The relevance of population-based longitudinal studies for science and social policies
Population-based longitudinal studies form the backbone of empirical research in the social, economic and behavioural sciences as well as in epidemiology and health research. As “largescale instruments” in these fields of science, they serve to test theories and make new observations, while also constituting the basis for evidence-based policy advice. Such large-scale instruments or indeed “research infrastructures” are extensive and complex research instruments with an at least national, if not international relevance for the respective fields of science. They are intended for long-term use (without striving to thwart scientific creativity), and are available to a large number of users, mainly for scientific purposes. Germany has increasingly promoted population-based longitudinal studies in the past few years. Due to structural barriers in the research funding system and to the lack of harmonisation of data collection and analysis along with shortcomings in the necessary interdisciplinary education and training, Germany is nevertheless struggling to keep pace with global research excellence. Against this background and drawing on a stock-taking of the current situation as well as pioneering international expert knowledge, the present statement aims at providing recommendations for the structure of research funding and the harmonisation of future research efforts.
Population-based longitudinal studies are indispensable for the world of research and for society due to three major functions:
- Firstly, human societies are subject to constant change. Longitudinal studies enable the documentation of both stable patterns and changes over time as well as the identification of new trends and the analysis of links between socioeconomic and biomedical mechanisms.
- Secondly, population-based longitudinal studies that examine the same individuals repeatedly over time are, under clearly defined conditions, ideal to test hypotheses about cause-effect relationships. This is usually not possible with administrative and process- generated data (“big data”).
- Thirdly, model-based forecasts of potential future developments can be derived from many of these analyses. This prognostic knowledge provides important guidance in social, economic and health policy decisions.
The relevance of population-based longitudinal studies with multi- or interdisciplinary survey programmes in the above-mentioned fields of science results from the fact that the experimental methodology prevailing in the natural and life sciences is of very limited use for the study of processes at population level. Hence, the social, economic and behavioural sciences as well as epidemiology and health research (in short: social and health sciences) exploit the great variety of biomedical, personality-related and socioeconomic factors within a population and their changes over time. They use theory-based statistical analyses to identify causal relationships and to make projections. Examples are: the assessment of the care and nursing requirements of an ageing population; evidence on positive effects of educational investments on an individual‘s lifelong health; the quantification of the influence of social class on life expectancy. The pioneering insights of birth cohort studies testify to the success of this research strategy. They allow for the analysis of biomedical and socioeconomic processes over an individual’s life course. These studies also show the importance of a long-term perspective, since the value of longitudinal data increases exponentially with the number of study waves.
On the one hand, the working group agrees that Germany as a research location is fortunate to have a diverse as well as intensely and widely used landscape of population-based longitudinal studies which also compares well internationally. This includes, for instance, the Socio- Economic Panel (SOEP), the National Educational Panel Study (NEPS), the Panel Analysis of Intimate Relationships and Family Dynamics (pairfam) as well as several epidemiological cohort studies along the lines of Multinational Monitoring of Trends and Determinants in Cardiovascular Disease (MONICA)/Cooperative health research in the Augsburg region (KORA) and, in particular, the recently initiated National Cohort (NAKO). Germany is also involved in a number of internationally comparative studies, such as the cross-sectionally replicative European Social Survey (ESS) and the longitudinal Survey of Health, Ageing and Retirement in Europe (SHARE). Finally, Germany has a considerable number of well-qualified graduates in the relevant fields of study as well as a large pool of academic staff.
On the other hand, the working group finds that this potential is currently not sufficiently exploited. Knowledge potentials remain untapped for three reasons: firstly, the existing financial and organisational infrastructures (e.g. funding instruments) suffer from contradictions and shortcomings that hamper the sustainability of population-based longitudinal studies. This also holds, secondly, for the intellectual infrastructure (for example, education and training). Thirdly, social science and biomedical research approaches are insufficiently linked and coordinated at virtually all levels.
Therefore, the central aim of this statement is to provide key elements of a long-term and coordinated promotion strategy for population-based longitudinal studies. This should complement and improve the currently prevailing short-term planning and ad hoc management in a number of financially and organisationally inadequately equipped institutions. In addition, biomedical and socioeconomic contents should be combined better in research and education.
A need for action is indicated first of all with regard to the funding instruments and the career opportunities for senior employees. The working group considers such measures to be primarily the responsibility of the German Federal Ministry of Education and Research (BMBF) in cooperation with the non-university research institutions and the German Research Foundation (DFG). Similar to the large-scale facilities in the natural sciences, the research infrastructures in the social, behavioural and health sciences are dependent on long-term stable funding and supervision by senior staff members. However, in contrast to the natural sciences, this fact has hitherto been inadequately recognised in the social, behavioural and health sciences. Firstly, therefore, instruments should be created that will allow for long-term funding (e.g. scheduled funding extensions). Secondly, management staff of major research infrastructures should be able to acquire the relevant skills and to aspire to defined career paths (e.g. tenure track for senior study staff or jointly appointed S-Junior professorships, a German form of junior professorship). Currently, both are not the case for many population-based longitudinal studies.
The working group is aware of the inherent tension between creativity and stability. The group therefore continues to advocate a decentralised approach which restricts neither the creativity of new initiatives nor scientific innovations within already successful infrastructures. However, every already established population-based longitudinal study requires a stable financial base complete with relevant funding instruments, allowing for its continuation and further development on the basis of repeated evaluations (as has, for instance, been successfully established with the Leibniz Association’s seven-year evaluation rhythm).
Action is likewise indicated with regard to universities and non-university research institutions. The current education and training programmes are deficient in several aspects. For one thing, only few locations offer a targeted training of methodological skills in the core subjects of this field of research; this is true both for undergraduate and postgraduate programmes. Furthermore, there exist no training programmes for transdisciplinary research cooperation, in particular for young researchers from the biomedical and the socioeconomic sciences who can be indispensable for obtaining innovative insights. There is also an appalling lack of coordination between the curricula of biomedical and socioeconomic studies, so that frequently, no comparable data can be generated and instructive synergies consequently remain unrealised. Finally, training opportunities in survey management are scarce.
Read the full statement by the German National Academy of Sciences Leopoldina, the Union of the German Academies of Sciences and Humanities, and acatech – National Academy of Science and Engineering, including recommendations at the link below.