Symptom Clusters in Patients with Chronic Obstructive Pulmonary Disease in China
: A Mixed Methods Study

Student thesis: Doctoral ThesisDoctor of Philosophy

Abstract

Background:
Patients with COPD typically experience multiple concurrent symptoms which have a detrimental impact on patient outcomes. Research in symptom management has predominantly focused on investigating single symptoms in isolation. Although the literature on symptom clusters is growing, the composition of symptom clusters differs depending on the various clinical variables as well as the methods of assessment. Effective management of symptom clusters in COPD patients remains a clinical challenge across healthcare settings. No attention has been given to gaining a broad and in-depth understanding of the experiences, perspectives and attitudes of COPD patients with symptom clusters. 
 
Aim:
To (i) identify symptom clusters, their associated factors, and impact on health-related outcomes and (ii) explore patient’s perspectives of symptom clusters in order to develop a theoretical understanding of their meanings and impact among patients with COPD in China, to inform the development of appropriate interventions and service developments. 
 
Methods:
A mixed-methods study with a concurrent triangulation design examined symptom clusters in patients with COPD. First, in the background chapter, a systematic review was conducted to examine and appraise evidence of symptom clusters, their compositions, and associated factors and appraise the methodologies of studies that report symptom clusters in patients with COPD. Second, also in the background chapter, a secondary analysis of existing data was performed to empirically test the Theory of Unpleasant Symptoms and examine evidence of synergistic impact of the symptom cluster on health-related quality of life in patients with COPD. Third, a quantitative cross-sectional study was conducted to identify symptom clusters and develop a symptom cluster model to investigate the interactions among symptom clusters, associated factors and health-related quality of life in COPD patients (n=450). Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) were used for the identification of the symptom clusters. Finally, structural equation modelling was involved in quantifying the potential relationships among symptom clusters, associated factors and health-related quality of life. Last, qualitative face-to-face, semi-structured interviews among 30 COPD patients explored the meaning patients attributed to their symptom clusters and their impacts on the health-related quality of life, and strategies they adopted to live alongside their COPD-related symptom clusters. 
 
Results:
The systematic review: Ten studies were eligible for inclusion in this study. Four common symptom clusters were identified. Two theoretical frameworks, four statistical methods, and various symptom assessment tools were used to identify symptom clusters. Factors associated with symptom clusters included demographic, clinical, and biological factors. No studies examined the subjective experiences of symptom clusters. Overall, inconsistencies were identified in the composition of symptom clusters across studies. This may be due to variations in study design, assessment tools, and statistical methods. Future studies should attempt to arrive at a common definition, especially that is theoretically derived, for symptom clusters, standardise the criteria for symptoms for inclusion in the clusters, and focus on patient’s subjective experience to inform which clusters are clinically relevant. 
 
The secondary analysis: The sample included 106 COPD patients from whom three symptom clusters were identified comprising of dyspnoea-depression, anxiety-sleep and depression- anxiety. Depression-anxiety (psychological symptom cluster) was significantly associated with poorer health-related quality of life (β=13.88, 95%CI:7.94-19.82), while no significant associations were detected with health-related quality of life for either depression or anxiety alone (β=6.66, 95%CI: -2.99-16.31; β=7.29, 95%CI: -0.78-15.35). Therefore, this study identified that the psychological symptom cluster (anxiety-depression cluster) had a synergistic impact on health-related quality of life. This finding empirically supported the ‘Theory of Unpleasant Symptoms’ that assumed symptoms in a cluster were shown to exert synergistic effects on health-related quality of life in COPD patients. 
 
The quantitative study: Respiratory related symptom cluster, psychological symptom cluster and cough-insomnia related symptom cluster were identified. The final model demonstrated a good fit with the data (c2/df = 2.982, CFI =0.970, TLI =0.920, RMSEA =0.066 and SRMR=0.026). Gender, stage of disease and monthly income were significant factors associated with symptom clusters.
Respiratory related and cough-insomnia related symptom clusters had a direct negative impact on health-related quality of life, while the psychological symptom cluster was found to have a direct and indirect negative effect on health-related quality of life. The final COPD symptom cluster model should serve as a framework to guide intervention research targeting symptom clusters to improve health-related quality of life of people living with COPD. Healthcare professionals should be especially attuned to identify those at most risk of facing a higher symptom burden in this case those who are female, have advanced stage COPD and/or lower income. During the clinical symptom assessment, healthcare professionals should pay attention to the close relationships among symptoms within a cluster to identify any ‘trigger’ symptom that could cause the development or exacerbation of other symptoms. 
 
The qualitative study: Thirty participants were included, and three themes were generated: (1) meaning of symptom clusters (i.e., the typologies and interrelationships among symptoms, variations in concurrent symptoms, factors contributing to symptom burden). (2) impact of symptom clusters on health-related quality of life (i.e., decreased physical functioning, social isolation, limitation in aesthetic needs and feelings of being stigmatised). (3) symptom coping strategies (i.e., mental-spiritual, self-management and medication control). This exploratory qualitative study provided insight into how people with COPD perceive and interpreted their lived experiences of symptom clusters. The early identification of COPD patients with symptom clusters and recognition of the central role that ‘trigger symptom’ play in experiences of multiple concurrent symptoms may provide insight into the future development of ‘trigger-based’ symptom cluster interventions. The studied patient’s symptom experiences were found to be complex and dynamic. Healthcare professionals, therefore, need to assess symptoms on an ongoing basis, be aware of ‘trigger’ symptom and explore underlying mechanisms. The broad range of multifaceted impacts of symptom clusters on patient’s quality of life identified in this study can help healthcare professionals develop holistic symptom management interventions. The several mechanisms of coping (e.g., spiritual support) and cultural perspectives identified in this study must be considered, developed, and tested in future healthcare professionals-led symptom management interventions. 
 
Conclusions:
This novel study, conducted in China, aimed at understanding the phenomenon of symptom clusters in patients living with COPD represents a major contribution to the evidence of a clinically important issue in that it adopts a mixed-methods design. Future research needs to target translating the growing body of symptom cluster literature into clinically relevant guidelines for managing symptom clusters and interventions across settings, to support a move towards advancing healthcare professionals and research for symptom management in COPD patients in China and elsewhere.
Date of Award1 Oct 2022
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorWei Gao (Supervisor), Jonathan Koffman (Supervisor) & Richard Siegert (Supervisor)

Cite this

'