Incidence and spatial distribution of cases of dengue, from 2010 to 2019: an ecological study

ABSTRACT BACKGROUND: Dengue is an arbovirus that has caused serious problem in Brazil, putting the public health system under severe stress. Understanding its incidence and spatial distribution is essential for disease control and prevention. OBJECTIVE: To perform an analysis on dengue incidence and spatial distribution in a medium-sized, cool-climate and high-altitude city. DESIGN AND SETTING: Ecological study carried out in a public institution in the city of Garanhuns, Pernambuco, Brazil. METHODS: Secondary data provided by specific agencies in each area were used for spatial analysis and elaboration of kernel maps, incidence calculations, correlations and percentages of dengue occurrence. The Geocentric Reference System for the Americas (Sistema de Referência Geocêntrico para as Américas, SIRGAS), 2000, was the software of choice. RESULTS: The incidence rates were calculated per 100,000 inhabitants. Between 2010 and 2019, there were 6,504 cases and the incidence was 474.92. From 2010 to 2014, the incidence was 161.46 for a total of 1,069 cases. The highest incidence occurred in the period from 2015 to 2019: out of a total of 5,435 cases, the incidence was 748.65, representing an increase of 485.97%. Population density and the interaction between two climatic factors, i.e. atypical temperature above 31 °C and relative humidity above 31.4%, contributed to the peak incidence of dengue, although these variables were not statistically significant (P > 0.05). CONCLUSION: The dengue incidence levels and spatial distribution reflected virus and vector adjustment to the local climate. However, there was no correlation between climatic factors and occurrences of dengue in this city.


INTRODUCTION
In the second half of the twentieth century, dengue fever spread throughout the tropics, threatening one-third of the world's population. It caused feverish illness in around 50 to 100 million people, with records of 500,000 cases of severe illness. 1 Dengue is caused by an arbovirus that is transmitted by the mosquitos Aedes aegypti and Aedes albopictus. Its symptoms range from an acute fever to a hemorrhagic condition, and can be caused by four different virus serotypes. 2 Once an Aedes female has become infected, it can transmit the virus to humans through blood transfers for the rest of its life, which leads to greater potential for spreading the disease. 3 Aedes aegypti also transmits other high-impact arboviruses such as chikungunya. 4 Circulation of different virus serotypes has increased the number of infected patients, especially with the severe form of the disease. 5 There is no specific therapy for dengue infections and supportive treatment can save lives. 6 The first cases of dengue in Brazil were recorded in the state of Roraima, in the northwestern area of the Amazon region, in 1981. 7 Dengue has become a serious public health problem in the city of Garanhuns, state of Pernambuco, northeastern Brazil. Over the last five years, the incidence of dengue has increased by 485.97%. Epidemiological studies have confirmed that the dengue virus, which first appeared in this region in 1986, presents high intensity of transmission. This research has therefore characterized a situation of lack of knowledge about the behavior of the virus and its vector. 8 In Pernambuco, the dengue virus serotype three (DENV-3) was associated with the most severe symptoms of the epidemic, from 1995 to 2006. 9

OBJECTIVE
To describe the incidence and spatial distribution of dengue cases in a medium-sized city with a seasonal climate and high altitude, located in Brazil's northeastern region. Over the period studied (2010-2019), outbreaks and epidemics were observed, with an increased in the incidence rate of 485.97% over the last five years of that period.

Study location
This descriptive ecological study was conducted in the city of The average rainfall over the study period was 68.90 mm per annum and the average relative humidity was 40.1%. 12,13 The climate of Garanhuns is influenced by meteorological systems that cause rainfall mostly in March, June and July. 12 The 2010 census registered a population of 129,408 inhabitants in this municipality, and a total of 139,788 inhabitants was estimated for 2019. 14

Data collection
Information about dengue cases was obtained from the follow-    (Figure 2).
The year 2016 represented the peak of dengue momentum in the municipality, with an incidence rate of 2,199 dengue cases per 100,000 inhabitants. The rainfall was not significantly higher than normal; the average temperature was 31.4 °C and the average humidity was 31.4%. These acted as ideal conditions for proliferation of transmitting agents and influenced the rate of occurrence of dengue, can be seen in Table 1.
The influence of demographic density on the incidence of dengue can be seen in Table 2. This depicts the relationship between the number of dengue cases and the density of inhabitants according to neighborhood. Annual data on average temperature, accumulated precipitation and average relative humidity were individually compared with the number of dengue cases. These correlations were not statistically significant (P > 0.0) ( Table 3).

DISCUSSION
The incidence recorded over the 10 years of the survey,     Figure 2. The incidence rate was 2,199 cases per 100,000 inhabitants, from a total of 3,031 cases recorded.
High spatial distribution in certain neighborhoods was also observed. Table 2 shows that some neighborhoods with high population densities usually had more cases of dengue than did low-density neighborhoods.
Temperature, relative humidity and precipitation did not show any associations with occurrences of dengue in the municipality. These climatic correlations were not statistically significant (P > 0.05). Some studies carried out in northeastern, southeastern and northern Brazil and abroad 16,18,19,20 have shown correlations contrary to those of the present study, given that they reported associations between climatic factors and the incidence of dengue. On the other hand, some other studies 28,29 have confirmed the lack of correlation of climatic factors with the incidence of dengue.  A study carried out in Pakistan showed that the incidence of the disease was influenced by climatic factors, such that the transmission rates among mosquitoes were higher within a favorable temperature range from 28 °C to 32 °C. 21 Laboratory data on larval development of Aedes aegypti have confirmed that these temperatures favor multiplication of these larvae, consequently enabling greater production of vectors and producing increased incidence of dengue. 22 In an urban area of the city of São Paulo, Brazil, a study showed that the oviposition rates of Aedes aegypti and Aedes albopictus were influenced by the maximum and minimum temperatures. 23 Temperature influences mosquitos' life cycles and plays a crucial role in the incidence of dengue. Analyzing the effects of temperature variations in cities can lead to preventive identification of thermal comfort zones favorable to the survival of mosquito populations. 24 Knowing how environmental conditions influence the dynamics of dengue epidemics is important for responding to its epidemics and for predicting the geographical and seasonal spread of the disease. 25 Although there is no statistical association between temperature and dengue cases, it appears that dengue peaks coincide with temperature spikes. 26 This hypothesis was reinforced through a wintertime study carried out in Taiwan, which is a in subtropical region, where a low temperature of 13.8 °C resulted in the near disappearance of Aedes aegypti. 27 A study on the impact of dengue in the state of Tocantins, Brazil, revealed that climatic conditions did not influence proliferation of dengue but, rather, the conditions that would be ideal for reproduction of the vector. 28 An ecological study carried out in Araguaína, Tocantins, also did not find any correlation with climatic variables and concluded that these variables contributed to vector proliferation, but did not influence the spread of dengue. 29 Data from 13 weather stations in Delhi, India, over the period from 2006 to 2015, indicated that there was a strong association between the incidence of dengue and the temperature, humidity, wind speed, summertime, settlement density and vegetation. 30 In China, results from sensitivity analyses indicated that temperature can be an effective or facilitating barrier for vector-borne diseases and can result in complex disease control. 31 Variations in daytime temperature, precipitation and relative humidity have had statistically significant results in multiple linear regressions for the number of dengue cases. 32 The association between climatic factors and dengue incidence suggests that application of any prospective dengue early warning system should be done on a local or regional basis rather than on a national scale. 33 The only 5% were suspected of being caused by the Zika virus. 35 The limitations of this study comprised its inclusion criteria, i.e. the subjects needed to have an address in the municipality of Garanhuns, their cases needed to have been notified to SINAN and a diagnosis of dengue needed to have been made. Patients who did not meet these inclusion criteria and those whose addresses could not be georeferenced due to lack of information were excluded: these exclusions corresponded to 13.78% of the total number of notified cases.

CONCLUSION
The climate and local geography of the study area, characterized by wide variations in temperature and precipitation, with prolonged periods of drought and densely populated neighborhoods, may have contributed to greater reproduction and dissemination of the transmitting vector. This may have led to differences in dengue incidence rates over the last five years, thereby increasing the number of outbreaks and even epidemics.
These results should serve as the basis for the creation of new control and continued prevention strategies. They also demonstrate that there is a need for greater in-depth study of the spatial distribution of dengue, using regression analysis.