 Welcome to the conference. I'm very excited today because this is our first in-person conference since 2019. Yay! So we decided to hold it in person this year for a change because we wanted to see all your lovely faces. So we've held it online for the last three years. We may possibly hold it hybrid in the future, we'll see. So welcome everybody, it's lovely to see you. So I did have a slide with the programme on but we've had a few tech problems so I'll just run through it. I hope you've all picked up the programme from Reception and we'll start with our keynote from Tamsin. So I'm really pleased to introduce Tamsin, our keynote for today. So Tamsin, new love down garden, is a senior clinical lecturer, armory consultant in public health with the children and young people's mental health research collaboration, also called CHIME, at the University of Exeter. She's an academic consultant on the NHS England mental health of children and young people in England's survey series and she co-leads the academic input to the survey consortium and she's been involved with the series since 2017. Influenced by her clinical experience in child and adolescent psychiatry prior to entering public health, Tamsin's research concentrates on the mental health of children and young people with a particular interest in the application of epidemiological methods for service planning. She currently holds a national institute for health and care research advanced fellowship which studies time trends in child and adolescent mental health and mental health related service contacts which is known as the changes project and the title of Tamsin's presentation is on the slide. I'm really delighted to be here. This is actually one of the first sort of longer in-person talks I've done since pre-Covid so we've had a bit of a struggle getting to see the notes view and all those kind of things. Luckily I can now see my notes so hopefully we'll be able to tell you everything I wanted to say. It's really great to come along to a conference that's focused on the really excellent work being done on a range of health studies so this morning in terms of the mental health of children and young people in England surveys but also the other studies that people are going to talk about. So as Vanessa did the introduction just kind of briefly sort of maybe repeat or expand on a few of those things so I've been working with the data from the MHCYP series for most of my academic career. I initially trained in medicine, psychiatry, child psychiatry and then public health so a lot of my work is around understanding children and young people's mental health at a population level and that's really where my interest is. So during my PhD I focused on studying the data from the 1999 and 2004 surveys. I was then a clinical rater on the 2017 surveys and for the past four years I've been a member of the survey consortium along with some colleagues in the room working on those national surveys. So today I wanted to talk about some, I spoke some insights that those survey data have given us both from research that's being carried out using them and also from the findings themselves from the reports and how they've moved our understanding forward in terms of children and adolescents mental health. Also I want to talk about some of the challenges and the opportunities moving forward. So although I'm a member of the survey consortium I'm also going to be presenting some research that I or colleagues have done and so I'm also kind of speaking here in the capacity as an academic rather than on behalf of NHS England so please also do bear that in mind. So to start with I think I'm probably preaching to the converted in this particular room but a reminder of why population studies and surveys are so important. So we do have increasingly good access to high quality activity data and outcomes data in terms of routine and administrative data including on outcomes now as well which allows us to track children's pathways through services and also examine their outcomes. However they crucially can't allow us to track trends in prevalence because access to services and activity is influenced by so many other factors what's on offer, barriers to services, historical service configurations and any manner of other factors. So it's really crucial that we have population surveys which allow us to understand population need and capture data on those who aren't in contact with services. So I think that's the value of these particular types of data. Internationally I think historically children young people's mental health research and services has perhaps been neglected certainly given its importance. So we know for example that around three quarters of mental health problems emerge before the age of 24, 25 or before the mid-20s and that there are opportunities to intervene but there has been historically quite disproportionately low funding potentially for services and for research and that's internationally. So I think that the national survey series is kind of an exception to that and really a world class resource that we can be using. And that's just on the slide a capture of the three national surveys and the first one from 1999 which is actually quite hard to get hold of on the internet if you look for it. So why is it so valuable? There's obviously probability samples which are designed to be representative of our population and they also contain detailed socio-demographic data so that we can characterise children young people and this is not just in respect of their mental health but also contains the number of wider topics such as education, socioeconomic status and so on. But also I say unique with caution they're fairly unique in terms of using clinically rated standardised diagnostic assessments so I think there are some other international surveys with children and young people that may also use these but it's quite rare. Most studies use symptom scales or other kinds of measures so I'll say a little bit more about that in a moment when I talk about the measures so they use both clinically rated standardised diagnostic measures and also dimensional measures. And there are largely identical measures contained in the baseline surveys as well I say largely identical because they are not completely identical sometimes some particular variables come in and out but in general that does allow us to track trends over time and also to look at things which may drive trends. And finally they contain multiple informants so that's the parent, the young person themselves and the teacher again pretty crucial really for children's mental health because the teacher may be best placed to inform on some aspects the parent on others and the young person on others. So I'm not going to go into much depth about the methods and I know my colleague Drity is going to talk more in depth about the measures and the content of the follow-up survey series that's been happening since 2017 but I just wanted to give you a little bit of background about what the standardised diagnostic assessment is that we use to derive diagnosis. So this significance of this that allows us to identify the prevalence of diagnosable psychiatric disorders in a community or population sample and the types of disorder according to international classification criteria and so that allows us to compare over time and also to provide estimates of children who could benefit from an evidence-based service or an intervention and just to note here that I am using the term disorder throughout the presentation because that is the term used in the surveys we are also quite often talking about DSM or ICD-10 mental health criteria the disorder but obviously I am aware that there are debates over the kind of terms that we should be using in mental health so I'm just using disorder in order to be consistent and because we are talking about quite often those of DSM or ICD-10 diagnosis so just to note on that. So the way that the doorbell works that was originally developed by Robert Goodman has been used in all three surveys is that it includes structured multiple informant interviews that also have sort of free text or verbatim elements where the parent or the young person can actually describe their problem in more detail and all of that information is taken together and then reviewed by a team of clinical raters who are trained in child psychiatry and they are then able to go ahead and assign DSM or ICD diagnosis and that produces our prevalence estimates of diagnosis in the baseline surveys and we also have a dimensional measure the strengths and difficulties questionnaire which is a measure of, as you might suggest, largely difficulties but also includes some strengths for example it has a pro-social subscale and valuably it also has an impact supplement so the impact supplement asks about whether the child's difficulties are impacting on their life at home, at school or with their friends so again that adds additional information the strengths and difficulties questionnaire is used in all of the surveys and in the follow-up surveys it's our main measure of mental health used to estimate whether a child is likely to have a probable disorder or not and again I think Dreetie is going to talk a little bit more about that later so I won't go into detail about the SDQ algorithm so I wanted to start off by presenting some insights from the baseline surveys because I think we've all got personally I have got very excited about the follow-ups that we've been doing during COVID but there's a lot of value to be had from the baseline surveys that we've previously been running so I wanted to focus on some of the key survey findings but also how they've been used in research and then I will move on to talk about the COVID-19 follow-ups so just a reminder when I refer to the baseline surveys these are the ones I'm talking about the 2017 is the most recent and is slightly different from the previous two in using the national patient database as its sampling frame covering England as opposed to Great Britain and also having a wider age range of 2-19 whereas the two prior surveys covered a narrower child age range ooh sorry that's not moving there we go so again it's easy to forget I think the significance of the 2017 figures when these were first released so prior to the 2017 survey we hadn't had data for around 13 years on children and young people's mental health in terms of prevalence or diagnoses at a national level so in effect thinking about things like service planning there was an element of working in the dark and I think the other thing that's important is there was a lot of concern about rising rates of referrals to child and adolescent mental health services attendances at A and E for self harm and a lot of media attention around children and young people's mental health which were suggesting there was I suppose a major crisis and what we did find is that actually there was only a fairly slight increase overall when you look at the prevalence of any disorder that's the blue line on the graph from 10.1 in 2004 to 11.2 in 2017 and that's in 5 to 15 year olds however even a small increase at a population level can translate to a lot of extra demand or need for services when you think about it proportionally and what we also saw is this appeared to be driven by an increase in emotional disorders in particular so that's the red line on the graph and this tallies with other international research which has suggested that many of the changes in children and young people's mental health that we've seen may be driven by increases in emotional difficulties and quite often in adolescent girls and so this has stimulated a lot of interesting interventions around particularly anxiety or depression and school based interventions so these were very significant findings when they were first released in terms of how we used them they've been used in various ways so this particular graph is from a study led by Laura Panacci which examined both physical long term conditions and mental health conditions in over the three baseline surveys and what this allowed us to do is to take a holistic view of child health over time and what she found here was a slight increase in the proportion of children who were healthy and that's very much an inverted commas but was defined as those who didn't have a physical long term condition or a mental health condition an increase in those with a physical long term condition but either sort of steady or perhaps a slight trend to an increase in those who had a co-occurring physical or mental health problem and I think this really speaks to the need to continue to focus on that group of children who have both the physical and the mental health condition and really take a holistic view of child health and I think that's hopefully something we're increasingly doing and there are examples of that in terms of things like the Cambridge Children's Hospital or other examples One of the other advantages of the surveys is that they also ask children, well young people and parents about any mental health related contact they've had with the service in the past year so this allows us to examine the proportion of children and young people with a disorder who meet criteria for a diagnosis who report having sought help from mental health concerns and one of the things these surveys have demonstrated over the years is that the proportion of children who report contact with the mental health specialist so for example child and adolescent mental health services has remained relatively constant over the three surveys hovering around a quarter and that's been the same in 1999, 2004 and in 2017 so around three quarters of those in the population who would meet criteria for a diagnosis don't report having had contact with the mental health specialist what they do more commonly report is contact with the teacher and the professional and this graph is actually for 5 to 19 year olds but if you looked at the 5 to 16 year olds you would find the proportion there is higher I don't have it at the top of my head I think it's something more around the lines of two thirds possibly but it's around half of those aged 5 to 19 who report having had contact with the teacher about their mental health in the past year and if you look at those who report having contact with any professional service it's around two thirds so that's still a third of a children who report not having had contact with any professional about their mental health this is of those who would meet criteria for a diagnosis and therefore could potentially benefit from some kind of intervention the other thing that lies behind this is of course the nuances of who doesn't who doesn't get contact with services amongst these so research by our team in Exeter has been examining predictors of contact with services using the 2017 data and we have found that children and young people from lower from more disadvantaged backgrounds from more deprived backgrounds and who might be of different black and minority ethnic groups appear to be less likely to access services given their level of difficulty so really unless we're addressing this there is the issue of entrenching inequalities so I do know there's a lot of work going on around this and around the development of culturally sensitive services services that are more inclusive and that are outreaching how important that is so finally in terms of the baseline surveys I want to share some findings from a paper we're about to submit so this is still in submission it's not published yet and it may change slightly but what we've done here is we've examined changes in the profile and characteristics of children who meet criteria for a diagnosis across all three surveys this is quite a complex analysis and we have quite a lot of findings we're working through but the headline is that children with a disorder in 2017 appeared to have higher levels of difficulty on the strengths and difficulties total difficulty score and higher levels of impact on the impact supplement than children from the earlier surveys in particular the 1999 surveys and this was the case for self rated difficulties and for parent rated difficulties but not for teachers and interestingly when we examined changes in the profile of children in the criteria for a disorder we did not find the same thing we found that the levels of difficulties remain relatively constant so this suggests that these changes aren't so easy to change in recognition or awareness but that actually children with a disorder may be struggling more there may be something going on which means those with problems are struggling more to thrive so something happening with the environment and of course you can speculate and there's other research going on about being able to start exploring with these surveys and we've only been able to do that because we have identical diagnostic measures from 1999 to 2004 and 2017 so I think that just illustrates the value so moving forward to the COVID-19 follow ups and the follow on surveys just to note these data are becoming available for research access but aren't currently available so I'm presenting here findings and reports themselves rather than my research so some of you in the room may be very familiar with these so please excuse me so just a reminder of where we were in 2019 early 2020s we were actually in discussions about follow up to the 2017 survey about a three year follow up and then these plans had to adapt very rapidly to going into lockdown and of course we have the opportunity to use cobalt of children from 2017 and go back to them and see how they were getting on in 2020 but we needed to mobilise quite rapidly to do that and the 2020 survey was able to get in the field in August of 2020 and we're now kind of fast forwarding to four years later I think Drity is going to give you more details on this particular slide which I've pinched from her because it was very good I think it's very clear and what you can see here is that we have now done four follow ups from the 2017 survey the latest, the 2023 the report is being written at the moment so we're not able to share those findings but we can look at the findings from 2020, 2021 and 2022 and again just to note that when I talk about probable disorder here we're talking about using the SDQ algorithm to estimate that rather than the standardised diagnostic assessment and you'll also note the nature of our cohort here which has advantages and disadvantages which again I can talk about and just flashing up the survey partners that have been involved in the follow ups and so this is quite familiar now this graph to those of us with an interest in child and adolescent mental health but this was a very concerning and I think personally quite surprising finding at the time this very sharp increase we saw between 2017 and 2020 with a jump in the proportion of children with a probable disorder from 1 in 9 in 2017 to 1 in 6 in 2020 and then perhaps just as significant in a way the fact that this high level has been maintained so we haven't seen a drop in this proportion we wait to see what 2023 shows us but we still see the sustained high rates which really show that I suppose this increase wasn't a blip and we still have alarm bells ringing and we have similar findings here when we examine 17 to 19 year olds and in fact what we see here is that things may be going in the wrong direction so this is tentative it is just a one off finding from 2022 we don't know what we'll see in 2023 but based on this we actually saw a rise between 2021 and 2022 in the prevalence of probable disorder in our 17 to 19 cohort so because these are cross sectional data what this might imply is that though we're thinking about those aged 17 to 19 now maybe those who have gone through some of the most significant developmental stages of transition during the pandemic and this may be why we perhaps see this happening in this particular age group but I think this does require further investigation and it is a single data point so triangulation with other surveys and looking to see what happens in our follow up is going to be really important and I think related to this is also something that's been happening around what the pandemic did to help seeking so we know that during the early part of the pandemic there is a decrease in referrals to child and adolescent mental health services generally primary care so a decrease in our consultations and prescribing which the recent research has shown and the 2020 survey also took the opportunity to ask about help seeking and about whether the pandemic had affected help seeking and what we see in this graph is that I think around a fifth of young people aged 17 to 22 said that they decided not to seek help for a mental health concern because of the pandemic during this time and similar proportions that they decided not to seek help for both the physical and the mental health concern so related to this I think we can see there are likely to be a lot of opportunities that were lost for early intervention and problems that may well have become less entrenched sorry, more entrenched and of course we have seen now sort of a then a rise in referrals and potentially a backlog but it's interesting to kind of see what young people themselves were reporting I think another point to make about the survey series and the follow ups is that they've tried to be responsive to concerns over time concerns of stakeholders and of the population so one of the things that was happening during the pandemic and it's still happening was concerns about eating difficulties and eating disorders in the population and so for the 2022 survey actually and the 2021 we were able to be responsive to this and include the screening questions from the doorbell eating disorder module and again I can talk a little bit more about this if anybody wants to ask questions about it but I think the really key thing here is that this is not rates of eating disorders what the screening questions do is they indicate that a child or young person may have an increased likelihood of a broader problem with eating, not a clinical diagnosis so just to put that in context a little bit a very small proportion of those who would be screen positive would actually have a disorder so if you look at the 2017 figure on this graph around 60% were screen positive for possible eating problems but only between 1% to 2% of young women in that year actually did have an eating disorder what this does show is however is that there's been a clear increase in those who may have broader difficulties between 2017 and 2021 so I think that's the important point to make we can't say there's been an increase in those with eating disorders from this data but we can say there's certainly been an increase in those with perhaps who might have broader problems and in the latest round of the survey we are going to include a sub-study which actually applies the full door to eating disorders module to children and young people and we will be able to come up with the prevalence estimates to share and that will be well hopefully later on this year sometimes we're working on it now so I also just wanted to point out a few other findings from the surveys and I think this also example is an example of how we were able to be responsive to concerns around the cost of living crisis so we included questions about whether families had fallen behind with bills, rent or mortgage had not been able to buy enough food or had to use a food bank or couldn't afford to keep their house warm enough and what we see here is that those with a probable disorder were much more likely to live in households that were experiencing those adversities so more than one in ten of those with a probable disorder lived in households that hadn't been able to buy enough food or had to use a food bank that was warm enough and almost one in five had fallen behind with bills, rent or mortgage so what this demonstrates really is that I suppose a stacking of risks or a stacking of inequalities together so again this isn't causal because these are cross-sectional data but it's just showing the different factors affecting children with probable disorder I also wanted to include a slide about some of the more positive things that we have asked about the stakeholders including the Department for Education on the survey questions and we have included questions about children and young people's feelings about school and you can read the different statements that were asked at the bottom there so this question was included in 2022 and what we found is that the majority of children were reporting positive experiences of school in terms of feeling safe being able to be themselves or having a friend they can turn to for support and by probable disorder we do see some differences in that those with probable disorder may be less likely to feel positive in some of these domains but I think it's important to show that we are also looking at things that may be working well for children and young people so finally I wanted to move on to talk about some challenges and opportunities of the data and moving forwards so in terms of the baseline surveys first there are a number of considerations that are using the data so in common with probably some of the other surveys and studies that people are working on in this room the MHCYP surveys have seen a decrease in response rates over time from 83% in 1999 to 52% in 2017 so although there is waiting available to attempt to adjust for this non-response it's still something to bear in mind about the difficulties of engaging families children and young people in these kind of surveys and perhaps also with the follow-up surveys the risk of survey fatigue so there's I think work to be done with communications there if you're thinking of using them to look at trends or to do comparisons over time also to be aware that there are some minor differences in measures used across cobalts although the key mental health measures are consistent and also that there are different core age ranges and geographical coverages so as I mentioned before many covers England and the follow-up theories using those data also has a number of considerations and challenges so we have seen significant attrition between 2017 and 2020 there wasn't any sort of keeping in touch done between 2017 and 2020 so when you think about it it's actually quite impressive that we were able to go back to these families three years later and actually achieve I think what was a 45% response rate which is really quite respectable and we've managed to keep a significant number of children and families engaged over the follow-ups and again Drity will say more I think about the response rates there have also been some inconsistencies in the questions that we've asked over the series but that's largely been due to the responsive nature of sort of regularly meeting with stakeholders and thinking about what questions we should include or refining the questions so in a sense it's been a fact that we have to be quite responsive, we've had to move fast and we've had to adapt questions and so there have been some new questions asked there have been others that have been amended so again if you're thinking of tracking questions over time it's helpful to have a look and see exactly what was included in each of those follow-up surveys sometimes there's been a new topic and sometimes something has been taken out there are again small numbers of children and young people in some groups in the follow-up surveys which can limit our analysis so most notably in children and young people from different black and minority ethnic backgrounds it's not always been possible to break those down by different groups and therefore if you were planning to do analyses sort of examining children and young people in particular ethnic or cultural groups that may be quite difficult to do in terms of the power and similarly in the age groups as the cohort moves forward so no thinking about that as well we do have I suppose relatively speaking an age in four so we have lost our five to seven year olds in the later survey of course they're not lost they've just got older but if you were wanting to do analyses on a younger age group lots of those have obviously aged out of a particular age band on the positive side we're now able to explore the experiences of young people aged 23, 24, 25 in the later surveys and I think given some of the concerns about transition to adulthood and the challenges facing our young people I think that is very valuable and the other thing to mention is the use of the SDQ algorithm to assign probable disorder so when you are looking to make comparisons in terms of prevalence over the follow-up series it's important to always use the SDQ algorithm and that goes to be using the 2017 survey when you're comparing it with the later surveys too that's probably quite a lot of detail but just something to be aware of so opportunities so there is the new release of the follow-up surveys data for analysis and what these will enable us to do is both longitudinal analyses of individual children over time and also the cross-sectional analyses that the survey reports have largely consisted of. I think a really important opportunity is that there may be linkages becoming available to routine and administrative data which will allow us to follow up these children in more detail or to think about what services they use look at education outcomes thinking about for example the national pupil database another opportunity is the use of those data with triangulating them with other surveys and other resources and the role of qualitative data so I think qualitative data is really important in putting the meat on the bones so to speak of the quantitative figures that we're showing an example of this is you may remember I showed you the slide about help seeking and about decisions not to seek help for problems during the pandemic so we did we have some funding to carry out some follow-up qualitative studies with young people who participated in the surveys and to ask them about their experiences of help seeking during the pandemic and again we found that many reported they may not have been seeking help because they felt that mental health wasn't a priority at the time that Covid was more important they weren't sure if a service was available for them or they had concerns about interacting online or they've been put off by phoning a GP surgery and listening to a long recorded message about Covid for example of course they had positive experiences to report as well but I think the value of qualitative data is that it can show us I suppose why we see the survey findings that we do so that's I think another opportunity the 2023 survey report is due this autumn as I think I've already mentioned with the addition of the eating disorders sub-study so I think that provides an opportunity to study the disorders and eating difficulties in more detail so I also wanted to talk a little bit about surveys an example of response of epidemiology and I'm not saying that the MHCYP surveys are unique in this respect I think a lot of health studies and a lot of research are doing this but it's certainly something that I've been quite struck by is that the pandemic has really demonstrated the value of mobilising to try and answer questions in real time that are relevant and to provide answers in a way that can actually inform policy and this does however involve some trade-offs so it involves having to make decisions quickly it involves having to make decisions about what you're going to ask about and what you aren't and whether it's feasible to ask more detailed or perhaps ethically difficult questions so there may be some topics that don't always get included and they are very dependent upon the availability of resource and expertise so with the surveys we've been lucky in terms of a consortium or bringing together Natsen and the Office for National Statistics and NHS England and we've also had the expertise of Professor Tamsen Ford who actually two Tamsens working on the surveys which can get slightly confusing but yes, Professor Ford has been working on the survey since 1999 and so she really has quite an unrivalled overview of this survey series so we've had a lot of expertise to draw on and I think that's why it's so important that we also develop capacity in sort of survey design and being able to analyse these surveys so that we can mobilise people going forward and I think surveys also should strongly reflect the need of those who are using data so again it was an example in the survey series of engaging with various stakeholders but I think we also need to think about those who are contributing their data to the surveys so it's quite a lot to ask to go back to children, young people and families every year as we have been doing for the past few years thank you and so what we wanted to do in the most recent round was to do some engagement activities so these were led by Natsen but also involved the University of Exeter and the University of Cambridge so we for example went back to the young people taking part they offered a thank you gift, we had a newsletter and we also held some activities where we asked them what are the topics that you want to see involved in the surveys how can we keep you engaged what would you like to know and they did highlight to us quite important things that they wanted to be asked about that they thought were important to them and that did include for example cost of living so even some of the younger children wanted to be asked about that they said it was affecting us you should ask us about it don't be afraid to ask us about it and also things like climate change and I think the other thing I wanted to highlight was that there was kind of a challenge from the young people that we spoke to in the engagement activities so they wanted to know how their data was being used that it was being used and quite crucially what difference it had made so I put some of their quotes up there so you can read them these are from young people who took part I think most of them had taken part in the survey some of them were general population young people but they all took part in our engagement and I think it's a challenge to all of us who are involved in collecting analysing or using data to work with it to make sure that it is benefiting children, young people and families so that's probably a good note to finish on given we're nearly at time so thank you for your attention today I've put some additional acknowledgements on that slide and I'd be really happy to take any questions Thank you so much Tamsyn that was brilliant, what a brilliant illustration of the power of these datasets and really important striking results