Eder et al. Infectious of Poverty (2018) 7:90 https://doi.org/10.1186/s40249-018-0475-7

SCOPING REVIEW Open Access Scoping review on vector-borne diseases in urban areas: dynamics, vectorial capacity and co- Marcus Eder1,2, Fanny Cortes3, Noêmia Teixeira de Siqueira Filha4, Giovanny Vinícius Araújo de França5, Stéphanie Degroote6, Cynthia Braga2, Valéry Ridde6,7 and Celina Maria Turchi Martelli2*

Abstract Background: Transmission dynamics, vectorial capacity, and co- have substantial impacts on vector-borne diseases (VBDs) affecting urban and suburban populations. Reviewing key factors can provide insight into priority research areas and offer suggestions for potential interventions. Main body: Through a scoping review, we identify knowledge gaps on transmission dynamics, vectorial capacity, and co-infections regarding VBDs in urban areas. Peer-reviewed and grey literature published between 2000 and 2016 was searched. We screened abstracts and full texts to select studies. Using an extraction grid, we retrieved general data, results, lessons learned and recommendations, future research avenues, and practice implications. We classified studies by VBD and country/continent and identified relevant knowledge gaps. Of 773 articles selected for full-text screening, 50 were included in the review: 23 based on research in the Americas, 15 in Asia, 10 in Africa, and one each in Europe and Australia. The largest body of evidence concerning VBD in urban areas concerned dengue and . Other covered included and West Nile , other parasitic diseases such as and trypanosomiasis, and bacterial rickettsiosis and . Most articles retrieved in our review combined transmission dynamics and vectorial capacity; only two combined transmission dynamics and co-infection. The review identified significant knowledge gaps on the role of asymptomatic individuals, the effects of co-infection and other factors, and the impacts of climatic, environmental, and socioeconomic factors on VBD transmission in urban areas. Limitations included the trade-off from narrowing the search strategy (missing out on classical modelling studies), a lack of studies on co-infections, most studies being only descriptive, and few offering concrete public health recommendations. More research is needed on transmission risk in homes and workplaces, given increasingly dynamic and mobile populations. The lack of studies on co-infection hampers monitoring of infections transmitted by the same vector. Conclusions: Strengthening VBD surveillance and control, particularly in asymptomatic cases and mobile populations, as well as using early warning tools to predict increasing transmission, were key strategies identified for public health policy and practice. Keywords: Arboviruses, vectors, Coinfection, Urban population, Epidemiology, Review

* Correspondence: [email protected] 2Aggeu Magalhaes Institute (IAM) / Oswaldo Cruz Foundation (Fiocruz), Avenida Professor Moraes Rego, s/n. Cidade Universitaria. CEP 50, Recife, Pernambuco 740-465, Brazil Full list of author information is available at the end of the article

© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 2 of 24

Multilingual abstracts trypanosomiasis (), leishmaniasis, and Please see Additional file 1 for translations of the filariasis, have affected hundreds of millions of people abstract into the six official working languages of the globally [12]. United Nations. Approximately half of the world’s population currently live in cities. The United Nations projects that 2.5 billion Background people will be added to the urban population by 2050, According to the World Health Organization (WHO), mostly on the Asian and African continents [13]. This vector-borne diseases (VBDs) account for more than rapid and increasing urbanization has posed a great 17% of all infectious diseases and cause more than 1 mil- challenge to nations, especially those less developed [14]. lion deaths annually [1]. Vector-borne diseases are trans- Urbanization has had an impact on the epidemiological mitted from person to person via a competent vector, pattern of infectious diseases. The main factors are such as mosquitoes, midges and . urban sprawl into forested areas, overcrowding, and pre- Transmission dynamics describes a range of factors carious urban infrastructures and housing in urban areas influencing how effectively transmission occurs over of developing countries. The absence of necessary in- space and time, and in a specific population. These vestments in infrastructure in these countries poses a factors include basic reproduction number, host im- serious threat to human health, including the (re-)emer- munity, travel and human behaviour. Transmission gence and adaptation of infectious agents in urban areas dynamics are determined by the interaction between such as dengue in South East Asia or, Chagas in Latin , vector, host (human, and in many cases America in areas where poor housing is hindering effect- also other animals, serving as reservoir or amplifier) ive [14–16]. and other environmental factors [2]. Basic knowledge about VBD transmission includes Vectorial capacity refers to the ability of a population susceptibility, vectorial capacity, and inter- population’s to transmit the pathogen to a new suscep- action of infectious agents. The understanding of VBD tible population [3]. transmission and persistence is essential for establishing The term co-infection describes human infection effective prevention and control interventions. Of similar through more than one organism, either by different importance is to know key aspects of introduction, strains of the same (e.g. two genetically different falcip- maintenance, and spread of VBDs, as well as the role of arum malaria protozoa), or entirely different environmental and climate factors, the urbanization (e.g. falciparum malaria protozoa and intestinal hel- process, socioeconomic conditions, population dynamics minths). Here, also co-circulation is considered, when and mobility [2, 17–20]. more than one different pathogens are present in an This scoping review evaluated the current state of vector (e.g. in mosquito populations of a specific knowledge on transmission dynamics, vectorial capacity, region) [4]. and co-infection regarding VBDs in urban areas from Malaria is a VBD that caused over 400 000 deaths in 2000 to 2016, to identify research gaps and implications 2015, most of them in children under 5 years of age [5]. for public health policy and practice. Traditionally associated with rural transmission, malaria is increasing found in urban and peri-urban areas [6, 7]. Main text An entomological marker of malaria transmission is the Research question entomological inoculation rate (EIR). It describes the We conducted a scoping review adapting Arksey and number of infected bites per unit of time, and a function O’Malley’s[21] methodological framework. A three-round of the so-called ‘man biting rate’ (MBR, the number of eDelphi survey was used to select six topics considered bites per person per time unit) and the sporozoite rate highest priority by a panel of 109 international VBD ex- (rate of infected mosquitoes, i.e. those carrying malaria perts, the majority of them being from Brazil, Burkina parasites ready to infect humans). Faso, Canada, Colombia, France, Spain and the United Currently dengue, a virus transmitted through Aedes States of America (43% researchers; 52% public health mosquitoes, threatens a half-billion people globally [8]. decision-makers; 5% from the private sector). The three Unlike , where sylvatic (forest) mosquito spe- rounds were: 1) suggestions of research topics; 2) cies and non-human primate reservoirs play a critical ranking of topics identified (more than 80 topics, role in the transmission, dengue only requires humans, a rated from “1–eliminate” to “5–top priority”); and 3) fact that explains its rapid spread in populated urban final selection of highest-priority topics (the 20 subjects areas [9]. Dengue has increased dramatically rated 4 or 5 by more than 65% of the participants). By the in the Americas, and recent introductions of chikun- end of the third round, the present topic—the impact of gunya and Zika have resulted in serious in transmission dynamics, vectorial capacity, and co-infections these regions [10, 11]. Other VBDs, such as American on the burden of vector-borne diseases in urban areas— Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 3 of 24

had obtained the mean rating of 3.90 ± 0.92 and was ranked Data extraction, summary, and analysis fourth. It was therefore among six top rated topics taken An extraction grid was created allowing to record for forward for research carried out by the consortium groups. each of the selected studies the following information: general information, key objectives and methods; over- view of results; methodological limitations and chal- Search strategy lenges encountered in lessons learnt/recommendations; We used the following key concepts: [“transmission dy- future research avenues; and, public health policy or namics” OR “vectorial capacity” OR “co-infection”]AND practice implications. Similarly, the methodological and “vector-borne” AND “urban areas” AND “epidemiology”. quality aspects of each study were evaluated using the All possible word variations and MeSH terms (as appro- modified Mixed Methods Appraisal Tool (MMAT; for priate) were added to the search command and validated description of qualitative, quantitative, and mixed by a librarian (see Additional file 2) for the following da- methods studies) [22] and parts of the TIDieR (Template tabases: PubMed, Embase, Global Health, Cochrane for Intervention Description and Replication) checklist Database of Systematic Reviews, OpenGrey, the Grey [23]. Summary tables and graphs were produced. Ini- Literature Report, and WHOLIS. Additional articles tially, the three contributors (FC, ME, NTSF) independ- were identified by screening the references of papers that ently extracted data from the same five articles, to met our inclusion criteria. As part of the protocol devel- ensure harmonization. Any remaining difficulties were opment the consortium members considered the 2014 resolved in a discussion with the remaining two partici- World Urbanization Prospects issued by the Population pants. Subsequently, the remaining 45 articles were sum- Division of UNDESA [13]. marised with quality assessed by the same three The literature search was undertaken from August to contributors and results recorded in the extraction grid. September 2016. We used Mendeley and Endnote soft- ware to manage references and remove duplicates. Results Description of included studies and their funding sources The search strategy initially identified 9239 records. After Inclusion and exclusion criteria removing duplicates and articles published before 2000, We included all articles and reports published in we screened 3365 articles by title and abstract and re- peer-reviewed journals or grey literature written in trieved 773 of them. After full-text screening, 50 articles English, French, Portuguese, Spanish, German, or Italian were selected for the scoping review (Preferred Reporting and published between 2000 and 2016. We excluded: Items for Systematic Reviews and Meta-Analyses articles focused on clinical or laboratory characteristics, [PRISMA] flowchart, Fig. 1). vector or seroprevalence only; reviews; con- Most of the 50 studies retained were conducted in ference papers; articles without research data; articles the Americas (n= 23; 46%), followed by Asia (n= 15; not addressing human disease; articles reporting 30%), Africa (n= 10; 20%), Europe (n= 1; 2%) and water-borne diseases or diseases without insect vector; Australia (n= 1; 2%) (continents, Table 1;countries, studies conducted in rural areas; and interventional Additional file 3; map, Fig. 2). Selected articles were orga- studies, such as mass drug administration, intermittent nized into three groups of diseases: 1) dengue (n= 20; preventive treatment, and vector control programs. 40%), 2) malaria (n= 15; 30%), and 3) others (n= 15; 30%), which included parasitic diseases: leishmaniasis (n= 4) and Chagas disease (n= 2); other arboviruses: Study selection chikungunya (n= 2), (n= 2), yellow We performed a pilot round of study selection to evalu- fever (n=2), and (n=1); and two bacterial ate consistency in the application of the above criteria diseases: plague (n= 1) and rickettsiosis (n= 1) (Table 1). and discuss discrepancies with 20 randomly selected ref- Two studies reported on co-infections, one on multiple erences. For both abstract and full-text screening, two Plamodium falciparum strains, the other on combined independent reviewers (FC and NTSF) selected the stud- malaria, helminth, and human immunodeficiency virus ies through the title and abstract/full-text, and a third (HIV) infection in pregnant women. Studies are summa- reviewer (ME) resolved discordances. rized in Table 2. After completing full-text screening for 205 articles, Studies were funded mostly through national (n= 21; an additional step was introduced to retain references 41%) and international (n= 15; 29%) government that combined at least two elements of the search strat- sources, followed by universities, non-government or- egy: transmission dynamics and vectorial capacity or ganizations, and global funding sources (< 10% each). transmission dynamics and co-infection. This last step Only one study was funded through pharmaceutical was done manually by the reviewers. companies, but did not involve clinical trials (hence Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 4 of 24

Fig. 1 Prisma Chart showing references retrieved at different stages of the search (full text review) was not excluded); some studies had several funding group shows that nearly half of the studies on dengue sources (Fig. 3). used either spatial (30%) or dynamic (15%) modeling, We transferred information for the 50 included studies followed by one third (20% and 13%, respectively) on into an extraction grid. All studies were descriptive. An studies on malaria.. The remainder of malaria research overview of study methods employed in each disease included mostly cross-sectional (n= 4; 27%) and cohort

Table 1 Final selection of N = 50 references: Group of diseases: dengue, malaria and others (ordered by parasitic, viral and bacterial diseases) by continent Continent Dengue Malaria Leishmaniasis Chagas disease West Nile Chikungunya Yellow fever Ross River virus Plague Rickettsiae Total Americas 12 2 3 2 2 0 1 0 0 1 23 Europe 0 0 0 0 0 1 0 0 0 0 1 Africa 1 8 0 0 0 0 1 0 0 0 10 Asia 7 5 1 0 0 1 0 0 1 0 15 Australia 0 0 0 0 0 0 0 1 0 0 1 Total 20 15 4 2 2 2 2 1 1 1 50 Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 5 of 24

Fig. 2 Distribution of final selection of N=50 references by geographic regions studies (n= 6; 40%). Studies on other pathogens were variation in relation to vector abundance, and the role of mostly outbreak (n= 7; 47%) and other case-control socioeconomic conditions. The role of imported dengue (n= 3; 20%) studies (Table 3). fever cases in triggering outbreaks in non- cities We applied the MMAT to evaluate study quality [22]. was highlighted [24]. Human movement due to eco- All studies had clear objectives set out, which were ad- nomic development and/or tourism was considered a dressed in 90% (n=18) of dengue studies and 87% (n=13) determinant for the spread of dengue infection [24–26]. of studies on other VBDs. Relevant sampling strategy for studying human or vector characteristics was present in Incidence of dengue cases or dengue virus infection approximately 65% of studies on dengue and other patho- The heterogeneity of dengue transmission in inner cities gens, but to a lesser degree (n= 4, 27%) in malaria work. seemed to be a common feature of the studies. Population Representation of the population under study was also bet- immunity and asymptomatic infection play an important ter addressed in studies on dengue and other pathogens role in dengue transmission dynamics, resulting in higher (around 70%) than in malaria studies (n= 6; 40%). Appro- incidence of dengue infection in previously lower preva- priate measurement was captured well in both dengue and lence areas [25]. Also, the intensity of transmission in malaria studies (n=17, 85% and, n=13; 87%, respectively). highly urbanized settings may not be perceived as an epi- Response rate (where appropriate) was clearly reported demic due to asymptomatic infection, suggesting the oc- only in about 25% of dengue studies and even less in other currence of a “silent ”, as shown in Salvador city, work (Fig. 4). Given the absence of any intervention stud- Brazil (2008–2009) [25]. Another prospective study dem- ies, the TIDieR tool was only applicable to very limited as- onstrated that dengue infection in the was pects of the included studies. Due to the limited added mainly spread by asymptomatic adults [27]. In concord- benefit, it was therefore agreed to not consider TIDieR fur- ance with these studies, asymptomatic dengue cases were ther in the extraction. also a potential source of subsequent outbreaks, as seen in four cities of Valle del Cauca, Colombia [26]. A study con- Description of findings of the scoping review ducted in Ho Chi Minh city, Vietnam, provided evidence Dengue transmission dynamics and vectorial capacity for some household dengue risk clustering, but on a short studies temporal scale rather than as sustained chains of localized Dengue research was identified mostly in the Americas transmission [28]. These findings are important for sur- and Asia. The majority of the studies (n= 17) evaluated veillance and control strategies [25–27]. the relationship between the incidence of dengue cases and vector density in endemic areas, and/or mobility of Mobility of human populations as a source of outbreaks the human population. A large number of studies also and/or disease persistence addressed asymptomatic infection as a factor related to The study conducted in Mexico City (2011–2012) pointed the spread of dengue virus infection, the climatic out that, besides asymptomatic individuals, human mobility Table 2 Description of included studies, by disease Eder

Authors, Date VBD Study type Objectives (outcomes Results and conclusions Learning points and Main limitations, Future research, public Funding Poverty of Diseases Infectious al. et Country measures) recommendations: comments health policy & practice Region transmission dynamics, City vectorial capacity, co- infection Dengue AMERICAS Araujo et al., Dengue Spatial modelling To assess incidence of Dengue incidence was Incidence mostly affected None reported. Importance of evaluating NR 2015 [39] Brazil dengue and its association higher in areas of high by land surface LST (environmental inner Southeast with multiple land surface temperature temperature (LST), due to city conditions), slum-like São Paulo environmental and (28–32 °C) and in slum-like Ae. aegypti proliferation. areas and human activities socioeconomic factors areas. Most cases occurred Recovering vegetation on (man-made reservoirs) for

in areas with low vegetation urban heat island (UHI). DENV transmission. (2018)7:90 cover. Translational study in Laboratory vector dengue. experiments indicated that higher temperature was favourable for Ae. aegypti proliferation. Dibo et al., Dengue Ae. aegypti female Almost 20 000 specimens Relationship between Using information about Widespread application in NR 2008 [35] Brazil abundance (collection) of culicids were caught, ~ climatic factors, vector and reported planning and undertaking Southeast and eggs (oviposition 10% were Ae. aegypti. The disease. Use of information (DF) cases may represent surveillance and dengue Mirrasol traps); sensitivity of both ovitrap caught 165 000 about climate for early only part of the total control activities methods; correlation of eggs in the period and detection of epidemics number of dengue virus entomological indices 173 autochthonous cases and for establishing more infections, rainfall data (positivity and averages of of dengue were recorded. effective prevention were obtained for a females and eggs) dengue strategies. location not exactly at the incidence and climate. mosquito collection site, and temperatures were available only for a neighbouring city. Teixeira et al., Dengue Ecologic spatial modelling Dengue spatial distribution Direct association was Significant direct A methodological Need to investigate and NR 2011 [41] Brazil and its relationship to found between dengue association between limitation was data analyze the association Southeast environmental and incidence and rainfall, in dengue incidence and georeferencing, due to the between dengue Rio de Janeiro socioeconomic variables both final generalized rainfall +lag, Gini index, incomplete notification incidence and explanatory linear model and some and Breteau index for Ae. data. variables through more monthly CAR models. albopictus, but only the Use of Breteau index as an complex models Direct association was also Gini index showed strong indicator of vector (complete Bayesian model) found between dengue association. infestation since it was capable of simultaneously incidence and a 1-month Importance of socio- considered precarious and capturing the spatial and rainfall time lag. environmental variables in not a reliable indicator. temporal autocorrelation the dynamics of structures. dengue transmission. Corrêa et al., Dengue Ecologic October Link between vector Ae. aegypti infestation was Higher building infestation Building infestation rates To reduce and maintain NG 2005 [34] Brazil 1997– May 2001 infestation index and positively associated with rates were associated with have limitations for the house infestation Southeast dengue incidence dengue incidence. higher risk of disease. This estimating vector index below the 1% Belo Horizonte correlation was significant infestation. threshold.

but weak. 24 of 6 Page Teixeira et al., Dengue Prospective populational Relationship between Wide range of prevalence Increase in Some years after the virus’ Further studies are NG 2002 [25] Brazil study, seroprevalence and intensity of viral circulation of previous dengue to two different serotypes introduction, transmission required to identify the Northeast incidence after 1 year of and standard of living, infection (16.2% to 97.6%) of dengue, and continues to occur, but behavioural and Salvador follow-up environmental quality and in the area. Areas with a maintenance of the with a decrease in the environmental risk factors Table 2 Description of included studies, by disease (Continued) Eder

Authors, Date VBD Study type Objectives (outcomes Results and conclusions Learning points and Main limitations, Future research, public Funding Poverty of Diseases Infectious al. et Country measures) recommendations: comments health policy & practice Region transmission dynamics, City vectorial capacity, co- infection vector density lower prevalence at environmental conditions perception of the problem (public and private baseline presented higher necessary and sufficient for by individuals and health domains) that have the incidence of dengue transmission of new professionals and hence greatest influence on infection. High serotypes, created reduced sensitivity of the transmission. seroincidence was conditions for occurrence surveillance system. observed in areas with low of epidemic haemorrhagic vector indices. dengue. Méndez et Dengue Prospective study Study dynamics during Infections in students: Virological surveillance for Asymptomatic infections Need to understand a) NG al., 2006 [26] Colombia and after epidemics; asymptomatic cases detecting infected Aedes potential source of inter-outbreak virus (2018)7:90 Valle del Cauca mosquito infection rates, outnumber symptomatic mosquitoes as early subsequent outbreaks; to circulation and b) Ae. albo- Cities: Cali, cases, serotypes (infection (silent circulation of DENV). warning system for include pooled mosquito pictus vectorial capacity. Palmira, Tulua, rates) Infections in mosquitoes: outbreaks. infection rates in Buenaventura Ae. aegypti adult mosquito entomologic surveillance. and larvae house indexes were not found to be associated with increased burden of disease. Estallo et al., Dengue Descriptive Spatio-temporal dynamics Spatio-temporal cluster of Significant risk factor for Applications of vector Need to develop NG, U 2014 [29] Argentina of DF outbreak 2009 dengue cases. Outbreak dengue emergence: travel, control measures (focal innovative strategies of Cordoba started because of migration and intervention at dengue vector control and imported cases from displacement within and positive houses) may have arboviral surveillance to neighbouring provinces. outside the country. interrupted DENV prevent future outbreaks. Ae. aegypti infestation Dengue virus spread may transmission. levels were not associated be related to human with occurrence of the movements. cases. Martinez- Dengue Prospective cohort with To determine risk in At baseline, seroprevalence Asymptomatic subjects Spread of infection within Clarify the role of NG, PI Vega et al., Mexico City spatial modelling individuals living near of infection was ~ 44% could participate in the community mainly asymptomatic individuals 2015 [26] 2 endemic dengue cases (exposed (5–14 yrs) and 92% (> transmission; peri-domestic depended on human in transmission. Private communities cohort); vector data 35 years) indicating high transmission was import- mobility; limited role of health centres should exposure levels. 3-fold in- ant for ~ 3 months. vector (50-m ratio) for contribute to notifications. creased risk of dengue in- dengue spread. fection among the exposed cohort. Reiter et al., Dengue Two serosurveys Investigations of 1999 Incidence of dengue was Prevalence of dengue in Need to explore role of Growing economy NG 2003 [37] USA: outbreak affecting cross higher in Mexico (16%) Texas was primarily due to other animals as feeding supporting use of air Laredo border urban area of than on USA side (1.3%). economic, rather than source for Aedes conditioning likely to Mexico: Laredo, USA, and Nuevo Vector was more climatic, factors. mosquitoes. decrease dengue rates. Nuevo Laredo Laredo, Mexico. abundant in Texas where dengue transmission was lower. Barrera et al., Dengue Longitudinal modelling Influence of weather and Containers for water Weather and human Passive surveillance system Consider using ovitraps for IG

2011 [32] Puerto Rico study human factors on Aedes storage and discarded tires factors driving Ae. aegypti captures small number of surveillance. 24 of 7 Page San Juan vector and cases were considered dynamics and oviposition. symptomatic cases and (numbers) important breeding sites. misses most asymptomatic Peak in mosquitoes’ ones. density preceded increase in dengue incidence. Table 2 Description of included studies, by disease (Continued) Eder

Authors, Date VBD Study type Objectives (outcomes Results and conclusions Learning points and Main limitations, Future research, public Funding Poverty of Diseases Infectious al. et Country measures) recommendations: comments health policy & practice Region transmission dynamics, City vectorial capacity, co- infection Sanchez et Dengue Case control study Study Aedes larval indices Effect of Breteau index (BI) Analysis of BI at local level, Underestimating larvae; Similar studies in future NG, IG al., 2006 [33] Cuba and risk of Havana within a radius of 100 m at with human defined increased capture after epidemics and in other Havana outbreak (2000) 2-month intervals. boundaries, could be dengue reported; ‘missing’ settings are necessary to Larval indices used in introduced in control cases reported; lack of verify general applicability. study predicted programs to identify high transferability to other transmission risk areas. epidemic settings or with 78% sensitivity and locations. 63% specificity.

Fouque et al., Dengue Descriptive Epidemiology of dengue DENV1, 2 and 4 were Vertical transmission of Females deposit small Understanding of endemic NG (2018)7:90 2004 [40] French Guiana after the 1st epidemic of isolated from dengue DENV indicated that the numbers of eggs in several circulation of dengue is a dengue hemorrhagic fever cases. DENV4 from vector can be a reservoir breeding containers. Not prerequisite for (DHF) (1991–1993) and mosquitoes. of dengue virus. possible to estimate development of a dengue during endemic period Widespread circulation of fluctuations in Ae. aegypti early warning system. (1993–1995) DENV in urban and rural populations. areas. Small outdoor containers were most common vector breeding grounds. ASIA Wu et al., Dengue Time series modelling Association between Incidence of DF was Weather variables Change in: disease Include development and NG 2007 [43] China weather variability and negatively associated with (temperature and diagnostic or classification verification of analytical Taiwan dengue fever monthly temperature humidity) were criteria, urbanization models appropriate for Kaohsuing deviation and relative significantly associated status/land use, vector predicting influences of humidity; time lag of with dengue incidence control program, and global warming on 2 months. rate. personal protection. changes in disease Vector density did not patterns at regional level. appear to be a good predictor of disease occurrence. Chang et al., Dengue Time series regression Temporal relation between Risk of increased number Proposed 2-stage model Complex relation between Suggestion for other NG 2015 [42] Chian modelling study local weather, entomology, of cases was associated suitable to identify best set human hosts, countries to apply the Taiwan and confirmed cases with entomologic indices of lag effects to environmental factors and authors’ proposed 2-stage Kaohsuing, on a lag of 2 weeks or generate outbreak dynamic changes in modelling system. Tainan (non- 1 month. prediction. mosquito density. endemic areas) Difficult to extrapolate to other settings with distinct weather conditions, human immunity profiles and vector distributions. Ali et al., Dengue Spatial modelling study Association of spatial Dengue clusters far less Case clusters and Ae. Hospital-based surveillance Controlling the Aedes NG, IG, 2003 [36] Bangladesh clusters and vector during observed in major albopictus were linked in underestimates disease mosquito; improving case NGO Dhaka large outbreak hospitals than within or densely populated area. burden; self-reporting management.

(distribution; relative risk) around the households; cases limit precision. 24 of 8 Page importance of Ae. albopictus in the transmission. Table 2 Description of included studies, by disease (Continued) Eder

Authors, Date VBD Study type Objectives (outcomes Results and conclusions Learning points and Main limitations, Future research, public Funding Poverty of Diseases Infectious al. et Country measures) recommendations: comments health policy & practice Region transmission dynamics, City vectorial capacity, co- infection Anders et al., Dengue Prospective spatial Distribution of dengue risk At baseline, there was Evidence for some Short follow-up: longer Reactive peri-focal U 2015 [28] Vietnam modelling study (prevalence, odds ratios) 2-fold higher risk in case household dengue risk focal transmission chains insecticide spraying Ho Chi Minh clusters than controls. clustering, on short missed? Self-reporting: unlikely to prevent many City Follow-up (14 days) temporal scale rather than underestimating additional infections. suggested no excess risk sustained chains of symptomatic infections? for dengue infection in localized transmission. Vector control applied clusters. Prevalence of after onset of illness DENV infection in Ae. causing reduced clustering aegypti was similar in cases observed? (2018)7:90 and control houses, and low (1%). Yang et al., Dengue Descriptive of dengue Factors determining the Dengue incidence had no Social factors and hygienic Not reported. Public education to NG 2009 [30] China outbreak dengue outbreak and relationship with local conditions for endemic include handling of Cixi measures to avoid meteorological factors. Ae. villagers and immigrant environmental factors additional epidemic albopictus was the main workers. (artificial containers with outbreaks vector. stagnant water). Socioeconomic conditions should be taken into account to interrupt transmission. Peng et al., Dengue Descriptive of outbreak Describe epidemic, risk : ~ 51/100000 Urbanization, population Underreporting and/or Importance of a NG 2012 [31] China factors and control Ae. albopictus was only density, habitats and misclassification of surveillance system for Dongguan measures vector species responsible habits favouring Ae. dengue. infectious diseases control. for the outbreak. DENV1 albopictus; high infestation, Surveillance needed in serotype. poor housing. rapidly urbanized areas Dengue outbreak initiated and among immigrants. by imported case from Southeast Asia. Sang et al., Dengue Modelling study Predicting local dengue Seasonal pattern. Imported DF cases, Relation between herd Establishment of an early NG 2014 [24] China transmission in Imported dengue cases, mosquito density were immunity, mosquito- warning system Guangzhou Guangzhou, China, mosquito density and critical for DENV human interaction, virus (incorporating imported influence of imported climate variability are transmission. strain and mosquito daily cases, mosquitoes density cases, vector density, important in dengue survival rate. and climate variability), climate variability transmission. usingexistingsurveillance datasets will help to control and prevent dengue locally. AFRICA Seidahmed Dengue Longitudinal study Spatial and temporal Spatial distribution of Main determinants of Not reported. Further research is needed IG et al., 2012 Sudan patterns of dengue dengue was irregular in dengue outbreaks were to define whether there is [44] Port Sudan transmission the city. IgM increased maritime traffic a true disappearance of Ae. seroprevalence ranged and weather variation. aegypti and, if so, how and between 3 and 8% among It should be feasible to from where the vector is

healthy residents and carry out timely vector reintroduced. 24 of 9 Page incidence rate was 35 new control measures to Study impact of climate clinical cases per 10 000 prevent or reduce dengue and socioeconomic individuals. transmission in this changes on dengue coastline area. emergence in the Red Sea region. Table 2 Description of included studies, by disease (Continued) Eder

Authors, Date VBD Study type Objectives (outcomes Results and conclusions Learning points and Main limitations, Future research, public Funding Poverty of Diseases Infectious al. et Country measures) recommendations: comments health policy & practice Region transmission dynamics, City vectorial capacity, co- infection Malaria Authors, Date VBD Study type Objectives (outcomes Results and conclusions Learning points and Main limitations, Future research, public Funding Country measures) recommendations: comments health policy & practice Region transmission dynamics, City vectorial capacity, co- infection AFRICA

Woyessa et Malaria Descriptive Study malaria transmission Parasitaemia in 3.7% of Malaria risk in urbanized Low case numbers in dry Address vulnerability of U (2018)7:90 al., 2004 [7] Ethiopia in Akaki town, at 2110 m 2136 blood films/3 months African highlands season possibly a result of highland population of altitude; parasitaemia (69% vivax, 31% increasing, especially bednet use/interventions. short period malaria prevalence; malaria/ falciparum), none in last during rains. Possible link Malaria transmission low in transmission and mosquito species month (dry season). An. to climate change. Long cities (water pollution) but associated epidemics. frequency arabiensis/An. chrystyi periods of non- short-term increase from Apply sustainable and predominant species (fairly transmission reduce popu- extra breeding sites during integrated vector control low quantities), suggesting lation immunity, with risk rainy season. (breeding sites)/case indigenous transmission. of recurrent epidemics. management to prevent epidemics. Ebenezer et Malaria Descriptive, modelling, Correlation of An. gambiae Man-biting rate high (6.9) MBR differs across different Methods unclear whether When assessing efficacy of NR al., 2016 [56] Nigeria spatial entomological inoculation in mangrove coastal water, eco-vegetational zones of any malaria species or FM transmission control rate (EIR) and malaria vs. fresh/brackish swamp Bayelsa State. only. measures, both prevalence and incidence water; An. gambiae s.s. Authors propose EIR is a EIR not suitable for inter- entomological (EIR) and rates in various ecozones 13% = most more direct measure of age/population compari- clinical (prevalence and efficient and malaria transmission sons, due to variable fac- incidence) data need to be anthropophilic vector (EIR intensity, compared to PR tors such as host considered. = 70); PR and IR can or IR alone. biomass. predict PR-EIR (71%) or IR- EIR (64%). El Sayed et Malaria Descriptive Transmission in low- An. arabiensis (only vector Transmission in semi-arid Low-income or proximity Need for improved malaria IG al., 2000 [59] Sudan income peri-urban Ed species) showed higher Khartoum unstable, with to agriculture responsible control to reflect dekheinat vs. suburban El MBR and indoor density in seasonal pattern and for increased transmission increasing urbanization manshia, in Khartoum low vs. high-income areas, higher risk in low-income in Ed dekheinat, or and changing malaria (prevalence rates, especially in rainy season. peri-urban areas of urban interaction? Authors epidemiology in Africa. mosquito density) Parasite screening: FM in expansion (bringing resi- describe major climate, Aim for sustained decrease high-income areas only, dential areas closer to cul- economic, social and in malaria morbidity and also other species in low- tivated and irrigated land). political changes to mortality from epidemics. income areas; most FM Sudanese capital in recent rates seen in < 15 year decades. age. Ivan et al., Malaria and Descriptive Malaria (Plasmodium Malaria prevalence lower Possible differential effect Cross-sectional study: no Complex immunological NG, IG 2012 [51] helminth co- species), helminth or dual in peri-urban (12%) vs. of ART regimen type on P. temporal/causality; Study profile of co-infection, ef- infection (HIV in- infection; cross-sections of rural (30%) areas; also hel- falciparum/Trichuris co- too small to measure true fect on anaemia (caused

fected women peri-urban and rural preg- minth infections (33% vs. infection. differences, especially by each of the three infec- 24 of 10 Page > 12 m ART in nant women (n = 338); 45%; n.s.) high co-infection potential differential effect tions). Rwanda) prevalence/OR rates (5% vs. 15%), Nema- of some ART regimens. Malaria/helminth co- todes found: Ascaris (21%), infection might facilitate Trichuris (9%) and hook- HIV acquisition. worm (1%); Table 2 Description of included studies, by disease (Continued) Eder

Authors, Date VBD Study type Objectives (outcomes Results and conclusions Learning points and Main limitations, Future research, public Funding Poverty of Diseases Infectious al. et Country measures) recommendations: comments health policy & practice Region transmission dynamics, City vectorial capacity, co- infection Müller et al., Malaria Descriptive Multiplicity Plasmodium 61% of Riboque Multiple P. falciparum Only passive follow-up for Investigate NG 2001 [50] São Tomé falciparum in (FM) population cross-section infections protect against febrile illness as pathophysiology of protective effect in PCR positive. The msp-2 super-infecting parasites; possible source of bias multiplicity infections: moderate- transmission genotype multiply-infected FM prevalence highest in (non-attendance, or long- frequently found in West Africa (IR) had less FM/non-FM over 5–10 year-old children. term antimalarial asymptomatic infection; subsequent 3 months. treatment). protective against superinfection? Peterson et Malaria Descriptive, modelling, Small-scale spatio- Kulldorff scan: spatial Benefits of identifying Patient may have self- Small-scale mapping or NR al., 2009 [49] Ethiopia spatial temporal study in 2003 cluster ≤350 m of temporally stable clusters treated or visited other prediction models to make (2018)7:90 epidemic in Adama (inci- breeding site. Other risk and risk factors to target health centres; no data urban malaria control in dence rate ratio) factors included: poor cost-effective interventions collection in temporary Africa more effective. housing; proximity to in breeding sites. vegetation; ↗temp/rain. transmission hotspots. Sissoko et al., Malaria Descriptive, modelling, Prevalence of density/malaria Mosquito density and Study areas fairly small and Efforts to maximize impact NG, 2015 [48] Mali spatial (a)symptomatic malaria/ parasitaemia spatial parasitaemia spatial mosquitoes possibly of interventions by NGO mosquito in 2 areas with clusters observed in dry clusters in small villages outreaching boundaries. targeting areas of more different trans-mission/vec- season, sometimes (low or mesoendemic Study done only one intense transmission likely tor distribution associated; but high malaria transmission), best month after rainy season. limited by lack of Anopheles density or detected during dry suitability, since high parasite carriage also season. parasite prevalence was found outside the detected outside hotspots. hotspots. Ye et al., Malaria Descriptive, modelling Predict FM in cohort of Model represented well Model predicted seasonal Model of parasitological Local-scale FM prediction IG, NGO 2009 [60] Burkina Faso children ≤5 years, living in FM incidence for three variation increasing vector data in children ≤5 years, beneficial to guide control; holendemic area (weather/ ecological settings, abundance with increasing while entire population incidence depends on vector data modelling) including 595 person-years temperature 14 days after contributing to malaria daily vector mortality and (n = 676) over 1 year. rainfall; in P. falciparum transmission. human parasite clearance malaria infection rate, both targets of incidence. control measures. ASIA Dev et al., Malaria Descriptive Fever surveys for malaria Malaria throughout the Areas with low-to- Potential confounding: Health planners and policy NR 2004 [54] India incidence and risk factors: year, more in rainy season; moderate EIR could health centres likely based makers to consider distance-breeding sites, mainly FM; incidence reduce malaria significantly in plains. characteristics of malaria healthcare facility (IR/RR); higher near streams and by using campaigns and Higher P. falciparum rates transmission and risk vector EIR and weather foothills/forest areas; lower other tools in combination during monsoon (likely factors in trials and data in areas < 5 km to nearest with GIS methods to tar- due to increased other, newer approaches healthcare facility. EIR and get intervention and save temperatures promoting for malaria control in this % malaria among fever costs. parasite development in part of the world. cases not correlated. vector). Dhiman et Malaria Descriptive Impact of altitude on ↗ temperature coinciding Reduced transmission Malaria incidence based Highland urban areas to NR al., 2013 [47] India monthly malaria incidence with peak malaria windows and different on convenience data from be considered vulnerable

and vector density incidence. Malaria vector composition in the health centres (rather than for malaria transmission, 24 of 11 Page transmission window highlands. cross-sectional study of especially due to decreased by 1 month parasitaemia population). environmental changes. with 400 m increase in altitude. Table 2 Description of included studies, by disease (Continued) Eder

Authors, Date VBD Study type Objectives (outcomes Results and conclusions Learning points and Main limitations, Future research, public Funding Poverty of Diseases Infectious al. et Country measures) recommendations: comments health policy & practice Region transmission dynamics, City vectorial capacity, co- infection Lee et al., Malaria Descriptive Malaria trends, 1983–2007 incidence > 90% of cases imported Medical practitioners to Singapore vulnerable to NR 2009 [53] Singapore epidemiological ranged from 3 to 11 per from other Asian countries; highlight risk of malaria to reintroduction of malaria, characteristics, local 100 000 population, with migrants often associated travellers visiting endemic requiring high vigilance transmission and control deaths in 92% due to FM, with larger outbreaks in areas and also to consider (e.g. screening; educating measures (incidence) and 8% vivax malaria, and relapsing P. vivax cases. possibility of simian on prophylaxis). Consider a sharp decline after 1997. malaria. simian malaria if no travel One P. knowlesi outbreak. history. Zhang et al., Malaria Descriptive, modelling, Time series regression of Strong seasonal pattern; Increasing An. sinensis Predictive model does not Direct public resources to NG 2012 [55] China spatial 2006–2010 vivax malaria, peak during 2nd half of density likely to contribute account for human control infection, rather (2018)7:90 vector density, weather year; visual spatial only little to malaria interventions since difficult than vector, when variables. association with average incidence in low to measure. incidence is low. Use temperature; Tmax/ transmission areas. surveillance and vary average humidity (1 m lag); control efforts according previous month’sincidence. to incidence. Zhao et al., Malaria Descriptive Epidemiology of malaria in 95% of cases were Domestic endemic areas No data on floating Future studies to NR 2013 [52] China Ningbo city. Data from imported vivax malaria are important source for population (might have determine impact of case and vector from domestic endemic limited, local transmission significant impact on floating population on surveillance, local weather areas, leading to limited of vivax malaria. incidence). Study failed to dynamics of local malaria and 2008 outbreak analysis local transmission, Strengthen monitoring for show rain as precipitating incidence. determined by An. sinensis imported malaria, ensure factor for monthly density Focus on timely case vector density. timely diagnosis/ of female An. sinensis. detection, diagnosis and treatment. treatment. AMERICAS Girod et al., Malaria Descriptive Explore malaria vectors An. darlingi density high at Variable relationships Low numbers of infected Not specifically reported; NR 2011 [58] French Guiana and associated (though variable between) between malaria An. darling mosquitoes of note is the strong transmission dynamics 3 study locations; only few incidence, An. darling due to traps located at entomological focus of Plasmodium positive and density, rainfall, and places away from where this work. none at the village with nearest river water levels. transmission was highest rates of human occurring. cases. Moreno et al., Malaria Descriptive/ longitudinal Determine anopheline Transmission throughout Asymptomatic carriers are Both Anopheles species are To identify bionomics of NG, IG 2007 [57] Venezuela study mosquito characteristics the year, with malaria important reservoir of indoor and outdoor biters Anopheles species relevant and climate factors/ prevalence between 12.5 to parasites for persistently but displayed strong to malaria transmission in malaria incidence 21.4 per 1000 population; high levels of transmission. exophagic behaviour in Venezuela for planning (prevalence) An. darlingii and An. villages located in forested and implementing vector marajoara important areas. control programs. vectors, more abundant during rainy season. Other diseases Leishmaniasis

Camargo- Visceral Spatial Spatial analysis surveillance Heterogeneous Benefit of investigating Limited information on Need for new spatial NR 24 of 12 Page Neves, 2001 leishmaniasis tools for VL; Araçatuba, transmission: human cases vector distribution and sampling methods; no analysis tools and [64] (VL) São Paulo, 1998–1999 more frequent in areas covariates, in a field study modelling of vector redefinition of protocols Brazil (incidence) with high canine rates. based on house sampling. density. for control of endemic Vector dispersion restricted disease in urban areas. to a few houses. Table 2 Description of included studies, by disease (Continued) Eder

Authors, Date VBD Study type Objectives (outcomes Results and conclusions Learning points and Main limitations, Future research, public Funding Poverty of Diseases Infectious al. et Country measures) recommendations: comments health policy & practice Region transmission dynamics, City vectorial capacity, co- infection Salomón et Tegu-mentary Outbreak investigation Distribution of vectors and 31 cases (25 ± 14 years Urban transmission risk at Possible over-matching Local-based surveillance NR al., 2006 [62] leishmaniasis cases and the risk factors old), m:f sex ratio = 1.8; ecotone-border: changes (controls selected in same system for entomological (TL) during the 2003 TL compared to matched in human distribution and block of cases) and small surveillance, diagnosis, and Argentina outbreak in Bella Vista, controls. Clusters in 2 activities, patches of sec- sample size: interpret risk treatment at sentinel sites. Corrientes. contiguous ondary vegetation, peri- factors Risk assessment needed neighbourhoods; risk urban streams, rainfall, and associated with for any project involving higher in peri-urban (96%) river period floods. leishmaniasis infection change in land use at city than in peripheral (4%) with caution. borders. areas. (2018)7:90 Thomaz- Cutaneous Ecological Epidemiological and 61% (n = 100) were male Diversity of risk factors Future exploration could Further eco- NR Soccol et al., leishmaniasis entomological aspects of rural workers ➔ including domestic dogs, emphasize sample size epidemiological/ 2009 [63] (CL) CL endemicity extradomiciliar occupational hazards; and consider distributional biogeographical Brazil transmission. 29% of CL deforestation to increase studies. approaches needed to houses had antibody agricultural and pastoral clarify relationships found positive dogs. areas likely important in the present study, to intermedia s.l. most factor. guide targeted prevalent vector for interventions. (V.) braziliensis. Uranw, 2013 VL Outbreak investigation Investigate possible urban 115 VL cases (448 controls) Transmission of Retrospective nature of Appropriate surveillance IG [65] Nepal transmission of VL in strongly clustered in 3/19 anthroponotic VL in urban study: possible recall bias and control measures to Dharan town, Eastern (70%) neighbourhoods; houses; P. argentipes (> 10 years) during be adopted not only in Nepal, 2000–2008 independent risk factors vector and L. donovani interview in 2009. rural areas but also in include poor housing and parasite both identified urban areas. low income. inside town. Chagas Medrano- Chagas disease Ecological Demographics of Chagas High seroprevalence in Evidence of significant T. cruzi infection needs to NGO Mercado et Bolivia infections in 5–13 year-old both South (25%) and exposure leading to severe be considered an urban al., 2008 [67] children (n = 2218) in North (19%) zones, 3× disease, already early in health problem, which is Cochabamba, Bolivia, higher OR in children aged life. not restricted to rural areas 1995–1999 10–13 years in NZ: higher and small villages of triatomine vector infection Bolivia. rate (79% in South and 37% in North), but not density. Salazar et al., Chagas disease Ecological T. cruzi antibody 14/1544 samples Active vertical transmission Call for improved housing, IG 2007 [66] Mexico seroprevalence and confirmed positive. Risk of infection confirmed, vector control measures associated risk factors in factors included seeing with 0.91% seroprevalence and surveillance to be put individuals < 18 year in reduviid bugs inside house in people <18y, which in place. Veracruz, Mexico, and fissures in roof. Bug required attentative 2000–2001 infestation index, follow-up on colonization, and natural seroprevalence at infection: 11%, 50%, and population level. 9%, respectively. 24 of 13 Page Table 2 Description of included studies, by disease (Continued) Eder

Authors, Date VBD Study type Objectives (outcomes Results and conclusions Learning points and Main limitations, Future research, public Funding Poverty of Diseases Infectious al. et Country measures) recommendations: comments health policy & practice Region transmission dynamics, City vectorial capacity, co- infection West Nile virus Godsey et al., West Nile virus Outbreak Comparison of Higher Cx. quinquefasciatus Cx. quinquefasciatus main Vector blood meal host Similar multi-faceted stud- NG 2012 [69] (WNV) entomology within and abundance and proportion vector for transmission, preference needs cautious ies during periods of in- USA outside WNV outbreak of human blood meals with blood meal host interpretation: represents creased WNV area in Arizona, 2010 detected in 6 outbreak vs. preference and availability snapshot during outbreak; transmission to aid (incidence) 6 control areas (similar increasing risk of WNV also, no bird/mammal development of demographics). infection. census was conducted. effective outbreak intervention strategies. (2018)7:90 Nielsen et al., WNV Spatial modeling Risk factors during 2006 16/1355 (1.2%) pools of Residential areas had Mosquito trapping not Need for spatial detection U 2012 [68] USA outbreak in northern . WNV positive. warm night temperatures conducted throughout the and emergency adult California residential Significant spatial-temporal and WNV-positive Cx. tar- urban landscape. control in locations with community (incidence) clustering of infected dead salis, likely critical factors WNV activity: interrupt birds, positive Culex and during transmission before areas of human cases. initiation of the outbreak. human infections occur. Chikungunya Ho et al., Chikungunya Outbreak study Epidemiology of 2006–2009: 812/1072 Initially sporadic cases Introduction of mutant Identify and address NG 2011 [71] Singapore chikungunya establishing (76%) were indigenous imported into Singapore, viral strain (A226V) in 2008 demographic vulnerability itself as endemic disease cases, imported from India then local transmission by resulted in rapid spread by (international travellers and and Malaysia; high risk Ae. aegypti as predominant Ae. albopictus as primary foreign workers). Vigilance group: foreign contract urban vector. vector. needed to detect and workers. respond early. Rezza et al., Chikungunya Outbreak study Identify primary source of Presumed was High similarity between Outbreaks in non-tropical NR 2007 [70] Italy infection and modes of a symptomatic man from strains found in Italy and areas transmission (risk ratio and India visiting relatives in those identified during unexpected; need for pre- attack rate) Italy. Laboratory earlier outbreak in Indian paredness and confirmation was obtained Ocean islands. response to emerging in- for 175/205 cases (32 PCR fectious threats in era of only; 135 serology only; 8 globalization. both) Yellow fever Gould et al., Yellow fever (YF) Outbreak investi-gation Confirm cause and further 605 cases (163 deaths; CFR Serology evidence for Outbreak diagnosis and Need for timely/adequate NGO 2008 [73] Sudan describe outbreak 27%); in 177 (29%) illness both chikungunya and YF response delayed/limited. response in future (incidence) suggesting YF. Positive IgM- during outbreak. Migration True YF and chikungunya outbreaks; improved antibodies in 5/18 recently and drought likely cases likely surveillance for YFV/other symptomatic, and 4/16 contributors to outbreak; underestimated. arboviruses; asymptomatic Ae. aegypti most abundant, Limited clinical Promote YF unvaccinated persons. Ae. luteocephalus information and possible in endemic areas and Also chikungunya IgM (important sylvatic YF to recall bias (retrospective for both detected in a few cases, both monkeys and study). arboviruses and malaria. (both ill and humans. asymptomatic). ae1 f24 of 14 Page Vasconcelos YF Outbreak study YF outbreaks in Goias and 77 cases reported in 8 Populations at risk: Unclear what other factors, Need for YF vaccination in NR et al., 2001 Brazil Bahia states, 2000 Brazilian states with agricultural workers, besides rainfall, were all areas recently affected, [72] Haemagogus janthinomys tourists, carpenters, associated with spread of as well as Ae. aegypti as main vector. Climate fishermen, truck drivers. YF in the area. control programs for the changes with increasing Further virology studies whole country, to avoid Table 2 Description of included studies, by disease (Continued) Eder

Authors, Date VBD Study type Objectives (outcomes Results and conclusions Learning points and Main limitations, Future research, public Funding Poverty of Diseases Infectious al. et Country measures) recommendations: comments health policy & practice Region transmission dynamics, City vectorial capacity, co- infection rainfall associated with needed to determine virus re-urbanization. epidemic and epizootic single vs. multiple episodes. genotypes. Ross River virus Carver et al., Ross River virus Ecological Relationship between Mouse index data (proxy Mice and RRV notifications Cause and effect (mice More research needed on NR 2008 [74] (RRV) amplifying host (M. for abundance) at Symes significantly related in and RRV cases) not proven: potential causal Australia musculus mouse) Farm matched with RRV autumn and autumn possible other factors relationship (mice and

abundance and RRV incidence in rural +summer when Culex required to enable RRV) to justify targeted (2018)7:90 notifications townships (postcode) in annulirostris mosquitoes transmission between the public health interventions cereal-growing Walpeup are abundant. two hosts. to reduce mice and region, Southern Australia. arboviral disease burden. Plague Pham et al., Plague Ecological Environmental factors and 472 plague cases; 4 main index, rodent density, No wider animal sampling; Plague outbreaks likely NG 2009 [76] Vietnam human plague in Vietnam flee and 3 rodent species. and rainfall could be used hence, collected rodents in due to multiple ecological central highland plateau, Increasing risk during dry as ecological indicators of confirmed cases might not factors linked to classical 1997–2002 (risk ratio) and hot months; plague risk in Vietnam’s be those causing outbreak. reservoir and vector of decreasing rainfall; central highlands plateau. Ecological investigation: bubonic plague; further increasing monthly flea no inference to individuals. research warranted. index/rodent density. Rickettsia Souza et al., Brazilian spotted Case-controlled Risk factors associated with Confirmed vs. discarded High incidence areas in Potential misclassification Control measures needed NG 2015 [75] fever (BSF) multivariate regression BSF, 2003–2013, Piracicaba BSF cases were associated São Paulo State reinforce through use of discarded focusing on elimination of Brazil River basin, São Paulo with increasing age, urban trend of urbanization of cases as controls: false dirty pastures and State (ORs) areas, presence of A. BSF in peri-urban peripher- negative cases (no timely prevention of human sculptum , ticks ies; presence of ticks and collection) ➔ OR biased contact with areas of host collected from horses, capybaras requires towards null hypothesis and vector species, to presence of capybaras, future field investigations. (no association). . control transmission of and dirty pasture BSF. environment. DENV: Dengue virus; NR: Not reported; NG: Non Governmental; IG: International Government; U: University; PI: Pharmaceutical Industry; NGO: Non-governmental organization GIS: Geographic information system; IR: Incidence rate; RR: Risk Ratio; FM: Falciparum malaria; ART: Antiretroviral therapy; NR: Not reported; NG: Non Governmental; IG: International Government; U: University; NGO: Non-governmental organization ae1 f24 of 15 Page Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 16 of 24

Argentina [29], Brazil [34, 35] and Bangladesh [36]. In contrast, other studies showed inverse relationships be- tween entomologic indices and dengue incidence [26, 37–39]. It has been well established that water storage containers and discarded tires are important mosquito habitats in many countries, which was also reinforced by studies in Puerto Rico [32] and French Guiana [40].

Weather and climate variability and vector proliferation Studies conducted in different regions evaluated the Fig. 3 Studies specified according to funding sources, some studies role of weather on the proliferation of Aedes aegypti, have more than one study source. Funding: not reported, n=16; dengue incidence, and the seasonality of the disease national government, n=21; university, n=4; international [30, 32, 39, 41–43]. Some ecological studies observed a government/WHO, n=15; non-governmental organizations, n=3; positive association of temperature and rainfall with varia- pharmaceutical industry, n=1 (some studies had more than one tions in the Aedes mosquito population [32, 38, 39]. Some funding source). studies conducted in the Americas showed a positive correlation of rainfall and temperature with dengue is another important factor in the spread of dengue infec- incidence [32, 41],whileothersinAsiafoundnega- tions in urban settings [27]. The spread of dengue by tive or no associations between these meteorological imported cases has been reported in some countries, such variables and dengue incidence [30, 42, 43]. In an as Argentina during the 2009 outbreak [29], as well as interdisciplinary study (2010–2011) conducted in São ChinaintheoutbreaksintheoutskirtsofCixi[30]and Paulo, Brazil, using geographic information systems, Dongguan cities [31] and near Guangzhou Baiyun Inter- more dengue cases were clustered in areas of land national Airport (2006–2012) [24]. The reports from China surface temperature above 32 °C than in areas charac- highlighted the role of immigrants from Southeast Asia as terized as low socioeconomic, high population density the source of localized outbreaks in this region. areas, or slum-like areas [39]. That study’sfindings were interpreted as suggesting that the land surface Vectorial capacity and incidence of dengue infection or temperature of the inner city was a better predictor disease for dengue incidence than were other factors such as There was conflicting evidence around indicators of vec- population density or socioeconomic indicators. tor abundance and incidence of dengue infection/dis- Therefore, the influence of higher temperatures in ease. Positive associations between high vector density small urban areas in São Paulo, known as urban heat and high dengue incidence were reported in different islands, was correlated with high-risk areas of dengue settings, such as San Juan, Puerto Rico [32], Cuba [33], transmission during this period (2010–2011) [39].

Table 3 Type of study methods, by disease group Dengue n=20 % Malaria n=15 % Other n=15 % All n=50 Total % Outbreak 2 10% 0 0% 7 47% 9 18% Case-control (cluster) 2 10% 0 0% 3 20% 5 10% Cross-sectional 2 10% 4 27% 1 7% 7 14% Cohort (longitudinal) 3 15% 6 40% 0 0% 9 18% Analytical - multivariate 6 30% 5 33% 5 33% 16 32% Ecological (population study) 4 20% 0 0% 5 33% 9 18% Spatial modelling 6 30% 3 20% 2 13% 11 22% Time series regression, dynamic modelling 3 15% 2 13% 0 0% 5 10% Secondary data only 5 25% 4 27% 1 7% 10 20% Prevalence – Incidence 15 75% 13 87% 13 87% 41 82% Odds Ratio 2 10% 1 7% 3 20% 6 12% Risk Ratio or Similar 3 15% 2 13% 3 20% 8 16% Methods applied in n=50 quantitative descriptive studies in numbers and %, for dengue, malaria and other pathogens (some studies employ more than one different study method) Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 17 of 24

Fig. 4 Quality of studies according to modified MMAT tool in numbers (%), for dengue, malaria and other pathogens

Other social and environmental factors influencing or transmission in that region, indicated by seroprevalence in associated with the complexity of urban settings each population [37]. Other social and environmental factors associated with disease incidence and vector abundance include living and Malaria transmission dynamics and vectorial capacity working conditions, human behaviour, urban infrastruc- studies ture, and water and sanitation, which includes water stor- The malaria research retrieved was mostly conducted in age conditions and housing quality [30, 32, 37, 40]. In the African region, where yearly estimates indicate to be Sudan, a study conducted in the neighbourhoods of the the highest burden of malaria cases (191 million cases in city of Port Sudan (2008–2009) [44], observed that dengue WHO African Region vs. 21 million in other parts of the incidence was heterogeneously distributed and higher en- world) and deaths; and the highest proportion of Pl. fal- tomological density indices were found in lower- and ciparum (vs. other species) worldwide in 2016 [5]. At middle-class neighbourhoods. In that setting, climate vari- the same time there is the lowest level of health expend- ability, maritime traffic, and socioeconomic conditions iture in this compared to other regions [45], further ag- were suggested as being the main drivers of dengue out- gravating the impact of the disease on populations and breaks in the past decade, although further research will economies. be required to study the impact of long-term climate change on dengue emergence in that region [44]. In Rio Incidence of malaria cases or infection de Janeiro, Brazil, spatial analysis showed a positive associ- There has been controversy about a potential expansion ation between social inequalities (Gini indices) and the of malaria from rural areas into cities. Research evidence container (Breteau) index for [41]. A ser- suggests increased malaria risk to urban dwellers, and osurvey conducted in the US-Mexican border area found transmission in urban and periurban setting [46]. In the an abundance of vectors in US cities, but higher dengue studies selected in our review, increasing transmission risk incidence in neighbouring cities in Mexico. This lead the was described as part of climatic changes affecting urban authors to conclude that population living conditions (low areas in the highlands of Ethiopia, and India, respectively air conditioning rates, small living spaces, high numbers [7, 47]. A study using small-scale temporal-spatial scan- of occupants) were the main determinants of dengue ning identified “hotspots” of high Anopheles density and Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 18 of 24

parasite carriage in two villages in Mali. However, trans- Ningbo, China [52]. An important factor for malaria mission was also found to be high outside these hotspots, transmission in forested villages in Venezuela was the casting doubt on the effect of targeted control interven- presence of mosquito species displaying predominantly tions in hotspots [48]. A transmission hotspot detected exophagic (outdoor) biting behaviour [57]. within 350 m of a large Anopheles breeding site during the In contrast, no correlation was found between EIR and 2003 outbreak in Ethiopia highlighted the potential of tar- percentage of malaria among reported fever cases in geted vector control measures to reduce cases [49]. India [54], nor between Anopheles density and human With regards to the occurrence of co-infections, case rates in French Guiana, and the authors argue the Müller et al. [50] found a protective effect in children in- mosquito traps might have been located in fected by multiple, genetically different Pl. falciparum non-transmission areas [58]. malaria parasite co-infections. Further, there was a pro- tective effect against episodes of febrile illness during a Weather and climate variability and vector proliferation three-month subsequent (passive) follow-up. The group Increased transmission during the rainy season has been found an 0.84 (95% CI: 0.71–0.99) hazard associated observed in Africa and Asia [7, 54, 59]. More specifically, with each additional Pl. falciparum genotype detected at increasing temperature and rainfall were followed by baseline and called for more research into this higher vector prevalence transmission models based on phenomenon, in particular the effects on the host in four different areas of Burkina Faso, with peak vector chronic infections [50]. Another group studied malaria prevalence occurring 2 weeks after peak rainfall [60]. In and helminth dual infections among pregnant women northwest China, monthly An. sinensis vector density with HIV infections on treatment for > 1 year. The study (relevant for Pl. vivax malaria transmission) was design was merely descriptive, showing co-infection rates strongly correlated not only with temperature (R =0.958, to be higher in women living in urban areas than among P < 0.001), but also with humidity and rainfall (R =0.746, those in rural areas [51]. The authors hypothesized re- P =0.005;andR =0.725,P = 0.008, respectively) [52]. Fur- garding potential aggravating effects of co-infection on thermore, increases in average maximum and minimum anemia, which is a consequence of each of these three temperatures (at 1 month lag) and rainfall (10-week lag) types of infection (malaria, helminths, HIV). in Ethiopia had malaria incidence risk ratios of 1.4 (for maximum temperature), 1.3 (for minimum temperature), Mobility of human populations as a source of outbreaks and 1.0 (for rainfall) [49]. Zhang et al. (2012) showed and/or disease persistence malaria epidemiology in China to have strong spatial Zhao et al. [52] found most malaria cases in Ningbo associations with average temperature. They proposed City,China,tobecausedbyPl. vivax imported from optimizing case management rather than vector con- domestic endemic areas, leading to local transmission trol for low-transmission areas [55]. through Anopheles sinensis. The authors suggested that more research was needed on the role of floating Other environmental factors, geography, and complexity of populations in local malaria transmission [52]. Migra- urban setting influencing malaria infection or disease tion and travel were also identified as important risk Peterson et al. (2009) identified not only proximity to a factors for malaria re-introduction in Singapore, evi- large Anopheles breeding site as a source of increased denced by an analysis of 25-year reporting data. The transmission, but also poor housing as a further import- authors advocated for screening, education, and good ant risk factor (malaria incidence risk ratio = 2.0) in case management. Finally, they suggested that also Adama, Ethiopia [49]. Similar observations were made simian Plasmodium knowlesi malaria would need to in semi-arid neighbouring Sudan, where transmission be considered as a possible source of fever in their was higher in rapidly expanding peri-urban low-income study population [53]. areas than in suburban higher-income areas of Khartoum [59]. Vectorial capacity and the incidence of malaria infection or Certain ecological areas show higher malaria transmis- disease sion than others. In Nigeria, transmission rates (MBR Several studies demonstrated changes in vector compos- and EIR) were higher near mangrove coastal water than ition according to geography and season to explain vari- in areas of fresh or brackish water [56]. In Ethiopia, ations in transmission [47, 52, 54, 55]. There was proximity to vegetation and to agricultural sites showed diversity among studies on the correlation between ento- higher vector density and more cases [49, 54, 59]. mological parameters and human malaria. The EIR as Further, changes to ecology and climate were consid- indicator for transmission was found to correlate with ered to be causing increasing malaria transmission in clinical prevalence and incidence data in Nigeria [56], urbanizedhighlandareasofAfricaandAsia[7, 47]. similar to Anopheles density and malaria cases in Capable Anopheles vectors and short-term malaria Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 19 of 24

transmission were observed during the rainy season Phoenix (US) [69]. A spatial study identified significant (due to short-term extra breeding sites) in Ethiopia, clustering of infected dead birds and positive Culex mos- where low population immunity between seasons quitoes near human cases occurring in residential areas causes epidemics [7]. Comparison of three villages at of California [68]; early detection was proposed as a key different altitudes in India showed both reduced to reducing the risk of outbreaks. Anopheles abundance and transmission windows for human cases with increasing altitude [47]. Geographic Role of tourism, migration and occupational exposure on expansion of mosquito vectors has been described as transmission of infection one of the possible effects of climate change [61]. Two studies reviewed chikungunya occurrence, one in Summarizing key insights from respective groups of Italy following virus introduction by a symptomatic indi- authors, there was an expression of need to address the vidual visiting from India [70] and the other in increased risk of transmission in vulnerable highland Singapore, where recent virus mutation allowed the in- areas [7, 47] and in spaces where specific risk factors are fection to be effectively transmitted by urban Aedes present, including proximity to breeding sites, poor albopictus mosquitoes [71]. Both research groups housing [49], low income [49], and floating populations highlighted the role of migrants in the spread of disease [52, 53]. They recommended better ways to monitor and and the need for effective disease surveillance to prevent address risk factors via spatial studies and forecast outbreaks. models that include entomological parameters and me- In Brazil, researchers identified a yellow fever trans- teorological factors. Further, the significance of asymp- mission link to tourism and occupational exposure (agri- tomatic infections was expressed in reports on cultural workers, carpenters, fishermen, truck drivers) parasitaemia and floating populations, identifying the and to Haemagogus janthinomys as the main mosquito need to detect and treat such cases to prevent trans- vector [72]. In Sudan, drought, migration, and the lack mission [52, 53, 59]. Finally, the complexity and po- of diagnostic capabilities or adequate response contrib- tential impact of co-infections on the host have been uted to a yellow fever outbreak where there was concur- recognised [50, 51]. rent transmission of chikungunya [73].

Other diseases Influence of disease ecology on transmission Four studies focused on leishmaniasis: two on cutaneous A field survey found seasonal abundance of amplifying leishmaniasis (in Argentina and Brazil) [62, 63] and two mouse populations to increase Ross River virus (RRV) on (in Brazil and Nepal) [64, 65]. transmission in Australia, combined with the presence Agricultural male workers were identified as a risk of the Culex annulirostris vector. The authors proposed group; further associated factors included peri-urban liv- more specific research on the causal relationship be- ing environment, low socioeconomic status, poor hous- tween mice and RRV, along with possible interventions ing, and domestic dogs. All reports demonstrated the to control the disease [74]. need for improved surveillance and control measures, to Bacterial diseases studied included -transmitted reduce infection risk both in urban and peri-urban areas, Brazilian spotted fever (BSF) in Brazil [75] and human with specific focus on dog populations. bubonic plague in the Vietnam Central Highland plateau Chagas disease was detected in young individuals in [76], for which multiple ecological factors were identi- Mexico (1% of people aged < 18 years) [66] and Bolivia fied, and the authors proposed using rodent density and (> 20% of school children aged 5–13 years). Poor hous- rainfall as ecological risk indicators. ing and high infection rates among transmitting vectors were identified as important risk factors; the authors Discussion suggested prioritizing detection and control programs in Dengue and malaria studies constituted the largest these urban areas [67]. groups of published research in our review—dengue pre- West Nile virus (WNV) is a VBD in which bird popu- dominantly in Asia and the Americas, and malaria in lations such as corvids serve both as important reservoir Africa. Dengue has the highest burden and vectors cap- and amplifiers, whereas migratory birds are involved in able of transmitting in urban and peri-urban areas of global transmission [68]. Researchers in the United these regions. The urbanization of the population in Af- States (US) correlated mosquito and local bird popula- rica has also reflected in malaria transmission that can tion WNV status with human incidence. They identified be currently considered an urban problem [77]. Despite Culex mosquitoes as important vectors in the Arizona being different pathogens (protozoa vs. virus), both outbreak where their abundance and (bird) host prefer- VBDs, despite spread by different mosquito species can ence increased the risk of human transmission, and be framed with regards to the importance ofurban heat compared to control sites in the metropolitan area of islands and eco-zones, human habitat (proximity to Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 20 of 24

breeding sites), host behaviour and mobility, the role of basic: questionable sampling techniques (convenience asymptomatic infections, and association with increased sampling), no indication of the proportion of temperature and rainfall (albeit more evidence is re- non-responders, offering only passive rather than active quired on the associations between climate variability follow-up, location of mosquito traps not matching with and dengue incidence to explain the discrepancies in areas of human transmission—all of which created risks recent studies) promoting vector abundance and associ- of bias. Also, most studies were descriptive (i.e., using ated disease incidence. Since the beginning of 1900 mal- measures of occurrence) rather than providing robust aria research employed the theory of Ross-MacDonald figures of transmission risk (measures of effect). Further, for the dynamics and control of mosquito-transmitted a multidisciplinary approach, as suggested above, pathogens. This transmission model has now been could have provided essential insights into the role of adopted for dengue research, which has become more asymptomatic infections, especially among floating intensified in recent years as the disease is becoming a populations. global problem [19]. Discussing other infections than dengue and mal- Dengue transmission and vectorial capacity have aria, we detected recommendations on specific sur- been studied mainly using the conventional frame- veillance and control measures that were included in work of interaction between human and mosquito most studies. For example, the need for entomological populations. A bulletin of the World Health surveillance and control in detecting risk areas for Organization highlighted the importance of increas- Leishmaniasis [62, 63, 65], Chagas’ disease [66, 67], ing residents’ knowledge regarding dengue transmis- and arboviruses (West Nile Virus [68, 69], Chikungunya sion, which was associated with a measurably lower [71]), and plague [76]. Also, the need of targeted surveil- mosquito reproduction in the respective areas [78]. lance and interventions focusing on important animal res- This was presented as an example of how broader ervoirs for Leishmaniasis (dog population) [63, 64], West public health efforts (beyond larvicide and focal Nile Virus (clustering of dead birds) [68], Ross River Virus spraying) can contribute to effective vector control (abundance of house mice), and plague /BSF (rodents) [78]. There is a lack of translational research and a [75, 76]. The importance of increasing such measures spe- need to combine multiple knowledge areas involving cifically in urban and periurban areas was highlighted in urban planners, travel and border agencies, transport relation to Leishmaniasis [62, 64, 65], Chagas’disease [67], authorities, environmentalists [79]. Such integration West Nile virus [68]andBSF[75]. In addition, enhancing would be an useful approach to better understand vigilance around migration and travel are needed to and respond to the complexity of dengue dynamics reduce risk for spreading of Chikungunya [70, 71]and in urban settings. Only a few studies addressed this Yellow Fever [72, 73]. For the latter, the importance of using information on previous dengue serotype im- vaccination programs was mentioned [72, 73]. Consider- mune status to understand disease spread and ing that two thirds of the studies were funded by govern- persistence. ment sources (national or international) the an integrated None of the selected studies assessed the approach including human and animal health, and ento- co-circulation of VBDs transmitted by the same vector, mology should be reinforced. The Joint external evalu- such as dengue, chikungunya and Zika, which coexist in ation tool by the World Health Organisation as part of many regions across the globe [11]. A syndromic ap- Global Health Security is an example for multi-sector and proach focusing on patients’ main symptoms, such as multi-disciplinary effort. This agenda considers multiple fever and rash (equally common symptoms for dengue, hazards, including detection and control of priority epi- chikungunya, Mayaro, Zika, etc.), rather than only on demic diseases, border surveillance, using an integrated isolated pathogens, might help to adapt VBD research ‘One Health’ approach including human, animal and en- more effectively to the clinical-epidemiological reality. vironmental health [80]. Combining such an approach with broad diagnostics This scoping review has some limitations. (e.g. testing for a panel of common vector-transmitted Performing a detailed data extraction on all 205 papers parasites, , and bacteria) would allow easy detec- was considered as unfeasible by the consortium. There- tion of and response to co-circulating vectors, including fore, an additional step to include only studies that cov- newly emerging pathogens. This is particularly true for a ered at least two of the key concepts (i.e. “transmission coordinated international response to new pathogen in- dynamics and vectorial capacity” and, “transmission dy- troductions or epidemics, such as Zika in the America. namics and co-infection”) limited the number of papers. Harmonization of syndrome-based protocols would in- An additional benefit of this approach was a more crease the effectiveness of such efforts. comprehensive picture that combined at least two Similarly, malaria has been studied largely in conven- components of infectious diseases in urban areas. At the tional frameworks. In some studies, quality was very same time, we acknowledge limitations arising from this Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 21 of 24

final step, which may have excluded important papers Box 1 The main implications for future research and public health reporting only one key concept. That way, classical mod- policy and/or practice elling studies, (particularly those dealing solely with Knowledge gaps and priority needs for future research mathematical models (for prediction of outbreaks, 1. Assess the magnitude of asymptomatic dengue infection at spread of infection, and/or long term sustainability of population level (surveillance of symptomatic dengue cases is transmission) may have been missed [17, 18, 81]. insufficient to evaluate the persistence of infection). Further limitations relate to quality and comparabil- 2. Improve indoor and outdoor vector density parameters for more ity of selected work. Only a few studies went beyond accurate modeling of transmission. description, as shown by MMAT evaluation of the 3. More studies are needed on climate and other environmental quality of studies. Clear objectives were set out in all (e.g. land surface temperature) changes and their effect on vector studies and were addressed to a fairly large extent. proliferation and dengue transmission. However, concerns about study quality arose regard- 4. The impacts of human mobility within and between cities and countries should be prioritized in future research. ing 10–20% of studies, which did not report the rele- vant sampling strategy. Compared to the studies on 5. Enhance research and seek scientific consensus on the benefit of simple, ready-to use forecasting tools to predict human VBD other pathogens, the malaria studies were less repre- risk (using entomological, meteorological, and other parameters). sentative of true population (therefore producing less 6. Promote research on co-infections with different pathogens, on generalizableresults),duetotheirdesigns,which immunology mechanisms and their effect on clinical outcomes and were mostly smaller-volume cross-sectional and co- onward transmission, and on means of effective diagnosis and treatment. hort studies. Implications for public health policy and/or practice Also, there was no attempt to stratify by population 1. In dengue-endemic areas, monitoring areas of low transmission may size. Stratification would have allowed to identify chal- be necessary to prevent spread of infection. lenges specific to highly populated urban areas as op- 2. Surveillance and control strategies focused on index cases should be posed to smaller urban areas like villages. Dengue is a timely to avoid time lag between outbreak onset and response. VBD amplified by humans (rather than other 3. Asymptomatic individuals contribute to persistence of dengue non-human hosts) which contributes to large scale and malaria transmission, reinforcing the need for population screening (e.g. biological marker laboratory testing blood banks, transmission in cities. In contrast, transmission dynam- sentinel sites), in low and high seasonality. ics for a number of other pathogens included (e.g. 4. Need to assess multiple data sources regarding symptomatic and Leishmaniasis, West Nile Virus, Yellow Fever, Ross River asymptomatic cases. virus and plague) rely on non-human host species, such 5. Surveillance and control strategies focused on index cases should be non-human primates, dogs, rodents or birds. Presence of timely to stop transmission. those species will depend on different types of urban 6. Greater efforts must be made to translate knowledge about VBD and peri-urban environments and other factors. The transmission into practice. same applies for the type of insect vectors implicated. 7. Employ scientifically agreed-upon ready-to use forecast models to From this perspective, further work classifying between predict human VBD risk based on entomological and meteorological different urban environments will be useful. parameters. The difference in methods used by studies is one fac- 8. Increased rainfall and humidity, especially during the rainy season, affects VBD transmission; authorities need to collaborate to heighten tor limiting the comparability, particularly on study that vigilance and control measures. combined weather and entomological data to predict 9. Poor housing, low-income neighbourhoods are high-risk areas VBD incidence [24, 31, 42, 60]. Differences also arose in for VBD transmission; they should be focus of affordable and the researchers’ selections of the most appropriate ways sustainable vector control measures in homes, workplaces and to control for factors such as seasonality and schools, to lower transmission over the long term. non-linearity of weather dependence, as pointed out in a 10. Certain occupational groups have higher exposure to VBDs; labour technical paper on temporal modeling research [82]. Of and agricultural authorities must invest in efforts to increase awareness and safety in relation to specific disease risks. note is that, to our knowledge, there are currently no 11. Transport authorities and border agencies need to screen floating international standards to advise on the most appropri- populations at risk of infection. ate modeling approach for real-time prediction to in- 12. Using a syndromic approach instead of the classic single-disease form public health practice. surveillance would allow timely response to the introduction of Finally, another limiting factor was that only two new pathogens or early outbreak detection. studies reported on co-infections, and both of those 13. Harmonization of protocols are needed to facilitate a coordinated addressed malaria. This is concerning in view of how international effort to control disease threats of national/international little is known about this phenomenon, the immuno- importance. National government and academic institutions to promote an integrated multi-disciplinary approach (human and animal logical mechanisms involved, and what it means for health, vector control), focusing on detection and control of priority clinical outcomes; even less is known about transmis- epidemic diseases, border surveillance. sion dynamics. Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 22 of 24

Conclusions Authors’ contributions The present review identified significant knowledge gaps ME, FC, NTSF, CB, CMTM participated in the design, study selection, data extraction and analysis, and write-up of this study. SD and VR participated in in several areas, ranging from the role of asymptomatic the design of this study and revised the manuscript. GVAF participated in individuals to the effects of co-infection and various host the revision of the manuscript. All authors approved the final version. characteristics, climate, and other environmental and so- cioeconomic factors on VBD transmission in urban Ethics approval and consent to participate areas. There is much more to know about transmission The study protocol was approved by the University of Montreal Human Research Ethics Committee. risk in the homes and workplaces of increasingly dy- namic and mobile populations. Consent for publication The lack of studies on co-infection is hampering the Not applicable. monitoring of infections transmitted by the same vector. A broad, syndromic approach including pathogen panels Competing interests would allow more flexibility in detecting new and The authors declare that they have no competing interests. co-circulating pathogens and in applying more effective Author details control. It would be useful to combine this with harmo- 1Public Health England Sierra Leone Country Office, Freetown, Sierra Leone. nized protocols and to define sentinel areas in order to en- 2Aggeu Magalhaes Institute (IAM) / Oswaldo Cruz Foundation (Fiocruz), able a well-coordinated international response where Avenida Professor Moraes Rego, s/n. Cidade Universitaria. CEP 50, Recife, Pernambuco 740-465, Brazil. 3Universidade de Pernambuco (UPE), Recife, needed. Due to the complexity of VBD transmission, fund- Pernambuco, Brazil. 4Liverpool School Of Tropical Medicine (LSTM), London, ing for translational research is especially recommended. UK. 5Secretariat of Health Surveillance, Ministry of Health, Brasilia, Brazil. 6University of Montreal School of Public Health (ESPUM), Montreal, Quebec, Canada. 7IRD (French Institute For Research on Sustainable Development), Additional files CEPED (IRD-Université Paris Descartes), Universités Paris Sorbonne Cités, ERL INSERM SAGESUD, Paris, France. Additional file 1: Multilingual abstracts in the six official working Received: 8 February 2018 Accepted: 2 August 2018 languages of the United Nations. (PDF 390 kb) Additional file 2: Literature search: Associated keywords, MESH and search terms. (DOCX 25 kb) References Additional file 3: Final reference selection by country. (DOCX 15 kb) 1. World Health Organization. Vector-borne diseases [Internet]. 2016 [cited 2017 Aug 17]. Available from: http://www.who.int/en/news-room/fact- sheets/detail/vector-borne-diseases. Abbreviations 2. Anderson R, May R. Infectious diseases of humans: Oxford Science ART: Antiretroviral therapy; BSF: Brazilian spotted fever; CAR: Conditional Publications; 1991. Autoregressive; CFR: Case fatality rate; CL: ; 3. Kramer LD. Complexity of virus - vector interactions. Curr Opin Virol. 2016; EIR: entomological inoculation rate; FM: Falciparum malaria; IG: International 21:81–6. Government; IR: Incidence rate; LST: Land Surface Temperature; MBR: Man 4. Heesterbeek H, Anderson R, Andreasen V, Bansal S, De D, Dye C, et al. biting rate; MMAT: Mixed Methods Appraisal Tool; n.s.: Not significant Modeling infectious disease dynamics in the complex landscape of global (statistically); NG: Non Governmental; NGO: Non-governmental organization; health. Science (80- ). 2015;347:aaa4339. NR: Not reported; OR: Odds ratio; PCR: Polymerase Chain Reaction; 5. World Health Organization. World Malaria Report. Geneva. 2016:2016. PI: Pharmaceutical Industry; PR: Prevalence rate; RF: Risk factors; RR: Risk Ratio; 6. Wilson ML, Krogstad DJ, Arinaitwe E, Arevalo-Herrera M, Chery L, Ferreira RRV: Ross River virus; TDR: Tropical Diseases Research and Training; MU, et al. Urban malaria: understanding its epidemiology, ecology, and TIDieR: Template for Intervention Description and Replication; transmission across seven diverse ICEMR network sites [Internet]. 3, Suppl. TL: Tegumentary Leishmaniasis; U: University; UHI: Urban Heat Island; ed. Spec. Issue Int. Centers Excell. Malar. Res. background, progress, ongoing VBDs: Vector-borne diseases; VL: Visceral Leishmaniasis; WHO: World Health Act. USA; 2015. p. 110–23. Available from: https://www.ajtmh.org/content/ Organization; WNV: West Nile virus; YF: Yellow Fever; YFV: Yellow Fever Virus journals/10.4269/ajtmh.14-0834. 7. Woyessa A, Gebre-micheal T, Ali A. An indigenous malaria transmission in the outskirts of Addis Ababa, Akaki town and its environs. EthiopJHealth Acknowledgements Dev. 2004;18:2–8. We would like to thank Sylvie Fontaine, public health research librarian at 8. Brady OJ, Smith DL, Scott TW, Hay SI. Dengue definitions the University of Montreal, for her careful review of the search strategy and are implicitly variable. Epidemics. Elsevier B.V. 2015;11:92–102. Mariam Otmani del Barrio, from the Unit on Vectors, Environment and 9. Halstead SB. Dengue Virus–Mosquito Interactions. Annu Rev Entomol. Society at WHO, for her comments on our final manuscript. This study was [Internet]. 2008;53:273–91. conducted as part of the VERDAS consortium project, funded by the Special 10. Fauci AS, Morens DM. Zika virus in the Americas—yet another Program for Tropical Diseases Research and Training (TDR) of the World threat. N Engl J Med. 2016;374:601–4. Health Organization. Partial support from the National Advisory Board of 11. Musso D, Cao-Lormeau VM, Gubler DJ. Zika virus : following the path of Scientific and Technological Development (Conselho Nacional de dengue and chikungunya ? Lancet Elsevier Ltd. 2015;386:243–4. Desenvolvimento Cientifico e Tecnologico [CNPq]; scholarship to CMTM 12. World Health Organization. Neglected tropical diseases - Fact sheets 306708/2014-0 and to CB 304174/2014-9). relating to NTD. 2017. 13. United Nations. World Urbanization Prospects: The 2014 Revision, Highlights Funding (ST/ESA/SER.A/352). New York, United. 2014. VERDAS Consortium. 14. Alirol E, Getaz L, Stoll B, Chappuis F, Loutan L. Urbanisation and infectious diseases in a globalised world. Lancet Infect Dis Elsevier Ltd. 2010;11:131–41. 15. Hassell JM, Begon M, Ward MJ, Fèvre EM. Urbanization and Disease Availability of data and materials Emergence : Dynamics at the Wildlife – Livestock – Human Interface. Trends Not applicable. Ecol Evol. The Author(s); 2017;32:55–67. Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 23 of 24

16. Struchiner CJ, Rocklöv J, Wilder-smith A. Increasing dengue incidence in densities, dengue fever and climate in Mirassol, state of Sao Paulo, Brazil. Singapore over the past 40 years : population growth. Climate and Mobility Mem Inst Oswaldo Cruz. 2008/10/25. 2008;103:554–60. PLoS One. 2015:1–14. 36. Ali M, Wagatsuma Y, Emch M, Breiman RF. Use of a geographic information 17. Amaku M, Azevedo F, Burattini MN, Coelho GE, Coutinho FA, Greenhalgh D, system for defining spatial risk for dengue transmission in Bangladesh: Role Lopez LF, Motitsuki RS, Wilder-Smith A, Massad E. Magnitude and frequency for Aedes albopictus in an urban outbreak [Internet]. 6th ed. United States: variations of vector-borne infection outbreaks using the Ross–Macdonald American Society of Tropical Medicine and Hygiene (60 Revere Drive, Suite model: explaining and predicting outbreaks of dengue fever. Epidemiol 500, Northbrook IL 60062, United States); 2003. p. 634–40. Available from: Infect. 2016;144(16):3435-50. https://www.researchgate.net/profile/Mohammad_Ali55/publication/ 18. de Castro Medeiros LC, Castilho CAR, Braga C, de Souza WV, Regis L, 8904823_Use_of_a_geographic_information_system_for_defining_spatial_ Monteiro AMV. Modeling the dynamic transmission of dengue fever: risk_for_dengue_transmission_in_Bangladesh_Role_for_Aedes_albopictus_ investigating disease persistence. PLoS Negl Trop Dis. 2011;5:e942. in_an_urban_outbreak/links/0f31752f1dd631f6fe000000.pdf. 19. Perkins TA, Reiner RC, Rodriguez-Barraquer I, Smith DL, Scott TW, Cummings 37. Reiter P, Lathrop S, Bunning M, Biggerstaff B, Singer D, Tiwari T, et al. Texas DAT. A review of transmission models of dengue: a quantitative and lifestyle limits transmission of dengue virus [Internet]. 1st ed. United States: qualitative analysis of model features. Dengue and dengue hemorrhagic Centers for Disease Control and Prevention (CDC) (1600 Clifton Road, fever. 2nd ed. Wallingford: CABI; 2014. p. 99–114. Mailstop C-12, Atlanta GA 30333, United States); 2003. p. 86–9. Available 20. Reiner RC, Stoddard ST, Forshey BM, King AA, Ellis AM, Lloyd AL, et al. Time- from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2873752/. varying , serotype-specific force of infection of dengue virus. Proc Natl Acad 38. Teixeira MG, Costa MCN, Natividade MS, Carmo EH, Carneiro DDT. New Sci U S A. 2014; epidemiologic profile of visceral leishmaniasis in an urban endemic/ 21. Arksey H, O’Malley L. Scoping studies: towards a methodological framework. epidemic area of Brazil [Internet]. Blackwell Publishing Ltd; 2015. p. 220. Int J Soc Res Methodol. 2007;8:19–32. 39. Araujo R V, Albertini MR, Costa-da-Silva AL, Suesdek L, Franceschi NCS, 22. Pace R, Pluye P, Bartlett G, Macaulay AC, Salsberg J, Jagosh J, et al. Bastos NM, et al. Sao Paulo urban heat islands have a higher incidence of Testing the reliability and efficiency of the pilot mixed methods dengue than other urban areas [Internet]. 2nd ed. Brazil; 2015. p. 146–155. appraisal tool (MMAT) for systematic mixed studies review. Int J Nurs Available from: http://www.sciencedirect.com/science/article/pii/ Stud. 2012;49:47–53. S1413867014002074 23. Hoffmann TC, Glasziou PP, Boutron I, Milne R, Perera R, Moher D, et al. 40. Fouque F, Garinci R, Gaborit P. Epidemiological and entomological Better reporting of interventions: template for intervention description and surveillance of the co-circulation of DEN-1, DEN-2 and DEN-4 viruses in replication (TIDieR) checklist and guide. BMJ. 2014;348 French Guiana. 1st ed. UK; 2004. p. 41–46. 24. Sang S, Yin W, Bi P, Zhang H, Wang C, Liu X, et al. Predicting local 41. Teixeira TR, Cruz OG. Spatial modeling of dengue and socio-environmental dengue transmission in Guangzhou, China, through the influence of indicators in the city of Rio de Janeiro, Brazil. Cad Saude Publica. 2011/04/ imported cases, mosquito density and climate variability. PLoS One. 27. 2011;27:591–602. 2014/07/16. 2014;9:e102755. 42. Chang FS, Tseng YT, Hsu PS, Chen CD, Lian Ie B, Chao DY. Re-assess Vector 25. Teixeira MDG, Barreto ML, Costa MDCN, Ferreira LDA, Vasconcelos PFC, Indices Threshold as an Early Warning Tool for Predicting Dengue Epidemic Cairncross S, et al. Dynamics of dengue virus circulation: A silent epidemic in a Dengue Non-endemic Country. PLoS Negl Trop Dis. 2015/09/15. 2015;9: in a complex urban area. Trop Med Int Heal. 9th ed. United Kingdom: e0004043. Blackwell Publishing Ltd (9600 Garsington Road, Oxford OX4 2XG, United 43. Wu PC, Guo HR, Lung SC, Lin CY, Su HJ. Weather as an effective predictor Kingdom); 2002;7:757–62. for occurrence of dengue fever in Taiwan. Acta Trop. 2007/07/07. 2007;103: 26. Méndez F, Barreto M, Arias JF, Rengifo G, Muñoz J, Burbano ME, et al. Human 50–7. and mosquito infections by dengue viruses during and after epidemics in a 44. Seidahmed OME, Hassan SA, Soghaier MA, Siam HAM, Ahmed FTA, dengue-endemic region of Colombia. Am J Trop Med Hyg. 2006;74:678–83. Elkarsany MM, et al. Spatial and Temporal Patterns of Dengue Transmission 27. Martínez-Vega RA, Danis-Lozano R, Díaz-Quijano FA, Velasco-Hernández J, along a Red Sea Coastline: A Longitudinal Entomological and Serological Santos-Luna R, Román-Pérez S, et al. Peridomestic infection as a determining Survey in Port Sudan City [Internet]. 9th ed. United States: Public Library of factor of dengue transmission. PLoS Negl Trop Dis. 2015;9:1–18. Science (185 Berry Street, Suite 1300, San Francisco CA 94107, United 28. Anders KL, Nga LH, Thuy NT V, Ngoc T V, Tam CT, Tai LTH, et al. Households States); 2012. p. no pagination. Available from: http://journals.plos.org/ as Foci for Dengue Transmission in Highly Urban Vietnam [Internet]. 2nd ed. plosntds/article?id=10.1371/journal.pntd.0001821. United States: Public Library of Science (E-mail: [email protected]); 2015. p. no 45. World Health Organization. WHO Global Health Expenditure Atlas [Internet]. pagination. Available from: http://journals.plos.org/plosntds/article?id=10. 2014 [cited 2018 May 17]. Available from: http://www.who.int/nha/atlas.pdf 1371/journal.pntd.0003528. 46. Keiser J, Utzinger J, Caldas de Castro M, Smith TA, Tanner M, Singer BH. 29. Estallo EL, Carbajo AE, Grech MG, Frias-Cespedes M, Lopez L, Lanfri MA, et Urbanization in sub-saharan Africa and implication for malaria control. Am J al. Spatio-temporal dynamics of dengue 2009 outbreak in Cordoba City, Trop Med Hyg. 2004/08/28. 2004;71:118–27. Argentina. Acta Trop. 2014/05/06. 2014;136:129–36. 47. Dhiman RC, Yadav YK, Shweta S, Poonam S. Altitude, temperature, and 30. Yang T, Lu L, Fu G, Zhong S, Ding G, Xu R, et al. Epidemiology and vector malaria vectors in Nainital and Udham Singh Nagar districts of Uttarakhand, efficiency during a dengue fever outbreak in Cixi, Zhejiang Province, China India: an evidence-based study [Internet]. 3rd ed. India; 2013. p. 220–224. [Internet]. 1st ed. USA; 2009. p. 148–154. Available from: http://www3. Available from: https://s3.amazonaws.com/academia.edu.documents/ interscience.wiley.com/cgi-bin/fulltext/122509729/HTMLSTART 34068470/paper.pdf?AWSAccessKeyId=AKIAIWOWYYGZ2Y53UL3A&Expires= 31. Peng HJ, Lai HB, Zhang QL, Xu BY, Zhang H, Liu WH, et al. A local outbreak 1533631438&Signature= of dengue caused by an imported case in Dongguan China. BMC Public RqMJv%2BivXXzRWHJdbAx%2B8zOvoKw%3D&response-contentdisposition= Health. 2012/01/27. 2012;12:83. inline%3B%20filename%3DAltitude_temperature_and_malaria_vectors.pdf 32. Barrera R, Amador M, MacKay AJ. Population dynamics of Aedes aegypti 48. Sissoko MS, van den Hoogen LL, Samake Y, Tapily A, Diarra AZ, Coulibaly M, and dengue as influenced by weather and human behavior in San Juan, et al. Spatial Patterns of Plasmodium falciparum Clinical Incidence, Puerto Rico. PLoS Negl Trop Dis. 2011/12/30. 2011;5:e1378. Asymptomatic Parasite Carriage and Anopheles Density in Two Villages in 33. Sanchez L, Vanlerberghe V, Alfonso L, Marquetti MDC, Guzman MG, Bisset J, Mali. Am J Trop Med Hyg. 2015/09/02. 2015;93:790–7. et al. Aedes aegypti larval indices and risk for dengue epidemics [Internet]. 49. Peterson I, Borrell LN, El-Sadr W, Teklehaimanot A. A temporal-spatial 5th ed. United States: Centers for Disease Control and Prevention (CDC) analysis of malaria transmission in Adama, Ethiopia. Am J Trop Med Hyg. (1600 Clifton Road, Mailstop C-12, Atlanta GA 30333, United States); 2006. 2009/12/10. 2009;81:944–9. p. 800–6. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/ 50. Müller DA, Charlwood JD, Felger I, Ferreira C, do Rosario V, Smith T. PMC3374431/. Prospective risk of morbidity in relation to multiplicity of infection with 34. Correa PR, Franca E, Bogutchi TF. [Aedes aegypti infestation and occurrence Plasmodium falciparum in Sao Tome. Acta Trop. 2001/03/07. 2001;78:155–62. of dengue in the city of Belo Horizonte, Brazil]. Rev Saude Publica. 2005/01/ 51. Ivan E, Crowther NJ, Rucogoza AT, Osuwat LO, Munyazesa E, Mutimura E, 18. 2005;39:33–40. et al. Malaria and helminthic co-infection among HIV-positive pregnant 35. Dibo MR, Chierotti AP, Ferrari MS, Mendonca AL, Chiaravalloti Neto F. Study women: prevalence and effects of antiretroviral therapy. Acta Trop. 2012/09/ of the relationship between Aedes (Stegomyia) aegypti egg and adult 04. 2012;124:179–84. Eder et al. Infectious Diseases of Poverty (2018) 7:90 Page 24 of 24

52. Zhao XF, Zhang JN, Dong HJ, Zhang T, Bian GL, Sun YW, et al. [Internet]. 9602nd ed. UK; 2007. p. 1840–1846. Available from: https://s3. Epidemiological characteristics of malaria in Ningbo City, China 2000–2011 amazonaws.com/academia.edu.documents/42727896/Infection_with_ [Internet]. 2nd ed. Malaysia: Malaysian Society of Parasitology and Tropical chikungunya_virus_in_Ital20160216-9308-1w1b1g6.pdf?AWSAccessKeyId= Medicine (Jalan Pahang, Kuala Lumpur 50588, Malaysia); 2013. p. 267–76. AKIAIWOWYYGZ2Y53UL3A&Expires=1533632155&Signature= Available from: http://msptm.org/files/267_-_276_Guo_Zhang_Xu.pdf YYpBM9q9qeF4UOvquP2kpM%2B9lsA%3D&response-contentdisposition= 53. Lee YCA, Tang CS, Ang LW, Han HK, James L, Goh KT. Epidemiological inline%3B%20filename%3DInfection_with_chikungunya_virus_in_Ital.pdf. characteristics of imported and locally-acquired malaria in Singapore. Ann 71. Ho K, Ang LW, Tan BH, Tang CS, Ooi PL, James L, et al. Epidemiology and Acad Med Singap. 2009;38:840–9. control of chikungunya fever in Singapore [Internet]. 4th ed. United 54. Dev V, Phookan S, Sharma VP, Anand SP. Physiographic and entomologic Kingdom: W.B. Saunders Ltd (32 Jamestown Road, London NW1 7BY, United risk factors of malaria in Assam, India. Am J Trop Med Hyg. 2004/11/02. Kingdom); 2011. p. 263–70. Available from: http://ovidsp.ovid.com/ovidweb. 2004;71:451–6. cgi?T=JS&PAGE=reference&D=emed13&NEWS=N&AN = 51319986. 55. Zhang Y, Liu Q, Luan R, Liu X, Zhou G, Jiang J, et al. Spatial-temporal 72. Vasconcelos PFC, Costa ZG, Travassos Da Rosa ES, Luna E, Rodrigues SG, analysis of malaria and the effect of environmental factors on its incidence Barros VLRS, et al. Epidemic of jungle yellow fever in Brazil, 2000: in Yongcheng, China, 2006–2010 [Internet]. 544th ed. UK; 2012. Available Implications of climatic alterations in disease spread [Internet]. 3rd ed. from: http://www.biomedcentral.com/1471-2458/12/544/abstract United States: Wiley-Liss Inc. (111 River Street, Hoboken NJ 07030–5774, 56. Ebenezer A, Noutcha AEM, Okiwelu SN. Relationship of annual United States); 2001. p. 598–604. Available from: http://citeseerx.ist.psu.edu/ entomological inoculation rates to malaria transmission indices, Bayelsa viewdoc/download?doi=10.1.1.543.2170&rep=rep1&type=pdf. State, Nigeria [Internet]. 1st ed. India: Malaria Research Center (E-mail: 73. Gould LH, Osman MS, Farnon EC, Griffith KS, Godsey MS, Karch S, et al. An [email protected]); 2016. p. 46–53. Available from: http://www.nimr.org.in/ outbreak of yellow fever with concurrent chikungunya virus transmission in assets/531046.pdf South Kordofan, Sudan, 2005 [Internet]. 12th ed. Netherlands: Elsevier (P.O. 57. Moreno JE, Rubio-Palis Y, Paez E, Perez E, Sanchez V. Abundance, biting Box 211, Amsterdam 1000 AE, Netherlands); 2008. p. 1247–54. behaviour and parous rate of anopheline mosquito species in relation to 74. Carver S, Sakalidis V, Weinstein P. House mouse abundance and Ross River malaria incidence in gold-mining areas of southern Venezuela. Med Vet virus notifications in Victoria, Australia. Int J Infect Dis. 2008/05/13. 2008;12: Entomol. 2007/12/21. 2007;21:339–49. 528–33. 58. Girod R, Roux E, Berger F, Stefani A, Gaborit P, Carinci R, et al. Unravelling 75. de Souza CE, Pinter A, Donalisio MR. Risk factors associated with the the relationships between Anopheles darlingi (Diptera: Culicidae) densities, transmissionof Brazilian spotted fever in the Piracicaba river basin, state of environmental factors and malaria incidence: understanding the variable São Paulo. Brazil Rev Soc Bras Med Trop. 2015;48:11–7. patterns of malarial transmission in French Guiana (South America). Ann 76. Pham H V, Dang DT, Tran Minh NN, Nguyen ND, Nguyen T V. Correlates of Trop Med Parasitol. 2011/03/15. 2011;105:107–22. environmental factors and human plague: an ecological study in Vietnam. 59. El Sayed BB, Arnot DE, Mukhtar MM, Baraka OZ, Dafalla AA, Elnaiem DE, et Int J Epidemiol. 2009/07/09. 2009;38:1634–41. al. A study of the urban malaria transmission problem in Khartoum. Acta 77. Hay SI, Guerra CA, Tatem AJ, Atkinson PM, Snow RW. Urbanization, malaria Trop. 2000/03/10. 2000;75:163–71. transmission and disease burden in Africa. Nat Rev Microbiol. 2004/12/21. – 60. Ye Y, Hoshen M, Kyobutungi C, Louis VR, Sauerborn R. Local scale prediction 2005;3:81 90. of Plasmodium falciparum malaria transmission in an endemic region using 78. Arunachalam N, Tana S, Espino F, Kittayapong P, Abeyewickrem W, Wai KT, temperature and rainfall. Glob Heal Action. 2010/01/07. 2009;2. Tyagi BK, Kroeger A, Sommerfeld J, Petzold M. Eco-bio-social determinants 61. Wertheim HFL, Horby P, Woodall JP. Aedes aegypti and Aedes albopictus. of dengue vectorbreeding: a multicountry study in urban and periurban Atlas Hum Infect Dis. Wiley-Blackwell; 2012. p. Section 1: Infectious Disease Asia. Bull World Health Organ. 2010;88:173-84. Available from: https://www. Drivers; 33–34. scielosp.org/pdf/bwho/2010.v88n3/173-184/en. 62. Salomon OD, Orellano PW, Lamfri M, Scavuzzo M, Dri L, Farace MI, et al. 79. Stoddard ST, Morrison AC, Vazquez-Prokopec GM, Soldan VP, Kochel TJ, spatial distribution asssociated with a focus of tegumentary Kitron U, et al. The role of human movement in the transmission of vector- leishmaniasis in Las Lomitas, Formosa, Argentina, 2002. Mem Inst Oswaldo borne pathogens. PLoS Negl Trop Dis. 2009;3:e481. Cruz. 2006/07/25. 2006;101:295–9. 80. World Health Organization. Joint external evaluation tool: International 63. Thomaz Soccol V, de Castro EA, Schuhli GS e, de Carvalho Y, Marques E, de Health Regulations (2005) [Internet]. 2016. Available from: http://apps.who. Fatima Pereira E, et al. A new focus of cutaneous leishmaniasis in the int/iris/handle/10665/204368. central area of Parana State, southern Brazil [Internet]. 3rd ed. Netherlands: 81. Rodriguez-Barraquer I, Cordeiro MT, Braga C, de Souza WV, Marques ET. Elsevier (P.O. Box 211, Amsterdam 1000 AE, Netherlands); 2009. p. 308–15. Cummings D a T. From re-emergence to hyperendemicity: the natural – 64. de Camargo-Neves VLF, Spínola R, Lage L. A Leishmaniose Visceral history of the dengue epidemic in Brazil. PLoS Negl Trop Dis. 2011;5:1 7. Americana no estado de São Paulo: situação epidemiológica em 2001-2002. 82. Schwartz J, Spix C, Touloumi G, Bachárová L, Barumamdzadeh T, le Tertre A, Rev Soc Bras Med Trop. 2003;36:27–9. et al. Methodological issues in studies of air pollution and daily counts of deaths or hospital admissions. J Epidemiol Community Health [Internet]. 65. Uranw S, Hasker E, Roy L, Meheus F, Das ML, Bhattarai NR, et al. An outbreak 1996 [cited 2017 Sep 14];50 Suppl 1:S3–11. Available from: https://www. investigation of visceral leishmaniasis among residents of Dharan town, ncbi.nlm.nih.gov/pmc/articles/PMC1060881/pdf/jepicomh00187-0005.pdf. eastern Nepal, evidence for urban transmission of Leishmania donovani [Internet]. 21st ed. UK; 2013. Available from: http://www.biomedcentral.com/ 1471-2334/13/21 66. Salazar PM, Rojas G, Bucio M, Cabrera M, García G, Ruiz A, et al. Seroprevalencia de anticuerpos contra Trypanosoma cruzi y su asociación con factores de riesgo en menores de 18 años de Veracruz. México Rev Panam Salud Publica. 2007;22:75–82. 67. Medrano-Mercado N, Ugarte-Fernandez R, Butron V, Uber-Busek S, Guerra HL, Araujo-Jorge TC, et al. Urban transmission of Chagas disease in Cochabamba, Bolivia. Mem Inst Oswaldo Cruz. 2008/09/18. 2008;103:423–30. 68. Nielsen CF, Armijos M V, Wheeler S, Carpenter TE, Boyce WM, Kelley K, et al. Risk factors associated with human infection during the 2006 West Nile virus outbreak in Davis, a residential community in northern California. Am J Trop Med Hyg. 2008/01/12. 2008;78:53–62. 69. Godsey Jr. MS, Burkhalter K, Young G, Delorey M, Smith K, Townsend J, et al. Entomologic investigations during an outbreak of West Nile virus disease in Maricopa County, Arizona, 2010. Am J Trop Med Hyg. 2012/10/31. 2012;87: 1125–31. 70. Rezza G, Nicoletti L, Angelini R, Romi R, Finarelli AC, Panning M, et al. Infection with chikungunya virus in Italy: an outbreak in a temperate region