The reporting of studies as a preprint prior to peer-review has gained popularity in the scientific community in the last decade. Preprints are complete scientific manuscripts that are publicly accessible on a preprint server and also citable, but have not yet been peer-reviewed for publication in a scientific journal . Preprint servers, for example medRxiv, Research Square, bioRxiv, PsyArXiv or arXiv, are the result of the rapid digitalization of science allowing researchers to share findings quickly, discuss them on social media, attract potential reviewers, and assure immediate access to content that would otherwise be delayed or even lost.
During the COVID-19 global health crisis we observed an exponential growth of preprints with more than 15,000 preprints related to COVID-19 , see also Figure 1 in the Appendix. Some of these preprints suffer from substantial shortcomings  such as poor methodology, poor reporting and unjustified conclusions. Given the potential policy implications during a pandemic, it is no surprise that preprints receive a huge amount of attention on social media and the national and international press. Often preprints are accompanied prematurely by press releases as well, but the danger is that newly uploaded preprints or preprints in the wild (not on a preprint server with well-defined publishing policies) often received no checking or vetting at all. If such preprints go straight to the press offices or beyond, unreliable findings are spread without a safety net and could potentially get overblown in the media. Many preprints servers therefore added a disclaimer for their COVID-19 related preprints, or sometimes even reject a manuscript when the study makes claims that may affect clinical practice . In the worst case, unvetted preprints may contribute to the distribution of misinformation and may ultimately undermine trust in science.
We recently discussed three ways in which the COVID-19 pandemic could help to change science for the better in terms of implementation of open science, enhancing team science with larger and better quality studies as well as rethinking peer-review . Careful peer-review is key to identify the best of research in terms of quality that will improve the understanding of this new disease. Admittedly, peer-review has some major limitations: it is very time-consuming, it is not particularly rewarding for researchers and may even be flawed . When research is under a stress test, such as at the moment, other forms of quality control if peer-review is not yet available for a preprint have to be considered.
Quality control is indeed crucial given the current rate of journal publications during this pandemic. Such rapid research will undoubtedly contribute to more research waste [6,7] and retractions. As of November 9 2020, RetractionWatch.com listed 46 COVID-19 studies that have been permanently retracted, temporarily retracted, or were given an expression of concern . However, a preprint is usually withdrawn only at an author's request. It is therefore vital to identify flawed research papers already at the preprint stage before damage is done .
For example, on May 18 2020 a preprint was uploaded on MedRxiv claiming that treatment of COVID-19 patients with hydroxychloroquine (HQ) and antibiotics was associated with significantly better clinical outcomes . The study was retrospective in nature despite serious concerns about the low methodological quality of studies with similar design . No study protocol was mentioned nor did the authors adhere to the STROBE reporting guidelines  for observational studies. The authors withdrew their manuscript on June 14 2020 “because of the controversy about HQ treatment”. WHO discontinued the HQ treatment arm in their SOLIDARITY randomized trial soon thereafter and as of today, treatment with HQ and/or antibiotics is not recommended by the WHO for COVID-19 patients.
The aim of the present project is to provide simple and clear guidance in form of a checklist to assess the quality and credibility of a preprint. Some guidance already exists, for example, the ResearchSquare preprint server offers data and methods reporting checklists for preprints. Additionally, an interdisciplinary survey among researchers found that preprints provide only little information to enhance their credibility . However, the context of the current pandemic and the rapid response of the scientific community through preprints puts us in a new and urgent situation to provide specific guidance systematically and comprehensively. To start, we target undergraduate students, whom we want to advise to critically assess the information given in preprints. Additionally we propose to work with university press officers to develop easy-to-use guidance in the quality assessment of preprints. To this end we will collaborate with the press offices at both universities (willingness to collaborate confirmed by Marco Cattaneo at UNIGE and Roger Nickl at UZH). Training sessions at both universities will inform about the usefulness of the checklist and may lead to further improvements. A large scale roll-out is beyond the scope of this project but could be the object of a subsequent larger grant application, e.g., from the newly founded Swiss Reproducibility Network (www.swissrn.org)
The collaboration is between the Center for Reproducible Science (CRS) at UZH led by Prof. Dr. Leonhard Held and the group of Prof. Dr. Evie Vergauwe at UNIGE, co-founder of the Swiss Open Psychological Science Initiative. Complementary efforts of both partners to improve the reliability of research results in general and their different scientific backgrounds in biomedicine and psychology form an excellent basis for this project. Specifically, the development of a tool to check the scientific validity of studies by the Higgs.ch magazine was based on a cooperation with the CRS (https://www.higgs.ch/science-check/), and a lecture providing guidance on how to read the scientific literature and the lay literature on research themes critically and sensibly (Scientific Skills and Knowledge in Psychology) has been developed for psychology students at UNIGE. Combining the background of both groups will not only positively influence the development of the material but also allow us to easily reach a wide audience at both universities.
In summary, PRECHECK aims to deliver important insights and training at a critical moment when the trustworthiness of science is at stake. In times of rapid digitalization, information overload, and a global-health crisis, we prepare the future scientists by targeting the youngest members of the academic community. The collaboration between the two partners will contribute substantially to the transformation of both universities towards a culture of reliability, integrity, transparency and true societal impact.
Covid-19 has not just caused a severe global health crisis but also a severe global economic crisis, by manymeasures the worst since World War II. Once the health crisis is under control, the attention will inevitably shift to the economic crisis and two questions will be at the center of the public debate: What happened tothe economy? And what are the lessons learned?
In this project, we aim to provide answers to these broad questions, using the concept of socio-economic resilience as our unifying theme. This ensures that there is a direct point of contact between our thinking group and the others, given that building resilience is the expert dialogue’s central goal. Our target audience are policymakers and other stakeholders in policymaking, most notably journalists and interested members of the public. We therefore plan to make our result accessible to this audience by summarizing them in a policy brief, underpinned by a collection of explanatory notes.
Our first task will be to establish what we mean by socio-economic resilience and why it is a desirable policy goal. In particular, we need to clarify how it compares to more standard economic policy goals such as efficiency, equity, prosperity, and stability. Is it just a new term for an old concept, or does it describe a genuinely new policy goal? Once this task is completed, we can turn to our two main questions of what happened to the economy and what are the lessons learned.
What happened to the economy? What is already clear is that Covid-19 had very heterogeneous effects on socio-economic systems and our second task will be to document this heterogeneity. Which countries, which industries, which firms, which households, and which individuals were particularly resilient to the pandemic? And has this all increased or decreased existing socio-economic inequality? Moreover, what kind of resilience was even required? Which parts of the economy experienced only temporary disruptions and which had to adjust permanently?
What are the lessons learned? Here, the task is to identify what kind disruptions we even need to becomeresilient to and then compare these disruptions with the Covid-19 shock. To organize our thinking, we find it useful to distinguish between “known unknowns” and “unknown unknowns”. The known unknowns are unknown disruptions associated with known challenges, particularly climate change, the digital revolution, and the emerging great power rivalry between China and the US. The unknown unknowns are unknown disruptions associated with unknown challenges, i.e. risks we are not even aware of.
We plan to approach these questions by aggregating existing knowledge and conducting original research. For this, we need the help of a full-time research assistant who can review the existing literature and perform basic statistical analyses.
Background: The Covid-19 pandemic confronts each government in the world with the same questions: how do we protect the population and save lives today, and how do we address the long-term challenges and prepare our societies from outbreaks in the future. Before the pandemic, evaluations of national epidemic preparedness had been done and epidemic response strategies had been developed (and were strongly connected).
However, many countries have done worse than would have been predicted although they seemed well prepared and with the capacities to prevent and control an outbreak of infectious disease. The Global Health Security (GHS) index report published in late 2019 identified 6 categories designed to assess a country’s capability to prevent and mitigate epidemics and pandemics: 1. Prevention; 2. Detection and reporting; 3. rapid response; 4. Health system (sufficient and robust health system to treat the sick and protect health workers); 5. Compliance with international norms (commitments to improving national capacity, financing and adherence to norms); 6. risk environment. The index attributed more weight to the first two categories. However, the pandemic has revealed weaknesses and identified domains that should be improved in order to render our societies more resilient in the near and far future.
Objectives: The project aims to develop knowledge concerning fields that should be improved in order to render our societies more resilient in the short-, medium- and longer-term future and approaches that should be adopted. Two areas of improvement should be examined based on a human rights and global health ethics approach: the first concerns the national (health) systems and the improvement of the resilience and preparedness of societies through adequate responses to current and future situations; the second concerns the global level and the improvement of collaboration and solidarity through the development and/or reform of legal instruments and of governance structures in global health.
Concerning the first approach, it has already been shown in the literature that life chances differ greatly depending on where people are born and raised, and where people work and live. Within countries, the differences in life chances are also important. This is the conclusion of the WHO Commission on social determinants of health from 2012 and the current pandemic has illustrated this situation perfectly, by focusing attention on the susceptibility of poor and vulnerable populations to disasters.
Concerning the second approach, the specific focus will be on particular agreements designed to protect and promote population health (e.g. the International Health Regulations adopted in 2005) and on the role of WHO as the central international organization governing collective responses in the time of a pandemic and holding the constitutional authority to adopt legal instruments, standards, strategies, guidelines and provide technical support.
The collaboration between our two institutions aims to propose answers solidly grounded on human rights and medical ethics and encouraging the evolution of institutional arrangements and the use of legal instruments for a better preparation for the future. The questions to be discussed are:
1.From human rights, ethical and legal perspectives, what is at stake in the current pandemic:
The ongoing COVID-19 pandemic is generating massive adverse health and socio-economic impacts for societies around the globe. Most importantly, this pandemic has also opened a window of opportunity to reflect on and to transform societal processes in order to prevent shocks originating from health, biodiversity, and climate as well as to reduce their impacts on economies across scales. Thus, there is aunique opportunity to upgrade resilience-based efforts and equitable investments into health, physical and social systems, and to build forward-looking sustainable transformations. Growing evidence shows that together the health, biodiversity and climate crises are challenging systems within and across sectors of society and nature, which may be less or more resilient to shocks, of varying duration, intensities, and frequencies.For example, it is becoming increasingly clear that the spread of disease agents with pandemic potential is linked to natural ecosystem fragmentation, as human encroachment in pristine ecosystems leads to more opportunities for human-wildlife interactions and spread of disease agents. There is also evidence that air pollution contributes to the spread of virus, while the measures to deal with the pandemic, i.e. global lockdowns, have had beneficial, yet temporary, effects on greenhouse gas emissions (but not concentrations), biodiversity and Earth system functioning.
To be able to deal with these concurrent stressors and shocks, we posit that there is a need for updating integrated risk assessments and management, to better prepare and recover the resilience of natural and social systems to sudden or gradual shocks as well as to the inherent interconnectedness of these pressures. Therefore, herein we propose to better understand the necessary conditions to devise a novel Integrated Risk Assessment and Management (IRAM2.0) to address the concurrent risks and shocks from health, biodiversity and climate. This approach deals with the opportunities emerging from tackling multiple shocks concurrently, leveraging on synergies and on the creation of co-benefits for the design of effective response systems. More specifically we propose to investigate:
i) What are the key drivers, stressors and vulnerabilities in the health, biodiversity and climate systems and how have they acted individually and synergistically in relation to the pandemic?
ii) What social, environmental and economic trade-offs, synergies and co-benefits can emerge from the adoption of IRAM2.0 measures?
iii) What type of data can better inform and what actions can be implemented to support IRAM2.0?
To answer these questions, we build on a scoping exercise that the PIs organized for the kick-off meeting of “Shaping Resilient Societies”, an initiative of the University of Geneva and University of Zurich in partnership with the World Economic Forum. During a series of “thinking group” meetings prior to and during the kick-off event (held on November 12, 2020), a group of experts provided their visions about key issues related to pandemics, climate, biodiversity and sustainability, including inputs on integrated risk assessment and management. Here we propose the obvious follow up to operationalize the gaps identified by the thinking group.
We will achieve this by (i) triangulating desk-based research and semi-structured interviews with stakeholders in the public, private and academic sectors; (ii) co-designing a citizen science and communication platform to collect, share and provide information necessary for the proposed IRAM 2.0; and (iii) analyzing whether citizen science data meets the theoretically identified needs for achieving IRAM2.0. We will focus on the integration of data to improve process-based understanding, causal explanations (e.g. links between environmental and health factors, as the core theme of the “OneHealth” approach) as well as the identification of social and environmental needs and the evaluation of risk management decisions. The on-line citizen science platform will be co-designed with the thinking group. We will then test the prototype platform in two areas of Switzerland (Geneva and Zurich). We will involve residents in these two cities, with focus on students from the Universities of Geneva and Zurich. Options for upscaling will also be discussed. We will also leverage on the expertise and support of the Citizen Science Center Zurich (https://citizenscience.ch/en/).
Our vision is that this project will provide a discussion forumin tandem with providing research evidence and societal embeddedness through citizen science, improved communication strategies and stakeholder engagement. Ultimately, this will contribute to improve social capacities to live with shocks like pandemics or extreme climate events, as well as ongoing cumulative shocks that require a new narrative on the interlinked health, biodiversity and climate stresses. We will also focus on measures to put in place to quickly respond to shocks and to shape resilient and agile societies. We believe that our results will allow us to identify potentially relevant leverage points for working towards true transformational change to build resilience in societies and for addressing the concurrent global change problems of our and future times.
The emergence of new viral epidemics continues to be a major challenge for public health and entire societies, as illustrated by the ongoing COVID-19 outbreak. To integrate lessons learned from the past is beneficial in this situation, as for example all public health interventions currently being implemented are adapted from experiences gained from past pandemics since the 19th century. However, past pandemic experiences are not sufficiently present in Switzerland. This knowledge of experience would be particularly valuable in terms of strengthening the resilience and responsiveness of the society with regard to on-going and future crises.
The literature on past influenza pandemics has shown how to integrate lessons from the past into pandemic planning. We identified the following research gaps: i. Interdisciplinary projects are still the exception rather than the rule; ii. Systematic comparisons and connections between two or more influenza pandemics are rare: iii. For Switzerland, the quantitative evidence basis on past influenza pandemics is still very thin, especially for the outbreaks of 1889 and 1957. For the Spanish flu 1918, outbreak patterns have hardly been reconstructed on the basis of incidence numbers.
In our proposed project, we build on the piloting work on the temporal and regional spread of the Spanish flu in the Canton of Bern in 1918, which was gradually expanded over the last 5 years at the Institute of Evolutionary Medicine at UZH (iem.uzh.ch/en/pastpandemics.html). A first publication on these pilot data from Bern is currently provisionally accepted in the medical top-journal Annals of Internal Medicine (and confidentially accessible for the reviewers of our proposal here. In this paper, we show for example that during the second wave, cantonal authorities initially reacted hesitant and delegated the responsibility to enact interventions to the municipalities, and reductions in incidence associated with interventions were less pronounced. In the present project, we now want to advance the state of quantitative knowledge on past influenza pandemics in Switzerland to a whole new level. For feasibility reasons, we will focus on the first three and still strongest pandemic influenza outbreaks in Switzerland, 1889, 1918, and 1957.
Our main hypothesis is that the patterns of pandemic spread, its determinants, and effects of public health interventions are similar across pandemics. The aim of our project is to use epidemiological methods to reconstruct and connect the temporal and regional spread of the 1889, 1918 and 1957 pandemic outbreaks in Switzerland based on incidence and mortality data, and to explore the determinants of the spreads as well as the change in epidemic growth associated with public health interventions. Our project is intended to digitise and make previously unavailable archive material accessible. This quantitative assessment also includes the critical handling of this data, its background, and uncertainties. Last but not least, we also want to ensure the transfer of these past pandemic experiences to the public and policy makers.
We will work with incidence and mortality data of/from influenza-like diseases, mostly aggregated on a weekly level for all cantons, districts, and (if available) municipalities in Switzerland. We have identified the following archival sources which provide rich information: For 1889, we will digitize what is probably the most comprehensive statistical overview ever compiled on a pandemic in Switzerland (Schmid 1895). This book consists of over 300 pages full of statistics and tables and is currently only available in printed paper form. For 1918 and 1957 we will digitize officially reported cases of influenza-like illness by physicians and municipalities/districts which were listed in the weekly “Bulletin des Eidgenössischen Gesundheitsamtes".
Based on these data, we will a) reconstruct the spatial and temporal spread of influenza like illness across the three pandemics, b) assess ecological determinants of the pandemic spread (environmental, socio-demographic, cultural characteristics, connectivity, etc.) as extracted from historical sources like the Swiss Federal census or existing HistGIS tools; and c) analyse the associations between public health interventions and the pandemic course. We will calculate incidence and mortality rates, estimate basic reproduction rates (R), and calculate space-time scan statistics (SaTScan). In addition we will use general additive models (GAM) to model daily or weekly incidence numbers as a function of covariables (population density, connectivity, share of agricultural workforce, regional GDP per capita, language region, weather data, etc.). The biggest challenge is reporting bias (non-participating physicians, misdiagnosis, etc.), which will be addressed by appropriate methods for presenting statistical uncertainties. Our project will result in joint scientific publications, policy briefs, and a larger grant application to the SNSF. Our project is interdisciplinary (we combine History, Evolutionary Medicine, Epidemiology, and Public Health), collaborative, and it is a continuation of the successful links between the two involved institutes and the involved applicants.
Quantifying past pandemic experience is important because knowledge about past pandemic experiences should be evidence-based. In our work we showed that in Switzerland, due to a lack of direct experience in recent decades, awareness of this immediate pandemic risk has been lost (smw.ch/op-eds/post/the-pandemic-gap). Even in today's Swiss Federal pandemic plan, the explicit knowledge gained from past experience does not go back any further than to the Swine flu in 2009. We will further open the doors to this valuable past and make these past pandemic experiences accessible again, especially for the public and to policy makers, who are key-players in coping with a future pandemic.
1. Objectives of the project
Access to data to inform decision making is of utmost importance in a public health crisis. As the current crisis evidences, access to and effective use of relevant data is not always possible. To remedy this unfortunate situation, this policy-oriented project has three objectives:
(1) Identifying existing barriers for access and use of data in public health crises.
(2) Developing alternative governance mechanisms to facilitate access to and use of the data needed to meet public health crises.
(3) Publication of a high-level policy paper presenting policy proposals for the implementation of alternative governance mechanisms that allow for better access and use of data in public health crises while protecting individuals from privacy breaches and personal or financial harm.
2. Existing barriers
In our previous exchanges, we identified three barriers to access and use of data in public health crises:
(i) Cultural barriers
There is a lack of data culture in our society. Namely personal data is not sufficiently perceived as a key resource for research, innovation and further development of the society on all levels, including (but not limited to) the health sector. Instead, the use of personal data is mainly considered a threat, not an opportunity. While the preservation of privacy is key, the traditional concepts of privacy must be re-thought. Most importantly, the creation, collection and processing of personal data does not necessarily interfere with privacy as it can be carried out without exposing information about individuals to anyone else. Appropriate governance mechanisms for access and use of personal data may lead to a shift in the public perception leading to a more opportunity-oriented data culture.
(ii) Infrastructural barriers
Switzerland has a high level of infrastructure compared to most other countries. However, the recent crisis has demonstrated that this infrastructure cannot be activated properly when required for the monitoring of the health situation, namely due to difficulties to report and aggregate key indicators. In addition to the need of a robust infrastructure supporting fast responses to provide all data resources needed to monitor, understand, and react to a crisis, additional barriers exist regarding data formats. To overcome these barriers, we need to distinguish between shared data, which is shared between different players in the health sector on a constant basis, and shareable data, which can be shared in emergency situations. To allow for the sharing of data in a state of crisis, today’s infrastructure must be adapted towards a robust and interoperable framework that allows to switch between “normal” and emergency situations.
(iii) Legal barriers
There is a lack of an enabling legal framework for access and use of data, especially with regard (but not limited) to personal data. Data protection laws are based on the assumption that the use of personal data is always a risk and a threat for data subjects. Accordingly, the EU’s General Data Protection Regulation (GDPR) and the Swiss Data Protection Act (DPA) establish important barriers to the use of personal data. Additional barriers for the access and use of data stem from factual ownership of data which is backed-up by law, namely the protection of trade secrets. As opposed to these restrictive rules, the law contains hardly any provisions that enable and foster the access to data collected and stored by businesses and government agencies. To allow for the necessary access and use of data in a health crisis, an enabling legal framework is needed.
3. Future Solutions
In order to overcome these barriers, substantial progress on three levels is required. While the focus of this project is to improve the access and use of data in public health crisis, some fundamental shifts are necessary that also apply to “normal” times to enable appropriate reactions to future health crises:
(i) Data Literacy
The use of data can only be promoted if politicians, businesses, and the public at large perceive data as a key resource for decision making, research, innovation, and further development in the digital society. To this aim, data literacy, that is, the ability to critically collect, manage, evaluate and use data, must be promoted at all levels of the society. A data literate society would be able to make informed decisions about how data handling should be organized, to assess the implications of data policy decisions, and to evaluate the validity of inferences drawn from data. In the future, data literacy will be as necessary for citizenship as the ability to read and write.
Data can only be shared in a public health crisis if the systems and formats used for storing relevant data are interoperable. To ensure interoperability, we need to investigate the introduction of mandatory interoperability standards for shared data as well as mechanisms for best-effort, ad-hoc interoperability for shareable data previously not deemed relevant for interoperability. This can only be ensured by establishing mandatory interoperability standards applicable to all data needed to meet a public health crisis (shareable data).
(iii) Rethink Data Protection Law
To use the potential of personal data, novel approaches to data protection law are needed. Such approaches need to abandon the risk-based catch-all concept of the GDPR and the Swiss DPA. Instead, novel rules should focus on providing appropriate means to remedy actual harm that may be caused by the processing of (personal and non-personal) data if such harm materializes. Theses remedies should not be limited to the recovery of damages (e.g., financial losses) but include suitable legal measures to avoid discrimination and manipulation that may result from the processing of personal data. While these novel approaches could apply both under normal circumstances and in public health crises, one could also restrict the application of these rules to a state of crisis, thus allowing a much wider use of data in a crisis while ensuring that no individual gets harmed by such uses. In addition to amendments of substantial law, novel dispute prevention and dispute resolution mechanisms are needed to ensure that disputes about access and use of (personal) data can be solved efficiently.