Physical activity as an intervention in severe mental illness

Summary

Physical activity is a modifiable risk factor for several physical and mental health conditions. It is well established that people with severe mental illness have increased risk of physical health complications, particularly cardiovascular disease. They are also more likely to be physically inactive, contributing to the elevated cardiovascular and metabolic risks, which are further compounded by antipsychotic medication use. Interventions involving physical activity are a relatively low risk and accessible way of reducing physical health problems and weight in people with severe mental illness. They also have wider benefits for mental health symptoms and quality of life. However, many barriers still exist to the widespread implementation of physical activity interventions in the treatment of severe mental illness. A more concerted effort is needed to facilitate their translation into routine practice and to increase adherence to activity interventions.

LEARNING OBJECTIVES

After reading this article you will be able to:

• • understand why physical activity is clinically important for people with severe mental illness

• • recognise the possible barriers to and facilitators of engagement in physical activity by people with severe mental illness

• • consider the next steps for commissioners, researchers and practitioners in this area.

Physical activity refers to any bodily movement that results in some degree of energy expenditure (Caspersen 1985). The importance of physical activity for treating and preventing many chronic diseases is well established. For example, population-level research shows that higher levels of physical activity are associated with lower incidence of cardiovascular disease, diabetes, cancer and common mental health disorders (Wilmot 2012; Kyu 2016; Bennett 2017; Schuch 2018, 2019). A recent umbrella review of 27 systematic reviews found that physical activity is effective for treating symptoms of depression, anxiety and stress disorders, schizophrenia, and alcohol or substance use disorders (Ashdown-Franks 2021). These results are primarily based on data from randomised controlled trials (RCTs), with a total of 152 studies included in the review.

The importance of physical activity to public health is exemplified by the widespread adoption of standardised physical activity guidelines (World Health Organization 2015). In the UK, the Chief Medical Officer's report recommends that adults engage in at least 150 min of moderate-intensity aerobic activity or 75 min of vigorous-intensity aerobic activity per week (Department of Health and Social Care 2019). Moderate-intensity activity might include brisk walking or riding a bicycle, whereas vigorous-intensity activity refers to running or sports such as football and athletics.

Among other health benefits, aerobic activity improves cardiorespiratory fitness and helps with weight management. Cardiorespiratory fitness reflects the efficiency of the circulatory and cardiovascular system during activity in adults (Blair 1996). Low cardiorespiratory fitness is associated with a greater risk of several chronic conditions, cardiovascular disease and common mental disorders (Lee 2011; Kandola 2019). Guidelines also recommend engaging in strength training at least twice per week. Strength training refers to exercises that strengthen muscles, such as using resistance bands or lifting weights. These exercises improve muscular fitness, which is another important marker of chronic health conditions, common mental disorders and cognitive performance (Celis-Morales 2017; Wu 2017; Firth 2018c; Kandola 2020).

Physical inactivity refers to not meeting national physical activity guidelines (Kohl 2012). Physical inactivity is widely recognised as different from sedentary behaviour, even though they both sit on the same activity spectrum. Sedentary behaviour refers to any activity in a sitting, lying or reclining position with very low energy expenditure (Tremblay 2017). Driving or watching television while seated are examples of sedentary behaviours. Physical activity guidelines in the UK recently included recommendations for reducing sedentary behaviour, where possible (Department of Health and Social Care 2019).

Despite sustained global health campaigns, physical activity levels in the general population are still lower than recommended. Worldwide, over 27% of adults (around 1.4 billion individuals) and 81% of young people aged 11 to 17 were estimated to not be achieving national physical activity guidelines in 2016, with little improvement since 2001 (Guthold 2018, 2020). Global cardiorespiratory fitness appears to have decreased in adults and children over the past few decades (Lamoureux 2019; Tomkinson 2019). Estimates suggest that around 5 million deaths worldwide are directly attributable to physical inactivity each year (Lee 2012).

Reducing physical inactivity has the potential to improve population physical and mental health. Approaches for increasing physical activity are generally safe, have few to no side-effects, are widely accessible and are cost-effective in the population (Roux 2008). The following sections will explore how and why people with severe mental illness (SMI) represent a high-risk group that would benefit from interventions to increase activity. We use SMI in this article to indicate a diagnosis of any bipolar disorder, non-affective psychoses or major depressive disorder. However, much of the current literature focuses on people with a schizophrenia diagnosis specifically.

Physical activity and SMI

People with SMI are generally less physically active than the general population. A recent meta-analysis of 69 studies (n = 35 684) found that people with SMI engaged in an average of 38.4 min of moderate to vigorous physical activity per day, around 13 min fewer than comparison groups without SMI (Vancampfort 2017b). These findings are mostly from self-report questionnaires, and data from the UK Biobank suggest that these measures likely overestimate physical activity in SMI populations (Firth 2018d). Data from activity monitors in 1078 people with schizophrenia from the UK Biobank suggest that their total physical activity is 80% lower than that of the general population. There are also differences between diagnoses. For example, people diagnosed with bipolar disorder had higher sedentary behaviour but higher moderate to vigorous physical activity per day than those reporting a diagnosis of schizophrenia (Vancampfort 2017b).

The lower activity levels in SMI populations have implications for physical health (Firth 2019). People with SMI have lower levels of cardiorespiratory fitness (g = −1.01) than healthy controls, with no difference between those with bipolar disorder or schizophrenia (Vancampfort 2017a). Physical inactivity and high levels of sedentary behaviour tend to cluster with other behavioural risk factors for physical ill health in SMI populations, such as smoking (Jackson 2015) and poor diet (Firth 2018a). These unhealthy lifestyle factors likely contribute to the 1.4–2 times increased risk of cardiovascular and metabolic disease in people with SMI compared with the general population (Firth 2019). Data from primary care records indicate that the risk of premature mortality is elevated in people with SMI (HR = 1.79 for bipolar disorder and HR = 2.08 for schizophrenia) and this mortality gap from the general population appears to be widening (Hayes 2017). The development of comorbid physical health conditions in people with SMI creates an enormous healthcare challenge for the UK's National Health Service (NHS) (Naylor 2016). As set out in its long-term plan, the NHS is committed to improving physical health outcomes for people with SMI and reducing this mortality gap (NHS England 2019).

Physical inactivity is a major component of an unhealthy lifestyle that is likely to contribute to the physical health risks associated with SMI. It may also affect symptoms of mental illness. Physical activity conveys a range of mental health benefits for people with SMI, including reducing depression and anxiety symptoms and improving cognitive functioning, sleep, quality of life and social adjustment (Rosenbaum 2014; Firth 2015, 2017; Ashdown-Franks 2020). Increasing physical activity and reducing sedentary behaviour in SMI populations may have transdiagnostic benefits that improve psychiatric symptoms and well-being while reducing the risk of physical health complications over time.

The effectiveness of physical activity interventions in SMI populations

The British Association for Psychopharmacology (BAP) recommends the use of lifestyle interventions for managing weight and cardiovascular disease risk in treating SMI (Cooper 2016). A recent European Psychiatric Association statement highlighted that interventions involving physical activity also have wider physical and mental health benefits in SMI populations, and there is an increasing focus on increasing their acceptability and effectiveness (Stubbs 2018). These interventions use various types of physical activity, such as stationary cycling, walking, resistance training, jogging and group-based aerobics sessions. Protocols are sometimes based on national activity guidelines or specific recommendations from authoritative bodies, such as the American College of Sports Medicine. But the content of these interventions may be perceived as unrealistically high for people with low activity. Some interventions consist of multiple short bouts of activity, such as two or three bouts of 10 min per day (Linke 2011). These approaches are less intense and allow greater flexibility to incorporate activity throughout the day.

Mental health

There is evidence that these physical activity interventions can improve mental health outcomes. A meta-analysis of 20 studies suggested that aerobic exercise interventions involving at least 90 min of moderate to vigorous physical activity per week can reduce positive (s.m.d. = −0.54, 95% CI −0.95 to −0.13) and negative (s.m.d. = −0.44, 95% CI −0.78 to −0.09) symptoms in people with schizophrenia (Firth 2015). The exercise intensities for these studies are typically based on individual factors, such as baseline fitness, activity levels, heart rate and other health factors. For example, exercises may be set at 40–60% of a participant's maximum capacity. A systematic review of six studies found that strength training also reduced positive and negative symptoms in people with schizophrenia (Keller-Varady 2018). The strength training interventions in these trials typically focused on resistance exercises for large muscle groups, such as the legs, chest, shoulders and back. Some interventions also included an aerobic component, such as running or walking. Interventions were delivered by a study investigator or trained exercise professional. Another meta-analysis of 10 studies involving with 20–60 min of aerobic exercise 2–4 times per week were associated with improved global cognitive function (g = 0.33, 95% CI = 0.13–0.53) in people with schizophrenia (Firth 2017).

Far fewer studies exist that investigate the relationship between physical activity interventions and psychiatric symptoms in people with bipolar disorder (Ashdown-Franks 2020). One systematic review with 31 studies of varying design found that higher exercise levels were associated with fewer depressive symptoms in people with bipolar disorder (Melo 2016). Associations with mania symptoms were inconsistent. However, this review included observational studies and some interventional studies, but no RCTs. Evidence for the utility of exercise in bipolar disorder is typically based on extrapolated evidence from trials in people with unipolar depression (Thomson 2015). For example, several meta-analyses of RCTs using physical activity interventions have consistently found moderate to large effect sizes for reducing symptoms of depression and anxiety disorders (Bridle 2012; Herring 2012; Cooney 2013; Josefsson 2014; Kvam 2016; Schuch 2016; Stubbs 2017). The role of exercise in bipolar disorder could be more complex than in unipolar depression. For example, preliminary evidence suggests that exercise can help people with bipolar disorder manage their excess energy, but may exacerbate manic and hypomanic symptoms (Thomson 2015).

Physical health and quality of life

There is evidence that exercise can also improve physical health outcomes in people with SMI. A meta-analysis of four RCTs suggests that around 12 weeks of aerobic exercise 2–3 times per week can improve cardiorespiratory fitness in people with schizophrenia (g = 0.43, 95% CI = 0.05 to 0.82) (Vancampfort 2015). A systematic review of ten studies found that walking-based interventions can produce small reductions in body fat or body mass index (BMI) in people with schizophrenia (Soundy 2014b). Another systematic review of seven RCTs found that combined aerobic and resistance training for around 95 min twice per week was effective for improving strength and overall mental health in people with schizophrenia (Martin 2017). Although less research is available for bipolar disorder, regular aerobic exercise in people with mild to severe depression improves various aspects of physical health, including cardiorespiratory fitness, body fat percentage and blood glucose levels (de Souza Moura 2015; Vancampfort 2017a).

Physical activity interventions also provide broader of range benefits to people with SMI. For example, two RCTs found that at least 120 min of moderate to vigorous physical activity per week improved quality of life and reduced disability (Firth 2015). A meta-analysis of eight RCTs using various types of exercise found large improvements in sleep (g = 0.73, 95% CI 0.18–1.28) in people with SMI (Lederman 2019).

Aerobic exercise can also stimulate a multitude of changes in the brain that are relevant to SMI (Kandola 2016). For example, aerobic exercise can increase the volume (Firth 2018b) and functioning of the hippocampus (Erickson 2009), including in people with schizophrenia (Pajonk 2010; Woodward 2018). These improvements in hippocampal integrity may contribute to reductions in overall symptoms and cognitive deficits in people with schizophrenia (Kandola 2016).

Evidence from 160 RCTs also suggests that both aerobic and strength training over a median of 12 weeks can reduce several pro-inflammatory markers in predominantly healthy participants (Lin 2015). There is also some early evidence of this in psychiatric populations (Euteneuer 2017; Lavebratt 2017). Pro-inflammatory cytokine levels are elevated in many people with SMI and could contribute to both psychiatric symptoms and physical health risks (Khandaker 2015; Rosenblat 2016).

Safety, dose–response and cost-effectiveness

Interventions to increase activity are a relatively low-risk approach for improving the mental and physical health of people with SMI. Exercise interventions do not substantially increase the risk of adverse events in the treatment of mental illness, although more consistent data on safety would be beneficial (Czosnek 2019). Despite promising initial findings, there remains a lack of large-scale RCTs of activity interventions in SMI populations (Vancampfort 2019).

There remain unanswered questions regarding dose–response (frequency, intensity and duration of intervention) and safety. For example, people with SMI may experience balance or other musculoskeletal problems that increase the risk of falling during exercise (Hamera 2010). Tailored exercise programmes that incorporate balance and coordination exercises could be a useful method of addressing these concerns (Vancampfort 2016a). The cardiovascular and metabolic benefits of exercise likely outweigh the risk of falls, which only represent a concern for a minority of people with SMI (Firth 2016).

Achieving 150 min of moderate to vigorous physical activity is a viable target in many populations, but different targets may be necessary for people with SMI (Vancampfort 2015a). For example, people with SMI may benefit from engaging in frequent bouts of light activity throughout the day to break up extended periods of sitting. Large RCTs are necessary to compare the relative benefits of different intensities, durations and frequencies of activity in people with SMI.

Although community-based exercise interventions are increasingly demonstrated as being cost-effective in non-SMI populations (Kelly 2021), similar data in SMI groups are lacking (Czosnek 2019).

Barriers and facilitators for implementation

Although there is a growing evidence base for their transdiagnostic benefits, research on the clinical implementation of physical activity interventions for people with SMI is lacking (Vancampfort 2015a; Lederman 2017). However, available research has indicated a number of barriers and facilitators in implementation (Box 1).

BOX 1 Engagement in physical activity interventions by people with severe mental illness: barriers and facilitators

Barriers

• • Medication side-effects

• • Co-existing physical health conditions

• • Lack of motivation

• • Illness-related symptoms (e.g. negative symptoms of psychosis, social anxiety, low self-esteem)

• • Lack of access to a suitable environment

Facilitators

• • Use of a qualified professional to deliver the activity

• • Social support of peers and healthcare professionals

• • Providing information about the activity

• • Personalised goal setting and activity plans

Barriers

Modifying physical activity is a challenge in any population and requires varying approaches for different people (Heath 2012). There are several physical, psychosocial and environmental barriers to increasing physical activity and decreasing sedentary behaviour in people with SMI (Soundy 2014a; Firth 2016b). These factors likely underlie the substantial drop-out rates from activity interventions in SMI populations, which are estimated to be around 27% in people with schizophrenia (Vancampfort 2015b). Nevertheless, this is within the range of drop-out rates for activity interventions found in other populations (18–35%), such as obese individuals (Linke 2011).

Medication side-effects and co-existing physical health conditions are examples of possible physical barriers to increasing activity in people with SMI. Physical activity is typically lower in adolescents and adults with SMI who are taking second-generation antipsychotic medications (Cuerda 2014; Vancampfort 2016b; Perez-Cruzado 2018). A qualitative study of 151 physical therapists working in psychiatric services suggests that medication side-effects are among the most commonly cited barriers to increasing activity in people with schizophrenia (Soundy 2014a). However, it remains unclear whether the medication or other factors explain the lower activity (Cuerda 2014).

Nurses and other healthcare staff also commonly report a lack of motivation and negative symptoms as common barriers to implementing physical activity interventions in people with SMI (Harding 2013; Robson 2013; Soundy 2014a). Related symptoms may also affect motivation. For example, social anxiety may reduce the likelihood of exercising outside or in a gym. Other motivational factors may relate to SMI treatment. For example, medication side-effects can affect motivation to exercise through sedation and neuromuscular side-effects (Soundy 2014a; Firth 2016). Understanding factors that increase motivation for activity in people with SMI is an essential but understudied area. A systematic review of 79 studies in SMI populations was able to identify only one that included motivation for physical activity as its main outcome (Farholm 2016). The review suggests that interventional studies commonly use motivational interviewing and goal-setting techniques without directly evaluating their effectiveness. Related factors include a reduced sense of self-esteem and confidence in people with SMI that limit motivation to increase activity (Soundy 2014a).

Facilitators

A number of factors facilitate the successful implementation of activity interventions in SMI populations. A meta-analysis of 19 activity interventions in people with schizophrenia found that having a qualified professional delivering the intervention moderated drop-out (Vancampfort 2015b). For example, including a physical therapist or exercise physiologist to lead the activity may increase adherence. A qualified exercise professional may increase the quality and enjoyment of exercise sessions. Social support is another facilitator of activity in SMI populations (Soundy 2014a ). Peers and healthcare professionals can play an active role in fostering a supportive and encouraging environment to encourage adherence. These factors may help with psychosocial barriers of self-esteem and motivation.

The provision of information about the activity is another possible facilitator of activity (Soundy 2014a). People with SMI may lack knowledge about the benefits of the activity or how to perform activities that most interest them (Matthews 2018). Providing this information contributes to a supportive environment and promotes self-efficacy, which can lead to increased activity. Goal setting and creating personalised activity plans are also likely to provide complementary benefits for promoting sustainable improvements in activity (McEwan 2016). For example, the PRIMROSE trial for reducing cardiovascular risk in SMI groups obtained good adherence to the intervention through the use of behavioural change techniques that included goal setting, action plans, progress tracking, providing positive feedback and dealing with setbacks (Osborn 2018).

Future directions

Adherence

Implementing physical activity interventions in the treatment of SMI will be an important step in achieving the NHS's long-term goal of improving physical health outcomes for SMI groups (National Health Service 2019). Studies of activity in SMI populations are increasingly moving beyond demonstrating efficacy to establishing effectiveness in real-world settings, including the use of behavioural change techniques (Farholm 2016). Increasing adherence to interventions will be key to improving the quality of physical healthcare in SMI populations in accordance with NICE guidelines (National Institute for Health and Care Excellence 2014). Further research must directly examine successful methods for designing and implementing activity interventions in SMI populations, particularly in people with bipolar disorder, where there is a paucity of work. The field may benefit from developing novel methods for motivating people with SMI to increase their activity. For example, ‘exergaming’ interventions target motivation through enjoyable video games that promote exercise and bodily movements. An exergaming study involving 16 out-patients with schizophrenia found good acceptability, feasibility and an attrition rate of around 19% (Campos 2015). However, no significant improvements in fitness, mobility or symptoms resulted from the intervention.

A recent trial in people with SMI also integrated self-management concepts and social cognitive theory to promote adherence to a cardiovascular risk reduction intervention delivered through community mental health programmes (Daumit 2020). For example, they used motivational interviewing and solution-focused therapy sessions tailored to minimise the impact of memory and executive functioning deficits. The intervention also included a points system to encourage participants to attend sessions and achieve goals. Further interventions in community settings that utilise novel techniques to promote adherence would be beneficial.

It will also be necessary to take a practical and realistic approach to developing better activity interventions in SMI populations. Maintaining good activity levels across the life course is necessary to improve long-term physical health outcomes. But activity interventions typically last between 12 and 24 weeks in SMI populations and follow-up periods rarely exceed 36 weeks (Firth 2017). It remains unclear whether increased activity levels during the study period reflect sustainable changes in activity habits.

Dose

Despite a growing number of studies in the area, there is still no consensus regarding the type, frequency, duration and intensity of activity most suitable for treating people with SMI (Cooper 2016; Lederman 2017). A recent systematic review of 32 studies found inconsistent and low-quality evidence that interventions can increase activity or reduce sedentary behaviour in people with SMI (Ashdown-Franks 2018). Large-scale RCTs are necessary to establish the type, frequency, duration and intensity of activity that will be most beneficial for people with SMI.

Safety, cost and cost-effectiveness

More research should focus on establishing safety and outlining the cost of activity interventions in mental health settings (Czosnek 2019). Although developing personalised activity plans and hiring exercise physiologists will likely promote adherence, they will also increase the cost. Managing cost-effectiveness will be essential, as activity interventions are most suitable as part of complex, multifactorial treatments approaches to mental illness (Firth 2019). Developing personalised approaches also requires accurate assessments of baseline activity and fitness. The use of devices to monitor activity levels could be prohibitively expensive in many clinical settings, where self-report measures would be more feasible. However, to our knowledge just one physical activity questionnaire has been specifically designed and validated for recording activity in people with mental illness (Rosenbaum 2020a).

A recent meta-review of 33 reviews found no available data on the cost of implementing activity interventions in mental health settings (Czosnek 2019). The long-term benefits of activity may outweigh the immediate costs of interventions. For example, physical activity may reduce cardiovascular risk, potentially offsetting future treatment costs for cardiovascular disease. More careful consideration of cost will be necessary to translate research into practice (Czosnek 2019).

Introducing lifestyle changes during routine care

The recent Lancet Psychiatry Commission on physical health in mental illness called for more routine implementation of lifestyle approaches in mental healthcare, including the involvement of physical therapists and exercise specialists (Firth 2019). This integrative approach to routine care reflects a need to manage the physical health of people with SMI more effectively, both with and without physical comorbidities. A greater integration of lifestyle approaches will contribute to the prevention or reduction of physical health problems over time and help to close the mortality gap between those with SMI and the general population. It will also help to mitigate some long-term medication side-effects, such as weight gain. Lifestyle approaches aiming to counteract weight gain in people with SMI are most effective when combining exercise with dietary changes, rather than exercise alone (Gurusamy 2018). Implementing these lifestyle approaches as part of routine care in SMI groups may require changing staff perspectives and culture, such as by offering brief lifestyle interventions to clinical and non-clinical staff (Rosenbaum 2020b). Practitioners’ lack of knowledge of physical activity interventions is a common barrier to their prescription in mental health services (Way 2018).

Systemic changes to reduce health inequalities

There has also been an increase in local and national initiatives to highlight systemic changes that facilitate the integration of physical and mental healthcare. For example, the World Health Organization has published guidance on changes at the individual, health-system and societal level for reducing health inequalities in SMI groups (World Health Organization 2017). Similar reports from Public Health England highlight local actions necessary to reduce these inequalities, including addressing the social determinants of poor health and early detection and intervention (Public Health England 2018). Co-producing these initiatives with SMI groups will be an important strategy for their successful implementation (Deenik 2020).

Long-term evaluation

It remains unclear to what extent these new ideas and best practice guidelines will produce meaningful improvements for SMI groups. However, the growing recognition of these issues may promote a wider-scale adoption and implementation of strategies that do result in sustainable improvements. Current interventions are based on theory and behavioural science and the BAP guidance highlights the limited evidence on long-term maintenance of lifestyle interventions (Cooper 2016). Demonstrating sustainable reductions in physical health risks may require further investment in long-term approaches with several years between follow-up assessments. The long-term maintenance of these interventions may also require environmental changes that facilitate greater incidental physical activity, such as accessible and comfortable walking and cycling areas (Vancampfort 2013).

Conclusions

The challenges of implementing successful activity interventions in SMI populations does not detract from their importance. People with SMI are at an elevated risk of physical health complications over time, including obesity, cardiovascular and metabolic disease, and premature mortality. Low physical activity, low fitness and high levels of sedentary behaviour are modifiable risk factors for many of these health complications. These risk factors are highly prevalent in SMI populations and likely compounded by other risk factors, including smoking and poor diet.

Activity interventions for people with SMI are an essential adjunctive to treatment. Promoting activity will reduce the risk of these physical health complications. It will also have a positive impact on mental health symptoms and have wider benefits, including improving sleep quality and quality of life. However, large-scale RCTs will be necessary to provide further information on dose–response and other fundamental aspects of treatment. There is also a paucity of research focusing on translation of trial evidence into practice. Details on practical considerations such as cost and safety are still lacking. One of the greatest challenges relates to improving adherence to activity in people with SMI. A more direct approach is necessary to identify methods to address poor adherence, such as developing novel ways of increasing motivation and enjoyment of activity.