A fundamental challenge of disease surveillance systems is how to transform data that is routinely collected into useful, actionable evidence that can inform control interventions. Disease surveillance systems typically focus on analyzing aggregate counts of cases over defined time periods to stratify the risk of populations (WHO, 2012). However, the use of raw case counts, or incidences, as a measure of disease risk can frequently be misleading because the quality of surveillance often differs significantly both across countries and within regions of a country making comparisons inappropriate. Areas with more complete reporting of disease may inaccurately appear to have more disease simply because reported cases scale linearly with completeness of reporting. Surveillance systems may change over time (e.g. improving in completeness over time) making comparisons difficult or even impossible. These problems are particularly troubling when examining disease trends or ranking regions according to their disease risk. Recent efforts have tried to improve quantification of disease burden by pooling numerous sources of data. For example, disease mapping methods that combine disease presence/absence data, environmental covariates and available incidence data (from cohort or cross-sectional studies) have been used to predict spatial limits and global case counts for diseases including malaria, dengue and several neglected tropical diseases. (Hay et al., 2009; Bhatt et al., 2013; Pigott et al., 2014; Hay et al., 2013) However, while these methods are successful in identifying boundaries of endemic areas, the robustness of these approaches to quantify transmission within endemic areas has not been validated.

For most immunizing infections (infections that generate long-lasting sterilizing immunity), serological surveys are regarded as the gold standard to measure the susceptible fraction and infer the extent of transmission, as they provide a direct measure of the proportion of the population that has been infected. This is particularly useful for diseases such as dengue, influenza or Zika, where the asymptomatic to symptomatic ratios are large or unknown. Methods to estimate transmission parameters from age-stratified serological data have been available for many years(Muench, 1959) and have been used to analyze trends in transmission for multiple diseases including measles (Grenfell and Anderson, 1985), hepatitis A(Schenzle et al., 1979), dengue (Ferguson et al., 1999; Imai et al., 2015; Rodriguez-Barraquer et al., 2011; Rodríguez-Barraquer et al., 2014), pertussis (Kretzschmar et al., 2010), influenza (Miller et al., 2010), malaria (Bretscher et al., 2013) and chikungunya (Salje et al., 2016). While well conducted age-stratified serological surveys can provide an unbiased measure of the susceptibility profile of a population and transmission parameters, they are generally not part of routine surveillance activities and are therefore only available for a limited number of locations and time points. For example, an extensive review of the dengue literature published recently found only three population-based serosurveys conducted in Brazil over the past 10 years, despite being the country that currently reports the largest number of cases worldwide (Fritzell et al., 2018). Similarly, only one and three studies were identified for the Philippines and Thailand, respectively, even though the burden of dengue is very high (Fritzell et al., 2018). Thus, the picture provided by serology is very incomplete and limited when trying to characterize transmission across time and space both within and between countries.

While aggregated case counts can be misleading when quantifying disease risk, the age distribution of incident cases contains a lot of information on the age-specific susceptibility of the population. Importantly, the age distributionof cases is also largely robust to under-reporting, facilitating the comparison between locations or over time. By combining age-specific incidence data and mechanistic models of how population immunity is acquired over time, it is possible to estimate key transmission parameters and obtain a much more accurate picture of the local and global burdens of disease. Since age-specific counts are routinely collected by surveillance systems as part of standard practice, they represent a missed opportunity to further our understanding of epidemic patterns for many diseases.

Here, we use dengue virus as an example to illustrate how age-specific incidence data can be used to quantify disease transmission and inform control interventions. Dengue is a relevant example because, despite being the most widely spread mosquito-transmitted virus, large gaps remain in our understanding of its global and local epidemiology (Fritzell et al., 2018). We present a model to estimate the transmission intensity of dengue from age-specific incidence data, and apply it to surveillance data from administrative units in four countries that suffer from endemic dengue transmission (Thailand, Colombia, Brazil and Mexico). We validate our estimates using serological data and show that they correlate well with known environmental drivers of dengue transmission at subnational level. Finally, we show how this approach could be used to guide the implementation of dengue control strategies such as vaccination.

We estimated the average forces of infection (FOI) over the last 20 years for 148 administrative level one units where age specific case data was available (Table 1, Figure 1B): 72/76 provinces of Thailand, 28/32 departments of Colombia, 21/27 states of Brazil, and 27/31 states of Mexico. These administrative units comprise 90%, 99%, 90% and 91% of the population at risk of these countries respectively. We also estimated forces of infection for 211 municipalities (administrative level two units) of Colombia where at least 200 cases were reported over the period covered by the data.

Figure 1 with 2 supplements see all

Download asset Open asset

The average FOI (the rate at which susceptible individuals that will become infected in a year by any dengue serotype) was 0.096 in Thailand (95%CI 0.092-0.100), 0.132 (95%CI 0.128-0.136) in Colombia, 0.124 (95%CI 0.116-0.128) in Brazil, and 0.052 (95%CI 0.048-0.056) in Mexico. This implies that on average 9.6% of the susceptible population in Thailand gets infected every year by any of the circulating serotypes ($1-e^{-FOI}$). Similarly, 12% of the susceptible population of Colombia and Brazil, and 5% of the susceptible population of Mexico get infected yearly.

However, as expected, transmission intensity varied greatly within countries, ranging between 0.04 and 0.15 (coefficient of variation (CV) = 0.27) across provinces of Thailand; between 0.02 and 0.20 across departments of Colombia (CV = 0.37), between 0.02 and 0.24 across states of Brazil (CV = 0.56); and between 0.01 and 0.092 across states of Mexico (CV = 0.45). Transmission was highest in the North East of Brazil (average FOI of 0.152) and in the Caribbean region of Colombia (average FOI of 0.156). There was also substantial heterogeneity within departments of Colombia. The mean CV for 15 departments where we had estimates for more than 5 municipalities was 0.4.

For the 16 locations where we had access to both age-stratified serological data (the gold standard) and case data, we found good correlation between the estimates of the FOI derived from both data sources (R2 = 0.73, 95% CI 0.51–0.87, Figure 1C). In contrast, we found no correlation between recent incidence of dengue in these locations (the average yearly incidence over the most recent 5 years of data) and the estimates of FOI derived from serological data (R2 = 0.002, Figure 1—figure supplement 2).

Since estimates of transmission intensity derived from seroprevalence data are only available for a small number of locations, to further validate our method we also explored the association between our estimates of the FOI for 211 Colombian municipalities (administrative level 2) with known environmental drivers of dengue transmission including temperature, elevation, precipitation a published metric of Aedes aegypti abundance and population density (Figure 2, Figure 2—figure supplement 1). Models were weighted by the number of cases used to estimate the FOI. On average, the FOI increased by 0.006 (95% CI 0.004–0.007, R² = 0.19) for each additional °C in temperature and, similarly, it decreased by 0.005 (95% CI 0.003, 0.006, R² = 0.21) for each 100 m increase in elevation (Table 2). While population density was not associated with FOI estimates in unadjusted analyses, a twofold increase in density was associated with a 0.007 (95% 0.004–0.009) increase in FOI in the best fitting adjusted model. This model included elevation, population density and precipitation, and explained 35% (95%CI 23–50%) of the variance.

Figure 2 with 2 supplements see all

Download asset Open asset

In contrast, the recent incidence of dengue in these municipalities was not correlated with temperature, elevation or Aedes aegypti abundance (R2 0.01, 0.01, and 0.00 respectively, Table 3). We did find a negative association between population density and incidence, indicating a 6% (95%CI 2–10%) decrease in log incidence for each 2-fold increase in population density.

Application: Guiding dengue vaccination policy

The only available dengue vaccine (Dengvaxia) has been licensed for use in children 9 years or older in 20 countries including Thailand, Brazil and Mexico. The WHO currently recommends confirmation (by virology or serology) of prior dengue infection at the individual level before vaccinating individuals, and therefore there is interest in identifying populations with high seropositivity to target pre-vaccination screening (WHO, 2018). In the absence of appropriate rapid serological assays that would allow implementing this individual screening strategy, an alternative that has been discussed is rolling out the vaccine in settings with 80% or greater seropositivity among 9 year olds. Using our estimates of the FOI, we calculated the proportion of the population expected to be seropositive at age 9 years for each of the subnational units represented in our data. Our results suggest that only a small minority of locations in Colombia and Brazil have > 80% seropositivity in this age group (Figure 3, Figure 3—source data 1). The expected proportion seropositive at 9 years of age ranged between 0.35 and 0.75 in provinces of Thailand; 0.13 and 0.85 in departments of Colombia; 0 and 0.88 in states of Brazil; and 0.07 and 0.56 in states of Mexico. The seroprevalence was estimated to be high enough in only 2/28 Colombian departments, 4/25 Brazilian states and none of the 72 Thai provinces or 27 Mexican states. Furthermore, even within the two Colombian departments that met the 80% seroprevalence threshold, only 9/13 (70%) of the municipalities evaluated reached this level. This proportion would probably be much lower had we considered all the municipalities in these departments, as those excluded had, on average, lower FOI (Figure 2—figure supplement 2).

Figure 3

Download asset Open asset

Figure 3—source data 1

Estimated forces of infection and seroprevalences for 148 spatial units.

https://doi.org/10.7554/eLife.45474.009

Download elife-45474-fig3-data1-v2.csv

In locations where seroprevalence among 9 year olds was estimated to be less than 80%, we calculated the age-group that could be targeted to ensure a seroprevalence > 80%. Our results suggest that, to comply with the WHO-SAGE recommendations, it would be necessary to target children 14 years of age or older in over 70% (108/148) of locations. Furthermore, in approximately 50% of the locations evaluated, it would be necessary to target individuals aged 18 years or older, precluding school-based vaccination strategies.

Sensitivity analyses

Since different surveillance systems have different reporting practices, we compared FOI estimates obtained when using data from severe cases alone (DHF or severe dengue) to estimates obtained when all dengue cases were considered jointly (Figure 4). This comparison was limited to Colombia, Brazil and Mexico because we only had DHF data from Thailand.

Figure 4

Download asset Open asset

In general, estimates derived from all cases tended to underestimate those obtained from DHF cases, but the correlation between the estimates was good (Figure 4A) as well as the correlation with the gold standard (Figure 4B). The correlation was highest for Colombia and Mexico (r = 0.74 and r = 0.78, respectively), but less so for Brazil (r = 0.48). However, 2/3 of the Brazilian locations that showed discrepant results had the smallest number of DHF cases (n = 217 and 265) suggesting that sample size may have played a role. Similarly, all the Colombian locations where there was a large discrepancy had very low counts of DHF cases (n = 41, 55, and 48, respectively).

In this paper, we use dengue virus as an example to illustrate how age-specific incidence data can be used to quantify disease transmission and inform control interventions. We present a method to estimate the FOI of dengue from age-specific incidence data, and use it to generate country-wide subnational estimates for Thailand, Colombia, Brazil and Mexico. Our estimates correlate well with those derived from seroprevalence data (the gold standard), capture the expected spatial heterogeneity in risk, and outperform all other metrics traditionally used by surveillance systems, such as the crude incidence.

Most surveillance systems use crude case counts or incidences to describe temporal and spatial trends of communicable diseases. Our results underscore the extent to which, for immunizing infections in endemic circulation, recent incidence may be a poor metric of transmission and may be misleading when ranking spatial units (Figure 1—figure supplement 2). Immunity of the population in high-transmission settings reduces the number of individuals that are susceptible to infection. As a result, incidence in places where transmission intensity is lower, but people remain susceptible for a longer period of time, may be roughly equivalent to that in higher transmission intensity areas. In contrast, metrics such as the FOI, that quantify the risk among the susceptible population, better reflect the underlying transmission potential. Since FOI estimates are derived from the age distribution of incidence, and not from the aggregate counts, they are more robust to differences in surveillance efficiency and can be obtained with relatively small numbers of yearly cases.

Our estimates of the average transmission intensity of dengue in Thailand, Colombia, Brazil and Mexico are consistent with large variation in dengue transmission between and within countries. Spatial heterogeneity was substantial not only between, but also within administrative level one units. While this heterogeneity is probably driven by multiple environmental, socio-economic and demographic factors, our results suggest that as much as 35% of the variance may be explained by differences in temperature/elevation and population density alone. Transmission intensity was highest in northeastern Brazil, northern Colombia and eastern Thailand. Northeastern Brazil was also the region that experienced the highest incidence of Zika during the 2015/2016 epidemic, and this is not surprising given that both viruses are transmitted by the same mosquito vector (A. aegypti)(Lowe et al., 2018).

While age-stratified serological data remains the gold standard to quantify dengue transmission, our results illustrate how inferences derived from age-specific surveillance data could be used to inform control interventions such as vaccination. For example, the high transmission intensities (and expected seroprevalences) found in Northeastern Brazil suggest that this region might be ideal to implement pre-vaccination screening strategies. In addition, the large heterogeneity estimated in Colombia suggests that decisions of where to deploy control interventions, including vaccines, should be made at least at the municipal/district (administrative level 2) level. Since age-specific counts are routinely collected by many surveillance systems, they represent a unique opportunity to further our understanding of dengue burden and risk at spatial scales at which serological data is rarely available. While here we focus on presenting average forces of infection over 20 years, the same methods can also be used to estimate yearly forces of infection (Hoang Quoc et al., 2016; Cummings et al., 2009).

There are several limitations of using age-specific surveillance data to estimate transmission parameters of dengue. Of concern are differences in case definitions and reporting practices between and within countries. While the case-definition of DHF is quite specific, milder forms of the disease are characterized by a non-specific febrile syndrome. Therefore, data from locations that report ‘all cases’ rather than more severe forms (DHF) are more likely to be misclassified, and could be less useful to infer transmission histories. Nevertheless, our results and sensitivity analyses are encouraging and suggest that that even data from all dengue cases can be useful to infer transmission patterns.

Our model also makes several assumptions that may be questionable. It assumes that the age distribution of cases represents the age distribution of secondary infections, thus ignoring the potential contribution of primary, tertiary and quaternary infections. It also assumes that risk of infection, symptoms, and health seeking behavior are not age dependent, even though there is some evidence that suggests that this may not be the case(Katzelnick et al., 2018; Guzmán and Kourí, 2002; Thai et al., 2011). It assumes equal circulation of all serotypes, despite the known dominance of specific serotypes for extended periods of time in the Americas (Katzelnick et al., 2018; Teixeira et al., 2009). Finally, it estimates transmission intensities over extended periods of time (20 years), averaging over variations in the FOI that may occur at shorter time scales. Despite these simplifications, validation of our estimates using age-stratified serological data from 16 locations is very encouraging, as is the good correlation with known drivers of dengue transmission. While further validation is desirable, it is important to note that some discrepancy between our estimates and those derived from seroprevalence data is expected as the two sources of data do not represent exactly the same period of time and location. For example, most of the serosurveys available were conducted in specific urban centers, while case data represents the full administrative unit.

Targeting control interventions against dengue and other communicable diseases requires good understanding of when and where transmission is occurring. Careful analyses of age-specific incidence data can provide very useful information to characterize transmission across time and space. While here we focus on dengue, the same approach should be generalizable to other immunizing infections including chikungunya and Zika. Open access to age-specific incidence data would greatly enrich and enhance existing efforts to quantify trends in the global burden of disease.

The code to implement the model described in our study is available at https://github.com/isabelrodbar/dengue_foi (copy archived at https://github.com/elifesciences-publications/dengue_foi). The case data used for the analyses is publicly available and can be accessed through the following links links: Brazil- http://tabnet.datasus.gov.br/cgi/deftohtm.exe?sih/cnv/mruf.def; Thailand - http://www.boe.moph.go.th/boedb/surdata/index.php; Colombia - http://www.ins.gov.co/lineas-de-accion/Subdireccion-Vigilancia/sivigila/Paginas/vigilancia-rutinaria.aspx and https://www.sispro.gov.co/Pages/Home.aspx; Mexico - http://www.epidemiologia.salud.gob.mx/anuario/html/anuarios.html.

1. 1. Amaya-Larios IY
   2. Ramos-Castañeda J
   3. Sepúlveda-Salinas KJ
   4. Mayer SV
   5. Falcón-Lezama JA
   6. Galeana-Hernández M
   7. Vasilakis N
   8. Comas-García A
   9. Martínez-Vega RA

   (2014) Seroprevalence of Neutralizing Antibodies Against Dengue Virus in Two Localities in the State of Morelos, Mexico

   The American Journal of Tropical Medicine and Hygiene 91:1057–1065.

   https://doi.org/10.4269/ajtmh.14-0145

   • Google Scholar
2. 1. Bhatt S
   2. Gething PW
   3. Brady OJ
   4. Messina JP
   5. Farlow AW
   6. Moyes CL
   7. Drake JM
   8. Brownstein JS
   9. Hoen AG
   10. Sankoh O
   11. Myers MF
   12. George DB
   13. Jaenisch T
   14. Wint GRW
   15. Simmons CP
   16. Scott TW
   17. Farrar JJ
   18. Hay SI

   (2013) The global distribution and burden of dengue

   Nature 496:504–507.

   https://doi.org/10.1038/nature12060

   • Google Scholar
3. 1. Bretscher MT
   2. Supargiyono S
   3. Wijayanti MA
   4. Nugraheni D
   5. Widyastuti AN
   6. Lobo NF
   7. Hawley WA
   8. Cook J
   9. Drakeley CJ

   (2013) Measurement of Plasmodium falciparum transmission intensity using serological cohort data from Indonesian schoolchildren

   Malaria Journal 12:21.

   https://doi.org/10.1186/1475-2875-12-21

   • PubMed
   • Google Scholar
4. Report

   1. Bureau of Epidemiology, Department of Disease Control

   (2019) National Disease Surveillance (Report 506) Entered Program 506

   Department of Disease Control, MoPH, Thailand.

   http://www.boe.moph.go.th/boedb/surdata/index.php

   • Google Scholar
5. 1. Burke DS
   2. Nisalak A
   3. Johnson DE
   4. Scott RM

   (1988) A prospective study of dengue infections in Bangkok

   The American Journal of Tropical Medicine and Hygiene 38:172–180.

   https://doi.org/10.4269/ajtmh.1988.38.172

   • PubMed
   • Google Scholar
6. 1. Cummings DA
   2. Iamsirithaworn S
   3. Lessler JT
   4. McDermott A
   5. Prasanthong R
   6. Nisalak A
   7. Jarman RG
   8. Burke DS
   9. Gibbons RV

   (2009) The impact of the demographic transition on dengue in Thailand: insights from a statistical analysis and mathematical modeling

   PLOS Medicine 6:e1000139.

   https://doi.org/10.1371/journal.pmed.1000139

   • PubMed
   • Google Scholar
7. Website

   1. DANE

   (2019) Estimación y proyección de población nacional, departamental y municipal por sexo, grupos quinquenales de edad y edades simples de 0 a 26 años 1985-2020

   Accessed January 1, 2019.

   https://www.dane.gov.co/index.php/estadisticas-por-tema/demografia-y-poblacion/proyecciones-de-poblacion
8. Website

   1. Direccion General de Epidemiología

   (2017) Sistema nacional de vigilancia epidemiológica (Sinave), Secretaría de salud

   Accessed July 7, 2017.

   http://www.epidemiologia.salud.gob.mx/anuario/html/anuarios.html
9. 1. Ferguson NM
   2. Donnelly CA
   3. Anderson RM

   (1999) Transmission dynamics and epidemiology of dengue: insights from age-stratified sero-prevalence surveys

   Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 354:757–768.

   https://doi.org/10.1098/rstb.1999.0428

   • PubMed
   • Google Scholar
10. 1. Fritzell C
    2. Rousset D
    3. Adde A
    4. Kazanji M
    5. Van Kerkhove MD
    6. Flamand C

    (2018) Current challenges and implications for dengue, chikungunya and Zika seroprevalence studies worldwide: A scoping review

    PLOS Neglected Tropical Diseases 12:e0006533.

    https://doi.org/10.1371/journal.pntd.0006533

    • PubMed
    • Google Scholar
11. 1. Grenfell BT
    2. Anderson RM

    (1985) The estimation of age-related rates of infection from case notifications and serological data

    Journal of Hygiene 95:419–436.

    https://doi.org/10.1017/S0022172400062859

    • PubMed
    • Google Scholar
12. 1. Guzmán MG
    2. Kourí G

    (2002) Dengue: an update

    The Lancet Infectious Diseases 2:33–42.

    https://doi.org/10.1016/S1473-3099(01)00171-2

    • PubMed
    • Google Scholar
13. 1. Hay SI
    2. Guerra CA
    3. Gething PW
    4. Patil AP
    5. Tatem AJ
    6. Noor AM
    7. Kabaria CW
    8. Manh BH
    9. Elyazar IR
    10. Brooker S
    11. Smith DL
    12. Moyeed RA
    13. Snow RW

    (2009) A world malaria map: plasmodium falciparum endemicity in 2007

    PLOS Medicine 6:e1000048.

    https://doi.org/10.1371/journal.pmed.1000048

    • PubMed
    • Google Scholar
14. 1. Hay SI
    2. Battle KE
    3. Pigott DM
    4. Smith DL
    5. Moyes CL
    6. Bhatt S
    7. Brownstein JS
    8. Collier N
    9. Myers MF
    10. George DB
    11. Gething PW

    (2013) Global mapping of infectious disease

    Philosophical Transactions of the Royal Society B: Biological Sciences 368:20120250.

    https://doi.org/10.1098/rstb.2012.0250

    • Google Scholar
15. 1. Hladish TJ
    2. Pearson CA
    3. Chao DL
    4. Rojas DP
    5. Recchia GL
    6. Gómez-Dantés H
    7. Halloran ME
    8. Pulliam JR
    9. Longini IM

    (2016) Projected impact of dengue vaccination in yucatán, Mexico

    PLOS Neglected Tropical Diseases 10:e0004661.

    https://doi.org/10.1371/journal.pntd.0004661

    • PubMed
    • Google Scholar
16. 1. Hoang Quoc C
    2. Henrik S
    3. Isabel R-B
    4. In-Kyu Y
    5. Chau NVV
    6. Hung NT
    7. Tuan HM
    8. Lan PT
    9. Willis B
    10. Nisalak A
    11. Kalayanarooj S
    12. Cummings DAT
    13. Simmons CP

    (2016) Synchrony of Dengue Incidence in Ho Chi Minh City and Bangkok

    PLOS Neglected Tropical Diseases 10:e0005188.

    https://doi.org/10.1371/journal.pntd.0005188

    • Google Scholar
17. 1. Imai N
    2. Dorigatti I
    3. Cauchemez S
    4. Ferguson NM

    (2015) Estimating dengue transmission intensity from sero-prevalence surveys in multiple countries

    PLOS Neglected Tropical Diseases 9:e0003719.

    https://doi.org/10.1371/journal.pntd.0003719

    • PubMed
    • Google Scholar
18. 1. Imai N
    2. Dorigatti I
    3. Cauchemez S
    4. Ferguson NM

    (2016) Estimating dengue transmission intensity from Case-Notification data from multiple countries

    PLOS Neglected Tropical Diseases 10:e0004833.

    https://doi.org/10.1371/journal.pntd.0004833

    • PubMed
    • Google Scholar
19. Website

    1. Instituto Brasileiro de Geografia e Estadistica

    (2017) Projeção da população do brasil por sexo e idade: 2000-2060

    Accessed July 11, 2017.

    http://www.ibge.gov.br/home/estatistica/populacao/projecao_da_populacao/2013/default_tab.shtm
20. Website

    1. Instituto Nacional de Estadistica y Geografia

    (2016) Censos y conteo de poblacion y vivienda

    Accessed January 7, 2015.

    https://www.inegi.org.mx/temas/estructura/default.html#Tabulados
21. Website

    1. Instituto Nacional de Salud Colombia

    (2016) Vigilancia rutinaria

    Accessed October 15, 2016.

    http://www.ins.gov.co/lineas-de-accion/Subdireccion-Vigilancia/sivigila/Paginas/vigilancia-rutinaria.aspx
22. 1. Katzelnick LC
    2. Ben-Shachar R
    3. Mercado JC
    4. Rodriguez-Barraquer I
    5. Elizondo D
    6. Arguello S
    7. Nuñez A
    8. Ojeda S
    9. Sanchez N
    10. Lopez Mercado B
    11. Gresh L
    12. Burger-Calderon R
    13. Kuan G
    14. Gordon A
    15. Balmaseda A
    16. Harris E

    (2018) Dynamics and determinants of the force of infection of dengue virus from 1994 to 2015 in Managua, Nicaragua

    PNAS 115:10762–10767.

    https://doi.org/10.1073/pnas.1809253115

    • PubMed
    • Google Scholar
23. 1. Kretzschmar M
    2. Teunis PF
    3. Pebody RG

    (2010) Incidence and reproduction numbers of pertussis: estimates from serological and social contact data in five European countries

    PLOS Medicine 7:e1000291.

    https://doi.org/10.1371/journal.pmed.1000291

    • PubMed
    • Google Scholar
24. 1. Lowe R
    2. Barcellos C
    3. Brasil P
    4. Cruz O
    5. Honório N
    6. Kuper H
    7. Carvalho M

    (2018) The zika virus epidemic in Brazil: from discovery to future implications

    International Journal of Environmental Research and Public Health 15:96.

    https://doi.org/10.3390/ijerph15010096

    • Google Scholar
25. 1. Miller E
    2. Hoschler K
    3. Hardelid P
    4. Stanford E
    5. Andrews N
    6. Zambon M

    (2010) Incidence of 2009 pandemic influenza A H1N1 infection in England: a cross-sectional serological study

    The Lancet 375:1100–1108.

    https://doi.org/10.1016/S0140-6736(09)62126-7

    • Google Scholar
26. Website

    1. Ministério da Saude

    (2017) Sistema de informações hospitalares/SIH/SUS

    Accessed January 20, 2017.

    http://tabnet.datasus.gov.br/cgi/deftohtm.exe?sih/cnv/mruf.def
27. Book

    1. Muench H

    (1959)

    Catalytic Models in Epidemiology

    Cambridge, Massachusets: Harvard University Press.

    • Google Scholar
28. Website

    1. National Statistical Office Thailand

    (2017) 2010 population and housing census

    Accessed July 11, 2017.

    http://popcensus.nso.go.th/en/
29. 1. Pigott DM
    2. Bhatt S
    3. Golding N
    4. Duda KA
    5. Battle KE
    6. Brady OJ
    7. Messina JP
    8. Balard Y
    9. Bastien P
    10. Pratlong F
    11. Brownstein JS
    12. Freifeld CC
    13. Mekaru SR
    14. Gething PW
    15. George DB
    16. Myers MF
    17. Reithinger R
    18. Hay SI

    (2014) Global distribution maps of the leishmaniases

    eLife 3:e02851.

    https://doi.org/10.7554/eLife.02851

    • Google Scholar
30. 1. Rodriguez-Barraquer I
    2. Cordeiro MT
    3. Braga C
    4. de Souza WV
    5. Marques ET
    6. Cummings DA

    (2011) From re-emergence to hyperendemicity: the natural history of the dengue epidemic in Brazil

    PLOS Neglected Tropical Diseases 5:e935.

    https://doi.org/10.1371/journal.pntd.0000935

    • PubMed
    • Google Scholar
31. 1. Rodríguez-Barraquer I
    2. Buathong R
    3. Iamsirithaworn S
    4. Nisalak A
    5. Lessler J
    6. Jarman RG
    7. Gibbons RV
    8. Cummings DA

    (2014) Revisiting rayong: shifting seroprofiles of dengue in Thailand and their implications for transmission and control

    American Journal of Epidemiology 179:353–360.

    https://doi.org/10.1093/aje/kwt256

    • PubMed
    • Google Scholar
32. Software

    1. Rodriguez-Barraquer I

    (2019) dengue_foi, version 83e039a

    Github.

    https://github.com/isabelrodbar/dengue_foi
33. Website

    1. SAGE

    (2016) Summary of the April 2016 meeting of the strategic advisory group of experts on immunization (SAGE)

    Accessed October 15, 2016.

    http://www.who.int/immunization/sage/meetings/2016/april/SAGE_April_2016_Meeting_Web_summary.pdf?ua=1
34. 1. Salje H
    2. Cauchemez S
    3. Alera MT
    4. Rodriguez-Barraquer I
    5. Thaisomboonsuk B
    6. Srikiatkhachorn A
    7. Lago CB
    8. Villa D
    9. Klungthong C
    10. Tac-An IA
    11. Fernandez S
    12. Velasco JM
    13. Roque VG
    14. Nisalak A
    15. Macareo LR
    16. Levy JW
    17. Cummings D
    18. Yoon IK

    (2016) Reconstruction of 60 years of Chikungunya epidemiology in the philippines demonstrates episodic and focal transmission

    Journal of Infectious Diseases 213:604–610.

    https://doi.org/10.1093/infdis/jiv470

    • PubMed
    • Google Scholar
35. 1. Schenzle D
    2. Dietz K
    3. Frösner GG

    (1979) Antibody against hepatitis A in seven European countries. II. statistical analysis of cross-sectional surveys

    American Journal of Epidemiology 110:70–76.

    https://doi.org/10.1093/oxfordjournals.aje.a112790

    • PubMed
    • Google Scholar
36. 1. Siraj AS
    2. Rodriguez-Barraquer I
    3. Barker CM
    4. Tejedor-Garavito N
    5. Harding D
    6. Lorton C
    7. Lukacevic D
    8. Oates G
    9. Espana G
    10. Kraemer MUG
    11. Manore C
    12. Johansson MA
    13. Tatem AJ
    14. Reiner RC
    15. Perkins TA

    (2018) Spatiotemporal incidence of zika and associated environmental drivers for the 2015-2016 epidemic in Colombia

    Scientific Data 5:180073.

    https://doi.org/10.1038/sdata.2018.73

    • PubMed
    • Google Scholar
37. 1. Stan development Team

    (2019) Stan: A C++ Library for Probability and Sampling

    Stan: A C++ Library for Probability and Sampling, 2.5.0, http://mc-stan.org/.

    http://mc-stan.org/

    • Google Scholar
38. 1. Teixeira MG
    2. Costa MdaCN
    3. Barreto F
    4. Barreto ML

    (2009) Dengue: twenty-five years since reemergence in Brazil

    Cadernos de Saúde Pública 25:S7–S18.

    https://doi.org/10.1590/S0102-311X2009001300002

    • Google Scholar
39. 1. Thai KTD
    2. Nishiura H
    3. Hoang PL
    4. Tran NTT
    5. Phan GT
    6. Le HQ
    7. Tran BQ
    8. Nguyen NV
    9. de Vries PJ

    (2011) Age-Specificity of clinical dengue during primary and secondary infections

    PLOS Neglected Tropical Diseases 5:e1180.

    https://doi.org/10.1371/journal.pntd.0001180

    • Google Scholar
40. 1. The R Foundation

    (2019) The R Project for Statistical Computing

    The R Project for Statistical Computing, R version 3.6.0, http://www.R-project.org/.

    http://www.R-project.org/

    • Google Scholar
41. 1. Vongpunsawad S
    2. Intharasongkroh D
    3. Thongmee T
    4. Poovorawan Y

    (2017) Seroprevalence of antibodies to dengue and Chikungunya viruses in Thailand

    PLOS ONE 12:e0180560.

    https://doi.org/10.1371/journal.pone.0180560

    • PubMed
    • Google Scholar
42. Report

    1. WHO

    (2012) Global Strategy for Dengue Prevention and Control, 2012–2020

    World Health Organization.

    https://www.who.int/denguecontrol/9789241504034/en/

    • Google Scholar
43. 1. WHO

    (2018) Dengue vaccine: who position paper, September 2018 – Recommendations

    Vaccine.

    https://doi.org/10.1016/j.vaccine.2018.09.063

    • Google Scholar
44. 1. Wilson S
    2. Booth M
    3. Jones FM
    4. Mwatha JK
    5. Kimani G
    6. Kariuki HC
    7. Vennervald BJ
    8. Ouma JH
    9. Muchiri E
    10. Dunne DW

    (2007) Age-adjusted Plasmodium falciparum antibody levels in school-aged children are a stable marker of microgeographical variations in exposure to Plasmodium infection

    BMC Infectious Diseases 7:67.

    https://doi.org/10.1186/1471-2334-7-67

    • PubMed
    • Google Scholar

Author details

1. Isabel Rodriguez-Barraquer

   Department of Medicine, University of California, San Francisco, San Francisco, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing

   For correspondence

   Isabel.Rodriguez@ucsf.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-6784-1021

2. Henrik Salje

   1. Mathematical Modelling of Infectious Diseases Unit, Institut Pasteur, Paris, France
   2. CNRS, URA3012, Paris, France
   3. Center of Bioinformatics, Biostatistics and Integrative Biology, Institut Pasteur, Paris, France
   4. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Johns Hopkins University, Baltimore, United States
   5. Department of Biology, University of Florida, Gainesville, United States

   Contribution

   Formal analysis, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-3626-4254

3. Derek A Cummings

   1. Department of Biology, University of Florida, Gainesville, United States
   2. Emerging Pathogens Institute, University of Florida, Gainesville, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

• 2,720

  views

• 404

  downloads

• 30

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.