Developing and validating a new national remote health advice syndromic surveillance system in England

Predicting Norovirus in England Using Existing and Emerging Syndromic Data: Infodemiology Study

BACKGROUND

Norovirus is associated with approximately 18% of the global burden of gastroenteritis and affects all age groups. There is currently no licensed vaccine or available antiviral treatment. However, well-designed early warning systems and forecasting can guide nonpharmaceutical approaches to norovirus infection prevention and control.

OBJECTIVE

This study evaluates the predictive power of existing syndromic surveillance data and emerging data sources, such as internet searches and Wikipedia page views, to predict norovirus activity across a range of age groups across England.

METHODS

We used existing syndromic surveillance and emerging syndromic data to predict laboratory data indicating norovirus activity. Two methods are used to evaluate the predictive potential of syndromic variables. First, the Granger causality framework was used to assess whether individual variables precede changes in norovirus laboratory reports in a given region or an age group. Then, we used random forest modeling to estimate the importance of each variable in the context of others with two methods: (1) change in the mean square error and (2) node purity. Finally, these results were combined into a visualization indicating the most influential predictors for norovirus laboratory reports in a specific age group and region.

RESULTS

Our results suggest that syndromic surveillance data include valuable predictors for norovirus laboratory reports in England. However, Wikipedia page views are less likely to provide prediction improvements on top of Google Trends and Existing Syndromic Data. Predictors displayed varying relevance across age groups and regions. For example, the random forest modeling based on selected existing and emerging syndromic variables explained 60% variance in the ≥65 years age group, 42% in the East of England, but only 13% in the South West region. Emerging data sets highlighted relative search volumes, including "flu symptoms," "norovirus in pregnancy," and norovirus activity in specific years, such as "norovirus 2016." Symptoms of vomiting and gastroenteritis in multiple age groups were identified as important predictors within existing data sources.

CONCLUSIONS

Existing and emerging data sources can help predict norovirus activity in England in some age groups and geographic regions, particularly, predictors concerning vomiting, gastroenteritis, and norovirus in the vulnerable populations and historical terms such as stomach flu. However, syndromic predictors were less relevant in some age groups and regions likely due to contrasting public health practices between regions and health information-seeking behavior between age groups. Additionally, predictors relevant to one norovirus season may not contribute to other seasons. Data biases, such as low spatial granularity in Google Trends and especially in Wikipedia data, also play a role in the results. Moreover, internet searches can provide insight into mental models, that is, an individual's conceptual understanding of norovirus infection and transmission, which could be used in public health communication strategies.

The Utility of Ambulance Dispatch Call Syndromic Surveillance for Detecting and Assessing the Health Impact of Extreme Weather Events in England

Extreme weather events present significant global threats to health. The National Ambulance Syndromic Surveillance System collects data on 18 syndromes through chief presenting complaint (CPC) codes. We aimed to determine the utility of ambulance data to monitor extreme temperature events for action. Daily total calls were observed between 01/01/2018−30/04/2019. Median daily ’Heat/Cold’ CPC calls during “known extreme temperature” (identified a priori), “extreme temperature”; (within 5th or 95th temperature percentiles for central England) and meteorological alert periods were compared to all other days using Wilcoxon signed-rank test. During the study period, 12,585,084 calls were recorded. In 2018, median daily “Heat/Cold” calls were higher during periods of known extreme temperature: heatwave (16/day, 736 total) and extreme cold weather events (28/day, 339 total) compared to all other days in 2018 (6/day, 1672 total). Median daily “Heat/Cold” calls during extreme temperature periods (16/day) were significantly higher than non-extreme temperature periods (5/day, p < 0.001). Ambulance data can be used to identify adverse impacts during periods of extreme temperature. Ambulance data are a low resource, rapid and flexible option providing real-time data on a range of indicators. We recommend ambulance data are used for the surveillance of presentations to healthcare related to extreme temperature events.

Describing the indirect impact of COVID-19 on healthcare utilisation using syndromic surveillance systems

Background

Since the end of January 2020, the coronavirus (COVID-19) pandemic has been responsible for a global health crisis. In England a number of non-pharmaceutical interventions have been introduced throughout the pandemic, including guidelines on healthcare attendance (for example, promoting remote consultations), increased handwashing and social distancing. These interventions are likely to have impacted the incidence of non–COVID-19 conditions as well as healthcare seeking behaviour. Syndromic Surveillance Systems offer the ability to monitor trends in healthcare usage over time.

Methods

This study describes the indirect impact of COVID-19 on healthcare utilisation using a range of syndromic indicators including eye conditions, mumps, fractures, herpes zoster and cardiac conditions. Data from the syndromic surveillance systems monitored by Public Health England were used to describe the number of contacts with NHS 111, general practitioner (GP) In Hours (GPIH) and Out-of-Hours (GPOOH), Ambulance and Emergency Department (ED) services over comparable periods before and during the pandemic.

Results

The peak pandemic period in 2020 (weeks 13–20), compared to the same period in 2019, displayed on average a 12% increase in NHS 111 calls, an 11% decrease in GPOOH consultations, and a 49% decrease in ED attendances. In the GP In Hours system, conjunctivitis consultations decreased by 64% and mumps consultations by 31%. There was a 49% reduction in attendance at EDs for fractures, and there was no longer any weekend increase in ED fracture attendances, with similar attendance patterns observed across each day of the week. There was a decrease in the number of ED attendances with diagnoses of myocardial ischaemia.

Conclusion

The COVID-19 pandemic drastically impacted healthcare utilisation for non-COVID-19 conditions, due to a combination of a probable decrease in incidence of certain conditions and changes in healthcare seeking behaviour. Syndromic surveillance has a valuable role in describing and understanding these trends.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-021-12117-5.

Adapting Syndromic Surveillance Baselines After Public Health Interventions

BACKGROUND

Public health surveillance requires historical baselines to identify unusual activity. However, these baselines require adjustment after public health interventions. We describe an example of such an adjustment after the introduction of rotavirus vaccine in England in July 2013.

METHODS

We retrospectively measured the magnitude of differences between baselines and observed counts (residuals) before and after the introduction of a public health intervention, the introduction of a rotavirus vaccine in July 2013. We considered gastroenteritis, diarrhea, and vomiting to be indicators for national syndromic surveillance, including telephone calls to a telehealth system, emergency department visits, and unscheduled consultations with general practitioners. The start of the preintervention period varied depending on the availability of surveillance data: June 2005 for telehealth, November 2009 for emergency departments, and July 2010 for general practitioner data. The postintervention period was July 2013 to the second quarter of 2016. We then determined whether baselines incorporating a step-change reduction or a change in seasonality resulted in more accurate models of activity.

RESULTS

Residuals in the unadjusted baseline models increased by 42%-198% from preintervention to postintervention. Increases in residuals for vomiting indicators were 19%-44% higher than for diarrhea. Both step-change and seasonality adjustments improved the surveillance models; we found the greatest reduction in residuals in seasonally adjusted models (4%-75%).

CONCLUSION

Our results demonstrated the importance of adjusting surveillance baselines after public health interventions, particularly accounting for changes in seasonality. Adjusted baselines produced more representative expected values than did unadjusted baselines, resulting in fewer false alarms and a greater likelihood of detecting public health threats.

Utility of emergency call centre, dispatch and ambulance data for syndromic surveillance of infectious diseases: a scoping review

BACKGROUND

Syndromic surveillance can supplement conventional health surveillance by analyzing less-specific, near-real-time data for an indication of disease occurrence. Emergency medical call centre dispatch and ambulance data are examples of routinely and efficiently collected syndromic data that might assist in infectious disease surveillance. Scientific literature on the subject is scarce and an overview of results is lacking.

METHODS

A scoping review including (i) review of the peer-reviewed literature, (ii) review of grey literature and (iii) interviews with key informants.

RESULTS

Forty-four records were selected: 20 peer reviewed and 24 grey publications describing 44 studies and systems. Most publications focused on detecting respiratory illnesses or on outbreak detection at mass gatherings. Most used retrospective data; some described outcomes of temporary systems; only two described continuously active dispatch- and ambulance-based syndromic surveillance. Key informants interviewed valued dispatch- and ambulance-based syndromic surveillance as a potentially useful addition to infectious disease surveillance. Perceived benefits were its potential timeliness, standardization of data and clinical value of the data.

CONCLUSIONS

Various dispatch- and ambulance-based syndromic surveillance systems for infectious diseases have been reported, although only roughly half are documented in peer-reviewed literature and most concerned retrospective research instead of continuously active surveillance systems. Dispatch- and ambulance-based syndromic data were mostly assessed in relation to respiratory illnesses; reported use for other infectious disease syndromes is limited. They are perceived by experts in the field of emergency surveillance to achieve time gains in detection of infectious disease outbreaks and to provide a useful addition to traditional surveillance efforts.

Comparison of statistical algorithms for daily syndromic surveillance aberration detection

Motivation

Public health authorities can provide more effective and timely interventions to protect populations during health events if they have effective multi-purpose surveillance systems. These systems rely on aberration detection algorithms to identify potential threats within large datasets. Ensuring the algorithms are sensitive, specific and timely is crucial for protecting public health. Here, we evaluate the performance of three detection algorithms extensively used for syndromic surveillance: the ‘rising activity, multilevel mixed effects, indicator emphasis’ (RAMMIE) method and the improved quasi-Poisson regression-based method known as ‘Farrington Flexible’ both currently used at Public Health England, and the ‘Early Aberration Reporting System’ (EARS) method used at the US Centre for Disease Control and Prevention. We model the wide range of data structures encountered within the daily syndromic surveillance systems used by PHE. We undertake extensive simulations to identify which algorithms work best across different types of syndromes and different outbreak sizes. We evaluate RAMMIE for the first time since its introduction. Performance metrics were computed and compared in the presence of a range of simulated outbreak types that were added to baseline data.

Results

We conclude that amongst the algorithm variants that have a high specificity (i.e. >90%), Farrington Flexible has the highest sensitivity and specificity, whereas RAMMIE has the highest probability of outbreak detection and is the most timely, typically detecting outbreaks 2–3 days earlier.

Availability and implementation

R codes developed for this project are available through https://github.com/FelipeJColon/AlgorithmComparison

Supplementary information

Supplementary data are available at Bioinformatics online.

Developing influenza and respiratory syncytial virus activity thresholds for syndromic surveillance in England

Influenza and respiratory syncytial virus (RSV) are common causes of respiratory tract infections and place a burden on health services each winter. Systems to describe the timing and intensity of such activity will improve the public health response and deployment of interventions to these pressures. Here we develop early warning and activity intensity thresholds for monitoring influenza and RSV using two novel data sources: general practitioner out-of-hours consultations (GP OOH) and telehealth calls (NHS 111). Moving Epidemic Method (MEM) thresholds were developed for winter 2017-2018. The NHS 111 cold/flu threshold was breached several weeks in advance of other systems. The NHS 111 RSV epidemic threshold was breached in week 41, in advance of RSV laboratory reporting. Combining the use of MEM thresholds with daily monitoring of NHS 111 and GP OOH syndromic surveillance systems provides the potential to alert to threshold breaches in real-time. An advantage of using thresholds across different health systems is the ability to capture a range of healthcare-seeking behaviour, which may reflect differences in disease severity. This study also provides a quantifiable measure of seasonal RSV activity, which contributes to our understanding of RSV activity in advance of the potential introduction of new RSV vaccines.

Can syndromic surveillance help forecast winter hospital bed pressures in England?

Background

Health care planners need to predict demand for hospital beds to avoid deterioration in health care. Seasonal demand can be affected by respiratory illnesses which in England are monitored using syndromic surveillance systems. Therefore, we investigated the relationship between syndromic data and daily emergency hospital admissions.

Methods

We compared the timing of peaks in syndromic respiratory indicators and emergency hospital admissions, between 2013 and 2018. Furthermore, we created forecasts for daily admissions and investigated their accuracy when real-time syndromic data were included.

Results

We found that syndromic indicators were sensitive to changes in the timing of peaks in seasonal disease, especially influenza. However, each year, peak demand for hospital beds occurred on either 29^th or 30^th December, irrespective of the timing of syndromic peaks. Most forecast models using syndromic indicators explained over 70% of the seasonal variation in admissions (adjusted R square value). Forecast errors were reduced when syndromic data were included. For example, peak admissions for December 2014 and 2017 were underestimated when syndromic data were not used in models.

Conclusion

Due to the lack of variability in the timing of the highest seasonal peak in hospital admissions, syndromic surveillance data do not provide additional early warning of timing. However, during atypical seasons syndromic data did improve the accuracy of forecast intensity.

Machine learning to refine decision making within a syndromic surveillance service

Background

Worldwide, syndromic surveillance is increasingly used for improved and timely situational awareness and early identification of public health threats. Syndromic data streams are fed into detection algorithms, which produce statistical alarms highlighting potential activity of public health importance. All alarms must be assessed to confirm whether they are of public health importance. In England, approximately 100 alarms are generated daily and, although their analysis is formalised through a risk assessment process, the process requires notable time, training, and maintenance of an expertise base to determine which alarms are of public health importance. The process is made more complicated by the observation that only 0.1% of statistical alarms are deemed to be of public health importance. Therefore, the aims of this study were to evaluate machine learning as a tool for computer-assisted human decision-making when assessing statistical alarms.

Methods

A record of the risk assessment process was obtained from Public Health England for all 67,505 statistical alarms between August 2013 and October 2015. This record contained information on the characteristics of the alarm (e.g. size, location). We used three Bayesian classifiers- naïve Bayes, tree-augmented naïve Bayes and Multinets - to examine the risk assessment record in England with respect to the final ‘Decision’ outcome made by an epidemiologist of ‘Alert’, ‘Monitor’ or ‘No-action’. Two further classifications based upon tree-augmented naïve Bayes and Multinets were implemented to account for the predominance of ‘No-action’ outcomes.

Results

The attributes of each individual risk assessment were linked to the final decision made by an epidemiologist, providing confidence in the current process. The naïve Bayesian classifier performed best, correctly classifying 51.5% of ‘Alert’ outcomes. If the ‘Alert’ and ‘Monitor’ actions are combined then performance increases to 82.6% correctly classified. We demonstrate how a decision support system based upon a naïve Bayes classifier could be operationalised within an operational syndromic surveillance system.

Conclusions

Within syndromic surveillance systems, machine learning techniques have the potential to make risk assessment following statistical alarms more automated, robust, and rigorous. However, our results also highlight the importance of specialist human input to the process.

Social patterning of telephone health-advice for diarrhoea and vomiting: analysis of 24 million telehealth calls in England

OBJECTIVES

Gastrointestinal (GI) infections are common and most people do not see a physician. There is conflicting evidence of the impact of socioeconomic status (SES) on risk of GI infections. We assessed the relationship between SES and GI calls to two National Health Service (NHS) telephone advice services in England.

METHODS

Over 24 million calls to NHS Direct (2010-13) and NHS 111 (2013-15) were extracted from Public Health England (PHE) syndromic surveillance systems. The relationship between SES and GI calls was assessed using generalised linear models (GLM).

RESULTS

Adjusting for rurality and age-sex interactions, in NHS Direct, children in disadvantaged areas were at lower risk of GI calls; in NHS 111 there was a higher risk of GI calls in disadvantaged areas for all ages (0-4 years RR 1.27, 95% CI 1.25-1.29; 5-9 years RR 1.43, 95% CI 1.36-1.51; 10-14 years RR 1.36, 95% CI 1.26-1.41; 15-19 years RR 1.59, 95% CI 1.52-1.67; 20-59 years RR 1.50, 95% CI 1.47-1.53, 60 years and over RR 1.12, 95% CI 1.09-1.14).

CONCLUSIONS

Disadvantaged areas had higher risk of GI calls in NHS 111. This may relate to differences in exposure or vulnerability to GI infections, or propensity to call about GI infections.

Retrospective Observational Study of Atypical Winter Respiratory Illness Season Using Real-Time Syndromic Surveillance, England, 2014-15

During winter 2014–15, England experienced severe strains on acute health services. We investigated whether syndromic surveillance could contribute to understanding of the unusually high level of healthcare needs. We compared trends for several respiratory syndromic indicators from that winter to historical baselines. Cumulative and mean incidence rates were compared by winter and age group. All-age influenza-like illness was at expected levels; however, severe asthma and pneumonia levels were above those expected. Across several respiratory indicators, cumulative incidence rates during 2014–15 were similar to those of previous years, but higher for older persons; we saw increased rates of acute respiratory disease, including influenza like illness, severe asthma, and pneumonia, in the 65–74- and >75-year age groups. Age group–specific statistical algorithms may provide insights into the burden on health services and improve early warning in future winters.

A methodological framework for the evaluation of syndromic surveillance systems: a case study of England

BACKGROUND

Syndromic surveillance complements traditional public health surveillance by collecting and analysing health indicators in near real time. The rationale of syndromic surveillance is that it may detect health threats faster than traditional surveillance systems permitting more timely, and hence potentially more effective public health action. The effectiveness of syndromic surveillance largely relies on the methods used to detect aberrations. Very few studies have evaluated the performance of syndromic surveillance systems and consequently little is known about the types of events that such systems can and cannot detect.

METHODS

We introduce a framework for the evaluation of syndromic surveillance systems that can be used in any setting based upon the use of simulated scenarios. For a range of scenarios this allows the time and probability of detection to be determined and uncertainty is fully incorporated. In addition, we demonstrate how such a framework can model the benefits of increases in the number of centres reporting syndromic data and also determine the minimum size of outbreaks that can or cannot be detected. Here, we demonstrate its utility using simulations of national influenza outbreaks and localised outbreaks of cryptosporidiosis.

RESULTS

Influenza outbreaks are consistently detected with larger outbreaks being detected in a more timely manner. Small cryptosporidiosis outbreaks (<1000 symptomatic individuals) are unlikely to be detected. We also demonstrate the advantages of having multiple syndromic data streams (e.g. emergency attendance data, telephone helpline data, general practice consultation data) as different streams are able to detect different outbreak types with different efficacy (e.g. emergency attendance data are useful for the detection of pandemic influenza but not for outbreaks of cryptosporidiosis). We also highlight that for any one disease, the utility of data streams may vary geographically, and that the detection ability of syndromic surveillance varies seasonally (e.g. an influenza outbreak starting in July is detected sooner than one starting later in the year). We argue that our framework constitutes a useful tool for public health emergency preparedness in multiple settings.

CONCLUSIONS

The proposed framework allows the exhaustive evaluation of any syndromic surveillance system and constitutes a useful tool for emergency preparedness and response.

Impact of extreme weather events and climate change for health and social care systems

This review, commissioned by the Research Councils UK Living With Environmental Change (LWEC) programme, concerns research on the impacts on health and social care systems in the United Kingdom of extreme weather events, under conditions of climate change. Extreme weather events considered include heatwaves, coldwaves and flooding. Using a structured review method, we consider evidence regarding the currently observed and anticipated future impacts of extreme weather on health and social care systems and the potential of preparedness and adaptation measures that may enhance resilience. We highlight a number of general conclusions which are likely to be of international relevance, although the review focussed on the situation in the UK. Extreme weather events impact the operation of health services through the effects on built, social and institutional infrastructures which support health and health care, and also because of changes in service demand as extreme weather impacts on human health. Strategic planning for extreme weather and impacts on the care system should be sensitive to within country variations. Adaptation will require changes to built infrastructure systems (including transport and utilities as well as individual care facilities) and also to institutional and social infrastructure supporting the health care system. Care sector organisations, communities and individuals need to adapt their practices to improve resilience of health and health care to extreme weather. Preparedness and emergency response strategies call for action extending beyond the emergency response services, to include health and social care providers more generally.

Correcting for day of the week and public holiday effects: improving a national daily syndromic surveillance service for detecting public health threats

BACKGROUND

As service provision and patient behaviour varies by day, healthcare data used for public health surveillance can exhibit large day of the week effects. These regular effects are further complicated by the impact of public holidays. Real-time syndromic surveillance requires the daily analysis of a range of healthcare data sources, including family doctor consultations (called general practitioners, or GPs, in the UK). Failure to adjust for such reporting biases during analysis of syndromic GP surveillance data could lead to misinterpretations including false alarms or delays in the detection of outbreaks. The simplest smoothing method to remove a day of the week effect from daily time series data is a 7-day moving average. Public Health England developed the working day moving average in an attempt also to remove public holiday effects from daily GP data. However, neither of these methods adequately account for the combination of day of the week and public holiday effects.

METHODS

The extended working day moving average was developed. This is a further data-driven method for adding a smooth trend curve to a time series graph of daily healthcare data, that aims to take both public holiday and day of the week effects into account. It is based on the assumption that the number of people seeking healthcare services is a combination of illness levels/severity and the ability or desire of patients to seek healthcare each day. The extended working day moving average was compared to the seven-day and working day moving averages through application to data from two syndromic indicators from the GP in-hours syndromic surveillance system managed by Public Health England.

RESULTS

The extended working day moving average successfully smoothed the syndromic healthcare data by taking into account the combined day of the week and public holiday effects. In comparison, the seven-day and working day moving averages were unable to account for all these effects, which led to misleading smoothing curves.

CONCLUSIONS

The results from this study make it possible to identify trends and unusual activity in syndromic surveillance data from GP services in real-time independently of the effects caused by day of the week and public holidays, thereby improving the public health action resulting from the analysis of these data.

Early outbreak detection by linking health advice line calls to water distribution areas retrospectively demonstrated in a large waterborne outbreak of cryptosporidiosis in Sweden

Background

In the winter and spring of 2011 a large outbreak of cryptosporidiosis occurred in Skellefteå municipality, Sweden. This study summarizes the outbreak investigation in terms of outbreak size, duration, clinical characteristics, possible source(s) and the potential for earlier detection using calls to a health advice line.

Methods

The investigation included two epidemiological questionnaires and microbial analysis of samples from patients, water and other environmental sources. In addition, a retrospective study based on phone calls to a health advice line was performed by comparing patterns of phone calls between different water distribution areas.

Results

Our analyses showed that approximately 18,500 individuals were affected by a waterborne outbreak of cryptosporidiosis in Skellefteå in 2011. This makes it the second largest outbreak of cryptosporidiosis in Europe to date. Cryptosporidium hominis oocysts of subtype IbA10G2 were found in patient and sewage samples, but not in raw water or in drinking water, and the initial contamination source could not be determined. The outbreak went unnoticed to authorities for several months. The analysis of the calls to the health advice line provides strong indications early in the outbreak that it was linked to a particular water treatment plant.

Conclusions

We conclude that an earlier detection of the outbreak by linking calls to a health advice line to water distribution areas could have limited the outbreak substantially.

Electronic supplementary material

The online version of this article (doi:10.1186/s12889-017-4233-8) contains supplementary material, which is available to authorized users.