Volume 197 - Issue 11
Birth of a cohort — the first 20 years of the Raine study
Authors: Charlotte M McKnight, John P Newnham, Fiona J Stanley, Jenny A Mountain, Louis I Landau, Lawrence J Beilin, Ian B Puddey, Craig E Pennell and David A Mackey
Med J Aust 2012; 197 (11): 608-610. || doi: 10.5694/mja12.10698
Published online: 10 December 2012
Published online: 10 December 2012
Participant engagement and careful planning are key to a successful cohort studyIn the beginning there was an idea ... and funding.
The project might never have started if not for a serendipitous meeting between an accountant for
the Raine Medical Research Foundation and a young obstetrician. The Foundation had been established 30 years earlier by the bequest of Mary Raine (Box 1), a successful businesswoman who ...
Participant engagement and careful planning are key to a successful cohort study
In the beginning there was an idea ... and funding. The project might never have started if not for a serendipitous meeting between an accountant for the Raine Medical Research Foundation and a young obstetrician. The Foundation had been established 30 years earlier by the bequest of Mary Raine (Box 1), a successful businesswoman who left her property empire to the University of Western Australia (UWA) for medical research. The accountant mentioned that the Foundation had decided to award a large sum of money to one big, visionary project, and the very next day the grant application was underway.
The obstetrician’s big idea, the Western Australian Pregnancy Cohort, had two objectives: to investigate the hypothesis that complications of pregnancy might be prevented by frequent ultrasound scans, and to develop a long-term cohort to study the role that early life events have on later health.
From 1989 to 1991, 2900 pregnant women were randomly assigned to either routine obstetric ultrasound or multiple scans.1 Extensive data were collected during pregnancy and the children were assessed at birth and at ages 1, 2, 3, 5, 8, 10, 14, 17, 18 and 20 years (Box 2). Questionnaire data, physical measurements and biological samples were collected looking at growth, cardiovascular, respiratory, immunological, musculoskeletal, nutritional, psychiatric, neurocognitive and ophthalmic health. The current dataset contains more than 85 000 measures on each participant, as well as 2.5 million genetic variants, with an exponential increase in publication output over time. Selected findings are summarised in Box 3.
Lessons learned: ingredients for a successful cohort study
Start with a good idea
The original “good idea” for this study was Mary Raine’s bequest to investigate “the nature, origin and cause of disease in human beings”. The obstetrician’s grant application made many references to fetal origins of adult illness, although it preceded the “developmental origins of health and disease” (DOHaD) theory by several years.
Initial funding
Projects requiring ongoing financial support, where the most important research outcomes may take decades, do not intrinsically appeal to funding bodies. How then to fund the establishment of a cohort? The Raine researchers designed the cohort as the follow-up of a randomised controlled trial (RCT) funded by the National Health and Medical Research Council (NHMRC). An RCT structure has funding appeal with potential for novel findings in a defined period, and allows simultaneous recruitment and collection of baseline cohort data. Participants in each arm of the RCT have had the same cohort involvement.
Planning for future work
The RCT was well designed, but the practicalities of cohort follow-up were initially ill defined. In the early years, volunteer staff maintained the cohort on a shoestring budget until ongoing funding was established. Planning for the future should engage not only investigators and institutions relevant to the initial phase, but also those who will be involved later as the cohort ages and the research focus changes.
Ongoing funding
A cohort cannot survive on short-term project grants; ongoing funding is necessary for staff and infrastructure. Initially, core funding was provided through NHMRC program grants to the Telethon Institute for Child Health Research. More recently, core funding has been obtained through “institutional buy-in”, whereby universities and research foundations associated with the study each contribute a set amount of funds over a 5-year period. Beyond the financial benefit, “investment” in the cohort encourages institutional research commitment. Multiple funding sources ensure that no single organisation “owns” the cohort or has exclusive access. Further funding is obtained by charging collaborators for data access.
Executive governance
The Raine Executive is chaired by the Dean of the UWA Faculty of Medicine, Dentistry and Health Sciences and includes the original investigators, a Raine Medical Research Foundation representative, an Emeritus Professor of Medicine and a scientific director. The Executive controls data access, and all proposed research projects, manuscripts and collaborations must be submitted in writing. The aim is a transparent, centralised approval process, led by an independent body prioritising the interests of the cohort. It ensures that research is not duplicated, all projects are of scientific merit, authorship issues are addressed early, and cohort data are protected. The Executive also determines which data are collected at each follow-up, and develops a long-term plan for the cohort. The Executive was formed when the cohort was aged 15, but the lesson from the Raine study experience is that it is important to establish a governance structure from the outset.
Avoiding participant loss to follow-up3,4
Initial data collection: As much demographic data as possible should be recorded. Many families will move house while their children are young, so grandparents’ details must be collected. A process for checking and updating contact information, including email, mobile phone and even Facebook, should be established.
Engagement: The families need to feel that they are “part of something”. The Raine study management maintains contact with the cohort through newsletters, birthday cards, a kids’ club and social networking websites. Large functions were organised for the cohort’s 10th and 21st birthdays. Engagement must be age-specific and relies on energetic staff involvement.
Cohort overload: The “burden” of research activity is carefully controlled, including length of questionnaires,5 use of invasive procedures, frequency of assessments (every 2–4 years) and even the number of attempts to contact participants to rebook appointments. Investigators must be aware of what else is happening in participants’ lives; an extensive follow-up in the year of the cohort’s final high school examinations may not be advisable.
Consultation: Feedback and consultation ensures research remains relevant and acceptable to participants. A small group of cohort members are selected as representatives, and they work closely with Raine study management (Box 4). Participant workshops are used to plan new research (such as projects in fertility and genetics), identify what information should be given back (retinal photographs and dual-energy x-ray absorbtiometry [DEXA] scan results) and gauge what the cohort would (or would not) be prepared to do.
Record linkage: Linking the cohort to total population datasets is an effective and inexpensive way to increase information on the cohort, including those lost to follow-up. These data may be obtained from hospitalisation, disability and mental health contacts and specific disease registers. Linking back into total birth information can assess how well the cohort reflects the general population. As the Western Australian linkage capacity includes data from educational, child protection and justice agencies, the cohort can be tracked in all contacts with these services, with further potential in linkage to national datasets such as the Pharmaceutical Benefits Scheme.
Impeccable ethics
Confidentiality must be maintained, and only de-identified data distributed to researchers. An auditing process should be established to ensure database integrity. When findings are potentially controversial, they must be managed carefully; the facts do not change, but presentation might. Every question must have a valid reason for being asked, with prioritisation of scientific rigour and research output.
What does the future hold? The next 20 years of the Raine study
Twenty-three research groups currently work with the Raine cohort, and international collaborations are expanding. The value of the Raine cohort will increase over time, particularly in later adult life when the major chronic diseases have developed, allowing assessment of genetic and environmental influences across the lifespan. At that point we will truly be fulfilling Mary Raine’s philanthropic request in “seeking, diagnosing and investigating the nature, origin and cause of disease in human beings ... and the prevention, cure, alleviation and combating of such diseases”.
1 Mary Raine (1877–1960) bequeathed her property empire to the University of Western Australia for medical research
2 Number of participants attending each follow-up assessment during the first 20 years of the Western Australian Pregnancy Cohort (Raine) Study
* Only volunteers for the Trier social stress test2 attended the 18-year follow-up; no overall cohort assessment was held at this time.
3 Selected research findings from the Western Australian Pregnancy Cohort (Raine) Study and related publications
Competing interests
No relevant disclosures.
Acknowledgements
We wish to sincerely thank the Raine study participants and their families, and the Raine Study Team for coordination and data collection. We acknowledge UWA, the Raine Medical Research Foundation at UWA, Telethon Institute for Child Health Research, UWA Faculty of Medicine, Dentistry and Health Sciences, the Women and Infants Research Foundation and Curtin University for providing core management funding. For further information about the Raine study, please contact Craig Pennell (craig.pennell@uwa.edu.au) or see http://www.rainestudy.org.au/.
References
- Newnham JP, Evans SF, Michael CA, et al. Effects of frequent ultrasound during pregnancy: a randomised controlled trial. Lancet 1993; 342: 887-891. 0_CHDIFBDH
- Kirschbaum C, Pirke KM, Hellhammer DH. The ‘Trier Social Stress Test’—a tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology 1993; 28: 76-81. 0_i1115733
- Booker CL, Harding S, Benzeval M. A systematic review of the effect of retention methods in population-based cohort studies. BMC Public Health 2011; 11: 249. 0_i1115735
- Hunt JR, White E. Retaining and tracking cohort study members. Epidemiol Rev 1998; 20: 57-70. 0_i1115738
- Edwards P, Roberts I, Clarke M, et al. Methods to increase response rates to postal questionnaires. Cochrane Database Syst Rev 2007; (2): MR000008. 0_i1115740
Provenance: Not commissioned; externally peer reviewed.