Empirical advances with text mining of electronic health records.

BACKGROUND: Korian is a private group specializing in medical accommodations for elderly and dependent people. A professional data warehouse (DWH) established in 2010 hosts all of the residents' data. Inside this information system (IS), clinical narratives (CNs) were used only by medical staff as a residents' care linking tool. The objective of this study was to show that, through qualitative and quantitative textual analysis of a relatively small physiotherapy and well-defined CN sample, it was possible to build a physiotherapy corpus and, through this process, generate a new body of knowledge by adding relevant information to describe the residents' care and lives. METHODS: Meaningful words were extracted through Standard Query Language (SQL) with the LIKE function and wildcards to perform pattern matching, followed by text mining and a word cloud using R® packages. Another step involved principal components and multiple correspondence analyses, plus clustering on the same residents' sample as well as on other health data using a health model measuring the residents' care level needs. RESULTS: By combining these techniques, physiotherapy treatments could be characterized by a list of constructed keywords, and the residents' health characteristics were built. Feeding defects or health outlier groups could be detected, physiotherapy residents' data and their health data were matched, and differences in health situations showed qualitative and quantitative differences in physiotherapy narratives. CONCLUSIONS: This textual experiment using a textual process in two stages showed that text mining and data mining techniques provide convenient tools to improve residents' health and quality of care by adding new, simple, useable data to the electronic health record (EHR). When used with a normalized physiotherapy problem list, text mining through information extraction (IE), named entity recognition (NER) and data mining (DM) can provide a real advantage to describe health care, adding new medical material and helping to integrate the EHR system into the health staff work environment.

[Syndromic surveillance: review and prospect of a promising concept]. / La surveillance syndromique: bilan et perspective d'un concept prometteur.

Syndromic surveillance appeared in the field of public health surveillance in the late 90's. Initially proposed for public health identification of bioterrorism events, the method failed to provide convincing evidence of its usefulness and potential benefits. The definition which is proposed today by the Centers for Disease Control and Prevention (CDC) of Atlanta is the most commonly accepted. It defines syndromic surveillance as an automatic process that goes from registration to transfer of data recorded within the framework of a professional rather than public health goal. Systems operating today have integrated a public health approach through routine surveillance procedures with a broader focus than bioterrorism, implying active participation of the official public health surveillance structures. Syndromic surveillance offers several advantages including quick access to a large volume of data in real time, no extra-work for data registration and construction of a historical dataset useful as an historical baseline. Nevertheless, the limitations of this type of surveillance should not be forgotten (sometimes limited sensitivity, specificity, important technical burden…). Today, recorded experience shows that there is no opposition between syndromic surveillance and classical surveillance. On the contrary, they should be presented as complementary procedures. Syndromic surveillance should be analyzed from a temporal perspective, examining its short-term use as an alert mechanism, mid-term use for constitution of historical time series, and long-term use for a description of human health in the 21st century.

[Winter surveillance of cold exposure effects on health among the homeless population in the Paris area: data from the Coordinated Health Surveillance of Emergency Department network (Organisation de la surveillance coordonnée des urgences [Oscour(®)])]. / Surveillance hivernale des effets du froid sur la santé des populations sans-domicile en région Île-de-France : utilisation des données du réseau d'Organisation de la surveillance coordonnée des urgences (Oscour(®)).

BACKGROUND: A program for helping homeless individuals in winter is implemented from November 1(st) to March 31(st) each year in France. Its aim is to prevent morbidity and mortality in this population during cold spells and periods of severe cold. A health surveillance system of the homeless population in the Paris area has been proposed to evaluate the effectiveness of the program and to alert decision-makers if an unusual increase in cold-weather effects is observed. The goal of this study was the creation of an indicator for the proposed surveillance system based on emergency department activity in the Paris area (Oscour(®) Network - Organisation de la surveillance coordonnée des urgences). METHODS: The winter 2007-2008 computer medical files of 11 emergency departments in the Paris area were examined to confirm diagnosis and ascertain patient-homelessness for each patient visit which was selected from the Oscour(®) database by the patient chief-complaint or diagnosis code referring to hypothermia or frostbites. The proposed indicator is based on the maximization of three criteria: the positive predictive value, the proportion of people identified as being homeless and the number of emergency department visits. A Shewhart control chart was applied to the indicator for the four winters between 2005 and 2009 in the Paris area. Values beyond the statistical threshold would indicate a need for an adjustment to the program strategy. RESULTS: Two hundred and sixteen medical files were analyzed. An indicator was created, "number of emergency department visits of 15 to 69-years-old persons with chief-complaint or diagnosis code referring to hypothermia". It had a positive predictive value estimated near 85 % and identified 61.7 % people as being homeless. In the winter of 2008-2009, the statistical threshold was reached in December during the first cold spell, and again at the beginning of January during a period of severe cold. CONCLUSION: Our results support the use of this health indicator, alongside social indicators, for optimizing the strategy for helping the homeless population during winter.

Burden of infant bronchiolitis: data from a hospital network.

In France, surveillance of bronchiolitis is based on a pilot network of hospital emergency departments. The study was a 1-year observational study (2007-2008) carried out in a central region of metropolitan France. The hospitalization rate for bronchiolitis was 17.7/1000 children aged <1 year and the estimated prevalence of bronchiolitis ranged from 17.7% to 34.4% in children aged <1 year. Such a network constitutes a valuable tool to estimate the dynamic and the burden of infant bronchiolitis.

Performance of a syndromic system for influenza based on the activity of general practitioners, France.

BACKGROUND: In France, as in other industrialized countries, syndromic surveillance systems for the early detection of illnesses have proliferated, but few validation studies on these systems performances exist. In Bordeaux, a south-western city in France, a system using a network of general practitioners house calls, such as SOS Médecins, provided local health data used to guide health service response, in particular in case of flu-like pandemic. We explored the capacity of SOS Médecins system to identify and follow influenza outbreaks using data from the Sentinel network, considered as being a gold standard for tracking seasonal influenza in France. METHODS: Data from SOS Médecins were analysed and compared with data from the Sentinel network. The sensitivity and specificity of SOS Médecins system were evaluated for different simulated thresholds. RESULTS: A relationship between the number of visits for influenza from SOS Médecins and the number of influenza cases from the Sentinel network was observed; data from the two systems were highly correlated. We showed the capacity of SOS Médecins system to identify outbreaks with a sensitivity and specificity of 93%. CONCLUSION: The sensitivity and specificity of SOS Médecins for early outbreak detection showed the value of these data in monitoring influenza activity.

High-flow nasal cannula use for bronchiolitis treatment in French intensive care units: A transversal study.

Over the past 10 years, continuous positive airway pressure (CPAP) has revolutionized the prognosis and management of bronchiolitis patients hospitalized in pediatric intensive care units (PICUs). High-flow nasal cannula (HFNC) is emerging as an alternative to CPAP. Despite encouraging results of several clinical and physiological studies, HFNC use remains controversial and its indications heterogeneous. To better define the place of HFNC in severe bronchiolitis respiratory support, we investigated the different ventilation assistance techniques used for severe bronchiolitis over 3 days at the peak of a bronchiolitis epidemic in December 2015. We conducted an observational cross-sectional study in 27 French university hospital PICUs. Fifty-nine patients were included. The results show that HFNC already accounts for nearly half of the respiratory support techniques used for severe bronchiolitis in French PICUs with no significant difference between the CPAP group and the HFNC group of patients.

Validation of a syndromic surveillance system using a general practitioner house calls network, Bordeaux, France.

A new syndromic surveillance system has been developed in Bordeaux City, South West France, using a general practitioners' house calls network. Routinely collected, sociodemographic data, patients' complaints and medical diagnoses made at the end of the visit were monitored using syndrome groups such as influenza syndromes, bronchiolitis, gastrointestinal, respiratory syndromes and others, based on International Classification of Primary Care (ICPC)-2 codes. A process control chart was implemented in order to distinguish signals of interest from "background noise". In 2005 and 2006, a total of 303,936 visits were recorded. Seasonal epidemics of influenza-like illness, bronchiolitis or gastrointestinal were identified. The automated and real time nature of the system also allowed the early detection of unusual events such as an acute increase in the number of heat syndromes during the heat-wave that occurred in France in July 2006. This new system complements existing surveillance programs by assessing a large part of episodes of illness that do not require hospital admissions or the identification of an etiologic agent. Attributes and advantages of the system, such as timeliness and diagnostic specificity, demonstrated its utility and validity in term of syndromic surveillance purposes, and its extension at the national level is in process.

[Smoking in France: What is the situation?] / Le tabagisme en France : quelle est la situation ?

Tobacco consumption in France is still important. Several reasons explain this: a great proximity between the State and the tobacco industry which allows an important lobby up to the highest level of the State, valorization by the elites. Yet things change gradually. The image of tobacco is less positive and the epidemiology of consumption has changed since the early 2000s. While adult consumption remains around 30% of daily smokers, youth consumption is steadily decreasing (41% in 2000 vs. 25% in 2017). Another positive point is the decrease in the number of smokers among general practitioners, 1994: 34% vs. 16% in 2015. These points hide the current reality of smoking in France that kills 73,000 people/year and lung cancer rates that literally explode in women. In 20 years, the incidence of lung cancer in women has increased from 5/10,000 people/year to 12/10,000 people years making lung cancer in women the leading cause of cancer deaths from breast cancer. At the same time, men went from 50/10,000 to 40/10,000. Finally, this health impact is responsible for a major financial impact: the social cost of tobacco in France is estimated at € 120 billion a year. Faced with these figures, the French policy to fight against tobacco has been reinforced since 2016 with the adoption of the plain packet and a significant increase in the price (+30% in 2 years).

A satellite infrastructure for health early warning in post-disaster health management.

The risk of epidemics and emerging or re-emerging diseases such as avian flu, tuberculosis, malaria and other vector-borne diseases, is rising. These risks can be contained with prevention, early warning, and prompt management. Despite progress in information technology, communication is still a bottleneck for health early warning and response systems in post-disaster situations. This paper presents Satellites for Epidemiology (SAFE), a component-based interoperable architecture for health early warning that employs satellite, radio, and wireless networks, geographic information systems, integration technology, and data mining to promptly identify and respond to a disease outbreak. In a post-disaster situation, a mobile health emergency coordination center is established and integrated to public health services for health monitoring. The added-value of SAFE for post-disaster health management will be demonstrated as part of an earthquake readiness exercise regarding a typhoid fever epidemic, in the island of Crete. Advanced communication and data mining techniques in SAFE offer new tools to the "Epidemic Intelligence" and contribute to advanced preparedness and prompt response by lifting communication barriers, promoting collaboration, and reducing the isolation of affected areas.

Syndromic surveillance based on emergency department activity and crude mortality: two examples.

Recent public health crises have shown the need for readily available information allowing proper management by decision-makers. One way of obtaining early information is to involve data providers who already record routine data for their own use. We describe here the results of a pilot network carried out by the InVS (Institut national de veille sanitaire) which gathered data available in real time from hospital emergency departments and register offices. Emergency departments data were registered from patients' computerized medical files. Mortality data were received from the national institute of statistics (Insee). Data were transmitted automatically on a daily basis. Influenza data from outbreaks in 2004/05 and 2005/06 were compared with data from the sentinel network for the same periods. Environmental health data were compared with meteorological temperatures recorded in Paris between June and August 2006. A mortality analysis was conducted on a weekly basis. Correlation between influenza data from emergency departments and data from Sentiweb (sentinel network) was significant (p<0.001) for both outbreaks. In 2005 and 2006, the outbreaks were described similarly by both sources with identification of the start of the outbreaks by both systems during the same weeks. As for data related to heat, a significant correlation was observed between some diagnoses and temperature increases. For both types of phenomena, mortality increased significantly with one to two weeks lag. To our knowledge, this is the first time that a study using real time morbidity and mortality data is conducted. These initial results show how these data complement each other and how their simultaneous analysis in real time makes it possible to quickly measure the impact of a phenomenon.

Syndromic surveillance based on emergency department activity and crude mortality: two examples.

Recent public health crises have shown the need for readily available information allowing proper management by decision-makers. One way of obtaining early information is to involve data providers who already record routine data for their own use. We describe here the results of a pilot network carried out by the InVS (Institut national de veille sanitaire) which gathered data available in real time from hospital emergency departments and register offices. Emergency departments data were registered from patients' computerised medical files. Mortality data were received from the national institute of statistics (Insee). Data were transmitted automatically on a daily basis. Influenza data from outbreaks in 2004/05 and 2005/06 were compared with data from the sentinel network for the same periods. Environmental health data were compared with meteorological temperatures recorded in Paris between June and August 2006. A mortality analysis was conducted on a weekly basis. Correlation between influenza data from emergency departments and data from Sentiweb (sentinel network) was significant (p<0.001) for both outbreaks. In 2005 and 2006, the outbreaks were described similarly by both sources with identification of the start of the outbreaks by both systems during the same weeks. As for data related to heat, a significant correlation was observed between some diagnoses and temperature increases. For both types of phenomena, mortality increased significantly with one to two weeks lag. To our knowledge, this is the first time that a study using real time morbidity and mortality data is conducted. These initial results show how these data complement each other and how their simultaneous analysis in real time makes it possible to quickly measure the impact of a phenomenon.

Smoking in Maywood: the unhealthy southern migrant hypothesis among African Americans.

OBJECTIVES: This paper examines the "unhealthy Southern migrant" hypothesis with regard to cigarette smoking among African Americans. METHODS: Using data collected in 1992 from a sample of 1,518 African Americans in Maywood, Illinois, as part of the International Collaborative Study of Hypertension in Blacks, logistic regression analysis was conducted to examine and compare smoking behavior and sociodemographic characteristics of Southern and Midwestern-born respondents. RESULTS: African Americans born in the South were less likely (OR = .69, CI = 95%, 0.53, 0.90) to be smokers than those born in the Midwest, after controlling for other sociodemographic variables, prior smoking status, and age of arrival to Maywood. CONCLUSIONS: The results do not support the "unhealthy Southern migrant" hypothesis, with regard to cigarette smoking, and indicate the need to identify factors that protect Southern-born African-American migrants from smoking.

Smoking behavior and opinions of French general practitioners.

This report examines smoking prevalence, sociodemographic factors, and the opinions of French general practitioners (GPs) about tobacco control policies. Data from the CFES (Comité Français d'Education pour la Santé) national survey on general practitioners included 1013 respondents. The questionnaire was administered by telephone and a response rate of 65% was attained. Instrumentation included variables related to medical practice, sociodemographic characteristics, and opinions about health behavior. Thirty-four percent of physicians were current smokers. A higher proportion of males smoked compared to women (36.1% vs. 24.9%, p < 0.01), and they consumed on average more cigarettes per day (11.2 vs. 8 cigarettes/day, p < 0.05). Slightly more than 52% of physicians regarded their role in reducing nicotine addiction to be important. Doctors who believed that the physician's role was limited were less likely to advise pregnant women to stop smoking (odds ratio = 0.39, p < 0.001), and nonsmokers were more supportive of bans on smoking in public places. Despite the high prevalence of smoking among French physicians, they can still play an important role in reducing smoking among their patients. Medical school curriculum and continuing medical education programs focusing on prevention and cessation in France should be strengthened to help reduce smoking rates among physicians and the general population.

[History of continuing medical education in the United States]. / Histoire de la formation médicale continue aux Etats-Unis.

Continuing medical education came into being in the United States in the late twenties when the mediocrity of the initial medical training of practicing physicians was recognized. Medical schools created a classical system of continuing education; the first mandatory program was initiated in urology in 1934. By 1957, the first set of guidelines for good medical practice were published by the American Medical Association (AMA). The mandatory nature of continuing education was widespread by the end of the sixties with variable regulations from state to state. At the same time, the AMA created an honorary diploma for physicians who complete 150 h post-graduate training within 3 years. Starting in 1970, the political predominance of the AMA in continuing education was questioned by other professional associations (hospitals, medical schools). After much discussion and opposing debate, a common association for continuing education was created in 1981. The AMA remained a leader in the early nineties and now has started programs targeted to patients.

[Knowledge, opinions and tobacco consumption in a French faculty of medicine]. / Connaissances, opinions et consommation de tabac au sein d'une faculté de médecine française.

OBJECTIVE: Tobacco consumption among French physicians is regularly followed-up and well-known; however, there is little information concerning the smoking habits of medical students even though they are the physicians of the future. How do they behave? Do they smoke? More or less than the other populations of the same age? METHOD: All the students of a Parisian medical school were surveyed with a self-questionnaire completed and collected just before the exams at the end of the year. The questionnaire incorporated both past and present smoking habits and also the students' knowledge and opinions on tobacco consumption. RESULTS: A total of 681 students replied. More than one-third were smokers (34.6%) among which 21.0% smoked every day and 13.6% smoked occasionally. Gender had no influence on prevalence rate and both men and women smoked a comparable number of cigarettes per day (males 12.0 cig/day, females 10.4 cig/day). Eleven percent were former smokers and 68.4% would like to quit. Nearly 100% believed that cigarette smoke can bother others and 75% felt they were exemplary figures for others on the subject of tobacco use. Finally, two-thirds of the students smoked light cigarettes. CONCLUSION: The smoking habits of medical students are similar to those of the general population of the same age. It is necessary to develop specific prevention programs for medical students because they will play an important public health role in the future in reducing the prevalence of tobacco consumption in France.

[Fiscal revenues derived from tobacco sales to young people in France]. / Les revenus fiscaux issus de la vente de tabac aux jeunes en France.

This article examines the public revenues for France from tobacco sales to youths in 1997. The result is compared to the means of prevention the same year. Public revenues are estimated from: the number of adolescents divided into age groups (range: 12 to 19), the smoking rates by age group, the number of cigarettes smoked daily in each age group, the price of a cigarettes pack and the percentage of taxes. In 1997, almost 1.7 million youths were smokers and they smoked 240 million of cigarette boxes annually. This sale represented 4.7 billion of French Francs (FF): 3.615 billion of taxes, 380 million for tobacconists' and 762 million for the tobacco industry. The same year prevention represented less than FF 0.5 per year and inhabitant, one of the lowest ratios in developed countries.