An interdisciplinary investigation of the caffeine-physical activity-affect relationship in individuals with bipolar disorder

  • Sammy Eden

Student thesis: Doctoral ThesisPhD


Introduction: Bipolar Disorder is a complex, multifaceted, chronic mood disorder
that causes significant distress and disruption to everyday quality of life for
hundreds of millions of individuals worldwide. The condition has no cure and so
requires lifelong treatment and self-management. Alteration of lifestyle
behaviours such as caffeine and physical activity offer accessible ways
individuals could self-manage their condition alongside more traditional
pharmaceutical and psychological treatments. Specifically, caffeine and physical
activity behaviours are both stimulating behaviours that co-occur together on a
daily basis. Several theoretical biobehavioural and cognitive accounts of bipolar,
such as the Behavioural Activation System (BAS; Depue et al., 1987; Depue &
Iacono, 1989) and the integrative cognitive model of bipolar (Mansell et al., 2007)
infer the potential for caffeine and physical activity to interact with affective state,
for example acting as both a trigger and a coping response. At the same time
however, little research has directly conceptualised or explored the potential
relationships between caffeine, physical activity and affective state.
Correspondingly, the overall aim of this thesis is to further our understanding of
the caffeine-physical activity-affect relationship in individuals with bipolar through
a combination of interconnecting studies that adopt an interdisciplinary approach,
considering both caffeine and physical activity in isolation as well as jointly in
parallel. The ultimate intention was to be able to translate our findings to a reallife
setting so that individuals with bipolar can make better, more informed choices
about their caffeine consumption and physical activity behaviours and be
supported in this by health-care professionals

Study 1: First, we conducted qualitative interviews with individuals with bipolar
(n= 20). The aim of this study was to explore in depth the personal experiences
of caffeine consumption and engagement in physical activity behaviours in this
patient group in relation to affect and individuals’ bipolar mood symptoms.
Interviews were then analysed using interpretative phenomenological analysis
(IPA) to understand participants lived personal experiences of their caffeine use
and physical activity behaviours.

Study 2: Second, we sought to observe and examine the associations in a real
life setting over time, employing an experience sampling method (ESM)
approach. Individuals (n= 63) with bipolar, unipolar depression and unaffected
controls completed 7-days of lifestyle monitoring of their affective state, caffeine,
and physical activity behaviours. The primary aim was to compare three groups
in terms of their physical activity and caffeine use behaviour, and in terms of the
relationship between these behaviours and affect. It was hypothesised that the
relationship between caffeine use and affect, and physical activity behaviours and
affect would be more apparent in those with a history of bipolar, such that there
would be greater variability in these behaviours and a stronger positive
association across time in those with bipolar relative to those without. We used
ANOVAs to see if there were differences in variability in these behaviours
between the three groups, and also multilevel modelling to assess the association
between caffeine, physical activity and affect across time across the three groups.

Study 3: Finally, we conducted a series of focus groups with individuals with lived
experience of bipolar and clinical professionals. The aim was to inform the
potential development of resources for patients (with bipolar) and clinicians based
upon findings of the caffeine-bipolar relationship from Study 1. This will provide a
first step towards further research and future development of resources, which
could potentially be delivered as part of a larger complex interventions in the

Results and conclusion: Overall, the triad of interlinking studies conducted as
part of this thesis suggests that there is a meaningful relationship between
caffeine and physical activity and affect in individuals with bipolar, although which
(if any) aspects of this are unique to bipolar disorder is currently unclear. Findings
highlight how caffeine appears to have been largely overlooked by bipolar
patients and healthcare professionals. Furthermore, the research highlights the
complexities of these relationships, as both caffeine and physical activity can be
both helpful and harmful, specifically for some individual’s caffeine requires a very
delicate balancing act. Findings emphasize the importance of increasing
knowledge surrounding caffeine and physical activity behaviours for both patients
with bipolar and clinical professionals, as both these lifestyle behaviours have the
potential to influence mood state. Future research in this area is needed to further
increase our understanding of the caffeine-physical activity-affect relationship. This will assist us to build a body of knowledge that can inform the development
of future lifestyle interventions, assisting in symptom management and improving
quality of life for those with bipolar disorder.
Date of Award14 Sept 2022
Original languageEnglish
Awarding Institution
  • University of Bath
SponsorsEconomic and Social Research Council
SupervisorJavier Gonzalez (Supervisor), Kim Wright (Supervisor) & James Betts (Supervisor)

Cite this