Factors influencing use and choice of Core Outcome Sets and outcome measurement instruments in trials of interventions to prevent childhood obesity: A survey protocol

Introduction

Approximately 39 million children under the age of 5 are living with overweight or obesity globally¹. The aetiology of childhood obesity is multifaceted, with childhood obesity arising from a complex interplay of individual, behavioural, interpersonal, environmental, and structural factors^2,3. While some interventions developed to prevent childhood obesity and/or improve child health behaviours related to childhood obesity (e.g., diet, physical activity), have demonstrated benefits, findings are inconsistent overall. A potential reason for this inconsistency is the heterogeneity in outcomes being measured and reported in intervention trials^4,5, which limits our ability to compare findings and identify effective interventions⁶. For example, in a previous review of outcomes examined in the infant feeding literature, 126 studies reported 236 different outcomes⁵. Similarly, in a review of outcomes reported in early childhood obesity prevention interventions, 221 outcomes were reported in 161 studies⁴. Standardisation of what outcomes are measured, and how they are measured, can address this issue.

To address heterogeneity in outcome measurement in trials of interventions to prevent childhood obesity, two Core outcome sets COS have been developed^7,8 (see Extended data). Core outcome sets (COS) are agreed-upon standardised sets of outcomes representing what to measure and report in all trials in a specific health area⁶. The first COS in this area is a COS for infant feeding interventions conducted with children aged from birth to 1 year to prevent childhood obesity⁷. It includes 26 outcomes, such as ‘responsive infant feeding’, ‘portion size’, ‘duration of exclusive breastfeeding’, and ‘child weight’⁷. The second COS (COS-EPOCH) is for use in trials of early interventions in children aged up to 5 years to prevent obesity in childhood⁸. It includes 22 outcomes, such as child physical activity, screen time, meal patterns, and sleep duration⁸. Dietary outcomes in COS-EPOCH⁸ are related to children from 1–5 years, with infant feeding outcomes up to 1 year included in the infant feeding COS⁷. As such, outcomes from both COS can be used together or separately depending on the intervention focus and/or child age. These COS represent what to measure in childhood obesity prevention interventions, and work is currently ongoing to determine how to best measure these outcomes through the development of standardised sets of outcome measurement instruments for the COS outcomes^9,10.

These developments will only improve evidence syntheses and identification of effective interventions to prevent childhood obesity to the degree that they are actually used by researchers¹¹. There is evidence of low use of COS across different health areas^12–16. Barriers to the use of COS across general cohorts of health areas include knowledge about what COS are and how to identify and use them, measurement issues associated with COS use, and trialists’ own outcome preferences^12,14,17,18. While developing standardised measurement sets helps to address issues around the identification of measurement instruments for COS, it is important that recommendations of measurement instruments consider previously identified issues related to measurement properties (e.g., validity, reliability), feasibility, acceptability, cost, and participant and administration burden^19,20. While research has been conducted on factors influencing the use of broad cohorts of COS^12,14,18 and choice of measures, such as patient-reported outcome measures^19,20, such work has not been conducted in the area of childhood obesity.

It is important to understand attitudes of researchers working on trials of interventions to prevent childhood obesity and/or to improve child health behaviours related to childhood obesity, about existing COS developed for this area. It is also important to identify and understand factors influencing their use of these COS. Such understanding will inform approaches to maximise future COS use. Similarly, understanding the most important factors for trialists working in childhood obesity prevention regarding choice of outcome measurement instruments ensures that recommended outcome measurement instruments are appropriate, feasible, and acceptable. Therefore, the aim of the current study is threefold: 1) to examine trialists’ awareness and attitudes towards the two current COS for trials of childhood obesity prevention interventions, and factors influencing use of these COS; 2) to explore the characteristics of outcome measurement instruments that trialists currently use in obesity prevention interventions and/or interventions to improve child health behaviours related to childhood obesity; and 3) to better understand how trialists choose outcome measurement instruments in trials of childhood obesity prevention interventions, and/or interventions to improve child health behaviours related to childhood obesity, and the factors that influence those choices.

Methods

Conclusion

Examining the factors that influence trialists use of childhood obesity prevention core outcome sets, and their choice and use of outcome measurement instruments can inform approaches to maximising uptake of these resources in future trials. Use of core outcome sets and standardised outcome measurement sets in interventions to prevent childhood obesity can reduce research waste and outcome hetereogeneity and improve evidence syntheses in this area.

Ethical issues

Ethical approval will be obtained prior to study commencement from the University College Cork Social Research Ethics Committee. Due to the nature and focus of this study, there are no anticipated ethical issues, and it is not expected that participants will experience any form of distress. Members of the study team (e.g., KMS, LS, BJ, SR, HS) may have existing research collaborations with survey participants, however identifiable trialist and trial data will not be collected in the survey and raw data will only be accessed by KMS, RA, MD and EL to minimise any risk of identification based on data collected. In addition, the study will include the following ethical considerations: All participants will be provided with an online participant information leaflet prior to providing informed consent and completing the online survey. The participant information leaflet will include information about the aims and processes of the study. This includes that participation is voluntary and that individual participant data will be confidential and anonymous as no identifying information will be collected. It will also include information on participants right to withdraw from the study at any point up to survey completion as individual submissions will not be identifiable due to the anonymous nature of the survey. The information leaflet will also include data management and storage information, which will be conducted in line with the University College Cork Code of Research Conduct.

Dissemiantion: Findings from this study will be disseminated as a peer-reviewed publication and as a presentation at conferences and methodological seminars.