First responders intervening in crisis situations are likely to subsequently develop mental disorders. We aimed to identify factors associated with anxiety disorders after a terrorist attack in both the medium and long terms. We used data collected on 180 first responders (medical/psychological health professionals and emergency rescue teams) interviewed face to face at 6-10 months (medium term) and 18-22 months (long term) after the January 2015 terrorist attacks in France. Anxiety disorders were measured using the Mini-International Neuropsychiatric Interview V6 and several other variables including terror exposure (comprising perceived level of exposure and real exposure level), sociodemographic characteristics, social support, mental health history, and access to psychological support resources. We developed a structural equation model to examine the interactions between these different factors. Postattack anxiety disorder prevalence in the medium and long terms was 16% and 14%, respectively. The main associated factors in the medium term were barriers to social support, perceived level of exposure, and a lack of psychological support resources. In the long term, the presence of anxiety disorders in the medium term and barriers to social support were directly associated with having anxiety disorders, while reexposure was indirectly associated. Barriers to social support played a crucial role in the prevalence of anxiety disorders in first responders following this traumatic event, both in the medium and long terms. Promoting stronger social cohesion and providing more psychological support resources following a disaster could help prevent anxiety disorders in this population.

CONTEXT:Tropical areas and small islands are identified as highly vulnerable to climate change, and already experiencing shifts in their temperature distribution. However, the knowledge on the health impacts of temperatures under tropical marine climate is limited. We explored the influence of temperature on mortality in four French overseas regions located in French Guiana, French West Indies, and in the Indian Ocean, between 2000 and 2015.METHOD:Distributed lag non-linear generalized models linking temperature and mortality were developed in each area, and relative risks were combined through a meta-analysis. Models were used to estimate the fraction of mortality attributable to non-optimal temperatures. The role of humidity was also investigated.RESULTS:An increased risk of mortality was observed when the temperature deviated from median. Results were not modified when introducing humidity. Between 2000 and 2015, 979 deaths [confidence interval (CI) 95% 531:1359] were attributable to temperatures higher than the 90th percentile of the temperature distribution, and 442 [CI 95% 178:667] to temperature lower than the 10th percentile.DISCUSSION:Heat already has a large impact on mortality in the French overseas regions. Results suggest that adaptation to heat is relevant under tropical marine climate.

Introduction Structured clinical interviews are the gold standard for assessing mental health. However limited resources may allow the use of only self-report questionnaires. In the context of emergency, such as terrorist attacks, the performance and thresholds of such tools still unclear. Objectives We investigated the performance of the Posttraumatic stress disorder CheckList Scale (PCL-S) and of the Hospital Anxiety and Depression scale (HADS), both compared to the MINI Interview, among civilians and first responders involved in terrorist attacks. Methods The data came from the IMPACTS survey which was conducted from 6-10 months among civilians (N=190) and first responders (N=232) after the January 2015 terrorist attacks in the Paris Region, France. Sensitivity and specificity of the PCL-S and HADS were estimated by the ROC curve, and the optimal threshold was defined using the Youden index. Results Regarding the PCL-S: for civilians and first responders respectively, the overall AUC was 0.947 and 0.899, and the optimal threshold were 38.5 and 39.5. Regarding the HADS-D: for civilians and first responders respectively, the overall AUC was 0.908 and 0.617 and the optimal thresholds were 7.5 and 1.5. For the HADS-A for civilians and first responders respectively, the overall AUC was 0.823 and 0.717, the optimal threshold were 9.5 and 6.5. Conclusions In the context of a terrorist attack, compared to the MINI, our study underlined satisfactory performance of the PCL-S and the HADS-D in screening for PTSD and depression respectively, while the screening of anxiety using the HADS-A was unsatisfactory. Disclosure No significant relationships.

After spreading in the Americas, West Nile virus was detected in Guadeloupe (French West Indies) for the first time in 2002. Ever since, several organizations have conducted research, serological surveys, and surveillance activities to detect the virus in horses, birds, mosquitoes, and humans. Organizations often carried them out independently, leading to knowledge gaps within the current virus' situation. Nearly 20 years after the first evidence of West Nile virus in the archipelago, it has not yet been isolated, its impact on human and animal populations is unknown, and its local epidemiological cycle is still poorly understood. Within the framework of a pilot project started in Guadeloupe in 2019, West Nile virus was chosen as a federative model to apply the “One Health” approach for zoonotic epidemiological surveillance and shift from a sectorial to an integrated surveillance system. Human, animal, and environmental health actors involved in both research and surveillance were considered. Semi-directed interviews and a Social Network Analysis were carried out to learn about the surveillance network structure and actors, analyze information flows, and identify communication challenges. An information system was developed to fill major gaps: users' needs and main functionalities were defined through a participatory process where actors also tested and validated the tool. Additionally, all actors shared their data, which were digitized, cataloged, and centralized, to be analyzed later. An R Shiny server was integrated into the information system, allowing an accessible and dynamic display of data showcasing all of the partners' information. Finally, a series of virtual workshops were organized among actors to discuss preliminary results and plan the next steps to improve West Nile Virus and vector-borne or emerging zoonosis surveillance. The actors are willing to build a more resilient and cooperative network in Guadeloupe with improved relevance, efficiency, and effectiveness of their work.

Background In September 2017, the Hurricane Irma devastated the islands of Saint-Martin and Saint-Barthelemy (French West Indies). This was a particularly distressing time for the local healthcare staff in charge of rescuing the population. The aim of this study was to identify the explanatory factors of post-traumatic distress and burnout in hospital staff. Methods An anonymous questionnaire was sent to all 509 hospital workers of Saint-Martin and Saint-Barthelemy. Post-traumatic distress and burnout was assessed using the Post-Traumatic Stress Disorder Checklist (PCL-S) and Copenhagen Burnout Inventory (CBI) scales. Bivariate and multivariate analyses were used to determine the explanatory variables for these two psychological disorders. Results Two hundred and sixty-two questionnaires were completed (response rate of 51.7%). The explanatory factors of post-traumatic distress were female gender (OR = 12.93, 95% CI: 2.70-232.10), electricity shortages (OR = 2.92, 95% CI: 1.13-8.19) and home damage (OR = 1.16, 95% CI [1.02-1.33]). In parallel, the explanatory factors of burnout were posttraumatic distress (OR: 10.42, 95% CI: 4.72-25.58), female gender (OR = 2,41, 95% CI: 1.24-5.02) and paramedical staff (OR = 2,53, 95% CI: 1.15-6.21). In the multivariate analysis, only burnout was significantly associated with post-traumatic distress (OR = 9.26, 95% CI: 4.11-23.14). Conclusions Six months after Irma, post-traumatic distress among hospital staff was strongly linked to burnout. This study revealed the lack of electricity as a new factor related to post-traumatic distress. It also suggested that psychological intervention should be strengthened.

Background Non-participation and attrition are rarely studied despite being important methodological issues when performing post-disaster studies. A longitudinal survey of civilians exposed to the January 2015 terrorist attacks in Paris, France, was conducted 6 (Wave 1) and 18 months (Wave 2) after the attacks. We described non-participation in Wave 1 and determined the factors associated with attrition in Wave 2. Methods Multivariate logistic regression models were used to compare participants in both waves with those who participated in the first wave only. Analyses were performed taking the following factors into account: socio-demographic characteristics, exposure to terror, peri-traumatic reactions, psychological support, perceived social support, impact on work, social and family life, and mental health disorders. Characteristics of new participants in Wave 2 were compared with participants in both waves using a chi-square test. Results Of the 390 persons who were eligible to participate in the survey, 190 participated in Wave 1 (participation rate: 49%). The most frequently reported reason for non-participation was to avoid being reminded of the painful event (32%, n = 34/105). In Wave 2, 67 were lost to follow-up, 141 people participated, of whom 123 participated in Wave 1 (re-participation rate: 65%) and 18 were new. Attrition in Wave 2 was associated with socio-demographic characteristics (age, French origin) and location during the attacks, but not with terror exposure or mental health disorders. Compared with those who participated in both waves, new participants declared less social and psychological support since the attacks. Conclusions Attrition at 6 months was not associated with exposure to terror or mental health disorders, which indicates that any bias in future analyses on IMPACTS on mental health outcomes will be limited. Our findings suggest the importance of adapting similar surveys for people of foreign origin and of improving strategies to avoid attrition of younger people, for example by using social media, peers, and the educational environment. The present study also revealed that a high level of exposure to terror and a lack of social and psychological support after a terrorist event could impede individuals’ participation in similar surveys in the short term.

ObjectiveDescribe short-term health effects of the Hurricane using the syndromic surveillance system based on emergency departments, general practitioners and dispensaries in Saint-Martin and Saint-Barthélemy islands from September 11, 2017 to October 29, 2017.IntroductionIn Saint-Martin (31 949 inhabitants) and Saint-Barthélemy (9 625 inhabitants) islands in the French West Indies, the surveillance system is based on several data sources: (1) a syndromic surveillance system based on two emergency departments (ED) of Saint-Barthélemy (HL de Bruyn) and Saint-Martin (CH Fleming) and on mortality (SurSaUD® network [1])); (2) a network of sentinel general practitioners (GP’s) based on the voluntary participation of 10 GPs in Saint-Martin and 5 in Saint-Barthélemy; (3) the notifiable diseases surveillance system (31 notifiable diseases to individual case-specific form); (4) the regional surveillance systems of leptospirosis and arboviruses based on the biological cases reported by physicians and laboratories of two islands.On September 6, 2017, Hurricane Irma struck Saint-Martin and Saint-Barthélemy islands. Both islands were massively destroyed. This storm led to major material damages, such as power outages, disturbance of drinking water systems, road closures, destruction of medical structures and evacuation or relocation of residents.In this context, the usual monitoring system did not work and life conditions were difficult. The regional unit of French National Public Health Agency set up an epidemiological surveillance by sending epidemiologists in the field in order to collect data directly from ED physicians, GP’s and in dispensaries. Those data allowed to describe short-term health effects and to detect potential disease outbreaks in the aftermath of Hurricane Irma. This paper presents results of the specific syndromic surveillance.MethodsBefore Irma, ED data were collected daily directly from patients’ computerized medical files that were filled in during medical consultations at ED. Among the collected variables, the diagnosis was categorized according to the 10th revision of the International Classification of Diseases (ICD-10). This surveillance system was completed by aggregated data of Emergency Medical Services (EMS), also including medical diagnosis coded using the ICD10.Because of the sudden disruption in hospital departments due to hurricane, electronic transmission was stopped. To replace it, ED data collection turned temporary into paper-forms and several epidemiologists were sent in Saint-Martin and Saint-Barthélemy to collect data directly from the ED physicians. This system remained until the end of October when connections and data transmission were restored.Because of destruction of medical structures, dispensaries were opened in different strategic areas of the island, 3 in Saint-Martin and none in Saint-Barthélemy. General practitioners have progressively reopened their practice (8 GP’s in Saint-Martin and 5 in Saint-Barthélemy) and patient's data were collected and integrated into the surveillance system.Based on a literature review and former experience, the main pathologies identified for the health risk assessment were: (1) somatic pathologies directly or indirectly related to the hurricane (trauma, wounds, cuts, burns, secondary infection); (2) infectious diseases related to the lack of hygiene partly due to damaged water and electricity networks and unavailable health care structures (gastroenteritis, food infections, respiratory diseases, skin infections, tetanus and other pathologies that may occur in the longer term linked to the incubation period especially leptospirosis and hepatitis A); (3) chronic pathologies by discontinuity of care (renal insufficiency, diabetic, cardio-respiratory decompensation, etc.); (4) pathologies related to animal bites and mosquito bites (vector-borne diseases); (5) psychological and / or psychiatry disorders.Then in the French West Indies, from September 11 to October 29,2017, data were routinely analyzed to detect and follow-up various expected or unusual variations of one or more pathology of the above list.ResultsThe following week after Irma (2017-37), the weekly number of ED visits compared to the mean activity observed in normal situation has increased: 1225 ED visits vs. 313 in 2017-35 in Saint-Martin and 227 ED visits vs. 94 ED visits in 2017-35 in Saint-Barthélemy. ED activity has gradually decreased to finally return to a based-activity as observed before the hurricane at the end of October.From September 11 to October 29, 25% of recorded emergency consultations in Saint-Martin island were trauma, wounds, burns and cuts. As in Saint-Martin, 42% of emergency visits in Saint-Barthélemy were pathologies directly or indirectly related to the passage of Irma (trauma, wounds, etc). Others major causes of ED visits were for treatment renewal (diabetes, renal insufficiency, etc.) and gyneco-obstetric activity because general practitioners had stopped their activity.In dispensaries and general practitioners, the most common pathology was gastroenteritis (11% in Saint-Martin) over the entire period of surveillance. At the beginning of the surveillance, skin infections were the most frequently found (20%) in Saint-Martin and psychological disorders (3%) in Saint-Bartélemy, while at the end respiratory infections were the most frequent (6%) in both islands.No increase in visits for chronic diseases, food-borne diseases, acute respiratory or diarrhea illness were detected. No autochthonous confirmed cases of cholera, leptospirosis, vector-bone disease, hepatitis A or typhoid fever had been reported, due to the destruction of the laboratory.ConclusionsSyndromic surveillance in the French West Indies allowed the epidemiologists to assess rapidly the health impact of hurricane in Saint-Martin and Saint-Barthélemy.The well-established relations between French National Public Health Agency and local professionals of both affected islands allowed to temporary switch from an electronic into a paper-based data transmission without any interruption of data analysis.Although several cluster suspicions have been investigated (especially of gastroenteritis, scabies, etc), no massive outbreak was detected. Then even with a degraded system, syndromic surveillance allowed to reinsure authority of the absence of major health impact due to Irma.References1-Caserio- Schönemann C, Bousquet V, Fouillet A, Henry V. Le système de surveillance syndromique SurSaUD ®. Bull Epidemiol Hebd 2014 ;3-4 :38-44.

BACKGROUND:Terrorist attacks occurred in Paris in January 2015. Aims To assess the mental health impact and the access to psychomedical care of people exposed to the attacks.METHOD:We implemented an open-cohort design 6 and 18 months after the attacks. Exposed civilians and rescue workers were included according to the exposure criteria A for post-traumatic stress disorder (PTSD) in DSM-5. A face-to-face questionnaire conducted by trained psychologists was used to collect sociodemographic characteristics, exposure level, scores on psychometric scales, an international neuropsychiatric interview and access to care.RESULTS:Six months after the attacks, 18% of civilians reported symptoms of PTSD, 31% had anxiety disorders and 11% depression. Among rescue workers, 3% reported symptoms of PTSD and 14% anxiety disorders. During the 48 h following the attacks, 53.2% of civilian had access to psychomedical care v. 35% of rescue and police staff.CONCLUSIONS:We found severe psychological consequences, even in people who were less exposed. Declaration of interest None.

This retrospective case study analysed the incidence and symptoms of ciguatera fish poisoning (ciguatera) in Guadeloupe (French West Indies) between 2013 and 2016. Cases attending the emergency departments of the two public hospitals and the reports received by the regional health authority in charge of monitoring (ARS) were compiled. Two hundred and thirty-four cases of poisoning were observed, with a mean annual incidence of 1.47/10,000 (95% CI): 1.29–1.66), i.e 5 times higher than the previously reported incidence (1996–2006). The main species described as being responsible for poisoning were fish from the Carangidae family (n = 47) (jack), followed by fish from the Lutjanidae family (n = 27) (snapper), Serranidae family (n = 15) (grouper), Sphyraenidae family (n = 12) (barracuda), and Mullidae family (n = 12) (goatfish). One case of lionfish ciguatera was observed. 93.9% of patients experienced gastrointestinal symptoms, 76.0% presented neurological signs (mainly paresthesia, dysesthesia and pruritus) and 40.3% presented cardiovascular symptoms (bradycardia and/or hypotension). A high frequency (61.4%) of hypothermia (body temperature <36.5 °C) was observed. This study reports for the first time the relatively high frequency of cardiac symptoms and low body temperature. The monitoring of ciguatera poisoning throughout the Caribbean region must be improved, notably after reef disturbance due to Irma and Maria major cyclones.

Background: Plasmodium falciparum is responsible for most malaria cases on Mayotte Island, in the Comorian Archipelago. Malaria is endemic and a major public health problem in the archipelago with an intense, stable and permanent transmission. This study reports results of 8 years of malaria surveillance from 2007 to 2014 after the strengthening of malaria control activities in Mayotte and the neighbouring islands.Methods: Surveillance was based on physicians' reports of malaria cases between January 2007 and December 2014. Malaria cases were confirmed by at least a positive rapid diagnostic test and/or demonstration of Plasmodium sp. in a blood smear. The date, and the patients' age, sex, address, presentation of symptoms, biology, treatment and recent history of travel were collected by verbal questioning during consultation and/or hospitalization. Monthly rainfall data were also compiled during the study period.Results: From 2007 to 2014, 2073 cases were reported on Mayotte Island: 977 imported cases, 807 autochthonous cases and 289 cases of unknown origin. The total malaria annual parasite incidence lowered from 3.0 in 2007 to 0.07 per 1,000 inhabitants in 2014 as the autochthonous malaria incidence decreased from 1.6 to 0.004 per 1,000 inhabitants in the same period and in all age groups. Most of the imported cases came from Comoros (94 %). Severe forms represented approximately 11 % of cases, and only two deaths have been recorded among the imported cases. Approximately 19 % of cases were hospitalized (3 % in an intensive care unit). There is clearly a decrease in malaria transmission in Mayotte since 2007 and the goal of elimination seems more achievable than ever. In 2011, Mayotte entered the elimination phase when P. falciparum API passed under 1 case per 1,000 people at risk.Conclusions: The combination of vector control measures, active surveillance and case management, including effective treatment with artemisinin-based combination therapy, has been essential to achieve a present status of low and decreasing malaria transmission on the island. Mayotte has entered the elimination phase, but some goals remain to be accomplished before a programme re-orientation toward malaria elimination is contemplated. Moreover, a regional management policy is crucial because this would allow control measures to be targeted and based on a regional surveillance-response system rather than isolated.