Philippe Quenel

Time series analysis of dengue incidence in Guadeloupe, French West Indies: Forecasting models using climate variables as predictors

BackgroundDuring the last decades, dengue viruses have spread throughout the Americas region, with an increase in the number of severe forms of dengue. The surveillance system in Guadeloupe (French West Indies) is currently operational for the detection of early outbreaks of dengue. The goal of the study was to improve this surveillance system by assessing a modelling tool to predict the occurrence of dengue epidemics few months ahead and thus to help an efficient dengue control.MethodsThe Box-Jenkins approach allowed us to fit a Seasonal Autoregressive Integrated Moving Average (SARIMA) model of dengue incidence from 2000 to 2006 using clinical suspected cases. Then, this model was used for calculating dengue incidence for the year 2007 compared with observed data, using three different approaches: 1 year-ahead, 3 months-ahead and 1 month-ahead. Finally, we assessed the impact of meteorological variables (rainfall, temperature and relative humidity) on the prediction of dengue incidence and outbreaks, incorporating them in the model fitting the best.ResultsThe 3 months-ahead approach was the most appropriate for an effective and operational public health response, and the most accurate (Root Mean Square Error, RMSE = 0.85). Relative humidity at lag-7 weeks, minimum temperature at lag-5 weeks and average temperature at lag-11 weeks were variables the most positively correlated to dengue incidence in Guadeloupe, meanwhile rainfall was not. The predictive power of SARIMA models was enhanced by the inclusion of climatic variables as external regressors to forecast the year 2007. Temperature significantly affected the model for better dengue incidence forecasting (p-value = 0.03 for minimum temperature lag-5, p-value = 0.02 for average temperature lag-11) but not humidity. Minimum temperature at lag-5 weeks was the best climatic variable for predicting dengue outbreaks (RMSE = 0.72).ConclusionTemperature improves dengue outbreaks forecasts better than humidity and rainfall. SARIMA models using climatic data as independent variables could be easily incorporated into an early (3 months-ahead) and reliably monitoring system of dengue outbreaks. This approach which is practicable for a surveillance system has public health implications in helping the prediction of dengue epidemic and therefore the timely appropriate and efficient implementation of prevention activities.

Short-Term Effects of Air Pollution on Mortality in Nine French Cities: A Quantitative Summary

Abstract Between 1990 and 1995, 9 French cities provided data on daily air pollution, total mortality, cardiovascular mortality, and respiratory mortality. Personnel in individual cities performed Poisson regressions, controlling for trends in seasons, calendar effects, influenza epidemics, temperature, and humidity, to assess the short-term effects of air pollution. The authors describe results obtained from the quantitative pooling of these local analyses. When no heterogeneity could be detected, a fixed-effect model was used; otherwise, a random-effect model was used. Significant and positive associations were found between total daily deaths in these cities and the 4 air pollution indicators studied: (1) Black Smoke, (2) sulfur dioxide, (3) nitrogen dioxide, and (4) ozone. A 50-μg/m3 increase in Black Smoke (24 hr), sulfur dioxide (24 hr), nitrogen dioxide (24 hr), or ozone (8 hr) was associated with increases in total mortality of 2.9% (95% confidence interval [Cl]) = 1.3, 4.4), 3.6% (95% Cl = 2.1, 5.2), 3.8% (95% Cl = 2.0, 5.5), and 2.7% (95% Cl = 1.3,4.1), respectively. Similar results were obtained for cardiovascular mortality. Except for sulfur dioxide, positive—but not significant—associations were found with respiratory mortality. The internal consistency among the cities studied, as well as consistency with previously published results, favors a causal interpretation of these associations.

Short-term effects of air pollution on total and cardiovascular mortality: the confounding effect of influenza epidemics.

Background: Air pollution is associated with total mortality. This association may be confounded by uncontrolled time-varying risk factors such as influenza epidemics. Methods: We analyzed independent data on influenza epidemics from 7 European cities that also had data on mortality associated with particulates (PM10). We used 10 methods to control for epidemics (5 derived from influenza data and 5 from respiratory mortality series) and compared those results with analyses that did not control for these epidemics. Results: Adjustment for influenza epidemics increased the PM10 effect estimate in most cases (% change in the pooled regression coefficient: −1.9 to 38.9 for total mortality and 1.3 to 25.5 for cardiovascular mortality). A 10-μg/m3 increase in PM10 concentrations (lag 0–1) was associated with a 0.48% (95% confidence interval = 0.27–0.70%) increase in daily mortality unadjusted for influenza epidemics, whereas under the various methods to control for epidemics the increase ranged from 0.45% (0.26–0.69%) to 0.67% (0.46–0.89%). The corresponding figures for cardiovascular mortality were 0.85% (0.53–1.18%) with no adjustment and from 0.86% (0.53–1.19%) to 1.06% (0.74–1.39%) with the methods of control. Conclusions: The association between air pollution and mortality is not weakened by control for influenza epidemic irrespective of the method used. To adjust for influenza epidemics, one can use methods based on respiratory mortality counts instead of counts of influenza cases if the latter are not available. However, adjustment for influenza by any method tested did not markedly alter the air pollution effect estimate.

Predictors of severe manifestations in a cohort of adult dengue patients

BACKGROUND Key symptoms observed during the febrile phase of dengue may identify patients who are likely to progress to severe disease. OBJECTIVES To test this hypothesis, we examined the relationships between symptoms reported by patients at presentation and the development of severe outcomes. STUDY DESIGN Retrospective analysis of data recorded prospectively in 560 adult dengue patients admitted to an emergency department. A logistic regression analysis was used to quantify the association between symptoms reported at presentation and outcome. RESULTS Plasma leakage was observed in 95 patients (17%), severe thrombocytopenia (platelet counts <20 x 10(9)/L) in 93 patients (16.6%) and acute hepatitis in 42 patients (7.5%). Severe thrombocytopenia developed in 57% of patients with plasma leakage and 40.5% of patients with hepatitis. Patients who developed a plasma leakage syndrome were older, mainly male, and reported more often an abdominal pain and a cough. Diarrhea and taking paracetamol >60 mg/kg/day before admission were associated with the development of acute hepatitis. Seven patients died. The mortality rate was 6/95 (6.3%) in patients who developed plasma leakage, 3/42 (7.1%) in patients who developed hepatitis, 5/93 (5.4%) in patients with severe thrombocytopenia, and 3/12 (25%) in the patients who demonstrated together all these severe manifestations. CONCLUSION Plasma leakage, severe thrombocytopenia and acute hepatitis identified subgroups of adult dengue patients with increased mortality rates. Key symptoms reported by the patients at presentation such as abdominal pain, cough or diarrhea were significantly associated with the development of severe manifestations and should be considered as warning signs.

Short-term effects of air pollution on mortality in the cities of Rouen and Le Havre, France, 1990-1995.

Abstract In this study, the authors examined the short-term effects of ambient air pollution on mortality across 2 French cities: Rouen and Le Havre. In Poisson regression models, which controlled for day-of-week effects, the authors used nonparametric smoothing to control for temporal trend, weather, and influenza epidemics. In Rouen, an interquartile range increase of 60.5–94.1 μg/m3 of ozone was associated with an increase of 4.1 % (95% confidence interval = 0.6, 7.8) of total mortality. Daily variations in sulfur dioxide (interquartile range increase = 17.6–36.4 μg/m3) were also associated with an 8.2% increase (95% confidence interval = 0.4, 16.6) in respiratory mortality. An increase of 6.1% (95% confidence interval = 1.5, 10.9) of cardiovascular mortality was also observed with an interquartile range increase of nitrogen dioxide (i.e., 25.3–42.2 μg/m3). With respect to Le Havre, an interquartile range increase in daily levels of sulfur dioxide (11.3–35.6 μg/m3) was associated with an increase of approximately 3% (95% confidence interval = 0.8, 5) of cardiovascular mortality. For particulate matter less than or equal to 13 urn in diameter (interquartile increase = 21.5, 45.4 μg/m3), an increase of 6.2% (95% confidence interval = 0.1,12.8) was observed. The estimates of pollutant effects and their standard deviations were slightly affected by the degree of smoothing temporal variations in this study. When low collinearity was present, the 2-pollutant models provided acceptable estimates of pollutant effects. They suggested that the ozone effect was independent of the Black Smoke effect, and that the effects of sulfur dioxide and nitrogen dioxide were unlikely to be confounded by ozone concentrations. However, high collinearity leads to large estimates of the pollutant coefficient variances and, therefore, leads to inaccurate estimates of pollutant effects. The analysis of the contributory effects of different pollutant mixtures requires further investigation in those instances in which high collinearity between pollutants is present.

Outbreak of Leptospirosis after a Race in the Tropical Forest of Martinique

Three athletes who participated in a race in the tropical forest of the Caribbean island of Martinique were subsequently diagnosed with leptospirosis using polymerase chain reaction (PCR). We investigated an outbreak to evaluate possible risk factors, and to determine the appropriate public health recommendations. Of 230 athletes, we contacted 148 (64%) and 20 (13.5%) met our case definition. Five were hospitalized and none were fatal. Ten (91%) of the 11 ill athletes who were tested were confirmed by PCR or serology. Serogroup Pyrogenes was commonly found. Cutaneous cuts, reported by 14 (73.7%), was the only potential risk factor using univariate analysis. Sporting event participants in tropical areas should be made aware of specific warnings and recommendations concerning the risk of leptospirosis, especially after periods of heavy rainfall or flooding. Rapid diagnostic assays such as PCR are particularly appropriate in this setting for early diagnosis and for formulating public health recommendations.

Short-term effects of airborne pollen on the risk of allergic rhinoconjunctivitis.

The authors carried out a pilot study in Clermont-Ferrand (France) over the period from 2000 to 2001 to analyze the short-term relationship between daily variations in rhinitis and allergic conjunctivitis and daily variations in pollen counts. Data from the ERASME database made it possible to identify all the individuals for whom at least one drug was prescribed to treat rhinitis or allergic conjunctivitis. Ten allergenic pollens were studied, and a Poisson regression model with penalized spline functions was used. A positive and significant association was found for pollens of Betula, Cupressaceae, Quercus, Fraxinus, and Poaceae. The relative risks varied between 1.3 (95% confidence interval [CI] = 1.0-1.5) and 1.2 (95% CI = 1.1-1.5) for an interquartile range increase of daily pollen counts of Cupressaceae (13 grains/m3) and of Betula (45 grains/m3), respectively. Further analyses, in particular, those using longer time series, are necessary to confirm these results.

Prospective and descriptive study of adult dengue cases in an emergency department, in Martinique.

OBJECTIVE Knowing about the clinical aspects of dengue in endemic zones is essential to implementation of appropriate case management protocols and public health interventions. PATIENTS AND METHODS The authors made a 4-year prospective, observational study of dengue-infected patients admitted to the emergency department of the Fort-de-France University Hospital. RESULTS Two hundred and sixty-three male and 297 female patients were included. The median age was 37 years (range: 14-91). The diagnosis was based on a positive RT-PCR (463 patients) or on the presence of specific IgM (97 patients). Two hundred and seventy-seven patients (49.5%) presented with dengue fever without complications. According to WHO criteria, 95 patients (17%) developed plasma leakage, including 39 patients (7%) diagnosed with DHF, and 10 (1.8%) diagnosed with DSS. Among the other patients without plasma leakage, 84 (15%) had isolated thrombocytopenia, 14 (2.5%) had internal bleeding, and 90 (16%) had unusual manifestations. Seven patients died (1.3%): fulminant hepatitis (two), myocarditis (one), encephalitis (one), acute respiratory failure (one), gangrenous cholecystitis (one), and post-traumatic intracranial hemorrhage (one). The other patients recovered. Seven patients were pregnant (1.3%) from 6 to 27 weeks of amenorrhea and carried their pregnancy to term without complications. CONCLUSION With this experience, we were able to develop appropriate case management protocols for patients during dengue epidemics.

Virological Surveillance of Dengue in Saint Martin and Saint Barthélemy, French West Indies, Using Blood Samples on Filter Paper

To strengthen active dengue surveillance in Saint Martin and Saint Barthélemy, two French Caribbean islands, we evaluated the epidemiological usefulness of collecting blood samples from NS1-positive dengue patients on filter paper to identify the dengue serotypes circulating in these regions during a 27-month period. This approach allowed dengue serotypes to be identified by reverse transcriptase-polymerase chain reaction in 90.1% of the total set of 666 samples analyzed and, in 95.5% of the samples collected during the acute phase of the disease. This prospective virological surveillance using blood samples absorbed onto filter paper, which were stored at 4°C and shipped at ambient temperature to a specialized laboratory for analysis, allowed us to avoid the logistic and financial costs associated with shipping frozen venous blood samples. This surveillance system offers a low-cost alternative for reinforcing dengue prevention in areas where specialized laboratories do not exist, notably by facilitating the early detection of potentially new dengue serotypes.

First description of a dengue fever outbreak in the interior of French Guiana, February 2006.

BACKGROUND A dengue fever outbreak occurred in the interior of French Guiana from November 2005 onwards. An investigation, with epidemiological, entomological and public health inputs, was initiated. Its objectives were to confirm the outbreak, to describe the emergence of dengue fever in the High Maroni area and to initiate a specific public health response. METHODS The investigation was conducted in Maripasoula in February 2006, the biggest community in that part of the country. Definition criteria were used for suspected, probable and confirmed cases of dengue fever. An entomological evaluation for larvae and adult mosquitoes was carried out. Some personal and collective vector control measures were set up by the vector control team. RESULTS This survey identified 127 suspected dengue fever cases, whereas the epidemiological surveillance system detected only six probable and confirmed cases from the same place and for the same period. The proportion of dengue fever was higher in those people who had not travelled (23.5%) than within the population that had travelled (15.3%) in the three previous months (P = 0.01). Larvae of Stegomyia aegypti were found throughout the town, and adults were captured in 90.9% of the houses. CONCLUSION This is the first time that a dengue fever outbreak has been described beyond the coastal region of this French overseas Department.