Introduction

Dengue fever is a significant cause of morbidity and mortality especially among children, in several Asian and Latin American countries and has globally gained in epidemiological importance in recent decades [1] with estimates of annual dengue infections as high as 390 million and 128 countries reporting cases [2]. Community participation is vital for success in dengue control, dengue prevention and dengue research [3]. Initiatives prevention program in Latin Americas such as the integrated management strategy for dengue prevention and control (IMS-Dengue) and integrated vector management (IVM) required community participation [3]. A community based environmental management with community participation embedded in a routine control programme was effective as indicated by a significant reduction of dengue vectors in some regions [4,5]. Research on dengue also requires community participation in many ways. For example, active school absence based surveillance in Thailand [6,7] and school-based versus community-surveillance program in Peru [8] were completed because an adequate community participation. In addition, projects such as the establishment of biobanks, genetic and molecular epidemiology studies and dengue vaccine trials all depend on adequate numbers of participants. However, the recruitment may not be a trivial task. The difficulty in recruiting participants in clinical study is well documented [9–11] and this has resulted in cost increases, study completion and implementation delays, and statistical inconclusiveness [9–14]. Although general support of medical research seems high, less than half of individuals are willing to donate blood when a specific request for donation and storage is made [15].

Willingness to participate (WTP) surveys measure the willingness of research participants to take part in a particular study and the associated factors [16]. The WTP surveys have been conducted in various setting either in actual [17–20] or in hypothetical condition [18,21] and for non-infectious [20,22–25] or infectious diseases [21,26–29]. However, most of those studies did not require phlebotomy. A study found that the WTP rate in genetic research among patients after with myocardial infarction varied greatly between 24 US hospitals, from 40% to 100% [23]. In addition, another study found that public WTP to donate the blood for long-term storage and genetic research was 42% out of 3,130 research participants [30]. A study in South Korea found that only 25% of the research participants were willing to participate in clinical trials [31].

One factor that influences the WTP is the ethnic background of the participants. Studies found that African-Americans and White Americans differ in their WTP in medical research [19,32–34]. African-Americans were less likely to give a blood sample compared with White Americans [19,23,35]. Chinese participants were also less likely to donate blood for genetic research compared to non-Chinese participants [36]. A systematic review of ten clinical intervention studies revealed that Hispanics had a statistically significant higher WTP than non-Hispanic whites [37]. In addition, it is well accepted that motivations for research participation vary between participants of different cultural backgrounds [38,39]. Therefore an effort have been made to identify the factors that promote and hinder the WTP in clinical research among participants in the particular heritage [40]. As far to our best knowledge there is no study conducted in Indonesia to assess the WTP in particular in the dengue research setting.

Up to now, WTP has only very rarely been assessed for research on dengue. The only record we found, from Puerto Rico, assessed the interest to participate in a dengue vaccine efficacy trial [28] and revealed that barriers for participating in a vaccine trial included lack of information on dengue vaccine candidates, lack of trust in new vaccines and trial procedures, and fear of infection or other side effects.

In Indonesia, a dengue hyper-endemic country, all four dengue virus serotypes circulate in at least 433 regencies or municipalities (84.7%) and more than 200 million people are at risk of infection [41]. A recent nationwide study found that over half of children have been infected by the age of 5 [42]. The incidence of dengue fever has rapidly increased from 0.05 in 1968 to approximately 39.80 per 100,000 population in 2014 [41,43]. It is estimated that 30 million dengue infections occur in Indonesia annually [2] with approximately 1% case-fatality rate [44]. In Aceh, one of the high-prevalence provinces, there were 2,208 registered dengue cases in 2014 (46.6 per 100,000 population) [41]. In the wake of coordinated dengue prevention programmes planned for this province, this study aimed to determine the factors that influence WTP in dengue research that would require phlebotomy procedures among participating healthy community members.

Material and Methods

Results

Participant characteristics

In this study, 535 healthy community members from nine regencies and municipalities of Aceh province were surveyed. Their characteristics are summarized in Table 1. The average age of participants was 30.8 (17–70) years. The majority (66.9%) were living in suburban areas. No participant was illiterate. More than a third (38.7%) had completed senior high school (12 years) and approximately a fifth (27.9%) had a university degree. Half of the respondents earned less than 1 million Indonesian Rupiah (US $ 81) per month. Although only one-tenth of the participants declared having personally experienced dengue fever, one-fifth declared that family members had previously suffered from dengue fever. In addition, this study found that more than 60% of participants had poor knowledge regarding dengue and approximately 70% had poor attitude to dengue and poor preventive practices against dengue.

Download:

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

Table 1. Characteristics of participants and univariate logistic regression analysis showing predictors of willingness to participate (WTP) in a dengue study (good vs. poor) (n = 535).

https://doi.org/10.1371/journal.pone.0159139.t001

Willingness to participate in dengue study and associated factors

Only 102 participants (19.1%) had a good WTP in the proposed dengue study. Univariate logistic regression analysis revealed that being female, working as a civil servant, private employee, entrepreneur or student/university student, having high SES, and good KAP domains were associated with good WTP (P < 0.05) (Table 1). Neither age group, education, religion, marital status, monthly income, types of residence nor previous personal or family experience with dengue fever were associated with WTP in dengue research.

Being female increased the odds of having a good WTP approximately twice. Compared to farmers, the odds of having a good WTP increased if the participants were working as civil servants (OR: 2.84; 95% CI: 1.11–7.23), private employees (OR: 3.91; 95% CI: 1.48–10.35) or entrepreneurs (OR: 3.44; 95% CI: 1.25–8.57). Participants grouped in the richest quintile had increased odds of having a good WTP compared to the poorest quintile (OR: 2.71; 95% CI: 1.04–4.35). As expected, the odds of having a good WTP were increased among participants who had good KAP domains (OR: 3.36 [95% CI: 2.15–5.24], 6.07 [3.83–9.63], 1.72 [1.72–2.70]). However, multivariate logistic regression analysis revealed that only knowledge and attitude towards dengue were independent predictors of WTP (OR: 2.30 [95% CI: 1.36–3.90] and 3.73 [2.24–6.21], respectively) (Table 2).

Download:

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

Table 2. Multivariate logistic regression analysis showing predictors of willingness to participate (WTP) in dengue study (good vs. poor) (n = 535).

https://doi.org/10.1371/journal.pone.0159139.t002

Analysis with Spearman’s rank correlation confirmed that there was a weak association between SES and WTP (r_s = 0.09, P = 0.032) (Table 3). Spearman’s rank correlation analysis also confirmed that the strongest predictor associated with WTP in dengue research were attitude and knowledge regarding dengue with r_s = 0.49 and r_s = 0.19, respectively (P < 0.001), while preventive practice against dengue had a weak association (r_s = 0.13).

Download:

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

Table 3. Correlation between willingness to participate (WTP) in dengue research, socioeconomic status and knowledge, attitude and practice (KAP) regarding dengue (n = 535).

https://doi.org/10.1371/journal.pone.0159139.t003

Discussion

The recruitment of participants for medical research is complex and challenging [14,57,58], especially when studies are interventional or involve invasive sampling methods. Inability to recruit required numbers of participants within a defined geographical setting or timeframe may delay study completion and incur additional costs, or result in statistically inconclusive outcomes [12–14]. For example, a study found that only 31% out of 114 UK trials funded by two UK funding agencies (Medical Research Council and Health Technology Assessment) achieved original recruited target and 45% recruited <80% of their target and 53% of trials required an extension [14]. One of the reasons was fewer patients agreeing to participate than expected [14]. In contrast, there are studies that have been and continue to be successful in gaining the support and participation of the community in research on dengue for example active school absence–based surveillance for dengue in Thailand [6], hospital-based participation in Hawaii [59] and school-based versus community-surveillance program in Peru [8].

Several factors including site-specific dengue epidemiology and cultural characteristics of the study population are important for community participation [8]. Understanding the factors that are correlated with the WTP of a particular population is thus essential before one sets out to conduct research that requires the participation of the members of that community. The main aim of this study was to explore the extent of community support for dengue research in Aceh, Indonesia. This support was measured by participants’ stated WTP in dengue research that would involve donating 2.5 ml of their blood and allowing their family member(s) to do the same if infected with dengue virus. The findings reported here have both practical and public policy implications. From a practical perspective, they can be used to address modifiable factors that are correlated with poor WTP in order to increase participation. From a policy perspective, public participation in dengue research is important because it facilitates the evidence-based decision-making process and public acceptance of health policy derived there from. This can only be achieved if community members are willing to participate in research.

This is the first study conducted to specifically explore WTP in dengue research in Indonesia and its related factors. The first study regarding WTP in dengue-related research anywhere was reported from Puerto Rico where WTP in a dengue vaccine efficacy trial had been assessed [28]. Basically, the concept of our study is similar to those on willingness to donate a blood sample for genetic research. Therefore, the results can also be compared with those from such studies [18,20,23,35,36,40,60,61].

Our study revealed that the percentage of participants with good WTP in dengue research was very low, less than 20%. This willingness rate is much lower than those reported from European and North American countries (60 to more than 90%) [18]. However, these comparisons should be interpreted with caution because the willingness rate represents the percentage of respondents who had good WTP. The good WTP in the present study, however, is defined as a cumulative WTP rate between the willingness of the participants to donate their own blood and their willingness to allow their family member(s) to do the same if participants or their family member(s) suffered from dengue infection. In addition, “undecided” participants in our study were also classified as “poor WTP” because a cut-off point of 80% had been set. In Asian countries, Singapore for example, approximately only 50% of participants were willing to donate blood samples for genetic research [36]. Nationwide survey in South Korea found that 25% of the participants were willing to participate in clinical trials [31]. This may be an indication of differences in race, ethnicity or culture in the study participants [32–34,62].

In general the factors associated with WTP in dengue research were sex, occupation, SES, and KAP domains. The univariate analysis found increasing odds of being willing to participate in dengue research if participants were female, working as civil servant, private employee or entrepreneur, had high SES or good KAP domains. While in multivariate analysis revealed that only knowledge and attitude towards dengue were the independent predictors of WTP. The observed high WTP in dengue research among women may be associated with more pronounced altruism of women compared to men in Indonesia [63]. It is well known that altruism is one of the most important factors for WTP in a medical research study [40,64–67]. A previous study also found that male was associated with high refusal in a genetic study that required phlebotomy in US [20]. Interestingly, studies consistency found that gender had no association with WTP toward blood donation and storage for genetic research in US [30,68,69], Sweden [70] and in Singapore [36]. A study found that male however had higher WTP in cancer genetics research in US [71]. These inconsistency findings explain, in part, why gender was not a robust independent factor in the present study.

As expected, good knowledge and attitude regarding dengue were identified as positive predictors of WTP in dengue research. Our previous study revealed that there was a strong association between good attitudes regarding dengue and good preventive practices against dengue, and that there was a good translation of attitudes into preventive practice in Aceh [72]. Therefore, the high rate of WTP among participants who had a good attitude regarding dengue might reflect their efforts for dengue prevention. In addition, it has been reported that participants who have a high level of understanding (good knowledge) regarding a medical study are more likely to have a positive attitude towards participation in a medical study [73]. A study in South Korea found that good perception and high awareness concerning medical research were identified as positive predictors of WTP in medical research [31]. Our study revealed that good attitude regarding dengue was the strongest factor for good WTP. Similarly, a study on willingness to donate an organ found a significant positive correlation with attitude [74]. Studies found that good attitude toward genetic research [70] and donation and storage of blood specimens for genetic research [30] had strong association with high willingness to donate a blood sample for research in Sweden and US, respectively. In Singapore, attitude characteristic was also associated with willingness to donate blood for genetic study [36]. In addition, a good attitude toward clinical trials had strong association with WTP in medical research [31]. Furthermore, it has been well known that the lack of knowledge and awareness about medical study is a significant barrier to participate in medical study [31,75]. Altogether, this underscores that knowledge and attitude of participants are the cornerstones for their successful recruitment to a medical study. Therefore, higher enrolment rates could be achieved through positively changing individual knowledge and attitude toward dengue fever and also attitude toward participation dengue research studies such as a targeted educational approaches using an educational video [29].

It was expected that the experience of participants or their families with dengue fever would serve to increase a perception of one’s own vulnerability and therefore willingness to participate in research on the topic. For example, a study showed that knowing someone who suffered from a particular illness was a very strong motivator for participation in related medical research, stronger even than a personal history of that illness [16]. In addition, the history of genetic disease had strong correlation with the willingness to donate a blood sample in Sweden [70]. However, our study revealed that having a family member with a history of dengue fever or a personal history of dengue fever did not increase the WTP in dengue research. A similar finding was found from a couple studies in US [20] and Singapore [36] that family history of the disease had no association with willingness to donate blood for genetic research. A possible explanation from our previous study is that participants in Aceh who had personally suffered from dengue fever had no significantly better knowledge regarding dengue fever [72].

Surprisingly, in this study, formal education level was not associated with WTP in dengue research. A previous study also found that education was not an important factor for willingness to participate in medical research in USA [16] and or to donate blood for genetic research in some Singapore [36], US [20,35,68], and Sweden [70]. Furthermore, study also found that there was no association between education and WTP to participate in vaccine clinical trial in Asian countries such as China [29], India [21] and South Korean population [31]. One of the possible reasons is that, although formal education is associated with knowledge regarding dengue, it was not associated with attitude to dengue [72]. However, other studies did find that formal education was correlated with WTP in medical research [76,77].

Limitations of this study include that WTP is only a behavioral intention and therefore may not reflect or predict actual enrolment. Participants might be more likely to answer positively about WTP due to a cultural tendency to exhibit pleasing behavior. Second, WTP in dengue research may depend on factors other than demography and KAP regarding dengue. Therefore, factors not measured in this study (e.g., financial incentive, trust, altruism) could affect actual participation rates. Third, this study was unable to assess psychological barriers that might affect WTP in dengue research.

To increase the WTP in dengue or other medical research, various strategies can be used. First, efforts to increase the knowledge and attitude of community members regarding dengue and basic information related to a specific medical study are necessary. In addition, it is important to implement pre-consent education [78] during the enrolment of participants to enhance pre-existing knowledge and improve WTP. Second, (financial) incentives should be considered for increasing participation rates because this is a robust motivator for WTP [67]. In Singapore, healthcare-related incentives like free medical check-ups seemed more preferred than money [36]. In Indonesia, based on our dengue serosurveillance study (unpublished), we found that the participation was highly increased when incentives were offered. Therefore, an offer of incentives should be considered in dengue research study recruitment. Third, training for recruiters is also important and should include how to frame the study information in a positive way. Finally, if possible, the dengue research should be embedded in a government program because more community members may participate if the research is conducted by the government compared to other parties [36].

Conclusion

This study found that the willingness to participate in dengue research was very low among community members in Aceh. The two most important associated factors are the knowledge and attitude of participants regarding dengue. Therefore, efforts to increase the knowledge and attitude regarding dengue in the communities prior to conducting dengue-related research may be critical for achieving sufficient participation rates.

Supporting Information

Acknowledgments

We thank the participants of this study, Dr. Mudatsir (Microbiology Department, School of Medicine, Syiah Kuala University) for assisting in ethical application processes, and Panji Fortuna Hadisoemarto (Department of Public Health, Faculty of Medicine, Padjadjaran University) for providing the SES questionnaire.