What do we want to know?

The UK Millennium Cohort Study (MCS) is a longitudinal interdisciplinary study following the lives of 19,000 children born in the UK in 2000/1. The study recruited families of children born in randomly selected electoral wards, disproportionally stratified to boost representation of children from disadvantaged and ethnic minority families; further booster data sweeps have resulted in additional data being collected for Scottish and Welsh children. Information has been collected at 9 months, 3, 5, 7 and 11 years, with the next sweep of data collection underway among study members who are aged 14 years. A wide range of data have been collected from children, parents and guardians, the partners of parents/guardians, older siblings and teachers, as well as sub-studies that collected data from health visitors; these include self-reported and objectively measured/verified data.

This study sets out to examine how MCS data are utilised. To fit within the remit of the study, we hone in on ten priority question areas (Strengths and Difficulties Questionnaire, Child Social Behaviour Questionnaire, Diet, BMI, Immunisations, School Dis/like, Self-reported Friendships, Self-reported feelings, Screen Time, Hobbies).

Who wants to know?

This project was commissioned by investigators of the MCS but the results may also be useful for investigators working with other datasets.

What did we find?

• In total we found 481 unique studies that were using MCS data and undertaking primary analysis. 
• Data that are collected through a recognised scale with defined thresholds or cut-off points for identifying constructs of interest and/or data that can provide a unique insight into a policy-relevant issue, are those most widely used in the MCS data.
• Measures that have been collected across sweeps – diet, BMI, SDQ and screen time - are all comparatively well used. Those measures that have started to be collected at age 7 (and first made available in 2010) have had lower usage.
• Data that were collected from the child’s own reports (e.g. friendships and feelings) have seldom been utilised in comparison to data collected through parental reports (e.g. SDQ).
• Collection of data from multiple informants did not always correlate with higher levels of usage.
• Data on immunisations at age 3 and 5 did not feature in a great number of publications. One of the unique strengths of the MCS immunisation data is that they were able to directly reflect and address the research needs of policy-makers in terms of understanding antecedents of MMR uptake.
• Imposing thresholds on data was found to be problematic in some cases, for example for BMI, where a number of different thresholds for overweight and obesity were in use. The use of different thresholds can lead to substantial differences in the results obtained.
   

What are the implications?

This is the first review using systematic methods that has explored MCS data use. We set out ideas for good practice around the use of and reporting of MCS data. These include:

• Encouraging notification of publication: Emphasising the importance of notifying CLS of publications using MCS data at the stage of accessing data may boost the number of studies that are traced.
• Further development of a community of users (requires further funding/investment): Establishing a searchable database of MCS users on the CLS website could help to further foster the community of users.
• Recording publications with study level meta-data (requires further funding/investment): Enhancing the functionality of the CLS library could allow for the recording of a greater number of fields on publications and a more efficient means of searching the library.
• Publishing the case for variables: Most variables/measures included in MCS surveys go through a process of consultation which involves a written case being made by the research team/users for their inclusion. We recommend that a record of this process is published for new variables to allow users to understand why variables have been suggested (and included).
• Nominating variable champions: Each variable that is included in MCS surveys is included usually after a written case has been made. ‘Variable champions’ could be nominated to provide some support or engage with in discussion around the use of those variables.
• Making MCS studies more identifiable: Further guidance or emphasis of the importance of naming of MCS in publications’ titles/abstracts/keywords may facilitate other reviews of data usage in future, and may give additional prominence to the study in the literature.

How did we get these results?

We achieved these results by undertaking systematic searches across 30 databases which generated over 4000 results. We then searched these records, first on title and abstract and then on the full text and extracted data on studies that fell within our specific areas of interest.

This report should be cited as:

Kneale D, Patalay P, Khatwa M, Stansfield C, Fitzsimmons E, Thomas J (2016) Piloting and producing a map of Millennium Cohort Study Data usage: Where are data underutilised and where is granularity lost? EPPI Centre, Social Science Research Unit, UCL Institute of Education, University College London