Feasibility and acceptability of developing core outcome sets in effectiveness trials for universal school-based public health interventions

Public Health (PH) interventions are unlikely to ‘work’ the same way widely. One consequence of this for PH trial design has been the use of discipline-specific, ad hoc or unvalidated outcome measures. This results in inconsistency of outcomes across trials of a similar intervention1, which in turn prevents meaningful evidence synthesis in systematic reviews due to heterogeneity of outcomes and high risk of bias. Systematic reviews are central to the development of new PH interventions2. Core outcome sets (COS) have been developed in clinical medicine as an agreed minimum set of outcomes to be measured and reported in all effectiveness trials.

Established methods for developing COS may not be optimal for PH interventions due to important methodological challenges: PH interventions have multiple active components, they are delivered to diverse participants who are typically healthy and outcomes are longer-term than clinical trials. Compared to clinical settings there is a wider range of settings in which the same intervention can be delivered and a wider range of stakeholders (e.g. GPs, Local Authorities, social workers, teachers, pupils, parents and PH consultants). Interventions designed to target the same health issue may be delivered to an individual, a group or be environmental/ structural. These factors are likely to cause greater diversity of relevant outcomes and gaining consensus for a minimum outcome set may be more difficult due to the competing interests.

Conversely, relevant stakeholders in a given PH setting (such as schools) are likely to remain the same regardless of disease area, hence we may have the opportunity to answer multiple questions in one consensus process (e.g. a core set for all school health interventions with a series of add-on modules for different disease areas within that setting). This would enable better efficiency of limited resources and time, and improve the design of future PH trials.

What studentship will encompass
This PhD considers school-based PH interventions for mental health, obesity and sexual health to answer the following methodological questions: (i) Can consensus of a PH COS be achieved across diverse stakeholder groups (ii) Is a modular COS system feasible and acceptable; can it be developed within one COS process? Are core outcomes generic to all preventive interventions in a given setting?

The PhD will follow the stages of COS development recommended by COMET3: (i) Phase 1- identification of a ‘long list’ of potential outcomes through systematic reviews of PH school-based interventions in the three disease areas (including trials, population-level surveys and local authority school surveys; (ii) Phase 2 - prioritization of outcomes using a Delphi survey; this will include 3 questionnaires or ‘rounds’, asking participants to rate the importance of different outcomes to be measured in future trials, average scores are fed back to participants in subsequent rounds, and; (iii) Phase 3 - consensus meeting to finalize core items and disease-specific modules. The degree of consensus achieved will be explored in each round by examination of: the percentage of discordant items (where there is disagreement between stakeholder groups); differences in scores between stakeholder groups; variability in scores across stakeholders4. Participant acceptability of the process and of a modular COS will be sought via qualitative interviews with a purposeful sample of stakeholders (approximately 20).

Detail of supervision
The supervisors have expertise in PH trials, systematic reviews and statistics. Dr Caldwell will support the systematic review aspects. Prof. Campbell is co-director of DECIPHer (centre for the Development and Evaluation of Complex Interventions for Public Health Improvement)5 which has a strong track record of evaluating PH interventions in RCTs. Dr. Brookes is a medical statistician, joint lead of the ConDuCT-II Hub outcomes theme and the HTMR Working Group on Outcomes in trials; she has contributed substantially to development of COSs.