Abstract
Background: Better self-management control in cancer survivors would benefit their functional status, quality of life, and health service utilization. Factors such as self-efficacy, social support, and coping style are important predictors of self-management behaviors of cancer survivors; however, the impact of these factors on self-management behaviors has not yet been empirically tested in Chinese cancer survivors.
Objectives: The aim of this study was to examine how self-efficacy, social support, and coping style affect specific self-management behaviors.
Methods: A secondary data analysis was completed from a cross-sectional study. A total of 764 cancer survivors were recruited in the study. Validated instruments were used to assess patients' self-efficacy, social support, and coping style. Structural equation modeling (SEM) was used to test the hypothesis.
Results: The SEM model fits the data very well, with root mean square error of approximation (RMSEA) of 0.034; close-fit test cannot reject the hypothesis of root mean square error of approximation of 0.05 or less, comparative fit index of 0.91, Tucker-Lewis index of 0.90, and weighted root mean square residual of 0.82. For the measurement models in the SEM, all items loaded highly on their underlying first-order factors, and the first-order factors loaded highly on their underlying second-order factors (self-efficacy and social support, respectively). The model demonstrated that self-efficacy and social support directly and indirectly, via coping style, affect 3 self-management behaviors (ie, communication, exercise, and information seeking).
Conclusion: Our results provide evidence that self-efficacy and social support impose significant direct effects, as well as indirect effects via copying style, on the self-management of cancer survivors.
Implications for Practice: Our findings may help nurses to further improve their care of cancer survivors in terms of their self-management behaviors, specifically communication, exercise, and information seeking.