A Non-Randomised Controlled Study of Interventions Embedded in the Curriculum to Improve Student Wellbeing at University

Round 1

Reviewer 1 Report

Thank you for the opportunity to read your paper. I found it thorough and very well presented.

The literature was up-to-date, the expression accessible, the analysis thorough and the explanation of the non-significant findings convincing.

It might be worth adding a sentence or two more about the qualitative and quantitative methods that could be adopted. Averaging over classes may conceal some marked individual shifts and methods that look at critical incidents or similar may be useful.

Additionally while it may be difficult to positively affect whole cohorts’ well-being we do know that our practices, of say failing people or marking that is perceived to be unfair, can detrimentally affect well-being. It may be that the way to improve students well-being is not to teach them about well-being but to create a positive safe and effective learning environment.

My only other suggestion is about the tables – the tables are very detailed so it is difficult to grasp the messages you are trying to convey.  It would help your reader if summary tables were embedded in the text.

Line 233 – delete ‘a’

Author Response

Please see the attachment. 

Author Response File: Author Response.pdf

Reviewer 2 Report

Introduction:

My only suggestion is to give a bit more depth around your approach to curriculum-embedded interventions and what this adds given the existing evidence is limited. Is it simply the case that you are aiming to more rigorously evaluate an intervention like this, or have you looked to the existing research and taken what 'active ingredients' seem most promising to inform the intervention itself (or both?)

Methods: 

Also a bit more detail on the design (e.g., a 2 x 2 mixed design with the between-groups variable being intervention vs control, much in the same way you have clarified the repeated measures variable). This is a simple change that I think would be helpful!

Having now read about the interventions it is clearer that these were not developed for the study and have a diverse range of approaches and underpinning mechanisms. It would be good to ramp up discussion of this in the introduction, and some of the background and relevant theory about the interventions that is currently in the method (such as alignment to 5 ways to well-being) would be better-placed in the introduction. As part of this, rationale for the outcome variables is needed. 

Table 2 is great and helped me understand overlap in module content/students involved as well as differences in approach. 

Please clarify the choice of control conditions a bit more i.e., was there any attempt to match with the intervention modules in terms of structure, teaching approach?

Why were some outcome measures dropped for time 2? Even with overlap identified from the time 1 data, it would have been better to have asked the same questions and potentially used the components identified in the PCA as outcome variables. 

Discussion:

It's good to note that qualitative research is needed where students can actually talk about what helps, what does not, what they would like etc. Co-production would also be good to inform novel curriculum-embedded interventions - if this is in fact something that students want. If they don't, and approaches have limited evidence, then it's also important to consider that these are potentially not what we should be pursuing!

Limitations:

The heterogeneity of the interventions, but also the modules they are part of, must be discussed. It may be aspects of the module that aren't to do with the well-being focused elements that are influencing outcomes. 

It's good to see you acknowledge the small proportion of students taking these modules completing the surveys, and I think a bit more about what this selection bias might look like and how it could be mitigated in future would improve this section. 

I'd also like some more information about why the control condition for some interventions is a different module, whereas for others it is comparing students on the same module based on attendance. I have concerns about this inconsistent approach to control groups, and the comparison with non-attenders vs attenders in particular. 

Author Response

Please see the attachment. 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments have been satisfactorily addressed and this paper now sets the scene much better, and more  adequately addressed the methodological concerns I had. Well done!