Lin Du1, Hai-Yan Shi2, Yan Qian3, Xiao-Hong Jin3, Yang Li4, Hai-Rong Yu5, Xiao-Man Liu2, Xue-Lei Fu1, Hong-Lin Chen4. 1. School of Medicine, Nantong University, Nantong, China. 2. Department of Thoracic Oncology, Rugao People's Hospital, Nantong, China. 3. Department of Oncology, Affiliated Hospital of Nantong University, Nantong, China. 4. School of Public Health, Nantong University, Nantong, China. 5. Department of Thoracic Oncology, Affiliated Hospital of Nantong University, Nantong, China.
Abstract
OBJECTIVE: To quantitatively examine the relationship between social support and suicidal ideation (SI) among patients with cancer and identify the moderators that influence the magnitude of this association. METHODS: Publications were searched in PubMed, PsycINFO, EMBASE, Cochrane Library and Chinese National Knowledge Infrastructure from database inception to May, 2020. Correlation coefficients (r) were chosen as the effect size with a random model to evaluate the overall effect size between social support and SI in patients with cancer. To assess statistical heterogeneity, we examined potential moderator variables on the social support and SI. RESULTS: A total of 881 studies were identified in initial search, and twelve studies were eligible. A negative, small but significant correlation (r = -0.22, 95% CIs: -0.30,-0.14, p < 0.001) was observed between social support and SI in patients with cancer, with a significant heterogeneity (I2 = 95.24%, Q = 231.27, p < 0.001). Moderator analyses indicated that race/ethnicity (Q(1) = 8.4, p < 0.05) and measurements of social support (Q(3) = 9.78, p < 0.05) and SI (Q(3) = 9.69, p < 0.05) significantly moderate the effect size between social support and SI. CONCLUSION: Taken together, we found a negative yet significant association between social support and SI in patients with cancer, which supported the importance of social support for the prevention of SI in patients with cancer.
OBJECTIVE: To quantitatively examine the relationship between social support and suicidal ideation (SI) among patients with cancer and identify the moderators that influence the magnitude of this association. METHODS: Publications were searched in PubMed, PsycINFO, EMBASE, Cochrane Library and Chinese National Knowledge Infrastructure from database inception to May, 2020. Correlation coefficients (r) were chosen as the effect size with a random model to evaluate the overall effect size between social support and SI in patients with cancer. To assess statistical heterogeneity, we examined potential moderator variables on the social support and SI. RESULTS: A total of 881 studies were identified in initial search, and twelve studies were eligible. A negative, small but significant correlation (r = -0.22, 95% CIs: -0.30,-0.14, p < 0.001) was observed between social support and SI in patients with cancer, with a significant heterogeneity (I2 = 95.24%, Q = 231.27, p < 0.001). Moderator analyses indicated that race/ethnicity (Q(1) = 8.4, p < 0.05) and measurements of social support (Q(3) = 9.78, p < 0.05) and SI (Q(3) = 9.69, p < 0.05) significantly moderate the effect size between social support and SI. CONCLUSION: Taken together, we found a negative yet significant association between social support and SI in patients with cancer, which supported the importance of social support for the prevention of SI in patients with cancer.