[MUSIC PLAYING] We are the Association for Child and Adolescent Mental Health, or ACAMH for short. And this is ACAMH Learn. Hello and welcome to the Mind the Kids podcast series. I'm Mark Tebbs. I'm your host for today. I spent my whole career working in mental health. I'm currently chief exec of a charity working with other charities, and my interest is really around grassroots organisations in how we improve community wellbeing. For this podcast, probably the most important qualification is also being a parent, so really understanding some of the issues that we're going to be talking about today. So today we're looking at early childhood risk, like poverty, family stress, or developmental delays, and how children do later in school. So how positive experiences in kindergarten, like supportive teachers and social play, can help protect children and young people as they grow older. I'll be speaking to Katharina Haag, who's recently published her JCPP paper entitled Navigating Early Risk, Differential Outcomes in Middle Childhood and the Compensatory Role of Kindergarten Experiences. Katharina, thank you for joining me. Lovely to be speaking with you today. Thank you very much for having me. Great stuff. So let's start with some introductions. So it'd be great to hear a little bit about yourself, where you work, maybe some of your research interests. And if there's been any key collaborators on the paper, then it's an opportunity to give them a little bit of a name check. Yeah. So my name is Katharina Haag. I am currently a senior researcher at the Norwegian Institute for Violence and Traumatic Stress Studies, but this work was done as part of a postdoctoral position I did at the Norwegian Institute for Public Health until last year. And the work was done in collaboration with a group that is called the Mobile Early Education project, which is a number of researchers from different countries, mostly from Norway, but one key collaborator is, for instance, also Tyler Watts, who is at Columbia University who was involved with this paper, as well as several Norwegian researchers. Great stuff. Part of my role is I work in a charity in [INAUDIBLE], London and we work with a lot of other charities who are working with vulnerable children, children that are high risk. And I guess childcare provision is a really hot topic in the UK. So really interested to your studies Norwegian. So I'm really interested to hear and compare with some of the stuff that's happening over the UK. But I guess what we hear a lot is childcare is really expensive. There's long waiting lists, the quality of childcare can be really variable. And I guess a lot of parents also feel quite guilty about leaving their children and the economic necessities of going back to work. So I'm really interested in this conversation, this topic because I think it'd be great to hear a little bit more about what your research has found. So let's kick off with just a little bit of an overview. So could you tell us what you set out to explore, why the topic matters, and what you discovered along the way. Yeah. So with this study, we mostly had two main aims. Firstly, we wanted to look at profiles of early risk exposure and see how they link to long-term outcomes. And that is based on the thought that we know quite well that the more adversity children are exposed to, the worse their outcomes on the long term. But what we don't know that much about is actually if the type of early adversity matters for what type of difficulties children get later on in life. And there is two strands of literature that seem to suggest this. And there's the more theoretical literature which is covered by dimensional models of early adversity. And those models, in their original form, they've since been expanded upon, suggest two pathways from early adversity to long-term outcomes. There's first deprivation experiences, like, for instance, neglect or also lack of cognitive stimulation in low resourced households, and they propose that can lead to poorer cognitive development and then long-term outcomes like poorer academic outcomes. On the other hand, they said that threat experiences like abuse or violence in the most extreme form are linked to poor development of emotion processing and regulation and that, that in the long term predisposes children to have more mental health problems. They have these two main pathways. But there's also a second strand of research that has developed recently more independently, which uses a particular type of statistical models, so-called mixture models. And I think the most well-known one is latent class analysis, which we're also using here, and they assume that risks tend to co-occur. So they try to derive these profiles of early risk exposures for children based on many different risk factors and see how they link to long term outcomes. And they are somewhat similar to the dimensionality models, find a so-called resource risk group, so a group that has parents with low resources like low education, low income that mostly struggles academically, but they also find what we call a family psychological risk group. So that is a group where the parents have high conflict. The parents might suffer from mental health problems, substance use difficulties, parenting problems, or the like, and these children tend to have more long-term mental health problems. But a lot of that research is based on what we call convenience samples, so samples that have been collected for a different purpose. Today, I have only very specific indicators of early risk, like limited numbers, and often also smaller samples. And we thought we would be in quite a good position because we have this large longitudinal cohort study, the Norwegian mother, father, and child study, that we could really look at a broad range of different indicators and derive profiles across all of them and see how they link to long term outcomes. And that is important because if we find that different early risks are linked to different long-term outcomes, then we might be able to actually look more in detail into the pathways towards these risk outcomes and might be able to provide more targeted support for children with different types of early risks. And the second aim is what you spoke about earlier, is looking more into the effects of kindergarten, so a bit more on the intervention side. And maybe to give some background about this, in Norway, almost all children attend kindergarten. So in 2024, I think it was 94% of children. Its declared aim by the government that as much as possible, participants should attend. And there's quite strict guidelines in place. So there's some guidelines, for instance, on the ratio you should have between staff and children, on the type of education that staff should have, but also on the contents. So there are regulations, what should be covered in kindergarten. And the idea is really that there should be good quality kindergarten available for all children across the country. And that is because-- I guess the broad idea is that kindergarten could really level the playing field for children from different backgrounds with different risks before they come into school, and set them out on a better pathway. So our aim was that we wanted to see if that is really possible, because there is some indication in the literature that kindergarten can support children from different risk backgrounds, even though there is some discussion about, is it possible for all children or only for some children, and to what extent can they improve? And it has also been proposed that specific experience in kindergarten might be able to buffer against particular risks. So for instance, one of the factors that we looked at is social play. It has been proposed that social play, by teaching children how to relate to others, how to regulate their emotions and their needs and situations with others, might actually improve their emotional processing and could, on the long term, maybe lead to better mental health and other social emotional outcomes. So there's this potentially more specific pathways. So we tried to understand if kindergarten experiences might be able to disrupt this association between early risk and later outcomes, and also if specific experiences matter for children with different risk profiles. And that is, of course, important because if kindergarten is able to buffer against these early risks, especially in countries like Norway where almost everyone attend, it's a really good opportunity to help different children catch up. Yeah, so that was a slightly long introduction. No, it was a great overview, great overview. OK, I think we're probably trying to do the podcast in those two halves then. So look at the early risk bit first and then we'll look at the protective factors of kindergarten. So if I've understood it, the early research was really focused on identifying the impacts, for example, trauma, on later development. But the later models are more mixed categorization of risk. So how do you go about that? How do you go about identifying that more broader risk profile? Well, in our case, we have this really big cohort study for whom data has been collected across many years, and especially lots of data in the early childhood phase, which is what we focused on for the risk factors, because that is really a period where lots of changes happen for the child, lots of brain development happens. So it's a really good period to look at in terms of how risk affect long-term outcomes. And how we went about it is that we looked at general models of child development, trying to see what is common factors that are highlighted as potential risk for poor long-term outcomes, and we went through our measures and tried to see what of these factors do we actually have available. And we discussed it in our group of authors and the wider group of people that we work with to come down to a list of 30 factors that we looked on in the end. OK, and then what were the groups that you identified, because it'd be really good to bring that to life for people so that they can-- maybe some examples of what those different categories kind of mean. Yeah, so specifically [INAUDIBLE]. Yeah, so the biggest group that we found, which is, I guess, quite positive, is a group of low risk children. So that was about 41% of our sample. And across all the different risk indicators that we looked at, these children score quite low. So they have relatively little risk exposures early on in childhood. And the main takeaway is that this group performed the best across all time points, across all measures of outcomes. So they did really substantially well, often more well than the other groups. The second group that we found is what in the literature is often termed as a resource risk group. And that was about 30% of a sample. That is mostly a group where parents have low income, low education, and also some slightly elevated genetic risk for ADHD and academic difficulties. And this group, at eight years, had some elevated risks of teacher suggesting that they might need to work harder in school. But then at 11 years, you could really, on the national test scores, see substantial differences compared to other groups, especially the low risk group, in terms of how they support. So their average points was at 49, and a test with the mean of 50 and a range up to 100, and the low risk group comparatively scored 56. So there's quite a bit of a point difference there, both in maths and in Norwegian. The other three groups were somewhat smaller, and the first of which is a preterm birth group, so a group of children who are preterm born and who often have low birth rates as well and associated family risks. And this group didn't have the poorest outcomes across the risk groups anywhere, but they sort of consistently lacked a bit behind the low risk group in all outcomes as well. We did also find a family psychological risk group like in the previous literature. So that was a group where there was high conflict between the parents themselves, relational conflict, but also conflicts on how to actually rear the child. There was lots of parental mental health problems, especially maternal depression early on, and mothers indicated low self-efficacy when it came to parenting. And this group showed elevated, both internalising and externalising, symptoms at eight years, but especially at 12 to 14 years, where we had registry data on how often children had been to specialist services to get mental health supports, they were almost double as likely as the other groups to have been to specialist mental health support for internalising problems. So they really had persistent mental health problems. And finally, we had a group of children that we called developmental risk group. So that was a group of children with early motor problems, language problems, and social problems. And they interestingly showed quite strong difficulties in eight years. So they were the group that the teachers were most concerned about. And they also were amongst the highest of internalising and externalising symptoms. But by the age of 11 to 14, so later on, these differences had mellowed out a little bit. So they were not ahead of other risk groups. So sometimes even better than these groups. So they managed to catch up a little bit. So I just want to be really super clear about the kind of methodology. So you assessed or categorise risk at about three years old and then you studied the kindergarten experience at 5 and then measured outcomes at 8 and 11. Is that right? Yeah, 8 and 11 to 14 years, because the registry data were available at slightly different times. OK, brilliant. So was there anything there that was particularly surprising to you or did it feel very consistent with the existing research literature? So the findings on the children with a resource risk, doing poor academically and the psychological risk, doing poor in terms of mental health, that is also found in the other literature. So that was quite consistent. What surprised me maybe a little bit is how persistent the difficulties actually are, because we can see these risk factors are measured between birth and three years of age. And at 11 to 14 years, we still see quite substantial group differences. So that is quite a long time frame that we see differences, which I maybe would have expected would have mellowed out a little bit more over time. What was interesting also is that in the literature that's mostly from the US, there's often a high risk group that really scores poorly across all areas. And we didn't find that here. And we were discussing it a little bit. Why did we not find this? And partially we think the US is often more high-risk samples. So this is children, for instance, who attend the Head Start programme, which is meant for children from low income families, whereas we have a broader community sample. So our group might have been just comparatively smaller compared to the US. And finally, I guess we didn't really have clear expectations what additional groups we would find. But it was interesting to see those preterm born children come up, which I think are quite a lot, discussed in the literature as risk groups, and the developmental risk group, which hadn't really been discussed that much before, but I think makes an interesting addition to the literature that we should maybe consider more. Yeah. OK. And then turning to the second part of your paper and about the experience of high quality kindergarten, so could you describe what impact that had on your findings. Yeah. So what we found in brief, I guess, is that we looked at three factors in kindergarten that we were interested in. We looked at if the children had close relationships with their teachers. So that's relationships characterised by warm, supportive relationships where the child feels quite open to express emotions or to tell the teacher about what's going on in their life and to get support from them. And we looked at social play in kindergarten. So the extent to which children engage with other children, and are they confident to approach them, to play with them. And finally, exposure to pre-academic activities. So that is teacher providing activities that make children familiar with early letters, early numbers, some geometric shapes, some scribbling so that they're a little bit prepared for schools. And that's usually in small groups. And what we found is basically that having close relationships with the teacher and social play, but not the academic activities, were linked to better outcomes across the board. So it didn't matter what risk group the children were in, and also didn't matter what kind of outcome we were looking at. So they seem to have quite a broad effect. OK, so the impact of close social relationships had a universal benefit to all children irrespective of the risk that they were experiencing? Yeah, OK. So be really useful just to really understand what the characteristics of that kind of kindergarten experience were. So in terms of the structural quality, so the quality of the experience of the children, the two factors that we found most effective is having these close student-teacher relationships and having high extents of social play. With the close student-teacher relationships, it has been hypothesised in the literature that having these discrete basis can be effective in many ways for children, because having this warm, supportive relationship allows children to feel safe and exploration and learning. So they have this attachment figure that they can work from. It provides teachers with an opportunity to model and to maybe use scaffolding techniques to help children improve their learning, and it's also an opportunity for feedback for children who struggle in various areas. So the idea is really that these relationships might be able to help both the emotional development of the children, but also the cognitive development, and that's how they can buffer against negative long-term outcomes. And with social play, it teaches children a lot about different aspects of relating to other people. So it teaches to better understand the other people's experience, to learn from them, and also how to regulate oneself in a social situation. And that, based on the literature, we particularly expected would help children in their social and emotional development so that it could buffer against long-term mental health difficulties. I'm just wondering whether the experience in Norway is replicable or applicable in other contexts, whether the benefits that were found were-- this might be a difficult question to answer, I don't know, but I wondered whether the impact that you found was felt particular to the circumstances in Norway and the kind of high quality childcare there. I mean, it's a good question. I would think the general protective effects of having good relationships and having good social play, I think especially for close student-teacher relationship, that has been shown across the literature, that is it's really useful. It's been shown in the US, it's been shown in different countries. So I think generally these relationships would have the same outcome. But I guess the difference is really in how kindergarten is implemented across different countries. In Norway, it is highly subsidised. It's very regulated and it's really built with the intent of helping different children catch up. So the question is maybe more, in different countries, do other children get the same opportunity to have these experiences? And at least from what I know about the UK, which is a bit limited, but I guess it's much more selective who goes into kindergarten. It's not as in a Norway that everyone attends. So I think, yeah, there might just be less opportunities for children to have these experiences. Yeah, yeah. No, that makes sense. So I'm just interested in the data set that you use. How did you conduct that kind of longitudinal study? Well, the study itself has been around for many years. I think it was started in 1998. And how it was set up is a longitudinal cohort study. So all women who are pregnant in Norway between 1998 and 2009, over a 10-year period, were invited to participate in this study. And around 40% accepted to participate. So now there is a sample of approximately 114,000 children. 92,000 mothers, and 75,000 fathers. So it's one of the biggest data sets like this in the world. And it's a really amazing data set with so many opportunities. It's kind of a researcher's dream, I would say, to look at. And basically these families have been followed all throughout the child's life. So there's three questionnaires, for instance, in pregnancy that look at all different kinds of factors like maternal living situation, her diet, her mental health, how she's doing in pregnancy, et cetera. And then after birth, the children have been followed in quite regular intervals with questionnaires asking about how is the child developing, what are their childcare arrangements, but also how is them other mother doing. There's some questionnaires to the father as well. So there's a really rich data over a long time. And these children are now 16 to 26 years old. So there's a really long term follow-up over many years. And what's also special with this study in particular is that there's genetic information available, which we also use in our study. So blood samples were collected from mothers, fathers, and children. It has been quite extensively genotyped and there's many groups working with that. And the data that we used in our study is that for the last part of the cohort, for the last three years, questionnaires were sent to all the kindergarten teachers, asking them about the kindergarten situation, the experiences of the children, how the children were doing. So this is also quite detailed data. And for many of them, also questionnaires were sent to their school teachers later on. So we have this continuous data on how they're doing in the academic environment. So it's a really rich data set. Yeah, amazing. That's very comprehensive. So let's turn a little bit to broader implications I guess. So I'm just curious around surround what the findings suggest from a policy perspective. So I guess I'm thinking about policy makers who are designing early childhood education systems or kindergarten experience. What are the findings suggest should be the priority for policy makers? But I guess there's just two main things. On one hand, you do find that kindergarten experiences can be effective over quite a long time frame. So it is, I think, really worth investing into, for instance, teacher trainings or other sort of interventions to really improve kindergarten quality, because we see that, at least based on our study, and of course, that needs replication and extension, but we see that it can really make a difference for children from many different backgrounds. So I do think it is worth investing into this. But at the same time, we also only find smaller effects of kindergarten on the long term. So there is protective effects, but there is also many forces pulling children back towards their starting points. Like many of the risks that they were exposed to continue. There's different home environments, difficult home environments, and other factors. So I think there also needs to be some more knowledge on how we can support these children from different risk backgrounds. So if I could make a recommendation, I think it would be very useful to collect more detailed data on these groups to understand a bit more how they move from these early risks to the later difficulties, and where we could maybe intervene to make a difference for children. Yeah, because I guess the differences still exist. It's just that the kindergarten experience is a protective factor across all of the risk profiles. So I guess, is there a risk that we over rely on kindergarten as a way of fixing those deeper inequalities that impact on that kind of risk profile in the first place? Yeah, I think that's a question that you could talk about for a long time. And so my thoughts, I guess, is twofold about that. On one hand, I think we have to disrupt these inequalities that often perpetuate across generations somehow. And we do know that early childhood is a good time to do so. It's when children's brains are the most malleable, where they develop key competencies. We do find that kindergartens can be quite effective in this. And especially in countries like Norway, they pose a really unique opportunity to reach almost all children at an early time point, which is not the case for many other interventions that we try to give, because families might not show up to appointments or might not choose to take up offers. So kindergarten can have a really important role, and I think should also take part of this role. At the same time, I do think that kindergarten alone cannot be the fix for the problems. As I said, we find small effects. We find that effects fade out, that other factors take over. So I do think there need to be, as you said, broader systemic changes as well, and maybe also more wide ranging changes, where we look more at the home environment or we look at how do schools follow up on the positive kindergarten experiences. So that it's a little bit more of a comprehensive approach. And that said, I think even if we do that, it will probably take at least a couple of generations to really level out these differences because humans are so complex. So it's very difficult to make wide ranging changes very easily. Yeah, absolutely. And this might also be quite a difficult question to answer, but I'm just wondering how this study fits in with that wider research on childhood resilience and early intervention. Yeah, so I mean, the findings that good relationships are protective against poor outcomes. I think that's been quite well established. It does also fit in with the findings, though, from the early childhood education literature. That effects are good in the beginning and they sort of fade out over time. It does match with previous studies have found in terms of what kindergarten can do and maybe also cannot do. And there's actually quite an interesting body of literature that, for instance, looks into what they call sustaining environments. So that children might not only need this early support, but really over time, we need to have environments that support them developing good skills and good outcomes, et cetera. Yeah, brilliant. Look, we're coming to the end of the podcast. It's been super interesting talking with you. I'm just wondering, is there one take home message? So I'm just thinking, if you had one minute with the national education minister of Norway, what would you say? I think I would say on one hand that I think Norway is doing a lot of things right. We do have great kindergartens that are quite effective in supporting lots of different children. So I would say continue investing in these kindergartens and continue with the plan of supporting children that way. But as I said earlier, I also think I would suggest to them to maybe collect some additional data and look a little bit more deeply into children with different risk backgrounds and how we can support them more effectively. So if we have these data, researchers might just be able to provide better recommendations on how to differently support these children so that they can have better long term outcomes. Brilliant. Katharina, it's been lovely speaking to you. It's been a really interesting podcast. Thank you very much. Thank you for your time. Great stuff. And listeners, if you've enjoyed the podcast, please leave a review or rating on your podcast channel. Thank you very much. [MUSIC PLAYING]