SEARCH

SEARCH BY CITATION

Keywords:

  • Dengue;
  • hazard rate;
  • Cox's proportional hazard function;
  • time-dependent covariates;
  • survival function;
  • logistic hazard model

ABSTRACT:

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED
  9. Appendix

The survival rate of mosquitoes is an important topic that affects many aspects of decision-making in mosquito management. This study aims to estimate the variability in the survival rate of Ae. aegypti, and climate factors that are related to such variability. It is generally assumed that the daily probability of mosquito survival is independent of natural environment conditions and age. To test this assumption, a three-year fieldwork (2005–2007) and experimental study was conducted at Fortaleza-CE in Brazil with the aim of estimating daily survival rates of the dengue vector Aedes aegypti under natural conditions in an urban city. Survival rates of mosquitoes may be age-dependent and statistical analysis is a sensitive approach for comparing patterns of mosquito survival. We studied whether weather conditions occurring on a particular day influence the mortality observed on that particular day. We therefore focused on the impact of daily meteorological fluctuations around a given climate average, rather than on the influence of climate itself. With regard to survival time, multivariate analyses using the stepwise logistic regression model, adjusted for daily temperature, relative humidity, and saturated vapor pressure deficit (SVPD), suggest that age, the seasonal factor, and the SVPD were the most dependent mortality factors. Similar results were obtained using the Cox proportional hazard model, which explores the relationships between the survival and explanatory variables.


INTRODUCTION

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED
  9. Appendix

In most tropical and subtropical cities (Hopp and Foley 2001, Diallo et al. 2005), the yellow fever mosquito Aedes aegypti (L.) is also a vector of dengue virus (DEN), causing dramatic epidemics (Gubler 1997). The control of mosquito populations remains the only means to prevent epidemics, but it is actually very difficult to maintain at a suitable level, especially when resistance to insecticides occurs (Carvalho et al. 2004, Lima et al. 2003).

In Brazil, except for earlier reports published when only clinical diagnostics were available (Pedro 1923), DEN reemerged beginning with the epidemics in Boa Vista (RR) in 1982, and Rio de Janeiro (RJ) in 1986 (Degallier et al. 1998, Degallier et al. 1996). Despite Brazilian climate conditions being favorable for DEN transmission, seasonal and interannual variations of DEN epidemics are clearly linked to seasonal and interannual variability of regional climate (Tauil 2002). Yearly peaks of dengue cases are generally noted during and just after the rainy season (Degallier et al. 1996). That is the case for Fortaleza (Ceará State, CE), a city with three million inhabitants located at the oceanic border of the semi-arid Brazilian Northeast Region (Degallier et al. 2012), and where several epidemics occurred these last decades (Vasconcelos et al. 1989, Vasconcelos et al. 1995).

The present work reports the results of a three-year field study (2005–2007) designed to investigate whether local climate can influence aging and mortality of the dengue vector Ae. aegypti. The main objective of this research was to test the direct relationship, if it exists, between some meteorological variables, such as air temperature (AT), air relative humidity (RH), or the saturated vapor pressure deficit (SVPD), and the daily mortality of the Aedes aegypti mosquito. The SVPD is the difference between the amount of moisture in the air and how much moisture the air can hold when it is saturated. Unlike relative humidity, vapor pressure deficit has a nearly straight-line relationship to the rate of evapotranspiration and other measures of evaporation. In order to develop an early warning system for preventing dengue epidemics, a mechanistic model based on statistics was designed using the results of previous works (Degallier et al. 2006, Favier et al. 2005).

However, some relations between climate parameters and mosquito survival (or mortality) needed both field evaluation and regional validation. Here a difficulty arose, as some authors have shown that, contrary to the generally accepted assumption, the mosquito mortality rate does not remain constant throughout its life (Styer et al. 2007, Degallier et al. 2012). Indeed, these last authors showed that young mosquitoes have a lower mortality rate (also known as hazard rate) than old ones and thus, may have a higher vectorial capacity. An ensemble of questions related to the Ae. aegypti age and daily mortality at Fortaleza-CE were highlighted in Degallier et al. (2012).

The study by Degallier et al. (2012) estimated the variability in the survival rate of Ae. aegypti and the environmental factors, such as experimentation conditions, site and season, that are related to such variability. In the present work, we studied the same experimental data to estimate the variability in the survival rate of the mosquitoes, taking into account the typical meteorological attributes that may be related to such variability.

MATERIALS AND METHODS

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED
  9. Appendix

The meteorological observations consisted of temperature and RH recorded each ten min. We summarized these data by computing the daily mean and the daily range, considering the minimum and maximum record values for two meteorological observed variables, air temperature and air RH, and one calculated SVPD, obtaining ten typical weather variables for each day.

Climate data for Fortaleza at the city scale

The region of Fortaleza has a typical tropical wet and dry climate (Figure 1) with high temperatures and high RH throughout the year. December and January are the warmest months, with a mean range of 25° C to 31° C. On a yearly average, the RH is 77% and the total amount of rainfall is 1,378 mm. The precipitation occurs during the first six months of the year, when relative humidity is high. A strong rainy season spans from February to May, with rainfall particularly intense in March and April. The climate is generally dry during the last six months of the year with very little rainfall in that period (Hastenrath and Greishar 1993).

image

Figure 1. Fifty years of monthly climate descriptions for Fortaleza-CE, Brazil.

Download figure to PowerPoint

The monthly number of DEN cases recorded for the city of Fortaleza remained very low during the first three months of the year (Figures 2a, 2b). This number grew to 4,000 cases/month in July, 2005 and August, 2005, a few months after the end of the rainy season. The year 2007 was somewhat different, with a slower temporal distribution and an attenuation of the seasonal peak (about 2,000 cases/month from May to July). Therefore, we distinguish the temporal pattern of DEN notification for the city of Fortaleza-CE (Figure 2c).

image

Figure 2. (a) and (b) 10-years monthly Dengue notification (DATASUS reported cases) in Fortaleza-CE, Brazil, (c) the monthly evolution of the Dengue notifications during the experimental procedure: 2005–2006–2007.

Download figure to PowerPoint

Measured at the meteorological station of the Universidade Federal do Ceará (UFC) (Site #0 in Figure 1 from Degallier et al. (2012)), weekly air temperature (AT) and air RH were relatively stable during 2005–2007, with a yearly average of 27° C and 78%, respectively. The seasonal variation, with ranges of about 8° C for AT and 35% for RH, followed the same seasonal variation as that of rainfall, with lowest (highest) temperature and highest (lowest) humidity observed during the wet (dry) season. The saturated vapor pressure deficit (SVPD) showed a similar seasonal variation with low values (average ∼ 6 mb) during the four months of the rainy season, and higher values (up to 12 mb) during the remaining eight months.

Experimental procedure at the local scale

During the three years of the study, we conducted twelve experiments (EXP1–12) distributed on seven sites (from #1 to #7, Degallier et al (2012)) in Fortaleza, separated by 3 to 10 km. Sites #1 to #6 were operated in private or institutional locations within a natural air environment, while Site #7 was operated inside a closed air-conditioned room at the Ceará State Secretary of Health. AT and RH were recorded each ten min. Over three years (2005–2007), 12 experiments were conducted, equally distributed during the wet and dry seasons, respectively. Additional information about local conditions and temporal experiment difficulties is provided in Degallier et al. (2012).

Briefly, 40–50 mosquitoes (35–45 females and approximately five males) were released in netted wooden cages (30×30×30 cm) at the Laboratory of Entomology of SESACE (Secretaria Estadual de Saúde do Ceará). The mosquitoes were taken from their rearing cage two days after hatching. Just before release in the experimental cages, the females were offered anaesthetized quail for blood-feeding for two hours. On the same day as blood-feeding, the cages were installed at the experimental sites. Cotton plugs with ten percent sugar were changed daily and dead mosquitoes were counted. A small tube with filter paper and tap water was renewed daily in each cage to collect eggs. Most of the twelve experiments lasted until the death of the last mosquito in the last cage. As there was no large-scale outbreak of dengue during the experiments, the cages were not exposed to insecticides.

Significant seasonal differences were recorded at all sites where mosquitoes were exposed to natural air conditions. The daily variations recorded a higher variability for the wet season, during which rainy episodes of a few days could induce a drop in AT which is associated with higher humidity. On the contrary, the climate parameters were quite stable during the dry season. Independent of a specific season, the daily variations of the measured variables were somewhat different between the sites. For instance, the daily temperature range was systematically higher for Site #2 than for Site #1, as for the daily relative humidity range, on behalf of the dry season and similarly for the wet season (Degallier et al. 2012).

Analysis of the influence of weather conditions and age on mortality

In order to investigate the relationships between local climate and mosquitoes survival rates, we combined three statistical methods which are described below. These methods were chosen for their power and suitability in survival analysis, related to the time series of explanatory weather variables.

Discrete hazard and survival functions

We used the framework of survival analysis to analyze the influence of weather conditions and age on mortality. While analyzing the influence on mortality of factors that vary in time, lifetime observations could not be used as a target variable to be explained and a regression model was required. Hence, two statistical analyses were performed in this study: (1) using as a response the dummy variable representing death occurrence of a given individual on a given day; (2) using as a response the discrete variable representing the number of death occurrences on a given day. We then attempted to explain these variables using the ten climate factors as predictors.

To model the effect of some continuous variables X1,…,Xk on the dummy variable Y, associated with the mosquito's risk of death, we used the binary logistic regression (Agresti 2002). as well as proportional hazard modeling. To analyze the influence of climate and age on mortality, we used the framework of survival analysis (Lawless 2003), the branch of statistics dealing with death in biological organisms and in particular the concepts of survival and hazard rate functions. The survival function, conventionally denoted by s, is defined as s(t) = 1-F(t), where F(t)-P(T ≤ t) is the cumulative probability function, with t the random variable denoting the time of death. The hazard function (also known as force of mortality or hazard rate), is defined as the event rate at time t, conditional on survival until time t or later: P(t ≤ T < t +δtTt. Implementing this framework, we estimated an empirical hazard function.

The proportional hazards models are a class of survival models in statistics. Survival models relate the time that passes before some event occurs to one or more covariates that may be associated with that quantity (Comfort 1979). In a proportional hazards model, the unique effect of a unit increase in a covariate is multiplicative with respect to the hazard rate. Survival models can be viewed as consisting of two parts: the underlying hazard function, often denoted h0(tj), describing how the hazard (risk) changes over time at baseline levels of covariates, and the effect parameters, describing how the hazard varies in response to explanatory covariates. The proportional hazards condition (Breslow 1975) states that covariates are multiplicatively related to the hazard. However, the covariates are not restricted to binary predictors, and in the case of a continuous covariate Xi, the hazard responds logarithmically. Thus, each unit increases in Xi results in proportional scaling of the hazard. The effect of covariates estimated by any proportional hazards model can thus be reported as hazard ratios. Cox (1972) observed that if the proportional hazards assumption holds (or is assumed to hold), then it is possible to estimate the effect parameters without any consideration of the hazard function. This approach to survival data is an application of the Cox proportional hazards model, sometimes abbreviated to Cox model or to proportional hazards model.

The logistic hazard model

Cox (1972) proposed an extension of the proportional hazards model to discrete time by working with the conditional odds of dying at each time tj given survival up to that point, the model is given by:

  • image

where hT(tjXi) is the hazard at time tj for an individual with covariate values Xi, h0(tj) is the baseline hazard at time tj, and exp {Xiβ} is the relative risk associated with covariate values Xi. Taking logs, we obtain a model on the logit (l) of the hazard or conditional probability of dying at tj given survival up to that time: l[hT(tjXi)] =αj+Xiβ, where αj=l[h0(tj)] is the logit of the baseline hazard and Xiβ is the effect of the covariates on the logit of the hazard. Note that the model essentially treats time as a discrete factor by introducing one parameter αj for each possible time of death tj. Interpretation of the parameters β associated with the other covariates follows along the same lines as in the logistic regression. Time-varying covariates and time-dependent effects can be introduced in this model along the same lines as before. In the case of time-varying covariates, note that only the values of the covariates at the discrete times t1 <t2 <…<tj-1 <tj are relevant. Time-dependent effects are introduced as interactions between the covariates and the discrete factor (or set of dummy variables) representing time (Therneau and Grambsch 2000).

The logistic regression analyses binomially distributed data, where the numbers of Bernoulli trials n are known and the probabilities of success p are unknown. The model proposes for each trial that there is a set of explanatory variables that might inform the final probability. The model then takes the form:

  • image

One can transform the output of a linear regression to be suitable for probabilities by using a logit link function. The logit, natural logs of the odds, of the unknown binomial probabilities are modelled as a linear function of the explained variables X1,…,Xk:

  • image

For a real-valued explanatory variable X1, the intuition is that a unit additive change in the value of the variable should change the odds by a constant multiplicative amount. The logit function is invertible, so

  • image

The parameters of the model {β01,…,βk} are estimated by the principle of maximum likelihood based on the data. So, the binary logistic regression is a useful way to describe the relationship between one or more independent variables and a binary response variable, expressed as a probability. The logistic function is defined as

  • image

The input is z and the output is f(z), which is confined to values between 0 and 1; f(z), represents the probability of a particular outcome, given that set of explanatory variables. The variable z is a measure of the total contribution of all the independent variables used in the model and is known as the logit. In this study the variable z was defined as z01X1+…+βkXk.

From a technical point of view, there is no error term in a logistic regression, unlike in classic linear regression. The logistic regression is useful when we are predicting a binary outcome from a set of continuous predictor variables. It is frequently preferred over discriminant function analysis because of its less restrictive assumptions.

The Cox proportional hazard model

Cox proportional-hazards regression allows analyzing the effect of several risk factors on survival. The probability of the endpoint (death) is called the hazard. The hazard function for the Cox proportional hazard model is modeled as:

  • image

where Xi is a collection of predictor variables and h0(tj) is the baseline hazard at time tj, representing the hazard for a sample unity (mosquito) with the value 0 for all the predictor variables. By dividing both sides of the above equation by h0(tj) and taking logarithms, we obtain:

  • image

One calls inline image

the hazard ratio. The coefficients {β01,…,βk} are estimated by Cox regression and can be interpreted in a similar manner to that of multiple logistic regressions. Suppose the covariate is discrete, then the quantity exp{Xi,β} is the instantaneous relative risk of an event, at any time, for an individual with an increase of one-unity in the value of the covariate compared with another individual, given both individuals are the same on all other covariates.

The Cox proportional regression model assumes that the effects of the predictor variables are constant over time. Furthermore there should be a linear relationship between the endpoint and predictor variables. Predictor variables that have a highly skewed distribution may require logarithmic transformation to reduce the effect of extreme values. This model is robust and a safe choice of a model in many situations. Because of the model form

  • image

the estimated hazards are always non-negative. Even though h0(tj) is unspecified, we can estimate {β01,…,βk} and thus compute the hazard ratio. The hT(tjXi) and ST(tjXi) can be estimated for a Cox model using a minimum of assumptions. In survival analysis, the Cox model is preferred to a logistic model, since the latter one ignores survival times.

The proportional hazard model is the most general of the regression models because it is not based on any assumptions concerning the nature or shape of the underlying survival distribution. The model assumes that the underlying hazard rate, rather than survival time, is a function of the independent variables (covariates), and no assumptions are made about the nature or shape of the hazard function. Thus, Cox's regression model may also be considered as a nonparametric method. The model may be written as:

  • image

where h(t) denotes the resultant hazard, given the values of the k covariates for the respective case (X1,…,Xk) and the respective survival time (t). The term h0(t) is called the baseline hazard, the hazard for the respective individual when all independent variable values are equal to zero. The baseline hazard is an unspecified function that does not depend on X but only on t. The exponential involves the X but not t; X are time-independent. Similar to ordinary linear regression in the logistic hazard model, the unknown parameters {β01,…,βk}are usually estimated by maximum likelihood.

Although the Cox model is non-parametric to the extent that no assumptions are made about form of the baseline hazard, there are still a number of important issues which need be assessed before the model results can be safely applied. First, they specify a multiplicative relationship between the underlying hazard function and the log-linear function of the covariates. This assumption is also called the proportionality assumption. In practical terms, it is assumed that, given two observations with different values for the independent variables, the ratio of the hazard functions for those two observations does not depend on time. The second assumption is that there is a log-linear relationship between the independent variables and the underlying hazard function.

An hypothesis of the proportional hazard model is that the hazard function for an individual depends on the values of the covariates and the value of the baseline hazard, h0(t). Given two individuals with particular values for the covariates, the ratio of the estimated hazards over time will be constant, hence the name of the method: the proportional hazard model. The validity of this hypothesis may often be questionable.

After the data compilation for all individuals, we submitted an application of the logistic regression to the dataset, using all available climate variables at a time as a predictor, to analyze their effect separately. We also considered potential delayed effects of predictors by applying the regression to lagged variables (Martinussen and Scheike 2006), with lag ranging from one to five days. Finally, we considered cumulative effects using the sum of the variable over the past two to five days as a predictor.

Following the statistical analysis, we estimated the logistic regression coefficients for the daily weather attributes on the complete sample. For this, we have chosen the model by the Akaike Information Criterion (AIC) in a stepwise algorithm (Venables and Ripley 2002), where the seasonal effect was statistically significant. It is worthwhile to note that multicollinearity in the logistic regression model (as well as in the Cox model) is a result of strong correlations between explanatory variables. The existence of multicollinearity inflates the variances of the parameter estimates. That may result in wrong signs and magnitudes of regression coefficient estimates and consequently, in incorrect conclusions about the relationships between independent and dependent variables. In this study, air temperature and relative humidity were involved in multicollinearity and have been combined into a single variable which was the saturated vapor pressure deficit.

RESULTS

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED
  9. Appendix

The influence of climatic variables is quite limited and difficult to capture. More specifically, this low predictiveness applies equally to air temperature, relative humidity, and saturated vapor pressure deficit: none of these three variables appear to be more predictive than the other in general (Appendix Table 1). After the model selection, concerning the Site #1 (where the hazard ratio was 1.39, which means that mosquito survival decreased about 39% in the dry season (Degallier et al. 2012)), for the TEMP (considering only the air temperature) model in Appendix Table 1a, TMIN and the seasonal effect were statistically significant for a confidence level of α= 0.05 for the RHUM and SVPD models (Appendix Tables 1b,1c) only the seasonal effects were statistically significant. In Site #2, where the hazard ratio was 1.27 (Degallier et al. 2012), none, including the meteorological attribute, was statistically significant at a significance level of α= 0.05 (Appendix Table 2). The outputs are not significant, indicating absence of evidence on climate on mosquito mortality records. In Site #7, where the hazard ratio was invariant considering the season (Degallier et al. 2012), and where the air conditioning was operated during the day time, MaxT and MaxSVPD were statistically significant to design mosquito mortality, at a significance level of α= 0.05 (Appendix Table 3).

Table 1.  The symbolic description of the R output for the Site #1 logistic linear model: (a) air temperature – TEMP, (b) air relative humidity – RHUM and (c) saturated vapor pressure deficit – SVPD. Significant relationships (p-values < 0.05) are in boldface. Thumbnail image of
Table 2.  The symbolic description of the R output for the Site #2 logistic linear model: (a) air temperature – TEMP, (b) air relative humidity – RHUM and (c) saturated vapor pressure deficit – SVPD. Significant relatioships (p values < 0.05) are in boldface. Thumbnail image of
Table 3.  The symbolic description of the R output for the Site #7 logistic linear model: (a) air temperature – TEMP, (b) air relative humidity – RHUM and (c) saturated vapor pressure deficit – SVPD. Significant relationships (p values < 0.05) are in boldface. Thumbnail image of

The application of the logistic regression to the weather attribute ranges is shown in Appendix Tables 4, 5, and 6. The three simplified models consider, for each location, the three risk factors ((XΔT), (XΔRH), and (XΔSVD)) and the seasonal effect to predict the daily risk of mosquito death. For Site #1, the parameters that fit the data are β0 (the intercept) and the β4 for the (negative) seasonal effect factor for each meteorological attribute (Appendix Table 4). The model can thus be expressed as

Table 4.  The symbolic description of the R output for the Site #1 logistic linear model: (a) air temperature range, (b) air relative humidity range and (c) saturated vapor pressure deficit range. Significant relationships (p values < 0.05) are in boldface. Thumbnail image of
Table 5.  The symbolic description of the R output for the Site #2 logistic linear model: (a) air temperature range, (b) air relative humidity range and (c) saturated vapor pressure deficit range. Significant relationships (p values < 0.05) are in boldface. Thumbnail image of
Table 6.  The symbolic description of the R output for the Site #7 logistic linear model: (a) air temperature range, (b) air relative humidity range and (c) saturated vapor pressure deficit range. Significant relationships (p values < 0.05) are in boldface. Thumbnail image of
  • image

where z=–βX4. For Site #2 (Appendix Table 5), the parameter which fits the data is β1∼ 0.10 for the daily air temperature range X1. So, in this model, increasing daily temperature range is associated with an increase of mosquito mortality (z goes up by β1= 0.098 for every day over the temperature range). For Site #7 (Appendix Table 6), the parameters which fit the data are β1= 0.09 for the daily range temperature and β1= 0.04 for the daily saturated vapor pressure deficit range. In these models, daily relative ranges are associated with an increased risk of mosquito mortality.

Then, if one wishes to use the Site #2 TEMP model to predict a particular mosquito's risk of death and considering a daily temperature range of 100 C (or the Site #7 SVPD range model for SVPD range of about 5 mb), the daily risk of mosquito death is therefore

  • image

with z = 0.10*10 = 1. This means that by these models, the mosquito's risk of dying on this specific day is augmented (in addition to aging) by:

  • image

We note that for a given meteorological variable (air temperature or air relative humidity), its daily minimum, maximum, and range seem to be equally predictive: none of these three summaries seem to have more predictive power than the others in general. However, they may have effects of opposite sign. This also corroborates the influence of local environment and seasonality on survival, as shown by Degallier et al. (2012).

In Appendix Table 7 we show the results of the Cox regression on the weather attribute and their ranges. It is worthwhile to emphasize that the three simplified models consider, for each location, the three risk factors, (XΔT), (XΔRH), and (XΔSVD) and the seasonal effect to predict the daily risk of mosquito death. For Site #1, the output indicates that climate attributes do not influence mosquitoes' life. Only the seasonal factor has highly statistical significant coefficients, at a significance level of α= 0.05. For Site #2, the covariates maximum daily air temperature, maximum daily relative humidity, and maximum SVPD have marginally significant statistical coefficients, at a significance level of α= 0.05. For Site #7, the SVPD covariates have highly significant statistical coefficients, at a significance level of α= 0.05. The exponentiated coefficients in the second column of the output are interpretable as multiplicative effects on the hazard. It means that holding the other covariates constant, an additional unit on increasing the daily temperature range augments the daily hazard of death by a factor of e0.216= 1.241 on average. Similarly, each additional unit on relative humidity range increases the hazard of death by a factor of 1.080. Having fit a Cox proportional hazard model to the data, it was of interest to examine the estimated distribution of survival times (Figure 3). One way to enhance the model is to use lagged (Martinussen 2006) values of covariates. Unexpectedly, the coefficients for the lagged weather covariates were statistically highly unsignificant. As a result of our survival analysis, the Cox model was preferred to the logistic model, since the latter one ignores survival times. This finding is also consistent with the results of Styer et al. (2007) showing that mosquitoes do senesce.

Table 7a.  Cox proportional hazard regression of mosquitoes' time to death on the time-constant local weather covariates. Significant relationships (p values < 0.05) are in boldface. Thumbnail image of
Table 7b.  Cox proportional hazard regression of mosquitoes' time to death on the time-constant local weather range covariates. Significant relationships (p values < 0.05) are in boldface. Thumbnail image of
image

Figure 3. The estimated distribution of survival times for the Cox proportional hazard regression of mosquito time to death on the time-constant local weather covariates as predictors.

Download figure to PowerPoint

Although daily weather and seasonal to inter-annual climatic variability influence mosquito vector biology and risk of vector-borne disease, this information is not readily employed in disease control programs. Note that mosquito activity may be unusually high or low, depending on weather conditions (Aitken et al. 1968). High temperatures, in the days following mosquito hatching, may have accelerated mosquito development and allowed many adults to emerge. Rainfall dictates when breeding sites will be flooded and when they will need to be inspected for mosquito larvae. In conclusion, the connections between the weather conditions and dengue transmission and outbreaks is not yet clear. Even if the weather conditions are favorable for transmission, the local human population may already be immune to the prevalent virus. Therefore, to develop a system for predicting and monitoring risk of mosquito vectors, factors such as recent rainfall, humidity, and temperature must be taken into account.

DISCUSSION

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED
  9. Appendix

Mosquito mortality is an important parameter of vectorial capacity, especially in the case of arboviruses for which no vaccine is available, and thus the elimination of the vectors remains the only means of prevention (Barbazan et al. 2010). Natural mortality of adult mosquitoes may be due either to intrinsic (aging, genetic) or extrinsic (predation, control, weather) factors. We showed that the mortality of Aedes aegypti adult mosquitoes in semi-natural conditions may no more be considered as a constant phenomenon during their life. It varies according to their age and to meteorological conditions, even in tropical climates. Such influences may be taken into account in models that are developed to forecast the risk of epidemics on a local scale.

As Degallier et al. (2012) showed, age, environment, and season may influence mosquito mortality. Complementing this study, we showed here that the mortality risk varies according to local daily weather conditions. However, this hypothesis is attenuated by the following result: “The total average of the experiment durations for Site #7, where the air conditioning was generally operated during the duty day time, is very close to the general average for Site #1 and Site #2 (49.5 days vs 48.9 days).” (Degallier et al. 2012). Ultimately, with these somewhat contradictory results in mind, it seems difficult to assert that the climatic daily conditions have a significant influence on the mosquito mortality, as was our initial working assumption.

Styer et al. (2007) reported that models based on mosquito populations with constant mortality are far easier to manage. Further, senescence has been poorly documented in natural or semi-natural conditions. Besides, mortality is lower for young females than for older ones (Degallier et al. 2012). Despite differences in the experimental protocols, we were able to confirm that mortality of mosquitoes varies with age, environment, and weather conditions. In addition, we have presently shown that under the climate of Fortaleza-CE-Brazil, only the deficit of vapor pressure (SVPD) may impact the mortality of the older mosquitoes that are potentially better able to transmit pathogens.

Including these results in models that comprise an early warning system would allow better predictions of risks of epidemics (Degallier et al. 2010a). As already mentioned by Degallier et al. (2010b), there are often statistically significant interactions between mosquito age and meteorological variables. For instance, the effect of the daily minimum relative humidity is positive for old individuals, but it is negative for young individuals. Again, older female mosquito survival was affected by minimum and range air temperature and relative humidity. The favorable conditions for younger mosquitoes may therefore turn out to be unfavorable conditions for older ones. Their combined effect on a natural population would therefore be ambiguous and dependent on its age structure.

Acknowledgments

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED
  9. Appendix

This work was part of the CNPq-IRD Project “Climate of the Tropical Atlantic and Impacts on the Northeast” (CATIN), CNPq Process 492690-2004-9. Additional funding came from a special project approved by the French Ministry of Foreign Office (MAE, Brasilia) via the IRD Representation in Brazil. JS thanks the FUNCAP for a grant (BPV-0025-00055.01.00/11). We thank all the engineers and technical people from FUNCEME, SESACE, and UFC who helped, often with difficulties, but always with good will, in the preparation and the realization of the experiments.

REFERENCES CITED

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED
  9. Appendix
  • Agresti, A. 2002. Categorical Data Analysis . John Wiley & Sons.
  • Aitken, T.H.G., C.B. Worth, and E.S. Tikasingh. 1968. Arbovirus studies in Bush Bush forest, Trinidad, W.I., September 1959 – December 1964. II – Field program and techniques. Am. J. Trop. Med. Hyg. 17: 237252.
  • Barbazan, P., M. Guiserix, W. Boonyuan, W. Tuntaprasart, D. Pontier, and J.-P. Gonzalez. 2010. Modelling the effect of temperature on transmission of dengue. Med. Vet. Entomol. 24: 6673.
  • Breslow, N.E. 1975. Analysis of survival data under the proportional hazards model. Int. Statist. Rev./ Rev. Int. Statist. 43: 4557.
  • Carvalho, M.d.S.L.d., E.D. Caldas, N. Degallier, P.d.T.R Vilarinhos, L.C.K.R. de Souza, M.A.C. Yoshizawa, M.B. Knox, and C. de Oliveira. 2004. Susceptibility of Aedes aegypti larvae to the insecticide temephos in the Federal District, Brazil. Rev. Saúde Públ. 38: 623629.
  • Chaturvedi, U.C., R. Shrivastava, and R. Nagar. 2005. Dengue vaccines: problems and prospects. Indian J. Med. Res. 121: 639652.
  • Cox, D.R. 1972. Regression models and life-tables. J. R. Statist. Soc. Ser. B 34: 187220.
  • Comfort, A. 1979. The Biology of Senescence. Elsevier, New York , 274 pp.
  • Degallier, N., C. Favier, C.E. Menkes, J.-P. Boulanger, J. Servain, W.M. Ramalho, and M. Lengaigne. 2006. Dengue transmission modeling and risk assessment under climatic changes. In: ICEID – International Conference on Emerging Infectious Diseases, March 19–22, 2006. pp. 112. American Society for Microbiology, Atlanta, GA, U.S.A.
  • Degallier, N., P.S. Lucio, J. Servain, A. Hannart, B. Durand, R. N. de Souza, and Z.M. Ribeiro. 2012. The influence of local environment on the Aedes aegypti (L.) aging and mortality. Case study: Fortaleza-CE, Brazil. J. Vector Ecol. 37: 428441.
  • Degallier, N., C. Favier, C.E. Menkes, M. Lengaigne, W.M. Ramalho, R. Souza, J. Servain, and J.-P. Boulanger. 2010a. Toward an early warning system for dengue prevention: modeling climate impact on dengue transmission. Climat. Change 98: 581592.
  • Degallier, N., J. Servain, A. Hannart, B. Durand, R.N. de Souza & Z.M. Ribeiro. 2010b. Impactos climáticos sobre a transmissão da Dengue no Nordeste do Brasil. In: J. Servain, J.N.B. Campos, E.S.P.R. Martins & D.S. Reis Jr. (eds). Clima do Atlântico Tropical e Impactos sobre o Nordeste (CATIN) Climat de l'Atlantique tropical et impacts sur le Nordeste (CATIN). pp. 331337. CNPq/FUNCEME/IRD, Gráfica e Editora Ronda Ltda.
  • Degallier, N., A.P.A. Travassos da Rosa, P.F.C. Vasconcelos, M. Moraes, L.T. Figueiredo, J.F.S. Travassos da Rosa, S.G. Rodrigues, E.S. Travassos da Rosa, B. Mondet, and P. Reiter. 1996. Le dengue et ses vecteurs au Brésil. Discussion. Ateliers, Cayenne, Guyane française (23/05/1995). Bull. Soc. Pathol. Exot., 89: 128136.
  • Degallier, N, J.P. Hervé, A.P.A. Travassos da Rosa, E.S. Travassos da Rosa, P.F.C. Vasconcelos, H.Á.O. Monteiro, G.C. Sá Filho, and J.F.S. Travassos da Rosa. 1998. Entomological studies on Dengue fever vectors in Brazil: the epidemics of Boa Vista, Roraima, 1982, Niteroi, Rio de Janeiro, 1986, and Ceara State, 1986, 1994. In: A.P.A. Travassos da Rosa, P.F.C. Vasconcelos, and J.F.S. Travassos da Rosa (eds.). An overview of Arbovirology in Brazil and neighbouring countries. Instituto Evandro Chagas, Belém , Pará , Brasil . pp. 261271.
  • Diallo, M., A.A. Sall, A.C. Moncayo, Y. Ba, Z. Fernandez, D. Ortiz, L.L. Coffey, C. Mathiot, R.B. Tesh, and S.C. Weaver. 2005. Potential role of sylvatic and domestic African mosquito species in dengue emergence. Am. J. Trop. Med. Hyg. 73: 445449.
  • Favier, C., N. Degallier, M.A. Dubois, J.-P. Boulanger, C.E. Menkes, and L. Torres. 2005. Dengue epidemic modeling: stakes and pitfalls. Asia Pac. Biotech. News 9: 11911194.
  • Gubler, D.J. 1997. Dengue and dengue hemorrhagic fever: its history and resurgence as a global public health problem. In: D.J. Gubler and G. Kuno (eds.). Dengue and Dengue Hemorrhagic Fever. pp. 122. CAB International, Wallingford , Oxon , UK .
  • Hastenrath, S. and L. Greishar. 1993. Circulation mechanisms related to Northeast Brazil rainfall anomalies. J. Geophys. Res. 98: 50935102.
  • Hopp, M.J. and J.A. Foley. 2001. Global-scale relationships between climate and the dengue fever vector, Aedes aegypti. Climat. Change 48: 441463.
  • Lawless, J.F. 2003. Statistical Models and Methods for Lifetime Data, 2nd ed. John Wiley and Sons, Hoboken , N.J. 664 pp.
  • Lima, J.B.P., M.P. da Cunha, R.C. da Silva Jr., A.K.R. Galardo, S.d.S. Soares, I.A. Braga, R.P. Ramos, and D. Valle. 2003. Resistance of Aedes aegypti to organophosphates in several municipalities in the State of Rio de Janeiro and Espírito Santo, Brazil. Am. J. Trop. Med. Hyg. 68: 329333.
  • Luna, J.E.D., M.F. Martins, A.F. dos Anjos, E.F. Kuwabara, and M.A. Navarro-Silva. 2004. Susceptibilidade de Aedes aegypti aos inseticidas temephos e cipermetrina, Brasil. Rev. Saúde Públ. 38: 842843.
  • Martinussen, T. and T.H. Scheike. 2006 Dynamic Regression Models for Survival Data. Springer. 470 pp.
  • McCullagh, P. and J.A. Nelder. 2000. “Chapter 13: Models for Survival Data”. Generalized Linear Models (2nd ed.). Boca Raton , FL . Chapman and Hall/CRC.
  • Pedro, A. 1923. O Dengue em Nicteroy. Brazil-Medico 1: 173177.
  • Styer, L.M., J.R. Carey, J.-L. Wang, and T.W. Scott. 2007. Mosquitoes do senesce: departure from the paradigm of constant mortality. Am. J. Trop. Med. Hyg. 76: 111117.
  • Tauil, P.L. 2002. Aspectos críticos do controle do dengue no Brasil. Cadern. Saúde Públ. 18: 867871.
  • Therneau, T. M. and P.M. Grambsch. 2000. Modeling Survival Data: Extending the Cox Model . Springer, New York , 350 pp.
  • Vasconcelos, A.W., J.S. Cavalcante, M.J.M Timbo, and A. Sleigh. 1989. Dengue in Ceara, Brazil 1985–1989. Impact and prospects. In: M.F. Uren, J. Blok and L.H. Manderson (eds.) Arbovirus Research in Australia. Proceedings 5th Symposium, August 28 – September 1, 1989, Brisbane, Australia. pp. 8690. CSIRO, Brisbane .
  • Vasconcelos, P.F.d.C., D.B. de Menezes, L.P. Melo, E.T.F.P. Pessoa, S.G. Rodrigues, E.S. Travassos da Rosa, M.J. Timbó, I.C.B. Coelho, F. Montenegro, J.F.S. Travassos da Rosa, F.M.O. Andrade, and A.P.A. Travassos da Rosa. 1995. A large epidemic of dengue fever with dengue hemorragic cases in Ceará State, Brazil, 1994. Rev. Inst. Med. trop. São Paulo 37: 253255.
  • Venables, W.N. and B.D. Ripley. 2002. Modern Applied Statistics with S. New York : Springer 4th ed. 496 pp.