SYMPTOM CLUSTERS IN CANCER AND PALLIATIVE CARE: Edited by Andrea M. Barsevick and Aynur Aktas
Clinicians and researchers interested in palliative care have a keen interest in the art and science of symptom management. As the field has grown, the concept of the ‘symptom cluster’ has come into its own as a problem worthy of attention in its own right. Previously, symptoms were understood as distinct entities, each with a separate management plan. Although we recognized that some symptoms influenced other symptoms – such as, fatigue and sleep disturbance or pain and depression – the evidence base for symptom management relied primarily on studies of individual symptoms. It has become clear to clinicians and researchers that a more holistic approach is needed to better understand the patient experience.
Our clinical and scientific understanding of symptom clusters has increased as the quantity and quality of research has grown. This issue of Current Opinions in Supportive and Palliative Care puts the spotlight on what we have learned about symptom clusters: how the construct is defined conceptually and methodologically; analytical methods for identifying clusters; proposed biological mechanisms; approaches to intervention; and the application of this construct to children as a special population. Each author has been charged with reviewing the most recent literature within the domain of their topic and presenting their opinions of the contribution of that literature to our overall understanding of symptom clusters. By asking a multidisciplinary group to examine the most recent literature, we hope to provide you with a current picture of this problem and ideas for moving forward clinically and scientifically.
Despite more than 20 years since the term ‘symptom cluster’ was applied to the study of multiple symptoms related to cancer [1,2], there are still conceptual and methodological controversies about it. Dr Aynur Aktas addresses the issue of whether two or three symptoms are required to meet the definition of a symptom cluster. She advises that ‘it is prudent to view symptom pairs as a cluster when they interact… in predicting outcomes (pp. 38–44)’. The author also demonstrates that there is substantial evidence, both qualitative and quantitative, that the psychoneurological cluster – fatigue, insomnia, pain, depression, and cognitive changes – is a very real and prevalent cluster that can negatively impact morbidity and quality of life. She also draws attention to the problem of symptom cluster stability over the course of a disease. In the changing landscape of symptoms over time, she highlights research identifying core and sentinel symptoms [3–5] that has moved the discussion of stability from a theoretical concern to an analytical challenge.
A wide variety of methods have been used for statistical modeling of symptom clusters. Dr Hee Ju Kim and coworkers (pp. 45–53) describe and provide examples of powerful techniques for identifying symptom clusters including the use of latent class analysis (LCA) and structural equation modeling (SEM). LCA is used for grouping individuals with similar symptom patterns; it has similarities to cluster analysis, but it provides stronger evidence of model fit than cluster analysis. SEM can evaluate relationships among multiple variables, mediators, and outcomes and thus, determine both direct and indirect effects on outcome variables. The authors also describe the use of longitudinal methods (e.g., latent transition analysis), never before applied in oncology, that could be used to evaluate the directional or causal relationships among symptoms. This could provide clues about the underlying mechanism of a symptom cluster.
There is increasing interest in the underlying biology of symptom clusters [6,7]. Dr Lisa Wood and coworkers (pp. 54–59) use the model of cancer patients exposed to cytotoxic chemotherapy to examine the underlying biology of sickness behavior, a collection of symptoms including malaise, sleep disturbance, cognitive difficulties, pain, depressed mood, and decreased appetite. They describe the relationship of cancer-related symptoms with inflammatory processes based on both laboratory and clinical research. This carefully conducted review summarizes the evidence to support the proposition that IL-1β, a proinflammatory cytokine, plays a central role in the induction of sickness behavior. This knowledge may lead to the development of novel treatments targeting IL-1β.
Dr Ann Berger and coworkers (pp. 60–66) provide perspective on 24 intervention studies published since 2009, 18 in early stage and six in advanced cancer. While acknowledging the wide variety of interventions addressing many different symptom clusters, the authors highlight a number of different approaches to symptom cluster management. Examples include cognitive behavioral therapy, yoga, relaxation acupressure, and mindfulness-based stress reduction. Importantly, these nonpharmacological interventions seemed to have the most beneficial effects alleviating multiple symptoms possibly due to their lack of specificity and broad spectrum of influence.
The vast majority of clinical research on symptom clusters has used adults as the model. Work on symptom clusters in children is in early development. Dr Marilyn Hockenberry and coworkers (pp. 67–72) focused their literature review on a variety of disorders including cancer to provide a broad view of the theoretical and empirical underpinnings of symptom clusters in children. This includes describing the most common and distressing symptoms children experienced at different points in the disease trajectory. The authors suggest that ‘assessing symptom clusters provides a holistic depiction of the child's experience’. This article brings a developmental science perspective to research on symptom clusters. This is particularly important because multiple symptoms could have deleterious effects on childhood development. Undoubtedly, this would necessitate evaluating symptom interactions and developmental changes in symptom severity over time.
We hope this collection of papers provides the reader with a broad picture of symptom clusters as we understand them today. Based on this review, we believe that this is a promising area for future research with the potential to yield valuable evidence for translation into clinical practice.
Cleveland Clinic The Harry R. Horvitz Center for Palliative Medicine and Supportive Oncology, Cleveland, Ohio, USA is a WHO Demonstration Project in Palliative Medicine and ESMO Designated Center of Integrated Oncology and Palliative Care.
Conflicts of interest
There are no conflicts of interest.
1. Dodd M, Janson S, Facione N, et al. Advancing the science of symptom management. J Adv Nurs 2001; 33:668–676.
2. Dodd MJ, Miaskowski C, Paul SM. Symptom clusters and their effect on the functional status of patients with cancer. Oncol Nurs Forum 2001; 28:465–470.
3. Skerman HM, Yates PM, Battistutta D. Identification of cancer-related symptom clusters: an empirical comparison of exploratory factor analysis methods. J Pain Symptom Manage 2012; 44:10–22.
4. Skerman HM, Yates PM, Battistutta D. Cancer-related symptom clusters for symptom management in outpatients after commencing adjuvant chemotherapy, at 6 months, and 12 months. Support Care Cancer 2012; 20:95–105.
5. Brown JK, Cooley ME, Chernecky C, et al. A symptom cluster and sentinel symptom experienced by women with lung cancer. Oncol Nurs Forum 2011; 38:E425–E435.
6. Illi J, Miaskowski C, Cooper B, et al. Association between pro- and anti-inflammatory cytokine genes and a symptom cluster of pain, fatigue, sleep disturbance, and depression. Cytokine 2012; 58:437–447.
7. Wang XS, Fairclough DL, Liao Z, et al. Longitudinal study of the relationship between chemoradiation therapy for nonsmall-cell lung cancer and patient symptoms. J Clin Oncol 2006; 24:4485–4491.