In this Papers Podcast, Dimitris Tsomokos discusses his JCPP Advances paper ‘Chronotype and depression in adolescence: Results from a UK birth cohort study’ (https://doi.org/10.1002/jcv2.12245). Dimitris is the first author of the paper.
There is an overview of the paper, methodology, key findings, and implications for practice.
Discussion points include:
- The bidirectional association between sleep duration and sleep quality and depressive symptoms in adolescence.
- The reason behind using the Millennium Cohort Study (MCS), a large, population-based longitudinal birth cohort, in the study.
- The cross-sectional association between chronotype and depressive symptoms and the differences between the sexes.
- Can a ‘sleep catch-up mechanism’ mitigate risk for depression and are adolescence that are in tune with their circadian rhythms at less risk of depression?
- The implications for policymakers and child and adolescent mental health professionals.
- Gender differences and eveningness.
In this series, we speak to authors of papers published in one of ACAMH’s three journals. These are The Journal of Child Psychology and Psychiatry (JCPP); The Child and Adolescent Mental Health (CAMH) journal; and JCPP Advances.
#ListenLearnLike
Subscribe to ACAMH mental health podcasts on your preferred streaming platform. Just search for ACAMH on; SoundCloud, Spotify, CastBox, Deezer, Google Podcasts, Podcastaddict, JioSaavn, Listen notes, Radio Public, and Radio.com (not available in the EU). Plus we are on Apple Podcasts visit the link or click on the icon, or scan the QR code.
Dimitris Tsomokos holds MPhys and PhD degrees in physics and is currently completing a conversion MSc degree in psychology (University of Glasgow). He is a founder-director of Social Cognition Ltd and Alphablocks Nursery School in north London, conducting research in developmental psychology and related disciplines.
Other resources
- Blog ‘Chronotype and Depression in Adolescence’ by Dimitris Tsomokos
Transcript
[00:00:10.000] Jo Carlowe: Hello, welcome to the Papers Podcast series for the Association for Child and Adolescent Mental Health, or ACAMH for short. I’m Jo Carlowe, a Freelance Journalist with a specialism in psychology. In this series, we speak to authors of papers published in one of ACAMH’s three journals. These are the Journal of Child Psychology and Psychiatry, commonly known as JCPP, the Child and Adolescent Mental Health, known as CAMH, and JCPP Advances.
Today, I’m interviewing Dr. Dimitris Tsomokos, Lead Researcher at Alphablocks Research Lab, funded by Alphablocks Nursery School, and Research Affiliate at the Institute of Education at University College London. Dimitris is the First Author of the paper, “Chronotype and Depression in Adolescence: Results from a UK Birth Cohort Study,” recently published in JCPP Advances. This paper will be the focus of today’s podcast. If you’re a fan of our In Conversation series, please subscribe on your preferred streaming platform, let us know how we did, with a rating or review, and do share with friends and colleagues.
Dimitris, thank you for joining me. Can you start with an introduction about who you are and what you do?
[00:01:17.792] Dr. Dimitris Tsomokos: Yes, and thank you very much, indeed, Jo. I’m a Researcher in the CUBIC Lab. That’s led by Professor Eirini Flouri, the Senior Author of this paper, in the Department of Psychology and Human Development at the Institute of Education at UCL. So, I’m currently also completing a MSc conversion degree in psychology from the University of Glasgow, because my research background is actually in physics, in the physics of complex systems. And roughly a decade ago now, I co-founded Alphablocks Nursery School here in North London, where I currently work as a Director. And I also work with Professor Stephen Scott from King’s College London, and other experts, bringing to market a new parenting programme called “Parenting Matters.”
[00:02:04.299] Jo Carlowe: Thank you very much. So, today, we are looking at your JCPP Advances Paper, “Chronotype and Depression in Adolescence: Results from a UK Birth Cohort Study.” Dimitris, can you start with an overview of the paper? What did you look at and why?
[00:02:19.390] Dr. Dimitris Tsomokos: So, we know that there is a bidirectional association between sleep duration and sleep quality and depressive symptoms in adolescence. So, particularly for adolescent populations, we already know that poor sleep, so that’s suboptimal sleep duration, say consistently sleeping under seven hours a night, including weekends, or very disturbed sleep with many night awakenings or taking a long time to fall asleep, so half an hour or more, we know that all of these aspects of poor sleep are associated with depressive symptoms. Whether that be subclinical, sort of, an adolescent depression or above the clinical threshold.
Now, for 14/15-year-olds, for example, having more signs of poor sleep has been linked with having more symptoms of depression. So, this is true for adult populations, as well, but in the case of adults, depressive symptoms are also predicted by a later, or so-called evening chronotype. In this paper, what we have focused on is we have examined the association between a later chronotype and depressive symptoms in middle adolescence, so at age 14 years. That is independently of poor sleep and any prior mental health difficulties.
[00:03:44.388] Jo Carlowe: Okay, thank you, and can you tell us a little about the methodology that you used for the study?
[00:03:49.638] Dr. Dimitris Tsomokos: Yes. So, we used the Millennium Cohort Study. That was our secondary dataset, and the reason for using the Millennium Cohort Study was that we wanted to investigate these associations between chronotype and depressive symptoms in middle adolescence, but doing so within a ecological model. So, accounting for many other things that can affect teenagers’ sleep. So, from socioeconomic conditions to their neighbourhood sense of safety and air pollution, which is a proxy for traffic, as well, to other important aspects that are sometimes overlooked. So, for example, whether they moved schools or if they drink alcohol, they have a poor diet or don’t exercise. Body mass index and all these other factors.
So, the study was primarily cross-sectional, using the age 14 wave of the survey, but I say primarily cross-sectional, because, of course, we did take into account prior mental health. So, we controlled for the participants’ total internalising and externalising problems at age 11-years, which was the previous survey wave. And yeah, the Millennium Cohort Study affords us with these incredible opportunities, really, because it tracks the lives of something like 19,000, just over 19,000 families. Of course, there’s attrition and through the survey, waves, but still the final sample included more than 11,300 teenagers. Which ensures adequate power, of course, and also, the scales and the different perspectives from which you can approach these associations.
So, for example, we used – we measured depressive symptoms through the Mood and Feelings Questionnaire. So, that was self-report by the teen – by the adolescents. It’s a very well validated scale. It has 13 items, things like, “I felt miserable or unhappy,” or “I thought I could never be as good as other kids.” So, those 13 items corresponded to 13, sort of, major symptoms. And finally, as a proxy of chronotype, we used a few measures, actually, including the well-established measure of the midpoint of sleep on non-school nights, so nights that the next day, you don’t have to go to school. But we also used a much simpler measure of the approximate time of falling asleep, within an hour slot, on school nights.
[00:06:19.354] Jo Carlowe: Right, and what did you find? What were the findings that you would like to highlight?
[00:06:25.217] Dr. Dimitris Tsomokos: Yes, so the main finding is a very robust cross-sectional association between chronotype and depressive symptoms at approximately age 14. Using the sleep midmor – midpoint on non-school nights, we find that every one to two hours of later sleep time, so more eveningness, there is an additional symptom of depression, or a higher intensity point in an existing symptom. So, that’s – that is after accounting for all that – arrange all those factors that I mentioned earlier and including prior mental health, poor sleep and low sleep quality and shorter sleep duration. So, after all those, there’s still that clear association.
The second aspect was that this association was moderated by biological sex, and females were actually more susceptible to a shift towards an evening chronotype. So, it’s important to remember that after puberty, between early and middle adolescence, there is a natural shift to later chronotype. So, all teenagers, sort of, naturally shift towards sleeping later, but the association we find is that this shift towards eveningness seems to affect teenage girls more than boys, in its link with depressive symptoms.
[00:07:52.787] Jo Carlowe: Right. Can we just dig into that a little bit? What – how do you interpret that association, so, both the positive association with depressive symptoms and this evening chronotype and also, that this was particularly the case for females?
[00:08:08.295] Dr. Dimitris Tsomokos: So, as I mentioned – so, let’s start, first of all, actually, with the chronotype and the definition. So, we have a natural preference based on circadian rhythms, an internal clock that tells us when our body prefers to go to sleep. So, if that is earlier rather than later, we refer to it as morning chronotype and later, as in evening chronotype. And the difference between them is roughly two to three hours, on average, between the morning types and the evening types. And when I referred earlier to the ‘midpoint’ of sleep and why that’s an accurate measure, it’s well, because if you don’t have school, or for adults, work, the next day, then you align more to your natural rhythm. Of course, you do catch up on sleeping on the weekends and there could be social jetlag, as well, so catching up with other things you want to do, particularly socially, for adolescents.
So, it’s not a clear-cut measure by any means, but nevertheless, there is – unless you measure it with actigraphy and very, very specific measures in the lab, in the wild, let’s say, in everyday life, the simple measure is you take the time you tend to fall asleep, say, in the – on a Saturday or a Friday, and then the time you wake up in the morning. So, if, for example, you sleep at 12 – midnight and you wake up at eight in the morning, then the midpoint is 4:00am. There is two to three hours of difference between the morning chronotypes, where the midpoint of sleep is maybe one to 2:00am and the evening chronotypes, that is maybe four or five.
[00:09:46.348] Jo Carlowe: It’s a lot of difference, isn’t it?
[00:09:49.043] Dr. Dimitris Tsomokos: It is, and it does mean that there are physiological differences between the two, because, of course, that midpoint is related to so many things, from REM deep sleep to temperature in the body and so on. So, essentially, what we found is that the preference to go to sleep later in the evening, say, for example, after 11:00pm, compared to going earlier, say nine to 10:00pm, is associated with the likelihood of more depressive symptoms in middle adolescence.
Now, that isn’t saying anything about causal direction. It’s simply an observation of the existence of an association of a correlation based on this large representative dataset, which incidentally, this birth cohort, who were born between the end of 2000 to early 2002, they’re actually in the middle of what is normally called Gen Z.
We did take into account, as well, of course, things like screentime, because these matter, as well. So, we control for those because there is also currently quite a bit of a conversation going on in the public sphere around smartphones, social media use and screens and how much this generation has been affected, and our paper has results on that front, as well, around screentime. But again, as I said, even if you account for all of those extra factors, there still remains that association.
So, the interpretation is quite complex, causally. We can’t make any causal statements there, but I do have a couple of comments that are, as I said, more of a speculation at the moment. They’re not supported by the current data, but we can speculate, perhaps, that the association is due to a confounder, so things that we haven’t been able to account for. So, a cause outside of our observations, which causes both an evening chronotype, a later chronotype and depressive symptoms.
Such a cause could be, for example, a genetic predisposition to both later sleep times and internalising problems, for example, and that speculation comes from recent studies. So, for example, a paper published in Translational Psychiatry last year, by Odessa Hamilton and others, they used the UK Biobank in adults and uncovered a common polygenic predisposition to both short sleep and depression. So, this is different from chronotype, of course, but I mention it by way of showing that genetic predispositions could play a role.
Another way to look at it is that there is actually a bidirectional association in which one feeds the other. So…
[00:12:31.880] Jo Carlowe: Hmmm, I wondered that.
[00:12:32.976] Dr. Dimitris Tsomokos: …could be – right, so it could be the – indeed, that depressive symptoms, even at the subclinical level, sometime before or during the change to a later chronotype, can make that change even more dramatic. So, whereas a 12/13/14-year-old would’ve slept an hour later under ideal circumstances, they actually sleep two or three hours later, because of prior mental health. Again, as I said, we accounted for that, but it’s not possible from this study to be able to untangle that.
[00:13:02.409] Jo Carlowe: It’s so complicated. Did you account for the pandemic and lockdown in the…?
[00:13:08.048] Dr. Dimitris Tsomokos: Right. So, I – as I mentioned, this is – this birth cohort falls in the Generation Z and they – being born 2000/2002, means that the timeframe of the study, when they were 14 and the interviews and neuropsychological assessments and the questionnaires were completed, that was before the pandemics. And in that sense, the benefit of that is that it shows us how things were, let’s say before the impact of the pandemic.
[00:13:40.339] Jo Carlowe: In your paper, you found that sleeping late on non-school nights predicted fewer depressive symptoms. I was wondering, does this suggest that there is some kind of sleep catch-up mechanism that can mitigate risk, or that when adolescents sleep, their sleep patterns are in tune with their circadian rhythms, so they get up when they like, go to bed when they like, that they are at less risk of depression?
[00:14:07.451] Dr. Dimitris Tsomokos: Hmmm, yes. So, that’s a very interesting interpretation on both counts, and again, we’re obviously speculating. The data doesn’t quite support one way or another, but our interpretation, indeed, has been that when teenagers align in the weekend more with their natural rhythm, which is to go to bed later, then that may be acting as a protective factor. And indeed, as you said, of course, there is the element of catching up, as well, on sleep over the weekends. So, that was a – well, in retrospect, it can make sense in that way, but a surprising finding that the teenagers who were able to sleep later and wake up later in the week – in weekends, were actually experiencing fewer symptoms.
[00:14:58.736] Jo Carlowe: So, given what you’ve said about this possible protective factor in following your natural sleep patterns, what are the implications of that? Does that mean, for example, that policymakers should think about changing school start times?
[00:15:15.783] Dr. Dimitris Tsomokos: Hmmm, well, in practice, sleep is not, obviously, a unidimensional factor. It’s a complex, very complex aspect of our lives and for adolescents’ lives, in particular. So, it’s really crucial to consider chronotype alongside sleep duration, sleep quality and many other factors. So, on the basis of our study, it appears that asking, simply asking adolescents what time they fall asleep on a typical school night is a valid proxy of their chronotype or of their tendency to sleep earlier or later. And on the basis of the – of our results, I don’t think it is straightforward to make policy recommendations. However, this very fact that there is an association with depressive symptoms, it is quite valid to capture a teenager’s preference by simply asking, “What time do you normally go to bed?” There’s a difference between say, nine to ten, as I said, and midnight to 1:00am.
And given those two things put together, that, sort of, again, hard recommendations that are difficult, however, there is a clear effect and association, to me, that says, let’s, sort of, go easy on adolescents that may seem very tired in the morning or very – I have two daughters, one of them is 14-years-old, and there are Teachers that, as I said, go easy on them. There are other Teachers who consider it, whatever it may be, from, you know, being disinterested in class, to laziness, to whatever. Of course, there are all these other factors that may play a role, but indeed, sleep is an important one. So, bearing that in mind, I would say in schools and for Teachers, it would probably be a good thing.
So, assuming that their sleep hygiene is good overall, sleeping later rather than earlier seems to be associated with an increased risk of depressive symptoms. And that’s a message in the educational setting, so I think it’s important to bear in mind.
[00:17:27.268] Jo Carlowe: What are the implications of your findings for CAMHS professionals?
[00:17:31.937] Dr. Dimitris Tsomokos: I would say that, again, it is important to, of course, to take into account all the other factors. I think CAMHS professionals and Clinicians know this very well already, in practice from their work, that it’s a very complex and multifaceted factor. But I think the take home message, perhaps, for the professionals and the Therapist side, is that all other things being equal, so sleep hygiene is good and diet and exercise are okay and, you know, the use of social media and screens and all these other aspects, if those are along the right lines, then, in that case, it is quite protective, let’s say, if an adolescent is able to sleep according to their natural rhythms. So, if, for example, Monday to Friday, that’s not possible because your chronotype is an evening one, you do tend to go after 11:00 and midnight, then at least, have enough, sort of, freedom and self-care for the weekend catch-up and to align with that chronotype in the weekend.
[00:18:39.515] Jo Carlowe: Yeah, so perhaps it’s a message for parents to take that onboard. Let’s go – returning to the paper, what – is there anything else that you would like to highlight?
[00:18:49.239] Dr. Dimitris Tsomokos: Perhaps to talk about that wide range of covariates and confounders that we use. So, all these other factors that we took into consideration to make sure that they didn’t confound that relationship that we found. So, we did that very carefully, so entering them in a, essentially, multiple regression model, in a careful way so that it didn’t introduce bias. And what we found there is that there are certain factors that are known. As I said, screentime, for example, impacting sleep and sleep duration by staying up later. And I believe it could be instructive for those interested to look into that detail, as well, and all these various factors, either as a risk or as a protective factor.
[00:19:38.388] Jo Carlowe: I suppose it might be worth returning to that difference you saw in gender, with eveningness being strongly related to depressive symptoms in females. Any thoughts as to why it particularly affects females?
[00:19:54.133] Dr. Dimitris Tsomokos: There is a gender difference in middle adolescence and internalising problems more broadly. So, it does seem to be the case that at round age 14, teenage girls are more – the prevalence of depressive symptoms is higher, anyway. So, that is partly the reason.
[00:20:16.789] Jo Carlowe: Are you planning any follow-up research, or is there anything else in the pipeline that you would like to share with us?
[00:20:22.817] Dr. Dimitris Tsomokos: Yes, I think what I mention around the impact of social media use and screentime, in a longitudinal sense, though, not cross-sectionally, that is something that I would like to pursue further. And again, the Millennium Cohort Study lends itself for this sort of analysis. Preliminary results, as I’ve already started on this, show that in fact, social media use at age 11, so early adolescence, impacts sleep at age 14. Which then impacts longitudinally, three years later, at age 17, a range of outcomes, both psychological distress and psychiatric outcomes, such as self-harm, suicidality and clinical diagnosis of depression or anxiety. So, there does seem to be an influence of social media use and screentime in early adolescence on sleep patterns in middle adolescence, that then carries forward into these adverse outcomes later in adolescence.
[00:21:27.940] Jo Carlowe: Interesting area. Finally, Dimitris, what is your take home message for our listeners?
[00:21:34.057] Dr. Dimitris Tsomokos: Adolescents, especially the age that we studied, so around 14-years-old, with an evening chronotype, a later chronotype, report on average, a higher number of depressive symptoms, in the way that was measured in the Millennium Cohort Study through the Mood and Feelings Questionnaire, at least, which is a broad scale. And eveningness was more strongly associated with depressive symptoms in females.
And the simplest way, perhaps, of measuring this preference, this chronotype preference, for this age group, may be asking teenage boys and girls what time they typically go to bed and fall asleep on a school night. So, simply asking them for the crude hour slot, “Is it nine to ten, ten to 11:00?” and that already encapsulates much of this preference, it appears. So, it is very correlated with the midpoint of sleep on non-school nights, which is encouraging, because it means that it’s a very simple way of understanding their chronotype.
Catching up with sleep and aligning with the body’s natural rhythm over the weekend seems to be the protective factor against depressive symptoms at that age group. And finally, yes, remembering that potentially, teenage girls are more susceptible to late sleeping, especially when that comes along with poor sleep.
[00:23:00.373] Jo Carlowe: Dimitris, thank you ever so much. For more details on Dr. Dimitris Tsomokos, please visit the ACAMH website, www.acamh.org, and Twitter @acamh. ACAMH is spelt A-C-A-M-H, and don’t forget to follow us on your preferred streaming platform, let us know if you enjoy the podcast, with a rating or review, and do share with friends and colleagues.
