Sensitivity of the Dengue Surveillance System in Brazil for Detecting Hospitalized Cases

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

We evaluated the sensitivity of the dengue surveillance system in detecting hospitalized cases in ten capital cities in Brazil from 2008 to 2013 using a probabilistic record linkage of two independent information systems hospitalization (SIH-SUS) adopted as the gold standard and surveillance (SINAN). Sensitivity was defined as the proportion of cases reported to the surveillance system amid the suspected hospitalized cases registered in SIH-SUS. Of the 48,174 hospitalizations registered in SIH-SUS, 24,469 (50.7%) were reported and registered in SINAN, indicating an overall sensitivity of 50.8% (95%CI 50.3–51.2). The observed sensitivity for each of the municipalities included in the study ranged from 22.0% to 99.1%. The combination of the two data sources identified 71,161 hospitalizations, an increase of 97.0% over SINAN itself. Our results allowed establishing the proportion of underreported dengue hospitalizations in the public health system in Brazil, highlighting the use of probabilistic record linkage as a valuable tool for evaluating surveillance systems.

Author Summary

This manuscript address essential issues regarding the current and future challenges of a dengue surveillance system. Dengue fever is one of the major public health threats in a large area of the world, including Brazil which represents around 70% of the cases in the Americas. The dengue surveillance system in Brazil was established in 1986, but evaluations of this surveillance system were rarely conducted. The need for accurate data is of paramount importance due to not only the increase of severe cases in the past decade, but also to monitor the impact of a vaccine as soon as it occurs. The evaluation of the sensitivity of surveillance system is a challenge, due to the lack of a gold standard. In our study, we took advantage of two very well structured and independent information systems with nationwide coverage. The hospitalization information system was defined as the gold standard for hospitalized cases in the public health system and compared to the notifiable diseases information system using a probabilistic record linkage. Therefore, we were able to evaluate the sensitivity of the dengue surveillance system in detecting hospitalized cases in ten capital cities from 2008 to 2013 in Brazil.

Related collections

Most cited references 26

Record: found
Abstract: found
Article: found

Is Open Access

Economic Impact of Dengue Illness in the Americas

Donald S. Shepard, Laurent Coudeville, Yara A. Halasa … (2011)

The growing burden of dengue in endemic countries and outbreaks in previously unaffected countries stress the need to assess the economic impact of this disease. This paper synthesizes existing studies to calculate the economic burden of dengue illness in the Americas from a societal perspective. Major data sources include national case reporting data from 2000 to 2007, prospective cost of illness studies, and analyses quantifying underreporting in national routine surveillance systems. Dengue illness in the Americas was estimated to cost $2.1 billion per year on average (in 2010 US dollars), with a range of $1–4 billion in sensitivity analyses and substantial year to year variation. The results highlight the substantial economic burden from dengue in the Americas. The burden for dengue exceeds that from other viral illnesses, such as human papillomavirus (HPV) or rotavirus. Because this study does not include some components (e.g., vector control), it may still underestimate total economic consequences of dengue.

0 comments Cited 181 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: not found

The Economic Burden of Dengue

Duane J. Gubler (2012)

Historically, dengue has been considered an unimportant public health problem because mortality rates were low and epidemics occurred only infrequently. In the years after World War II, great progress was made in controlling infectious diseases of all kinds, especially vector-borne diseases, and the war on infectious disease was declared won in the late 1960s.1 Coincident with this success in disease control was the economic recovery in some countries of Southeast Asia, and urban growth increased as people from rural areas migrated to the cities to find work. The lack of planning, inadequate housing, water, sewage, and waste management in these cities created ideal conditions for dengue viruses and their mosquito vector Aedes aegypti, both of which had been spread widely in the region during World War II, to thrive.2 The result was the emergence of epidemic dengue hemorrhagic fever. During the 1950s and 1960s, the disease was limited to a few countries in Southeast Asia, but as economic growth expanded, the cities and epidemic dengue also grew, because little effort was made to control the mosquito vector. In the 1970s and accelerating in the last two decades of the 20th century, epidemic dengue expanded regionally and globally. Epidemics increased in both frequency and magnitude, and the viruses became hyperendemic (cocirculation of multiple virus serotypes) in most major cities of the tropics. Despite this alarming emergence of a severe and fatal form of epidemic dengue, the disease was still considered a minor public health problem by policymakers. In 2012, dengue is the most important vector-borne viral disease of humans and likely more important than malaria globally in terms of morbidity and economic impact. The latest studies estimate 3.6 billion people living in areas of risk, over 230 million infections, millions of cases of dengue fever, over 2 million cases of the severe disease, and 21,000 deaths.3 In addition to the public health and economic costs, there is a major social impact in those countries where large epidemics occur, often disrupting primary care for hospitalized patients.4 Given the dramatic urban growth and lack of adequate surveillance for dengue in tropical developing countries in the past 50 years, it is likely that even these figures underestimate the true disease burden of dengue. Thus, ignored for many years, only recently has the potential magnitude of the dengue problem been acknowledged by policymakers and funding agencies. This acknowledgment has resulted in excellent progress in understanding dengue virus biology and development of dengue vaccines and antiviral drugs but not economic impact of the disease. As the pipeline of dengue vaccine candidates, a number of which are already in clinical trials, has grown, it has become apparent that vaccine access will require more and better studies on the actual cost of dengue to endemic communities. In recent years, many studies have been conducted in different Asian and American countries in attempts to measure the economic impact of dengue on a community. Although most have contributed valuable information on the subject, none has provided comprehensive data on the actual cost of dengue disease. The work by Halasa and others5 provides one of the best studies to date on that subject. Working in collaboration with the Puerto Rico Department of Health and the US Centers for Disease Control and Prevention Dengue Branch in Puerto Rico, the investigators conducted a comprehensive review of 100 laboratory-confirmed dengue patients who experienced the disease between July of 2008 and March of 2010. The study included both adults and children and hospitalized and ambulatory care patients. Records were reviewed for each patient, who was then subjected to a comprehensive interview about their illness and how it impacted their activities. The work by Halasa and others5 measured the cost, both direct and indirect, of dengue in six categories of cases and went to considerable effort to validate their data. The results, perhaps not surprisingly, show the economic burden of dengue in Puerto Rico to be very high. This study is the first study to show the societal distribution of the economic cost of dengue, with the individual household bearing the largest burden (48%) compared with only 24% by the government and 22% by insurance.5 The total annual cost of dengue between 2002 and 2010 was $46.45 million ($418 million during the 9-year period).5 Clearly, there is a need to conduct more comprehensive studies of this kind in other dengue-endemic countries of the world. These kinds of data are needed to allow policymakers and public health officials to make informed decisions on the cost efficacy of dengue control programs. The private sector also needs comprehensive economic analyses to guide decision-making in vaccines and antiviral drug development. Finally, international funding agencies require these kinds of data to establish priorities in public health funding. Studies of this kind will always have limitations, and this study was no exception. However, the work by Halasa and others5 clearly identified the most important limitations, such as recall bias and not being able to estimate the cost of tourism and deaths associated with dengue. The latter two limitations, however, would add to the overall cost of dengue, driving the cost even higher. The fact that the study was supported by a private vaccine manufacturer might raise questions about conflict of interest by some people. In my mind, however, this possibility is not an issue because of the detailed methodology used and the reputation of the Brandeis group as the leader in research on the economic impact of dengue.

0 comments Cited 105 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: found

Is Open Access

Epidemiological Trends of Dengue Disease in Brazil (2000–2010): A Systematic Literature Search and Analysis

Maria Glória Teixeira, João Siqueira, Germano L C Ferreira … (2013)

Introduction Dengue disease is an escalating public health problem [1]. Approximately 2·5 billion people live in over 100 endemic countries, predominantly in tropical areas where dengue viruses (DENV) can be transmitted [2]. DENV are arboviruses that are transmitted to humans by infected Aedes aegypti (Linnaeus) mosquitoes – the primary vector. Infection with any one of four DENV serotypes (DENV-1, -2, -3, or -4) can produce a spectrum of illness ranging from a mild, non-specific febrile syndrome, to classic dengue fever (DF), or severe disease forms, such as dengue haemorrhagic fever (DHF) and dengue shock syndrome (DSS), that can be fatal. The World Health Organization (WHO) estimates that >50 million dengue infections and >20,000 dengue-related deaths occur annually [1], [3], [4]. A recent disease distribution model has estimated there to be 390 (95% credible interval 284–528) million dengue infections per year, of which 96 million are apparent (i.e., cases manifest any level of clinical or sub-clinical severity) [3]. During 2001–2007, >4 million cases were notified in the Americas, and during 1995–2002, >75% of these cases were reported from Brazil [5], [6]. Ae. aegypti was eradicated from Brazil as a result of a Pan American Health Organization (PAHO) programme to control the spread of yellow fever. Additionally, DENV transmission was also suppressed in the Americas during the eradication programme. South American countries became re-infested with Ae. aegypti after the programme was discontinued and this, combined with the co-circulation of multiple DENV serotypes, led to the spread of dengue disease across the continent [5], [7]–[9]. In 1982, there was a dengue outbreak in a small city in the northern region of Brazil (Boa Vista/Roraima), which was quickly brought under control and the virus did not spread [10]. In 1986, the re-emergence of DENV-1 in Rio de Janeiro state [11] resulted in over 60,000 reported cases in 1987 and the subsequent spread of DENV increased national public health concerns [12]–[14]. Since the late 1980's the incidence of dengue disease continued to increase; 204,000 cases were reported nationally in 1999 [15], [16]. By 2000, DENV transmission was reported in 22/27 Brazilian states, and the mosquito vector was present in all states [17]. Much of Brazil is affected by a tropical wet and dry climate with high temperatures, high humidity and seasonal variations in rainfall; climate patterns that can provide appropriate conditions for breeding and survival of the Ae. aegypti mosquito. The country is divided into five regions (North, Northeast, Central-West, Southeast, and South) comprising 26 states and the federal district containing the capital city, Brasília. In 2000 there were nearly 170 million inhabitants of Brazil, increasing to more than 190 million in 2010 [18], the majority of whom live in the large cities of the Southeast and Northeast regions [19]. The National System for Surveillance and Control of Diseases (SNVS) of Brazil, operates as part of the national health system (Sistema Único de Saúde, or SUS). All reported cases from public health services or private health providers are included in the notification database (Sistema de Informacoes de Agravos de Notificacao [SINAN]), which is openly accessible via the internet [20]. Until 2011, the SNVS adopted the case definitions outlined in WHO guidelines [21], [22]. In 1997, the WHO categorized symptomatic dengue disease as: undifferentiated fever, DF and, DHF [21]. DHF was further classified into four severity grades, with grades III and IV being defined as DSS. However, difficulties in applying the criteria for DHF [23], led the WHO to suggest a new classification based on levels of severity: non-severe dengue disease with or without warning signs, and severe dengue disease [22]. During 2000–2011, both surveillance and hospitalization reporting systems in Brazil used DF and DHF; the surveillance system used an additional classification designated ‘DF with complications’ (DFC) [24]. Importantly, the articles included in this literature analysis that were based on secondary data used these surveillance sources. Our objectives of this literature search and analysis were to describe the epidemiology of dengue disease (national and regional incidence [by age and sex], seroprevalence and serotype distribution and other relevant epidemiological data) in Brazil during 2000–2011, and to identify gaps in epidemiological knowledge requiring further research. Methods A literature review group, including authors of this contribution, developed a literature survey and analysis protocol based on the preferred reporting items of systematic reviews and meta-analyses (PRISMA) guidelines [25]. Our protocol prescribed well-defined methods to search, identify, and select relevant research, and set predetermined inclusion criteria. The protocol was registered on PROSPERO, an international database of prospectively registered systematic reviews in health and social care managed by the Centre for Reviews and Dissemination, University of York (CRD42011001826: http://www.crd.york.ac.uk/prospero/display_record.asp?ID=CRD42011001826; protocol: http://www.crd.york.ac.uk/PROSPEROFILES/1826_PROTOCOL_20130401.pdf) on 9 December 2011. Search strategy and selection criteria Between 31 July 2011 and 4 August 2011, we searched databases of published literature (Table 1) for epidemiological studies of dengue disease in Brazil. Search strategies for each database were described with reference to the expanded Medical Subject Headings (MeSH) thesaurus, encompassing the terms ‘dengue’, ‘epidemiology’, and ‘Brazil’. Google and Yahoo searches (limited to the first 50 results) were used to identify national and international reports and guidelines, congress abstracts, and grey literature (e.g., Ministry of Health data, lay publications). 10.1371/journal.pntd.0002520.t001 Table 1 Databases searched for citations relating to dengue disease epidemiology in Brazil. Database Website United States National Library of Medicine and the National Institutes of Health Medical Database http://www.ncbi.nlm.nih.gov/pubmed/ Excerpta Medica Database (EMBASE) MedLine Scientific Electronic Library Online (SciELO) – a consolidated electronic publication project that makes available the full text articles from more than 290 scientific journals from Brazil, Chile, Cuba, Spain, Venezuela and other Latin American countries http://www.scielo.org/php/index.php?lang=en Virtual Health Library (VHL), an initiative by Brazil-based BIREME (the Latin American and Caribbean Center on Health Sciences Information) that facilitates searches of the Latin American and Caribbean Health Sciences Database (LILACS) and the PAHO Headquarters Library database and other regional health resources http://regional.bvsalud.org/php/index.php?lang=en WHO Library database (WHOLIS) http://dosei.who.int/uhtbin/cgisirsi/3foptRgmQT/7440030/38/1/X/BLASTOFF Brazilian Ministry of Education: Theses Bank (CAPES) http://capesdw.capes.gov.br/capesdw/ To reduce selection bias, peer-reviewed contributions in English, Portuguese, or Spanish published between 1 January 2000 and 4 August 2011 were included; no limits by sex, age, ethnicity of study participants, or by study type were imposed. Single-case reports and articles only reporting data prior to 1 January 2000 were excluded. Unpublished reports were included if they were identified in one of the sources listed above. Data from grey materials supplemented that from peer-reviewed literature. Publications not identified in the target databases by the search strategy (e.g., locally published papers) and unpublished data sources meeting the inclusion criteria (e.g., theses, Ministry of Health data) were included if recommended by members of the literature review group. Editorials and data from literature reviews of previously published peer-reviewed studies were excluded. Duplicates and articles not satisfying the inclusion criteria were removed following review of the titles and abstracts. A further selection was made based on review of the full text from the first selection of references. Included publications were summarised using a data extraction instrument developed as a series of spreadsheets. Due to the expected heterogeneity of eligible studies in terms of selection, and number and classification of cases, a meta-analysis was not conducted. For the purposes of the analysis we defined national epidemics as those years with an incidence/100,000 above the 75th percentile for the period. A trend analysis was conducted on the national incidence and case number data. Results and Discussion We identified 714 relevant citations, 51 of which met the inclusion criteria and were entered into the data extraction instrument (Figure 1; Table S1). 10.1371/journal.pntd.0002520.g001 Figure 1 Result of literature search and evaluation of identified studies according to the preferred reporting items of systematic reviews and meta-analyses (PRISMA). All references identified in the on-line database searches were assigned a unique identification number. Following the removal of duplicates and articles that did not satisfy the inclusion criteria from review of the titles and abstracts, the full papers of the first selection of references were retrieved either electronically or in paper form. A further selection was made based on review of the full text of the articles. National epidemiology During the period 2000–2010, the incidence of dengue disease in Brazil varied substantially, reaching a peak in 2010 of >1 million cases (538/100,000 inhabitants) and the lowest value was approximately 72,000 cases in 2004 (63.2/100,000 inhabitants) (Table 2, Figure 2A–C, Table S2) [6], [15], [16], [26]–[31]. Despite the yearly variations and cyclical epidemics, trend analysis of the incidence of dengue in Brazil in the period 2000–2010 showed an overall increase in incidence over time that was not statistically significant (β = 12·9/cases per 100,000, p = 0·49). Analysis of the number of cases of dengue disease over the review period shows a growth trend that was not statistically significant (β = 47·984 cases/year, p = 0·25). Nevertheless, the trend analysis suggests a worsening of the problem over time. 10.1371/journal.pntd.0002520.g002 Figure 2 Trends in epidemiology of dengue disease Brazil, 2000–2010. (A) Reported number and average incidence per 100,000 population of probable* cases of dengue disease. (B) Reported number of cases of dengue fever with complications (DFC) and dengue haemorrhagic fever (DHF). (C) Reported number of dengue disease related hospitalizations (DFC+DHF) and deaths due to DFC and DHF. (D) Number of Ae. aegypti-infested municipalities. (E). Average incidence of dengue disease per 100,000 population, by region. The epidemiology of dengue disease in Brazil during the review period suggests that incidence and disease severity increased over the decade, although the situation is complicated by national epidemics in 2002, 2008 and 2010. The incidence of dengue disease over the review period reflects the wide distribution of Ae. aegypti nationally. In most regions the dengue disease incidence followed national trends. (Adapted from Teixeira 2009 [15] and Siqueira 2010 [26]; additional data supplied by Teixeira MG and Siqueira JB, 2012). 10.1371/journal.pntd.0002520.t002 Table 2 Incidence of dengue disease in Brazil: Summary of national dengue disease incidence data and case numbers and DHF case numbers extracted from source documents. Year Parameter Range Source of data 2000 Dengue disease (n) 138,388–231,000 6, 16, 27, 31 Dengue disease (Incidence per 100,000 inhabitants) 92.3–150 15*, 28–30 DHF (n) 40–888 6, 16, 27, 29 2001 Dengue disease (n) 381,718–413,000 6, 16, 27, 31 Dengue disease (Incidence per 100,000 inhabitants) 225–254 15*, 28–30 DHF (n) 630–682 6, 16, 27, 29 2002 Dengue disease (n) 684,527–794,219 6, 16, 26, 27, 31 Dengue disease (Incidence per 100,000 inhabitants) 335.3–470 15*, 26, 28–30 DHF (n) 2608–2714 16, 26, 27, 29 2003 Dengue disease (n) 280529–342000 16, 26, 27, 31 Dengue disease (Incidence per 100,000 inhabitants) 195–200 15*, 29 DHF (n) 650–913 16, 26, 27, 29 2004 Dengue disease (n) 71,847–113,000 16, 26, 27, 31 Dengue disease (Incidence per 100,000 inhabitants) 75 15* DHF (n) 81–159 16, 26, 27 2005 Dengue disease (n) 134,298–204,000 16, 26, 27, 31 Dengue disease (Incidence per 100,000 inhabitants) 150 15* DHF (n) 463–1395 16, 26, 27 2006 Dengue disease (n) 252725–347000 16, 26, 27 Dengue disease (Incidence per 100,000 inhabitants) 200 15* DHF (n) 642–910 16, 26, 27 2007 Dengue disease (n) 501666–560000 16, 26, 27 Dengue disease (Incidence per 100,000 inhabitants) 300 15* DHF (n) 1541–1907 16, 26, 27 2008 Dengue disease (n) 637,663–806,036 16, 26, 27 Dengue disease (Incidence per 100,000 inhabitants) 120–336.3 15*, 26 DHF (n) 647–4502 16, 26, 27 2009 Dengue disease (n) 407,000–411,500 16, 26 Dengue disease (Incidence per 100,000 population) 205,5–214,9 15* DHF (n) 2679 26 2010 Dengue disease (n) 1,027,100 26 Dengue disease (Incidence per 100,000 inhabitants) 538.4 26 DHF (n) 3807 26 Empty cells indicate data not reported in source documents. * Dengue disease incidence data from Teixeira 2009 [15] were estimated from Figure 2. Dengue fever incidence rates (per 100,000 inhabitants) according to geographic regions and year of occurrence. Brazil, 1986–2007. There were three national epidemics (years with incidence above the 75th percentile for the period [279.95]) in 2002, 2008 and 2010. In 2002 there were 684,527 to 794,219 probable cases of DF, in 2008, 637,663 to 806,036 cases [16], [26], [27], and in 2010 there were over 1 million reported cases (Table 2; Figure 2A) [26]. A trough occurred in 2004 (71,847 to 113,000 cases) [16], [26], [27], [31], representing 18,000) is striking when compared with data from the previous decade: during the 1990s 94,000 hospitalizations in 2010 (Figure 2C) [26]. The incidence of dengue-related hospitalization was 31·6/100,000 population during the 2002 national epidemic, approximately 40·8/100,000 during the 2008 national epidemic, and 49·7/100,000 during the 2010 national epidemic [26]. These increases in hospitalization rates during epidemic years might suggest an increase in the severity of dengue disease in Brazil, although an increased awareness during epidemics and a lower threshold for hospitalization might also account for these increases. The number of dengue-related deaths followed the same patterns as the other epidemiological indices of dengue disease. In 2010, of 13,909 cases classified as DFC and 3807 classified as DHF, there were 370 and 308 fatal cases, respectively. The overall number of DHF- or DFC-related deaths was 678 compared with only 19 in 2004 (Figure 2C) [26]. A seasonal pattern of dengue disease was observed in those studies with available seasonal case distribution data. The highest incidences occurred during January–June [34]–[38], corresponding to the period of highest rainfall and humidity, providing suitable conditions for Ae. aegypti breeding and survival. The study by Goncalves Neto et al. [35] showed that 83·3% of dengue disease cases occurred during the rainy season and demonstrated a positive Pearson correlation with the amount of rainfall (r = 0·84) and relative humidity (r = 0·76) and a negative correlation with temperature (r = −0·78). Regional epidemiology We found published regional data for part of the study period from four of the five Brazilian regions [6], [28], [34], [35], [39]–[51]. No published data were recovered for the North region. The available data show that incidence rates varied greatly throughout the country (data not shown; Table S3). In a study of 146 Brazilian cities in October 2006, incidence rates (per 100,000 population) in the 61 cities that reported >500 dengue disease cases ranged between 24·70 (Sao Paulo) and 6222·71 (Campo Grande) [52]. By the end of 2006, 25 of the 27 states had reported local dengue epidemics [15]. The geographic distribution of the Ae. aegypti vector has widened over the 11-year review period, involving an increasing number of municipalities (Figure 2D) and this has resulted in a broader regional distribution of dengue disease. In most regions the dengue disease incidence followed national trends (Figure 2E). In the early years of the survey, the Southeast and Northeast regions were most affected by DENV infections, whereas from 2009 more cases were reported from studies within the Central-West region. Incidence rates reported in the South region were consistently lower than in other regions. The distribution of reported cases of dengue disease during the 2010 national epidemic was different from that in the 2002 and 2008 national epidemics with high attack rates observed over larger areas of Brazil [26]. These regional variations in dengue disease incidence are unsurprising given the geographically diverse nature of Brazil with its large variations in climate and population density. Demographic patterns of dengue disease in Brazil A change in the age distribution of dengue disease over the survey period was evident from the available data. Young adults were most affected by DF and DHF during 2000–2007 and 2000–2005, respectively (i.e., DHF was coincident with the highest incidence of DF). However, in 2006 the incidence of DHF among children aged 53% of DHF cases occurred in children 40 years; all cases (n = 18) are virologically confirmed and from one hospital. Slightly more women than men are affected by dengue disease throughout Brazil [36], which is similar to the sex distribution of reported cases in other Latin American countries [9]. During 2001–2010 the male∶female ratio of reported cases ranged from 0·75–0·82 [9], [26]. Regional data were more variable. In 2000 the ratio was 1·09 in the city of São Luís [35], and 0·5 in the City of Santos in 2010 [54]. Women with dengue disease were slightly older than men (mean age 33·7 years versus 30·2 years, respectively; p = 0.019) [37]. DENV distribution Seroprevalence Seroprevalence data provide further information to illustrate epidemiological trends (see Socio-demographic factors below). Population seroprevalence estimates varied throughout Brazil during the decade analysed. In individuals aged 18–65 years, the highest seroprevalence rates were reported in the cities of Mossoró and Caruaru (97·8% and 94·5%, respectively) with lower seroprevalence reported in Rio Branco (69·2%) and Macapá (48·4%) [55]. In serological surveys of volunteers without DF symptoms in Goiânia, seroprevalence was 29·5% in 2001 and 37·3% in 2002 [56]. In Recife, a large urban centre, during 2004–2006, 354 (53·8%) of 658 patients with suspected DENV infections had antibodies to DENV, of which 175 (49·4%) were characterized as primary infections and 179 (50·6%) as secondary infections [36]. In 2002, the seroprevalence in Recife was 76·3% (45 cases) [44]. Few age-specific seroprevalence data were reported in studies included in our analysis. Seroprevalence data also reveal that dengue disease is under-reported. Current passive surveillance systems do not report on mildly symptomatic and non-specific febrile cases and do not represent the true rate of infection and transmission. Based on the findings of a seroepidemiological study in Recife conducted between August and September 2006, Rodriguez-Barraquer et al. calculated that 100/100,000 inhabitants from 66·10% in 2001–2002 (before PNCD implementation) to 48·97% in 2003–2006 (after implementation) [77]. Strengths and limitations of this survey and analysis Despite some gaps, our literature survey and analysis provides a comprehensive overview of the evolving epidemiology of dengue disease in Brazil over the period 2000–2011. This study has several important strengths. Our survey was thorough; we screened >700 articles to identify relevant publications and we developed a comprehensive data extraction instrument to facilitate the capture of all relevant data. Nevertheless, the lack of comprehensive and continuous data for the survey period limits our ability to make comparisons and draw firm conclusions over the years, across regions, and among different ages. For example, age-stratified data were not reported systematically and age range boundaries differed by study. Therefore, although we can suggest trends in age distribution, it is not possible to directly compare data from the selected publications. The inclusion of publications in three languages reduced selection bias in our literature review and analysis. However, despite the inclusion of PhD dissertations and theses there is a bias towards published articles. An assessment of quality of evidence was not carried out and potential weaknesses of some studies such as inadequately described case selection, small sample sizes, and unspecified statistical methods were not reasons for exclusion. Consequently, any limitations of the original studies are carried forward into our review. Many of the studies relied on data reported by passive surveillance systems, which can vary between regions and over time [33] and may misrepresent the number of cases due to changes in reporting behaviour and misclassifications. Avenues for future research Our literature survey and analysis identified several knowledge gaps, which indicate potential avenues for future study. In particular, there are gaps relating to the regional incidence of dengue disease in Brazil, national and regional age-related data, and national and regional serotype information. Further epidemiological studies may help to clarify and define regional differences. The large increase in the number of DHF cases and the shift in age distribution of DHF towards younger age groups that occurred during the 2007–2008 national epidemic warrant explanation. One possibility is that the change in circulating DENV serotypes over time may have affected the pattern of dengue disease epidemiology in Brazil [78]. Age-stratified seroprevalence studies will improve assessment of the level of transmission and inapparent infection, as well as providing information relating to the age shift. Further studies into the risk factors for dengue disease and its severity are also important. For example, in Southeast Asia, DENV infection has been more widespread for a longer period of time than in the Americas, creating a large group of individuals likely to experience a second or third infection [32]. These secondary infections carry an increased risk of severe dengue disease. The data in this review do not address the Southeast Asian experience and further examination as to whether this phenomenon is replicated in Brazil is required. In addition, few studies in the review specifically measured the effects of urbanization in Brazil, with effects only inferred from studies of other socio-demographic factors. The diversity of ethnic backgrounds within the population suggests that further genetic studies are warranted to determine whether ethnicity affects the clinical expression of dengue disease and the risk for severe outcomes. Studies are also required to clearly define associations with other diseases if comorbidity screening is to be used to identify patients at a greater risk of developing DHF. We acknowledge that there are gaps in our epidemiological knowledge of dengue disease in Brazil, due, in part (as in many other countries) to the inherent weaknesses of passive surveillance systems. The majority of infections are clinically non-specific consequently dengue disease is often mis-diagnosed during inter-epidemic periods [8]. The findings presented here are in broad agreement with those of Honório et al. [79], who found only 23·3% of infections were symptomatic, and with Lima et al. [80], who showed that the number of cases reported for the Southeast region of Brazil under-represented the number of infected individuals. This was also found in studies conducted in other countries [81]. Only when an epidemic occurs is the full spectrum of the disease recognised. Consequently, the disease is likely to be under-reported during inter-epidemic periods but over-reported during epidemics [82]. Overall, we believe the national surveillance data under-estimate the true incidence of DENV infections. However, extensive representative serological surveys are required to estimate the true rate of infection and transmission and, thus, despite its drawbacks, passive reporting is important for the identification of disease trends over time. Conclusions Our review and analysis of the epidemiology of dengue disease in Brazil during the past decade suggests an overall increase in the distribution and severity of dengue disease. During the last decade (2000–2010), a total number of 8,440,253 cases were reported (the highest figure in the history of dengue disease in this region) with the highest number of severe cases (221,043; 2.6%) and fatal cases (3058; 0.036% of the total reported cases and 1.38% of the severe cases) [83]. The 1588 cases of severe dengue disease and 163 deaths reported as of epidemiological week 8 in 2011, represent 67% and 73%, respectively, of the total cases registered in the Americas [84]. The co-circulation of multiple DENV serotypes and high dengue disease endemicity may be responsible for the increased occurrence of severe forms of dengue disease and increases in the numbers of dengue disease-related hospitalizations. In addition, the increase in the number of severe cases of dengue disease and a shift in age group predominance of severe forms observed during 2007/08 confirm that dengue disease must remain a public health priority in Brazil. Even though the studies included in this literature review have improved our understanding of the epidemiology of dengue disease in Brazil, further studies are required to clarify the epidemiological pattern and to understand regional epidemiological differences, the diversity of genotypes of circulating serotypes and the extent of herd immunity by age group. Our review has highlighted the main epidemiological characteristics of dengue in Brazil in the first decade of this century and revealed that the epidemiological pattern of dengue disease in Brazil is complex. The changes observed are likely to have been the result of multiple factors, which still require elucidation. Supporting Information Checklist S1 PRISMA 2009 checklist. (PDF) Click here for additional data file. Table S1 Citations used in the literature analysis. (PDF) Click here for additional data file. Table S2 Incidence of dengue disease in Brazil: national data. (PDF) Click here for additional data file. Table S3 Incidence of dengue disease in Brazil: regional data. (PDF) Click here for additional data file.

0 comments Cited 100 times – based on 0 reviews      Review now

Bookmark

All references

Author and article information

Contributors

Duane J. Gubler: Role: Editor

Journal

Journal ID (nlm-ta): PLoS Negl Trop Dis

Journal ID (iso-abbrev): PLoS Negl Trop Dis

Journal ID (publisher-id): plos

Journal ID (pmc): plosntds

Title: PLoS Neglected Tropical Diseases

Publisher: Public Library of Science (San Francisco, CA USA )

ISSN (Print): 1935-2727

ISSN (Electronic): 1935-2735

Publication date (Electronic): 18 May 2016

Publication date Collection: May 2016

Volume: 10

Issue: 5

Electronic Location Identifier: e0004705

Affiliations

[1 ]National Dengue Control Program, Ministry of Health, Brasilia, Brazil

[2 ]Institute of Tropical Pathology and Public Health, Federal University of Goiás, Goiania, Brazil

Duke-NUS GMS, SINGAPORE

Author notes

The authors have declared that no competing interests exist.

Conceived and designed the experiments: GEC PLL MdPC ACRS JBS. Performed the experiments: GEC PLL MdPC ACRS JBS. Analyzed the data: GEC PLL MdPC ACRS JBS. Wrote the paper: GEC PLL MdPC ACRS JBS.

* E-mail: giovanini.coelho@ 123456gmail.com

Article

Publisher ID: PNTD-D-15-01682

DOI: 10.1371/journal.pntd.0004705

PMC ID: 4871568

PubMed ID: 27192405

SO-VID: a772b408-188c-4891-944a-96cb7fc25844

License:

This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

History

Date received : 27 September 2015

Date accepted : 21 April 2016

Page count

Figures: 2, Tables: 3, Pages: 12

Funding

The authors received no specific funding for this work.

Custom metadata

Data Availability All relevant data are within the paper.

ScienceOpen disciplines: Infectious disease & Microbiology

Data availability:

ScienceOpen disciplines: Infectious disease & Microbiology

Comments

Comment on this article

scite_

Cited by 12

See all cited by

Most referenced authors 638

See all reference authors

Sensitivity of the Dengue Surveillance System in Brazil for Detecting Hospitalized Cases

Read this article at

Abstract

Author Summary

Related collections

SciELO Brazil

Most cited references 26

Economic Impact of Dengue Illness in the Americas

The Economic Burden of Dengue

Epidemiological Trends of Dengue Disease in Brazil (2000–2010): A Systematic Literature Search and Analysis

Author and article information

Contributors

Journal

Affiliations

Author notes

Article

History

Page count

Funding

Categories

Custom metadata

Comments

Comment on this article

Similar content 429

Cited by 12

Most referenced authors 638