COVID-19 Interview series – Gustav Nilsonne
Meet Gustav Nilsonne, a researcher in neuroscience and metascience at Karolinska Institutet and Stockholm University.
In light of the ongoing COVID-19 pandemic, CESSDA launched a new series of interviews in 2020. These interviews are targeted at researchers and we aim to focus on a specific topic each time, hearing from a top researcher in the field to gain insight into the current situation.
This time, we talk with Gustav Nilsonne, a researcher in neuroscience and metascience at Karolinska Institutet and Stockholm University. Both of these institutions are a part of the Swedish National Data Service (SND) consortium and SND is the Swedish Service Provider of CESSDA.
Gustav Nilsonne is also the coordinator of SND’s domain specialists, a position that includes coordination of the national domain specialist activities, as well as being a bridge between SND’s headquarters and the domain specialists in the SND consortium universities.
In his research in neuroscience, Gustav focuses on sleep and diurnal rhythms, and communication between the brain and immune system. Topics which indeed appear relevant after one year of living during a pandemic!
CESSDA asked Gustav Nilsonne to answer a few questions.
How are you coping with the COVID-19 crisis and how do you make sure that you get work done?
Generally I am quite happy with working from home, because of the increased flexibility and freedom. I have had to learn how to plan breaks and things to do outside, so I don’t spend all day in front of the screen, which easily happens if I don’t make a conscious effort. We have a cat at home, who helps by coming over a few times a day, requesting that I come from the keyboard and pet her instead.
Can you point out one benefit and one drawback of your change in routine as a result of the pandemic?
Less travel, due to conferences and meetings moving online, is a considerable improvement for me. With more control over my day, I find it easier to keep up regular exercise, sleep, and meal habits. A drawback is missing out on the informal conversations with colleagues in the corridors and coffee rooms.
Can you tell us about your ongoing research projects and how COVID-19 may affect the road ahead in terms of your research?
We had to stop data collection prematurely in the spring of 2020 in a major research project of mine, involving MRI scanning of sleep deprived volunteers. The funding period was close to ending, and we were not going to be able to resume collecting data for this project. For some time now, I have been moving towards gathering less data and doing more secondary analysis of existing data, in order to reach larger sample sizes and more reliable findings. There are so many datasets out there just waiting to be combined and analysed in new ways. I think that the pandemic will accelerate this move.
Sweden is one of twelve selected countries in European Social Survey new web panel. Wave 1 of the survey includes questions on vaccinations against COVID-19. What are you looking forward to as a result of this web survey?
Comparative data on a large scale are an indispensable research tool for investigating differences between countries and populations. While I am not very knowledgeable about details such as the representativeness of this survey, I do expect it will be of great value. One question that will be particularly interesting to me concerns how answers regarding attitudes to vaccinations will compare to actual uptake between countries.
How can the data collected by large international surveys such as the ESS, ISSP, EVS, WVS and Eurobarometers service (or others) the research community in the aftermath of the coronavirus?
The scientific community has responded to the pandemic with a great flurry of activity, and we are seeing a steeply rising wave of publications. However, high-quality investigations of community interventions to combat COVID-19 have been few and far between. Important public health and social questions are being debated quite hotly, but on a shaky foundation when it comes to evidence. Which kinds of public health advice, recommendations, and restrictions are most effective and cause the least harm?
I hope that in the aftermath of this pandemic we will build a readiness to test societal interventions. When the next pandemic comes, I would love to see “sleeper protocols” for scientific studies ready to be activated, with ethics approval and other administrative obstacles already cleared. Proper comparison studies of different interventions, such as recommendations on distancing, hand-washing, masks, limits on commercial and cultural activities, and so on, could then be started immediately. This kind of studies should be coupled with large-scale data collection efforts, and preferably tie in with existing registries and population-based surveys.
Do you think that the importance of social science data will be better recognised as a result of the current COVID-19 pandemic?
The morbidity and mortality burden of COVID-19 has turned out to strongly follow a socioeconomic gradient. Risks of hospitalisation and death are considerably higher in deprived areas. Factors such as income and migration status have emerged as strong determinants. This highlights that public health efforts against COVID-19, and by extension likely against future pandemics, need to consider a wide range of social factors. Important data will continue to come out for years to come, and will hopefully be able to disentangle to some extent the mechanisms behind this inequity, e.g. crowded living conditions, precarious employment, health and protective behaviours, etc.
The COVID-19 pandemic has also shown that near-real-time availability of high-quality data is crucial for monitoring not only case numbers and mortality, but also social and economic consequences of the pandemic. We need better access to data from health records, registries, and administrative databases. Challenges include interoperable IT systems, semantics and harmonization, e.g. standardized variable names and metadata, administrative hurdles, and a lack of consensus on appropriate safeguards for privacy. The COVID-19 pandemic showcases that good data can save lives, and I hope this experience will provide an impetus for better solutions.
Your research in metascience concerns reproducibility and openness in science. You are also one of Sparc Europe's European Open Data Champions. Can you tell us more about this?
I take great interest in reproducibility and openness, just as you say. For instance, I have been involved in several studies of analytical and empirical reproducibility in science on humans. Currently I am particularly interested in building a theory of the choices researchers make when they analyse data. They traverse a space of possible paths, all of which may be reasonable, but which make different assumptions and which may yield different results.
Fundamentally, I regard data as the building block of cumulative knowledge, and I believe it is very important that we make our data open, both when we create new data and with regard to data which already exist, scattered about the face of the earth, but which is at risk of erosion and loss. We need knowledge and skills (the CESSDA management guide is an excellent resource). We also need infrastructures. Data repositories are relatively well developed, but further support is needed for curation and sharing of data. Universities need to invest in data stewards, data librarians and similar functions to ensure that research can be supported with skills and with enough hands to do the work of preserving and sharing data. Funders need to explicitly support data management and sharing as a part of the research cycle, and we also need dedicated funding targeted at saving data from old and existing research, which is at risk.
What software platforms and tools for research data sharing and analysis are you currently using?
For sharing research data, I use different repositories depending on the needs: the Swedish National Data Service, with which I am affiliated of course, but also openneuro.org for brain imaging data, Zenodo for its integration with GitHub, and sometimes others such as the Open Science Framework (OSF). For analysis, I like to use R because it is open and also very powerful, and for code sharing I prefer GitHub/Zenodo.
What are the main barriers for the reuse of research data and code at an international level and what is for you the key to improving research data management?
We need to shift the perspective on what is a scientific contribution. Instead of counting publications and impact factors, we need to value data and code as scientific outputs. The technical means are already in place. It is fully possible to publish data and code with permanent identifiers so they can be properly cited and attributed. Funding agencies, when they assess grant applications, need to develop ways to ask researchers to show the value of their data and code, alongside or instead of publication lists. A conversation has started on responsible assessment and metrics, but much remains to be done.
What are you most looking forward to achieving in 2021 given the current circumstances?
Now I have settled in to working from home, and I do not anticipate travelling to meetings and conferences this year. I hope to be able to find more time to write instead. For instance, I am working on a theoretical framework for the relationship between data loss and scientific inference, which I would love to put into print. I also hope to be able to write more about open science and reproducibility for the general public and/or by writing blogs.
Gustav Nilsonne is on Twitter!
Recommendations for Covid-19 data from the Research Data Alliance (RDA): Rekommendationer om Covid-19-data från Research Data Alliance (RDA) (interview in Swedish)
Find out how CESSDA is supporting researchers on our COVID-19 page. Our flagship product, the CESSDA Data Catalogue is a platform for researchers wanting to find and reuse social science and humanities research data, including on issues related to the pandemic. All COVID-19-related metadata will be harvested to the Data Catalogue, as they become available by the Service Providers.
See the previous article in this series: COVID-19 Interview series – CESSDA COVID-19 Ambassador – Helena Laaksonen.