What have we learned from longitudinal studies in Aotearoa New Zealand?

This second part of this Special Issue includes six papers. Edwards et al discuss how longitudinal research grounded in Māori concepts of lifecourse is urgently needed in Aotearoa New Zealand, as it will lead to better understanding of the lived realities of whānau and help to design strategies to improve Māori wellbeing (Edwards et al. 2023). Next, Dunedin Multidisciplinary Health and Development Study researchers reflect on lessons from 50 years of longitudinal research and emphasise their gratitude to the participants who have willingly given so much to make the study successful (Poulton et al. 2023). Following this are papers from three clinical studies, the New Zealand Parkinson’s Progression Programme (MacAskill et al. 2023), The Dementia Research Prevention Clinic Study (Tippett et al. 2023), and the New Zealand Genetic Frontotemporal Dementia Study (Ryan et al. 2023); all aim to characterise biomarkers that will allow early diagnosis and treatment of dementia, an intervention that we know drastically improves outcomes for dementia patients and will significantly aid in improving the health of the New Zealand population as we age. Finally, a report from the Health, Work and Retirement Study (Allen et al. 2023) describes impacts on people’s capabilities (particularly cognition) as they age. Together, these studies highlight the potential that we have as a community of longitudinal researchers to work collaboratively to markedly improve health and wellbeing outcomes for the New Zealand population.

As we draw together this two-part Special Issue showcasing the breadth of longitudinal research in Aotearoa New Zealand, here the Guest Editors of the special issue, together with three leading NZ longitudinal researchers reflect on the implications of longitudinal research for health, education, and social policy in New Zealand. We also reflect on what longitudinal research has taught us, and how we can use these findings to improve the lives of future generations.

The benefits of cohort-based longitudinal studies to science and society

Longitudinal cohort studies are unique and remarkable in their ability to directly track and capture the lived realities of their participants, in the changing contexts of their families and social and environmental factors over time. Appropriately designed, and with a suitable population cohort that is both recruited and retained to preserve study power, such studies can provide information about choice, churn, and change over time at an individual level; other study designs cannot do this.

Increasingly the sustainability and utility of longitudinal studies has been challenged. Linked administrative datasets (some of which already exist) are becoming increasingly available and often have the advantage of speed, in that they do not require long lag times to realise their potential. An example of this in the NZ context is the Integrated Data Infrastructure (IDI) resource, the availability and curation of which represents a major innovation in bringing together information from across sectors. The IDI’s innovation lies in the rich information it provides that is increasingly being used not only to inform policy decisions, but also it is being used directly by researchers, who are using these data to undertake a new form of ‘longitudinal research’ on wellbeing.

However, it would be short-sighted to relinquish the significant investment in existing longitudinal studies in NZ (or internationally) in favour of relying on the administrative datasets alone to inform policy evaluation, to inform new strategy development, and to measure successful implementation of policy. The reason for this is that each type of study and data resource has both strengths and limitations. Administrative data have important limitations, as they are biased towards inclusion of those who interact with government services, hindering our understanding of those who eschew mainstream services. Longitudinal cohorts are expensive to maintain, but allow investigators dive deeply into a relatively small number of lives, and, as well as the broad coverage of administrative data that traditionally has information collected from the majority of the population, complement each other to maximise strengths and minimise limitations of either data type on its own. As such, the sum of the information when these data resources are combined is greater than the information in any single dataset type. We regard it as crucial that, in the design phase of any new longitudinal study, as well as in any established study, explicit examination of the utility of data linkage to administrative datasets is an important design consideration. It is critical that this feature adapts over time, respecting data governance agreements, as administrative datasets evolve, and as longitudinal cohorts mature.

Preserving the integrity, relevance and success of longitudinal studies in Aotearoa New Zealand

Inevitably, building and maintaining the trust of study participants remains key to sustaining the utility of longitudinal cohort studies. In the NZ context, the operational management of studies within University environments, rather than within government agencies, has been an important contributor to maintaining participant trust, and this is evidenced via the successful recruitment and retention of diverse groups of participants over time. Having universities at the helm as allowed cohort studies, like the Dunedin and GUiNZ studies, to follow some vulnerable and hard to reach children who do not routinely engage with government services and whose data is often missing or incomplete in administrative records. Further, data guardianship and appropriate data access protocols and processes are essential adjunct components that support the ‘contract of trust’ that exists between researchers and the children and families who participate in these studies (e.g. Reese et al. 2022). These features are unique to the NZ environment, preserve the integrity and relevance of such studies, and are key drivers of success of longitudinal studies in Aotearoa New Zealand.

Longitudinal studies are expensive to set up and to sustain, and often funding is provided for short periods of time. This impacts study productivity, as it means that seeking funding can become an important activity in maintaining longitudinal studies, rather than investing in analyses or translation of findings. Thus to preserve the success of longitudinal studies, funding tranches that cover multiple years and multiple data collection waves (of whatever mode) could help facilitate the most efficient retention of participants over time, and efficient use of the resulting information. Moreover, multi-year funding affords researchers the ability to invest in explicit longitudinal planning, to collect information about trajectories over time (not just cross-sectional, repeated measures), and to communicate expectations of timing of follow up with participants who are increasingly geographically mobile, preserving study impact and visibility. A barrier to study success, however, are frequent and repeated reviews to determine whether to maintain a longitudinal study. While it is important to routinely assess that costs invested are being utilised appropriately to address agreed study objectives, often, study reviews are driven by political rather than scientific motives, which can undermine the morale of a committed study team and directly threaten the viability of the research.

Integrity and relevance is also evidenced by the robust, multi-disciplinary information obtained from participants in longitudinal birth cohort studies, which also creates a national treasure (taonga) with the ability to improve outcomes for individuals born in similar epochs as well as to compare with other studies internationally to understand what works in different socio-political contexts. When appropriately designed, implemented and supported, they retain the ability to provide a unique perspective on ordinary individual lives that can have extraordinary consequences for populations today and in the future.*

The value of longitudinal research for health and social policy

Longitudinal research is essential for informing health and social policy. This is because health and social policy requires a strong evidence base, that can take many years to build, and longitudinal data are unique in their ability to provide this. This ability is because longitudinal studies enable the modelling of individual pathways across the lifecourse and across multiple domains, providing a depth of knowledge that other study designs do not allow. For instance, longitudinal studies reveal how different factors such as development, early childhood, education, and health, to name a few, act across the lifespan, meaning that the vast and comprehensive knowledge base afforded by longitudinal research affords the provision of robust evidence to help determine whether a particular policy change or amendment, or even intervention, will be impactful. Longitudinal research designs using repeated measures, such as the Dunedin and Christchurch studies, are particularly valuable in terms of being able to identify and test questions of causality (Hill 1965; Vandenbroucke et al. 2007). In this way, longitudinal research can be used to identify events, factors and conditions that are potentially modifiable, providing targets for intervention at various levels (individual; population). For example, longitudinal data can inform early intervention strategies, which can help to prevent problems that are harder and more expensive to treat in later life; for instance, those in the pre-school period aimed at improving language acquisition and self-regulation (e.g. Reese et al. 2022). These outcomes are only possible because of the particular strengths of lifecourse longitudinal studies, highlighting why long-term investment in longitudinal studies is vital; it gives researchers the power to determine how and why problems occur.

This two-part Special Issue of the Journal of the Royal Society of New Zealand demonstrates that Aotearoa New Zealand has an abundance of both well-established and emerging longitudinal studies, and that excellence in longitudinal research is a key feature of New Zealand science. The opportunity to combine findings from New Zealand’s longitudinal studies together has allowed us to showcase the breadth of longitudinal research being undertaken in New Zealand and the ability of these studies to impart real benefits to the New Zealand population.

*with thanks to reflections by Helen Pearson (The Life Project, 2015)