Dr Lucy Foulkes
@lucyfoulkes.bsky.social
Psychologist at University of Oxford | Adolescence + mental health
Linktr.ee/lucyfoulkes
Linktr.ee/lucyfoulkes
I think there is something distinct about schools which makes en masse interventions particularly problematic (in terms of social cog development but also the intensity of time spent with peers) - but I also suspect adults wouldn't want to do en masse MH interventions w colleagues...
October 23, 2025 at 3:10 PM
I think there is something distinct about schools which makes en masse interventions particularly problematic (in terms of social cog development but also the intensity of time spent with peers) - but I also suspect adults wouldn't want to do en masse MH interventions w colleagues...
link to that paper here: link.springer.com/article/10.1...
October 23, 2025 at 3:08 PM
link to that paper here: link.springer.com/article/10.1...
yes for sure - it's a good point. We have written elsewhere about how many young people don't feel safe or comfortable at school/around their peers and that therefore it's a poor environment to expect them to be vulnerable about their MH.. this would fit in to that
October 23, 2025 at 3:07 PM
yes for sure - it's a good point. We have written elsewhere about how many young people don't feel safe or comfortable at school/around their peers and that therefore it's a poor environment to expect them to be vulnerable about their MH.. this would fit in to that
At the same time, we can find out if there are subgroups especially likely to benefit from interventions and why. Ultimately, this will make school approaches to MH more safe and effective for everyone
more info in paper - open access here:
www.sciencedirect.com/science/arti...
more info in paper - open access here:
www.sciencedirect.com/science/arti...
Potential harm from universal school-based mental health interventions: Candidate mechanisms and future directions
Universal school-based mental health interventions involve lessons delivered to whole classes of young people irrespective of need, with the overall a…
www.sciencedirect.com
October 23, 2025 at 11:54 AM
At the same time, we can find out if there are subgroups especially likely to benefit from interventions and why. Ultimately, this will make school approaches to MH more safe and effective for everyone
more info in paper - open access here:
www.sciencedirect.com/science/arti...
more info in paper - open access here:
www.sciencedirect.com/science/arti...
We therefore suggest a compromise: if more trials *are* run, they should be optimally designed to test for possible mechanisms and individual differences. We give detail about how to do this
This will enable us to understand why negative effects sometimes happen in trials and who is most at risk
This will enable us to understand why negative effects sometimes happen in trials and who is most at risk
October 23, 2025 at 11:54 AM
We therefore suggest a compromise: if more trials *are* run, they should be optimally designed to test for possible mechanisms and individual differences. We give detail about how to do this
This will enable us to understand why negative effects sometimes happen in trials and who is most at risk
This will enable us to understand why negative effects sometimes happen in trials and who is most at risk
BUT, the best quality trials suggest that these interventions will either show null or negative effects. It might therefore be unethical to run more trials - a potential waste of money that puts some young people at risk of negative effects
October 23, 2025 at 11:54 AM
BUT, the best quality trials suggest that these interventions will either show null or negative effects. It might therefore be unethical to run more trials - a potential waste of money that puts some young people at risk of negative effects
We write that the research field therefore now faces a dilemma
The ideal way to answer the two remaining questions – why negative effects happen and who is most at risk – would be to run more trials of universal mental health interventions...
The ideal way to answer the two remaining questions – why negative effects happen and who is most at risk – would be to run more trials of universal mental health interventions...
October 23, 2025 at 11:54 AM
We write that the research field therefore now faces a dilemma
The ideal way to answer the two remaining questions – why negative effects happen and who is most at risk – would be to run more trials of universal mental health interventions...
The ideal way to answer the two remaining questions – why negative effects happen and who is most at risk – would be to run more trials of universal mental health interventions...
Re individual differences:
The literature to date has shown many possible subgroups that might be at increased risk, but nothing consistent enough to be sure
The literature to date has shown many possible subgroups that might be at increased risk, but nothing consistent enough to be sure
October 23, 2025 at 11:54 AM
Re individual differences:
The literature to date has shown many possible subgroups that might be at increased risk, but nothing consistent enough to be sure
The literature to date has shown many possible subgroups that might be at increased risk, but nothing consistent enough to be sure
Re mechanisms, there are a few possibilities:
1. Intervention is making ppts more accurately report previously underreported symptoms (best case scenario)
2. Intervention is making ppts feel worse - eg encouraging rumination on symptoms, awareness not paired with support
So far we don't know
1. Intervention is making ppts more accurately report previously underreported symptoms (best case scenario)
2. Intervention is making ppts feel worse - eg encouraging rumination on symptoms, awareness not paired with support
So far we don't know
October 23, 2025 at 11:54 AM
Re mechanisms, there are a few possibilities:
1. Intervention is making ppts more accurately report previously underreported symptoms (best case scenario)
2. Intervention is making ppts feel worse - eg encouraging rumination on symptoms, awareness not paired with support
So far we don't know
1. Intervention is making ppts more accurately report previously underreported symptoms (best case scenario)
2. Intervention is making ppts feel worse - eg encouraging rumination on symptoms, awareness not paired with support
So far we don't know
However, two questions remain unanswered:
1. Why do universal interventions sometimes have negative effects? (mechanisms)
2. Are some young people more vulnerable to negative effects from these lessons than others? (individual differences)
1. Why do universal interventions sometimes have negative effects? (mechanisms)
2. Are some young people more vulnerable to negative effects from these lessons than others? (individual differences)
October 23, 2025 at 11:54 AM
However, two questions remain unanswered:
1. Why do universal interventions sometimes have negative effects? (mechanisms)
2. Are some young people more vulnerable to negative effects from these lessons than others? (individual differences)
1. Why do universal interventions sometimes have negative effects? (mechanisms)
2. Are some young people more vulnerable to negative effects from these lessons than others? (individual differences)
It is now well-established that universal school MH interventions can have a range of negative effects - i.e. something we hoped would improve ended up getting worse
(We found at least one neg effect in 33% of high-quality CBT and mindfulness trials; Guzman Holst et al 2025)
(We found at least one neg effect in 33% of high-quality CBT and mindfulness trials; Guzman Holst et al 2025)
October 23, 2025 at 11:54 AM
It is now well-established that universal school MH interventions can have a range of negative effects - i.e. something we hoped would improve ended up getting worse
(We found at least one neg effect in 33% of high-quality CBT and mindfulness trials; Guzman Holst et al 2025)
(We found at least one neg effect in 33% of high-quality CBT and mindfulness trials; Guzman Holst et al 2025)
thank you for all your expert help 🙏
October 17, 2025 at 11:18 AM
thank you for all your expert help 🙏
October 10, 2025 at 3:40 PM
Instead, we may be better off focusing our efforts on targeted support (1:1 or small group) for those who want or need it
I argue this further in the paper below
acamh.onlinelibrary.wiley.com/doi/full/10....
(with thanks to Carolina et al for all their great work getting this paper out 🙏)
I argue this further in the paper below
acamh.onlinelibrary.wiley.com/doi/full/10....
(with thanks to Carolina et al for all their great work getting this paper out 🙏)
October 10, 2025 at 3:36 PM
Instead, we may be better off focusing our efforts on targeted support (1:1 or small group) for those who want or need it
I argue this further in the paper below
acamh.onlinelibrary.wiley.com/doi/full/10....
(with thanks to Carolina et al for all their great work getting this paper out 🙏)
I argue this further in the paper below
acamh.onlinelibrary.wiley.com/doi/full/10....
(with thanks to Carolina et al for all their great work getting this paper out 🙏)
We have to be cautious: of course, there are far too many YP that do have these problems and aren’t getting the help they need
But it’s another reminder that universal interventions aren’t a good idea, because many people in each class don't need help (and these interventions don't help MH anyway)
But it’s another reminder that universal interventions aren’t a good idea, because many people in each class don't need help (and these interventions don't help MH anyway)
October 10, 2025 at 3:36 PM
We have to be cautious: of course, there are far too many YP that do have these problems and aren’t getting the help they need
But it’s another reminder that universal interventions aren’t a good idea, because many people in each class don't need help (and these interventions don't help MH anyway)
But it’s another reminder that universal interventions aren’t a good idea, because many people in each class don't need help (and these interventions don't help MH anyway)
For the measure of depressive symptoms, 5381 (69.8%) were in the low-low-low group
For total SDQ score, 4938 (64.2%) were low-low-low
This is important and encouraging. It tells us that, across about a year, the majority of young people do not have mental health problems
For total SDQ score, 4938 (64.2%) were low-low-low
This is important and encouraging. It tells us that, across about a year, the majority of young people do not have mental health problems
October 10, 2025 at 3:36 PM
For the measure of depressive symptoms, 5381 (69.8%) were in the low-low-low group
For total SDQ score, 4938 (64.2%) were low-low-low
This is important and encouraging. It tells us that, across about a year, the majority of young people do not have mental health problems
For total SDQ score, 4938 (64.2%) were low-low-low
This is important and encouraging. It tells us that, across about a year, the majority of young people do not have mental health problems
Let's go back to this graph and look at the pink line. This line represents the students who were low in anxiety before the study, low in anxiety after it, and low 12 months later
76.6% of participants (5514 students) were in this
low-low-low group
76.6% of participants (5514 students) were in this
low-low-low group
October 10, 2025 at 3:36 PM
Let's go back to this graph and look at the pink line. This line represents the students who were low in anxiety before the study, low in anxiety after it, and low 12 months later
76.6% of participants (5514 students) were in this
low-low-low group
76.6% of participants (5514 students) were in this
low-low-low group
Why does this matter?
It contributes to the growing argument that universal school interventions are not the answer to improving MH in young people
But this study highlighted something else that I think is important... it showed that the majority of young people do not have mental health problems
It contributes to the growing argument that universal school interventions are not the answer to improving MH in young people
But this study highlighted something else that I think is important... it showed that the majority of young people do not have mental health problems
October 10, 2025 at 3:36 PM
Why does this matter?
It contributes to the growing argument that universal school interventions are not the answer to improving MH in young people
But this study highlighted something else that I think is important... it showed that the majority of young people do not have mental health problems
It contributes to the growing argument that universal school interventions are not the answer to improving MH in young people
But this study highlighted something else that I think is important... it showed that the majority of young people do not have mental health problems
This suggests that the average overall findings found in MYRIAD are not masking specific hidden subgroups that experience positive or negative effects
We don't know if this applies to other universal interventions that show null/negative effects - Carolina is looking at this now
We don't know if this applies to other universal interventions that show null/negative effects - Carolina is looking at this now
October 10, 2025 at 3:36 PM
This suggests that the average overall findings found in MYRIAD are not masking specific hidden subgroups that experience positive or negative effects
We don't know if this applies to other universal interventions that show null/negative effects - Carolina is looking at this now
We don't know if this applies to other universal interventions that show null/negative effects - Carolina is looking at this now
Our key finding was that having mindfulness lessons (vs being in the control group) didn’t predict which of these trajectory groups a student would end up in
There was variety in students’ mental health trajectories, it just didn’t link to whether they'd been in the mindfulness vs control group
There was variety in students’ mental health trajectories, it just didn’t link to whether they'd been in the mindfulness vs control group
October 10, 2025 at 3:36 PM
Our key finding was that having mindfulness lessons (vs being in the control group) didn’t predict which of these trajectory groups a student would end up in
There was variety in students’ mental health trajectories, it just didn’t link to whether they'd been in the mindfulness vs control group
There was variety in students’ mental health trajectories, it just didn’t link to whether they'd been in the mindfulness vs control group