Purpose
Psychological constructs independently affect health outcomes and quality of life in cancer patients, and exercise can influence these constructs. However, exercise oncology research tends to focus on negative psychological constructs (e.g., depression), with limited evidence exploring positive psychological constructs (e.g., optimism). This systematic review and meta-analysis examined which psychological constructs have been utilised, and which have been found to effectively respond to exercise.
Methods
A systematic search across four databases was performed following PRISMA guidelines up until February 2024. Randomised controlled trials (RCTs) analysing the effects of exercise on any psychological constructs in any oncological population were included, limited to full-text articles published in English in peer-reviewed journals. Full texts were screened by two independent reviewers, and an iterative process was performed to determine the conceptual categories. Random-effects meta-analyses were conducted using Jamovi, and study quality was assessed using the Delphi list tool.
Results
A total of 189 RCTs were included. Forty-three tools measuring psychological wellbeing were thematically consolidated into three positive (social, quality of life (psychological), generic (wellbeing)) and three negative psychological constructs (anxiety, depression, generic (distress)). Significant small effects favouring exercise were found for psychological wellbeing using EORTC (emotional: SMD = 0.20, p<0.001; social: SMD = 0.21, p<0.001), FACT (only emotional: SMD = 0.21, p<0.001 only), HADS (only depression SMD = 0.25, p<0.001), CES-D (SMD = 0.19, p=0.005), and SF-36 (only mental component score SMD = 0.36, p=0.002) tools. However, pooled analysis for generic (well-being) and anxiety were limited, warranting further research.
Conclusions
Considerable variability was observed in the use of unidimensional and multidimensional tools, alongside inconsistent reporting of results across exercise oncology trials. Although findings varied across psychological constructs, cancer characteristics, exercise prescription, and the content and psychometric properties of the measurement tool should collectively inform the selection of tool when assessing mental health outcomes.