The Administration of Web-Based Questionnaires in UK Biobank, a Prospective Cohort Study of 500,000 UK Adults

Jo Holliday, Thomas Littlejohns, Naomi Allen, Nicola Doherty, Cathie Sudlow

Introduction: In an increasingly digital environment, web-based questionnaires are a convenient and cost-effective way of collecting data from participants in large-scale health research studies. This study reports on the administration of web-based questionnaires in UK Biobank, a prospective cohort study of 500,000 men and women aged 40-69 years.

Methods: UK Biobank is an open-access resource designed to investigate the genetic, environmental and lifestyle determinants of a wide range of diseases of middle age and later life. During baseline assessment (conducted between 2006 and 2010), participants provided extensive self-report data, physical measures and biological samples, and consent for long-term follow-up through electronic medical records and to be re-contacted by UK Biobank. Enhanced phenotyping is ongoing and includes the administration of web-based questionnaires to those participants with an e-mail address (approx. 330,000 participants). Questionnaires have focused on the collection of more detailed exposure data (e.g., a series of four 24-hour dietary recalls over a 16-month period; a questionnaire on occupational history) and on health outcomes that are difficult to assess through linkage to health records (e.g., cognitive function; mental health).

Results: The response rate for UK Biobank’s series of web-based questionnaires is 36%-42%, with higher response rates obtained with the use of reminder emails sent to non-responders. Participants who have responded to one or more questionnaires are more likely to be older, female, more educated, non-smokers, drink alcohol on a daily basis, have a lower body mass index and to live in less socioeconomically deprived area compared to non-responders.

Conclusion: The use of web-based questionnaires in UK Biobank has been successful at obtaining in-depth information from participants on a large-scale and at low-cost. The implications of observed differences in characteristics between responders and non-responders for the long-term follow-up of UK Biobank participants will be considered. In particular, what approaches can be developed to encourage the participation of certain subgroups with low response rates. The use of web-based approaches for collecting research data more generally will be discussed.

This abstract was submitted to the 2017 Society for Prevention Research Annual Meeting.