B3423 - Characterising the ALSPAC mothers who are also UKBiobank participants - 29/11/2019

B number: 
B3423
Principal applicant name: 
Andy Boyd | University of Bristol
Co-applicants: 
Alison Teyhan, Mark Mumme, Richard Thomas, Nic Timposn
Title of project: 
Characterising the ALSPAC mothers who are also UKBiobank participants
Proposal summary: 

Many of the ALSPAC mums and fathers/partners may have also volunteered to take part in the UK Biobank cohort study. It is important that the study Data Managers and the researchers understand who is in both studies. This is because studies such as ALSPAC and UKBiobank are often used together in order to study rare events or small associations (where you need large numbers of participants for the statistical tests to work) or they are used to check and confirm whether findings in one study are also seen in another study. Finding the same patterns means there can be more confidence the findings are genuine, rather than occurring by chance or due to error. In both cases, the statistical tests assume the people in one study are different to the people in the other study. However we now know there is substantial overlap between participants in ALSPAC and UK Biobank (and possibly other studies).

ALSPAC and UKBiobank are ensuring that any duplication is flagged so researchers can take account of this (without knowing the identities of the participants). To inform thinking on how to best deal with this issue, it is necessary to produce descriptive statistics describing the characteristics of the ALSPAC participants who are in UKBiobank and how these differ from ALSPAC participants who are not in UKBiobank.

Impact of research: 
To inform users of the ALSPAC and UKBiobank resources and to aid study managers plan additional research questions, methodological research, data collection and participant communications & engagement strategies based on the overlapping sample.
Date proposal received: 
Wednesday, 27 November, 2019
Date proposal approved: 
Friday, 29 November, 2019
Keywords: 
Epidemiology, Statistical methods, Methods - e.g. cross cohort analysis, data mining, mendelian randomisation, etc.