(i) Resources: Ensure capacity-building before the pandemic hits
The analysis of our data shows that HIS grounded on sufficient capacity outside of pandemics are able to expand and create strong surveillance systems during crises. These include as key elements a legal basis for all aspects concerning HIS, an adequate number of health and care staff, and the necessary infrastructure with regard to health care institutions and ICT infrastructure. The experience with the COVID-19 pandemic in European countries shows that capacity-building should take place before rather than during a pandemic. Sustainability in funding of routine data collection mechanisms, automated communication flows between institutions and government levels are key elements, as well as compliance with data protection laws. In this context, governance structures and governance documents (e.g. an HIS strategy or written protocols on patient safety and hygiene) also play a role. The capacity-building refers mainly to direct health information, i.e. dimension (1) of our conceptual framework, but could also extend to the holistic dimension (2) (cf. Table 2). For instance, financing of health information infrastructure and data linkage are a precondition for ensuring sufficient capacity, as is availability of trained legal experts to deal with questions related to data protection. Sufficient resources are a key for faring strongly on the domains mentioned in Table 1 (legislation, staff, and infrastructure) in order to ensure strong HIS in times of crisis or a pandemic.
(ii) Indicators: Collect longitudinal information on vulnerable groups
Monitoring the wider effects of COVID-19 requires longitudinal data from health systems and surveys with reference points from before the time of the pandemic, using validated questions. Building on high-quality data allows a broad understanding both of the drivers of pandemics and the impact of pandemics on health and non-health related aspects. For example, in Belgium, COVID-19 specific health surveys, which can be benchmarked with previous surveys, are organised. The Belgian COVID-19 health surveys help scientists and policymakers to estimate the impact of the COVID-19 crisis on health. They allow mapping trends and studying various public health topics. Attention is given to physical, mental and social well-being, and the use of health services. As another example, the Finnish Institute for Health and Welfare (THL) prepared a rapid impact assessment about the effects of COVID-19 epidemic on the population's service needs, the service system and the economy. As part of a sero-prevalence study, a short web-based questionnaire was administered to learn more about quality of life, functional capacity, use of health and social care services for non-COVID-19-related health issues (e.g. chronic conditions), smoking or alcohol use, to name a few. In addition to surveys, a participatory process is desirable in which users (e.g. from the research community) are involved in evaluating the usefulness of the indicators collected, and in identifying potential gaps.
For example, and relating to the holistic level of health information [see (2) in Table 2], it has become apparent that data on certain socio-economic groups may be of high relevance, more so during a pandemic. Socially vulnerable groups (e.g. homeless people, drug users, sex workers) need to be monitored closely, as precarious working and living conditions have shown to foster the occurrence of outbreaks of COVID-19 (Takian et al., 2020; Frazer, 2020). Administrative data (e.g. registries) that allow for identification of specific social settings or socio-economic characteristics of the population, in addition to health-related characteristics, are particularly valuable in this context. Countries with a health-in-all-policy approach that is reflected in the data collected and reported, are in a better position to monitor the health of the total population with specific attention to the situation of vulnerable groups.
Relating to direct health information (level (1) in Table 2), people in need of long-term care are another vulnerable group in the current pandemic, as a large share of those most at risk of dying from COVID-19 are older people and/or live in care homes (Comas-Herrera et al., 2020). Countries with HIS that integrate information from the health and the long-term care sector are likely better able to respond to the crisis. Furthermore, other vulnerable groups are those with specific health conditions associated with lower socio-economic conditions that may increase the risk for severe COVID-19 disease. For example, people with comorbidities (e.g. obesity) have been shown to be at risk of experiencing severe COVID-19, with prevalence of obesity being significantly elevated in lower socio-economic groups. In combination with a lack of health literacy often associated with lower socio-economic status, people at risk may not always realise they are. It is therefore crucial that HIS collect information on the general population including the people with co-morbidities who are not necessarily seeking medical care for these conditions. In general, countries with a focus on using routine mechanisms for data flow may be in a better position to monitor specific groups, e.g. flows between administrative data and vital statistics, flows across settings such as hospitals, laboratories and care homes.
(iii) Data sources: Establish strong reporting systems on staffing numbers and in primary care
Referring to direct health information (level (1) in Table 1) our analysis shows that countries which have comprehensive and complete reporting systems for staffing numbers in hospitals and care homes may be better able to manage and create separate teams. Equally, going beyond direct health information but more towards a holistic approach (level (2) in Table 1) HIS that systematically collect information on available materials (e.g. in hospitals) including medication and collaborate with other sectors (e.g. trade, economic sector), are in a better position to respond to the crisis, and clarify logistical questions more rapidly than other countries.
As a second point to consider under level (1) relating to direct health information, countries with sufficient capacities in primary health care (incl. documentation systems) will be better able to monitor suspicious cases for COVID-19 in the phases following the peak. Countries like France or Spain monitor people with COVID-19 symptoms from the onset, including collection of information on time needed between occurrence of first symptoms until testing and/or diagnosis. In addition, where health literacy of primary health care providers is strong, they may be better able to communicate potential risks during the pandemic to their patients, or reach out to patients at risk proactively. Similarly, countries with the respective reporting systems to monitor suspicious cases (also e.g. via inbound telephone hotlines or in primary care) have been able to develop early warning systems (e.g. France, Spain) that may help prevent future waves of the pandemic. Similarly, the capacity to collect mortality data varies across countries depending on the ability to test. This would require logistic capacity and information systems that link mortality reporting from different sources (within countries and internationally).
(iv) Data management: Link data sources to identify unmet needs for essential health care timely
It is likely that during the peak of the crisis, demand for essential health care may have decreased. People may postpone medical care, including the use of preventive services thus making adequate and reliable data management a crucial element. For instance, a crowding out of essential, regular health services is likely to have taken place in many countries (see e.g. Desai et al. 2020, Metzler et al. 2020). Relating to strengthening direct health information (level (1) in Table 1), foundations for excellent data management procedures need to be laid before the outbreak of a pandemic. Trust relationships with data providers need to already be established, making data providers more likely to be engaged and, for instance, ready to put in the extra time during a crisis. Also, data managers will already be comfortable with the software tools in use. Investing in training staff on a regular bases before a crisis, will ensure that staff can process data more easily when a crisis hits.
Countries with strong HIS (direct health information, cf. level (1) in Table 1) allow for detailed reporting on health care use in all settings, including ambulatory care, and reporting on continuity of care between primary and secondary care. These are likely to be in a better position to build up capacities for maintaining and managing regular health services (not directly related to the pandemic) during a crisis. WHO recommends an “expanded (dual) dashboard of service coverage and delivery indicators and the use of key tracer indicators on utilization patterns and mortality on both COVID-19 and non-COVID-19 conditions to manage a dual-track health system” (WHO, 2020:8). In the Netherlands, foresight studies are conducted, in which scenarios are developed to provide insight into longer-term health impacts, accounting for the current effect of the pandemic. Also, in the Czech Republic, a comprehensive COVID-19 monitoring system has been established, including information on COVID-19 confirmed cases, their medical history, general inpatient / intensive care, and outcome, ultimately allowing to monitor not only status of the epidemic, but also potential long-term health outcomes of recovered COVID-19 patients. In general, coordination of data collection across sectors, e.g. via a unique identifier or a centralised data storage platform for all health care sectors may be a crucial element in handling the pandemic, as it is done for instance in Denmark or Norway. In addition, data quality checks are important too, e.g. when it comes to doing cross-checks and taking out duplicates, thus generating reliable statistics (e.g. on COVID-19-cause mortality).
(v) Information products: Introduce room for innovation and digitalisation
According to the country experiences in our data, having the legal basis clarified for the use of video or phone consultations beforehand enabled some countries to rapidly reduce human contacts between patients and doctors substantially during the pandemic. Also, countries where innovation - in both direct health information systems and those not directly linked to health - is incentivised, specifically funded or promoted, were better able to develop flexible HIS to deal with the crisis (incl. telemedicine solutions). As an example, Austria had laid the foundations for e-prescriptions already before the crisis and was therefore able to facilitate implementation during the pandemic.
The crisis has also shown the potential for the use of mobile services and digital technologies in the health sector, as well as difficulties in harmonised data protection legalisation across EU countries. In fact, many countries have seen limited use e.g. of contact tracing apps, as they did not comply with the EU’s general data protection regulation (GDPR). Coordinated efforts at EU level may bear some potential in this respect, e.g. an app developed by the EC (DG SANTE). Contact tracing may however be context-specific and bear limited success in European countries as compared for instance to Asian countries. Also, non-digital solutions may be found that can be highly innovative. For example, in Wales, a collaboration between academia public health agency, health service and the Government was created to respond to the COVID19 challenge and provide intelligence to guide policy and practice interventions to minimise direct and indirect harms to the population.
(vi) Dissemination and use: Create public trust for health information
The WHO (2020) recommends “simple , timely, effective, evidence-based and honest communication, transparency to build public trust, manage infodemics”. Transparent dissemination of data on health care utilisation (e.g. using dashboards) tends to create public trust. Good practice examples in this respect are Portugal and France. As the WHO recommends, data analytics on public health decisions must be data driven at all times. WHO also recommends “to monitor the mental and physical impact of various measures on the population, as well as the public’s willingness and ability to adhere to those measures.” WHO particularly focuses on countries with public policy levers outside the health sector and HiAp approaches (WHO, 2020).
This pandemic has shown that reporting on mortality numbers varies widely across European countries, and sometimes even at subnational level. It is therefore required that a strong HIS communicates in a clear and transparent language. This includes for example communicating on complex differences in reporting methodologies, transparent information on how numbers are being produced, and also the transparency with which evidence supporting re-opening after lockdown is communicated. Important questions for evaluating the HIS in this respect are: How does the government interact with experts in the field? How do experts report evidence to policy-makers? Are evidence-building documents available to the general public? How does interaction with public opinion take place as a consequence of measures taken (e.g. social media, press conferences)?