due to low mosquito density, whereas the reason for not using bednets in the forest was inconvenience.
Conclusion

Despite that overall coverage and usage of bed nets was high, only one third reached the standard level specified by the policy. Overnight in the forest, the dissatisfaction with the quality of free bed nets, insufficient number of bed nets, discomfort from heat, perception of no benefits of bed nets due to low mosquito density, and inconvenience were factors influencing bed net use. Maintaining high coverage and utility rate of bed nets should be a priority for the malaria high-risk population.


Page 3/24
Background Based on the 2020 world malaria report, there were 87 countries and areas with ongoing malaria transmission and approximately 229 million malaria cases in 2019 [1].Thailand reported a substantial decline (19%) in total reported cases between 2018 and 2019 [2]. Malaria in Thailand is patchy in its distribution and can be typi ed as 'border malaria' and 'forest malaria', with the highest transmission along international borders and in rural forested areas [3]. The western border with Myanmar has had the highest burden of malaria and has been the focus of malaria control programs for decades [3]. Motivated by the continuous decrease in malaria burden, the Thai government has declared a national malaria elimination plan intending to achieve this by the year 2024 [4].
To accomplish this goal within the time frame, a key strategy is to increase the bed net coverage and utilization, especially insecticide-treated nets (ITN) or long-lasting insecticidal nets (LLIN). The goal is to achieve 90% LLIN coverage among populations in high transmission areas (designated as A1 or A2 areas). A1 is village with reported indigenous malaria cases in current nancial year and A2 is village without indigenous malaria cases for past 1-3 years [5]. Previous studies among the general population in Thailand and among the populations on the Thai-Myanmar border (Prachuap Khiri Khan Province) identi ed poor coverage and poor utilization of ITN/LLINs [6,7]. However, actual coverage and utilization of bed nets as well as factors attributing to bed net use have not been investigated in this study area.
The objective of this study was to assess the ownership, accessibility, and utilization of both treated (ITNs and LLINs) and untreated bed nets in a remaining malaria transmission focus in western Thailand.
A cross-sectional study was conducted and questionnaires were used to determine predictors of bed net use. The results are useful for formulating appropriate policies for the control programs and for the promotion of LLINs and long-lasting insecticide-treated hammock net (LLIHN) within the context of populations living along the malarious borders.

Study design
This is a mixed-methods cross-sectional study that included a questionnaire, an inspection form, and semi-structured interviews among selected participants. The household survey was conducted from August to October 2019.

Study site
A community-based cross-sectional survey was conducted in the areas under the International Center of Excellence for Malaria Research (ICEMR) project in Tha Song Yang District, Tak Province, northwestern Thailand (Fig. 1). Tha Song Yang is situated in the northwestern region of Tak, on the Moei River bank near the Myanmar border. The climate is tropical with an annual average temperature of 26.4˚C. The rainy season is between May and October, with an average yearly rainfall of 1540 mm. The inhabitants of this area are approximately 30% Thai and 70% ethnic minorities. Normally, there are two peaks of malaria transmission, one at the beginning of the rainy season (May-August) and the other at the end of the rainy season (October) [8]. Plasmodium vivax and P. falciparum, respectively, are the predominant species in this region, although all human malaria parasites, as well as the simian malaria species P. knowlesi, have been identi ed in this area [9]. This study was conducted in 5 hamlets including Nong Bua, and Tala Oka (Mae Usu sub-district), Suan Oi, Pha Man, and Ko Ma Nae (Tha Song Yang sub-district).

Study population, sampling and sample size
For the quantitative component of the study, households residing in the study site were randomly selected from the study villages. Out of a total of 918 households in the 5 hamlets, 335 were selected by probability proportional to hamlet size, using an online statistical calculator [10]. Most of the household members are the Karen ethnic group [11]. From each household, the questionnaire and inspection form were administered to heads of household or representatives who look after the household. For the qualitative component, 24 potential respondents including 2 community leader from each sub-district, and 4 heads of household from each hamlet were selected to participate in the semi-structured interview. Purposive sampling was used to recruit participants for this part of the research. The target participation and the potential participants were selected based on geography, age, gender, and reported bed net use (including both those who do and don't report using bed nets). Important community leaders as "gatekeepers" were consulted and engaged in order to identify and invite the participants to join this study.

Household surveys
A community-based survey was conducted to assess the ownership, accessibility, and utilization of both treated (ITNs and LLINs) and untreated bed nets using a questionnaire, inspection form, and semistructured interviews. Respondents from 331 households in completed the interviews. In the selected households, face-to-face interviews were conducted with heads of household by a trained study team who speak and understand both Thai and Karen languages.

Data analysis and statistics
Data were extracted from the survey database and imported into the SPSS program version 22.0 [12] for analysis. Proportions (with 95% con dence intervals) were used to summarize categorical variables related to ownership, accessibility, and utilization of bed nets. Log-binomial regression was used to explore the sociodemographic factors associated with bed net utilization (1 = yes, used bed net; 0 = no, did not use). Since populations in the same household are likely to have the same pattern of either using or not using bed nets, the log-binomial regression model included a random intercept for household to account for confounding and difference in response variation within and between households. An adjusted prevalence ratio (PR) (with 95% con dence intervals [CIs]) was used to measure the strength of this association. We used log-binomial regression (calculating adjusted PRs) rather than logistic regression (and calculating model-adjused odds ratios) because the proportion of bed net use was high and may have resulted in misleading model adjusted odds ratios.
The qualitative analysis employed a thematic analysis approach [13]. The tentative code categories were given in the conceptual framework. Data were analyzed using thematic analysis of content to allow bringing together of similar views from different respondents together.

Household-level characteristics
For the 331 households participating in the study, the majority had bamboo/wood walls (94.56%), bamboo/wood oors (82.18%), and terracotta/galvanized iron roofs (71.60%) (Fig. 2   Bed net ownership, access, and utilization  The most common pattern of bed net deployment by the household heads included inspecting for holes (86.32%) and checking for mosquitos trapped inside (83.06%). However, only 36.81% tucked in the nets fully before sleeping, and 28.99% slept away from the edges of the bed. Only 103 (33.55%) of the household heads deployed the bed nets properly (Table 4). In the nal multivariable log-binomial regression, all three explanatory variables including "not stay overnight in the forest or the eld", "like to use free bed nets", and "su cient numbers of bed net to cover all sleeping spaces" showed a signi cant association with daily bed net use (Table 5). Log-binomial regression was also used to compare the bed net use among respondents who stayed overnight outside the house in the previous year, but only family income showed a signi cant association with daily bed net use in the nal multivariable log-binomial regression (data not tabulated).
Reasons for use or non-use of bed nets in the household and the forest A total of 22 respondents were interviewed by semi-structured interview. The results organized into key themes that emerged from the discussions.

Reasons for use of bed nets in the household
Most participants reported using bed nets to prevent mosquito bites and malaria infection. Most of them remembered using bed nets from a very young age, whereas some of them started to use bed nets when they had children or grandchildren. Bed net use had become a habit for them and they could not sleep without it.
"I use bed nets to protect myself against mosquito bites. I have used bed nets since I was born and becoming my habit I cannot sleep without the net" (Female villager, Mae Usu subdistrict) Additionally, free bed net distribution campaigns were mentioned as a factor that supported them to start and continue using bed nets in the last 20-30 years.
"After I immigrated to Thailand 20 years ago, I started to use bed nets and they were free bed nets distributed from health providers. The net is good and prevents mosquitoes from biting and not getting sick with malaria." (Female villager, Tha Song Yang subdistrict)

Bed net use in farms and forests
Forest goers in this area had not used any kinds of bed net while they were in the forest. Most forest goers used bed net only in the subsistence farm huts, their temporary residence nearby the forest. Both forest goers and farmers, if they had su cient bed nets for use in their households, they would take the old bed nets for use in the subsistence farm huts. If they had limited bed nets, they would carry bed net to the subsistence farm hut and carry it back home, especially when children accompanied them to the farms. However, many of them used bed nets in a subsistence farm hut only in the rainy season when mosquitoes are abundant.
"When children accompanied us to the farm plots or forest, we always carried bed net to use in the farm hut. Our children need to get protection from mosquito bites." (Female villager, Mae Usu subdistrict) "I sleep under bed net every day. And even when I was going to overnight in the farm, I took an old bed net to use in the farm hut." (Male villager, Tha Song Yang subdistrict)

Reasons for non-use of bed nets in the household
The major reason that participants reported for not using bed nets every day at home was discomfort from the heat and perception of unnecessity due to low mosquito density, especially in the hot season.
"I am still afraid of malaria, but I could not sleep it was too hot to sleep under the nets in the summer" (Female villager, Tha Song Yang subdistrict) The second most commonly reported reason was feeling complacent and not wanting to expend the effort needed to set up and use nets in the evenings, especially for adolescents who did not share sleeping space with their parents.
"My son and daughter rarely use bed net because they are so lazy to hang the bed net up. I sometimes have to hang it for them and tell them to sleep under bed net." (Female villager, Mae Usu subdistrict) Other reasons mentioned for not using a net at home included inadequate number of bed nets, being not habitual of sleeping under the bed nets, use of alternative mosquito control methods, too small size of the free bed net, rough texture of material of free bet nets, strong smell of insecticide, and inadequate space to hang a bed net.

Reasons for non-use of bed nets in the forest or farms
The major reason for the non-use of bed nets in the forest was inconvenient to carry the bed nets and prioritized other essential items. Even in the subsistence farm hut, some felt lazy to unpack and hang the bed nets up before sleeping in. Although someone received an LLIHN to use in the forest, they tried to use the hammocks in the forest, they found problems from using the net.
"Going to forest, we travel light with a few necessary items so we can carry lots forest products when we return. It is inconvenient to bring the net to the forest, it takes space to carry. It also can get entangled with other items, obstructed hunting and gatering process, in particular to hunting we need to be quick to shoot the targets. Sometime at night, animals such as elephants approach our sleeping area, we have to run away, we have no time to pack the net." (Male villager, Tha Song Yang subdistrict) The second reason was the inadequate number of the nets for use when visiting the forest. Other reasons for not using a net in the forest included being not habitual of sleeping under the bed nets, discomfort from heat, and use of alternative mosquito control methods.

Discussion
This is the rst cross-sectional study assessing the ownership and utilization of bed nets among the border population in Tha Song Yang District of western Thailand.  [14], and 50-65% coverage of ITN/LLIN among households of migrant population [15,16]. The WHO advocates for universal coverage of ITN/LLINs with one net for every 2 persons [17]. In spite of the high-level ownership of and access to bed nets in this study because of the LLIN distribution campaign, only 30.82% of all households had su cient numbers of ITN/LLIN for one ITN/ LLINs per two persons. This insu ciency might be due to the operational challenges of the LLIN distribution system, as some studies identi ed operational barriers to continuous LLIN distribution [18,19]. This study suggests that the insu ciency of ITN/LLIN is also a concern and can be used as an indicator for the LLIN distribution program. Plucinski et al. reported that the access indicators of LLINs were high in Mozambique when the bed net distribution campaigns used a novel distribution model and used LLINs designated for each sleeping space [20]. Thus, the LLIN distribution systems in Thailand should be evaluated in the future.
One of the major factors in not using bed nets in our data was the use of different sleeping sites. Work and evening activities in the forest impact where community members choose to sleep on any particular evening. Together with the previous ndings in Prachuab Khiri Khan province [7] and Southern Thailand [32], sleeping elsewhere especially at temporary shelters in the plantation is another non-use reason.
Another major factor in bed net use was net manufacturing materials and size are other factors affecting net utilization among households with more family members [7]. Many other studies reported under usage of nets due to the lack of su cient nets to cover all household members [7,15,26,27,[33][34][35].
Household socio-economic status has likewise been shown to in uence bed net access and is the strongest determinant of net use [28,[36][37][38][39][40][41][42][43]. Our study showed that wealthier families were signi cantly more likely to use bed nets every day than the poorer families, likely because of the ability to purchase bed nets for use outside the home.
Consistent with previous studies [44][45][46][47],the perceived role of bed nets as a means to protect against mosquitoes and malaria was a signi cant predictor of seasonal bed net use. In our research, participants reported higher bed net usage during the rainy season. Aside from malaria prevention, the main reason for sleeping under nets was because they provided comfortable sleep and protected against biting insects (including mosquitoes).
Other reasons for the not using bed nets were in line with other research [7,16,[48][49][50][51][52][53][54][55][56][57] suggesting that discomfort from heat and low mosquito density are common reasons for non-adherence. Complacence, the inadequate number of of bed net, rough materials of bed net, not having a routine of bed net use, use of alternative mosquito control methods, and too small size of free bed nets were the other most commonly cited reasons for not using a net at home. Some studies have shown that bed nets decrease air ow, making it feel hot and sti ing under a net [48,58,59]. In seasons of low and/or variable mosquito nuisance, education will need to emphasize that the risk of malaria is not necessarily diminished when mosquito numbers are perceived to be low.
The reasons for not using bed net in the forest were also consistent with some past studies in the greater coverage and utilization of bed nets in this study, the bed net use behavior of people may also affect the effectiveness of bed nets. Some did not use bed net every day, especially in summers and when they stayed overnight in the forest or the eld. Some did not deploy bed nets properly, used damaged/torn nets, and used untreated nets. Nevertheless, given the broader impacts of ITNs for preventing all vectorborne diseases such as malaria [71,72], Japanese encephalitis [73], and leishmaniasis [74], proper net distribution and encouraged use still need to be strengthened, especially for the high-risk populations such as the forest goers.
Outdoor transmission remains the main challenge for malaria elimination program in Thailand. This transmission is in uenced by human behavior, including efforts at preventing exposure to mosquito vectors. At least one study (from Tanzania) has shown that relatively modest coverage (53.13% ITNs use) in the community can lead to community-wide protection against malaria [75]. However high risk groups such as Karen ethnic groups living along the border and forest goers may not receive the same bene ts of the protection. In our study, low use of bed nets when overnighting in the forest was related to perceptions of inconvenience of carrying and using bed nets, insu cient numbers of bed nets to use outside the home, and low perception of bed net necessity. Therefore, interventions among these high risk groups should pay attention to factors beyond bed net coverage, and use and look for long-lasting new tools or approaches for preventing exposure to mosquito vectors to decrease outdoor malaria transmission.
Study limitations included 63.14% of heads of households were female. However, female heads of households should be in better position to respond about ownership and use of bed nets of every household member. This study was conducted in Thasongyang district; therefore, the results could represent population living along Thai-Myanmar border and Karen ethnic groups, but not represent Karen groups from other areas in which lifestyle, culture, or health operation were different. Due to the limitation of budget and time, this study was conducted in one season (rainy season with more abundant of mosquitoes) which may result in high rate of bed net use in the previous night. Nonetheless, adding question of everyday use of bed net, should reduce the bias from season differences.

Conclusions
This study showed that high overall coverage and usage of bed nets in the study area; however, only one third reached the standard level speci ed by the policy. Overnighting in the forest or the farm plots, the dissatisfaction with the quality of free bed nets, insu cient number of bed nets, discomfort from heat, perception of no bene ts of bed nets due to low mosquito density, and inconvenience were factors in uencing bed net use. Moreover, only one third of heads of households deployed bed net properly.
Further health promotion programs should ensure to maintain high coverage and utility rate of bed nets and further studies should prioritize the investigation of new preventive tools for effective outdoor protection in particualt ot the forest goers.