Arboviral diseases such as dengue, Zika, and chikungunya, transmitted by Aedes mosquitoes, remain an important public health concern in Colombian communities. Digital health tools such as WhatsApp may provide an opportunity to strengthen preventive behaviors and community engagement in vector control efforts. Therefore, a quasi-experimental implementation study was conducted in two endemic municipalities, Villa del Rosario and Los Patios, in Colombia, to evaluate a WhatsApp-based digital health strategy designed to support the prevention and control of Aedes-borne diseases and to promote the application of a protective coating (PC) in laundry tanks, one of the main breeding sites of Aedes mosquitoes.

The main questions it aims to answer are:

Whether a WhatsApp-based digital health intervention, added to community-based strategies, can improve household preventive practices against Aedes-borne diseases, compared with community strategies alone or routine vector control activities.

Whether the combined use of WhatsApp messaging and community-based promotion of protective coating in laundry tanks can reduce Aedes entomological indices, compared with clusters not receiving the full intervention.

Whether the intervention is feasible and acceptable for households and community participants in endemic urban settings.

The study was conducted in three geographically separated clusters of approximately 3,000 - 3,500 households each. Cluster 1 received community strategies plus WhatsApp messaging, cluster 2 received community strategies only, and cluster 3 served as the control group. Protective coating was applied in clusters 1 and 2. The study included three phases: a pre-intervention baseline assessment, an intervention phase with interim assessment, and a post-intervention final evaluation and follow-up. Household surveys and entomological inspections were conducted to assess preventive practices, vector indices, and acceptance of the intervention.

Arboviral diseases such as dengue, Zika, and chikungunya, transmitted mainly by Aedes aegypti, remain a major public health problem in Colombia and other tropical and subtropical countries. In endemic urban communities, prevention and control are challenged by persistent vector breeding in domestic water containers, limitations of routine vector control programs, insecticide resistance, and insufficient community participation in sustained source reduction activities. In the metropolitan area of Cúcuta, north-eastern Colombia, laundry tanks are among the most productive breeding sites for Aedes mosquitoes. In this context, innovative and acceptable community-based strategies are needed to strengthen preventive practices and support vector control interventions.

Recent evidence suggests that mobile health tools may help reinforce health behaviours and improve community engagement in disease prevention. WhatsApp is widely used in Colombia and represents a feasible communication channel for delivering reminders, health education, and behaviour-oriented messages at household level. In parallel, a protective coating (PC) applied to laundry tanks had been developed and evaluated as a novel vector control tool in the study area. Because successful application of the PC depends in part on household preparation and acceptance, this study assessed whether a WhatsApp-based digital health strategy could support its uptake and reinforce preventive practices for the prevention and control of Aedes-borne diseases.

The purpose of this implementation study was to evaluate a WhatsApp-based mHealth intervention designed to encourage preventive health behaviours and strengthen community engagement in vector control activities, particularly in relation to the application of protective coating in laundry tanks. The study was conducted in the municipalities of Villa del Rosario and Los Patios, both located in the metropolitan area of Cúcuta, Norte de Santander, Colombia. These municipalities are characterized by endemic transmission of dengue and other arboviral diseases, widespread use of domestic laundry tanks, and socioeconomic conditions that make sustained prevention especially relevant.

A quasi-experimental study with three geographically separated clusters was conducted. Cluster 1 received the full intervention, consisting of community strategies plus WhatsApp messaging (CS+WAM). Cluster 2 received community strategies only (CS), without WhatsApp messaging. Cluster 3 served as the control group and continued to receive routine vector control activities. PC was applied in clusters 1 and 2. Each cluster included approximately 3,000 households and was selected from highly endemic urban communities with similar sociodemographic characteristics to reduce contamination and enhance comparability across groups.

The study population consisted of households located in the selected clusters. For household surveys and follow-up evaluations, one adult household member, usually the household head or another adult resident, was recruited in each selected household. Systematic sampling was used for baseline and final evaluations, with a target of 300 households per cluster, for a total of 900 households. Baseline and follow-up data collection included household questionnaires and entomological inspections. Additional recruitment activities in the intervention clusters were supported by female community leaders, community volunteers, and vector control staff, who also assisted with the implementation of the PC.

The study was implemented in three main phases. First, a pre-intervention phase was conducted to collect baseline information on household sociodemographic characteristics, preventive practices related to dengue, Zika, and chikungunya, and entomological indices. Second, the intervention phase included delivery of the WhatsApp-based communication strategy in cluster 1, promotion of the protective coating in clusters 1 and 2, and an interim assessment approximately four months after application of the coating. Third, a post-intervention phase was conducted to evaluate final changes in preventive practices, entomological indicators, and acceptance of the intervention. Baseline data collection took place in September 2021, the WhatsApp intervention started in October 2021, the protective coating was applied in December 2021, and the final evaluation was completed in September 2022.

The WhatsApp-based component was developed as a behaviour reinforcement and health promotion strategy. Participants in cluster 1 who consented to receive messages were sent periodic WhatsApp messages, generally on a weekly basis, focused on prevention of dengue, Zika, and chikungunya and on preparation for the application of the protective coating. Reminder messages were sent before household visits for coating application. The intervention was informed by the Communication for Behavioural Impact (COMBI) framework and by national dengue prevention guidance. Messages were brief, written in Spanish, and designed to be understandable and engaging for community users. When needed, additional voice messages were provided by a facilitator to answer questions or clarify intervention procedures. This communication strategy was intended for health promotion and behaviour support only and did not provide medical care.

The partial intervention in cluster 2 consisted of conventional community strategies without WhatsApp messaging. Community volunteers promoted the protective coating through household visits, flyers, stickers, and direct practical instructions about how to prepare the laundry tank before application. In both intervention clusters, trained community volunteers and vector control staff supported the application of the coating and monitored implementation. Households in the control cluster did not receive WhatsApp messages or PC and continued with standard routine vector control activities.

Household surveys collected information on demographic characteristics, water storage practices, preventive behaviours, and acceptance of the intervention. Entomological inspections were conducted using standardized forms and World Health Organization operational procedures to identify water-holding containers and detect larvae and pupae. The main entomological indicators included the House Index (HI), Container Index (CI), Breteau Index (BI), and Pupae per Person Index (PPI). Acceptability of the WhatsApp component was assessed using Likert-scale items, while acceptance of the protective coating was assessed through household uptake and willingness to reapply the product.

Data analysis was conducted using RStudio. Descriptive statistics were used to summarize household and entomological data. Changes in preventive practices before and after the intervention were assessed using McNemar tests. Multivariable logistic regression models were used to identify factors associated with post-intervention preventive behaviours. A difference-in-differences approach was used to compare changes over time between intervention and control clusters. Entomological indicators were calculated per cluster and compared between study phases to evaluate changes in vector density.

All participants were informed about the purpose of the study and provided informed consent before participation. Only adults were interviewed. Household data were anonymized by replacing personal identifiers with codes. The study protocol received ethical approval from the Albert-Ludwigs-Universität Freiburg, Germany (application number 142/19), and from the National Institute of Health in Bogotá, Colombia, with support from local health authorities in Cúcuta and Norte de Santander.

Inclusion Criteria:

• Households/participants resided within one of the study clusters.
• The household had a laundry tank.
• One adult household member (household head or another adult resident) was available to participate and provide informed consent.
• For participation in the WhatsApp component, the participant owned a smartphone with WhatsApp installed and was able to read and respond to simple WhatsApp messages.

Exclusion Criteria:

• Uninhabited households.
• Households without a laundry tank.
• Households already using biological vector control methods in the laundry tank, particularly larvivorous fish.
• For the WhatsApp component, households or users without access to a smartphone or without WhatsApp installed.

De-identified household-level data generated during this study may be shared partially through publications and supplementary materials, and additional data may be made available upon reasonable request to the corresponding author. Because the unit of analysis was the household, no direct personal identifiers, including names, phone numbers, addresses, or WhatsApp contact information, will be shared. Shared data may include household survey variables, preventive practices, acceptability measures, and entomological indicators used in the analyses.

Up to 12 months after publication.

Access will be limited to de-identified household-level data and will be granted only for research purposes, upon reasonable request to the corresponding author and in accordance with the informed consent provided by participants and the ethical approvals of the study. No directly identifiable participant, household, or WhatsApp contact information will be released.