What is the Relationship Between Physical Activity and Mental Health?


An interview with Brendon Stubbs, PhD.




“Speaking of Lifestyle Psychiatry” is an interview series with prominent researchers and clinicians who have made meaningful contributions to understanding the role of lifestyle interventions in mental health care. As this field continues to grow and evolve, the goal of this series is to facilitate a critical discussion of research methods and existing evidence, as well as highlight any evidence-based, practical interventions clinicians can utilize presently.

Brendon Stubbs, PhD, is a clinical-academic physiotherapist who studies physical activity and mental health, the mind-body interface, and healthy aging. He has a Bachelor of Science in Physiotherapy, a Master of Science in Neurological Rehabilitation, and a PhD in Pain Medicine and Rehabilitation. He works locally, nationally, and internationally with collaborators, with whom he has published more than 500 academic papers, many of which focus on physical activity in people with mental disorders. He is a top 1% cited researcher worldwide.

In addition to academic papers, Dr Stubbs has written a book on the evidence-based use of physical activity in the treatment of mental illness and contributed to the European Psychiatric Association Guidelines and Position Statement on the use of physical activity for severe mental illness and Lancet Psychiatry Commission on improving physical health and wellbeing in individuals with mental disorders. His work has been featured in a number of media outlets. He is currently a researcher and clinical lecturer at the Institute of Psychiatry, Psychology and Neuroscience (IoPPN) at King’s College London.

Young: A physiotherapist is not a specialist I commonly see as part of mental health care teams. Tell us more about your background, including what roles you have played clinically and how you got involved in the research you currently do.

Stubbs: So, I ended up in mental health by mistake. I wish I could say it was a noble plan for a long period of time, but it wasn’t. But it turned out not to be a mistake at all.

Soon after graduating—this was in 2003—I worked in a mental health hospital. I was really endeared to all of the patients I was working with, and I just saw this huge opportunity to make a real difference. How people’s bodies could affect their brain and their mental health was not a particularly hot topic at the time, especially in inpatient psychiatry. Yet, I noticed the food people were eating was rubbish, and people were not prioritizing moving around and walking—all things that would be within the usual purview of a physiotherapist. I just saw huge potential to help.

So, I started there and then worked in different roles within mental health services—inpatient, in the community, working with older adults with mental health issues—and I started to realize that we were doing some things that no one else had written about. And some of the stuff was quite basic: just getting people out walking and increasing people's activity levels and how this related to mental health. I thought it'd be a good opportunity to write out some of the things that we were doing, and I started in research in that way. Then I sought more formal training in research and carried on from there.

Young: It sounds like the research you went on to do was really informed by the clinical experience you had first.

Stubbs: Absolutely. I really just view myself as an inquisitive clinician. A lot of the research was things I was seeing in clinical practice and was trying out, and patients were telling me what they liked or didn't like. Coworking and codeveloping with people with lived experience was something very intuitive. All of the work was driven by the thought, What can we do to make people's lives better today and in the future?

Young: You’ve done a great deal of research since that early clinical time. What has been the focus of the research?

Stubbs: When it started, it was really led by local issues, which turned out to not just be local issues, but really broad issues. So, the initial research was focused on, How can we get this very sedentary population moving? What health promotion initiatives can we do? How does this relate to people’s mental health, and how can track and monitor all of this?

And these themes have continued to this day regarding how we can use physical activity and structured exercise programs to help people move and improve clinical outcomes.

More recently in this area, we are doing large clinical trials on people in the community with severe mental illnesses—mainly psychotic disorders including schizophrenia and bipolar disorder—and looking at how we can increase physical activity in this population compared to usual care and how this relates to their recovery.

We are doing trials looking at exercise as an intervention in people who are having the onset of subjective memory complaints, but who don't have any objective change in that cognitive function. And we are looking at how this relates to the gut microbiome and how this relates to hippocampal neurogenesis. So, we are working with basic lab scientists and looking at fecal samples and hippocampal neurogenesis.

The other focus has been on the physical health of those people who use mental health services. For example, one of the early projects we did was looking at the risk of osteoporosis, and some of that work continues today. We looked at people’s mental health conditions and what medications they were taking, and we have a DEXA scanner to measure people’s bone density. What was not really well-known at the time we started and what we have found is this population has a high risk of osteopenia and osteoporosis.

Similarly, we are also looking at pain in people with mental illness and how this may present later for people with mental health conditions and may be more advanced, and asking how we can understand and address this.

Young: So, you have played a really large role in understanding many of the physical health complications that someone with mental illness faces, which I think was and still is underrecognized. In addition to shedding light on this physical health gap, your research has also looked at how physical activity can be used an intervention for mental illness. It’s really 2 distinct important research topics that are connected.

As a clinician interested in lifestyle interventions, what I hear from a lot from other clinicians with similar interests is how hard it can be to get buy-in from their health system, or where they are working, in terms of seeing this as an important area and getting funding. Has this been a struggle for you at all? And if so, what are some ways you have dealt with this?

Stubbs: Yes, it has been hard. And there are a number of different reasons it’s difficult. I think one of the most obvious reasons is there was not a very good evidence base for a long time, particularly in the context of mental illness and mental health, and it has been seen as a luxury and not really part of people’s core, standard treatment.

For example, people are given medications, which really helps their mental health, but those medications may make someone more tired and more sedentary as well as impact their metabolic health. In my view, we have a responsibility to help people ameliorate those risks. And, even in the absence of that medication, we see some of those same changes just related to the mental illness itself, so we have a responsibility to help these people lead a healthy lifestyle, which wasn’t recognized for a long time.

But what I found really helpful is, thankfully, there is lots of evidence and really robust trials outside of the mental health population that show the efficacy of physical activity in reducing some of the risks we are really concerned about in those with mental health issues. Cardiovascular disease is an obvious example, as it is a leading cause of early deaths in people who use mental health services, which is now virtually unequivocally accepted. And we have good evidence that exercise can impact these conditions in the general population.

One study I often refer to was published in The British Medical Journal (BMJ) by John Ioannidis, MD, from Stanford University. He showed quite some time ago that physical activity in the general population can be just as effective as some of the common pharmacological agents for cardiovascular disease when he compared them observationally. Since then, they have done network meta-analysis looking at theoretical comparisons of exercise and some of the common pharmacological agents for the prevention and management of cardiovascular disease and death.1

So, we know from observational data as well as trial data that exercise can really help with a primary concern in our patients. Now, do we need to go and do similar massive clinical trials specifically in people with severe mental illness? Probably not. Waiting for that could be perceived as being potentially unethical.

And cardiovascular health is just one area that can improve. We are not even talking about people’s ability to function or their mental health symptoms. But I have found that it has been really good leverage to rely on the expansive evidence outside of mental health services to demonstrate the potential relevance of physical activity within mental health services.

Young: What I find difficult in the mental health field sometimes is separating mental wellness and the wellness industry from mental illness and mental health care for people. I think we are getting to a place where many people are able to say, “Yeah, I can see how exercise makes me feel better or makes me feel more energized as a general wellness tool.”

But maybe we are still seeing some skepticism when it comes to exercise being useful in a more pathological or clinical state. What are some of the findings when it comes to research of physical activity and exercise in mental illness?

Stubbs: I think that's an important distinction to make just before I get into the nuts and bolts of the research. Physical activity is good for the general population. It does help to make people feel good and healthy. The evidence base from large randomized controlled trials unequivocally says, particularly in major depressive disorder (MDD) and depressive symptoms (less so in bipolar disorder, and not because of lack of evidence, but because of lack of research), that exercise can improve people's mental health symptoms, including quite profound symptoms.

Let’s focus on 3 different conditions we treat in mental health care: depression, schizophrenia, and bipolar disorder. First, let’s consider people with MDD and depressive symptoms. Large-scale, randomized controlled trials and big meta-analyses have shown that when you add exercise to usual care, you get significant and clinically meaningful reductions in people's depressive symptoms. This is consistently shown, whether the physical activity be yoga, resistance training, or aerobic training.

One of the trials I often like to refer people to who may be a bit skeptical and may be interested in our research methods or who may have some reasonable concerns (such as with the short follow-up in many trials or the relatively small size of trials) is a study done by Mats Hallgren, PhD, of the Karolinska Institute. His trial studied more than 300 people with MDD in the community using a 3-armed randomized controlled trial with a 12-month follow-up. So, it had a large number and a long follow-up.

In this randomized controlled trial, they compared 3 groups: usual care, clinician-supported internet-based cognitive behavioral therapy (CBT), and supervised group exercise. What they found is that exercise and CBT were just as effective as each other in terms of reducing depressive symptoms. There was no clinically meaningful difference between them, and they were both better than usual care.2

And then, if I circle back around to a common comorbidity in depression, cardiovascular disease, and the knowledge that inflammation and cardiovascular health are issues in depression, exercise can really be a double win in depression.

There is also evidence for schizophrenia. There is not nearly as much evidence as depression, but there is evidence. And, like depression, there is evidence on a variety of activities, including aerobic, yoga, and growing evidence around resistance and strength. These studies—a mixture of observational trials and randomized controlled trials—show that physical activity can improve people’s psychotic symptoms, overall functioning, quality of life, as well as various other metrics, including cardiorespiratory fitness.

The one area where evidence is not so clear on is bipolar disorder. This is not so much because of negative evidence, but because of lack of research. There just has not been much funding in this particular area.

And I’m not saying, nor have I ever said that, physical activity will work for everyone with mental health issues. That is nonsense. But, as part of an overarching plan to improve symptoms as well as physical health in someone with mental disorders, it can definitely play an important role.

Young: Thank you. This is a great overview. And you answered what I was going to ask next, which is: Which conditions have the most research? It sounds like depression has the most evidence, not just because it has a lot of positive evidence, but also because it is the most researched condition in general. Other conditions, like bipolar disorder, might be just less well-researched at this time, but it is not that there are trials that are negative or do not show any response to exercise. Is that correct?

Stubbs: Yes. When you look at the aggregate, and that's why meta-analysis is good—because it’s an aggregate of all the controlled trials and seeing what the evidence says in an unbiased way—what the aggregate evidence says is that there is an overall positive effect that exercise has on symptoms of depression and schizophrenia as well as physical health markers and overall quality of life.

Young: What about in anxiety? That is a very common disorder. Is there evidence for physical activity as an intervention to help clinical anxiety? I know there is, again, in more the general wellness sense, support for reducing stress. But in a more clinical sense, in terms of panic disorder or generalized anxiety, is there evidence for exercise in those?

Stubbs: Yes, there is. And this probably sits alongside schizophrenia in terms of the quality of the evidence base. So, there is a lot of evidence in the form of observational data on the future risk of anxiety symptoms or anxiety disorders. And it could be any of the ones you mentioned, including posttraumatic stress disorder (PTSD), showing that people who are more active in their general life have a reduced risk of those anxiety or stressor-related disorders in the future.

When we look at exercise as an intervention or management approach, there is some evidence that resistance exercise and aerobic exercise can reduce symptoms in people with elevated anxiety symptoms using a screening tool. And there's a few studies looking at people who've got a confirmed disorder, such as a DSM-V or ICD-10 diagnosis. The field is quite messy in terms of the classification of different disorders.

So, there are a few studies for obsessive compulsive disorder (OCD), PTSD, panic disorder, but there are not lots for each. In terms of anxiety or stressor-related disorders, there is probably the most research on PTSD, and there is a good evidence base within that.

Young: So, when it comes to using exercise as an intervention, one question clinicians often ask then is: What does the research say about how often or what type of exercise might work best? If I have someone in my office who has MDD and I want to counsel them on an exercise intervention, what do I say? And maybe the research is a bit young for this type of specificity.

Stubbs: So, while the evidence base may be a bit premature, I would refer to a great paper that was published in the BMJ discussing that there are no randomized controlled trials that show parachutes work when evacuating a plane.3 There are none.

But clearly, we are not going to wait for a randomized controlled trial. And I’m having a bit of fun with this—I’m not saying exercise is like a parachute. I’m just drawing an analogy that we don’t have randomized controlled trials for everything.

But we know from the World Health Organization, the American College of Sports Medicine, and others that physical activity is generally very, very safe even in people with conditions like heart failure, and the risks of not doing it and not promoting it in almost all cases is higher than people doing it. So, while the evidence base is improving, I don't think we need to necessarily wait 10, 20, or 30 years until we start recommending it for people.

Young: So, just to summarize so far: Exercise may play a role in improving the physical health of people with mental illness and to target some of their symptoms, with some of the best evidence being in depression just because of the number of trials that have been done in depression. We don't yet have the specific evidence to say how often or what type might be best for each disorder, but the existing evidence suggests that we shouldn't wait for that to recommend patients start some sort of exercise program.

Stubbs: Yes. Around type, generally aerobic is accepted, and that could be anything that gets your heart rate up and going. It could be running, playing tennis—whatever you as an individual want to do. And resistance training is an emerging evidence base.

And, in terms of how often, we're just leaning toward what is in general recommendations of the World Health Organization, which is trying to accumulate up to 150 minutes of vigorous physical activity per week or up to 300 minutes of moderate-intensity activity, with starting small and doing something you enjoy as the main message.

Young: There are so many activities that someone could choose from. And I imagine this may make studying physical activity even more challenging. You can exercise inside or out in nature. You can do resistance or aerobic training. There are so many different factors to account for in studying this. I wonder if we have become a little pill-centric in medicine because the gold standard trials really lend themselves better to pill studies. For example, you can blind pills—you obviously can't blind exercise.

While you are on this path to laying a foundation that this is a legitimate intervention (as you mentioned, it has been compared to other well-accepted interventions), what are some of the research barriers you have come across?

Stubbs: I have been involved in double-blind randomized control trials, not with exercise for obvious reasons we will discuss, but with medication. And yes, doing a medication trial, comparatively speaking, not factoring in the safety risks, is very easy in terms of what you do. You have certain doses, and you can have somebody whose only involvement in the study is overseeing the administration of that. With exercise, or other lifestyle interventions like nutrition or even talk therapy to an extent, it is very difficult to blind the participant as well as the assessor. So, it’s a bit of an unfair yardstick to compare.

The only way we have thought about potentially doing this is something perhaps boring, like going on a specialized exercise bike with an internal randomization based on the participant’s number. A person could be randomized to either high frequency and output or low frequency and output, which would not be known to the participant or the assessor.

Still, you could probably guess quite comfortably if you are the participant which group you are in. Also, this would be more like comparing a therapeutic dose of a medication to a low dose, instead of an inert compound, which is what people typically compare new agents to. With exercise, you are not really able to have an inert dose.

Even outside the study, in people’s daily lives, people are still moving around, and what we have found in our trials is, once people start moving and start talking about exercising, they become more interested in it, even those in the control group. So, it becomes even harder to control and implement.

Those are some of the challenges we face.

Young: I know the science on this is still young, but what are some of the presumptive mechanisms by which exercise could theoretically improve mental health?

Stubbs: This is really an exciting area for exercise, particularly looking at brain health and mental health, because there is unequivocal evidence, again outside of a psychiatric context, that exercise promotes healthy aging, impacts telomere length, lengthens lifespans, and improves people’s physical capacity. So, the effect is really multifaceted and multisystem.

But, if we want to look at individual systems, we see an impact of exercise on many levels within the brain. First, we can look at specific regions of the brain. One area that has been of great interest for psychiatrists, neuroscientists, and clinicians is the hippocampus. The hippocampus is reduced in many people with mental illness, including schizophrenia, depression, and bipolar disorder. This is ubiquitously decreased, and then decreases over the time of many people's illnesses.

And it's difficult to alter the size of the hippocampus, butresearch, which has built on observational data over time, has shown that exercise can increase the actual volume of the hippocampus. That is particularly exciting.

Another area of interest is the prefrontal cortex, which has been implicated in a number of different mental disorders as well. And we have also shown that, in the context of depression, people who have greater muscular strength and do more resistance training have less white matter hyperintensities within the brain.

Next, if we look at connectivity in the brain, we see improvement in connectivity between key areas of the brain, some of which I have just mentioned. And if you look at functional magnetic resonance imaging (fMRI), you can see changes as well.

There is a nice study, not within mental health services, that demonstrates a potential impact of exercise on mental health. This was a randomized controlled trial studying the effects of just 10 minutes of lite cycling, mimicking the effects of light exercise, on fMRI compared to sitting still. This study showed that just 10 minutes of lite cycling resulted in significant stimulation in the hippocampus and the prefrontal cortex, and other areas as well.4 Overall, it seems exercise creates immediate effects on the brain as well as effects that take place over time.

And then we know that there are various neurobiological and immunological effects of exercise. For example, growth factors, such as brain-derived neurotrophic factor (BDNF), are released. Muscles are also a great mediator of growth factors as well as immunological factors within the body. So, we get changes in many of inflammatory biomarkers we know are raised in many mental disorders, including depression, schizophrenia, and bipolar disorder. With exercise, we see decreases in these markers, including IL-2, IL-4, IL-6, and c-reactive protein (CRP). We also see stimulation of the endocannabinoid system, which is kind of a pleasure system within the brain.

Additionally, there are psychosocial benefits of exercise—for example, a sense of fulfillment, sense of achievement, and self-efficacy. And a lot of those benefits are the ones people really care about.

So, my kind of subjective interpretation of this is that we are just kind of scratching the surface at the moment and that it is way more complicated; it would be kind of dumbing it down to just relate it to 1 molecule or 1 area of the brain.

Young: One thing that I have struggled with a bit as a clinician who utilizes evidence-based lifestyle interventions in practice is the way evidence-based lifestyle interventions, like what you have presented regarding exercise, can be improperly used.

Sometimes I see it presented in treatment planning as an afterthought and lumped in very generally along with a laundry list of other potential lifestyle items that may or may not have evidence. Sometimes I see the evidence borrowed by others—some who may be against psychiatry or medication in general, and some who promote a “holistic approach,” but in doing so, use a lot of techniques and supplements that lack evidence.

As a result, the good evidence base ends up muddied by things that lack evidence or by overselling the existing evidence. As a researcher, how do you address this so that some clinicians don’t end up viewing exercise with doubt or skepticism because of this murkiness?

Stubbs: This speaks to an unfortunate fact as well as a positive fact. Physical activity and exercise are unregulated. Theoretically, anybody could start doing physical activity and exercise, and providing input to people. It doesn’t require that people have 6 years of training to deliver it to people.

In a way, this is a good thing because a lot of people have access to exercise and there are fewer barriers. But we need a certain level of proficiency and understanding of the core tenants of the evidence base as well as awareness of motivation. All these things can help make sure exercise is delivered in a reliable, credible way.

And the research evidence shows, much in line with that CBT, for instance, there are better outcomes when CBT is delivered by someone with a higher level of knowledge in doing CBT. Similarly, exercise interventions tend to be more effective and have better outcomes when they are delivered by someone who has like a bachelor’s degree or higher in an exercise or related field.

And how do we avoid all of the nonsense out there? It’s difficult. There are a lot of people out there flouting a lot of absolute nonsense and trying to amalgamate exercise with that, and it does have the potential to distract away from the credible science. But we just need to remember that the science is really credible, we have made huge strides in the field, and the implementation is good.

Young: It sounds like one way of navigating the murky waters is to make sure the interventions are delivered by someone with expertise and then, in the future, having practice parameters established by experts that clinicians can reference to guide their interventions.

Stubbs: Absolutely.

Young: Going deeper into how we can implement this—I think most of us in the mental health field would say a collaborative approach would be ideal. If we were able to have a psychiatrist, and a primary care doctor, and someone with expertise in exercise, and another expert in nutrition, all part of 1 team delivering care, that would ideal. But that’s not the reality for most situations.

Any advice you could give to someone in a more typical setting, such as a mental health clinic or private practice, who is looking to use exercise as an intervention without having experts in exercise in their office?

Stubbs: Yeah, I would recommend busy clinicians who may not have the expansive expertise to have a look at some of the available free resources, such as Moving Medicine, which offers free resources on how to have short conversations, using a motivational framework, to assess people’s activity level and help them find an activity that they would enjoy. And I think just doing that and having those conversations coming from a clinician is really powerful and important. And I have certainly seen a benefit from that.

Another helpful resource is the Simple Physical Activity Questionnaire (SIMPAQ), which is a way to understand what people are doing in a regular week. When you combine this with discussing activities people like and helping to dispel some of the myths surrounding physical activity (like the thought that you need to do something intense, like CrossFit), it can be really impactful.

Overall, when clinicians make some time in an appointment to talk about exercise and recommend it, and just become familiar with some of the evidence we just talked about to share with people, the recommendations to people do not need to be particularly complicated. You can help people start off small and build up from that.

Young: Thank you. We know that access to care is an issue. And this seems to be global, and not just in certain areas or countries. And when you see favorable comparisons of exercise with other interventions (like you mentioned CBT), it seems exercise could be a powerful way to increase access to evidence-based tools that can improve mental health. It also seems this type of intervention, if used correctly, could help improve health equity for those with mental health conditions.

Stubbs: I think there are a lot of ways that physical activity and exercise can be used in that way. So, if we think about physical activity and exercise as a form of inclusion and social support, it is a really powerful way for people to be brought together. Again, this does not need to be an aerobics class or something like that—it could be in a walking group. But this notion of inclusion is really powerful and important.

If we are talking about equity of care, and, more broadly, some of the other areas which we've talked about (including some of the leading causes of premature mortality), it’s not only giving people tools, but also giving people an opportunity build up trust with clinicians. I have seen wonderful things happen in walking groups.

People talk about substantial things that are going on for them in terms of their mental and physical health, and have an opportunity to build up trust, build rapport, and have open conversations to address some of the inequities they face.

Young: Thank you so much for sharing your insights on this topic. I just want to close out by asking you about the documentary you had come out this past winter called Mind Games. You are the scientist in this documentary, studying and following the impact of exercise on individuals who compete on high levels in fields such as chess, mahjong, memory games, and eSports, which require a high mental capacity.

What are the implications of what you found in terms of the positive impact of exercise on mental capacity in these “mind athletes” for the general population as well for individuals receiving mental health care?

Stubbs: I think exercise and physical activity can be a really powerful way to help people just to function and feel a bit better in the moment. And, looking at the people in the film who were functioning quite well mentally but just very sedentary, and seeing how exercise impacted concentration, memory, and focus in them, you can see how this would be a very key message to the general population.

I think this could translate really well into mental health. There are many patients struggling with cognitive functions, including memory and cognitive inflexibility, and exercise may be a nice addition for people to improve these.

Young: Yes, brain fog is a very common symptom that people with depression complain about, often even after some of their other symptoms improve. The data from this film seems to suggest that exercise could impact that functioning.

Stubbs: Yeah, absolutely. So, lots more to do and study.

Dr Young is an adult psychiatrist in private practice and adjunct professor of psychiatry at NYU Grossman School of Medicine, where she teaches a class on lifestyle medicine in mental health care. Additionally, Dr Young is a medical reviewer and regular contributor at Everyday Health.

The opinions expressed in the interviews are those of the participants and do not necessarily reflect the opinions of Psychiatric Times®.


1. Naci H, Ioannidis JP. Comparative effectiveness of exercise and drug interventions on mortality outcomes: metaepidemiological studyBr J Sports Med. 2015;49(21):1414-1422.

2. Hallgren M, Helgadóttir B, Herring MP, et al. Exercise and internet-based cognitive-behavioural therapy for depression: multicentre randomised controlled trial with 12-month follow-upBr J Psychiatry. 2016;209(5):414-420.

3. Smith GC, Pell JP. Parachute use to prevent death and major trauma related to gravitational challenge: systematic review of randomised controlled trialsInt J Prosthodont. 2006;19(2):126-128.

4. Suwabe K, Byun K, Hyodo K, et al. Rapid stimulation of human dentate gyrus function with acute mild exerciseProc Natl Acad Sci U S A. 2018;115(41):10487-10492.

© 2024 MJH Life Sciences

All rights reserved.