key: cord-0921427-q4kpbocd authors: Marbus, S. D.; van der Hoek, W.; van Dissel, J. T.; van Gageldonk-Lafeber, A. B. title: Experience of establishing severe acute respiratory surveillance in the Netherlands: evaluation and challenges date: 2020-05-30 journal: Public health in practice DOI: 10.1016/j.puhip.2020.100014 sha: 0a4be9f048521079c802d449143d88bbb140c230 doc_id: 921427 cord_uid: q4kpbocd Abstract The 2009 influenza A(H1N1) pandemic prompted the World Health Organization (WHO) to recommend countries to establish a national severe acute respiratory infections (SARI) surveillance system for preparedness and emergency response. However, setting up or maintaining a robust SARI surveillance system has been challenging. Similar to other countries, surveillance data on hospitalisations for SARI in the Netherlands are still limited, in contrast to the robust surveillance data in primary care. The objective of this narrative review is to provide an overview, evaluation, and challenges of already available surveillance systems or datasets in the Netherlands, which might be used for near real-time surveillance of severe respiratory infections. Seven available surveillance systems or datasets in the Netherlands were reviewed. The evaluation criteria, including data quality, timeliness, representativeness, simplicity, flexibility, acceptability and stability were based on United States Centers for Disease Control and Prevention (CDC) and European Centre for Disease Prevention and Control (ECDC) guidelines for public health surveillance. We added sustainability as additional evaluation criterion. The best evaluated surveillance system or dataset currently available for SARI surveillance is crude mortality monitoring, although it lacks specificity. In contrast to influenza-like illness (ILI) in primary care, there is currently no gold standard for SARI surveillance in the Netherlands. Based on our experience with sentinel SARI surveillance, a fully or semi-automated, passive surveillance system seems most suited for a sustainable SARI surveillance system. An important future challenge remains integrating SARI surveillance into existing hospital programs in order to make surveillance data valuable for public health, as well as hospital quality of care management and individual patient care. *score: per evaluation criterion the following scores were attributed: 3 points if rated ''good", 2 points if rated ''moderate", 1 point if rated "poor". The total evaluation score was expressed as the sum of all individual evaluation criteria and percentage of the total maximum score. • Multiple surveillance systems or datasets are available in the Netherlands with potential use for SARI surveillance. • There is currently no gold standard for SARI surveillance in the Netherlands. • A potential sustainable SARI surveillance system for the long-term is a fully or semiautomated, passive surveillance system. • SARI surveillance data should be valuable for both public health and individual patient care. • An important future challenge remains integrating SARI surveillance into existing hospital programs. Surveillance is a vital tool to monitor shifts in the occurrence and burden of infectious diseases in the population, which is necessary for prevention and control. 1 In Europe, these are mainly eastern European countries, while in western Europe, Germany 12 and Belgium maintain a syndromic sentinel SARI surveillance system, which is complemented with influenza testing in Belgium. 13, 14 Since 2013, the United Kingdom (UK) publishes weekly national influenza reports, including incidence estimates of influenzaconfirmed hospitalisations. 15 Other European countries merely report the absolute number of influenza positive patients admitted to general wards or intensive care units (ICUs), i.e. without any denominator data. [16] [17] [18] Insight in the spectrum of surveillance systems and datasets that could potentially be used for SARI surveillance is scarce. In addition, studies which focus on challenges and main lessons learned for establishing a robust SARI surveillance system are lacking. In the Netherlands, several initiatives aim at setting up a severe infectious disease surveillance system. 19, 20 Our objective is to provide an overview, evaluation, and challenges of the available surveillance systems or datasets in the Netherlands, which could potentially be used for near real-time surveillance of severe respiratory infections. Our lessons learned could be valuable to other countries aiming to establish a robust SARI surveillance system. 19 In this pilot study different strategies were tested to assess which hospital data were best suited for a sustainable real-time SARI surveillance system. In JBH, an active, case-based surveillance system was set up, with registration by medical staff of any patient fulfilling the SARI case definition. A SARI case is defined as a hospitalised patient with at least one systemic symptom or deterioration of general condition and at least one respiratory symptom and symptoms started within a week from admission. In Leiden University Medical Center (LUMC), SARI surveillance was embedded in an automated cluster detection system, which was operational since 2013. 27 This passive, prospective, syndromic SARI surveillance system was based on financial claim codes corresponding to diagnoses related to the clinical syndrome SARI. These clinical syndromes include upper respiratory infections, lower respiratory infections and other respiratory infections The aggregated data were reported real-time by LUMC (catchment population 183,000 persons). Ambulance dispatch calls data could be used as syndromic data for an early warning system for respiratory infectious disease. 28 Examples of each surveillance system or dataset are given in supplemental file, figure S1-7. Available infectious disease surveillance systems or datasets in the Netherlands were reviewed for evaluation as a potential SARI surveillance system. The 4 authors assessed the The selected evaluation criteria, based on United States Centers for Disease Control and Prevention (CDC) and European Centre for Disease Prevention and Control (ECDC) guidelines for public health surveillance 30, 31 are: data quality: the completeness and validity of the data recorded in the surveillance system, including the addition of microbiological diagnostics; timeliness: the speed between steps in a surveillance system, from event occurrence, recognition, report, to control and prevention activities; representativeness: the ability to accurately describe the occurrence of an event over time, place and person; simplicity: the system's structure and ease of operation; flexibility: the ability to adapt to changing information needs or technological operating conditions; acceptability: the willingness of persons and organisations to participate in the system; stability: the system's reliability (ability to collect, manage and provide data without failure) and availability (ability to be operational when needed); We added one additional evaluation criterion: sustainability; the ongoing maintenance and support of a routine epidemiologic and/or microbiologic surveillance system; To asses each evaluation criterion, the 4 authors (2 infectious disease consultants, 1 medical doctor/epidemiologist, 1 senior epidemiologist) independently assigned the qualification 'good', 'moderate', or 'poor'. A semi-quantitative score for the surveillance system or dataset was obtained by attributing three points for each evaluation criterion rated ''good", two points for each evaluation criterion rated "moderate" and one point to each evaluation criterion rated "poor". If the opinions of the 4 experts diverged, the assessment was re-evaluated in order to reach consensus. The total evaluation score was calculated as the sum of 8 evaluation criteria scores. Descriptive statistics were used for reporting the score per surveillance system or dataset, such as total number and the percentage of the maximum score. Crude mortality monitoring scored the best if comparing the 7 surveillance systems or datasets based on the evaluation criteria ( Table 1 ). The SARI sentinel surveillance had the lowest score. In all surveillance systems or datasets data quality and sustainability was moderate to good. The largest contrast in in evaluation scores (poor versus good scores) between surveillance systems or datasets were seen for timeliness. The best evaluated surveillance system or dataset currently available for SARI surveillance is crude mortality monitoring, although it is still not sufficient. This system is well-established in the EU region with weekly country reports on the EuroMOMO and on the RIVM website. 26,32 However, crude mortality surveillance reports only all-cause mortality and therefore lacks specificity for SARI. Within the EuroMOMO network, models are now being developed to attribute mortality to influenza. 33 Crude mortality monitoring is also providing crucial data in the COVID-19 pandemic, in addition to the reported deaths from laboratoryconfirmed SARS-CoV-2 infection. If disease-specific mortality, such as respiratory mortality, could be reported, a more sensitive endpoint for SARI surveillance would be reached. However, it is not expected that cause-of-death statistics will become available near realtime in the foreseeable future. In ICUs a large amount of patient data is collected for quality assurance. In the Netherlands, these data are available in the NICE database and selected variables could provide a robust, syndromic SARI surveillance system, if timeliness could be improved together with maintaining current coverage. An exploratory query indicated that reporting frequency could be improved to every six weeks. However, to be better optimised for a SARI surveillance system for preparedness and emergency control, timeliness has to be improved to at least a weekly reporting frequency. During the current COVID-19 pandemic NICE was quickly modified for COVID-19 monitoring with several updates per day. The financial coding system comprised a passive, syndromic SARI surveillance system, which is evaluated as good for timeliness, simplicity and acceptability. Passive surveillance systems, such as fully automated cluster detection systems, are the preferred design for SARI surveillance, because they minimize administrative burden and increase sustainability. National register of hospital discharge diagnoses with specific ICD-10 codes related to respiratory infections are available with a one-year time lag, which precludes its use for SARI surveillance problematic. Efforts are underway to improve timeliness, which could make these data potentially valuable for SARI surveillance. In Germany, weekly SARI surveillance was established based on ICD-10 discharge codes from a large number of hospitals. 12 The virological laboratory surveillance, as it is available in the Netherlands and several other European countries, lacks linked-patient data, catchment population estimate, and distinction between primary-and secondary care, which makes it less suitable for use as SARI surveillance on its own. However, it could be of potential value in complementing another surveillance system or dataset, such as syndromic SARI sentinel surveillance. In comparison, Denmark has a national microbiology database available, with national legislation that allows for linkage with other patient data. 37 Established public health surveillance systems may be more extensively evaluated based on other surveillance system attributes, such as level of usefulness, sensitivity, and positive predictive value. 8 We chose a limited amount of evaluation criteria which are applicable and available for the current surveillance systems and datasets that could potentially be used for SARI surveillance in the Netherlands. In addition, costs for developing a SARI surveillance system are not included in this evaluation. With limited public health funding in many countries, this might be a critical first obstacle in setting up SARI surveillance. 8 Based on our experience and evaluation, improving sustainability is crucial for establishing a robust SARI surveillance system. In terms of sustainability, several challenges play an essential role. Firstly, the administrative burden associated with surveillance should be addressed. In a demanding hospital setting with increasing registration burden for hospital staff 45 , our experience from SARI sentinel surveillance indicated that additional workload associated with surveillance should be decreased as much as possible. Thus, to improve timeliness, simplicity, and acceptability of a SARI surveillance system, we believe that implementation of a passive, fully or semi-automated, SARI surveillance system is required. This is underlined by high scores evaluations scores for mortality monitoring and virological laboratory surveillance, which are largely automated surveillance systems as well. Secondly, a different appreciation of the value of epidemiological surveillance data by data providers, such as clinicians, laboratories or hospitals, should be taken into account. We experienced that stakeholders withdrew their participation in SARI surveillance after a year, because of different appreciation of the value of epidemiological surveillance data. Therefore, we believe it is essential that a SARI surveillance system serves both a public health and a patient care goal. 46 This could be achieved by integrating SARI surveillance in existing hospital programs in order to make surveillance data valuable for public health as well as patient care. 48 SARI surveillance data could for example be utilised for monitoring antibiotic or antiviral use and resistance and lead to targeted antibiotic stewardship programs (ASP) interventions in patient care. 49 Embedding SARI surveillance in a quality of care program for SARI patients is a strategy that was pursued in our SARI sentinel surveillance. 8, 50 Being part of routine quality care helped improve efficiency of our SARI sentinel surveillance system and increased the commitment of the participating hospital. Our aim is establishing a fully or semi-automated passive SARI surveillance system in the Netherlands based on financial codes. The advantages are the limited administrative burden and the data availability based on financial coding is (near)-real-time. Based on our experience with SARI sentinel surveillance, it is currently not possible to easily combine syndromic SARI data with microbiological diagnostics due to information communication technology (ICT) difficulties. Therefore, we aim to establish a separate laboratory surveillance system for influenza, RSV, S. pneumoniae and SARS-CoV-2, parallel to passive syndromic surveillance. In the long term, our goal is establishing an integrated, automated, passive SARI surveillance system with laboratory outcomes in sentinel hospitals evenly geographically distributed across the Netherlands. Multiple surveillance systems or datasets are available in the Netherlands with potential use for SARI surveillance. In contrast to ILI in primary care, there is currently no gold standard for SARI surveillance in the Netherlands. Based on our experience from sentinel SARI surveillance, a potential sustainable SARI surveillance system for the long-term is a fully or semi-automated, passive surveillance system. In addition to increased timeliness, and simplicity of the surveillance system, the acceptability is improved by reducing unnecessary administrative burden of hospital staff. An important future challenge remains integrating SARI surveillance into existing hospital programs in order to make surveillance data valuable for both public health and patient care. Planning a public health surveillance system Public health surveillance: historical origins, methods and evaluation Flu News Europe Epidemiological Surveillance Standards for Influenza Surveillance and studies in a pandemic: fourth meeting of the SSiaP working group Evaluation of syndromic surveillance in the Netherlands: its added value and recommendations for implementation. Euro surveillance : bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin WHO. WHO surveillance case definitions for ILI and SARI Surveillance for severe acute respiratory infections (SARI) in hospitals in the WHO European region -an exploratory analysis of risk factors for a severe outcome in influenza-positive SARI cases Characterizing the epidemiology, virology, and clinical features of influenza in China's first severe acute respiratory infection sentinel surveillance system Establishing an ICD-10 code based SARI-surveillance in Germany -description of the system and first results from five recent influenza seasons Virological surveillance of influenza in Belgium: season Sentinel Network of Hospitals Improving regional influenza surveillance through a combination of automated outbreak detection methods: the 2015/16 season in France. Euro surveillance : bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin Improving regional influenza surveillance through a combination of automated outbreak detection methods: the 2015/16 season in France. Euro surveillance : bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin Ernstige acute luchtweginfecties: de ontbrekende bouwsteen in de surveillancepiramide Annual report Surveillance of influenza and other respiratory infections in the Netherlands: winter Critical Care Scoring Systems Estimating severity of influenza epidemics from severe acute respiratory infections (SARI) in intensive care units Winter Simulation Conference; 2015. 29. RIVM. Virological surveillance data Updated Guidelines for Evaluating Public Health Surveillance Systems Data quality monitoring and surveillance system evaluation -A handbook of methods and application. Stockholm2014. 32. EuroMOMO. Mortality monitoring in Europe Influenza-associated mortality determined from allcause mortality Annual report surveillance of influenza and other respiratory infections in the Netherlands: winter 2017/20182018. 35. Epiconcept. I-MOVE+ Influenza surveillance in Europe: comparing intensity levels calculated using the moving epidemic method. Influenza and other respiratory viruses The burden of pneumococcal pneumonia -experience of the German competence network CAPNETZ Co-infection in severe influenza: a new epidemiology? Clinical infectious diseases : an official publication of the Infectious Diseases Society of America Prevalence of respiratory viruses among adults, by season, age, respiratory tract region and type of medical unit Clinical practice of respiratory virus diagnostics in critically ill patients with a suspected pneumonia: A prospective observational study Examining the Phenomenon of the Administrative Burden in Health Care, Allied Health, and Respiratory Care Syndromic surveillance: two decades experience of sustainable systems -its people not just data! Epidemiology and infection Steps to a sustainable public health surveillance enterprisea commentary from the international society for disease surveillance Challenges and opportunities: builiding sustainable surveillance systems Surveillance of Antibiotic Use and Resistance in Intensive Care Units (SARI) Quality of information: a European challenge. The Journal of hospital infection Liselotte van Asten, Elisa Benincà and Susana Monge for sharing data for this manuscript. Zurlyte for sharing information about their national SARI surveillance system. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The authors declare that they have no competing interests. ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: