‘CAMHS around the Campfire’, is our a free live online journal club.
- Slides from the session
- Open Access JCPP paper by Dr. Faith Orchard and colleagues; Orchard, F., Gregory, A.M., Gradisar, M. & Reynolds, S. (2020), Self‐reported sleep patterns and quality amongst adolescents: cross‐sectional and prospective associations with anxiety and depression. J. Child Psychol. Psychiatr. https://doi.org/10.1111/jcpp.13288
- Sleep expert Professor Colin Espie’s blog on the paper.
- Dr. Faith Orchard video abstract 5 minute video abstract
- Dr. Faith Orchard discusses the paper in this 10 minute podcast,
- We have a research digest of this paper plus a whole section of research disgests on Sleep
ACAMH Sleep Topic Guide
All ACAMH content on sleep can be found via this search
Related Mental Elf content on sleep
Andre Tomlin – Hi, everyone. A very warm welcome to the event. It’s a real pleasure to be here. I’m Andre Tomlin from The Mental Elf. It’s lovely to see so many familiar names in the list. A very warm welcome to everyone. We’ve got people joining as we go, but hi, Chloe. Hi, David. Hi, Dom. Hi, Felicity. Hi, Maria. Lovely to see you all again. Thanks for joining us. I am really pleased that the first event in this CAMHS campfire series is going to be discussing whether a good night’s sleep can help prevent and treat youth depression, and anxiety, a really important topic, a topic that’s often overlooked and something that we all know about, something that’s really accessible.
We’ve got a brilliant panel of experts who we are going to be hearing from over the next hour. I’m going to just briefly introduce them so you know who’s on the call, I don’t know if we can get a kind of gallery view going so everyone can see the people who are on the panel while we introduce them. First of all, we have Nkem Naeche, who’s a student nurse and has lived experience of anxiety and sleep problems. She’s from the Young Person’s Network at the amazing McPin Foundation mental health research charity. Welcome, Nkem. We have Professor Colin Espie, who’s a professor of sleep medicine at the University of Oxford. And as Matt said, he’s a notable worldwide expert in this field. And he’s recently, just yesterday published a blog on the mental health website summarising the research that we’re going to be talking about tonight. Welcome to the webinar, Colin. We have Dr Faith Orchard. She’s a lecturer in psychology from the University of Sussex and props to Faith because she is the researcher who wrote the paper that we’re going to be talking about tonight, we’re going to be critically appraising.
So, she’s putting herself in the firing line. She’s going to answer lots of questions. She’s going to talk very enthusiastically about this area of research. It’s great that she’s here to answer the questions we have about her paper. Welcome, Faith, thanks a lot for joining us. We have Dr. Steph Lewis, who’s a psychiatrist and a researcher from ACAMH, she works in South London. And we also have my colleague, Douglas Badenoch, from the National Elf Service. Steph and Douglas are going to be our chat moderators tonight. So do post in the chat if you have any questions or comments. We really want to hear from you. We really want to see who we’ve got on the call. And because of that, I think the starting point is going to be a poll. So if we could launch the first poll, please, ACAMH people, we’ve got some ACAMH elves in the background, and hopefully we’ll all see that. So this is a really simple poll. Just tell us who you are, what professional group matches you best this evening. And we will let one hundred or so of you answer that? And then we will get a sense of who we’ve got here on the call today.
So if you want to stop that polling then please and publish the results for us. We can see we’ve got a really nice mix, actually. Clinical psychologists, I think, are the highest group alongside psychiatrists and researchers. But we’ve got lots of educational professionals. We’ve got students, other health professionals, social workers, nurses. Brilliant. Thank you very much. That’s great. We’ve got such a mixed group. So, yeah, let’s turn that off now. I think that’s lovely because I guess one of the things that we’re trying to do with these events is to facilitate some kind of democratic conversations that bring together all of these diverse groups of people so we can make some decisions about what we do with this evidence. And I guess we’re going to come to that at the end.
So, first of all, I just want to give you a sense of what we’re going to do. We’re going to spend 50 minutes taking you through this process that some of you may have done thousands of times and some of you may never have done. That’s looking at a research paper and critically appraising it and thinking through what are the strengths and limitations of this paper? Is it asking a question that’s relevant to front-line practise? Does it come up with some meaningful results? Do they have any implications for people with these issues, for young people with depression and anxiety and sleep problems? Do they lead us into future research? Or do they actually lead us into some things that we can do in practise?
So you’re going to be doing some work along the way, all of you, so get your thinking caps on. And as I said, pop your questions and any comments you have in the chat. It’s also worth saying, as Matt has already said, the webinar is being recorded. So we’re going to share everything with you afterwards. There’s going to be a video, the chat, slides. There’s going to be papers. We know how much you all love a pack from these sorts of events. So we will be preparing one and we’ll be sending it out. So don’t worry about asking about any of that stuff in the chat. Focus on what we’re going to be talking about.
So, we have some slides, can we put up the slide that’s got the kind of setting the scene, please, from our presentation. So this is the question for our campfire. “What does the evidence from a population-based birth cohort study tell us about the links between sleep in adolescence and anxiety or depression?” So I guess, as a starting point, we want to get a handle on what you think of this area, sleep as a target for treatment or research.
Can we put the second poll up, please, and just get a sense from people of where they are at? So this is a question that you can answer from one to ten. 1 is strongly disagreeing and 10 is strongly agreeing. You might need to scroll down to see 10. To what extent do you agree that sleep is potentially a helpful target for preventing depression and anxiety? So we’re seeing some people strongly disagreeing and we’re seeing a lot more activity, I think, at the top end.
OK, let’s switch that off now. We’ve got 90 percent of people having responded. So the biggest response is actually 10. So there’s allegiance bias here in this meeting, you’ve all come along because you think this is a good thing. So we need to be careful about that. Actually, quite a few people are saying seven and eight as well. So, you know, agreeing to some extent and a reasonable proportion of you, 10 percent in total, are at the sceptical end of the spectrum. Either that or you got it wrong and you put 1 thinking it was strongly agree. But I think we’re OK with that.
So let’s turn that off. Thank you. So let’s move on from that. I’d like to first of all turn to Nkem and I think it’s really great that we’ve got some lived experience of these issues. I guess I’d like you to share your perspective on this, Nkem, and tell us a little bit about your story. Do you think sleep is something that we can target, help people with their sleep problems that might then help people with their depression or their anxiety? What’s your perspective on this?
Nkem – I completely agree. I think sleep is a topic that… I’m a student nurse and sleep for us is something that we holistically look at towards health, mental health and all such. I think, when I was an adolescent, who didn’t really understand mental health because it wasn’t really discussed for us necessarily, I didn’t realise how sleep exacerbated that, and vice versa. It’s a funny link. I personally struggled because teenagers generally work as night owls, so I’d fall asleep really late and because of my anxiety, fall asleep really late, and we’re talking about one or two a.m. as a 13, 14 year old, then developing hourly interval sleep. So every hour on the hour waking up, not really getting any good quality sleep, and then struggling to wake up, and that resulted in periods of fatigue and drowsiness throughout the day, which you all can imagine would affect academic performance, make you irritable, and have you essentially creating really bad relationships with teachers because they think that you’re not putting in effort or things like that. And really, it’s a result of not being able to sleep. That in turn creates pressure. And so you start to feel even more anxiety and it’s just a really, really bad cycle. So, for me personally, I think it’s a very important aspect for us to understand and develop and hopefully build interventions and better ways of teaching people how to sleep better and helping anxiety.
Andre Tomlin – Thank you. Tell us, Colin, a bit about the clinical perspective, you’re the professor of sleep medicine here in the room. What’s your perspective on this?
Prof Colin Espie – Well, yeah, thank you. Yeah. And thanks to the Mental Elf, ACAMH, and of course Faith and team for doing this research and bringing this in front of us. I qualified 40 years ago last month, with a round of applause for that, or a vote of sympathy or something. When I first qualified, I don’t think people thought, you know, the basic assumptions I think were made about things, and sleep over most of those decades has been seen really as a symptom of something else, which is an extraordinary proposition when one considers that every every species sleeps. To think that it’s just an incidental phenomenon. So I think what we’re recognising from the literature, including the study that Faith and team have done, is that sleep is actually of fundamental importance. We spend 36 percent of our life on average across the years sleeping. We see not only a lot of people with sleep problems coming to us directly, but we see an increasing number of researchers and clinicians asking for training in sleep medicine. We’ve got people from 31 countries on our master’s or postgraduate diploma programme on sleep medicine at Oxford over the past few years.
So things are a burgeoning interest in recognition of the importance of sleep, not just in this silo, if you like, if you think of it that way, Andre, of sleep, you know, as a specialist area. But on the lateral, that sleep becomes relevant really at every age and stage in life and really with every disorder. The last thing to say there is increasingly we’re seeing that research funders are unwilling or reluctant to fund research on mental health now unless sleep is properly measured.
Andre Tomlin – We’re going to come on to the paper in a second and talk about this specific piece of evidence that Faith has recently produced, but just give us a picture of the general evidence landscape about mental health problems in young people and sleep. What else is there out there?
[00:11:29.560] – Colin
Yeah, well, what we tend to find in young people is there’s two main reasons why people struggle to sleep. One is what we might call the typical psychological difficulty of stress-related insomnia, where the mind is racing, or it’s hard to get into a pattern, hard to get to sleep. You’re waking up, can’t get back to sleep. But the other problem that teenagers have is something called circadian-phased delay. I think it was mentioned earlier, the body clock, a tendency to be a night owl during your teenage years. Having had three kids and four grandkids, I’ve seen this a few times, as well as remembering my own experience, that it’s actually difficult to fall asleep early and it’s difficult to get up too early because the body clock positioning is a factor. So in truth, sleep and circadian factors operate together. That is the amount of sleep we need at the timing we need it, and we see both of those difficulties: getting to sleep and getting to sleep at the right time are problematic for teenagers in particular.
Andre Tomlin – So this is encouraging in terms of this webinar, what we’ve heard already from Nkem and Colin. Faith, put us in the picture a bit with your paper, how does your research fit in? What questions specifically did you try and answer and what inspired you to do that in the first place?
Dr. Faith Orchard – Yeah, so hi, everyone. So I guess the thing to say is that this has been a kind of a growing body of literature for a number of years. So people trying to explore this longitudinal perspective relationship between sleep, anxiety or depression. So we were adding really to that body of literature that was already growing. Every study has its strengths and limitations. There were a few gaps in the literature that we were hoping we could fill with this really great resource from the Dahlback data. I think the main thing is really whether… Often, sleep is measured at a very basic level. So, I mean, Colin just touched on this already. But particularly what was quite common is you’d see just even one or two items on a questionnaire referring to sleep as the measurement, which, you know, you often can’t control what measures there are if you’re looking at longitudinal datasets. So that was kind of the nature of what was existing. Also, when measuring sleep, often people just report sleep problems rather than looking at the different facets of sleep problems.
So different things that might be affecting you, different things that might be a problem for you. There was very little known about whether different sleep problems might be more of an issue in terms of anxiety and depression. The other thing that is almost never looked at is weekday sleep versus weekend sleep. So are there differences in the week? And actually this is really relevant to one of Colin’s last points about this idea that, you know, the circadian clock, the body clock shifting is that we know in the week on a school night, it’s particularly problematic for young people because their body clocks want them to go to bed later, but they’re still getting up for school. So that kind of comparison between the weekday and the weekend is quite important. So we were really fortunate. We looked into the Ahlsback data and found that there was good measures on some of these things we were able to look at to address some questions that were perhaps underrepresented in the literature that existed already.
The other thing that I haven’t touched on yet is that there also isn’t very much known about the sleep problems that exist in individuals with depression compared to individuals with anxiety. So do they experience the same sleep problems or are their sleep problems different? And again, having the diagnostic data in Ahlsback and having the different questionnaire measures of depression and anxiety meant that we could do those comparisons. So that kind of led us on to want to address two key questions with this paper. So, first of all, do sleep problems or different facets of sleep differ in adolescents with anxiety compared to adolescents with depression? And those two groups compared to a non-clinical group, so a group that don’t have any anxiety or depression. And then are there specific sleep problems that predict future anxiety and depression, so across multiple time points. This is one thing I haven’t also touched on, which is we were able to look at multiple time points in the future, whereas most research tends to look at just a baseline and a follow-up. So we were able to see whether or not it predicted your future in later adolescence, but also in adulthood as well.
Andre Tomlin – Thank you. That’s great. There are some fantastic resources that we’ve shared already on the ACAMH website and some stuff that Faith has produced herself. So the paper is open access and you can read it. There’s an abstract, a video abstract, there’s a podcast. Can we put the link for that stuff into the chat now? Because I’m sure some of you will have read that and looked at the paper in advance. But if you haven’t then go and have a look at those things now or afterwards. There’s also a great blog that Colin wrote that we published on the Mental Elf yesterday. Mental Elf blogs are unusual because they don’t just summarise the research. They also talk about the strengths and limitations and the implications. So they say things that are critical of the paper, as well as highlighting the good things that come from it. That’s what we’re going to look at now. We’re going to look at validity. Is this evidence valid?
So we’ve got a poll, first of all, for those of you that have read the paper, we’d like to ask you what you think. Is this research valid? Can we rely on this research or is it so limited in its methods that actually there’s not much that we can tell from this? Tell us what you think and feel free, if you are saying unsure, to tell us why you’re unsure. Is that because you haven’t read the paper, or is it because you don’t know enough about research methodology to make a comment on whether it is or not reliable? Tell us what you think. Thank you, Stuart.
And there’s a link from Douglas to the Mental Elf Blog. If you want to have a quick read. The blogs are great because you can read in five minutes a summary and then you can read the paper. OK, so we’ve got 50 percent of our audience saying yes, actually slightly more than 50 percent. Fifty five percent. That’s great. Nobody’s saying no. That’s interesting. But a lot of you are saying unsure. So the other forty five percent saying unsure. So we’ll ask that question again at the end and we’re going to see what results we get, whether we get a higher number of yeses or whether we get some no’s, perhaps. Let’s see.
So, Colin, let’s turn to you now, I’m interested in what you thought of Faith’s paper. You had to critically appraise it when you wrote the Mental Elf blog. Tell us a bit about the methods that Faith and her team used. Tell us a bit about the strengths and limitations of this approach.
Prof Colin Espie – Yeah, so as Faith has mentioned, this is a longitudinal prospective study, which means that people are followed up over a period of time and questions are asked at various points. So you can see the temporal relationship between different experiences. In this case, we’re interested in the relationship between sleep and mental health and particularly pre-existing sleep status and forward-looking mental health to see if it’s a predictive variable. Now, this comes from the area of research called epidemiology. One of the limitations of epidemiology, traditionally, that Faith mentioned is that you don’t necessarily get a very good index measure of the thing you’re interested in, because maybe there’s one question on this and there’s one question on that, and this kind of thing, so nothing is measured in great definition. I think the particular strength of this paper is that there’s a very strong, I think, set of questions relating to not just sleep alone, but also this circadian factor of the timing of your sleep and the differences between weekdays and weekends, something called social jetlag, that I can explain further if people are interested.
I think that is a particular strength, because then when they look at what is the relationship between sleep as a predictor and mental health, they’re able to look at what aspects of sleep are the most important predictors. I don’t know if it’s a spoiler to say anything about validity, Andre, but I think validity actually is a strength from a personal point of view. The reason for that is that validity is about does this address the issue it says it’s addressing? And, where sleep is concerned, and where mental health is concerned, we are dealing mostly with self-reported experiences. We don’t say to somebody, “You’re not depressed, because you don’t look depressed to me.” If somebody says they’re depressed and feels depressed, or says they’re anxious, it doesn’t matter if we say, “Look, you look very confident. It doesn’t seem to affect what you’re doing.” People live with mental health experiences and continue to try and operate despite them. And sleep belongs to that category of things as well, where the experience of sleep is very, very important and, actually, the objective science of sleep study that we do in the sleep lab in Oxford is still to catch up properly with being able to properly measure the experiences that people have. So I think the strength of this study are in the robust data collection questions that are asked, the longitudinal nature of the study, and the fact that the analyses were conducted fairly conservatively.
When you ask several questions simultaneously, there’s always a good chance that one of them is going to give you the prize. What they did in this study was they did a correction, called Bonferroni correction, for multiple comparisons. And I think that gives greater confidence that the results are real rather than just artefacts of statistical analysis. The last point to say is that the overall predictive power of the relationship in the study is relatively small. That’s not unusual in epidemiological research. There’s many other things that will that will contribute to why someone might get depressed. The point of this study is that a small proportion of that explanation is accounted for by something that is a precursor, not just a consequence to getting depressed, it’s a precursor to it. And importantly, it’s a treatable and fixable one.
Andre Tomlin – Thank you. To everyone now in the meeting, it’s absolutely fine. I want to give you all permission to say, “I don’t know,” or “what does that mean?” Or “I have no idea what that thing is all about.” I have working on mental health research for the last 20 years and every single day that’s how I feel when I look at research papers. Most days I give Douglas, who’s here in the webinar, a message saying, “Please help me. I have no idea what this means.” And so we all feel that it’s fine. Oh, he’s saying don’t say that, OK, but I feel that I’m just projecting here now. So please do post your uncertainties in the chat and we will help you with that, or Douglas will help you with that and hopefully we will bring some of those in later on as we open up the conversation.
I wanted to turn to Nkem now to just tell us what you thought of this paper. And obviously you’re coming at it from a lived experience perspective. Having had sleep issues and anxiety issues yourself, what did you think of it? And did it feel like a robust piece of research to you?
Nkem – So I couldn’t speak too much on the science behind everything. I can only make basic understanding. So I think it is a very good piece of research, more also because it reflects what I’ve experienced and also what I’ve heard from my peers, who, when we were 15, 16, talking about our lack of good sleep and feeling drowsy throughout the day. I guess as I can see in the comments now, some people have made a point about the size. I would say on an initial basis, maybe the size and the area specifically, also the ethnicity element, that it was 98 percent white would speak to me in a different way only because I’d think, well, the UK is quite diverse, maybe would that represent ethnic minorities also, maybe socioeconomically, would we, myself as a BAME community, would we fare worse? And would we see higher levels of anxiety and depression related to bad quality of sleep? Overall, though, for my lived experience, it does reflect what I’ve gone through and what my peers have gone through, too. I think it’s a very good, robust piece of research. So, yeah.
Andre Tomlin – Great. Thank you. I want to bring in Faith now. I mean, respond to that specifically, Faith, because you’re using this ALSPAC birth cohort, this Children of the 90s study from Bristol. Bristol is a really diverse city, isn’t it? So why have we got 98 percent white in this paper?
Dr. Faith Orchard – Yeah, I mean, I think this was touched on by someone in the chat earlier, I spotted that, you know, the nature of signing up to these studies is that you are going to get people who want to sign up to the study. So, you know, the ALSPAC team did a really brilliant job of, you know, really going to all the hospitals and getting as many people as they could, and they’ve got amazing numbers of pregnant women, around 14,000 were recruited, which is really brilliant. But, obviously, Bristol is more than 14,000 people, more than 14,000 pregnant women, you know, so it is going to probably appeal to a certain population. Having said that, ALSPAC do publish information on their website about how representative their sample is of the UK as a whole. And broadly, they’re not too far off on a number of different kind of demographics and factors. But I think, you know, and again, this comes back to various comments, and Nkem mentioned this as well, that even if it’s relatively representative of the UK, it still means it doesn’t address other questions that we would want to know. We’d still want to know about different communities, different populations. I saw someone mentioned, you know, city versus rural areas. There’s going to be a lot of things that feed into this. I suppose you can’t do everything in one study. But I think this is a really important thing to highlight there about the representation.
Andre Tomlin – Tell us a bit about the outcomes that you measured in the study and also how it worked in terms of when you took these assessments and how many people kind of dropped out during that process because it covered a number of years, this study, didn’t it?
Dr. Faith Orchard – Yeah, that’s absolutely right. So we started with the age 15 data, and this was a subset of those participants that took part in a more intensive kind of sampling stage where they included the sleep and the diagnostic data. So of those I mean, obviously, 14,000 mothers to begin with are recruited by age 15. Obviously, you’re expecting a drop out. But this study also focussed on a subset of the remaining participants anyway, which was just over 5,000 at age 15. So at that point, we had, as Colin mentioned, a nice range of different sleep questions that we could report on, and diagnostic data for anxiety and depression so the researchers that put together the study used the Dorber, which is a diagnostic tool at age 15. So we had that data all available then and then we had multiple time points in the future from there that we were able to look at where there was more measures of mental health. So in the study, we particularly focussed on the ages 17 and 24, which had more mental health diagnostic data, and age 21, where we had questionnaires for anxiety and depression. So we had different types of mental health measures at different time points.
We unfortunately didn’t have detailed sleep data available to us at those later time points. So we couldn’t look at bidirectional relationships. We were only able to look at the one direction, but that was the main thing that we wanted to explore anyway. So that wasn’t too much of a major concern for us. So I can’t remember if you had other questions in there, Andre.
Andre Tomlin – I get the sense that we’ve now got 150 people saying, “Okay, tell us the findings. What did you find out?” So, tell us the findings.
Dr. Faith Orchard – So there are a lot, so I’ll try and be kind of brief. So the findings broadly branch into two types of analysis that we did. So we did what we call the cross sectional findings. Now, this was just the aged 15 data. So this was where we looked at, remember I said earlier we wanted to know whether young people with anxiety or depression, how their sleep might differ from each other, but also how it would differ from individuals or teenagers without anxiety or depression. So that was the first thing that we did. We also broke up the sleep items into what we call sleep patterns. So essentially your sleep time. So what time do you go to sleep, how long you sleep for and then your sleep quality. So does it take you a long time to fall asleep? Do you wake up in the night? Do you feel tired in the day? So Nkem touched on this as well. So that’s kind of daytime sleepiness experience.
And what we found was that for the sleep pattern, so the sleep time, the depressed teenagers had worse sleep than the anxious and the control group. So they were going to sleep later and getting less sleep overall compared to the other two groups. And interestingly, the anxious group didn’t really differ much from the control group. There was one variable that was very slightly different. I think it might have been wake up time on weekends, but the difference was in real terms, very, very small. Whereas for the sleep quality variables, both the clinical groups, the anxiety and the depression group, were worse on pretty much every sleep variable that we measured. So it seemed to be that quality was quite problematic in both groups, but the patterns were only really affected in the depressed group.
So those were our cross-sectional findings. Then we have the longitudinal prospective ones. So this is looking at the prediction of whether or not sleep predicts future mental health. And actually here we had a really nice, consistent pattern across each of our analyses. So we looked at whether sleep predicted the diagnoses at 17 and 24 and the symptoms at 21. And for every analysis, we got a really clear picture that your total sleep time on a school night, your daytime sleepiness, how often you woke up in the night, and the perception of how bad your sleep is were all predictors of future anxiety and depression at every time point. It was a really, really clear pattern that those things came out.
The one variable that was a bit inconsistent was how long it takes to fall asleep. So this predicted anxiety and depression in some analyses, but not others. And actually, I think the effects were very small for that variable, which is probably why we’re seeing a slightly more inconsistent picture, but quite a clear message that sleep problems were predicting future anxiety and depression, although, like Colin said, the effects were quite small. But we would expect that because of the number of things that would also be predicting future anxiety and depression.
There was one thing when I was sort of thinking about today and, you know, things to reflect on, there was a couple of points of interest that kind of stood up to me today when I was mulling over the results, and that is that there seem to be different facets of sleep or different sleep problems that were characteristic of anxiety and depression compared to predicting anxiety and depression. So, quite interestingly, how long it took to fall asleep was a really clear problem in teenagers with anxiety or depression, but it wasn’t a clear predictor of future anxiety and depression. And on the other side, the anxiety group didn’t seem to show problems with what time they went to bed or how much sleep they got. But it was a clear predictor of future anxiety. I was wondering whether this might actually tap into almost the kind of the chicken and the egg argument, perhaps there are different sleep problems that occur when you are anxious or depressed and different sleep problems that might predict future difficulties, which is a really interesting thing to consider, but it is important we measure different types of sleep and not just “do you have a sleep problem?”
Andre Tomlin – Yeah, I wonder if we could get into the appraisal of the research a little bit now. I’m thinking maybe we can bring in Steph and Douglas to summarise some of the questions that we’re having in the chat and some of the comments. It feels like there’s lots of comments coming in around outcome measures and how that was done, around the sample, not just the size of it, but the representativeness of it. Are there any key questions, Steph and Douglas, that you want to ask of Faith at this point?
Dr. Faith Orchard – So, there were a couple of questions about sample size. For example, Stewart Sadler said it’s a huge sample size. Is there a chance of a type one error? But there’s some disagreement about this because Annika commented that it was quite a small sample size. It would be great if you could clear that up and put it in the context of research.
Yeah, it’s a really great question and it’s something that we obviously have to think about. So you’re right, the sample size, well, so broadly the sample size is pretty big. But I guess it’s also worth saying, and I know someone spotted this in the chat, that the diagnostic groups are actually relatively small because obviously when you have a population size study, only a small sample of those individuals will have anxiety or depression. So those between group comparisons have some slightly smaller groups in them. So I think for those comparisons, you know, you’re probably unlikely to have the issue of the type one error because of sample size because actually those groups are quite small. But I think, as Colin touched on, and I know a couple of people mentioned it in the chat, because we were running multiple analyses, we did control for that multiple testing, which would also feed into, you know, seeing a statistical error, which is actually quite a conservative approach that we took. We wanted to be really cautious about how we handled the data. So that was something that we tried to do to accommodate that potential problem.
Douglas Badenoch – I was going to ask that as well, but the other one I was going to ask about was just about, Colin mentioned it briefly, the strength of the association, you know, how big is the effect? And you mentioned just now Faith. Actually, it’s probably there’s kind of crossing over effects that are more complicated than just a simple linear aspect of causation. So, what are the challenges in getting that across and unpacking that?
Dr. Faith Orchard – Let me check I understand your question. So, first of all, the size of the effect. Sometimes it’s easy to quantify this and sometimes it’s harder. And unfortunately, when you’re dealing with sleep diaries, it gets quite complex. Analysing sleep diaries is a bit messy, and that’s essentially what we’ve got here. We’ve got time. So it broadly looks in the range of as the scales, as you kind of go up a point on the scale, your risk is somewhere in the region of kind of you’re 1.6 times more likely to have a diagnosis of anxiety or depression. It varies obviously from variable to variable, but that’s broadly what we find. It’s hard to do that with the sleep times, as I said, because that’s quite complicated to interpret. So it is a relatively small effect, but then it’s still there. And I think, like Colin was saying, you know, any effect is something that we need to be aware of and especially over, you know, almost ten years to that age 24. The fact that you’re still seeing that this is playing a role in the prediction is something I think that’s noteworthy, even if it’s a small effect. Sorry, Colin, were you going to jump in there?
Prof Colin Espie – Yeah, I’m happy to just quickly comment on that if you want. I think we’ve got to bear in mind this is a population level study. So therefore, you’re interested in here, not clinical level effects in terms of the effect sizes, which you might expect to find 0.4 or 0.5. You might expect to find 16 to 25 percent of variance explained, which is the R squared, which is the value you have in your table, and it’s down about two or three, four percent or something like that. This is a population level study. So therefore, the people in here who don’t necessarily have major sleep problems or major depression problems, you’re looking at how these two measurements co-vary over time. A good parallel to this would be if we think about what decisions have we made in public health based on small but stable effect sizes? We wouldn’t have added fluoride to the drinking water supply to protect people’s teeth if we didn’t have effect sizes at this level. This is a population health level significance. That means we should be doing something about this for the population as a whole. For those people who have got specific problems, and people have been posting comments about, you know, people with neurodevelopmental disorders and other things like that, people who are more prone to have sleep difficulties and probably emotional regulation problems and maybe anxiety and depression.
If you did a study only in that group, you would probably find much higher effects. If you are only looking at the clinically at risk populations, you would probably find much higher effects. So I think it’s important to bear in mind this is a population level study. So therefore we should be looking for population level statistics that give some comfort around the question, “Is this worth addressing? Should we be doing something more than we’re doing at the moment about this?”
Andre Tomlin – That’s nice, Colin. Thank you. So let’s move on. I’m interested to get the audience’s views of this evidence, this specific study that Faith has produced and whether we think it’s relevant to clinical practise. Are the findings of this research something that we can use in clinical practise with young people affected by anxiety and depression? So let’s open up this poll, poll four, and let’s ask everybody just to quickly say what they think. It’s a yes/no/unsure question again.
Thank you. So you’ve had some time to think about it. Yay, we got a no! Was that you, Douglas, just on purpose, trying to…? Was it you Faith? Okay, so we’ve got a very, very convincing result here, we’ve got almost 90 percent of you saying, “Yes, this evidence is clinically relevant. Eleven percent saying unsure and one person, thank you, just to have a bit of balance, saying “no”. OK, that’s great. So, conclusion’s, let’s think about what we do with this. Let’s think about how we move forward with this. Over to you, Faith. What were your conclusions as the researcher from this work?
Dr. Faith Orchard – Yeah, so I think obviously I’ve gone over the results, so I won’t go over them again, I think there’s lots of other lovely discussions that we could be having. But I guess the key thing for us that came out from this, you know, myself and the other researchers involved was that there seemed to be, you know, a good suggestion here that firstly there might be a benefit to working on sleep problems in young people with anxiety and depression, people who already have diagnoses of anxiety, depression, symptoms. And actually that this might potentially support their recovery from some of those difficulties. And the other, I think potential impact of this is that there’s also potentially a benefit of supporting sleep in young people who don’t necessarily have anxiety and depression but are presenting problems with sleep in their teenage years. I think what we need to do is, is do work to build up the evidence around this, where if we target sleep in teenagers, does it not only benefit them in the immediate, but does it potentially prevent them from future problems with anxiety and depression? So that’s what we’d really love to see happening next in the field.
Andre Tomlin – I’m the first person to be muted when I’m talking. It’s always nice to have the chair being such a pro. I guess putting it into context, maybe Colin could do this for us. I’m thinking, Colin, as you say, public health interventions. There’s loads of stuff going on in young people’s lives. You know, we could be thinking about, you know, are you lonely? Are you being bullied? All sorts of unexpected things going on. Do you think repetitively in a negative way and all that kind of stuff? Do you have enough money? You know, how does it fit in with that? And for you, what kind of implications are there not just for clinical practise, but also for where we go next with research?
Prof Colin Espie – Yeah, well, I mean, just to say at the outset that there’s no question that social disadvantage and inequity in all its forms is probably the strongest predictor of mental health difficulties. And traditionally, I suppose these aren’t clinical areas to focus on. But just to say I’m totally aware that that’s a huge issue and we shouldn’t just skirt around that. It’s an important thing. And there are some public health measures that perhaps we can take around sleep that would improve that. So, for example, in our schools and PHE, we tend to have a curriculum connected to the Ministry of No Fun, right? You know, don’t take drugs, don’t drink too much, don’t get a girl pregnant, you know, you’re getting overweight, et cetera. A lot of the traditional health education topics are really quite negative to young people and they have that connotation, even if you’re interested in things that are a little bit closer to them, like your confidence level or how you’re feeling, these are quite difficult things to talk about, actually, at any age and stage in life. Sleep is something that is surprisingly unstigmatised as a topic to discuss. And it is a gateway, if you like, a Trojan horse into being able to discuss mental health issues with people, because it’s quite easy to ask even strangers or to say to a stranger, “I had a terrible sleep last night,” you know. Quite often when I do these kinds of things on radio or whatever, the interviewer at the end says, “Professor, how do you sleep yourself?” I can’t imagine that being appropriate for them to ask me almost any other personal question about my mental health. But one feels that that is OK to talk about that. And I think that’s one of the, you know, the secret powers of sleep, not just what it does for our brain, what it does for mental health, but also what it does for conversation. You know, so I don’t think when we think about our services, we should just be thinking about clinics and assessments and clinicians. I think all that is important and that should be prioritised for people who are really struggling. I think there’s a lot that can be done just in the everyday.
Prof Colin Espie – There is actually a working group commissioned by Nadine Dorries, a cabinet secretary, to look at the importance of sleep and health. I’m delighted to have been asked to be an expert adviser to that. And maybe, I don’t know what your political views are. That’s not my point. But maybe it’s getting the attention even of parliamentarians because there’s virtually no legislation about covering sleep in anything.
Andre Tomlin – Interesting. Nkem, do you want to come in with a kind of lived experience perspective for us now again, at this point? What do you think the implications are for young people of this evidence? And having read this and been part of this webinar, what do you think you would be saying to your peers and to young people struggling with mental health issues and sleep problems right now?
Nkem – My experiences of anxiety and sleep have been not great. I sought services when I recognised that maybe this wasn’t normal patterns of sleep behaviour. And unfortunately, I was turned away. I was told that I wasn’t eligible and that I was getting enough sleep, even though it was hourly and sort of if I did get a good night’s sleep, it was maybe six hours at a time, which, you know, we know that rhythms should be 7.5 hours plus for young people, adolescents. I think sometimes it can be hard as a young person to continue to seek out services when you’re turned away. There can be sometimes disconnect and you don’t feel as though you’re heard. But then it can be quite enlightening being part of things like this, where you see that people care and they do want to put in the effort and they do want to find interventions and things that will work. So what I’ve had to take into myself is to continue to be a part of this. Continue, and hopefully there will be change as we’re seeing Faith, Colin, being a part of all of this. It’s really good. It’s really nice to see work being done. A lot of us go through this, a lot of us have the same experience I have had. I’ve seen in the comments, some people commenting that and all I can say is keep voicing these concerns, keep addressing them, and we see people making these changes.
Andre Tomlin – Thank you. Yeah, really positive, but also really kind of empowering that you’re coming and saying that and telling us your story, so thank you. We’ve got one more poll now. We’re going to go back to the question we asked before and see if you’re all thinking the same as you were before, seeing if any of these uncertainties can be turned to yeses or no’s. So if you could launch the next poll, please. And while we’re doing that, also, think about any questions you’ve got that haven’t been answered yet, pop those in the chat. We’ll be turning to Douglas and Steph in a minute and having some final questions for our panellists.
So, boom, Faith, we’ve got 80 percent of people saying eight, nine or 10, so that’s a very certain result. 40 percent of people saying 10, one or two people now in the one to five category, we had over 10 percent before. We’ve got five percent now in that category. So people have certainly moved up to the strongly agree that sleep is potentially a helpful target for preventing depression and anxiety in young people. That’s a nice takeaway from today. Thank you.
So, questions, any further questions? Steph and Douglas, do you want to kind of tell us what’s been going on in the chat? Anything else you’d like to put to our panel?
Douglas Badenoch – A lot of really good questions in the chat and a lot of really positive feedback, so thank you to everyone who has taken the trouble to contact us. It’s really helpful for us to try and steer things in the right direction. I think Colin’s been doing a brilliant job at answering the questions as they come up often, which is excellent. So, managing to do that and talk as well. Amazing. I suppose one of the themes is treatment interventions. What can we do? Are there things we can do now to help teens who are experiencing difficulty, perhaps with all three: sleep, anxiety, and depression? Or where do we go to to look for that information? So I’ve posted a link to the ACAMH website, but it would be useful to get the panellists’ recommendations on where one might start looking for that information. We can include that in the pack, of course. And gender is another, gender differences is another thing that’s come up a few times, and a few people asking about specific people with things like autism spectrum disorders. What are the issues with how do we manage sleep there?
Andre Tomlin – Colin, do you want to come in on the general interventions question first?
Prof Colin Espie – On the gender side of things, for most sleep disorders, not for all, but for most sleep disorders, females outnumber males by about 60 to 40 in presentation at clinics. I think that’s partly due to gender factors or sex factors in terms of fluctuations in hormones that affect circulating rates of the brain hormone, melatonin, amongst other things. But it’s also partly to do with a kind of underreporting by men of difficulties that they have. So, I mean, there’s a lot one could go into on the gender issue. On the neurodevelopmental issue, I spent about seven or eight years as a clinical director of Intellectual Disability Services back in the day, back in the late ’80s, early ’90s. And I did quite a lot of research on people with those difficulties and have stayed interested in that area ever since. I think as clinicians, we should be doing the hard stuff, right? We should be trying to figure out how to solve the big problems, big clinical problems, as well as taking the population characteristics as a whole. I think services in general, obviously for people with neurodevelopmental difficulties, are sparse or highly variable. But I think sleep is something that would be welcomed as part of that.
And I think it’s important for a number of reasons. One, there’s often a circadian element to disruption in people with developmental difficulties. Plus, there’s a really important emphasis that needs to be placed, I think, on the need for order and pattern and being able to adapt to new circumstances or change, which may be intrinsically difficult, I think, for some individuals. So I think no doubt a psychological approach is better here. There’s a lot of talk about melatonin as a substance, as a pill to take. I have to say the evidence is pretty slim for that on any guideline reviews. Ways of influencing the brain production of melatonin include exposure to light, but there is a little bit of evidence that pill-format melatonin may be helpful in some developmental disabilities. If people are interested in, you know, digging deeper into some of the points here that I can’t really go into, I’m happy for folks to contact me on Twitter or whatever. I’ll try to direct them towards services. I see that Sleep Scotland are on there, just to put up a shout-out for them. I was delighted to speak at their, was it their 10th or 20th anniversary of the foundation Sleep Scotland, having been involved a little bit at the beginning. It’s a great organisation, taking the importance of sleep into the community, into our schools and into the homes of people with disabilities.
Andre Tomlin – That’s great. Thank you, Colin. Brilliant. I’m going to draw it to a close now because we’ve got a couple of housekeeping things to do just before we finish. There’s loads more questions. So thank you very much for asking those and thanks to Colin for offering to help with those. You’re going to be absolutely drowning on Twitter now, I’m sure. I just want to finish off by obviously thanking everybody for joining us, thanking Nkem and Faith and Colin particularly, but also thanking ACAMH for inviting Mental Elf, myself and Douglas to partner in this and to run these campfires.
We’re going to be doing this on a monthly basis throughout 2021, and we want your views now on future topics. There’s another poll. I hope you’re not getting poll fatigue, but we’d like you to tell us what you think we should be covering with these webinars and would you recommend them to your colleagues? So, again, this is a one to ten where ten is something that you would very much recommend, and one is something that we’re looking at the results here, so I doubt anybody’s going to say one, but go on. And tweet us your ideas. The hashtag is #camhscampfire and, you know, send us the things you would like us to cover with these next webinars. Thank you. Someone’s done a two and a four. Brilliant.
The next webinar is going to be on the 28th of January. It’s going to be looking at voice hearing in adolescence. That’s a paper that’s coming out in the Child and Adolescent Mental Health Journal by Dr Sarah Parry. A really interesting topic. Young people and voice hearing. That’s going to be our next topic. So please do sign up for that now on the ACAMH website if you’re interested. And yeah, get the message out, please, about this. We’ve been banging on about critical appraisal and research for twenty years and there’s been very little interest from anyone to actually sit down and look at papers in this way. But we think it’s a really fundamental part of practise in mental health, if we don’t know how to assess research, and whether it’s any good or not, and whether we should be acting upon it, we’re a bit lost, really.
So, tell your friends, tell your colleagues, get them along, and hopefully we can all learn together and discuss the evidence in a friendly and constructive way. So, thanks very much for joining us, and good night.