Effectiveness of Aedes-borne infectious disease control in Latin America and the Caribbean region: A scoping review

Background Aedes aegypti and Aedes albopictus are primary vectors of emerging or re-emerging arboviruses that threaten public health worldwide. Many efforts have been made to develop interventions to control these Aedes species populations. Still, countries in the Latin America and the Caribbean (LAC) region struggle to create/design/develop sustainable and effective control strategies. This scoping review synthesises evidence concerning the effectiveness of Ae. aegypti and Ae. albopictus prevention and control interventions performed in LAC (2000–2021). The findings can be used to evaluate, compare and develop more effective control strategies. Methodology The review is based on the methodology by Joanna Briggs Institute for conducting a scoping review. The MEDLINE (via PubMed and Web of Science), Cochrane Library, Scopus, EMBASE and ScienceDirect databases were used to search for articles. Grey literature was searched from governmental and non-governmental organisation websites. Four reviewers independently screened all titles and abstracts and full-text of the articles using the Rayyan web application, based on pre-defined eligibility criteria. Results A total of 122 publications were included in the review. Most studies focused on dengue virus infection and data on Ae. aegypti. Entomological data were mainly used to determine the intervention’s effectiveness. An integrated control intervention was the most commonly employed strategy in both regions. Biological control measures, environmental management, and health education campaigns on community participation achieved more sustainable results than an intervention where only a chemical control measure was used. Challenges to implementing interventions were insufficient financial support, resources, workforce, intersectoral collaboration and legislation. Conclusions Based on the synthesised data, an integrated vector (Aedes) management focused on community participation seems to be the most effective approach to mitigate Aedes-borne infectious diseases. Maintaining the approach’s effect remains challenging as it requires multisectoral and multi-disciplinary team engagement and active community participation. Future research needs to address the barriers to program implementation and maintenance as data on this topic is lacking.


Unfunded studies
Enter: The author(s) received no specific funding for this work.  Describe where the data may be found in full sentences. If you are copying our sample text, replace any instances of XXX with the appropriate details.
If the data are held or will be held in a public repository, include URLs, accession numbers or DOIs. If this information will only be available after acceptance, indicate this by ticking the box below. For example: All XXX files are available from the XXX database (accession number(s) XXX, XXX.).
• If the data are all contained within the manuscript and/or Supporting Information files, enter the following: All relevant data are within the manuscript and its Supporting Information files.
• If neither of these applies but you are able to provide details of access elsewhere, with or without limitations, please do so. For example: Data cannot be shared publicly because of [XXX]. Data are available from the • The submitted manuscript is a scoping review. The findings and conclusions are based on published data. XXX Institutional Data Access / Ethics Committee (contact via XXX) for researchers who meet the criteria for access to confidential data. dengue virus, chikungunya and Zika viruses are also of concern. They have rapidly spread 82 throughout the LAC region in the last eight years, resulting in epidemics and high levels of 83 morbidity, creating an added burden on the region's health systems [7]. 84 Mosquitoes of the Aedes genus are considered the most important vectors of the mentioned 85 arthropod-borne viruses [8]. In the LAC, Aedes aegypti is the primary mosquito vector of 86 dengue, chikungunya, and Zika viruses [9]. In addition, Aedes albopictus (the Asian tiger 87 mosquito) has also been shown to be a competent vector of the above-reported viruses [10, 88 11]. Aedes albopictus is of medical importance due to its aggressive daytime human-biting 89 behaviour, ability to adapt to colder climates and live in artificial and natural containers close 90 to humans, resulting in disease transmission in new geographical areas. Since Aedes 91 albopictus is also present in the LAC, countries in this region must consider this mosquito 92 species' possible implications for transmitting viruses [12,13] Study design: This scoping review shows evidence from primary studies or journal articles or 215 organisational reports, or dissertations (e.g., cross-sectional studies-analytical, case-control 216 studies, cohort studies, randomised and non-randomised controlled trials, impact evaluations, 217 qualitative and mixed-methods studies). We decided not to include systematic or scoping or 218 narrative reviews and meta-analyses because we were interested in the singular study 219 findings. All such 'reviews' reference lists were searched for additional primary studies on the 220 research topic during the full-text screening stage. Conference proceedings, letters to the 221 editor, editorials, and commentaries were excluded. 222 Search: Two team members (PP and VJ) experienced in searching databases designed the 223 search strategy (S2 Table). Keywords were identified based on three domains; (i) mosquito-224 borne infectious diseases, (ii) location, and (iii) Aedes aegypti and Aedes albopictus 225 prevention and control interventions using Medical Subject Headings [MeSH], existing 226 literature reviews, and subject experts' opinions. The keywords were combined with 227 appropriate boolean operators to search for articles in electronic databases, and similar 228 keywords were used for grey literature. Proximity operators, truncation and wildcards were 229 used for keywords to increase the sensitivity of the search. The search was initially conducted 230 on PubMed and then tailored to other databases. PRESS checklist was used to review the 231 search strategy's quality [28]. 232 Selection of studies: The compiled search results were de-duplicated using EndNote 20 and 233 exported to rayyan.ai to screen blinded [29]. The reviewers stored the articles and other data 234 files using cloud storage technology. The selection of studies was performed by four reviewers 235 independently (VJ, MG, PP, TL), and disagreements were resolved in discussion with the 236 team. The articles were screened twice (i) in the title and abstract screening stage and (ii) in 237 the full-text screening stage. The selection process is presented in the PRISMA flow chart 238 [30]. Three reviewers (VJ, PP, TL) piloted data extraction on a sample of the included studies 239 (5% of the complete list of retrieved studies) to ensure that the data extraction technique was 240 consistently applied. The data extraction was conducted by three reviewers (VJ, PP, TL) using 241 a charting table. Data were extracted for the following variables: study identifiers, locations, 242 study design, methods, demography information, intervention, comparison, outcome, 243 challenges and recommendations. 244 A second reviewer (VJ) cross-checked each entry. The reviewers discussed doubts until 245 consensus was reached or consulted a fourth reviewer to resolve the disagreements. 246 Data analysis: The data were analysed using Microsoft Excel using basic formulae and data 247 for the figure. Frequencies and percentages were calculated for the variables of interest for 248 summarising the data, and a 'characteristics of included studies' table was prepared from the 249 data charting sheet. Advanced analysis was not performed, and the quality assessment of the 250 included studies was not carried out as both are not processes suggested by the JBI guidelines 251 [25]. 252

253
The results are presented here by the following key themes: (i) characteristics of the included 254 publications, (ii) differences and similarities between regions, (iii) interventions, (iv) challenges 255 and lessons learned, and (v) strategies employed or recommended to improve the 256 effectiveness of Aedes prevention and control measures. 257 258 Characteristics of the included publications 259 A total of 11,222 (11,118 articles from databases and 104 articles from other sources, e.g., 260 google or reference list search) articles were identified by the reviewers in the initial search. 261 After eliminating duplicates, the remaining articles were screened by reading their title and 262 abstract. One hundred twenty-eight articles from databases and 101 articles from other 263 sources were qualified and further screened by reading the full-text. Eighteen of the 128 264 articles from the databases and 89 articles from other sources were excluded based on the 265 exclusion criteria. A total of 122 articles were included in the scoping review. Fig. 1  Note: The reviewers included one report documenting interventions performed in countries 284 located in both Latin America and the Caribbean region. This report (USAID, 2019) is written 285 in English, focusing mainly on Zika, but dengue was also mentioned. The following 286 interventions were performed: biological and chemical control measures, environmental 287 management, usage of traps and health education. The outcomes were: entomological data, 288 removal of mosquito breeding sites, and improvement of knowledge, attitude and practices. 289 The study designs were classified as non-randomized control trials. 290 Demographic information: study participants' age, gender, and income or employment status. 291 * Economic modelling assessment. 292 ** Application of larvicide (biological and chemical): Bacillus thuringiensis var. israelensis, 293 Diflubenzuron, Novaluron, Spinosad, Temephos and Pyriproxyfen. 294 *** Integrated approach refers to a control strategy with more than three components (e.g., 295 health education, environmental management and application of larvicide. 296 **** Type of outcome: More than three types of outcome (e.g., self-reported dengue symptoms 297 and anti-dengue IgM seropositivity rates combined with entomological parameters) or a type 298 of outcome that has not been categorised, including removal of mosquito breeding sites, 299 community participation, maintenance of activities through capacity building, Disability-  (Table 1). The Aedes albopictus 307 mosquito has received more attention in Latin America than in the Caribbean. Chemical 308 control measures were often applied in both areas compared to biological control measures. 309 Mostly entomological data were used to determine the intervention's effectiveness, and it was 310 more often used in Latin America than in the Caribbean region (Table 1) populations [59]. No significant reduction in the Aedes aegypti adult population was proven, 318 but a significant reduction in egg density was observed in the second year compared to the 319 first year. In Puerto Rico, Bti was applied at a rate of 500 grams/ Hectare using vehicle-320 mounted aqueous wide-area larvicide spray applications [148]. This study found that Bti was 321 successfully deposited into jars in both open and covered locations. After the intervention, 322 differences in ovitrap densities were observed between the intervention and control group 323 resulting in 62% (P = 0.0001) and 28% (P < 0.0001) reductions in adult female Aedes aegypti 324 mosquitoes [148]. In Colombia, treating street catch basin with 2 grams of pyriproxyfen 325 (approx. 0.05 mg/mL) was associated with a reduction in the incidence of dengue (rate ratio 326 0.19, 95% CI 0.12-0.30, P < 0.0001) [73]. Two studies conducted in Brazil used pyriproxyfen 327 to control Aedes mosquitoes [51,55]. One study found that the mentioned intervention led to 328 a ten-fold decrease in adult mosquito emergence from sentinel breeding sites [51]. The second 329 study also showed the beneficial effect of pyriproxyfen. After the intervention, Aedes juvenile 330 catch decreased by 79%-92%, and juvenile mortality increased from 2%-7% to 80%-90%. Costa Rica, the application of ultra-low volume at the front door and in each room provided 364 three weeks of significant control (P < 0.05) [82]. A study in Mexico that used metofluthrin 365 emanators to reduce indoor adult Aedes aegypti abundance reported substantial reductions 366 in abundance rate ratios for total Aedes aegypti, female abundance, and females that 367 contained blood meals (2.5, 2.4, and 2.3-times fewer mosquitoes respectively; P < 0.001) 368 [105]. 369 370 Insecticide spraying led to a decline in Aedes mosquito density and incidence of ABIDs. The 371 reported beneficial effect is up to seven months; thus, data on the longterm effect is lacking. 372 The usage of heavy equipment led to more beneficial effects than portable equipment. Scheduled and periodic container washing led to a decline in Aedes (larvae) index in Peru 390 [111]. In Venezuela, no significant difference in reducing the Breteau index (0.84 with a 95% 391 CI: −8.94-10.62 and pupae per person index: −0.023 and a 95% CI: −0.749-0.703) was 392 observed after covering drums with insecticide-treated nets [119]. A study in Brazil indicated 393 that a long-term decrease in adult female population density was achieved only when water 394 tanks and metal drums were covered with nylon net [44]. Another study conducted in Brazil 395 that placed concrete in the bottom of storm drains indicated that after the intervention, water 396 accumulated in 5 (9.6%) of the storm drains (P < 0.001), none (0.0%) had immature forms of 397 Aedes species (P < 0.001), and 3 (5.8%) contained adults' mosquitoes (P = 0.039) [58]. Mexico showed the long-term (more than two years) benefits of using insecticide-treated 414 screens combined with treating the most productive breeding sites of Aedes aegypti [96]. In 415 Mexico and Venezuela, a combined approach (using insecticide-treated curtains and treating 416 water containers with pyriproxyfen chips or cover water containers) was also applied [107]. In 417 both countries, entomological indices after the intervention were significantly lower than 418 baseline. However, no significant difference between the control and the intervention group 419 was observed due to the spillover effect (an indirect effect on a subject/ area not directly 420 treated by the experiment  [116,117]. One study indicated that the vector densities in the intervention group, 435 on average, increased less than those in the control group (from spring to autumn) after 436 implementing the interventions (collection of small containers and covering of large 437 containers). However, the difference was statistically not significant [116]. In the other study, 438 the average pupae per person index decreased in the intervention group 11 times and in the 439 control group four times (P < 0.05). Although the difference was not statistically significant, the 440 container index, house index, and Breteau index decreased in the intervention group more 441 than those in the control clusters [117]. 442

443
One study in Cuba and Haiti used long-lasting insecticide-treated curtains or bed nets to 444 control Aedes mosquitoes [134,137]. In Cuba, no effect of the insecticide-treated curtains on 445 Aedes infestation levels (house index and Breteau index) was observed (study period 18 446 months) [134]. In contrast, the study in Haiti demonstrated significant differences between the 447 intervention group and the control group. At one month post-intervention (usage of insecticide-448 treated bed nets), all entomological indices declined (house index in the intervention group 449 declined with 6.7 (95% CI -10.6, -2.7; P < 0.01) and Breteau index reduced by 8.4 (95% CI - In general, environmental management, especially combined approaches (e.g. using 456 insecticide-treated screens and treating the most productive breeding sites), led to beneficial 457 and even longterm effects (≥ two years). However, it is crucial to be aware that the community 458 perceptions/ participation and negligence of potential mosquito breeding sites can negatively 459 affect the approach mentioned above's effectiveness. 460 461 Traps 462 In Latin America, eight studies that used traps as control measures were identified [37, 41, 42, 463 49, 53, 66, 71, 72]. Five studies were conducted in Brazil, and three were performed in 464 Colombia. One study conducted in Brazil found that sticky traps (MosquiTRAP)  The evidence on traps suggests that sticky traps are less effective than ovitraps (combined 497 with a type of insecticide) and traps to capture adult mosquitoes. Ovitraps and traps to capture 498 adult mosquitoes led to a significant reduction in entomological indices and a decline in the 499 incidence of ABIDs. 500 501 Genetically modified mosquitoes 502 One study released transgenic male Aedes aegypti mosquitoes with the OX513A line to 503 assess the related changes in the distribution of infestation and abundance of Aedes aegypti 504 populations six and eighteen months after the intervention in two areas in Brazil [56]. An 505 average suppression of ± 70% of the wild population due to the release of transgenic males 506 was observed in the mentioned study. In one of the areas, the mosquito population remained 507 suppressed for 17 weeks, whereas in the other area, the suppression lasted 32 weeks [56]. 508 The reported results highlight the benefits of using genetically modified mosquitoes in Aedes 509 mosquito control. 510 511 Health education and community mobilisation/participation 512 513 In Latin America, schools were a popular setting to provide health education concerning 514 diseases transmitted by the Aedes mosquitoes discussed in this scoping review [35, 69, 76, 515 78, 79, 87, 88, 92, 95, 99]. The health education interventions implemented at schools 516 increased awareness of Aedes biological characteristics [35,95] and dengue prevention and 517 control practices [69,87]. Also, infestation rates of immature forms of Aedes mosquitoes in 518 schools were reduced [76,78,88]. A health education program implemented in Mexico also 519 influenced the parents' behaviour [92]. According to the mentioned study, the entomological 520 indices decreased significantly (P < 0.05) in houses in the intervention area, apparently 521 because parents acted on the comments of the children and eliminated or monitored mosquito 522 breeding sites [92]. A study that evaluated a health education campaign's sustainability (two 523 years post-intervention) to prevent dengue in schools reported that the intervention's effects 524 were not sustained. A study that combined mass communication campaigns with a school and museum-based 539 educational program reported that exposure to the intervention was associated with increased 540 knowledge about dengue. [138]. Also, the intervention was associated with an increased 541 proportion of tires protected from rain and a decreased proportion of water storage containers 542 positive for mosquito larvae [138]. However, the evidence also suggests that only education provision is not enough to control 561 Aedes mosquitoes in the long run. Health education campaigns must include community 562 participation/ mobilisation efforts to be successful. 563 564

Integrated interventions 565
A study that evaluated the impact of the Brazilian national dengue plan (information 566 campaigns, epidemiological surveillance and vector reduction interventions, e.g., usage of 567 larvicide temephos and spraying of insecticide) indicated that the goals concerning the 568 reduction of dengue incidence (50% reduction in dengue cases) and larval infestation (less 569 than 1%) were not achieved in most municipalities [43]. Another study that evaluated Brazil Brazilian study [39]. Significant changes in KAP were achieved, and the house index was 581 reduced (6.9% before and after 4.4%, with a significant difference, P = 0.040) [39]. 582 583 A study in Mexico that combined epidemiological surveillance, environmental management 584 through social mobilisation, and chemical control also demonstrated beneficial results [98]. 585 Statistical significant differences (p < 0.001) in the relative abundance of Aedes aegypti larvae 586 before and after applying larvicide temephos and removing water containers were reported 587 [98]. For example, in La Paz, the average house index was reduced from 16 -83% to 0 -5% 588 after the interventions for three years. Beneficial results were also observed in the positive 589 container index and the Breteau index in La Paz and other areas (Cabo San Lucas and San 590 Jose del Cabo), where the interventions were implemented [98]. The beneficial effects of 591 combining interventions were also documented in another study conducted in Mexico [93]. 592 Dengue incidence decreased from 81.4% in 2010 to 79.1% in 2011 [93]. 593   594 Other studies also demonstrated the benefits of combined interventions [32, 40, 50, 62, 83, 595 84]. In Brazil, one study that combined the application of insecticide temephos, environmental 596 management and health education campaigns did not observe positive results [34]. The lack 597 of effect is possibly caused by negligence in eliminating breeding sites observed in the group 598 that also applied temephos and the dilution of the insecticide in non-removable containers. A 599 study in Nicaragua also did not demonstrate a positive effect of temephos combined with other 600 interventions [109]. In the mentioned study, temephos exposure was not associated with a 601 reduction in any entomological indices, and in some cases, temephos exposure was even 602 associated with higher entomological outcomes [109]. (i) Community participation enhances the success of Aedes control interventions [77, 684 84, 110, 117, 121, 122, 145]. Schools are an important setting that needs to be 685 considered when designing community participation strategies since beneficial 686 results have been booked with health education campaigns employed at schools 687 [35,48,78,99]. 688 The use of a multi-sectoral approach or inter-sectorial integration [39, 88, 89, 126, 689 127, 138], integrated Aedes control approaches [96,110,146], and prioritizing 690 ABIDs (local government providing financial and administrative support) [120, 121] 691 were stated as common reasons for the success of an intervention. 692 The effectiveness of the intervention improved when qualified staff planned and 693 implemented an intervention [45,67], and the community accepted the intervention 694 technique and believed in its effectiveness [106]. 695 (iv) Interventions were effective when health education messages were repeated 696 throughout the intervention [112]. 697 The health workforce should be trained and capable of identifying the most 698 productive breeding sites [73]. The workforce should keep standard operating 699 procedures for entomological and epidemiological surveillance, and data should 700 be consistent and follow standardised measurement tools for inspection activities 701 [67,127]. Furthermore, the workforce should assess the short term and long-term 702 impact of the Aedes control program because these assessments facilitate 703 program implementation in the long run and sustainability. To assess immediate 704 outcomes, periodic monitoring and evaluation are recommended to inform what 705 works and needs to be amended [67]. Having a list of program indicators could be 706 beneficial for program implementers to quantify and draw inferences on the 707 outcomes against the input, which can be helpful for the allocation of resources. 708 (vi) It is also important to consider conducting a cost-benefit analysis as the 709 intervention progresses, considering the direct cost of implementing the 710 intervention and the potential consequences [90]. 711 (vii) Lastly, the workforce should consider factors (e.g., meteorological conditions such 712 as wind speed, direction, or temperature) contributing to differences in larval 713 mortality related to application data [148]. Another factor that can invariably affect 714 implementation if not counter for is the seasonal variation of pupal productivity [85]. Aedes albopictus prevention and control interventions performed in LAC in the last twenty-one 721 years . To our knowledge, this review provides the most updated overview of 722 evidence regarding the effectiveness of Aedes prevention and control interventions in the 723 region mentioned above. Most of the data were from studies in Brazil, followed by Mexico and 724 Colombia. Cuba has most of the publications in the Caribbean region, followed by Puerto Rico 725 and Trinidad. The synthesised data showed that combined interventions were more effective 726 than a single approach. Another review supported our findings by reporting that it is unlikely 727 that any single intervention will be fully effective against Aedes mosquitoes and encourages 728 the development and usage of an IVM program to control the mentioned mosquito species 729 effectively [153]. The most important question that remains is, how can we sustain the effectiveness of the IVM 744 program? One way to counter this is through routine monitoring and evaluation, building the 745 capacity of the community, building a network of relevant stakeholders (local and 746 international), and taking the local challenges of the health system and the community into 747 account when designing the IVM program. However, all these would require multisectoral and 748 multi-disciplinary team engagement and active community participation. An IVM without 749 community participation is not sustainable [154,156]. 750

751
The effectiveness of the interventions included in this review was reduced by the following 752 factors, the spillover effect [75,107,137], small sample size [116], providing educational 753 strategies to deploy control measures in the control group [67], parallel government campaigns 754 [75,84,85,87,128,136], insecticide resistance [149] or media freedom to change or stop the 755 distribution of information [70]. Also, most interventions were followed for a short period, which 756 led to limited evidence on the sustainability of the employed interventions. Future research 757 should focus on Aedes Albopictus and its active role in transmitting ABIDs. Also, researchers 758 might consider studying the impact of an IVM approach on various ABIDs (e.g., dengue and 759 Zika combined) since this evidence is also missing in the scientific literature. Furthermore, 760 more research and dialogue on the barriers of an IVM program are needed to work toward 761 more sustainable programs. This scoping review also aimed to synthesise evidence on how 762 other countries in the LAC region dealt with their challenges, but this information was not 763 available in all the included publications. Therefore, we recommend future research to address 764 this topic to provide more practical advice to countries struggling with their Aedes control 765 approaches. Lastly, we encourage countries with more advanced techniques and qualified 766 workforce to support countries with less developed health systems. 767 Limitations and strengths 768 Since the quality of the included studies was not assessed in this scoping review, the included 769 studies' results need to be interpreted with caution. Quality assessment is essential for 770 systematic reviews and meta-analysis, but it is not required for scoping reviews [25]. Reiter P, Fontenille D, Paupy C. Aedes albopictus as an epidemic vector of 822