Influenza epidemic surveillance and prediction based on electronic health record data from an out-of-hours general practitioner cooperative: model development and validation on 2003-2015 data

Construction of Influenza Early Warning Model Based on Combinatorial Judgment Classifier: A Case Study of Seasonal Influenza in Hong Kong

OBJECTIVE

The annual influenza epidemic is a heavy burden on the health care system, and has increasingly become a major public health problem in some areas, such as Hong Kong (China). Therefore, based on a variety of machine learning methods, and considering the seasonal influenza in Hong Kong, the study aims to establish a Combinatorial Judgment Classifier (CJC) model to classify the epidemic trend and improve the accuracy of influenza epidemic early warning.

METHODS

The characteristic variables were selected using the single-factor statistical method to establish the influencing factor system of an influenza outbreak. On this basis, the CJC model was proposed to provide an early warning for an influenza outbreak. The characteristic variables in the final model included atmospheric pressure, absolute maximum temperature, mean temperature, absolute minimum temperature, mean dew point temperature, the number of positive detections of seasonal influenza viruses, the positive percentage among all respiratory specimens, and the admission rates in public hospitals with a principal diagnosis of influenza.

RESULTS

The accuracy of the CJC model for the influenza outbreak trend reached 96.47%, the sensitivity and specificity change rates of this model were lower than those of other models. Hence, the CJC model has a more stable prediction performance. In the present study, the epidemic situation and meteorological data of Hong Kong in recent years were used as the research objects for the construction of the model index system, and a lag correlation was found between the influencing factors and influenza outbreak. However, some potential risk factors, such as geographical nature and human factors, were not incorporated, which ideally affected the prediction performance to some extent.

CONCLUSION

In general, the CJC model exhibits a statistically better performance, when compared to some classical early warning algorithms, such as Support Vector Machine, Discriminant Analysis, and Ensemble Classfiers, which improves the performance of the early warning of seasonal influenza.

Reorganisation of GP surgeries during the COVID-19 outbreak: analysis of guidelines from 15 countries

BACKGROUND

General practitioners (GPs) play a key role in managing the COVID-19 outbreak. However, they may encounter difficulties adapting their practices to the pandemic. We provide here an analysis of guidelines for the reorganisation of GP surgeries during the beginning of the pandemic from 15 countries.

METHODS

A network of GPs collaborated together in a three-step process: (i) identification of key recommendations of GP surgery reorganisation, according to WHO, CDC and health professional resources from health care facilities; (ii) collection of key recommendations included in the guidelines published in 15 countries; (iii) analysis, comparison and synthesis of the results.

RESULTS

Recommendations for the reorganisation of GP surgeries of four types were identified: (i) reorganisation of GP consultations (cancelation of non-urgent consultations, follow-up via e-consultations), (ii) reorganisation of GP surgeries (area partitioning, visual alerts and signs, strict hygiene measures), (iii) reorganisation of medical examinations by GPs (equipment, hygiene, partial clinical examinations, patient education), (iv) reorganisation of GP staff (equipment, management, meetings, collaboration with the local community).

CONCLUSIONS

We provide here an analysis of guidelines for the reorganisation of GP surgeries during the beginning of the COVID-19 outbreak from 15 countries. These guidelines focus principally on clinical care, with less attention paid to staff management, and the area of epidemiological surveillance and research is largely neglected. The differences of guidelines between countries and the difficulty to apply them in routine care, highlight the need of advanced research in primary care. Thereby, primary care would be able to provide recommendations adapted to the real-world settings and with stronger evidence, which is especially necessary during pandemics.

A method for estimating the transmissibility of influenza using serial cross-sectional seroepidemiological data

BACKGROUND

Seroepidemiological surveillance data has been demonstrated to be useful for estimating the cumulative incidence of influenza, and measures the difference between pre- and post-epidemic seropositive fractions. Despite this, such studies relied on a chosen cut-off value for seropositivity. The aim of the present study is to develop a method to analyze distributions of serial cross-sectional seroepidemiological surveillance datasets using an epidemiological model so that the transmission potential can be estimated without imposing a cut-off value.

METHODS

A mathematical model of influenza transmission with a discrete antibody titer level was constructed. The final size equation for pre- and post-epidemic titer levels was derived. Subsequently, using the estimated distribution of the dilution increase caused by infection and the measurement error distribution, the model parameters were optimized using the maximum likelihood method. A bootstrap-based confidence interval calculation and sensitivity analysis were also performed.

RESULTS

Without imposing a cut-off value, the cumulative incidence was quantified, thereby yielding an estimate of the basic reproduction number. For the purpose of exposition, the proposed method was applied to influenza A/Victoria/3/75(H3N2) data, and serological data between 1975 and 1976 were compared. The estimated reproduction number was greater than that using the cut-off value of the hemagglutination inhibition level with titer level 20 (dilution 1:20) or above to define positives.

CONCLUSION

The proposed method without a cut-off value offers an unbiased approach to estimating the cumulative incidence along with the reproduction number. If a cut-off value is required, the results imply that titer level 20 or above may better represent a reasonable cut-off value for calculating the incidence, but it could underestimate the basic reproduction number.

Influenza surveillance: determining the epidemic threshold for influenza by using the Moving Epidemic Method (MEM), Montenegro, 2010/11 to 2017/18 influenza seasons

Background: In 2009, an improved influenza surveillance system was implemented and weekly reporting to the World Health Organization on influenza-like illness (ILI) began. The goals of the surveillance system are to monitor and analyse the intensity of influenza activity, to provide timely information about circulating strains and to help in establishing preventive and control measures. In addition, the system is useful for comparative analysis of influenza data from Montenegro with other countries.

Aim: We aimed to evaluate the performance and usefulness of the Moving Epidemic Method (MEM), for use in the influenza surveillance system in Montenegro.

Methods: Historical ILI data from 2010/11 to 2017/18 influenza seasons were modelled with MEM. Epidemic threshold for Montenegro 2017/18 season was calculated using incidence rates from 2010/11–2016/17 influenza seasons.

Results: Pre-epidemic ILI threshold per 100,000 population was 19.23, while the post-epidemic threshold was 17.55. Using MEM, we identified an epidemic of 10 weeks’ duration. The sensitivity of the MEM epidemic threshold in Montenegro was 89% and the warning signal specificity was 99%.

Conclusions: Our study marks the first attempt to determine the pre/post-epidemic threshold values for the epidemic period in Montenegro. The findings will allow a more detailed examination of the influenza-related epidemiological situation, timely detection of epidemic and contribute to the development of more efficient measures for disease prevention and control aimed at reducing the influenza-associated morbidity and mortality.

Evaluation of the secondary use of electronic health records to detect seasonal, holiday-related, and rare events related to traumatic injury and poisoning

BACKGROUND

The increasing adoption of electronic health record (EHR) systems enables automated, large scale, and meaningful analysis of regional population health. We explored how EHR systems could inform surveillance of trauma-related emergency department visits arising from seasonal, holiday-related, and rare environmental events.

METHODS

We analyzed temporal variation in diagnosis codes over 24 years of trauma visit data at the three hospitals in the University of Washington Medicine system in Seattle, Washington, USA. We identified seasons and days in which specific codes and categories of codes were statistically enriched, meaning that a significantly greater than average proportion of trauma visits included a given diagnosis code during that time period.

RESULTS

We confirmed known seasonal patterns in emergency department visits for trauma. As expected, cold weather-related incidents (e.g. frostbite, snowboarding injury) were enriched in the winter, whereas fair weather-related incidents (e.g. bug bites, boating accidents, bicycle accidents) were enriched in the spring and summer. Our analysis of specific days of the year found that holidays were enriched for alcohol poisoning, assaults, and firework accidents. We also detected one time regional events such as the 2001 Nisqually earthquake and the 2006 Hanukkah Eve Windstorm.

CONCLUSIONS

Though EHR systems were developed to prioritize operational rather than analytic priorities and have consequent limitations for surveillance, our EHR enrichment analysis nonetheless re-identified expected temporal population health patterns. EHRs are potentially a valuable source of information to inform public health policy, both in retrospective analysis and in a surveillance capacity.

Comparison between influenza coded primary care consultations and national influenza incidence obtained by the General Practitioners Sentinel Network in Portugal from 2012 to 2017

Influenza is associated with severe illness, death, and economic burden. Sentinel surveillance systems have a central role in the community since they support public health interventions. This study aimed to describe and compare the influenza-coded primary care consultations with the reference index of influenza activity used in Portugal, General Practitioners Sentinel Network, from 2012 to 2017. An ecological time-series study was conducted using weekly R80-coded primary care consultations (according to the International Classification of Primary Care-2), weekly influenza-like illness (ILI) incidence rates from the General Practitioners Sentinel Network and Goldstein Index (GI). Good accordance between these three indicators was observed in the characterization of influenza activity regarding to start and length of the epidemic period, intensity of influenza activity, and influenza peak. A high correlation (>0.75) was obtained between weekly ILI incidence rates and weekly number of R80-coded primary care consultations during all five studied seasons. In 3 out of 5 seasons this correlation increased when weekly ILI incidence rates were multiplied for the percentage of influenza positive cases. A cross-correlation between weekly ILI incidence rates and the weekly number of R80-coded primary care consultations revealed that there was no lag between the rate curves of influenza incidence and the number of consultations in the 2012/13 and 2013/14 seasons. In the last three seasons, the weekly influenza incidence rates detected the influenza epidemic peak for about a week earlier. In the last season, the GI anticipated the detection of influenza peak for about a two-week period. Sentinel networks are fundamental elements in influenza surveillance that integrate clinical and virological data but often lack representativeness and are not able to provide regional and age groups estimates. Given the good correlation between weekly ILI incidence rate and weekly number of R80 consultations, primary care consultation coding system may be used to complement influenza surveillance data, namely, to monitor regional influenza activity. In the future, it would be interesting to analyse concurrent implementation of both surveillance systems with the integration of all available information.