Skip to main content

Transforming the understanding
and treatment of mental illnesses.

Celebrating 75 Years! Learn More >>

 Archived Content

The National Institute of Mental Health archives materials that are over 4 years old and no longer being updated. The content on this page is provided for historical reference purposes only and may not reflect current knowledge or information.

Let’s Play Ball: How Sex and Gender Effects Influence Sports Involvement, Hippocampal Volume and Depressive Symptoms in Children


>> OPERATOR:  Welcome to the National Institute of Mental Health’s Office for Disparities Research and Workforce Diversity Webinar entitled, “Let's Play Ball: How Sex and Gender Effects Influence Sports Involvement, Hippocampal Volume and Depressive Symptoms in Children.” At this time our participants are in a listen-only mode. Questions can be submitted any time during the presentation via the Q and A pod located on the lower right-hand corner of your screen. Please note that this call may be recorded. It's now my pleasure to turn the call over to Tamara Lewis Johnson. Please, go ahead.

>> TAMARA LEWIS JOHNSON: Well, good afternoon everyone. On behalf of National Institute of Mental Health, I welcome you to The Office for Disparities Research and Workforce Diversity Webinar.  This is the first of two webinars for the 2019 series. We offer the webinar series as an opportunity for program staff and the public to learn about recent scientific advances and sponsored research in the key areas that our office is focused on , which women’s mental health, the mental health of LGBTQ individuals,  the mental health of people living in rural communities, and  the mental health communities of color.
The second webinar will be held on Wednesday, the 18th of September at 10:00 AM. The title for that webinar is “Insights from Social Psychology in Neuroscience on Implicit Bias,” by Dr. Georgina Rippon.

Please allow me to introduce Dr. Deanna Barch. She is currently the Chair and Professor of the Department of Psychology and Brain Sciences and the Gregory B. Couch Professor of Psychiatry at the Washington University in Saint Louis. Her research program is focused on understanding normative patterns of functional brain connectivity across development as well as the mechanisms to give rise to the challenges in behavior and cognition found in mental illnesses, such as schizophrenia and depression, utilizing behavioral neuroimaging in computational approaches.

She received her undergraduate degree from Northwestern and completed her Ph.D. at the University of Illinois in Champaign-Urbana and completed a postdoctoral fellowship at Western Psychiatric Institute and Clinic. She is the deputy editor of Biological Psychiatry and is on the editorial boards of Schizophrenia Bulletin, Journal of Abnormal Psychology, and Clinical Psychological Science. Dr. Barch is on the Scientific Board of the Brain and Behavior Research Foundation, the One Mind Foundation and the Stanley Foundation and on the Executive Committee of the Association for Psychological Science.

Her research has been funded by the National Institute of Mental Health, The National Alliance for Research on Schizophrenia and Depression, the National Science Foundation, and the McDonell Center for Systems Neuroscience. She was a member of the National Institute of Mental Health Scientific Council and is a member of the National Institute of Mental Health Research Diagnostic Criteria Committee. She is a fellow of both the Association for Psychological Science and the American College of Neuropsychopharmacology, and a member of The Society for Experimental Psychology.
I welcome Dr. Barch to give a wonderful presentation this afternoon.
>> DR. BARCH: Thank you very much. I want to thank Tamara in the Office of Disparities Research, for offering me the opportunity to give this presentation and to thank her for that very generous introduction. I also want to make sure at the start to acknowledge my collaborators on this project, Lisa Gorham, Terry Jernigan, and Jim Hudziak. In particular, I want to highlight Lisa Gorham who was the very talented undergraduate in my lab who really spearheaded this line of research. She was a student-athlete herself and many of the ideas that we'll talk about were ones that she helped develop both from her lived experience as a student-athlete and also through reading the literature.

Let's get started. I think we need to start by talking a little bit about what we currently know about the relationships between sports and exercise and mental health in kids and adolescents. We do know already that active lifestyles in children and adolescents are positively associated with fewer symptoms of depression and anxiety, better self-concept and more effective coping with stress. In addition, active lifestyles promote a positive self-concept and more effective coping with stress. That's one starting point for the work we're going talk about today. We also know that being involved in sports through the high school years is predictive of more feelings of school engagement, doing better at school and having increased self-esteem.  While this is known for young people, and we also know this to be true in adults as well.   If higher levels of exercise are significantly associated with lower levels of depression, then that begs the question of what might be the mechanism that might be leading to some of these associations and might there be mechanisms in the brain that are contributing to these relationships. We know that children who are what we would refer to as higher fit children have better cardiovascular fitness and show greater bilateral hippocampal volumes.

The hippocampus is very important to think about in this context because the hippocampus is a region of the brain that is important for, in particular, stress responses, and things like memory and cognitive function. While bigger is not always better, in this case, much of the evidence in the literature suggests that larger hippocampal volumes are associated with more positive cognitive attributes such as doing better on cognitive tasks and having better stress responsivity. We also know in kids that greater aerobics fitness predicts larger left middle prefrontal cortex volumes. In addition, the prefrontal cortex is another part of the brain that may also be very important for cognitive control and emotion regulation.

Interestingly, this is an area of research where there are some nice homologies between the animal research and the human research. That's not always true but in this case, if you look in rodent models where you can experimentally manipulate things, we see that robust exercise leads new cell growth in the hippocampus, which is what people sometimes refer to as neurogenesis. There are some nice parallels between animal research that are consistent with the human research where we can't always so easily do experimental manipulations.

Now the third part of this is understanding depression and the brain. We've talked about the fact that previous literature suggests that among kids and adolescents, greater involvement in sports and more aerobic fitness are associated with less depression, less anxiety, and greater self-esteem. We've also talked about those same characteristics being associated with greater hippocampal volume in both humans and animal models and some evidence for greater prefrontal volume. How might that then link exercise and sports to depression?

Well interestingly, people who have major depression, pretty consistently show reductions in hippocampal volume relative to individuals who are currently healthy or don't have major depression. That's been seen in many different studies. It’s been seen in individual studies. It been seen in large meta-analyses. There's a relatively consistent relationship between people having experienced clinical level of depression and having smaller hippocampal volume. We also know that this relationship is even seen in very early onset depression, and adolescent depression, so it's not just in adults with depression or for people who've had many years of chronic courses of depression.

Interestingly, we actually see some asymmetry in terms of gender there where we see some evidence that smaller hippocampal volumes in depression are more prominent in males than females. Although not every study has looked at this question, many studies don't look at it separately for males and females, and not every study that has looked at it separately from males and females has seen consistent sex differences so it's more of a hint than anything else.

There are a number of theories or ideas about how hippocampal volume and depression might be related. One prominent model is that things in an individual's life, whether that be stress or other factors, that disrupt hippocampal development and function might lead to challenges with stress regulation. The hippocampus importantly contains many receptors for what are called glucocorticoids. Glucocorticoids are an important part of the brain’s pathway to respond to stress by helping to shut off the brain's response to stress. When we have a stress response and we activate something called the hypothalamic-pituitary-adrenal axis, one of the things that happens is that we release cortisol. In order to help shut down and regulate that response, cortisol needs to bind with glucocorticoid receptors, and those are highly present in the hippocampus. Disruptions to the hippocampus can disrupt the ability of these glucocorticoid receptors, which may make it more difficult to shut down these stress responses. It's been hypothesized that that  contributes to risk for depression.

It may also be the case that the experience of depression over time may be increasing people's experience of stress, and also contributing to subsequent changes in hippocampal volume. It may go in either direction. Given these three kinds of research that I've just talked about, research linking exercise and sports to better mental health, linking exercise and sports to hippocampal volume, and the evidence linking hippocampal volume to depression, we went into this project with a number of hypotheses.

First, we thought that among children in involvement in sports would be associated with a severity of depression.  To be concrete about that, we thought that children who were more involved in sports would be less likely to have depression or would have lower levels of depression. We also thought that involvement in sports in kids would predict larger hippocampal volume bilaterally, meaning both on the right and on the left.  More sports involvement, greater hippocampal volume. We also thought that larger hippocampal volumes would predict depression severity in children so that if you had larger hippocampal volume, you would have less depression severity.

We thought that the hippocampal volume would mediate, meaning be the pathway by which the sport is related to depressive symptoms. The hypothesis is that greater involvement in sports contributes to increased hippocampal volume, and that may be protective against depression. That hypothesis is a causal hypothesis, meaning that we're making predictions about what might be causing what. I will be very clear from the beginning that the data we have so far is going to tell us about relationships, and it's not going to establish what causes what. The underlying hypothesis is certainly that there is a causal relationship.

So far, I've mostly talked about exercise and sports in the brain. Much of the literature on sports and mental health has shown the potential of that sports involvement might have on the brain. There are other things about being involved in sports that might have positive mental health benefits for both boys and girls, but maybe even in particular for girls. That's the idea that involvement in sports or at least some types of sport may be an important form of social support for individuals.

This is where again my student Lisa who helped to spearhead this work. She had been a student-athlete.  She played volleyball. She ran. She felt like she had received a lot of social support from being involved in these team sports, from the other members of her team and from her coaches. She thought social support might be potentially an important pathway by which involvement in sports might be associated with reduced depression. If you think that's true, then you might predict that involvement in certain kinds of sports might be particularly strongly associated with less depression.

Team sports, for example, where you are part of a team might be a form of sports involvement that might be most likely to be associated with better mental health, better self-esteem, if the social support was part of the action on the brain. You might contrast that with sports that were more individualized, either where you were an individual athlete or even where maybe you were part of a larger team, but the way the sport is actually carried out is individually.

To give you a couple of examples of that, playing baseball, or softball, or soccer is something that's a team sport where you have to play as a team. That is the nature of the sport. There are other sports, say being involved in track and field where you may be part of a team, but you may compete individually, and then there are people who may be engaged in sports on their own with individual coaches, who are not necessarily part of the team. If there is an additional role for social support, we might see some differences across those types of sports and their relationship to depression. Then, the key question is, do any of these relationships differ for boys versus girls? Again, going into this, Lisa's hypothesis was that the social support component might end up being particularly important for girls versus boys, but that initially, we both had hypothesized that the pathway through hippocampal volume would be pretty similar for both boys and girls.

Let me start by telling you a little bit about the methods. We had the unique opportunity to work with a data set that I will tell you a little bit more about because I think it's a very important resource for the community, and so given that we have so many community members on the line, I think it will be useful to take a minute and talk about it. This particular set of analysis was done with 4,196 children, aged 9 to 11 from locations across the country, as part of a project called the Adolescent Brain and Cognitive Development Study.

I'll do a little sidebar here for a few minutes and tell you about this project, that I'll call ABCD for short, and then I'll come back and talk about the specific methods that we used for this project and the results. We have many wonderful federal partners for the ABCD project. It's a very collaborative project, so it's being funded by support from many of the different National Institutes of Health Institutes, including Drug Abuse; Alcohol Abuse and Alcoholism; Cancer; Mental Health; Child Health and Human Development; Heart, Lung, and Disease. I have them listed here so I won't read them all, but also partners like the National Institute of Justice, the Center for Disease Control and Prevention, and the National Endowment for the Arts. There are many federal partners. This is a very ambitious project.
The ABCD study has recruited approximately 11,875, it depends on exactly which count you say, children ages, 9 and 10 when they were enrolled in the study and we're hoping to follow them for up to 10 years. The goal of the project is to try to understand and identify individual development trajectories of things related to the brain, and emotion, and cognition, academics, and the various factors that can impact them both positively and negatively.

We have 21 sites across the country, I happen to be in the site at Washington University in St. Louis. We have 100-plus investigators, and the data is being released publicly. I will say a little bit more about that.  I think this is a very unique study because many studies that are about normative development maybe are not particularly large studies, and many studies that are about mental health development tends to focus more on the things that have negative impacts on children's development.

What I think is really unique and important about ABCD, is that's it's really trying to look at both the things that help support positive brain development, cognitive development in kids, and the things that may get in the way. As part of that then, we are looking at things like sports involvement, involvement in the arts, positive support in the community, all kinds of factors that might be resilience factors that help children develop in healthy positive ways, as well as the things that might get in the way.

 We will be able to and it is a large we because it won’t be just us it will be other ABCD partners and the larger scientific community. We'll be able to use the data both starting at baseline and throughout 10 years, to understand these factors in a way that might let us develop new interventions or new things that we could help to implement that would promote healthy brain development in kids. Just in case anybody's interested, the red dots here are our 21 sites across the country, so we stay on the East Coast and the West Coast. You will see fewer study sites in the middle South-West.

That is in part because we did need to have sites that were in major academic centers and had the imaging facilities necessary to carry out the study. We worked very hard to try to recruit a very diverse sample. It is not exactly representative of the United States population, but we tried as best as possible to approximate as much of the diversity in the US population of children as we could. It is diverse in terms of socioeconomic status, racial and ethnic minorities, and equally represented across genders. This just gives you a schematic of the kinds of things that children and their families do. We do ask them to come into our sites every single year.

Every other year they do neuroimaging where we get a number of different kinds of neuroimaging measures. We look at the structures of the brain, the size, and the shape. We look at connections among brain regions, both structural and functional connections. We look at brain activity when kids perform different kinds of tasks that measure cognitive and emotion and reward challenges, but we also get many other measures. We have children and their families tell us about their experiences, their communities, the kinds of things that they're involved in.

We collect some biomarker data in terms of being able to look at genetics, hormones and development, through a very broad array of measures on each child. We've also started to use what we'll call mobile technologies.  Children are wearing Fitbits so we can measure physical activity in a way which will be an important addition to the kinds of data that I'm going to talk about today. I just wanted to note here to really highlight the fact that the ABCD data is available to qualified investigators through the National Data Archives at the NIH support.

We've already done two data releases, so all of the baseline data is available for investigators and part of the follow-up data. I emphasize this because it really is a wonderful and rich resource for people to answer questions that maybe we never even thought of. In addition, this data might be really useful for many of the folks on the phone here. Now let me come back to our particular project that we're going to talk about today using the data from the ABCD study. As I mentioned before, we were looking at data from the first release, what we call 1.0, to measure depression and anxiety we use something called the child-behavior checklist.

This is a parent-report measure that looks at anxiety and attention, aggression, externalizing and internalizing depression, and other symptoms. Every parent or every caregiver in the study fills that out about their child. As an aside, as the children are getting older, we are increasingly asking them to self-report on their own emotional experiences, but at the younger ages we are still heavily relying on parents helping inform us. For the imaging data, as I mentioned already, we acquire a number of different kinds of imaging data.

For this project, we focused on the structural imaging data, and we used FreeSurfer, which is a program that will segment the brain and help identify the shape and the volume of different parts of the brain. We focused for this project on the hippocampus. Again, as a bit of an aside, this project actually uses multiple different scanner types from different companies. The fact that the images are coming from different scanners is taken in account in the analyses.

 And then we asked quite a bit about sports and activities and I going to go into in a little bit of  detail to give a sense of the kinds of things we asked about and  a sense of the range of activities.  We worked with a number of people in the field to develop a questionnaire that was as inclusive as we could be about the different kinds of sports and activities that kids might be involved in. It looked at lifetime involvement in a whole huge range of sports and then also activities like music and dance, and other hobbies. We tried to capture as many different things as possible. For example, the questionnaire asked about things like chess club and all kinds of music involvement, arts and crafts, painting, dance, acrobatics, every kind of sport that seemed like any kid could be involved in, arts and craft, painting, just a wide variety of things.

We asked parents to tell us about the frequency of their being involved, how long they were involved, the type of activity. Then we also followed up and asked whether it was being done as part of an organized activity at school or outside of school, whether they were getting private lessons or doing it as part of a group, was it an individual sport, was it structured so we tried to get more information about that but doing it this way gives a sense of what kinds of things kids are involved in that are not necessarily part of a structured activity as well as those things that are a structure activity.

Gathering the data in this manner allows us to capture things that kids might be very involved in, that aren't necessarily part of a structured activity, as well as those things that are part of a structured activity. The data is from baseline, but we continue to ask these questions every year when the kids and their families come in, so we are continuing to get updates about their involvement in sports and activities, so we can look at how that changes over time. As I noted before, in one of the follow-up years, we've also started asking children to wear Fitbits for several weeks so we can get other measures of physical activity.

 The different sports we asked about, you can see a big range here. We do update this list of sports as we get information about sports that we might have missed or about which we weren't aware, but we tried to be as inclusive as possible. Then what we did was develop some definitions about different kinds of sports to try to get at this question about social support. We defined team sports generally, as when a child engages in a sport at school or through an organized outside league. It could be any sport on that list, but it was when it was through an organized kind of group.

Then we had a more restrictive definition of team sports, where it had to be a sport where the actual playing of the sport occurs as a team. Things like baseball, basketball, field hockey, football, volleyball, all those things where you're actually playing it as a team. An individual sport would be something where the child engages in that sport on their own time or through private lessons. You could, for example, have a child who's very involved in skateboarding, and that would be an individual sport, or maybe they're involved in tennis, but it's through private lessons, and not through school, something like that.

We have another definition that we call structured sport, which overlaps a little bit with team sport, but was basically any sport that the child was involved in through school, or an organized league or through private lessons. That was kind of a broader category that included some of the others.

 Okay, let start by telling you about the results and let me orient you to what I am going to show you here. We started the study out by asking whether any of these different kinds of activities or sports were associated with depression and whether there were any gender differences in that association. I will go through and show you those different results.

We did look at not just sports, but we also looked at involvement in activities because it's certainly is possible that it's really not sports per se. It could be that involvement in any kind of organized activity really had positive mental health benefits for kids. We also looked at activities in addition to sports to try to ask that question and see how specific it was to sports involvement. All right let me start by showing you the results for what we call the "Main Effect." Is there a relationship between one of these activities and depression? The beta weight is from a regression, actually, a linear mixed model that looked at the relationship between these different kinds of categories of sports or activities and depression, taking into account the fact that some kids might actually be siblings or twins, which is important statistically. A negative value here means that, for example, greater involvement in sports was associated with less depression and then I'll show you the p-value. Now, you'll see that many of these p-values are wildly significant. This is a very large sample size and so it's not so difficult to get things to be significant. We should pay attention to sort of how big that beta weight is.
These are very robust relationships. They are not huge relationships, but they are very robust. When you look just at this question of what's related to depression, we see that pretty much everything is related to less depression. More involvement in activities, more involvement in sports, in general, being involved in a team sport whether it's the broad or the restrictive definition, being involved in an individual sport, being involved in a structured sport. Every one of those is associated with less depression.

 Again, as a reminder, since we are looking at the baseline data, what we can say is that the data are related, we can't tell you what's causing what. It could be that being involved in these activities and sports is preventing kids from developing depression or it could be the case that more depressed kids are less likely to engage in these things. We'll talk a little bit more at the end of what are the next steps in order to identify which is causing what. Then we wanted to look at were there things that showed an interaction with sex, and a little bit to our surprise many of these relationships actually showed an interaction with sex.

When we followed things up to say, "Was it significant in boys and girls or just one or the other?" what we found is that by far these relationships were much stronger in the males than the females. Pretty much across the board except for interestingly team sports, the restricted definition of team sports, so being involved in a sport where you are part of a team like baseball or football.  There we saw a significant relationship for both boys and girls. It was still stronger for boys than for girls, but it was significant for girls.

Whereas for everything else involvement in activities more generally, just general involvement in sports, individual sports, it was significantly related for boys but not for girls. I should also note these relationships were specific to depression, as opposed to anxiety, so when we looked at predicting anxiety, we did not see a relationship between involvement in sports and activities. It seems the relationships for depression held even if you included anxiety in the model.

It seemed to be more about depression than anxiety and it seemed to be much stronger for males than females. Just to illustrate this relationship, I've got two things here to illustrate it for you. On the left, I have a scatter plot that shows the relationship between the number of sports in which a child is involved and their depression score. What we have done here is shaded the dots separately, so the open dots are female and shaded in dots are males. Then I have some regression lines there, the solid one is for females and the doted one is for male.

What you can see there is that yes, indeed overall there is a relationship between more sports and less depression, but that relationship is a bit stronger for males than females. On the right is a different graph where we have grouped kids into participating in a team sport yes or no, the restricted definition and their parent-reported depression score. In that graph on the left are the kids who are not involved in the team sport and on the right are the kids who are involved in the team sport. What you can see there is that for both males and females but particularly for males, parent-reported depression levels are lower if the child is involved in a team sport than if they're not involved in a team sport.

Overall our kids are not showing super-high rates of depression although sadly there are some kids in this sample who do have some fairly high levels of depression but fortunately, there are many kids who do not have high levels of depression. It's still the case that those levels of depression are lower in the kids involved in team sports versus not. We then went on and asked about hippocampus and depression, and so I'm going to show you some data analyzed in a somewhat similar way where we use these linear mix models to look at whether or not the hippocampal volumes are related to depression.

Indeed, what we saw was that overall it was the case. Higher levels of depression were associated with smaller hippocampal volumes and there was an interaction with sex that was such that this relationship was stronger in males than females. Lower hippocampal volume was significantly related to greater depression in males but not in females. Then, what about the third leg of that triangle, involvement of sports in the hippocampus? I'm going to show you the data in the same way I did for sports involvement and depression. We've got the different kinds of activities and sports that are going to be by row, and we're going to look at what is related to hippocampal volume.

To start, what we see is that many forms of sports and activities were associated with, in this case, larger hippocampal volumes, which is the direction we would predict. There was some relationship between just being involved in activities in larger hippocampal volume and sports in general, but the strongest relationships we saw actually were for involvement in team sports and involvement in structured sports. Interestingly, here, we did not see any interaction with sex.

We saw that these relationships were present for both males and females. This next slide here will show this relationship for both males and females. If you look at the graph on the left, what you can see is that for both males and females, you have larger hippocampal volume being associated with a greater number of sports, in which a child is involved.

Then, if you look at the graph on the right, now that we've got it plotted for kids who are involved in a team sport versus kids who are not involved in a team sport, and you see for both males and females, being involved in a team sport is associated with a larger hippocampal volume than not being involved in a team sport.

We have this interesting association now where the relationship to depression, both for sports and hippocampal volume, are stronger for males and females for at least most forms of sports, but that their relationship to the brain is similar across males and females, so being involved in more sports, particularly team or restrictive team sports is associated with larger hippocampus volume in both boys and girls, but only in boys is it then also associated with less depression.

Now, remember, I raised the issue before saying, is the data consistent with the idea that the pathway by which sports may be associated with less depression might be through hippocampal volume? We did an analysis called the mediation analysis that tests that kind of hypothesis. As I showed you earlier, for boys being involved in team sports was associated with less depression and being involved in team sports was associated with greater hippocampal volume, and greater hippocampal volume was associated with less depression in boys.

Importantly, that direct relationship between being involved in team sports and having fewer depressive symptoms, that relationship is reduced when you take into account hippocampal volume. That's consistent with what we call partial mediation, meaning that part of the relationship between team sports and depressive symptoms is being accounted for by hippocampal volume.

Now, that direct relationship is still significant, meaning that there are things other than hippocampal volume that are contributing to the relationship between team sports and depressive symptoms. For example, it might partly be the social support involved in team sports or other factors that might be contributing. One important point I do want to bring up, there are lots of socio-demographic factors that influence the ability of kids to be involved in sports.

To some extent, this is more true when kids are younger where public schools don't necessarily always have options for team sports involvement for kids like in grammar school or middle school. For a lot of public schools, that may not start until you're in junior high or middle school or high school. Unfortunately, many times being able to be involved in sports or other activities when kids are younger requires families being able to afford to pay for that.

That is a potential important confound. All of these analyses though also included information about family socioeconomic status. All of these relationships were still present even when we accounted for family socioeconomic status, which itself has a relationship to the likelihood of being able to be involved in some of these sports and activities, as well as a child’s mental health. That's an important point I forgot to mention earlier.

To summarize, what I've shown you is that these data suggest that participation in sports was related to less depression in boys but not girls, somewhat to our surprise. Also, this is very consistent with our work on exercise and depression. We saw some very modest relationships to participation in non-sports. In some follow-up analyses we did, all of the relationships to sports involvement held even if we controlled the general activity involvement.

The activity measures I showed you before included all activities, sports, and non-sports. If we separate out just the non-sports, we did not see many relationships to either depression or to hippocampal volume.
As I noted here though, we weren't really necessarily anticipating sex differences at least in the direction that we saw, where we saw these relationships being stronger for boys than girls because these previous findings either had not looked at it or didn't see such sex differences. It is possible that there's something about the developmental stage of our kid that's important. These kids are younger, and the meaning and the role of sports may change quite a bit as kids evolve through puberty and into adolescence.

We know that girls enter puberty earlier than boys on average. There are, of course, individual differences. It is possible, we think, that the relationship of sports and exercise to depression in girls might become quite different as they enter into puberty. Also, there will be hormonal changes that could influence those relationships, but I think also the social meaning of involvement in sports might be different as girls enter into puberty and what that means in terms of self-esteem, sources of self-esteem, sources of social support for girls.

The meaning of that may change as they move into puberty. The hypothesis is that those relationships might become stronger for girls as they move into puberty, even though they're not as present in girls, these younger girls, most of whom have not entered puberty yet or are only at the various early stages. I think that these findings also raised some interesting potential questions about differences and cultural attitudes about sports for girls versus boys.

There certainly are differences to some extent in expectations historically for participation in sports for girls versus boys. Some of that is clearly changing. There are important changes that are happening in society in terms of the importance of girls playing sports and the visibility of girls playing sports. Certainly, the recent wonderful winning soccer team in the U.S. has very much highlighted the potential role for sports and girls, but historically, that has not always been the case.

Some of the differences in the relationship to depression might be accounted for by some of these cultural attitudes. Again, they may change as we move into puberty where there may be different interpretations about sports involvement and different impacts for girls versus boys. Interestingly, though, we did see that work was related to hippocampal volume in both boys and girls. That was true for involvement in any type of sport other than just the individual sport.

Although, truth in advertising, it was somewhat less likely for people to be involved just in individual sports. It could be that if we had greater variance in the intensity of individual sports involvement that maybe we would have seen more of a relationship there. We did not see these relationships when you pulled out just activities that were not sports. That would have been things like chess club or music or art. That is not to say that music is not good, or arts involvement is not good for children's mental health or self-esteem.

It just means that in these particular analyses, we didn't see those relationships here. Again, how those kinds of things relate to children's mental health and brain development may be evolving as the kids get older and move into puberty. This really extends the previous work in adolescents and adults to kids. It also suggests that exercise is associated with enhanced hippocampal volume, which again, many people interpret as a positive brain attribute.

We like to think though that there may be some added role for social support or engagement. The fact that we did see these relationships more strongly for team sports than individual sports means that there could be an added benefit in terms of hippocampal volume to social support and maybe stress reduction. One of the things we did to follow up on that is to say, "Well, does this relationship to team sports hold even if you account for involvement in other sports or if you do what we'll call a dose-response relationship?"

We did some follow up analyses where we tried to generate an overall estimate of how many hours on average a week a child was being involved in sports, because some children were involved in several different sports. One possibility was that those kids who were involved in team sports, maybe they were just more likely to be involved in sports, spending more hours being involved in sports.

Maybe it's not really anything about being in a team, maybe it's just the number of hours that you're engaged in sports. We created this dose measure of sports involvement and asked whether you still saw a relationship to being involved in team sports, even when you took into account the intensity of their sports involvement. We still saw that relationship. It does suggest there is some added benefit of the team context, and whether that is having some more direct impact on hippocampal volume development or if it's through something having to do with structure or social support, we're going to need to do further research to tease that apart.

We did see that smaller hippocampal volumes were related to depression in boys, though not girls. There is, as I noted, a little bit of previous work suggesting more evidence for reduced hippocampal volume and depressed males, but not depressed females. Although there are certainly other studies that have also seen reduced hippocampal volume in females with depression, so this is clearly an area of research where more work is needed.

It's also consistent with some prior animal work, looking at, for example, more impairments in male rodents after maternal deprivation interventions that seemed to be associated with, depression-like behaviors in animal models. There is a evidence that males may have some greater vulnerability or susceptibility to some early things that might be associated with depression.

There's also some very interesting literature that has suggested neuroprotective effects of estrogen, though the evidence for this is somewhat mixed. That is another possibility why early on, we see these effects more strongly for boys than girls. We also see some interesting evidence that testosterone might play a role in increasing vulnerability to neurotoxic processes in rats.

That would be consistent with the possibility that boys, at least early on, might have some greater susceptibility to things that might be depressogenic and may show more strong relationships between things that would modulate hippocampal volume and depression. It might be showing those relationships more strongly in boys than girls as these kids move into puberty. Again, that may change as these kids move into puberty.

Some important future directions here. I keep hammering on the fact that these were associations and they can't tell us about causal directions. We do need to determine causal directions because we don't want to make recommendations about what kinds of things kids should be involved in without evidence of causal relationships. I do think the longitudinal nature of ABCD will really help with that so we will be able to ask questions like, for example, "All right, how much a kid is involved with sports at baseline?"

If we look over time, does that predict their development of depression over time even when you control for a depression starting out? If there really were a causal relationship between involvement in sports and depression, we should see that those relationships predict over time and we can look at the opposite. We can say, "Okay. Well, is it the case that how depressed the kid is at time one, does that predict the likelihood that they'll be involved in sports at time two, which would be sort of more consistent with the opposite?"

Should we get more evidence for causal relationships, one could then think about potential intervention studies. It has a number of public health implications should we find evidence that involvement in sports causally has an influence on depression. As I noted before too, I think it's going to be important to continue to see whether these gender differences continue into adolescence.

As boys and girls go through puberty and have hormone status changes and the meaning of sports involvement may change, we would predict that we would start to see a stronger relationship between sports involvement and depression and self-esteem in girls as they move into adolescence where that might be particularly helpful and protective, but that is an empirical question. We think it will be important to look at the relationship of hormones and puberty status might modify or moderate the relationship between exercise and sports involvement in both brain development and mental health in kids. All right. I will stop there and wrap up, and then we can proceed to taking questions.

>> OPERATOR:  Any plans for younger subjects?

>> DR. BARCH: I'm assuming that means in the ABCD. Enrollment for the ABCD is complete, so the ABCD itself will not be recruiting younger kids. We will be following the same kids over time, but I think there is an interest in working with even younger children. You'll have to stay tuned as to whether there might be some new projects coming down the line that would start working with kids much younger.
There certainly are some other projects being done. There's a project called ECHO, which is Environmental Child Health Outcomes that is recruiting much younger kids and looking at a host of environmental factors that may be influencing development. That would be an opportunity to look at some of these similar questions.

>> OPERATOR:  Are you willing to share your PowerPoint slides?

 >> DR. BARCH: Yes, we'll be happy to share the PowerPoints with people.

>> OPERATOR:  Can you say more about the measurement and operationalization of the different categories. What was included as a non-sports activity, what were team sports, and what were individual sports?

 >> DR. BARCH: For non-sports activities, we tried to ask about as many different things as possible. We asked about involvement in all forms of art, so music, painting, anything like that at all.
We looked at things like debate club. They're a little young for this yet, but they are things that people can be doing. Chess, checkers, anything that was not related sports-- There were enough evidence that kids regularly engage in these sorts of things. We tried to measure those.

Then, for team sports, our restrictive definition of team sports was it had to be a sport that was played as a team, so baseball, football, volleyball, field hockey, soccer. Things where you're playing it as a team, ice hockey. Then, the more general definition was if you were engaged in a sport as part of a team even if the sport wasn't played as a team. Some kids might be, for example, on a swim team. You could do relays and we would consider that a team sport, but mostly, swimmers are swimming their individual activities, so they're not playing the sport as a team, but they are on a team.

That would be considered a team sport. Then, an individual sport would be if the kid is doing it on their own, not as part of a team. For example, maybe they're getting tennis lessons and doing tennis lessons, but they're not doing it as part of a team, or they're doing some biking activity that they're doing on a regular basis but they're not doing it as part of a team.

 >> OPERATOR:  Did you look at kids who are not engaged in organized team sports but exercise regularly on their own?

>> DR. BARCH: Yes, we did look at that. There were not that many parents who reported that their kids do that. There are opportunities, say, like they jog regularly on their own. We would certainly count it as being involved in an individual sport if a kid, for example, was really involved in skateboarding and did that regularly, but they didn't do it as part of a team.

That would be counted as an individual sport. That's where I think the Fitbits will be very useful because that will give us a measure at least over the time period that the kids are wearing the Fitbits of their level of exercise in a more objective fashion.

>> TAMARA LEWIS JOHNSON: "You mentioned the culture of sports differences between boys and girls. Are you able to ask questions related to race and ethnic differences as it relates to sports?"

>> DR. BARCH: We haven't yet but we could ask whether there are differences across individuals who self-described as being of different races or ethnicity in these relationships. That is certainly something we can look at. We don't currently have measures in the ABCD that explicitly asked about attitudes about sports involvement. That actually would be an interesting thing to consider.

Yes, we could ask whether these relationships vary as a function of race or ethnicity or other cultural characteristics that are being measured. What we don't currently have is information about exactly what those differences might be in terms of attitudes about sports involvement across genders or races or ethnicity.

>> TAMARA LEWIS JOHNSON: "Can you elaborate more on the measurement and operationalization of the different activity categories? For example, what was included as non-sport activities? What were team sports and what were individual sports?"

>> DR. BARCH: Non-sport activities, we tried to ask about as many different things as we could that kids were involved in. That would include all the different music options, all the different other kinds of parts involvement, painting or anything else that kids could be involved in. It would include things like debate team, it would include things like chess or checkers teams.

Anything that seems to be of enough frequency that we would have enough kids reporting on being involved in that. Our restricted definition of team sports was that the sport had to be played as a team, baseball, soccer, hockey. Our more general definition was that the kid was doing it as part of a team, but not necessarily playing the sport as a team. The best example of that is swimming.

At this age, a lot of kids are involved in swimming, and they may be part of a swim team, but they may be only swimming individual events. It's a general definition of a team sport because they are on a team, but they are not actually executing the sport as part of a team.

>> TAMARA LEWIS JOHNSON: "Sports involvement was related to fewer depressive symptoms for boys and not girls. Why do you think that is if we know that doing exercise releases endorphins, which could then be related to better mental health?"

>> DR. BARCH: I cannot say that that was a result that we anticipated. We know less about what it's like for kids, like do we know for sure that kids are releasing endorphins in the same way as adults when they're engaging in sports? No, we don't necessarily know that. Our data did suggest that sports were related positively to hippocampal volume in both boys and girls, it just wasn't further related to depression.

It's also the case that we are not measuring mood in the moment while the kid is engaging in sports. We're measuring more like does the kids show symptoms of depression more generally? An interesting question would be, to what extent do boys and girls experience mood elevations in the moment when they're engaging in sports? Is that different between boys and girls? Or is it the same between boys and girls? Because that might be different than whether it's then associated with an overall reduction in depression symptoms, which are not necessarily being experienced in the moment that they're playing sports.

>> TAMARA LEWIS JOHNSON: Are you thinking about looking at comparative models from international data that might be available in sports for kids in that age range?

>> DR. BARCH: There are a few other studies. Particularly, there's a study called Generation R that's going on in Europe that would have some similar data, and that might be an opportunity to do some comparative look at those relationships. That's a great idea.

>> TAMARA LEWIS JOHNSON: Are you measuring adverse childhood experiences (ACEs)? Might you be able to explore this down the line?

>> DR. BARCH: Yes, we are. We have a number of different ways of measuring at least some components of adverse childhood experiences. Parents do tell us about things that their child may have experienced that would constitute as adverse childhood experiences. In some of the follow-up years, we started asking both kids and parents about life events that they might have experienced.

The other thing that we also have data available on is geocoding, where if you have information about where someone lives, there are databases available like the American Community Survey that could tell you about characteristics of the neighborhood. That can also give you some information about, for example, poverty or crime in the neighborhood. You could look at that in relationship to child development. There will be a number of different measures that people can look at in the ABCD either separately or in relationship to how they impact sports involvement and its relationship to brain development or depression.

>> TAMARA LEWIS JOHNSON: "Do you think there is  any correlation between this data and young people who have higher ACE scores, not looking at racial differences, that someone else asked but looking at young people who experience high levels of trauma in their lives, knowing that perhaps there is higher incidence of depression and less access to sports?"

>> DR. BARCH: There's two ways you could look at it. One way you could look at it is to say, is involvement in things that we think might be beneficial, and again, we don't know from the data the causality, but I'm just going to say, might it be the case that kids who are particularly at risk for depression because of some earlier adverse experiences might show particularly strong benefit of being involved in sports or other things that might be positive for brain development and depression?

That's a really interesting question, and definitely, one that we could explore.  Another question, is it the case that kids who had an earlier adverse experience are less likely to be involved in sports? Is that one of the things that might be contributing to an increased risk for depression or less healthy brain development? One could certainly look at that as well. Those are really interesting questions that we haven't looked at yet.

>> TAMARA LEWIS JOHNSON: "Is there a sleep component measured? Individuals who participate in sports tend to focus on self-care related to sleep and nutrition, which can also be associated with depression."

>> DR. BARCH: Yes, so we have measures of sleep from the parents starting at baseline. We do ask about sleep. Then, as the kids got a little bit older, these data were from baseline, we're just about to finish with the year one follow up, and we've already started the year two follow up. In year two, we also started asking kids about what we call chronometrics, so as kids are moving into puberty, their sleep cycles change. We are asking questions about that.

With the Fitbits, we will be able to get some information on sleep as well. That's interesting in and of itself, but it could also be something that's part of the relationship between sports and brain development and depression.

>> TAMARA LEWIS JOHNSON: I think this might be one of our last questions. Relying on parent proxy, do you think there may be differences in how parents are perceiving depressive symptoms in their children based on gender norms?

>> DR. BARCH: Yes, it is certainly possible. As the kids get older, we are increasingly asking them to self-report on their own depression and anxiety symptoms. That is something we'll be able to ask about explicitly. Is it the case that you still see relationships when kids are reporting about their own depression levels, and do you continue to see any evidence for gender differences when kids are self-reporting?

>> TAMARA LEWIS JOHNSON: That concludes our questions. I want to thank everyone who submitted a question. Wait a minute, I have two more. Are there differences between contact sports, for example, football, rugby, men's hockey, and possibly soccer, and any non-contact team sports such as swimming, tennis, volleyball?

>> DR. BARCH: That's a good question. We did not break them up by contact versus non-contact. That's an interesting question. That is something we could look at, but we haven't so far.

>> TAMARA LEWIS JOHNSON: Does CBCL have normed data that this data could be compared with?

>> DR. BARCH: They do and in fact the data that I showed that is the T-score is the normed data.

>> TAMARA LEWIS JOHNSON: Are you thinking of recruiting younger kids?

>> DR. BARCH: ABCD will not be recruiting younger kids but the Environmental Child Health Outcome (ECHO) study will be recruiting younger kids for their study.  There are other projects in discussion that may include younger kids. 

>> TAMARA LEWIS JOHNSON: That concludes our questions. I want to thank everyone who submitted one. I want to thank Dr. Barch. I want to thank everyone who participated in making this webinar happen. You see here our contact information, if you want to contact either myself or Dr. Barch, you can do so by email. Thank you very much for participating in the webinar this afternoon. I hope that you'll be tuning into the webinar that we'll be having in September.

>> DR. BARCH: Thank you, everybody.

>> OPERATOR:  This will conclude today's program.

>> TAMARA LEWIS JOHNSON: Thanks. Bye-bye.

>> DR. BARCH: Bye-bye.