This is a quasi-experimental two-group intervention study, one composed of Endemic Disease Control Agents (ACEs) and one of college students. The study was conducted in the city of Campina Grande, being the second largest urban centre of Paraíba in north-eastern Brazil and currently has 355,000 inhabitants. In the city, there are about 350 ACEs linked to the Entomological Surveillance of the municipality.
The study population included 100 students and 50 ACEs who signed up for a voluntary distance learning course which would use mobile devices. The course was disseminated through the university's website and in partnership with the Department of Health of the Municipality of Campina Grande, which selected 50 agents to participate in the course.
Participation was in the form of a scavenger hunt where a cash prize was offered to those who completed all activities, including the course evaluation. The project was initiated after approval by the Research Ethics Committee of the State University of Paraíba (Protocol CAAE 67429517.5.0000.5187) and due consent of the participants gained by their signing the informed consent form.
For the purpose of statistical analysis, only participants who completed all of the elements of the programme, answering both questionnaires, before and after the intervention, and performing all activities of the intervention. Participants who did not complete these steps or who skipped the course were excluded. The final sample consisted of 58 participants, 31 students and 27 ACEs.
Intervention - Theoretical background
The World Health Organization (WHO) published in 2012 a document entitled “Health Education: Theoretical Concepts, Effective Strategies and Core Competencies” which outlines the main theories of behavioural change and those that can be used in educational interventions [12]. The WHO highlighted nine behavioural change theories that have been widely used in health interventions; three were selected whose constructs were studied in this investigation. The Theory of Rational Action (TRA) considers that knowledge contributes to changing attitudes and beliefs, which are essential for behavioural change [12]. The Health Belief Model (HBM) advocates that decision-making depends on the perception of susceptibility, disease severity, benefits and barriers associated with behaviour [12]. Later, the model added the concept of self-efficacy. Social Cognitive Theory (SCT) describes three main factors that affect a person's likelihood of changing health behaviour: self-efficacy, goals, and outcome expectations. Learning target behaviours can be facilitated by observing people performing the same [12].
Changing the frequency with which a person performs a particular behaviour requires a change in at least one of the following observables: ability, motivation, or opportunity to engage in the activity [13-14]. Ability refers to the physical and psychological condition to perform a behaviour. Motivation involves all processes that energize and direct behaviour, including not only goals, plans, and beliefs, but also "automatic" processes that involve emotions, habits, and impulses. Opportunity involves all factors external to an individual that can influence engagement in an activity, ranging from the physical environments in which people spend time to the social and cultural environment that determines how we perceive and think about actions. It includes reducing the distance between intention and action. Therefore, to maximize the potential benefit of behaviour change interventions, it is important to understand that capacity, motivation, and opportunity vary depending on specific behaviours, target populations, and social and environmental contexts [13-14].
Experience in performing a behaviour can make it habitual, so that intention and motivation over time become less important in determining individual performance. In order to foster change, it is necessary to show people performing the target behaviour that one wants to establish in the population, and transmitting personalized, interactive and accessible information. However, it is not necessary to link the preliminary or previous change in attitudes to the subsequent change in behaviour, as both may occur concomitantly [15–17]. Familiarity, observation of others performing the desired behaviour, and social norms are predictors of behaviour change. Table 1 summarizes the constructs of behaviour change theories selected for study in this investigation containing a brief description of the theory and concepts, and some questions for reflection.
Intervention – description and assessment tools
Using a distance-learning platform adapted for mobile devices, the intervention consisted of tasks or missions that were presented through short videos with people performing the desired target behaviour. To demonstrate the accomplishment of the tasks, participants produced audio-visual content (videos) and shared on social networks (Facebook). The posts had specific hashtags such that it was possible to track the post and interactions on social networks. In addition, the participant also included the link to their publication on the virtual platform. There were ten tasks to be performed in the 40-hour course, held between October and December 2018, with two face-to-face meetings, one to explain the objectives and operation of the platform and the other to reward participants. This strategy was previously described and tested by Mangueira et al (2019) [18].
During the course, the content of the missions was evaluated by the ACEs and students to identify possible errors in inspection procedures or elimination of mosquito breeding sites. Thus, the participants also answered some questions to evaluate the quality and relevance of each mission so that the content could be reproduced with students and teachers of Basic Education in the next stage of the project.
The assessment of perceptions and behaviours before and after the intervention was performed by applying a self-reported questionnaire containing questions with binary yes and no answers. Positive responses indicated that the participant had perception and preventive behaviour, or less risk of acquiring arboviruses, and they were worth 1 to 3 points; negative answers, or those that were not applicable to the participant's situation, were worth zero points.
Each of the behaviours target prevention and breeding elimination was considered a dependent categorical variable. The sum of the score for the ten questions related to prevention actions led to the “Prevention Score”, which could range from 0 to 10 points; which could be added to the score of the five answers regarding breeder elimination (“Breeding Elimination Score”) thus giving rise to the “Target Behaviour Score” (0 to 15 points).
Independent variables were grouped into the following subgroups:
Sociodemographic variables such as gender, age, marital status, income, occupation and whether the participant had children.
2) Variables related to environmental risk that encompassed conditions outside the participant's home such as the existence of vacant land or abandoned houses, offer of garbage collection services, running water and entomological surveillance. The sum of the score of the answers gives rise to the “Environmental Risk Score”, ranging from 0 to 5 points;
3) Home risk corresponds to conditions that may or may not facilitate the proliferation of mosquitoes in the participant's home. For example, if the person lives in rented accommodation, they are likely to be less interested in window screens or home improvements. Roofless houses with water tanks, water storage containers, gardens and plants are more likely to have breeding grounds. Positive statements were scored and the higher the value of this score, the greater the risk of having breeding grounds at the participant's residence. Thus, the “Home Risk Score” ranges from zero to ten points. The sum of the two previous scores gave rise to the “Environmental and Household Risk Score”, ranging from 0 to 15 points.
4) Constructs of Behaviour Change Theories: 25 assertions were proposed with which the participant could agree or not; being assigned one to three points for positive responses that revealed knowledge, perceptions, attitudes and beliefs that favour behaviour change. The sum of the scores gave rise to different scores: “Knowledge Score” (TRA) consisting of five assertions; “Self-efficacy and observation score” (SCT) that evaluated the constructs of self-efficacy, collective efficacy and observation; the “Facilitator Score” (SCT) which assessed familiarity, the existence of target behaviours prior to intervention, and behaviours that facilitate behaviour change. Finally, constructs of the Health Belief Theory (“Health Belief Score”) (HBT), such as perception of susceptibility, severity, barriers and benefits associated with behaviour change, were analysed. All positive answers meant having a positive understanding or perception of the concept examined. The “Behaviour Change Theories Score” resulted from the sum of all scores and ranged from 0 to 25 points. This score allowed us to measure whether the participant changed his perceptions after the educational activities.
Statistical analysis
Descriptive statistics were used to describe the population profile and the frequency of each questionnaire response before and after the intervention. To assess whether there were differences in the median (categorical variables) or mean (normal distribution scores) values between the groups analysed; the Chi-square test, Fisher's exact test and ANOVA Kruskal Wallis and Friedman tests for independent and paired samples were used [19], considering the significance level of 5% (p-value <0.05). The reliability analysis of the instrument was performed using Cronbach's alpha. The analyses were performed with the aid of the R statistical software [20].
Secondly, multivariate analysis techniques were used, adjusting the Principal Component Analysis (PCA), whose eigenvalues were greater than one (λ> 1), as suggested by Kaiser (1960) [21], in order to identify a smaller number of variables. uncorrelated alternatives that somehow summarize the main information of the original variables. Subsequently, these main components were presented in Biplot graphs for individuals and variables with their respective confidence ellipses (with 95% reliability). Biplot is a method that represents two-dimensional multivariate data, where each observation is represented by the pair of scores of the first two main components, representing each group (students and agents) in their respective confidence ellipses.