Cognition and functioning in bipolar disorder
: exploring treatment response and mechanisms of cognitive remediation therapy

Student thesis: Doctoral ThesisDoctor of Philosophy


Bipolar disorder (BD) is a debilitating mental health condition characterised by recurrent mood episodes. Cognitive and functional impairment are prominent features of BD affecting a substantial proportion of patients. Cognitive deficits often persist after the remission of acute mood symptoms, during phases of euthymia, while functional recovery following acute episodes is suboptimal and significantly associated with cognitive outcomes. However, pharmacological treatments exert few effects on cognitive deficits and even fewer on functional difficulties. Cognitive remediation (CR) is an intervention specifically targeting cognition and functioning, with initial findings indicating meaningful benefits in these outcomes for people with BD. Though, the existing literature presents limited examination and understanding in regard to mechanistic questions, such as for which patients or under which circumstances is CR more beneficial, whether and how treatment effects on cognition transfer to functional outcomes, and which are the characteristics or components of therapy contributing to treatment effects.

Using cross-sectional and longitudinal data from a randomised controlled trial assessing the effects of CR against treatment-as-usual (TAU) in euthymic patients with BD, this PhD thesis comprises five empirical studies. The overarching aims of these studies are: a) to characterise cognition and functioning in this cohort by estimating the proportion of cognitive and functional impairment and identifying the patient characteristics associated with baseline performance in cognitive and functional outcomes (Studies 1-2), and b) to explore differential treatment response (Study 3) and specific treatment mechanisms of CR (Studies 4-5).

A substantial proportion of the CRiB cohort (34%) presented with clinically significant cognitive impairment according to a validated definition, while a model including premorbid IQ, number of current medications, and previous psychological therapies was able to significantly differentiate between participants with and without cognitive impairment (Study 1). An even larger proportion of euthymic participants (54%) experienced moderate or severe functional difficulties. These difficulties were associated with higher residual depressive symptoms, poorer executive skills and more subjective cognitive complaints (Study 2). Factors associated with cognition and functioning at baseline, along with other factors previously examined in the literature, were considered as potential moderators of the CR effect on post-treatment outcomes. This effect was not moderated by any baseline factors for global cognition and psychosocial functioning. For personal recovery goals though, attainment was greater for patients with higher cognitive complaints at baseline and more previous experience with psychological therapies (Study 3). An examination of transfer mechanisms indicated that cognitive gains following CR translate into functional improvements at follow-up, but the model of transfer appears to differ between functional outcomes. Global cognition partially mediated the effect of CR on psychosocial functioning, while the cognitive level achieved after treatment moderated the CR effect on goal attainment (Study 4). Finally, session attendance and core therapeutic components of were explored as factors potentially underlying treatment effects. Achieving therapy completion according to a pre-defined cut-off of 20 sessions improved the effect of the therapy across outcomes. No therapy components were associated with cognitive gains, but spending time in contact with the therapist and implementing useful strategies were associated with greater benefits in functional outcomes (Study 5).

Taken together, these results contribute to our better understanding of the factors affecting response to CR, the way cognitive gains translate into functional benefits, and the therapy characteristics contributing to improved outcomes. Pending replication from future research, findings from this thesis provide promising indications that CR can produce meaningful benefits for most patients with BD, independently of their pre-treatment characteristics, and that changing cognition can improve functional outcomes in the long-term. Therapy engagement and certain therapeutic components might be vital for deriving greater benefits. Prospectively integrating CR to clinical practice may significantly facilitate the recovery process in patients with BD.
Date of Award1 Jan 2022
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorAllan Young (Supervisor), Til Wykes (Supervisor) & Matteo Cella (Supervisor)

Cite this