Professor Lucy Bowes, Magdalen College, University of Oxford, and Head of the oRANGE Lab, discusses her research on early life stress in relation to psychological and behavioural development, the impact of bullying in adolescents, together with exciting developments with virtual reality.
Lucy’s research focuses on the impact of early life stress on psychological and behavioural development. In particular, Lucy has focused on the effects of victimization on young people’s adjustment and wellbeing. Her research integrates methods from social epidemiology, developmental psychology and behavioural genetics in order to understand the complex genetic and environmental influences that promote resilience to victimization and early life stress. The aim of her work is to guide intervention work by identifying protective factors that promote positive outcomes among vulnerable children.
Watch Lucy’s lecture on ‘Adolescent peer relationships and mental health: an epidemiological perspective’, from our Emanuel Miller Memorial Lecture and Conference 2018 – Focusing on Adolescent Mental Health.
Introduction: This podcast is brought to you by the Association for Child and Adolescent mental health, ACAMH for short. You can find more podcasts and other resources on our website www.acamh.org and follow us on social media by searching ACAMH.
Interviewer: Hello, welcome to the in conversation podcast series for the association for Child and Adolescent mental health or ACAMH for short. I’m Joe Carlow, a freelance journalist for the specialism in Psychology. Today I’m interviewing Dr. Lucy Bowes of the Department of Experimental Psychology at the University of Oxford. Lucy’s work focusses on the impact of early life stress on psychological and behavioral development. Lucy, thank you for joining me. Can you start with an introduction?
Dr. Lucy Bowes: Thank you very much. Yes, I’m an Associate Professor as you mentioned in the department of experimental psychology at Oxford and I’m the Head of the oRANGE lab which is the Oxford lab of risk and resilience genes and environment.
Interviewer: And your research focusses on the impact of early life stress on psychological development. What prompted your interest in this specific area?
Dr. Lucy Bowes: I think I was fascinated because we know that one of the biggest predictors of poor mental health is exposure to stressful life events, in particular events that involve interpersonal violence have been shown to be particularly stressful. But there’s a lot of variation in the ways in which people both experience this and also how they cope with these types of stressful experiences. And I find it really fascinating that although these are major risk factors not everyone will go on to develop poor mental health, and I was also struck by the fact that traditionally in research and particular in my area and sort of social epidemiology, we focused a lot on risk and what goes wrong and we’re quite good now at characterising risk pathways and things that seem to make people more at risk of developing poor mental health. But we know far less about what goes right, in fact what we can do to promote resilience and I think that’s what really fascinated me. So it feels like a missed opportunity because I think we can probably learn as much if not may be more from individuals and communities and from situations in which people display a resilient processes in the face of adversity as we can from looking at the negative outcomes.
Interviewer: I want to ask you about those in more detail, so let’s start with the social, psychological and perhaps epigenetic factors that may put a young person more at risk let’s say of developing psychological problems and then afterwards we’ll look at the flip side.
Dr. Lucy Bowes: So we knew for example that being the victim of violence and aggression, so experience in child abuse or being bullied at school and puts people at risk of developing depression and that risk loss across the life course. So for example in one study, we found that potentially up to a quarter of depression in our sample might be attributable to childhood bullying, because it is unfortunately quite common and because the effect size is quite large, but things like being bullied at school is also associated with risk of anxiety, suicidal ideation and attempts so very serious outcomes.
But again, there’s a lot of variation in outcome and it’s true that certain vulnerabilities probably make a person more likely to experience victimization in the first place. So in particular here, I’m talking about bullying victimization, I should say, and those same vulnerabilities may also mean that they’re more vulnerable than more susceptible to the negative effects of experiencing things like peer victimisation, so, using social factors, I guess the first thing to say is I’m interested in different forms of victimisation, but they are very correlated. So, for example, children who have been maltreated or experienced physical abuse at home are twice as likely to be bullied at school, which I find desperately sad.
We know, you know, thinking of broader social factors, we do see some gradients in terms of victimisation. So you see a greater prevalence found in more deprived families. You also see for example greater levels of bullying in neighborhoods that are characterised by violence. But I think it’s really important to note that victimisation is experienced by children across the socio-economic spectrum. So it’s by no means just you know, children, for example, growing up experiencing poverty.
We also know that bullying occurs in schools that have a greater social hierarchy, which I find very interesting. So there’s a theory for example that kids may bully others in order to get social rewards, though they can actually gain social status by doing this, and actually that gives a window of opportunity for intervention. So if we can change those social norms and we can promote active bystander behavior and I’ll talk a bit later about interventions that do that, we can actually reduce the prevalence of bullying. So, some of these big social constructs like social norms actually give us a window for change.
Interviewer: I’m interested when you talk about hierarchy, are you’re talking about a hierarchy created by the way the school is structured or is this something that happens in the playground?
Dr. Lucy Bowes: That’s a very good question and it’s probably a little bit of both. I say both because there have been studies that sadly I wasn’t involved in but other studies that have found that in schools where teachers are struggling and where there are governance issues, so it’s not quite hierarchy, but certainly they’re feeling that they are not you know that their perhaps not equal, you find higher prevalence rates of bullying among the schoolchildren, which is fascinating. From the teacher perspective you also if you look like you say that the children themselves and where you have where you have stronger social hierarchies that perhaps are being maintained and supported implicitly by the school norm school structures then you tend to see higher rates of bullying.
Interviewer: So these are things that could be mitigated for?
Dr. Lucy Bowes: They are, they are. I think what I find fascinating with these broader social factors is although they can seem thorny and tricky to change, they’re is always something that’s going to you know from a public health point of view they’re going to have a bigger impact if we can change the social norms to be more pro-social. It’s going to have a bigger impact on more people than for example, individual psychological therapies with vulnerable young people.
Interviewer: Later I was going to ask you about any policy changes that you would like to see but I think this probably relates to what you’re talking about, you know what can be done really.
Dr. Lucy Bowes: So, in terms of social policy changes, again, if we think about the context of bullying, every school must have a policy to deal with bullying. Yet schools are given very little support and information about what are the best evidence-based interventions to stop bullying in their school. So they’re sort of left with a situation where they’ve got to have a policy but what do they do? You know, do you draw on the information from schools around you but that might not be the best way of tackling bullying. I think we can look to Finland as a brilliant example.
So Finland had issues with bullying, they had some high profile cases of suicide that had apparently been linked to bullying. The University of Turku and Christina Salmivalli got together with the Finnish government to develop the KiVa program, and that was taken across Finland, so over 90% of the public schools in Finland were implementing this anti-bullying intervention and they’ve tracked the progress year on year and they’ve seen rates of bullying coming down and they’re now in a situation where teachers that have just finished their training will actually go and work with staff using KiVa in the school. So they really instilling these values earlier on in teachers. That’s brilliant. We’ve got an in intervention that’s been shown to work has been shown to take them to scale. We need things like that here and that’s what we don’t have.
Interviewer: What is the intervention? Can you describe it?
Dr. Lucy Bowes: Yes, of course. So KiVa is an intervention that works not just in tackling the individual cases of bullying, although it does do that, but it really works on the bystander module. And that’s the finding that bullying doesn’t tend to occur in isolation, it tends to occur in front of me. And actually what you find is where bullying does occur in front of people, if those children who witness bullying do not do anything about it, perhaps being unable or do not you know don’t realize what’s going on. Actually, what happens is they’re providing implicit consent for the bullying making the impact of the bullying of the young person experiencing it can actually be worse. But it’s also a missed opportunity because they’re the majority of the group these bystanders and they probably feel like they’re not involved at all.
They will tell you quite rightly that they’re not bullying anyone. But actually they’re playing an important role and empowering kids, helping kids to understand that role and to challenge these behaviors when they see them in ways that they feel safe and able to do is a core part of this program. So it’s promoting active bystander behavior. It also involves the talk component as well to increase empathy. So understanding, you know for kids sat alone at break time or in the lunch hall what can you do to make them feel more supported and what might that do in terms of their experience at school.
Interviewer: And are similar programs happening here in the UK?
Dr. Lucy Bowes: Good question. It is a bit of a plug. I want them to be and so we are involved in the large trial funded by the NIHR, and there is a group of us led by Judy Hutchings in Bangor University in Wales, but with also number of collaborators including Tamsin Ford in Exeter and Richard Hastings in Warwick and we’re trying to have a pragmatic randomised control trial to see if KiVa might work here in the UK. And that’s really important because as I said, it’s been taken to scale in Finland, but Finland is quite different from the UK context. They have fewer social inequalities, for example relative to the UK. Teachers also have a minimum of Masters level training where we have a bit more of a mix in terms of the backgrounds of our teachers. So there are certain important differences that means, we mean that we need to be confident that this program will work as it is in the UK. There’s been a small pilot trial in Wales but this is a much larger trial to really test the effectiveness and also the cost-effectiveness of this program.
Interviewer: I want to take a step back because we talked about the social factors that may put a young person at risk of developing psychological problems. These are children that have experienced maltreatment and victimisation, but you also talked about the fact that not all children will go on to develop psychological problems and that some children are more resilient. So I’m wondering what are the genetic and environmental influences that are known to be protective and promote the resilience to victimization and early life stress.
Dr. Lucy Bowes: Great question. So this is an area where there’s sort of burgeoning research to try to uncover these different factors, and there are some key variables. So beyond the individual, things like having social support. So particularly close social support that’s characterised by heart high warmth seems to be protected against quite a few different forms of adversity. In fact, that seems to be a very important driver. We’ve also shown that that warmth may come both in terms of maternal wants in having a good relationship with mother. We can’t say about fathers because we didn’t measure this but I’m sure it would be the same. We also note again close sibling relationships can be protective. So there are sort of family level variables that seemed to be important. We also found variables that relate to how much time the parent spends and invests with the child seems to be protective. So for example, engaging and stimulating activities. And these were activities that were designed to be low cost, so we weren’t just looking at families with higher income for example. So these are the sort of the social variables that seem to be protective.
We know that there are individual level protective factors as well, so we know that certain coping styles may be more protective for example than others. Certainly the way we attribute different scenarios may also play a role in whether we are more or less likely to experience negative outcomes. As in like our ability to regulate our own emotions, for example may also be important. So these are individual level targets really for intervention as well.
In terms of the epigenetic factors I think we’re a long way really from knowing what they are. There has been some early work for example looking at the serotonin transporter gene and suggesting perhaps greater methylation might be associated with better outcomes, but I think we really need to see these sort of smaller studies replicated in independent samples. And before we can add weight to them. And I think for me I’m particularly interested in the modifiable environmental factors that may be protective or promotive. And we have to consider that children’s genes may make them more or less vulnerable because I think if you don’t include knowledge of genetic influence, then you’re missing a big part of the puzzle and that’s a shame. But really I will say I’m interested in knowing about genes almost so I can rule out genetic confounding almost so that I can say, okay, well, that’s what you know, that’s what the genes are. But how are they working and what you know, where do we need to work to effect change.
Interviewer: Given what is known about these factors, how can this knowledge be used to help predict which victimised children will go on to develop psychopathology?
Dr. Lucy Bowes: I think we have to be very careful because almost all of this research is correlational and you know that old adage is completely true ‘correlation is not causation’. When we look at prediction modeling really what we see is the biggest predictors are previous depressive symptoms. For example, if a young person is going to go on to develop depression and that’s something we probably knew already. So I think we want to be careful about trying to give someone a level of risk that throws everything in together.
The other reason I’m less confident is if you do throw all these risk factors in together, it may not tell you enough about what you actually need to do to effect change. We need to know what we can do to best support that person. The flipside and people would argue that you know, having a better understanding an individual level of all the different risk factors you have, may lead to personalised treatments, in the future, and that, of course, would be a positive thing. But I think we want to be careful about sort of these things running away with themselves and when people, you know finding out that they have a sixty percent likelihood of developing depression and what that means to a person what that actually even having that knowledge might do to a person so I’m very cautious really I suppose about this field.
Interviewer: You stated that the aim of your research is to guide intervention work by identifying these protective factors that we’ve talked about. How are your findings been translated into clinical practice? And I know you’ve already mentioned a few quite successful interventions, so perhaps you can go into those in more detail.
Dr. Lucy Bowes: Sure. So, one I wish was on my research it’s not but I why reinvent the wheel when there’s something good. So the first is the KiVA program, and that’s an evaluation of an existing program as I mentioned earlier. And for me, that’s the first part. We need to prevent bullying in the first instance in that we can prevent bullying and that’s important. It is tractable. It’s been described by Tamsin Ford as the most tractable risk factor of mental health, and it is we can do something. So KiVa is a preventative whole school intervention to reduce the overall prevalence rates of bullying in primary schools. But I think even with the best interventions there is always some bullying and there are some individuals who experience bullying even in the best schools.
What we are currently lacking is targeted support for those young people who’ve experienced perhaps repeated or chronic bullying victimisation and it’s impacting on their mental health and at the moment, we’re not really geared up. We don’t have any tailored intervention specifically for them and for their experience and that’s what I would like to see more of because I think that’s an important addition to what we and the work we’re currently doing to reduce the bullying and its impact. So that’s where we’re at the very early stages of this virtual reality study. At the moment we’re at the stage of creating scenarios that are very similar to the experience of verbal bullying which is the most common form that’s reported among young people. And at the moment we’re just looking at how young people react to this experience.
Interviewer: Could you just clarify about how it works then?
Dr. Lucy Bowes: So it’s immersive virtual reality which has been used actually in the treatment of lots of different mental health conditions, which is very exciting. It’s mobile technology so we can go into schools. We can collect Baseline assessment data with questionnaires. But then the young person enters the simulated classroom environment, so they wear an Oculus Rift headset and it feels they have a high level of immersion. So they feel like they’re in a virtual classroom and they’re in reality sat behind a desk in the virtual reality environment, they’re also sat behind a desk to really aid this immersive feel.
And then we have different scenarios that relate to verbal bullying so we are looking at adolescent girls at the moment who were aged between 11 and 15 and they’re sat behind the desk and then there’s a group of older looking adolescent girls three avatars in front of them holding the conversation. And in the bullying scenario, there is some pointed head turns and eye gazes, and some whispering, and giggling and we escalate to some mild name-calling. We want to make this as realistic as possible, but also not something that’s going to be traumatising, but is something that is going to be stressful so we can look at how people react in real time and start to better understand what we might want to do in terms of promoting resilience in the moment and also understanding the impact of bullying over the longer term.
Interviewer: And that project is working with secondary school children. I was wondering if the age that the child or young person is victimized or bullied makes a difference if it’s prolonged, if it starts younger?
Dr. Lucy Bowes: So the research certainly shows that chronic bullying particularly across the primary to secondary school transition is associated with worse outcomes. So we’ve looked at that to that earlier on in my studies and we can see greater levels of emotional behavioral difficulties in young people that have experienced chronic victimisation.
There’s also some evidence that being bullied and adolescents may be particularly harmful. I think theoretically for me that makes sense because in adolescence you find that the impact in pure influence is so much greater adolescents are very responsive to their peers views at that age much more so than in younger children. And you also see this interesting paradox where rates of bullying are higher in primary school, but it’s also somewhat more random. So you know, there’s higher rates lots of kids get bullied. When you move towards adolescence and secondary school context the overall prevalence goes down, but what you find is some vulnerable few are more repeatedly targeted and perhaps not just by one or two bullies by more people, so it becomes very targeted. And again, I think in terms of your mental health that probably also is likely to be more damaging over time.
So I’ve been very fascinated in adolescent bullying. It’s also a lot harder to intervene or it had been I should say so the KiVa program for example is very much a primary school and hasn’t yet been shown to be effective in Secondary School, although I’m really excited to see findings by Chris Parnell’s group of the inclusive trial where actually they were able to reduce rates of victimisation in secondary schools in kids in the UK, which is fantastic.
Interviewer: I was just wondering what, how social media feeds into this?
Dr. Lucy Bowes: I think that there’s a lot of concern about cyberbullying as we would call it which isn’t just social media, that’s any kind any bullying that takes place in an online format so that could be social media. It could also be text messages.
There’s a myriad of different ways this can take place and there’s very good reason why people are worried. For example, there’s you know, you could have a much larger audience than in more traditional face-to-face forms of bullying. It may go online. Definitely. It’s actually very hard to remove content if it’s been posted multiple times and different places. Whereas at least in face-to-face bullying is awful as it is. It’s usually time limited. In terms of the control ability which of course is a major factor when we think about how stressful something is, bullying at school though it’s not controllable may happen in specific places that the young person may be able to avoid and then maybe the option of changing school. Not that that’s always a brilliant option nor does it always solve the problems, but it is an option. Whereas when it happens online it’s actually very difficult to remove the young person from that environment and I think the older generation just thinks well go offline. You know don’t check your phone. But actually that’s like telling a young person not to talk and to isolate them more from their friends, you know, don’t don’t speak to anyone that doesn’t work. So it’s a bit tricky.
Then there’s accountability issues. You know, is it the app developer? Is it still school that has the main responsibilities, is it parents who should be accountable for this. So for all of those reasons people have been super concerned about cyberbullying and I think rightly. But I would say that at least the epidemiology in surveys has indicated at least at the moment that the prevalent in terms of prevalence. Actually, it’s face-to-face bullying that children and young people are reporting more commonly and the impact on things like mental wellbeing something we looked at in one of our papers the impact of cyberbullying appears to be somewhat less than face-to-face bullying at least at the moment in the samples that we have.
So it’s not to say that I think we shouldn’t worry. I think we should be concerned, but I think we also need to not forget that it’s face-to-face bullying that we also need to be tackling and that we can tackle and we need to protect the young people. We shouldn’t just assume that because it’s social media and its online that this is going to be terrible and actually, you know, if you do if you ask young people and in our study over a hundred and ten thousand young people told us that actually it was face-to-face bullying they were experiencing more regularly than cyberbullying and I think their voice should matter here.
Interviewer: I’m wondering what barriers are there that prevent victimised children from getting the help that they need?
Dr. Lucy Bowes: I think access is a major difficulty. So for bullying really is a lack of targeted evidence based support. So often what you find as a young person it’s not until lots of things are going on in their lives than they may reach the need for counselling services that they may get help but there’s a bit where people sort of fall through the gap. And it’s not just bullying, we’ve also looked at child maltreatment. So child abuse and in a review that I was part of back in 2016 and we actually found that you know, with the current evidence base it was very difficult to differentiate which interventions might be most successful. We couldn’t achieve whether any might actually cause harm which is deeply concerning and there really was a sort of a lack of evidence which I really think we need to change, you know, we know that bullying and treatment and these kinds of experiences are harmful and we know we can at least stop them and I think we really need to be doing more studies. So, we know exactly how to stop them and how to support young people.
Interviewer: Did I hear you right that you said there are some interventions that have been shown to cause harm?
Dr. Lucy Bowes: So I said, so in that study I said we couldn’t tell whether they were harmful or not. But that is a point that I really wanted to bring up and I think people know that drug trials can go wrong and can accidentally cause harm and people don’t talk enough about the fact that psychosocial interventions can go wrong too. I think anything that can go right also has the potential to harm.
So, you know, there’s a long history of this there was a study back in the 50’s and called the Cambridge Somerville Youth Study for young boys that were at risk of hunger problems. It’s a beautiful study. It had everything that even today we would want to do. It was a lovely design, it had targeted social work support, support for the parents, educational support, it was very holistic. You know it ticked every box. And it also was one of the components had an opportunity for these young people to meet other young people in the program during summer camps. At the end of this program was that it looked like it hadn’t been particularly effective.
However, when the results were analysed many years later what they found was that kids in the intervention group, were more likely to have been in prison to be a substance user or to have died and this is in a randomised control trials, very hard objective terrible outcomes. And I think for me this really highlights not that our cities are bad quite the opposite. We need them. We need to test this properly because we might not know unless we’ve properly evaluate we won’t know, if we have a before and after you know, it may look like something’s getting better, but that might just be what time does you know people tend to regress to the mean people often do get better over time and that might disguise real and actual harms.
So we absolutely need randomised control trials and we need to look very carefully about whether there’s any potential for harm. Now many trials are doing them and they report them which is wonderful, but I think it really is it’s a broader point that just because something doesn’t involve the drug does not mean that it can’t be harmful.
Interviewer: It’s really fascinating. I don’t think it’s something that’s really talked about very much at all.
Dr. Lucy Bowes: You know, it’s a careful message because what we don’t want is backlash so that people think we shouldn’t do a trial because you could cause harm. it really for me it’s quite the opposite and I think I say this as someone who’s particularly interested in school-based interventions and in the education context, it’s been one in which you know, there’s been quite a lack of RCT’s it is changing but traditionally schools didn’t really favour the randomised control trial approach, and there’s also just more broadly, you know, there’s lots of charities working with vulnerable young people who tend to evaluate their projects in terms of a before and after of this particular measure. We need to be open about the fact that we need to be very careful because these are people who’ve already experienced from do not want to do anything that could make the situation worse.
Interviewer: Sticking with research and hopefully looking at the potential for good as well as the potential for harm, what other research projects are you working on that you can share today?
Dr. Lucy Bowes: So I’ve mentioned already there that KiVa project and also our virtual reality project. There’s one other project that I’m really interested in and that’s we have a project funded by the Calleva Trust and it’s looking at the impact of adversity and adverse experiences across the life course and it’s using secondary data analysis. So we using older birth cohorts and we’re particularly interested in trying to identify cohorts in which and the social norms have varied greatly.
So just to give one example, if we think about today’s practice, physical punishment of a child is not just in some cases depending on the punishment and who’s delivering it, is not just illegal it’s also considered really against social norms. It’s considered harmful and our evidence now suggests that it is harmful in contemporary samples as well. So slapping and hitting a child is not considered a good disciplinary method and I absolutely agree with that. But what is fascinating is if you go, back even to the 1950’s, it was quite the opposite, you know that old expression ‘spare a rod spoil the child’, you know, the idea was actually if you were a parent that cared about your children and that wanted to discipline them, you know, you really should discipline your children, even if that involved hitting them.
So we’re interested in you know, this gives us an idea so we can actually test whether this might still be a causal risk factor for poor outcomes. If we have a situation in which it’s not necessarily the hitting that’s bad, but rather the context of which one is hit. So if usually it’s you know, it’s considered so wrong and actually the parents who are more likely to hit their child may be for example more impulsive or more stressed or have more things going on in their lives that tells us one thing. But if we find in these older samples actually that the correlation is still there. Even when the confounding structure is very different, even when you know, it wasn’t so, it wasn’t considered so negatively then that actually tells us that the history itself might be a causal factor for things like conduct problems. So that’s just one study, but we’re interested in using this approach to try to think about things that maybe considered risky now but weren’t then and also vice versa to try to unpick causal inference.
Interviewer: And if listeners want more information on some of the research projects you’ve mentioned is there a particular website they can go to?
Dr. Lucy Bowes: Yeah, so we have an oRANGE Lab website where we put all of our information and that’s orangelaboxford.com and that will post some of the information at least from research coming out in our lab and I’m very happy for people to get in contact because I’m fascinated by other research particularly interventions can cause harm.
Interviewer: Okay brilliant. Lucy what else is in the pipeline that you’d like to mention?
Dr. Lucy Bowes: Pipeline or perhaps I should say a very distant pipeline. One thing that’s on my mind a lot that I’d like to move towards working them is the fact that a lot of our interventions and here I’m particularly thinking about school-based interventions have been designed for kids who are in mainstream schools who don’t have for example any special educational needs or language difficulties. And yet it’s those children can do experience special educational needs actually who are even more likely to experience poor mental health yet, they’re precisely the group where there’s a lack of research into what works. So one thing that’s on my mind is thinking about how we can better support these different groups of young children and young people and tailor interventions according to their needs. It’s a dream at the moment but it’s something that I would really like to have some serious consideration of.
Interviewer: Finally Lucy. What is your takeaway message for those listening to this podcast?
Dr. Lucy Bowes: I guess that’s that, you know in a sense we have a gift here. We have some risk factors that we know are risky that we know are a causal factor. I think we can say safely say we know are a causal factor for poor mental health. Now we need to do something about it. We need to spend money and time evaluate, developing and evaluating interventions so that we can better support young people who’ve experienced adversity. So if people are interested in discussing any of my own research or indeed any of the other research that I’ve mentioned for example, the interventions that harm which i’m particularly fascinated by, please do get in touch and you’re very welcome to email me and my email address is firstname.lastname@example.org
Outro: This podcast was brought to you by the Association for Child and Adolescent Mental Health, ACMH for short.