In this blog, CLOSER Director Professor Rebecca Hardy argues that, longitudinal population studies need to fight for their place in an increasingly crowded data landscape, but advances in technology and infrastructure mean the future is bright for these unique resources.
In January, CLOSER had the pleasure of hosting representatives from international longitudinal population studies at our collaborative conference in London. Today we published the report of the day’s discussions – reflections on the key challenges facing these studies now and in the future. Here’s what I took away.
Telling our story
A key theme emerged: the need for longitudinal population studies to maintain visibility in a world of ever-expanding data resources. Longitudinal studies play a pivotal role in understanding within-person change over time, while giving voice to the experiences of different generations and consideration of the varying social contexts in which they live.
But have we been effective in articulating our unique contribution to science? The need for communication and connection with a range of stakeholders came through strongly from our delegates. We must communicate with participants to retain the diversity of voices within studies, and to gauge the acceptability of scientific changes, such as new forms of data collection and data linkage. We must connect with policymakers to ensure longitudinal evidence informs policy and practice, and to make the case for linkage to administrative records. And we must communicate the value of the data to the best researchers across disciplines through discoverability and training, to ensure these studies continue to be used for top-quality research.
Longitudinal studies and the open science movement
The open science movement is aiming to make scientific research more transparent and reproducible – and it’s having an impact on the infrastructures around longitudinal population studies.
There was discussion around the importance of FAIR data and the key role of documentation. For example, documentation is pivotal to the validity of any data harmonisation exercise, and to the utility of the outputs generated. CLOSER is committed to open science and last year became a stakeholder with the UK Reproducibility Network. Our contribution to metadata standards and documentation of harmonisation processes will prove vital for ensuring data are reusable and reproducible going forwards.
Country and cultural context matter
Participant engagement was one area where there were considerable differences in approach across countries and cultures. There is no substitute for local knowledge of the study population and, although some aspects can be generalised from one study to another, tailoring of engagement activities is important.
There was great enthusiasm for the possibilities for cross-national comparisons, particularly with respect to policy relevance. However, in terms of data harmonisation, for example, this adds a further layer of complexity as linguistic and cultural differences must be taken into account in such projects.
A recurring theme during the day was the huge potential of exciting technological innovation to transform studies and enhance their scientific potential. Innovation is, of course, vital, but in some cases the enthusiasm was tempered by caution and the need for evaluation.
There must be a strong scientific rationale for using innovative data collection methods. This is especially true if they are replacing tried and tested methods or repeated measures that are more easily harmonised. Studies also need to understand the validity of data collected by any new method, the extent of error, and bias in the responding sample.
Sharing information on failures in innovative approaches, as well as the successes, is necessary in order to avoid pitfalls and reduce costs.
Training the next generation
The rapid advances in technology mean that study teams cannot easily keep pace with the skills required.
The increasing diversity of data collected, from multi-omics to geo-spatial, means that researchers also require an increasing range of analytical skills.
The need to entice those with computing and data science skills into the longitudinal community is perhaps particularly pressing, while there is a clear need to ensure development of skills of existing study staff who are vital for the running of studies. A clear message from across sessions was the need for additional guidance, training and best practice models for harmonisation, data linkage and the design of discovery resources.
Despite the challenges of keeping a longitudinal population study going, the passion with which study teams spoke about their studies was inspiring. It reminded me of the need to value both the lifelong volunteers who participate in these studies, and the staff who run them.
Collaborative events such as the CLOSER conference are hugely valuable in enabling the sharing of approaches, methods and best practice. CLOSER is committed to providing more opportunities to the global longitudinal community to work together in future. But these opportunities will need to be paired with adequate funding and recognition of the incredible efforts involved in seizing the opportunities before us.
Professor Hardy is Director of CLOSER. You can follow her @rebeccajhardy.