Professor Andrew Lewis Okay, I’m Andrew Lewis,  and I’m a Clinical Psychologist and Academic   working at Federation University in Victoria,  Australia. I’ll be doing a presentation today   on “The Prevention of Adolescent Depression.” So,  before I do so, I just wanted to acknowledge the   traditional custodians of the lands and waters  where our – my university and myself are located   and recognise the continuity of their cultures  and caring responsibilities for these sites of   teaching and learning, and I also wanted to pay  my respect to their Elders past and present. So, I’m going to break my talk down into four  separate parts. First of all, we’ll talk a little   bit about what depression is, as it presents  itself in adolescents, and we’ll talk about   some of the epidemiological studies that tell us  more information about the nature of depression   in this period, how it presents itself and how  prevalent it is in different countries. And then,   I’ll move onto what we know about the causes  of depression in adolescents. It’s important   to have that, sort of, background before we move  on to talking about interventions and prevention,   because such interventions do need to be  targeted at what’s causing the problem.   And then, in the final section, I’ll talk a  little bit about the work that I’ve done around   prevention and some of the studies, intervention,  the models that we’ve developed and evaluated,   and most of that work has been working  with families an – who have adolescents. So, let me start off, then, with some discussion  of the epidemiology of adolescent depression.   One of the interesting things about depression  is that the prevalence in young children is   relatively low. So, some studies, for instance,  have found that prevalence estimates are around 2%   in children, but move up quite quickly  once you move into the adolescent period,   doubling, or in many cases tripling, in their  prevalence. So too, the male to female ratio,   which is one-to-one during childhood,  starts to flip around during adolescence and   we find a much greater prevalence of females  developing depression as compared to males. So,   by the time of 18-years-of-age, the cumulative  incidence is somewhere around 20% and that’s a   figure which is a little bit dated nowadays, and  you’ll see me going on to presenting some more   current data, which suggests that that cumulative  incidence rate is now considerably higher. Now, the lifetime prevalence of major depression  in females has a higher prevalence rate. It   actually has its origins in this adolescent  period, where the incidence for females tends to,   sort of, be double or triple the  incidence for males. And so, females   unfortunately, carry this lifetime higher  degree of susceptibility to depression. Now, I’ve been, sort of, speaking loosely, using  the term ‘depression’, but I think it’s important   to try and be more specific. Sometimes we can  make a useful distinction between depressive   symptoms and psychiatric disorders that  have depression as a component. So,   this figure on this slide is really depicting the  fact that depressive symptoms are a feature of   many different psychiatric disorders, major  depressive disorder being one of those. But   someone can present with depressive symptoms,  but the underlying problem might be something   more like, say, bipolar disorder, dysthymic  disorder, schizoaffective disorder or seasonal   affective disorder. All of those disorders have  depressive features as a component of them. So, what are the specific features, then,  of depressive disorders? So, here you’ll see   an instrument that’s used quite widely in  population level studies. This is called a   Short Moods and Feelings Questionnaire, and it  gives you a rough idea of the kinds of things   that people are rating when they’re indicating  that they are experiencing a high level of,   in this case, symptomatology. So, this  is not a diagnostic measure by any way,   shape or form. It’s really just a way of  a person self-reporting these kinds of   symptoms, and as you can see if you read  the sections in red here on this slide,   that a depressive disorder, it, kind of – it  more or less clusters into certain features. So, there’s one feature that could be called  anhedonia and that is, you know, the difficulty of   experiencing pleasure. There’s a number of things  there that you can see, “I didn’t enjoy anything   anymore.” “I felt miserable,” and so on. So,  that, sort of, lack of enjoyment of things or   losing interest in things. The other component  is this self-critical aspect of depression,   so someone has a very negative view of  themselves, of their place in the world,   thinking of themselves as unlovable and no good or  a bad person. And this crosses over with a general   negative set of cognitions and a very  negative view of their experience. So, the other feature that’s very important to  note around depression is that it’s a combination   of both, sort of, psychological and mental  phenomena and somatic, physical components. So,   the physical component is really important  to consider and for Clinicians to assess,   obviously, so restlessness, feeling tired,  and changes in sleep and appetite. So,   there’s obviously a biological process,  there’s a psychological process. The final item is, you know, incredibly important  from a clinical point of view, that suicidality,   thoughts of suicide, are a common feature  of someone experiencing depression. So,   that should be assessed and interventions  undertaken to address that, obviously. Now, as I mentioned previously, there is evidence  that there is increasing prevalence in general   of mental health problems in adolescents. And  just as I was preparing this talk, there was a   new paper released from a group that works here  in Melbourne, at an organisation called Orygen,   and they have released a paper in The Lancet,  as a commissioned study on youth mental health.   And this is a – some of the data that they  presented there of a range of studies, which   if you look at the dates here along the X axis,  you can see that over a period of – it varies,   you know, but generally between about, you  know, ten to 15 years ago, you can see that   there’s really quite marked increases across  a wide range of studies in the prevalence of   mental health problems for adolescents. So,  the timeframe here is really important to bear   in mind. This is a relatively short period of  time, and you have this quite sharp increase in   the prevalence of mental health problems  for both young men and for young women. Okay, so let’s move on, then, to the  second section of my presentation,   which is really to try to drill into this a  little bit more and understand what are some   of the causes of adolescent depression? What  do we know about it at this point in time?   And the first place to start is to really  examine the epidemiological findings in a   little bit more detail and scrutinise, you  know, what these might actually mean. So,   say, for example, if we just take that  last slide where you can see that there   is an increase in prevalence over the last ten  or 15 years in this Lancet Commission report,   I mean, this obviously suggests that we’re  talking about something that has significant   social and cultural factors at play. Human biology  doesn’t change that much in that period of time. So, we know that this very incre – very  marked increase that’s occurred suddenly,   it must be to do with, you know, broadly what  we could call sociocultural or socioeconomic   factors. But we also know that the increased –  the marked increase in incidence of depressive   disorders over adolescence is very closely related  to pubertal development. So, that’s this, sort of,   underlying biological change that occurs in  individuals. We know that as puberty progresses,   the further someone moves through puberty,   the more – the higher the rate of depressive  symptomatology that tends to be reported. There are significant gender differences, which I  mentioned previously, but they are not universal.   So, generally, it’s the case that this higher  incidence for females is typically in Western   countries. So, we also have adolescent-onset  of depression being a major risk factor for   subsequent episodes of depression. So, this  is – this refers to the fact that in its most   typical presentation, depression is an episodic  phenomenon. So it has a certain course and then,   remits and – however, you tend to find that  people who experience a depressive episode   in adolescence are more susceptible to a  repeated episode later on in their life. Okay, now, more than half of the first episodes  of depression occur in adolescence. So,   you can see from that further analysis of  some of these epidemiological findings,   it points very strongly to the need to  both understand the onset of depression   in adolescents, first of all, and  then to think very seriously about   interventions that are effective in adolescents  and interventions which are going to prevent or   reduce the first onset. And so, that’s really  the rationale in my, sort of, research focus   for focusing my work on adolescents. It  makes a lot of sense, I think, to do so. We know that the world of adolescents is changing  very rapidly, and one only needs to bring to mind   some of these major global issues, such as climate  change and globalisation, economic inequality, the   rapid rise of largely unregulated social media and  political instability, not to mention, of course,   the global pandemic in itself. So, these are  things that our – this generation of adolescents   are facing. A very complex, unpredictable  world. A world in which they are often, sort of,   economically disadvantaged by some of the  systems that we have in place, in terms of   home ownership and access to high quality jobs,  access to further education and training. But   there is also a very major impact of those sorts  of social and economic conditions on families,   and I will continue to make the case that the  family operates as a, kind of, a secure base,   not just for a child, but also for an adolescent.  So, when you have impacts on family functioning,   by implication, you tend to get spill-on  effects for children and adolescents. In some of the research that we’ve done,  we’ve really looked at that in close   detail. This is a paper that you’ll see at  the bottom of this slide around the use of   social media by depressed adolescents. And that  was a very interesting qualitative analysis of   parents talking about their depressed  adolescent. And, you know, you’d read   that comment from participant, “The difference  between what she,” her adolescent daughter,   “would look at on the internet as a healthy  and happy adolescent,” as compared to “a   healthy and happy adolescent would look  at is very different.” So, you know,   this parent is then expressing huge concerns  not only about her adolescent’s mental health,   but also about what is she looking at on the  internet and how are those two things interacting? So, to think, then, about, you know, what some  of the scientific literature tells us about   associations with adolescent depression, or if you  like, risk factors, as they’re often referred to.   So this was a report done a little while ago now,  but essentially, a very detailed review of the   risk factors for adolescent depression. Now, this  review identified 116 different risk factors. So,   you could hardly say that there’s a dearth of  research in this area. There’s obviously a great   deal of research and the point is not to just  accumulate risk factors. The point is to try to   pull those risk factors together into some  sort of meaningful themes and common elements. And so, from that work on the various risk  factors, this is a summary that I’ve produced   of both risk and protective factors. And you can  see that you can break it down into biological,   psychosocial, comorbidities, such as sleep  problems or other mental health issues. But   the two that have been of particular interest  to me and the interventions that I’ve tried   to develop and trial, is – are these last two.  There seems to be pretty consistent evidence that   there’s a relationship between stress  exposures and family social environment. This was a study that we did on quite a large  dataset to, sort of, tease out a particular   aspect of that. You can have a read of that, the  reference is down the bottom, there. But it shows   that for all adolescents, there’s a relationship  between the quality of parent-child relationships   and their vulnerability to depression,  but that’s exacerbated when they face   stressful circumstances. And it’s particularly  acute for females who are residing in families   that are reporting low closeness. They have a much  higher susceptibility to depressive symptoms. So,   starting to move away from just that  simple bivariate risk factor, into well,   how do all of these variables work together  to come up with a, sort of, model that   might be a little bit more causal in trying to  understand risk and vulnerability to depression? We also need to consider the fact that it’s not  simply just adolescence as the developmental   stage that we should be thinking about, and  particu – this is particularly pertinent to   taking a prevention perspective. Prevention  needs to intervene before someone starts   to develop the disorder that’s trying to be  prevented. And so, it’s important for us to   consider, well, what are the earlier adversities  in a child’s life that might make them vulnerable,   such that when they enter into adolescence and  perhaps experience some of those stresses that   are part and parcel of adolescence, what  makes some more vulnerable than others? And this is a slightly, sort of, complicated  slide, I apologise for that, but this is   our attempt to think some of that through. So,  some of these early vulnerabilities could entail   genetic factors, could entail prenatal development  or things experienced in very early infancy. And   some of the vulnerability factors are very likely  to be across these three domains. An affective   disposition, you know, there’s clearly a lot of  variance in young children in their ability to   experience pleasure, or conversely, not experience  it. There’s also different cognitive styles,   which seem to set in fairly early. So this, sort  of, negativity or negative bias in the experience   that someone has, or their interpretation  of their own experience, more to the point. And then, there are clearly temperamental  differences that have been researched for a long,   long time, and we know that the young  children that present with this reactive,   or sometimes it’s called neuroticism, kind of  temperament, have this degree of vulnerability.   And so, for those individuals, moving into  adolescence and going through the pubertal   development, it’s likely that perhaps only  a fairly low or negligible stressor might be   the catalyst for the development of depressive  symptoms. Those who don’t have that, sort of,   earlier vulnerability and have a low vulnerability  probably are also vulnerable to depression. But it   would take a much higher stressor for that  to occur and perhaps the depression might   be of a more straightforward kind of a kind  and more treatable in that particular way. This is a more complicated model and draws on  some work that a Researcher called Garber has   come up with, called the “Stress Reciprocity  Model.” Given that we know that vulnerability   to stressful experiences is important, we have  to distinguish between, on the one hand, stress   exposure, and on the other, the person’s reading  of a given situation as itself stressful, right?   And so, that might draw on, say, personality  or temperament factors, what one person finds   stressful, another person may not, and it’s the  interaction of how much stress you’re exposed   to and then your interpretation of that stress,  which is actually the subjective experience of   the stressor. And you put those kind of factors  against the backdrop of pubertal development and   you have changes in endocrine systems that makes  the stress response much more sensitised. And you   also have sexual maturation, which in part,  is a biological process, which culminates in,   you know, much stronger motivations to have close  relationships, intimate relationships and so on. And so, you have a young person, sort of, seeking  those kinds of close connections and interested in   others in a way that perhaps they haven’t  been before, but you also have them highly   sensitised to the stressors that go with those  kinds of interactions. And so, down the bottom   here you have some of the vulnerabilities that  go – that are inherent in the combination of   those factors, such as, you know, personal and  interpersonal challenges. Adolescents are doing   things interpersonally that they’ve never done  before. They’re also going through psychosocial   transitions where they’re individuating from  their parents and facing challenges that they had   never – they’d never faced previously. And they  have an increased vulnerability to interpersonal   loss, disappointments, negative self-appraisal,  and feelings of hopelessness and isolation. So, you know, that gives you a rough idea  of, kind of, a – some of the causal factors   and type – the modelling and interaction of  various components across development and then,   as a young person moves into adolescence. You  can see that there’s a strong emphasis, you know,   perhaps the emphasis that I’ve given is really  around the family environment as a potential,   sort of, both risk factor and/or protective factor  and also, exposure to stressors. So, that’s the   kind of framework that I’ve used in some of my  work on adolescent depression and its prevention. So, let’s have a bit of a think, then, about,  you know, what can we do? Very broadly speaking,   you know, we have these two sources of research.  On the one hand, we have longitudinal and   population level studies, both those that are  existing and also, the opportunity to develop new   longitudinal studies, you know. And the kind of  studies that I think will be very informative in   this space would be those that begin in pregnancy  and follow a cohort through into adolescence. And   so, you can really get that developmental pathway  really well mapped in these kinds of studies. That needs to be combined, on the other hand,  with studies that are intervention studies. So   developing new interventions, trialling  new interventions and then, conducting   meta-analysis of existing trials to, sort of,  see what seems to be working. And the combination   of those two sources of information is really  likely to drive the science forward in this area. Coming back to the interventions, so, well,  how do we begin to think about what we might   be able to do here? This is something  that I’ve used a lot in my teaching,   sort of, a spectrum of interventions,  to think about the fact that, you know,   we have options here to design interventions that  are focused on either prevention or treatment,   but we also should be considering interventions  designed to promote mental health. Each of   those kind – different sorts of interventions  is going to be targeted in a different way,   at different populations and so on. So, that’s  another thing to think about in terms of designing   prevention interventions, is to think about  the optimal setting for the intervention. Interventions can be designed to target a whole  population. So they may be changes in policy or,   you know, changes that are going to affect a  whole state or a whole nation or a whole county,   or what have you. Then you’ve got  particular communities where, you know,   working with the given community about what  some of the particular risk factors that are   going on in that particular community  can develop a much more targeted kind   of intervention to address what’s going  on there in that particular location. It’s quite a large literature and many attempts  to develop prevention interventions in schools.   This has had mixed findings. I think there have  been some interventions that have successfully   prevented mental health problems and depression  in schools, but there have been many that haven’t.   So, this is an area where, you know, we can learn  a lot from as much the successes as the failures. As I mentioned earlier, the kind of work  that I’ve done is focused more on the family,   as working with individual families, or  small groups of families where there are   vulnerable adolescents. So, that’s  more of a targeted intervention,   where working with that family is likely to have a  preventative, sort of, effect on the adolescent’s   susceptibility to depression. And nowadays, there  are quite a raft of different interventions that   are focused on the internet and social media as a  point of intervention, a lot of promise there. So,   we also have to then think about the level of  intensity of intervention and that tends to,   sort of, intersect with these different settings.  So, there can be low intensive interventions that   are developed – delivered in a population or  a community, but once you start, say, working   with an individual family, obviously, that’s  a much more intensive level of intervention. Okay, so when we start to think about the  level of the intervention and whether or not   we’re – we should be targeting a population,  a community or a school setting, for example,   it’s really important to think about  the fact that, you know, when we look at   how depression develops over time, you know,  this, kind of, slide here is presenting some   data analysis that we did on a longitudinal  dataset, and it’s using a technique called a   ‘latent class analysis’. It’s able to divide  up a large dataset into meaningful groups,   and you could see that the symptomatology  over on the Y axis, for a pretty large   percentage of kids going through these age  groups, from four and a half through to ten,   is just simply very low and is flat. So, 80% of  the children have no symptom and they have no –   they’re not vulnerable to an increase over time.  Whereas it’s really this blue line here, which is   a group of around 8% of these children, who show  this tendency to increasing symptoms over time. So, you can see that, obviously, from a prevention  point of view, that is the group that you would   want to target. And it’s very important to  note that from a population point of view,   so going into a school, not all of the young  people are going to be vulnerable. And so,   it’s very important when designing  these kind of interventions,   that for the people who are not vulnerable,  we’re not doing any harm to them. We’re not,   sort of, stirring things up, or we’re not alerting  them to symptoms that they don’t actually have,   and they don’t have any vulnerability to having  them, right? So, this is the tricky part,   I guess, of designing interventions, is that if  they’re at a population level or a school level,   they have to be of some benefit to everyone,  whether or not they carry a vulnerability or not. Now, I also mentioned that a way of working in  this area is to conduct reviews and meta-analyses   of existing studies and fortunately, we have some  great Researchers who have already done that,   kind of, very intensive work and then, this  publication here of a Cochrane review by   Sally Merry and her team, has been really  setting the standard of rigour around the   review of existing prevention and interventions.  The team that I work with have done some more   fine-grained analysis of the interventions  that were used in the – this Cochrane review. And you can see that the interventions, which  have generally been successful in preventing   depression, have certain characteristics  about them and these characteristics are   that they make efforts to include families,  that they have professional facilitators.   So that’s distinct from, say, going into a school  and asking the Teachers to do the intervention,   or asking community members to do an intervention.  The interventions that have been – were found to   be successful, generally have been run  by professional facilitators. The age at   which the interventions have targeted has  been pre-adolescence, this is prevention,   of course, so the eight to 12 year  range is really the optimal range.   And the final characteristic of the most  successful interventions that have been   reviewed in this Cochrane review are that they  have an interpersonal focus. So, they teach   children and young people skills about how to  have effective relationships with their peers,   with their parents and that sort of thing.  So, they have an interpersonal focus. Alright, so, in this final section  four, and I want to move onto some   of the specific work that I’ve done around  prevention, working with families. You know,   I’ve been fortunate to work with a  number of colleagues in developing and   designing an intervention called “Behavioural  Exchange and Systems Therapy.” And we’ve done   a number of clinical trials of this intervention  and written up descriptions of the programme logic   and the underlying theory and the particular  techniques that are used in this approach. So,   this is a family group intervention. You might  work with about four or five families who come   along and have an eight-week programme, usually  two hours per week, so that’s 16 hours of therapy   in total. And there’s a focus on working  with both the parents and the young person,   and the young person’s siblings, so it eventually  becomes a full family intervention model. It draws on, obviously, family systems theory,  it draws on some aspects of attachment theory,   and it, over time, has developed a number of,  sort of, specific techniques. There is a training,   it’s – so, it’s a two-day training for people  who wish to undertake and use that model. And   I’ll just go through, kind of, quickly, some of  the techniques that we do use in that approach   and speak to parents about the fact that,   you know, parenting an adolescent requires a,  sort of, shift in their developmental stage   themselves. So, parenting is conceptualised  as, itself, a developmental process. We know that young people, from infancy to  childhood, to adolescence, they themselves,   are going through a developmental process, but  parents, equally, need to change their parenting   a lot – across those different stages of  development. And so, on some occasions,   we find that parents are cha – struggling  with their parenting role of an adolescent   because they haven’t, sort of,  adjusted their parenting style   to some of the things that adolescents are  wanting to do, and that may be parenting   their young person more in a more childlike kind  of a way. So, we have a exercise where we present   the developmental lifecycle of families  and seek some discussion around that. Another part of the intervention, which draws  on what I was talking about earlier around   modulating stress reactivity, is – are a  number of interventions that are designed to   help parents to learn self-care techniques  and to modulate their own stress reaction,   and to take positive action for themselves  to deal with their own stresses in their   lives. Which has a modelling effect for their  young person to see that the parent is able to   regulate their own emotional states. And so we  sometimes use the metaphor of the ‘oxygen mask’,   that you’re all probably familiar with from flying  around in planes, that the host and the hostess   will tell you to put on your own oxygen mask  before seeking to help the young person. And so,   that’s a very, very important metaphor for a  lot of parents, who find themselves feeling   quite helpless and hopeless in the context  of their adolescent’s mental health issues. We have an emphasis drawing on more of the, sort  of, systems theory ideas of getting parents to   try and do things differently and to interact  with their young person to change communication   patterns. And we call this the ‘red buttons’  task, which is, sort of, designed to get parents   to speak with their young person about some of the  red buttons, or the trigger points, that occur in   the family’s communications and how they can, kind  of, unpack those and rethink how they actually   want to communicate and get along as a family. So,  that’s just, very, very briefly, a bit of a taster   of some of the techniques that are used in that  family-focused intervention that we’ve trialled,   and there are a number of papers there that you  could look up if you want further information. So, I’ll just say, by way of conclusion, then,  that I think you’ve had a bit of a taster of   the fact that depression in adolescents  is complex and that there are multiple   contributions and causes that we  could think about in terms of why   depression occurs and why adolescence is  a particularly vulnerable time for it.   Fortunately, there is evidence from pretty robust  reviews and meta-analyses, that prevention efforts   to intervene for adolescent depression can be  effective, and that is very important to know   that. It’s also important to go through those  studies and work out, you know, what are the   key ingredients, what are the elements that have  proven to be effective in those interventions? Interventions need to be very carefully designed,   thinking carefully about who the target  of the intervention is and what the level   of intensity of the intervention is going  to be. And you can see that I’ve presented   the fact that there are interventions that go  from the whole population, a community or a   family-based that – so, it’s important to think  through what’s required in given circumstances. Finally, prevention is a really critical part,  I think, of the health system and a part of the   mental health system which has been relatively  neglected, given the urgent needs of direct   clinical care. And that’s understandable, but it’s  – now that the evidence is accumulating, I think   there’s a more robust case to focus on prevention.  And the evidence supports that quite strongly,   suggesting that prevention can be effective and  it can also be a very efficient use of resources. Okay, so that’s really it from me.  Thank you very much for your attention.