How did lockdown impact children?
41:54 Share

This BBC podcast is supported by ads outside the UK.

Residents at Brightview Senior Living Communities enjoy enhanced possibilities, independence, and choice. Brightview Dulles Corner in Herndon and Brightview Great Falls offer vibrant senior independent living, assisted living, and memory care services through various daily programs and cultural events.

Chef-prepared meals, safety and security, transportation, resort-style amenities, and high-quality care. Everything you need is here. Discover more at brightviewseniorliving.com. Equal housing opportunity. BBC Sounds. Music, radio, podcasts.

Hello and welcome to a special edition of More or Less with me, Tim Harford. It's five years almost to the day that the Prime Minister announced that UK schools would be closed to prevent the spread of the COVID-19 virus. We're going to devote this extended episode to explaining what we know and what we don't about the effect of the COVID lockdowns on young people.

In the early stages of the first lockdown, it wasn't entirely clear how young people themselves were going to be hit by the virus. But it turned out that they would be spared from the worst impacts of the disease. Here's Alistair Munro, a doctor and lecturer in paediatric infectious diseases at the University of Southampton.

COVID itself actually had a relatively small clinical impact on young people. In terms of what we saw from respiratory infections, it wasn't anything that would seem unusual compared to normal respiratory viral infections in children.

For a few young people, Covid was a very serious disease. Long Covid has also been a problem for some. And young people did die. But almost always children with pre-existing conditions that made them very vulnerable. The total number of children who died was in the tens, not the tens of thousands.

So in general terms? COVID wasn't a particularly deadly pandemic for children. I think if adults hadn't been affected, it would have been quite easy for the COVID pandemic to have come and gone. And actually, most child health providers probably wouldn't really have noticed it. It's kind of an intergenerational transfer of harms in many ways. It was in order to protect children.

the more elderly and more vulnerable people from the infections. Unfortunately, we did have to implement quite a substantial number of restrictions on children and young people that for the most part were not in their benefit. Of course, children and young people didn't want to see parents and grandparents laid low by Covid-19.

But they paid a price for helping to stop the spread of the disease. They missed out on experiences that other generations would have considered essential rites of passage. Birthday parties, school plays, freshers' week and the first day in the office. Not to mention learning in a classroom, working face-to-face and hanging out with friends.

But for all the obvious strangeness and hardship of the time, have those restrictions had a lasting impact on the lives of young people? Is there a whole generation whose life prospects have been pushed backwards by the decision to lock down society? Five years on, as part of Radio 4's special coverage of the lockdown, it's time to take a long, hard look and see what we can figure out about the effect the lockdowns actually had on this part of society.

We're going to tell this story from the youngest kids up to people in their mid-twenties. We're going to focus here on the strongest evidence we can find, but there's a lot that's definitely missing from the picture. We're also going to focus on England, for the simple reason that it's where we found the best data.

Let's start with preschoolers. How did they fare during lockdowns? They didn't have the typical formative life experiences like playing in parks, going to shops, seeing family, seeing friends. They were very much confined to their homes with very limited ability to get out and about. That's Professor Claudine Bowyer-Crane from Sheffield University, who's studied this group during and since the lockdown.

their experience would greatly depend on who they were confined to their homes with and how those people were coping with the additional pressures of the pandemic. But whatever the circumstances, it seems likely that preschool children would just develop differently when confined to barracks for months. If you consider that children learned through social interaction...

and that social interaction was limited, then a reasonable assumption would be that children's early development may be delayed, particularly in areas like language and socio-emotional well-being. Children weren't able to develop their language skills in the way that they typically would, and also there weren't the opportunities for developing things like self-regulation.

It's hard to learn how to share when there's no one to share with. Now, we'd love to have a study that shows us one way or another whether these developmental changes actually happened for babies born in the lockdown. There is a study looking into that called the Bicycle Study, but it hasn't reported yet.

Claudine and colleagues looked at slightly older children, those who started reception classes in school in the autumn after the first lockdown, aged four or five. And for those children, we found a drop in the proportion of children achieving expected levels of development at the end of that year, particularly in language and communication and in literacy classes.

and maths. The metric they used was something called the Early Years Foundation Stage Profile, or EYFSP, a teacher assessment that every child had to go through before the lockdown. Pre-pandemic, 71% of four and five-year-olds met the expected standard. While these assessments were not compulsory during the lockdown, some schools still kept them going and shared their data with the researchers.

This showed that for the lockdown kids in the participating schools, the number meeting the standard fell from 71% to 59%. Claudine also helped run a separate study using a different methodology to try to establish what was happening more clearly. So we have data on a smaller sample of around 1,500 children across 69 schools across the country.

And we looked at both language and socio-emotional well-being in those children. And in that sample, we found little evidence of children having increased problems with language or increased problems with socio-emotional well-being. Ah, does this mean the EYFSP teacher assessments were wrong?

Not necessarily. It might just be that having the kind of parents who put their children up for scientific surveys during a lockdown skewed the sample in Claudine's study. It might be that the two different assessments were testing something slightly different. What both sets of data show is consistency in risk factors.

So while most of the children in our study were performing as expected, there were still risk factors such as deprivation. So what the research suggests is that rather than a clear negative impact in and of itself...

the lockdown was an amplifier of existing factors. And this reflects what teachers have been telling Claudine in her research. What teachers are telling us is that there's more variation in the classroom in terms of the amount of support that children need, not just for learning, but also for things like developing resilience, being able to focus for a period of time, developing those attention skills, etc.,

So there's likely a group of young children who missed out on some key development because of the lockdown, and many who are basically fine. If you want to know more about this group of young children, then Panorama are looking at this group in a programme called Lockdown Kids Five Years On, and you can find that on the iPlayer now.

Let's move on from reception and move up the age groups. Just to remind you of the timeline, schools closed for most children in the first lockdown in March 2020. Some year groups came back towards the end of June. Then schools were mostly open in autumn…

staying open in the second circuit breaker lockdown, but with various restrictions in school, depending on which part of the country you were in. It all depended on the tier, remember that? Including bubbles, masks and lots of reasons that chunks of school would be missed. There wasn't one universal pupil experience. There were many, even between year groups at the same school.

Then in January 2021, schools opened for one day, allowing a little bit of mixing and breathing on each other before a third lockdown was announced and the schools were closed for most children again. And then schools opened up for good in March 2021. Covid was still disrupting classes, but school life was closer to normal.

The most obvious place to start with trying to understand the effect of all that is with attainment, with how much children actually learn. And just as with preschoolers, when the lockdown kicked in, people were worried. What was the word everyone kept using at that time? Unprecedented. So I don't think anyone really knew what to expect today, but it just felt intuitively like it can't be a great thing.

for children not to be at school. This is Daisy Christodoulou, the Director of Education at a private testing company called No More Marking. More about them in a second. Intuitively, it felt like kids' education might suffer. But did it? Well, the problem, again, is pretty obvious. The data goes AWOL. Exams in the summer of 2020 didn't happen. And you had huge controversies about different ways of awarding students' grades.

GCSEs and A-level exams were cancelled in 2020 and in 2021, and the replacement assessments just don't let us answer the question of whether the lockdowns prevented students from learning. The other tests, by which you'd get a reliable picture of how pupils were performing, were also stopped. Yeah, we have quite a few blanks in our normal assessment data.

This is where Daisy's company, No More Marking, comes in. They run writing assessments for schools across England where the marking system is basically consistent over time. So you can see individual progress and compare marks between students. And as it happens, one of their assessments came at the perfect time for our purposes here. We assessed a large group of Year 6 students just before the first lockdown in March 2020.

And then we assessed a really large group of Year 7 students in September 2020. Year 6, by the way, is the last year of primary school. Year 7 is the first year of secondary. Now, these weren't the exact same kids, but they were in the same cohort. And Daisy's company administered tests to tens of thousands of pupils in these age groups.

The change in the median marks is as good as we've got as a measure for the lockdown effect. We found that there was a really big decline in their attainment and in their scores. Define really big. So the year sixes scored 550 on our writing scale. The year sevens scored just 533. And that's about where we'd expect year fives to be in November.

Right. So our year sevens were 22 months behind where we expected them to be in the September of year seven. Or very roughly, in about six months, they went backwards about 18 months. Yeah, another way of looking at it. We should say this level of learning loss, as it came to be known, was pretty extreme. In other areas, such as maths and reading, it seems fewer months were lost.

But it does look like that intuition, that lockdowns harmed education, has some more data behind it. However, we're not just looking at the immediate effect in this programme, we're looking at the lasting impact. So what happened next? When we published these results at the time, we did get a bit of pushback, people saying this is an enormous fall, can it really be right? And I think we stood by it then and we stand by it now. But one of the things we said at the time is we felt that this was something that students could recover from.

So just as people were predicting a V-shaped recovery in the economy, Daisy was predicting one in education too. The students could bounce back. They could end up actually even by the end of year seven, year eight, year nine, being actually back to where you might expect them to be. So that's what you predicted, but presumably we now know whether you were right or not. Absolutely. I wouldn't be saying it like this if I wasn't, would I? I suppose not. Yeah, we were right. So yes, by the end of year nine...

which was June 2023, those students were roughly back to where we might expect them to be. This basic picture seems to be broadly backed up by another set of tests, this time something called the National Reference Test. These examine a sample of kids in the spring before GCSE exams, and they're used by the Department for Education as a kind of survey to measure overall pupil performance, and they, again, are comparable over time.

What these tests show is that in English, there's been a slight decline at lower exam grades since the lockdown, but higher exam grades haven't really moved. And for maths? Really not very much has changed. You look at the stats, they've got these graphs and the line from 2017 to 2024 is kind of really just flat.

So from these data points, it looks like the lockdown effect may not have led to big, permanent damage on education levels. It doesn't look terrible. It doesn't look amazingly brilliant. You could argue, oh, they should be doing better than they are, but it looks pretty steady. It doesn't look like there's a huge, apart from that initial dip, there's not some huge disaster.

That's not to say that flat is a great result. Perhaps things would have improved if it hadn't been for the lockdown. But still, there has been no collapse in learning. However, it is time for some more caveats. The first thing to say is that this was what Daisy found for her tests on writing. Other private test providers did a similar thing for other subjects. And while they found there was a recovery, it didn't recover all the way.

John Andrews is head of analysis at the Education Policy Institute and he's been looking at other tests. When we look in maths, there was a fall that has been recovered to a certain extent, but we're still below where we would have been pre-pandemic.

What's more, the recovery was not distributed evenly across England. The Education Policy Institute also looked into regional variation. Those did tend to tie up with where there had been particular issues with school closure and lower attendance overall. So we saw greater learning losses in parts of the north and in the Midlands as well. So where kids missed more school, the recovery was slower, which certainly makes sense.

The second thing to say is that the relatively small changes in the average results might mask a bigger divergence in performance, because there was certainly a huge divergence in the pupils' experience of lockdown. That amplifier again.

Dr Rebecca Montague is the Acting Director of Research and Policy at the Sutton Trust, a charity which aims to improve social mobility. Depending on the income level of their family, children had very different experiences of the pandemic.

On one end, there were children who on day one had their own computer to use at home in a quiet room without distractions. On the other end of the spectrum, there were children who even far later in the pandemic were never provided with a laptop, who were perhaps completing their schoolwork on a phone shared between several siblings in overcrowded housing without access to a quiet space to work.

Rebecca collaborated with University College London on a big longitudinal study called the COVID Social Mobility and Opportunity Study, or COSMO, which followed thousands of young people over the course of the pandemic. They were 14 or 15 at the time of the first lockdown, and their different experiences led to very different amounts of learning when the schools were shut.

We know from that data that access to a device was absolutely crucial because those young people without a device reported working on average just eight hours a week in lockdown one. Those were just a mobile device, 10 hours, and those with a laptop or a tablet, 14 hours. So the amount of hours of schoolwork that children were completing was very closely associated to the kind of device that they had access to.

So there's good evidence that lower-income kids might have spent less time learning. But do we know that that led to worse results? We do when it comes to the gap between lower-income students and their better-off peers and how they have performed over time. So before the pandemic, what we call the attainment gap, they

The gap in how well young people from lower income backgrounds do in school compared to their better off peers had been narrowing, albeit very slowly. But we know that from results post-pandemic, there's been a reversal of that trend and a decade of progress in closing the gap has been wiped out.

However, this attainment gap analysis is a bit messy to say the least, which is why we're not going to give you any precise numbers. It's calculated using the number of kids who are eligible for free school meals, and this number has seen big increases because of changes in the benefits system. So it might partly be tracking population changes, not attainment changes.

Here's John again from the Education Policy Institute. We've actually done additional analysis that's looked at pupils that will move in and out of disadvantage. So we've attempted to control for these changes that have happened in the benefit system. And at the moment, we can be reasonably confident that that gap has widened. The attainment gap is also a relative measure. So it can be driven by either lower-income kids doing worse or higher-income kids doing better or both.

it's clear that there are reasons to think both of these things might have happened during the lockdown. And there certainly is some evidence that it's a widening distribution rather than just everyone shifting down to lower attainment. So when we look at the top decile of attainment and look at their reading scores, they've actually gained a bit. So their attainment in reading is higher than similar pupils pre-pandemic. And at the bottom of the distribution, the attainment's lower.

Empathy is our best policy.

This is Ashley Akedene from the Ben and Ashley I Almost Famous podcast. If you could lose 10.4 pounds in one month, would you try? Well, with Future Health, you can. Find out if weight loss meds are right for you in just three minutes at tryfh.com. That is tryfh.com. Tryfh.com.

Results may vary based on start weight and adherence to diet, exercise and program goals. Data based on independent studies sponsored by Future Health. Future Health is not a health care services provider. Meds are prescribed at provider's discretion. There's one area where we can clearly see a longer lasting impact. Attendance. Dave Thompson is the chief statistician at the FFT Education Data Lab.

Using the measure of persistent absence, which is being absent for at least 10% of the time that you're supposed to be in school, then up until 2019...

Persistent absence was fairly flat at around 10.5%. So that's just over 10% of children missing 10% or more of their lessons. Since the pandemic, it shot up. And in 2022, 2023, which is the most recent year for which national data has been published...

The persistent absence rate was 21.2% of pupils, so it had doubled since 2019. Then there's another measure of absence, severe absence, where a child is missing half of their lessons. In the autumn and spring term for 23-24, 3.5% of secondary age pupils died.

were severely absent, which doesn't sound a lot, but it's three times as many as in 2019. These numbers are high and it's pretty unclear what's driving them.

There might be a change in how we treat sickness and it might be related to mental health. There is a particular rise in kids with EHCP plans, which indicate that they have special educational needs. It's also possible that the idea that kids have to go to school every day has shifted, in the same way that the idea that you go to work has changed. School attendance is much worse on Fridays, just like going to the office is at its lowest on Friday.

Again, as we've seen elsewhere, kids in lower-income families are driving the trend. Disadvantaged pupils are three times as likely as other pupils to be persistently absent. However you cut it, five years on, school attendance seems to be one of the data points that does show a lasting impact of the lockdowns. And there's one last thing that seems to have changed.

Behaviour. Anecdotally, teachers talk about a decline in behaviour, and you can see that in the stats. The only two metrics we have that cover the entire school system are measures of permanent exclusion and suspension.

Permanent exclusions went down in the lockdowns and have edged up since to pre-pandemic levels. But suspensions have nearly doubled. Suspensions were at the highest recorded level in 22-23. So there was a rate of nine suspensions for every 100 pupils in 22-23. And that is much higher than what we were seeing back in 2019 when there were just over five suspensions per 100 pupils.

We don't have good research that shows why the behaviour in this group has declined. Mental health might be a factor here, and girls appear to be a concern. They're still a smaller fraction of suspensions, but they're growing as a proportion. As this behaviour works through the system, it is possible the rates will come down again. But this group of young people is certainly one to watch closely as they head out into the world. MUSIC

We're going to keep on growing up, and now we've reached university. When the first lockdown began, universities went online. With the clubs and pubs closed and social distancing in place, the student experience certainly changed. Perhaps that might have meant more time and less hangover to learn things. But while we've struggled to understand learning loss at school because of a lack of data, at universities, that lack becomes ridiculous.

Take what happened to the exam results. It's actually incredibly difficult to spot the extent to which there was real learning loss. And that's partly because what universities did in that first summer, and to some extent in the second summer, was allow students to have multiple attempts or to take longer in order to hit the required standard. This

This is Jim Dickinson, Associate Editor at Wonky, which is a higher education news website. Jim told me that these no-detriment policies were designed to make it fairer for the students whose lockdown life makes learning tougher than pre-pandemic. So, who move back to crowded family homes with other kids running around the kitchen.

But the students who had nice, calm circumstances to learn in with fewer distractions also got to have lots of goes or extra time. So overall, when you add up those two impacts, what actually was seen was significant grade inflation. More students got a first, more students got a 2-1 and so on. In terms of comparing outcomes with pre-pandemic, the results are useless.

And what happens next makes the post-lockdown results useless for a comparison too.

The regulator, the Office for Students, partly at the behest of government, was very concerned with the level of grade inflation and so has asked universities to adjust some of their criteria in order to get the number of firsts and two ones under control. So it's really, really hard to disaggregate that policy intervention from the overall assessment of whether or not students are kind of doing better or worse.

Perhaps then we can use data on the actual learning experience of students.

Did students do less learning? Who was collecting that data? No one, and they didn't before, they didn't during, and they don't now. Nobody collects the statistics on the number of hours that students spend on campus or the number of taught hours in a timetable that are experienced online or in person. Oh, so nothing there.

Now, the Office for National Statistics did start collecting specific survey data on the mental health of students during the pandemic. And this, as you might expect, showed pretty high levels of loneliness and anxiety during the lockdowns. Afterwards, the rates seemed to improve a little and then they got worse again. And there is a clear potential reason for that.

The maintenance loan hasn't been keeping up and certainly wasn't keeping up with the very high levels of inflation we saw a couple of three years ago. And what that means is that students are more likely now to have a part-time job and crucially are working significantly longer hours in those part-time jobs. So students might still be lonely because of a change in the university experience caused by lockdown, or it might be because they can't hang out with people because they have to work more.

It's really hard to disaggregate the long-term impacts of the pandemic and the bounce back from it from the wider cost of living pressures that everyone's been under. In general terms, the answer to our present question in regard to students is rather perplexing. The broad answer, I'm afraid, is we don't know.

The last group we're going to look at is those young people who left education and started work during the lockdown. Here, just as with education, there was plausible reason to think the effect would be bad.

Xiaowei Shi is a senior researcher at the Institute for Fiscal Studies, the IFS, who's been analysing the data. There's a long-established literature within economics on the scarring effect of recessions, which basically shows that young people who graduate during a recession tend to have poorer labour market outcomes, not just during the recession itself, but for years to come afterwards. As the lockdowns closed down large parts of the economy, the fear was that this scarring effect would happen in this case too.

However, there was also reason to think it might not. COVID was a very strange recession in that right after restrictions lifted, we had this huge jobs boom. So reasons for optimism, reasons for concern. So when you looked at the data, looking at young people who were starting work during the lockdown or shortly afterwards, what did that data show?

In short, it showed that there wasn't much effect on the labour market prospects of those who enter the labour market in 2020 and 2021, apart from in the very thick of the pandemic. So three to six months after graduation, only 58% of them were in work, which was nine percentage points lower than the previous cohort. However, we saw that they made a complete recovery just half a year later. So by the time we looked at six to 12 months employment rates,

They look no different from previous cohorts. So we're looking at people who are graduating either from school or from university in the summer of 2020. And what you're finding is they had a bad autumn, but by 2021, things were looking fine. Exactly. So their employment rates had completely recovered to what we would have expected at that point of their careers.

This isn't to say that their job prospects were fantastic, of course. It was pretty tough entering the job market as a young person before the pandemic. But the data from just after the lockdowns shows it was a similar kind of pretty tough.

So what about the next cohorts to enter the jobs market? Those who started work in 2022 or later? I'm afraid we don't know much because of issues with the data after 2023 that makes this hard to study. Ah yes, the issues with the post-2023 data. This is a question that comes back to the slow collapse of the Labour Force Survey, which is run by the Office for National Statistics –

We're going to look into that on our next regular episode of More or Less, so we're going to leave that hanging here. Bottom line, we don't know what's happened to the employment prospects of young people after 2023. There's reason to think that perhaps...

Things might be worse for the cohorts that graduated after them, right? So these are the cohorts that really suffered from huge disruptions to their education during the pandemic and then entered a more challenging labour market where you don't have this huge post-pandemic job boom. They might have suffered, they might not. Because we don't have the data, we cannot at this point actually tell you what is going on.

You're listening to a more or less special programme on the impact of the Covid lockdown on young people. It's part of a wide range of Radio 4 coverage, with our friends on Woman's Hour and Start the Week looking at this from different directions, as well as a series of documentaries hearing the experiences of the children themselves. Find it all, as ever, on BBC Sounds.

We've made it to an issue that's been sitting in the background of many of these other themes, mental health. However, in terms of the clarity of what the data can show us, there is something going on that you need to know about first. The demise of the U-shaped curve.

The U-shaped curve in well-being is really one of the most established facts in social science, I think. It's something we've seen across countries over time. So it's the idea that well-being or mental health or various measures start off quite high when people are young. They fall as people enter middle age and then they start rising again.

as people get into their 50s and 60s. So the midlife crisis is real? Precisely, the midlife crisis is real, possibly up until now. What we've seen over the past several years, not just over the pandemic,

is a really sharp increase in mental health problems among young people. So a big decline in the mental health of young people, which now means that the mental health of young people, people in their 20s, is worse than the mental health of those in their 30s and 40s. So this is not a lockdown trend. According to the survey data Xiaowei is using, the mental health of young people has been declining for years.

There's no longer a U-shaped curve. It's more like a steady increase in wellbeing that starts low and improves through your life. This is the context in which the lockdown takes place. And when it did, perhaps unsurprisingly, by this wellbeing measure, the mental health of young people declined again.

It's completely undoubtable that COVID did affect the mental health of young people. We saw a huge deterioration of mental health during the COVID lockdowns themselves.

However, young people seem to have bounced back from lockdowns very quickly. So they seem to have recovered back to the pre-pandemic trend. Right. So we've recovered to the previous terrible trend. Exactly. But it doesn't seem to have permanently shifted the trend, if that makes sense. I mean, obviously, we're only a couple of years from the lockdown, but it doesn't seem to have changed the trend. But the trend is down. Yes, the trend is very bad, especially for young women.

Now, we don't have the time here to get properly into the reasons for this long-term decline in the mental well-being of young people. There are lots of potential causes. Screen time and phone use, economic stagnation and concern about jobs and money, pessimism about the future, climate change. These are some of the prime suspects. The lockdown potentially amplified all of them.

However, there's another bit of evidence that seriously complicates that idea of a mental health bounce-back. And it starts, once again, with another survey. The mental health of children and young people, so it's affectionately known as MHCYP. It took place in 2017.

The important factors about it was that it was very carefully selected to be representative of the English population of children aged between 2 and 19. This is Professor Tamsin Ford, head of the Department of Psychiatry at the University of Cambridge. The MHCYP is run by the NHS, along with a load of other organisations, including Cambridge University, where Tamsin was involved.

The questions are more comprehensive than those in the surveys Xiaowei was using, and you can use them and some clever analysis to work out how many of the young people who took the survey are likely to be suffering from diagnosable mental health conditions, clinical depression, eating disorders, autism spectrum disorders, and so on.

They'd last run one of these MHCYP surveys in 2017. And after COVID hit, they decided to run follow-up surveys to measure how the lockdown was affecting young people's mental health.

The first follow-up survey took place during July of 2020, immediately after the first lockdown. And we went back in three subsequent years to these same people to ask them questions about their mental health, their activities of daily living and how they were coping.

So you had this large representative survey and you're able to go back to the same young people over and over again, or to their parents and teachers, and ask about their mental health. So what did they tell you about how they were feeling? Well, the bottom line to summarise is that there was an abrupt deterioration in the mental health of all children and young people on average at a population level between 2017 and 2020.

That was maintained over the subsequent three follow ups to 2023. Right. And it affected both children and emerging adults because our cohort, of course, was getting older. But the group that seemed to be doing the worst are young women in their late teens and early 20s who were doing particularly badly.

If they did manage to keep these follow-up surveys accurate and representative despite the lockdown, then the sheer scale of the deterioration is noteworthy.

In the initial 2017 survey, around one in nine children and young people had a probable mental health disorder. But the last survey in 2023 suggests that those under the age of 16, it's one in five, have a probable disorder. And in our high-risk 17 to 19-year-olds, it's one in four. So it's quite a big change and it's been sustained over all four follow-ups.

So you have a questionnaire that you trust that is indicating clinically significant psychiatric disorders and proportion went from one in nine to one in five and in the older teenagers to one in four. Yeah. Wow. Yeah. When you see the figures, it's really quite, it's almost a gut punch really. You know, that's a lot of young people who are really struggling.

Now, hold on, you might be saying. We've already seen how there was a declining trend in general mental wellbeing before Covid. As the MHCYP was in 2017, how do we know that this decline didn't happen before the lockdown? Well, the blunt truth is that we don't. And

And the same pre-COVID trend also came up in previous MHCYP surveys. And yes, we did document a significant deterioration over the first couple of decades of this century. But what seems to have happened is an increase in children with anxiety and depression numbers.

at disorder level. But the jump between 2017 and 2020 is an order of magnitude bigger and seems to have been maintained. And I think my work and other people's work suggests that it's not necessarily new risk factors. It's an amplification of what was already going on. The correlation with the timing of COVID is striking, even if we can't be sure about causality. So on balance,

I do think that the pandemic and resulting restrictions have had a fairly major impact. It's the fact that the effect appears to have been sustained which is the biggest concern, perhaps capturing a proportion of the young people who didn't or couldn't bounce back.

And there's one subgroup that is a particular worry. The other thing we've seen clinically and we're able to follow up in the last survey is a massive increase in the presentation of young people with eating disorders and mostly very severe, requiring emergency and urgent treatment. This survey doesn't tell us the whole picture.

While it does give us good grounds for thinking that the lockdowns left some young people with serious mental health conditions, which persisted after the lockdowns were over,

There's also evidence that some children thrived during the pandemic. The Oxwell survey repeated totally anonymous cross-sectional surveys. That found some really interesting reports on these children who actually life was better during lockdown. And it was often around problems with school that the lockdowns took them out of having to deal with.

This survey, run by Oxford University, found one third of school pupils were happier when the schools were closed. They could access education at their own pace. They didn't have to contend with bullying or a very social environment that they didn't understand.

Now, we don't know this for sure, but it's certainly plausible that this group connects to something we were talking about earlier, those high absence rates in schools. Maybe parents realised their kids were happier at home, and with the social norms broken by the lockdown, were happier letting them stay there more often. Maybe. That maybe has been a persistent problem in trying to figure all of this out. We've tried to answer a question that seems pretty important –

What was the long-term effect of closing schools and jobs and making everyone stay at home in the lockdown? It's tempting to tell a sad and simple story about a lockdown generation of young people permanently scarred by the UK's response to the pandemic. But that story doesn't match the data, patchy as that data is. And it doesn't match the rich and varied experience of young people across the UK.

Some young people lost the ability to develop their maths and reading skills. Some lost a fair chance to get a good job. Some became anxious or depressed. And a huge number of young people bounced back strong, thriving at school, in the job market and in themselves. There is no lockdown generation. There are millions of different stories and those stories aren't over yet.

And for those of us who like stories to be backed up by data, we are going to have to do a much better job of collecting that data than we have done over the past five years. That is it for this special programme. We'll be back later in the week for our Standard Edition. Until then, goodbye. More or Less was presented by me, Tim Harford. The producer was Tom Coles. Our production coordinator was Brenda Brown. The programme was recorded and mixed by Nigel Appleton. And our editor was Richard Varden.

This is LeVar Arrington from Two Pros and a Cup of Joe. The Toyota Tundra and Tacoma are designed to outlast and outlive, backed by Toyota's legendary reputation for reliability. So get in a Tundra with the available i-Force Max Hybrid engine, delivering exceptional torque and towing capacity. Or check out a Tacoma with available off-road features like

crawl control it can take you beyond the trails toyota trucks are built to last year after year mile after mile so don't wait get yours today visit buyatoyota.com for deals and more toyota let's go places

Residents at Brightview Senior Living Communities enjoy enhanced possibilities, independence, and choice. Brightview Dulles Corner in Herndon and Brightview Great Falls offer vibrant senior independent living, assisted living, and memory care services through various daily programs and cultural events.

Chef-prepared meals, safety and security, transportation, resort-style amenities, and high-quality care. Everything you need is here. Discover more at brightviewseniorliving.com. Equal housing opportunity.