Experience of establishing severe acute respiratory surveillance in the Netherlands: Evaluation and challenges.

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.

Severe acute respiratory infections surveillance for early signals in the community.

BACKGROUND: Surveillance of acute respiratory infections (ARI) in the Netherlands and other European countries is based mostly on primary care data, with little insight into the severe spectrum of the disease. We compared time-trends for ARI in secondary care with influenza-like illness (ILI), ARI and pneumonia in primary care, and crude mortality, in order to assess the value of routinely collected data on respiratory infections in hospitals and the added value of severe acute respiratory infections (SARI) surveillance. METHODS: We calculated incidence of ARI in secondary care, ILI, ARI, and pneumonia in primary care, and crude mortality using five historical databases (2008-2016). RESULTS: Over eight years, seasonal incidence peaks of ARI in secondary care occurred earlier than ILI and ARI incidence peaks in primary care, except during the 2009 influenza A(H1N1) pandemic and post-pandemic season. The median time-lag between ARI in secondary care and ILI, ARI and pneumonia in primary care was 6.5 weeks, 7 weeks, and 1 week, respectively. Crude mortality lagged a median 5 weeks behind ARI in secondary care. CONCLUSION: This observational study demonstrates that routinely collected data can be used for describing trends of ARI in secondary care and may be suitable for near real-time SARI surveillance. In most seasons, the incidence peaks for ARI in secondary care preceded the peaks in primary care and crude mortality with a considerable time-lag. It would be of great value to add microbiological test results to the incidence data to better explain the difference in time-lag between these surveillance systems.