G·A·B·B·A

This work represents the most comprehensive description of community perceptions regarding dengue prevention in a region of early contact with the disease. Lessons learned will undoubtedly strengthen the prevention efficacy in epidemic and endemic regions as well as contribute towards the preparedness of dengue risk areas worldwide. The methodologies used may be useful in developing similar tools for the prevention of other diseases.
Dengue is considered to be the main arboviral disease worldwide. Its prevention depends primarily on community involvement through the acquisition of preventive behaviours such as larvae source reduction activities. Since 2005, when Aedes aegypti (the main dengue vector) was reported for the first time in Madeira Island, the local authorities have largely invested in campaigns to promote source reduction activities in the community. Despite these campaigns the mosquito thrived through the Island and in 2012 the first dengue outbreak emerged.
The present study’s main objective was to assess Madeira Island’s community perceptions regarding dengue-preventive behaviours. This is a critical step to better understand how to promote these behaviours more effectively in the community.

In order to do so the study started by performing a cross-sectional survey which applied a questionnaire and a breeding site inventory on a random representative sample of 1276 subjects from the most aegypti-infested area in Madeira Island. Moreover, a new methodology was developed - Essential-Perception analysis (EP-analysis) which allows the quantitative assessment of community perceptions regarding dengue prevention in a 0-10 score. The EP-score level of the community was estimated through a categorization of the residents’ answers to the questionnaire. According to EP-analysis the perception level equal to ten (EP-score = 10) is the maximum value of perception that can be assessed, however it represents the essential perception required to attain a minimum understanding of the proposed behaviour. Results showed that most of the residents were under this minimum perception requirement (99.6%) and that the population had an EP-score average= 5. The EP-score was associated to the presence of breeding sites. This score was significantly higher when there are no breeding sites in the houses (p˂0.001), suggesting its ability to infer about the residents’ preventive practices. The EP-analysis also identified of 13 myths about dengue prevention which were believed by the population. Some myth examples are “mosquitoes only exist in dirty houses” or “insecticides are very effective to control mosquitoes”. The most frequent breeding sites found were compatible with a clean and organized urban environment, presenting an atypical pattern of infestation compared to dengue endemic regions.

The first dengue outbreak in Madeira, provided the sole opportunity to compare perceptions before and after this outbreak using the perception quantitative assessment, EP-analysis. For this purpose a second cross-sectional survey was performed to apply the same questionnaire to a subgroup of the population and the results from both studies were compared. A randomized block design methodology was developed which enabled the comparison by matching the two studied populations while ensuring their homogeneity in six determinant variables. A total of 88 pairs of female subjects living in Funchal were compared. Each pair was equal in educational level, age group, trips to dengue endemic countries and self-reported exposure to mosquito bites. This number was satisfactory for a comparison using Wilcoxon test (power~1.000). After the outbreak the community showed an improvement in their perception with an average increase of two points in the EP-score. However, there was not a significant alteration in the percentage of individuals who achieved an EP-score=10 which guarantees the minimum understanding of the proposed behaviour. Some myths disappeared such as “dengue only occurs in tropical/non-developed countries”, but curiously others myths emerged such as the false idea that the end of the outbreak corresponded to “A.aegypti eradication”. However, even after the outbreak most of the female community still believed in at least one myth (98.9% before the outbreak, 96.7% after it).

Finally this study complemented the previously mentioned quantitative perception assessment with a qualitative perception assessment through focal groups and subsequent deductive and inductive thematic analysis. The results were consistent with previous ones, confirming the presence of most of the previously identified myths. Moreover, more than twenty new perceptions, beliefs and feelings were identified (not detected in the questionnaire), such as “the demotivation about the preventive activities” since they seemed “hard to perform” in their procedures or even “in vain” when there is no supervision to ensure a joint effort within the neighbourhood.

In conclusion, this mixed-method study recognized what can, cognitively and emotionally, impair the community engagement in the proposed behaviours, indicating important messages and interventions that should be considered in subsequent local dengue campaigns. Results also showed for the first time how an urban female community experienced the outbreak, revealing that it provoked an improvement in perceptions regarding source reduction which had not been enough to guarantee a minimum understanding of it. Moreover, the study provided a score of the level of understanding of dengue-preventive activities. This allows the comparison of dengue-preventive perceptions between different demographic groups, their monitoring over time, and their evaluation after a preventive interventions/campaings.