The impact of COVID-19 will be long-term and far reaching for both academics and fellow citizens. We have much still to learn about this virus. Longitudinal cohort studies are especially valuable as they allow us to look at change from pre- COVID-19 to living with and through the pandemic.
Two cohort visions for 2020
Both Generation Scotland (GenScot) and the Avon Longitudinal Study of Parents and Children (ALSPAC) had a clear vision for 2020. Backed by recent awards under the Wellcome Trust Longitudinal Population Studies portfolio, this was the year that both planned to forge ahead with a new wave of recruitment and recontact of their longitudinal cohort studies. Like all active UK cohorts, that was abruptly put on hold as the COVID-19 pandemic reached the UK. Monday 23rd March 2020 was the official date for lockdown, but both GenScot and ALSPAC had independently anticipated the inevitable and already started to pivot towards online surveys and enhanced data linkage.
Independent, but aligned through common purpose
Although we started off developing our surveys independently, the shared themes of family and population based longitudinal studies with an emphasis on health and wellbeing naturally led to an overlap of validated core questionnaire content. Two University of Bristol questionnaires for ALSPAC participants gained more than 13,000 returns, while the University of Edinburgh survey to GenScot volunteers sent out in April 2020 gained 4,800 direct returns, with a further 14,000 (18,800 in all) through collaborative networks and social media channels.
GenScot and ALSPAC had already established academic links and were keen to collaborate through their Wellcome awards. A quick call to compare results made it crystal clear that our surveys highlighted the same major issues – increased depression, anxiety and worry under lockdown.
The scope for joint analysis was obvious and an initial cross-cohort collaborative study was undertaken. This was led by Senior Research Associate Dr Alex Kwong, who fortuitously has a joint appointment between Bristol and Edinburgh through the MRC Mental Health Pathfinder awards. If there was any need to confirm the value of joint appointments across cohorts and institutional, here was a clear case in point.
The impact on mental health during the COVID-19 pandemic in these two longitudinal UK population cohorts showed remarkable similarity in findings. Both depression and anxiety were high during the pandemic, even when accounting for pre-pandemic mental health. There was a strong age affect too, with younger members feeling the effects of lockdown more keenly. Finally, there was consistent evidence that women, those who had experienced financial problems before the pandemic and those with a prior history of mental health conditions were more likely to be at risk of higher depression and anxiety during lockdown.
Wellcome spot the opportunity
As soon the Wellcome Trust Longitudinal Population Studies portfolio managers, Mary de Silva and Bruna Galobardes, head of our individual and joint efforts, realised an opportunity, they got in touch and asked if we would consider working together to share our surveys more widely. We were keen to do so and the initial discussion led to a wider discussion, which they drove. Soon, several other cohorts were on board and through near daily online meetings and email exchanges, a plan emerged to adapt the GenScot and ALSPAC questionnaire for wider use in the UK and other cohorts.
Secretariat and Steering Group established and COVID-19 questionnaire developed
Life under COVID-19 has been a strange ‘out-of-office’ experience for academics like us too. The normal rhythm of work days and weekends, core time and downtime blurred. In rapid order, Wellcome suggested that they support a light-touch Secretariat, charged with coordinating and delivering questionnaires in a ready-to-use format, and convene a Steering Group to provide input, guidance and expertise on content. By 27th April 2020, the first questionnaire had been agreed and was ready to share with cohorts for ethics approval. The Secretariat made the survey available in both REDCap and Qualtrics format for ease of adoption, administration and analysis.
Take up and next steps
As of 5th August 2020, 50 cohorts or individual researchers, representing hundreds of thousands of participants have picked up the questionnaire for use in their COVID-19 related research.
Meanwhile, the Secretariat and Steering Group have designed the first follow-up that is now ready to share. Questionnaire Two repeats key questions in the first version so that changes over time can be captured, but also adds new questions that reflect both new information on COVID-19 risk factors, outcomes and changes on Government mitigation and suppression measures. Wellcome have provided start-up support for one year and we expect to run future questionnaires timed to reflect further changes in control measures and their relaxation as we ease out of lockdown and adjust to a new normal.
An analysis group has been formed to plan and conduct cross-cohort analyses with six initial areas of immediate research and policy impact.
- Coping mechanisms/risk perception/wellbeing and health outcomes.
- COVID-19 assessment, direct and by proxy, with probabilistic assignment of case and risk status.
- Longitudinal exposures (pre-Covid-19 and life course) and differential outcomes.
- Standardisation against representative population-based collections.
- Impact of COVID-19 on education.
- Environmental influences.
One simple approach
There is now no doubt that the direct and indirect impacts of COVID-19 will be widespread, long-term and far reaching. We need a longitudinal approach to adequately assess and seek to mitigate these effects. The UK has a rich diversity of long standing and contemporary cohorts that are well placed to provide the means to investigate and report on the wider impact of COVID-19. The Wellcome COVID-19 questionnaires provide one simple approach to ensure cross-cohort harmonisation and meta-analysis.
For longitudinal studies, the collection and linkage over time is critical, matched only in importance by participant retention. Under COVID-19, we have seen past barriers to health data linkage overcome through collective effort. It is what the public expect and we need to retain that capability, subject of course to robust and transparent governance, regulation and oversight. We have, through COVID-19, being given a glimpse of what the future could hold for longitudinal health and wellbeing studies.