MOTIVATION, PHYSICAL ACTIVITY, SOCIAL SUPPORT AND
MENTAL WELLBEING: INVESTIGATING DIFFERENCES BETWEEN
ONLINE AND FACE-TO-FACE SOCIAL SUPPORT
This chapter established that interventions which have incorporated an element of peer
support were generally reported as successful for improving PA levels and psychosocial
variables such as health-related quality of life and mental wellbeing. However, to date, there
has been little research considering peer support within community PA programmes for
MHSU, and in particular longitudinal research. Community settings provide opportunities for
MHSU to receive social support from peers to help overcome barriers towards PA
engagement. Social support can be offered in different ways which may positively or
negatively impact outcome variables such as PA, motivation, perceived social support and
mental wellbeing. Chapter 3 will quantitatively and longitudinally explore patterns of change
in psychosocial variables of MHSU within a community PA programme (Get Set to Go) and
explore differences between social support delivered via an online community platform or
face-to-face during PA sessions.
Abstract
There is a need to better understand how key features of community-based programmes such
as social support may encourage participation by MHSU over time (Quirk et al., 2017). This
study explored patterns of change in motivation, physical activity (PA), social support and
mental wellbeing of mental health service users (MHSU) and whether these patterns differed
between an online and face-to-face social support intervention condition. MHSU participants
(N=1324, 364 males, 350 females) registered to a community-based PA programme
responded to a multi-section questionnaire at baseline, 3 months and 6 months, which
measured variables of PA, motivation, social support and mental wellbeing. Multi-level
growth curve analyses revealed increased PA levels in the face-to-face condition versus
decreased PA levels in the online condition. However, non-significant differences were
observed for social support, motivation and mental wellbeing between the intervention
conditions. This study provides insight into different channels of social support within a
community PA programme for MHSU. An online community platform represents a feasible,
cheaper alternative to face-to-face social support for psychological outcomes; however face-
to-face social support was more effective for enhancing PA engagement.
Introduction
The evidence base for the positive relationship between PA and mental wellbeing is
well established (Rosenbaum, Tiedemann, Ward, Curtis, & Sherrington, 2015). Longitudinal
research investigating PA and anxiety globally found that engaging in low levels of PA which
did not meet the recommended levels to maintain and improve health was positively
associated with anxiety (Stubbs et al., 2017). Significant positive associations have also been
found between PA and mental wellbeing for those with depression (Schuch et al., 2017;
Stubbs, Rosenbaum, Vancampfort, Ward, & Schuch, 2016; Vancampfort et al., 2017) and
anxiety (McDowell, MacDonncha, & Herring, 2017). There is increased evidence for PA
having a positive impact on physical and mental health, as well as illness symptoms of
MHSU (Farholm, Sørensen, & Halvari, 2017). However, MHSU still experience barriers that
make it difficult to take advantage of the benefits associated with PA, and therefore engage in
less PA and significantly greater amounts of sedentary behaviour compared to the general
population (Vancampfort et al., 2016). Such findings suggest that attention should be given to
how to promote the uptake of PA in this population to bring about health benefits (Quirk,
Crank, Harrop, Hock, & Copeland, 2017).
One consistently reported barrier that impedes engagement in PA by MHSU is a lack
of motivation (Soundy et al., 2014). A lack of motivation can act as a major obstacle for both
starting and maintaining PA and is therefore an important research focus to consider. A
theoretical framework that has been applied to understand the role of human motivation is
Self-determination theory (SDT; Deci & Ryan, 2000). SDT is a contemporary theory of
motivation with rapid growth in applications for sport and PA (Ntoumanis, Quested, Reeve,
& Cheon, 2017). Central to SDT is the distinction between autonomous forms of self-
determined motivation and controlling or non-self-determined forms of motivation (Deci &
Ryan, 2000). Autonomous motivation is associated with positive behavioural outcomes (i.e.,
persistence to exercise), positive cognitive outcomes (i.e., attitudes and intentions), and
psychological wellbeing (Fortier, Duda, Guerin, & Teixeira, 2012).
As motivation presents one of the key determinants of PA engagement, attention
should be dedicated to motivation-related constructs of PA behaviour and the recognition that
motivation to initiate and persist with PA is multidimensional and dynamic in nature (Daley
& Duda, 2006). These motivation dynamics are characterised as a process in which
individual reasons to engage in a certain behaviour, such as PA, change over time
(Wasserkampf & Kleinert, 2015). Ideally, the dynamic changes become more strongly
internalised so the reasons for engaging in PA become more part of the ‘self’ (i.e., organismic
integration) (Wasserkampf & Kleinert, 2015). Therefore, in facilitating the engagement of
PA among MHSU, attention should be directed towards the dynamic nature of behaviour
regulations.
The Organismic Integration Theory, a subtheory of SDT (OIT: Deci & Ryan, 2000)
offers a framework for clarifying the inclination that individuals have towards integrating
subjective reasons for PA behaviour into themselves. Within OIT, different regulation forms
are aligned on a continuum of self-determination (Ryan & Deci, 2007). The continuum
contains three autonomous forms of motivation (intrinsic, identified, integrated), two
controlled forms of motivation (introjected and external) and amotivation which suggests
zero motivation towards that behaviour (Deci & Ryan, 2000). Research has produced
consistent support for a positive relationship between autonomous forms of motivation and
PA, with intrinsic motivation being predictive of long-term adherence to PA (Ng et al., 2012;
Teixeira, Carraça, Markland, Silva, & Ryan, 2012) and amotivation negatively associated
with PA (Vancampfort et al., 2016). However, whilst research may imply a progression from
least to most fully autonomous regulations along the continuum, this should not be
considered as a stage model nor a developmental continuum, but rather a conceptual
continuum (Ryan & Deci, 2007). For example, not all individuals start with external
regulation and move towards more autonomous regulations. Socio-contextual factors can
either support or undermine the internalisation process, inherent personal tendencies for
growth and development, leading to consequences of such processes on quality of motivation,
psychological wellbeing and PA engagement (Ntoumanis et al., 2017). As such, depending
on the social conditions in which individuals obtain a new regulation, they are able to start
anywhere along the continuum.
Optimising the quality of motivation is further associated with whether a significant
other who plays an instrumental role in shaping an individual’s experience within the
exercise setting supports the need for autonomy, competence and relatedness, and therefore
promotes intrinsic interest of PA (Teixeira et al., 2012). Effective communication is a crucial
element to support individuals’ successful engagement in PA and promote autonomous
motivation (Ntoumanis et al., 2017). Previous research has mostly focused on creating an
autonomous supportive environment through a figure of authority (e.g., an exercise
professional) (Moustaka, Vlachopoulos, Kabitsis, & Theodorakis, 2012; Rouse, Ntoumanis,
Duda, Jolly, & Williams, 2011). However, more recently, the satisfaction of psychological
needs can also be supported via different sources and/or programmes such as peers
(Kinnafick, Thøgersen-Ntoumani, & Duda, 2014) who can be approached to seek help with
motivation and PA engagement (Duda et al., 2014).
There has been increasing interest for research focusing on motivation and PA among
MHSU. Being active can be challenging for these individuals due to symptoms of illness or
medication, and a lack of motivation (Firth et al., 2016). However, non-uptake of PA by
MHSU does not necessarily reflect solely a lack of motivation to be active, as research has
shown high levels of interest in PA among MHSU but low levels of activity engagement
(Ussher, Stanbury, Cheeseman, & Faulkner, 2014). This implies a gap between individuals’
motivation and behaviour that can be addressed by exploring how the context and social
environment could help reduce this gap and facilitate PA engagement (Quirk et al., 2017).
Several factors contribute to the general poor physical health of MHSU including
social environmental determinants such as stigmatisation and a lack of social support
(McDevitt, Snyder, Miller, & Wilbur, 2006). Research on social relationships and social
support stand among the strongest social science contributions to our understanding of
experience and outcomes of mental illness (Smith & Christakis, 2008). The absence of
positive social relationships is a significant risk factor for morbidity and mortality (Cacioppo
& Cacioppo, 2014) with individuals who are more socially connected living longer and
experiencing better mental and physical health (Holt-Lunstad, Smith, Baker, Harris, &
Stephenson, 2015).
Socially supportive relationships could help with the initial barriers towards PA
engagement in MHSU who are known to experience social isolation and stigmatisation
(Soundy et al., 2014). It is therefore important to consider PA settings that can facilitate
social support for MHSU. Community-based approaches to PA involve community members
coming together to promote PA in an organised and integrated manner (Bopp & Fallon,
2008). Limited resources are needed to reach large numbers of individuals, often resulting in
greater improvements in outcomes and improved sustainability over time (Quirk et al., 2017).
Community-based approaches are appropriate for MHSU whose health is influenced by
complex individual-level and system-level factors, and who may have specific needs and
barriers to PA participation (Firth et al., 2016). Prescribing PA does not directly lead to
mental health and wellbeing benefits, but efficacy of these efforts are entirely reliant on
individuals’ current states, past and context (Rebar & Taylor, 2017).
Social support can mitigate the impact that stress and negative events have been
demonstrated to have on mental wellbeing (Lewandowski, Rosenberg, Jordan Parks, &
Siegel, 2011). This buffering role of social support is not only dependent on the amount of
social support an individual receives but is further affected by the type of communication the
individual uses to receive the support (Lewandowski et al., 2011). PA engagement can
facilitate interaction within natural settings (i.e., the community), and provides the
opportunity for social interaction (relatedness), mastery in the physical domain (perceived
competence) and independence (autonomy) (Lubans et al., 2016). Participating in a
programme that directly targeted social isolation and disconnection led to improved mental
health, mental wellbeing and social connectedness in young adults (Haslam, Cruwys,
Haslam, Dingle, & Chang, 2016). This demonstrates that by reducing social isolation and
building on social identification through programmes incorporating social support,
individuals can overcome the challenges faced as a result of mental illness (Haslam et al.,
2016).
Social support is conventionally delivered face-to-face, however advances to both
society and technology have led to the emergence and growth of the internet as a
communication tool providing ways for individuals to access social support online (Ziebland
& Wyke, 2012). As a result, individuals are increasingly turning to computers and technology
to seek out social support (Wright & Rains, 2013). Online social networking is a prominent
form of communication within Western populations (Naslund, Aschbrenner, Marsch, &
Bartels, 2016). MHSU can experience challenges with face-to-face communication due to
impairments in social functioning and therefore online methods provide alternative
opportunities to generate social interaction with peers (Naslund et al., 2016).
Reported benefits of online social support include social interactions being
independent of time or location (Trepte, Dienlin, & Reinecke, 2015). Informational support
can be offered to individuals, via a larger number of sources, in comparison to an offline
social context and therefore, the probability or receiving requested information appears
higher for online social settings (Trepte et al., 2015). Virtually omnipresent networks were
better able to provide social support as a result of greater availability with regards to time, in
comparison to a network that required interactants to be present at the same time (Trepte et
al., 2015). Additionally, the internet, social media and online community sites all have the
power and ability to provide social support and resources 24 hours a day, reach difficult-to-
serve populations (e.g., MHSU) and at a relatively low cost (Parikh & Huniewicz, 2015).
This provides a compelling reason to explore the impact of online social support for MHSU
within a community PA programme and consider how it compares to face-to-face social
support.
A study comparing face-to-face social support groups and online support groups for
cancer patients reported no significant differences on outcomes such as anxiety, depression
and quality of life (Huber et al., 2018). However, greater patient ratings were reported for
face-to-face social support being more effective for exchanging information, gaining
recognition and caring for others (Huber et al., 2018). Additional research investigating
student perceptions of a learning programme found that face-to-face learning perceptions
were reported higher in terms of social presence, social interaction and satisfaction compared
to online learning (Bali & Liu, 2018). However, statistically, there were no differences in
learning preference among individuals (Bali & Liu, 2018). There is limited generalisability
of research findings beyond a cancer patient sample or college students, and therefore
research is required to investigate face-to-face and online social support perceived by adult
MHSU, and to consider the differences between the channels of social support on outcomes
such as motivation, PA levels and mental wellbeing within a PA context.
Additionally, prior research that has considered associations between variables such
as PA, mental wellbeing, motivation and perceived social support have predominantly been
cross-sectional designs (Gunnell, Crocker, Mack, Wilson, & Zumbo, 2014; Kwag, Martin,
Russell, Franke, & Kohut, 2011). However, cross-sectional data results in paths that do not
represent causal relationships between predictors and outcomes (Wang et al., 2018).
Investigating models or patterns within the data longitudinally would obtain more reliable
results. Moreover, longitudinal investigations over multiple time points, could allow for
multilevel modelling to determine how trajectories of change influence variables such as PA,
social support, mental wellbeing, motivation of MHSU (Wang et al., 2018).
The current study
Modifiable behavioural risk factors such as low levels of PA have become an
important target in programmes to help improve the overall physical and psychosocial
health of MHSU (Stubbs, Williams, Gaughran, & Craig, 2016). PA in group community-
based settings have additional benefits of social support from peers that may be able to
encourage activity engagement, through connections and the creation of social identity
(Quirk et al., 2017). However, there is a need to better understand how key features of
community-based programmes such as social support may encourage participation by
MHSU (Quirk et al., 2017).
A systematic review concluded insufficient evidence for current community-wide
programmes, particularly highlighting issues with scalability. This is a common weakness
with many previous programmes which fail to reach a substantial portion of the community
involved (Baker et al., 2015). Further, a frequent limitation of research examining the
relationship between PA and mental wellbeing is small sample sizes and a lack of
longitudinal designs establishing longer term associations (Harandi, Taghinasab, & Nayeri,
2017; Wang et al., 2018). Little is known about the mechanisms that support successful
implementation of community-based PA programmes for MHSU (Harris, 2018). Therefore,
longitudinal research is needed to understand how the concept of, and differences between,
face-to-face and online support within a community PA programme may be associated with
outcomes such as PA, social support, motivation and mental wellbeing among MHSU
(Bellamy, Schmutte, & Davidson, 2017).
Aim and hypotheses
This study set out to aid our understanding of how online and face-to-face support in
the context of a community PA programme may differentially influence social support, PA,
motivation and mental wellbeing of MHSU. Thus, this study examined if online versus face-
to-face social support predicted changes in 1) PA, 2) motivation, 3) social support and 4)
mental wellbeing of MHSU.
Design
Method
Community-based PA programme. In November 2014, a national UK based mental
health charity implemented a three-year programme called Get Set to Go (GStG) to
encourage MHSU to become more involved in PA. Eight local charity organisations,
affiliated with the national charity, across four regions of the UK (North East, North West,
Midlands and London) organised PA taster sessions of a variety of sports along with social
support and one-to-one advice to MHSU. Face-to-face social support was provided during the
PA sessions via one-to-one and group-based support. Group-based activities depending on
what was available in the local area (e.g., badminton, bowls, yoga and tai chi) were provided
to introduce MHSU to PA. Peer volunteers provided the face-to-face social support and a
Sports Coordinator was employed to be in charge of the organisation and running of the
programme in each local area.
In addition to the face-to-face social support offered within PA sessions, the charity
also facilitated an online social support community for their members. As part of GStG, the
existing online community was developed to better support MHSU in sharing stories about
getting active and how to overcome barriers to encourage individuals to engage in PA.
Information focused on being physically active, getting started and was provided via short
videos. Comments were left on the interface to encourage discussions among the members.
MHSU members could offer advice and share stories of personal experiences and therefore
could interact on the topic of PA. Online social support could be accessed 24 hours a day,
7 days a week.
Participants and Procedure. 1324 MHSU (males =364, females =350) who were
either registered to the GStG programme (n=980) or members of the existing online
community platform (n=344) took part in the study. Participants had a range of diagnoses
including anxiety (n=394), depression (n=448), stress (n=247), PTSD (n=65), personality
disorder (n=69), bipolar disorder (n=78), schizophrenia (n=65) and OCD (n=54). 952
participants completed the questionnaire at baseline, 180 completed the questionnaire at 3
months and 192 at 6 months.
Ethical approval was obtained from a Research Degrees Board of a University in the
East Midlands of England. All individuals that had registered as a participant on the GStG
programme were invited to participate in research as part of the programme. Participants
involved in the study were provided with consent forms and written information about the
purpose of the study. Participants who were either registered to GStG or were online
community platform users completed a questionnaire at 3 time points (baseline registration, 3
months, and 6 months). Participants in the face-to-face condition (GStG) initially completed a
hard copy questionnaire before it was adapted to a digital form for the follow ups, and the
online condition completed the questionnaire online. The questionnaire was developed after
consultation with a Lived Experience Panel representing a variety of mental health
backgrounds. In accordance with ethical principles, participants were reassured about the
confidentiality of the responses and their right to withdraw at any time until the final report
had been written. The same procedure was followed across all 3 measurement timepoints.
Participant questionnaires were matched up across the 3 measurement occasions using
demographic information.
Measures
Motivation to exercise. A short version (18 items) of the Behavioural Regulation in
Exercise Questionnaire-3 (BREQ-3; Markland & Tobin, 2004; Wilson, 2006) was developed
to assess an individual’s motivation to exercise. This instrument has been used to measure
behavioural regulation according to SDT in the exercise domain. In SDT, regulatory
mechanisms indicate degrees of behavioural internalisation, reflecting the transitioning of
habits and requests, to endorsed values and self-regulations (Cid et al., 2018). The BREQ-2
has previously been validated with a mental health population (Vancampfort et al., 2013). A
shortened version of BREQ-3 was utilised in this study which used 2 items per motivation
regulation to assess overall motivation. Participants rated their motivation to exercise on a 5-
point Likert scale, (not true for me) to 4 (very true to me). Example items include “I exercise
because it is fun” (intrinsic regulation), “I value the benefits of exercise” (identified
regulation), “I feel guilty when I don’t exercise” (introjected regulation), “I exercise because
other people say I should” (external regulation) and “I don’t see the point in exercising”
(amotivation).
A Relative Autonomy Index was calculated by weighting the subscales of the self-
regulation questionnaire used to measure motivation (Williams, Grow, Freedman, Ryan, &
Deci, 1996). For example, intrinsic regulation was weighted 2, identified regulation weighted
1, introjected regulation weighted -1 and external regulation weighted -2. The controlled
subscales (external and introjected) were weighted negatively and the autonomous subscales
(identified, integrated and intrinsic) were weighted positively. The more controlled the
motivation regulation style represented by the subscale, the larger its negative weight and the
more autonomous the regulation style represented by the subscale, the larger its positive
weight.
Physical activity behaviour. Physical activity was assessed using the International
Physical Activity Questionnaire short version (7 items) (IPAQ-SF; Craig, Marshall, Sjöström,
Bauman, Booth, Ainsworth., et al., 2003). Participants reported the number of days in the
past week and the total time per day of walking, engaging in moderate-intensity and
vigorous-intensity physical activity in bouts of 10 minutes or more. Moderate and vigorous
intensity activities were defined using the standard IPAQ descriptions. Moderate activity
included activities that required moderate physical effort and made the participant breathe
harder than normal. Vigorous activity included activities that required hard physical effort
and made the participant breathe much harder than normal. Total PA on the IPAQ represents
all PA related to work, transportation, leisure, domestic or garden activities. Example items
include “During the last 7 days, on how many days did you walk for at least 10 minutes at a
time?” (responding with a number of days) and “How much time did you usually spend
walking on one of those days?” (responding with hours and minutes per day). Previous
research has validated the IPAQ-SF against objective measures of PA (i.e., accelerometers)
(Lee, Macfarlane, Lam, & Stewart, 2011).
Perceived social support. The Social Provisions Scale (SPS-10; Cutrona & Russell,
1987) assessed perceptions of social support. Participants responded to 10 items indicating
the extent to which they felt supported. Responses were given on a 4-point Likert scale from
1 (not at all true) to 4 (completely true) to descriptions of both the presence or absence of a
specific provision i.e. social support. An example item is “There are people in the group I can
depend on to help me if I really needed it.” Previous research has validated the SPS (Caron,
2013).
Mental wellbeing. Mental wellbeing was assessed using the Warwick Edinburgh
Wellbeing Scale (WEMWBS; Tennant et al., 2007) which was a 14-item self-report measure.
Participants rated their experience regarding each statement over the last 2 weeks. Each item
is scored using a 5-point Likert scale ranging from 1 (none of the time) to 5 (all of the time).
Example items are “I’ve been feeling optimistic about the future” and “I’ve been feeling good
about myself.” Previous research has validated the WEMWBS within the mental health
population (Bass, Dawkin, Muncer, Vigurs, & Bostock, 2016).
Data Analysis
Multilevel regression analysis employing MLWin 3.02 (Charlton, Rasbash, Browne,
Healy, & Cameron, 2018) was used to examine changes in perceived social support,
motivation, PA levels and mental wellbeing over the course of 6 months. This type of
analysis is particularly useful when there are missing observations since it does not assume
equal number of measurement occasions for all individuals (Hox, 2000). Two levels of
analysis were specified. Level 1 encompassed the repeated measures of perceived social
support, motivation, PA and mental wellbeing across 3 timepoints (baseline, 3 months and 6
months). These repeated measures were nested within participants which constituted level 2
in the analysis. The first part of the analysis was to examine whether the study variables
significantly changed over the 6-month intervention. This was done by examining
unconditional growth models with a linear time variable included, which was centred at the
beginning of the study (i.e., Time = 0).
The second part of the analysis aimed to ascertain whether any change over time
could be accounted for by the intervention variable of online versus face-to-face. This was
achieved by adding to the unconditional growth models an intervention variable (face-to-face
= 0, online = 1) as well as the interaction term between time and the intervention variable. A
statistically significant association between the intervention variable and the dependent
variable evidenced differences at baseline. A significant interaction term evidenced different
rates of change across the intervention types.
Results
Preliminary Analysis.
Table 3.1 presents the means, standard deviations and Cronbach’s alpha coefficients
for all variables on each of the 3 measurement occasions. Cronbach’s alpha coefficients
revealed that all subscales demonstrated adequate scale score reliability.
Table 3.1. Means and standard deviations for all variables at all 3 timepoints
Baseline 3 months 6 months
Variable α M SD M SD M SD
Perceived social support 0.93 2.85 0.67 2.93 0.65 2.92 0.7
Intrinsic motivation 0.85 44.3 197.53 25.34 147.2 44.92 199.46
Identified motivation 0.83 42.52 192.55 25.56 147.17 45.22 199.4
Introjected motivation 0.71 41.43 192.77 24.56 147.32 43.94 199.67
External motivation 0.77 44.82 202.64 29.5 164.34 43.27 199.81
Amotivation 0.76 45.82 205 23.99 147.4 43.33 199.8
Physical activity 0.23 64.2 62.27 115.69 189.81 157.85 248.34
Mental wellbeing 0.95 2.74 0.89 2.7 0.79 2.77 0.88
Note. α = median Cronbach's alpha coefficients over the three measurement occasions
Primary Analysis
Table 3.2 presents the conditional growth models which describe patterns of change
for social support, motivation, mental wellbeing and PA over time.
Table 3.2. Final conditional growth models describing changes in study variables over the 3
measurement occasions
Outcome variable
Social Support
β (SE)
Motivation Physical Activity Mental Wellbeing
Intercept 2.934 (0.024) -43.745 (9.820) 53.951 (4.592) 2.887 (0.030)
Linear time 0.053 (0.029) -0.341 (11.748) 71.318 (5.493) 0.020 (0.037)
Intervention -0.303 (0.049) -7.826 (19.826) 45.015 (9.271) -0.635 (0.064)
Intervention x Linear time -0.028 (0.056) 25.519 (22.906) -90.840 (10.710) 0.040 (0.072)
Note. Standard errors in brackets
Social support. The first part of the analysis indicated no significant linear effects of
time for levels of social support. Inclusion of the intervention variable as a predictor of the
intercept revealed participants in the online condition had lower levels of perceived social
support at baseline compared to the face-to-face condition. However, a non-significant
intervention × time interaction suggested no differences in the patterns of change in social
support between online and face-to-face conditions.
Motivation. The first part of the analysis indicated no significant linear effects of
time for motivation. Inclusion of the intervention variable as a predictor of the intercept
revealed no significant differences in motivation between the online and face-to-face
condition at baseline. Furthermore, a non-significant intervention × time interaction
suggested no differences in the patterns of change in motivation between online versus face-
to-face intervention conditions.
Mental Wellbeing. The first part of the analysis indicated no significant linear effects
of time for mental wellbeing. Inclusion of the intervention variable as a predictor of the
intercept revealed the online condition had lower levels of mental wellbeing at baseline
compared to the face-to-face condition. However, a non-significant intervention × time
interaction suggested no differences in the patterns of change in mental wellbeing between
online versus face-to-face intervention conditions.
Physical Activity. The first part of the analysis revealed significant linear increases in
time for PA. Inclusion of the intervention variable as a predictor of the intercept revealed the
online condition had higher levels of PA at baseline compared to the face-to-face condition.
Furthermore, a significant intervention × time interaction was observed, which indicated that
MHSU in the face-to-face condition increased their PA levels throughout the 6 months of
monitoring. Opposingly, the PA levels of those in the online intervention condition decreased
throughout the same time period (Figure 3.1).
Figure 3.1. Changes in PA levels over time between face-to-face and online intervention
conditions
Discussion
The aim of the current study was to examine if online versus face-to-face support
predicted changes in PA, motivation, social support and mental wellbeing of MHSU. Results
found that online versus face-to-face support did not predict changes in motivation, social
support or mental wellbeing. However, results did show that online versus face-to-face
support predicted changes in PA levels of MHSU, showing a significant increase in PA levels
for the face-to-face condition across the 6 months, and a significant decrease in PA levels for
the online condition.
Previous research has looked at the importance of MHSU’s priority for a healthy
lifestyle and found that this was a predictor of regular PA (Chapman et al., 2016). Those
individuals who considered a healthy lifestyle as high priority were more regularly physically
active, and those with a better self-perceived general health and fewer health problems which
limited PA, were also more regularly active (Chapman et al., 2016). MHSU participants in
the face-to-face support condition may have perceived PA as a higher priority in comparison
to participants in the online support condition as they had voluntarily signed up for the PA
programme. Individuals may have accessed the online platform for reasons other than to
encourage their own PA engagement. Face-to-face support may have generated greater social
integration of participants and facilitated a stronger sense of belonging for those experiencing
the supportive environment of the community programme, further encouraging greater levels
of PA engagement (Lubans et al., 2016). However, it is apparent that the decision for MHSU
to engage in PA is more complex than the nature of this quantitative data found. Further
Changes in PA over time between face-to-face and
online social support conditions
250
200
150
100
50
0
Baseline6 months F2F
Timepoint Online
Physical Activity (PA)
investigation via qualitative methods would allow for underlying factors, processes or
influencers that might be at play to explain the levels of PA in both online and face-to-face
support conditions.
With regards to social support, participants in the online condition had lower baseline
levels of social support compared to the face-to-face condition. For some individuals,
accessing the online platform may occur when they are in a low mood state. The decision to
reach out and connect with others to discuss personal, health-related issues may be at a time
when MHSU are facing significant life challenges and experiencing an increased level of
instability (Perry & Pescosolido, 2015). In support of previous research, online social support
can be readily accessed by MHSU with few other factors prohibiting individuals’ access
(Parikh & Huniewicz, 2015). Moreover, previous research found that individuls with
depression who percevied lower levels of social support had worse outcomes in terms of
recovery from mental illness, and social functioning (Wang et al., 2018). A lack of social
interaction was further linked with increases in depression outcomes (Wang et al., 2018).
MHSU’ perceptions of online social support in the current study findings may have been
tainted by their negative emotions linked to their mental illness. However, this goes beyond
the scope of the data collected for this study, and therefore explanations for differences at
baseline for social support between the two conditions cannot be explained. Future research
might find value in qualitatively exploring the motives behind why MHSU initially access
virtual social support via an online community platform.
No significant differences were found between the two conditions for perceived social
support over the 6 months. These findings oppose previous research which found clear
differences between online and face-to-face support. A study which addressed questions of
how satisfying and effective social support was perceived to be in online and offline contexts
by an adult population found that both environments positively influenced individuals’
satisfaction with social support over a longitudinal programme. Participants perceived more
emotional and instrumental support in face-to-face contexts compared to online (Trepte et al.,
2015). However, findings reported that users of online networks often have certain
expectations regarding the kind of support that can realistically be provided in an online
setting and know that deep emotional support should be sought from face-to-face contexts
(Trepte et al., 2014). A caveat of these study findings is the measuring of social support is
context sensitive, resulting in limited comparability across contexts (MHSU within a
community PA programme) and individual programme experiences. Again, this provides a
justification for the current study’s findings; however, the data obtained from quantitative
methods alone make it difficult to draw on any further explanations.
MHSU face a number of barriers to PA including physical health, a lack of money,
tiredness, disorganisation, exhaustion and embarrassment to engage in PA (Vancampfort, De
Hert, et al., 2017). Motivation is an additional barrier reported by MHSU with the initial step
to PA participation being MHSU greatest challenge (Firth et al., 2016). According to OIT, the
social environment can either facilitate or impede the internalisation process and
consequently affects both quality and quantity of the internalisation (Ryan et al., 2009).
Highly internalised behaviours typically relate to an empathic, supportive context such as the
environment created within the GStG programme. The current study found no differences or
changes in participant motivation between the face-to-face and online support conditions. The
lack of increase in motivation for either condition over time suggests that a combination of
barriers (e.g., physical health, fatigue, embarrassment, loneliness) may be present in this
complex relationship between MHSU’ motivation to engage in PA, which further research
needs to address in order to create a supportive environment that can facilitate motivation
towards PA engagement.
A further explanation for the lack of change in motivation over time between the two
conditions could be regarding the dynamics of behaviour regulations (Wasserkampf &
Kleinert, 2015). The function of introjected regulation in the context of behaviour adoption
needs to be considered. As originally proposed by OIT theorists, introjected regulation is
acknowledged as being less powerful in the changing process (e.g., changing behaviour) in
the long-term, given its controlled nature (Ryan et al., 2007). However, although introjected
regulation is perceived as controlling and pressurising, and thus associated with negative
affective states, this form of self-regulation might be helpful and supportive during the initial
process of MSHU adopting PA behaviour (Sabiston et al., 2010). Nonetheless, this type of
regulation may force individuals through feelings of guilt and shame to engage in PA
(Sabiston et al., 2010). Therefore, increased introjected regulation may be functional, but
only under certain conditions, individuals and contexts. Future research should consider the
dynamic nature of motivation regulations in order to facilitate the adoption of PA by MHSU.
Overall, there were no significant differences found for MHSU’ mental wellbeing
between the face-to-face and online support conditions. These findings within a mental health
population support previous research which found no differences in anxiety and depression
levels, or quality of life in MHSU between face-to-face and online support groups for cancer
patients (Huber et al., 2018). However, findings do not support the findings of a recent meta-
analysis which reported that positive communication reduces anxiety and develops feelings
of security (Harandi et al., 2017).
A lack of significant findings for MHSU’ mental wellbeing, across either support
condition, suggests a more complex underlying process. For example, the mental wellbeing
of MHSU neither increased nor decreased over time across the two support conditions. A
narrative review published evidence on the associations between mental health and
sociodemographic and economic factors at an individual-level and area-level (Silva,
Marques, & Teixeira, 2014). Across 78 studies, the main individual factors that had a
statistically significant independent association with worse mental health included a lack of
social support, female gender, low education, financial concerns and unemployment (Silva et
al., 2014). Associations were reported from 69 studies between area-level factors and mental
health, including social capital, geographical distribution, built environment and ethnicity
(Silva et al., 2014). This suggests that ameliorating economic situations of individuals by
enhancing community connectedness and combating neighbourhood disadvantages and social
isolation may enhance the populations’ mental health (Silva et al., 2014). However, findings
from this study demonstrate that the nuances of mental health cannot be captured in large
scale data collection methods such as the survey used in this study. Further research is
required to understand the deeper relationship between social support and the mental
wellbeing of MHSU.
Limitations and future directions. The inclusion of data from a large sample of
MHSU is a strength of this current study, with existing research predominantly recruiting
small sample sizes (Harandi et al., 2017). However, the depth of data collected via
quantitative methods restricts the ability to explain statistically insignificant findings without
drawing on assumptions. Current study findings should therefore be interpreted with some
caution due to methodological considerations. All measures used in this study were self-
report. Self-report measures are not fully reliable, as participants responses will be subject to
recall bias. In particular, the IPAQ PA measure required participants to report the amount of
time spent exercising (vigorously, or moderate intensity), walking and sitting in the past 7
days in hours and minutes. This may have led to inaccurate figures reported for duration and
intensity. Participants may also perceive PA differently based on their individual definitions
of, and what they feel should be classed as PA. Additionally, asking participants to report
based on the last 7 days may not have been entirely representative of their normal PA
behaviour, due to fluctuations in their mental health and symptoms associated with any
medication. This suggests that quantitative data alone cannot capture a true reflection of
MHSU behaviour. Future research should incorporate qualitative research to
understand MHSU individual experiences to better understand the mechanisms
and underlying processes, such as motivation, contributing to PA engagement.
The longitudinal design of this study was implemented to understand the
often- complex PA and mental health relationships where interlinked determinants
are present (Harandi et al., 2017). This was a strength of the current study, as
previous literature has predominantly implemented cross-sectional designs (Gunnell
et al., 2014). Despite this, longitudinal survey data can be prone to sampling errors
(Trepte et al., 2014). Recruiting for participants to complete a survey at multiple time
points is relying on participants time and commitment to the research once being sent
their follow-up survey. Although the use of online surveys to recruit within mental
health research is beneficial (i.e., to overcome barriers of face-to-face recruitment),
practice can be criticised due to the lack of environmental control compared to face-
to-face methods (Roivainen, Veijola, & Miettunen, 2016). For example inattentive
responding, defined as ‘responding without regard to the item content’ (Huang,
Bowling, Liu, & Li, 2015, pg. 157), has shown negative associations with high
emotional stability (Bowling et al., 2016). Moreover, higher levels of emotional
stability contribute to a degree of social competence which may in turn increase the
likelihood of accurate responding (Bowling et al., 2016). In the current study,
emotional instability which MHSU may have been experiencing, could have acted as
a distraction from careful and accurate responding. Therefore, future quantitative
research could consider inserting ‘bogus items’ as improbable statements that only
have one correct response to detect inattentive responders with the view of excluding
this data (Berry, Rana, Lockwood, Fletcher, & Pratt, 2019).
In alignment with previous research (Wasserkampf & Kleinert, 2015), this
current study assessed motivation using composite scorings such as the Relative
Autonomy Index (RAI: Ryan & Connell, 1989). To produce a RAI, the individual
regulation scores were weighted and then aggregated to form a numerical index
of the extent to which an individual’s behaviour is autonomously regulated or not
(Mullan & Markland, 1997).
However, is it likely that conceptual distinctions between single types of behaviour
regulation will be masked when using bundled scores (Mullan & Markland, 1997).
Applications of composite scorings fail to consider that the six forms of motivation
regulations are qualitatively different from one another (Wasserkampf & Kleinert,
2015). Therefore, future research should assess each regulation’s respective
contribution to MHSU motivation in order to make conclusions about PA behaviour
in the long term.
Despite the strengths that this longitudinal study adds to previous research,
there was a lack of opportunity for MHSU to express their personal dispositions or
experiences regarding social support within the community PA programme. As such,
the findings suggest that there are deeper social processes at play with MHSU
behaviour that are negated through the use of quantitative data alone. In light of
these concerns, the findings must be recognised as limited in their capacity to offer
generalisable recommendations or understanding of MHSU experiences. Instead,
they signpost areas for development and further study, and offer a foundation upon
which further inquiry regarding how social support impacts MHSU’ motivation to
engage in PA can be built.
Conclusion
The current study found a significant increase in PA levels of MHSU in the
face-to- face support condition. A significant decrease in PA levels was found for
the online support condition who had higher PA levels at baseline compared to the
face-to-face condition.
However, no significant differences were found between face-to-face and online
support conditions for motivation, perceived social support or mental wellbeing.
Additionally, there were no changes over time for motivation, social support or
mental wellbeing, but reportedly lower mental wellbeing at baseline for the online
condition compared to the face-to-face condition. This research considers the
complexity of studying the PA behaviour of MHSU including symptoms of their
mental illness or prescribed medication, as well as their differing needs and
perceptions of the social environment viewed as facilitative or inhibitive of PA
behaviour. Given the established benefits of PA participation for physical, mental
and cognitive health of MHSU, more effective programmes that target increasing PA
levels are required. There is a need for longitudinal studies with multiple time point
measures to clarify the relationship between social determinants, such as social
support, and mental health and wellbeing of MHSU. Research is required to develop
understanding of the barriers, motivation and preferences for PA in order to develop
programmes that aim to facilitate PA engagement among MHSU informed by their
needs and shaped by their priorities (Mishu et al., 2018). This requires both
quantitative and qualitative research moving forwards to ensure that programmes and
policy guidelines can be tailored to maximise positive effects among MHSU (White
et al., 2017).
References
Ainsworth, B. E., Caspersen, C. J., Matthews, C. E., Mâsse, L. C., Baranowski, T., & Zhu,
W. (2016). Recommendations to Improve the Accuracy of Estimates of Physical
Activity Derived from Self Report. Journal of Physical Activity and Health.
https://doi.org/10.1123/jpah.9.s1.s76
Aknin, L. B., Dunn, E. W., Sandstrom, G. M., & Norton, M. I. (2013). Does social
connection turn good deeds into good feelings? On the value of putting the “social” in
prosocial spending. International Journal of Happiness and Development.
https://doi.org/10.1504/ijhd.2013.055643
Ali, K., Farrer, L., Gulliver, A., & Griffiths, K. M. (2015). Online Peer-to-Peer Support for
Young People With Mental Health Problems: A Systematic Review. JMIR Mental
Health, 2(2), e19. https://doi.org/10.2196/mental.4418
Allen, J., Balfour, R., Bell, R., & Marmot, M. (2014). Social determinants of mental health.
International Review of Psychiatry, 26(4), 392–
407.
https://doi.org/10.3109/09540261.2014.928270
Andersen, S. M., & Berk, M. S. (1998). Transference in everyday experience:
Implications of experimental research for relevant clinical phenomena. Review of
General Psychology. https://doi.org/10.1037/1089-2680.2.1.81
Anderson, K., Laxhman, N., & Priebe, S. (2015). Can mental health interventions change
social networks? A systematic review. BMC Psychiatry, 15(1).
https://doi.org/10.1186/s12888- 015-0684-6
Andrew, S., & Halcomb, E. J. (2006). Mixed methods research is an effective method of
enquiry for community health research. Contemporary Nurse.
https://doi.org/10.5172/conu.2006.23.2.145
Ashida, S., & Heaney, C. A. (2008). Differential associations of social support and
social connectedness with structural features of social networks and the health
status of older adults. Journal of Aging and Health.
https://doi.org/10.1177/0898264308324626
Bellamy, C., Schmutte, T., & Davidson, L. (2017). An update on the growing evidence base
for peer support. Mental Health and Social Inclusion. https://doi.org/10.1108/MHSI-
03-2017- 0014
Berger, M., Wagner, T. H., & Baker, L. C. (2005). Internet use and stigmatized illness.
Social Science and Medicine. https://doi.org/10.1016/j.socscimed.2005.03.025
Berry, K., Rana, R., Lockwood, A., Fletcher, L., & Pratt, D. (2019). Factors associated
with inattentive responding in online survey research. Personality and Individual
Differences. https://doi.org/10.1016/j.paid.2019.05.043
Bidee, J., Vantilborgh, T., Pepermans, R., Huybrechts, G., Willems, J., Jegers, M., & Hofmans,
J. (2013). Autonomous Motivation Stimulates Volunteers’ Work Effort: A
Self- Determination Theory Approach to Volunteerism. Voluntas, 24(1), 32–
47. https://doi.org/10.1007/s11266-012-9269-x
Bledin, K., Loat, M., Caffrey, A., Evans, K. B., Taylor, B., & Nitsun, M. (2016). ‘Most
Important Events’ and Therapeutic Factors: An Evaluation of Inpatient Groups for
Peoplewith Severe and Enduring Mental Health Difficulties. Group Analysis.
https://doi.org/10.1177/0533316416675442
Bopp, M., & Fallon, E. (2008). Community-based interventions to promote increased
physical activity: A primer. Applied Health Economics and Health Policy, 6(4), 173–
187. https://doi.org/10.2165/00148365-200806040-00001
Bowling, N. A., Huang, J. L., Bragg, C. B., Khazon, S., Liu, M., & Blackmore, C. E.
(2016). who cares and who is careless? Insufficient effort responding as a reflection of
respondent personality. Journal of Personality and Social Psychology, 111(2), 218–
229. https://doi.org/10.1037/pspp0000085
Brand, T., Pischke, C. R., Steenbock, B., Schoenbach, J., Poettgen, S., Samkange-Zeeb, F.,
& Zeeb, H. (2014). What works in community-based interventions promoting physical
activity and healthy eating? A review of reviews. International Journal of
Environmental Research and Public Health. https://doi.org/10.3390/ijerph110605866
Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Using
Qualitative Research in Psychology, 3, 77–101. https://doi.org/The publisher’s
URL is: http://dx.doi.org/10.1191/1478088706qp063oa
Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative
Research in Psychology, 3(May 2015), 77–101.
https://doi.org/10.1191/1478088706qp063oa
Braun, V., & Clarke, V. (2014). What can “thematic analysis” offer health and wellbeing
researchers? International Journal of Qualitative Studies on Health and Well-
Being, 9. https://doi.org/10.3402/qhw.v9.26152
Braun, V., & Clarke, V. (2019). Reflecting on reflexive thematic analysis. Qualitative
Research in Sport, Exercise and Health.
https://doi.org/10.1080/2159676X.2019.1628806
Bryant, W., Tibbs, A., & Clark, J. (2011). Visualising a safe space: The perspective of
people using mental health day services. Disability and Society.
https://doi.org/10.1080/09687599.2011.589194
Burr, J. A., Han, S. H., & Tavares, J. L. (2016). Volunteering and Cardiovascular Disease
Risk: Does Helping Others Get “under the Skin?” Gerontologist, 56(5), 937–947.
https://doi.org/10.1093/geront/gnv032
Burton, E., Farrier, K., Hill, K. D., Codde, J., Airey, P., & Hill, A. M. (2018). Effectiveness
of peers in delivering programs or motivating older people to increase their
participation in physical activity: Systematic review and meta-analysis. Journal of
Sports Sciences, 36(6), 666–678. https://doi.org/10.1080/02640414.2017.1329549
Carey, E. C., & Weissman, D. E. (2010). Understanding and finding mentorship: A review
for junior faculty. Journal of Palliative Medicine.
https://doi.org/10.1089/jpm.2010.0091
Carless, D., & Douglas, K. (2008a). Social support for and through exercise and sport in a
sample of men with serious mental illness. Issues in Mental Health Nursing, 29(11),
1179– 1199. https://doi.org/10.1080/01612840802370640
Carless, D., & Douglas, K. (2008b). The contribution of exercise and sport to mental
health promotion in serious mental illness: An interpretive project. International
Journal of Mental Health Promotion, 10(4), 5–12.
https://doi.org/10.1080/14623730.2008.9721771
Carless, D., Douglas, K., & D., C. (2012). The ethos of physical activity delivery in
mental health: A narrative study of service user experiences. Issues in Mental
Health Nursing, 33(3), 165–171.
https://doi.org/http://dx.doi.org/10.3109/01612840.2011.637659
Carneiro, L. S. F., Fonseca, A. M., Vieira-Coelho, M. A., Mota, M. P., & Vasconcelos-Raposo,
J. (2015). Effects of structured exercise and pharmacotherapy vs. pharmacotherapy
for adults with depressive symptoms: A randomized clinical trial. Journal of
Psychiatric Research, 71(July), 48–55.
https://doi.org/10.1016/j.jpsychires.2015.09.007
Cattan, M., Kime, N., & Bagnall, A. M. (2011). The use of telephone befriending in low
level support for socially isolated older people - an evaluation. Health and Social
Care in the Community, 19(2), 198–206. https://doi.org/10.1111/j.1365-
2524.2010.00967.x
Charlesworth, G., Sinclair, J. B., Brooks, A., Sullivan, T., Ahmad, S., & Poland, F. (2017).
The impact of volunteering on the volunteer: findings from a peer support programme
for family carers of people with dementia. Health and Social Care in the Community,
25(2), 548–558. https://doi.org/10.1111/hsc.12341
