Recruiting, retaining and gathering complete data on participants in research projects, be they patients or health professionals, can be extremely difficult. These problems increase the risk that research will be abandoned before its true value is appreciated, or lead to delays in resolving uncertainty for decision makers, while further studies are done. Poor recruitment, retention and outcome collection frequently lead to many prospective studies being extended, increasing costs. Researchers need to use strategies that are themselves evidence-based. This study proposes to link with an existing longitudinal ageing study called NICOLA to provide evidence on what research participants prefer in relation to providing personal information through a self-completed questionnaire. NICOLA is a large study of people over the age of 50 that is being conducted in Northern Ireland. NICOLA is aiming to recruit 8500 people and will ask them questions about participation in social activities, including organised structured and informal activities; relationship quality; loneliness; stress; resilience; quality of life; alcohol intake; food poverty and assess their health and wellbeing. Participants in NICOLA agree to having an interviewer visit them at home to ask questions about their lives, complete questionnaires in their own time and attend a health assessment appointment. They also agree to being followed up over a course of at least 10 years. The research described here will examine the impact of differing times and formats of a self assessment questionnaire on completion rates, specifically:

• To assess the effect of being given a questionnaire during a face to face interview with a researcher compared to receiving it by post
• To explore the potential impact of interview fatigue on completion rates

The Northern Ireland Cohort for the Longitudinal Study of Ageing (NICOLA) is an opportunity to conduct methodology research relevant to many features of a large prospective study. NICOLA will begin in earnest in Northern Ireland in 2013 and is being conducted by a multidisciplinary team in the Centre for Public Health at Queen's University Belfast. It is an omnibus programme of research on ageing that will continue for at least 10 years and will recruit 8500 middle-aged people in Northern Ireland and follow them into old age, providing a comprehensive assessment of their physical and mental health, their lifestyles, and their social and economic decision making. NICOLA will look at how they perceive disability and health, and how this differs between well-off and disadvantaged groups. NICOLA will also study various genetic, biological and psychological factors, including how participants perceive risk and value their time, and the effect of this on their retirement behaviour (including how they manage their money and their health). People over the age of 50 will be invited to take part and asked to complete detailed interviews and questionnaires every 2 years and health assessments every 4 years. In this randomised trial, we will assess the effect of given them the self-assessment questionnaire at different time points, and on paper or online.

Cohort and other studies with longitudinal designs often struggle to maintain their recruited participants across all the study time points and to obtain complete data. The problems reduce the effective sample size quality of the evidence and robustness of the findings, and can undermine the often substantial investment in such studies. Researchers are devoting increasing amounts of time and resources to understand how participants decide to take part in research and what researchers can do to make their work appeal to potential participants. We will explore some of these issues within NICOLA and, to our knowledge, no similar work has been conducted solely within an over 50 population.

Establishing rapport with the participants has been mentioned as another method to increase response rate and we will explore this by examining whether receiving a questionnaire in a face-to-face exchange increases the likelihood of completion - a topic that was not addressed in the aforementioned systematic review.

Our randomised comparison of online versus paper questionnaires will gauge the interest in online methods among the over 50s, and the effects of the different modes on completion and completeness. Embedding this research at the beginning of a large cohort study allows the opportunity to test different methodologies to discover what works best at with regards to the delivery of questionnaires

The overall aim of this research is to examine the impact of differing times and formats of a self assessment questionnaire on completion rates. Specific objectives include:

• Assessing the effect of being given a questionnaire during a face to face interview with a researcher compared to receiving it by post
• Exploring the potential impact of interview fatigue on completion rates

The proposed study is a randomised trial of different timings and methods of offering a self-completion questionnaire. It will be embedded in an ongoing cohort study. All participants who complete the CAPI (Computer Assisted Personal Interview) will be invited to fill in a self-completion (paper) questionnaire whose contents will include measures of: participation in social activities, including organised structured and informal activities; relationship quality (partners/relatives/friends); loneliness; perceived stress; resilience; quality of life; alcohol intake and a measure of food poverty.

After participants have agreed to take part in the study, they will be randomised into one of two different trial arms:

1. Receive a paper based questionnaire immediately after their CAPI, handed to them by their Interviewer
2. Receive a paper based questionnaire in the post approximately one week after their CAPI interview

Inclusion Criteria:

• Participants must be aged 50 or over.
• Participants must be living in an non-institutionalised environment.
• Participants must be capable of providing informed consent.

Exclusion Criteria:

• Participants who are 49 or younger.
• Participants who are institutionalised.
• Participants who are not capable of providing informed consent.