WHAT IS ALREADY KNOWN ON THIS TOPIC

A recognised priority in mental healthcare is to increase physical activity and decrease sedentary behaviour in people with psychosis but current interventions have low uptake and high attrition rates. A psychosis-specific framework for increasing physical activity, which has been informed by patients, carers and staff, may lead to better interventions.

WHAT THIS STUDY ADDS

This study provides a new framework to understand physical activity in patients with psychosis. Five factors are important: ‘Purpose’ (personally meaningful reasons to move); ‘Predictions’ (internal expectations for movement and physical activity based on past experiences and projections of the future); ‘Present state’ (including tractable emotions and cognitions); ‘Provision’ (informed, tailored and joined up support); and ‘Process’ (reinforcing cycles that enhance or restrict movement and physical activity).

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

The framework, and an associated checklist, can facilitate assessment and personalised intervention to increase physical activity and decrease sedentary behaviour in people with psychosis.

Background

Lack of exercise and sedentary behaviour are both independently linked to poorer health and premature mortality.1 Many people given a diagnosis of psychosis exercise less and sit for longer than other people.2 3 Although the promotion of physical activity for patients with psychosis is clearly recognised as important, there is no model or framework that addresses the specific barriers to physical activity faced by people with psychosis. Furthermore, current exercise interventions have low uptake by patients with psychosis and high attrition rates.4 5 Theoretically driven exercise interventions may be more likely to result in success.6 We set out to develop a framework for increasing movement and physical activity in people with psychosis. This framework is developed to incorporate greater understanding of (1) psychological processes that impede movement, (2) psychosis specific experiences, (3) staff and carer perspectives and (4) drivers of sedentary behaviour.

Many people in the general population, especially people from underserved populations, face multiple barriers to physical activity.7 For people given a diagnosis of psychosis, these barriers may to some degree be qualitatively and quantitatively different because of the many additional challenges they face such as, but not limited to hearing voices, excessive mistrust, severe depression, anhedonia, stigma, side effects of medications, physical comorbidity, low self-efficacy and often very small social networks.8–13 For people with psychosis, the provision of information, resources and practical help to access exercise programmes has often been insufficient to promote behaviour change.14 This may be partly because the psychological barriers to physical activity have not been sufficiently addressed. Psychological barriers to exercise are consistently reported by patients to be among the most important. These include depressed mood, low self-efficacy, stress, anxiety and low motivation.10 11 Low motivation has been addressed in attempts to understand or increase exercise in people with psychosis. This has been understood by applying socio-cognitive models such as the transtheoretical model,15 self-determination theory16 and the health action process approach.17 These models focus on promoting motivation and translating this into action (see Romain and Bernard18 for a review of behavioural and psychological approaches). Cross-sectional research has identified relationships between factors in these models and exercise in people with psychosis19–21 but there is limited evidence from clinical trials that they can predict change in exercise behaviour in people with psychosis.22 23

One attempt to use general models in an intervention illustrates the need for psychosis-specific approaches. The motivational theory-based technique of using ‘if-then plans’ was tested in a feasibility study to increase physical activity in men with obesity and psychosis but reported an attendance rate of only 32%.5 Participants declined to participate in this non-specific intervention, citing their mental health symptoms (eg, paranoia and low mood) as the reason. Interventions may need to be guided by frameworks that pay attention to the specific difficulties patients are facing. In relation to the broader population of people with ‘severe mental illness’, Soundy et al 24 summarise some relevant psychosocial attributes, such as the development of an identity that fits with the chosen activity, and sensitive social support that influences physical activity.

Engagement in exercise programmes by patients with psychosis is also dependent on social support.11 12 25 26 Support by staff has been recognised as important27 and studies have highlighted the need for involvement of qualified exercise practitioners,4 8 28 as well as the challenges of achieving this.29 In qualitative studies, staff factors identified as helpful for increasing activity for people with psychosis include staff knowledge, staff modelling, staff prioritising patients’ physical health, creating good relationships and the provision of daily routine.30–32 However, a better understanding of what prevents staff from offering this support with this specific population may be helpful. Furthermore, the influence of family and friends, who are likely to spend more time with patients than staff, has rarely been examined. Studies examining the role of either staff or carers specifically in reducing sedentary behaviour in people with psychosis are very scarce.

Finally, sedentary behaviours need to be incorporated into frameworks that guide interventions. The understanding of sedentary behaviour in people with psychosis has been relatively neglected. Cross-sectional studies have found that sedentary behaviour is positively correlated with negative symptoms, cognitive difficulties and extrapyramidal medication side effects and negatively correlated with positive symptoms.33 However, intervention work for decreasing sedentary behaviour is at a very early stage.34 35

Objective

We set out to develop a framework, grounded in the experience of people with psychosis and those who support them, which can be used to guide future research and interventions for increasing physical activity and decreasing sedentary behaviour in people with psychosis.

MethodsStudy design

This qualitative study took a critical realist approach. This realist/subjectivist position postulates that there is an objective reality, but that this reality is observed through the researcher’s unique lens. Reflexive thematic analysis (RTA)36 was used to analyse the data. RTA is a qualitative research method used to identify and interpret patterns of meaning across a dataset. This method was a good fit for combining peer and non-peer researcher perspectives and for analysing data from focus groups and interviews together.

The study was designed and conducted in collaboration with people with lived experience and carers, including an expert advisory group and a peer researcher. The expert advisory group included people with lived experience of psychosis and a carer of a person with experience of psychosis. Including this perspective may facilitate greater depth and nuance in data collection and analysis and may increase the trustworthiness of the findings.37 38 The methodological rigour of the study was enhanced by using quality assessment and credibility tools.39 40 Detailed demonstrations of how we addressed quality, such as sensitivity to context, commitment and rigour, coherence and transparency, and impact and importance are provided in online supplemental materials.

Participants

The study included the perspective of three distinct groups of participants: patients with a diagnosis of psychosis, carers and staff. Participants were recruited via three mental health trusts in England (Oxford Health NHS Foundation Trust, Berkshire Healthcare NHS Foundation Trust and Humber Teaching NHS Foundation Trust). Carers were identified by mental health trust staff or carer support groups associated with the trust. Staff were identified by clinical research network staff or identified themselves.

Inclusion criteria for patients were: aged 16 years or older; able to give informed consent; clinical diagnosis of schizophrenia spectrum psychosis; and experience of psychosis in the past 2 years. Inclusion criteria for caregiver participants were: aged 16 years or older; able to give informed consent; and family, partners or close friends of people meeting the above criteria who have at least 5 hours contact per week. Inclusion criteria for staff participants were: aged 16 years or older, able to give informed consent and experience of working directly with patients meeting the above criteria in the community in the past year.

Purposive sampling was used to maximise variation in participant characteristics, including age, gender and ethnicity. Staff variation also included job role and type of mental health team (ie, early intervention or adult community mental health team) and for carers it included relationship to the patient they cared for. The characteristics for sampling were informed by an expert advisory group.

Procedure

Participants could choose to take part in a focus group or an individual interview. Focus groups were conducted by a clinical psychologist (RD). Seven were co-facilitated by a peer researcher (AH). Focus groups included a mixture of patients, carers, and staff, allowing interactions between people in the group to contribute to learning. A further 23 individual interviews were conducted by RD.

The topic guide (see online supplemental materials), developed in consultation with the PPI Expert Advisory Group, covered sitting, standing up to break sitting time, exercising and how people can be encouraged to move more. It was employed flexibly, to allow for the development of new ideas during participation. Follow-up questions and prompts were used as needed.

Field notes and audio recordings were taken, transcribed verbatim and pseudonymised.

Analysis

Analysis followed the six-step guidance provided by Braun36 and was underpinned by a critical realist approach (see online supplemental materials for details). Following familiarisation with the data, initial codes were created incorporating units of meaning in the data. These synthesised data from all three groups. Passages of data and their associated codes were discussed with members of the research team (including the peer researcher), facilitating additional codes and interpretations (eg, peer researcher reflections about choice and control feeling overwhelming or positive in different circumstances). Codes with similar meanings were grouped (e.g. combining ‘benefits of movement’ and ‘rewards’). The list of codes was refined and discussed through an iterative process, and codes incorporating more latent meaning were added.

Using a single sheet of paper, codes were grouped into broad categories. These early thematic clusters represented collections of codes which predominantly summarised a topic area. These were generated and regenerated iteratively, through a process of reflection and immersion in the data. Respondent validation was conducted with 14 participants, including 6 patients, 6 carers and 2 staff. This further refined and defined the themes. For example, an initial theme entitled ‘other people’ was understood to incorporate information about connecting with others and how this was done. It was renamed ‘connections with others’. Further reflections recognised that the connections with others were a part of the provision of support in a system.

The themes form a five-factor framework. Each factor represents an important area in which there can be barriers and facilitators to physical activity. The framework was translated into a checklist for encouraging physical activity.

Analysis was supported by the use of NVivo V.12.41

Findings

In total, 77 potential participants were identified. 51 participants took part, 14 declined (reasons unknown) or were not able to attend and 2 did not meet the inclusion criteria.

In total, 19 patients, 14 carers and 18 staff took part. See table 1 for characteristics.

Table 1

Participant characteristics

Focus groups lasted an average of 84·8 min (SD=12·5). The 23 individual interviews lasted an average of 54·9 min (SD=27·6). With the exception of one focus group, which included a patient who had asked to be with a particular staff member, no focus group participants knew one another.

Five themes (referred to hereafter as factors) were identified from the thematic analysis. These are summarised in figure 1. The elements of the factors that can help support clinical assessment were used to produce a checklist (see table 2).

Figure 1

Summary of the 5 P (Purpose, Predictions, Present state, Provision and Process Framework). PA, physical activity.

Table 2

The Purpose, Predictions, Present state, Provision and Process checklist*

Factor 1: Purpose

I think it’s something about having … like the pull factors…. and potentially like if people feel like they have things that they love or enjoy or are meaningful in some way to get up and do, that could be helpful opening up…. like the world getting a bit bigger and that becoming a bit more part of the day to day—Maria, staff.

As I said, I can sit for ages, but then I have stuff to do as well, the dogs need walking, I’ve got to go out, and that is the main one that gets me up at the minute—Julie, patient.

She [daughter] lacks a focus, she lacks a reason for doing something, uh, a bit like Jemima said it’s easier just to sit and put the TV on—John, carer.

It was described that both breaking up sedentary time and physical activity are done for a reason, and the meaning or purpose of the movement to that individual is critical. Moving for a valued reason, rather than just for the sake of moving, is best connected to a person’s sense of themselves, what they value and their identity. For some individuals, goals and measurable progress towards meeting these, are motivating and reinforce the sense of identity. For others, the person’s own goals are less important to them or less motivating than it is to simply help other people. A sense of collective identity, of belonging to something bigger than oneself, can also provide the purpose of the movement.

Though applied in relation to both physical activity and sedentary behaviour, the importance of a purpose and meaning to activity was more frequently discussed in relation to getting up from sitting than it was in relation to being physically active.

Factor 2: Predictions

At home ……the parents would tell you to stop wasting your time sitting down and you always hear them quote, ‘you might become a vegetable on the couch’ there and all of that. It’s always been there, like a knowledge you grow up with, just knowing you have to go for a walk or you have to do something—Celine, patient.

A lot of clients, they build up in their head it’s going to be horrendous, it’s going to be so scary, and I say to them that the reality of actually doing it isn’t as bad as you thought it was going to be in your head—Nicki, staff.

All the diabetes people could put [spouse] straight onto drugs for diabetes, rather than trying to expect that she would adopt a healthy lifestyle—Hugh, carer.

An individual’s predictions and expectations of being physically active were noted as an important determinant of movement. The predictions are shaped by the person’s own background in relation to activity, with implicit messaging about the value of movement beginning early in the family, culture and wider community. Past experiences of physical activity, from childhood or more recently, can make these predictions positive or negative. These predictions may be made not just in word form but also in the form of powerful images, which instil hope or fear.

Expectations of oneself and being physically active in the future were reported as being influenced by others’ expectations. Staff and carers alike can communicate explicitly and implicitly the hope and expectation that someone will be active. Movement may be implicit in the support that is offered or modelled by those around the person. Hope in patients is particularly inspired by physical activity being modelled by others who have experienced similar challenges.

Factor 3: Present state

Yes, it’s [TV] distracting. The noise obliterates or calms down what’s going on in his [son’s] head… just sitting and not talking and kind of being mesmerised by it because apparently, it helps with the voices—Valerie, carer.

You get consumed about things, like, “the voices are bad”, or “I don’t deserve to be doing these things because [voice] says I’m too fat to be doing them anyway”, and sometimes you sit and nothing goes through your head. It’s just blank time, as I call it. Just nothing—Julie, patient.

I guess one of the things I've noticed is worry and rumination, it can be quite powerful and people can get quite locked into it. And actually… or feel like by sitting and thinking about… kind of going over their worries, they're going to find the answer, or kind of figure it out. I wonder whether that can mean people sit and get quite lost in what’s going on in their heads… like lost in that rumination—Maria, staff.

A person’s ability or perceived ability to move was partially determined by internal psychological and physical experiences. Symptoms of psychosis such as persecutory fears and hearing voices were directly responsible for choices not to do a preferred activity. Attempts to manage these symptoms can also involve staying still for long periods of time. Furthermore, these symptoms can result in a person becoming so engaged by their internal experiences that they are not aware of time passing and therefore how long they have been sedentary.

The content of thoughts, both positive and negative, were influential in choices about physical activity at any one time. This was open to change, depending on environmental stimuli, such as the presence of other people. Subsequent changes in mood could then further facilitate or hamper physical activity.

Physical sensations and health-related problems such as pain, balance and co-ordination problems, fatigue and weight gain influenced the ability and choice to move, though sometimes providing motivation to move, rather than only impeding it.

Factor 4: Provision

But no, there’s nothing there to ensure… not even a sheet to say, “this is what you need to do or you shouldn’t be sitting down for too long”, So yeah, I suppose there was a bit of a lack of continuity there when you were released [from hospital]—Larry, patient.

But I don’t know how much she [care coordinator] knows and something that came back was that she didn’t know that he had been very sporty. And that’s the sort of information which is really good for care coordinators to have—Anne, carer.

Just having more time because I think there’s that period of time when someone’s got the motivation and the energy and things and then if you’re not able to harness it then and there it feels like it kind of just slips away and then it’s gone and you’re like okay I’m going to have to start from scratch again now—Jen, staff.

Participants described how the provision of the right support and resources can be an effective catalyst for movement. At a time when informal support from friends and social networks often diminishes, support from caregivers and staff is especially significant.

Clear and regular communication between patients, staff and carers was reported as important. For example, effective communication between staff and caregivers about a person’s purpose and preferences for certain activities assisted in quickly finding the right motivation for physical activity. Similarly, good communication between different services in the care system was essential, in order to allow transitions in care to facilitate physical activity, allowing it to be maintained or increased, rather than suddenly reducing.

Provision of accessible information to patients and those who care for them is important, allowing everyone in the system to understand the benefits of movement, how to make changes and what resources are available. Provision of resources and transport to access them was reported as fundamental.

Factor 5: Process

I guess it’s momentum isn’t it? That short term sensation of feeling good and feeling better and that kind of stays with you doesn’t it? The longer you leave it, the not exercising, you forget the short term feeling of it—Toby, patient.

I suppose one of the things would be that if you tried to help somebody and they don’t want to do it, you know, that can demotivate the motivator, as it were. So, perhaps something around how you can encourage that person or that body of people, whoever they are that’s supporting that service user—Henrietta, carer.

I think the habit forms as a lot of that responsibility is taken away. I think because of that, they don’t necessarily need to get up and cook and shop and things like that and we’re taking that responsibility away from them because again, we’re primarily looking at the mental health… people can become quite dependent on us and we often do a lot of thinking about how we can make them more independent because if not…. they will just physically sit there because they know that people are going to care for them and come in—Simon, staff.

It was found that psychological processes can serve to initiate, maintain or even interfere with physical activity. Momentum can be gained by forming habits and routines but is also lost when those habits are interrupted. Conversely, unhelpful habits or routines can be created which can be hard to break, creating and maintaining amotivation. Sudden changes in motivation were conceptualised as turning points. These were sometimes triggered by an interpersonal experience such as advice offered or an intrapersonal experience such as a realisation that staying in bed will impede recovery.

Processes between the patient and other people may also be impactful, with certain support styles serving to enhance motivation and others serving to reverse it. The need for a particular support approach was dynamic, where on some occasions one approach may work well to encourage movement but the next day this may serve the opposite effect.

Furthermore, those around the patient were also subject to motivational processes. Their own motivation and emotional availability to support movement may be influenced by the behaviour of the patient themselves, other people who are trying to support the patient or the system in which they are operating.

Discussion and clinical implications

The interviews with patients, carers and staff highlight how a large range and number of factors need to be in place for people diagnosed with psychosis to stand up more often and to be physically active. The Purpose, Predictions, Present state, Provision and Process Framework, and associated checklist, provides a way of grouping these factors. Key factors highlighted in the framework are the importance of understanding the meaning of movement and physical activity to each person and provision of support that can facilitate it. This study also identifies the importance of expectations for movement, based on previous experiences, current support and systemic factors. The framework provides a structure for conversations and assessment in day-to-day clinical practice and also a way to guide the development of new treatments.

The findings support previous research that indicates that motivation is a critically important factor for both exercise and movement.8 25 42 Our findings identify the importance of understanding the meaning of physical activities to patients. The meaning of activity needs to closely match a patient’s sense of their identity and what is of value to them. As previous studies have highlighted,11 12 30 43 a one-size-fits-all approach to increasing activity is unlikely to work. Tailoring the offer to patients is important. Our findings highlight one important reason why this might be—if the activity does not hold a salient meaning, relating to someone’s values and sense of self, it is unlikely to be enough to overcome the other barriers they face. Therefore, support offered will likely be more successful if it is provided on the basis of a sufficient understanding of the person and their values. The central importance of identity in relation to recovery in people with schizophrenia has previously been highlighted, with calls for healthcare professionals to pay attention to an individual’s past identities.44

Motivational processes in the care system are also crucial. As other studies identify,11 12 42 social support can be key to helping patients to overcome barriers to initiation and maintenance of physical activity. Our findings highlight that supporters are subject to motivational changes too. The motivation of patients, carers and staff appear to be inter-related. With the provision of support for the supporters, ‘giving up’ could be replaced by an understanding that progress is not steady and that someone declining an activity at one time does not mean it should not be offered again.

As identified by previous studies,30 45 46 patients’ low motivation, low mood and anxiety can be so overwhelming that consideration of any other activity is eclipsed. Our findings identify some of the specific thoughts about physical activity that people can experience. Mood-related cognitions (eg, “I don’t deserve to be doing these things”), psychosis-related cognitions (eg, “It’s not safe to go outside to exercise”) and cognitions based on previous exercise experiences (eg, “Exercise makes me feel humiliated”) were identified. Furthermore, participants described cycles between these thoughts, emotions and behaviour that resulted in physical activity being even less likely. For example, in some instances, where one element of the cycle was disrupted (eg, rumination was interrupted), physical activity became more likely. The content of thoughts will determine the strategy for increasing movement and physical activity. It is therefore important that there is development of assessments of movement-related cognitions.

Psychological processes also affect motivation through the predictions (eg, that exercise will be painful or humiliating) made about physical activity. These predictions may also be amenable to psychological therapy techniques such as imagery rescripting. However, the findings suggest that predictions can also be influenced by other people in the support structure, suggesting that there may be value in obtaining support from peer support workers. Peer support workers can provide hope for the future, and there is evidence that they can enable adherence to physical activity by patients with schizophrenia.47

Previous literature identifies that exercise programmes delivered by trained exercise professionals can enhance the adherence and success.4 8 28 Furthermore, where physical health comorbidities are so common, the safety of exercise programmes with input from exercise professionals is likely to be better, for example, through exercise capacity assessments.48 This issue was not raised in the focus groups or interviews and therefore does not feature in the results of this study. However, development of interventions will need input from exercise professionals. In terms of delivery, there is typically very limited provision of physiotherapy support in psychiatric community settings. The use of the framework and checklist in routine clinical care in the community may serve to help staff and carers to address some of the barriers that prevent patients from accessing this specialist support in the form of groups or programmes which are offered. Furthermore, the framework and checklist may enable conversations and planning about lower intensity movement, such as standing up more often, which may be safer to support by non-specialists.

There are a number of study limitations. Participants were recruited from only three sites across the UK and will not be representative of all patients, staff and carers. The sampling strategy was intended to maximise variation, but the majority of the carers who took part were parents of someone with psychosis, with few participants occupying other caring relationships. Background, culture and community were found to influence the predictions people made about physical activity. It is therefore important that further research with ethnic minorities and other under-represented groups is undertaken. There are advantages to conducting mixed focus groups for patients, staff and carers, but it is possible that some members of the groups felt inhibited by the presence of their co-attendees and contributed less, or differently, as a result. In order to minimise power imbalances in focus groups, we co-facilitated groups with a peer researcher where possible and invited people to contact the facilitator afterwards if there was anything additional they wished to say. This study addressed both sedentary behaviour and exercise in the same interviews. The five factors are each relevant to both types of physical activity but some may be more relevant to one type of inactivity than the other. For example, discussions around ‘purpose’ more often related to addressing sedentary behaviour than exercising. We consider that the next steps are to develop precise assessments of the identified processes, test their importance in quantitative studies and begin a process of intervention development to produce a treatment that is implementable by a range of mental healthcare staff.

Data availability statement

Participant data will not be shared due to the potential for participant identification. Details of the analytic process are provided in supplementary materials, along with extensive supporting quotations.

Ethics statementsPatient consent for publication

Not applicable.

Ethics approval

This study involves human participants and was approved by the Health Research Authority Wales Research Ethics Committee 6 (REC reference: 21/WA/0285). Participants gave informed consent to participate in the study before taking part.

Acknowledgments

The authors would like to thank the participants who took part in this study and the people who supported them to do so where necessary. The authors would also like to thank members of the expert advisory group for providing lived experience and carer expertise into the design and delivery of the study. The authors would like to thank the three NHS trusts that participated: Oxford Health NHS Foundation Trust, Berkshire Healthcare NHS Foundation Trust and Humber Teaching NHS Foundation Trust. The authors thank the members of their research and development teams and in particular Ceri Kirosingh, Samantha Morris and Emma Anderson. For help with recruitment, the authors thank Diane Hilson, Nicola Collett and Faisal Shaikh. Finally, the authors would like to thank the clinical teams in the participating NHS trusts.