Cancer is a major public health problem globally. Due to the aging population and the development of medical screening technology, the number of new cases of cancer diagnosed has been increasing year by year (Irwin, 2013). For example, in the United States, recent research reported that there were over 18 million cancer survivors nationwide, accounting for around 5 % of the total population (Miller et al., 2022). Cancer patients often experience impaired social functioning, reduced quality of life, and a loss of will to live, and thereby have an increased vulnerability to psychological problems (Mitchell et al., 2013). A meta-analysis revealed that the prevalence of depressive symptoms among cancer patients was up to 44.63 % (95%CI: 42.24 %–47.01 %) in China (Ding et al., 2023), which is six to seven times greater than the general population (Lu et al., 2021). Furthermore, insomnia symptoms are also common in cancer patients. For instance, a meta-analysis found that 45 % (95%CI: 33 %–58 %) of head and neck cancer patients globally had insomnia symptoms (Santoso et al., 2019), approximately three times higher than that in the general population (Wong et al., 2023). With the rapid improvement of diagnostic and treatment methods, the survival rate of cancer patients has steadily increased, with more than two-thirds of patients with cancer having an extended survival period (Miller et al., 2022). Hence, the mental health of cancer patients is critically important for both their recovery from cancer itself and their long-term quality of life (Chu et al., 2021; Cicchetti et al., 2022).

Psychiatric disorders or syndromes encompass distinct individual symptoms. For instance, depression consists of a cluster of individual symptoms such as hopelessness, worthlessness, sadness, anhedonia and impaired concentration (Thapar et al., 2022). Similarly, insomnia comprises various symptoms like difficulty falling asleep, difficulty staying asleep, and early awakening (Sutton, 2021). Neuropsychological mechanisms underlying different symptoms are often distinct and interrelated (Hackman and Farah, 2009). However, most studies on depression and insomnia in cancer patients only focused on the total or average scores of standardized rating scales, rather than examining the individual symptoms. This approach may mask the substantial differences among individual symptoms and the interrelationship between them (Fried and Nesse, 2015). Network analysis, an advanced data analysis approach, however, can be a useful tool to address this limitation.

Network analysis is a novel statistical approach that has been widely used in recent years to establish an ordered spatial network and elucidate the relationship between multiple symptoms simultaneously (Beard et al., 2016). In a network model, individual symptoms are defined as nodes, with the position of nodes reflecting the importance of each symptom (Mullarkey et al., 2019; van Borkulo et al., 2014). Interactions between symptoms are defined as edges, and the thickness of edges represents the intensity of symptom relationships (Epskamp et al., 2018). Additionally, node centrality indices reveal the connectivity of a variable with all other variables in the network. For instance, the expected influence (EI) is used to identify the most important symptoms of a network model, and the bridge EI is employed to elucidate the symptom influence and interconnection of symptoms cluster (Epskamp et al., 2018; Opsahl et al., 2010). Therefore, the application of network analysis can provide unique insights into potential causes and treatment of psychiatric disorders or syndromes that cannot be gleaned from overall symptom counts or severity scores (Cai et al., 2022).

To date, no studies have been published on the network analysis of depressive and insomnia symptoms among cancer patients. To address this gap and improve health outcomes for cancer patients, we compared the network structures of depression and insomnia symptoms between cancer patients and cancer-free participants (controls hereafter) based on the Health and Retirement Study (HRS). We hypothesized that the network structures of depression and insomnia symptoms in cancer patients would be significantly different compared to controls.