Impact of rumination on sleep quality among patients with non‑alcoholic fatty liver disease: a moderated mediation model of anxiety symptoms and resilience

Non-alcoholic fatty liver disease (NAFLD) is a condition in which the liver accumulates excessive fat in the absence of a minor cause such as alcohol or drugs. NAFLD encompasses the entire course of this disease, including non-alcoholic fatty liver, non-alcoholic steatohepatitis (NASH), fibrosis, and cirrhosis [1]. Over the last 30 years, NAFLD has been prevalent, affecting more than 25 percent of the world’s population, with the highest prevalence rate of 35.7% and 35.3% in South America and North America, respectively [2]. According to a recent systematic review and meta-analysis, the overall prevalence of NAFLD is 29.2% in China [3]. Despite the regional difference, the prevalence of NAFLD is increasing overall worldwide [2, 3]. There is evidence that NAFLD significantly increases overall mortality, and the mortality rate of NAFLD patients is higher than that of the general population [4, 5]. NAFLD places an increasing burden on patients and their families and poses a major public health challenge [6, 7].

NAFLD is a metabolic disorder with an unhealthy lifestyle [8], such as sleep insufficiency and poor sleep quality [9]. Epidemiological research has shown that sleep insufficiency and poor sleep increase the risk of NAFLD [9‐11]. Three biological mechanisms have been demonstrated, including poor sleep (1) may reduce the sensitivity of insulin and raise the incidence of type-2 diabetes [12, 13]; (2) can trigger inflammatory reactions (e.g., an increase in interleukin-6 and tumor necrosis factor-α) [14, 15]; (3) can activate the hypothalamic–pituitary–adrenal axis to secrete stress hormones (e.g., catecholamine, cortisol), raising the risk of metabolic syndrome [16, 17]. Moreover, poor sleep can aggravate NAFLD’s progression [18]. Therefore, identifying risk and protective factors of sleep quality and clarifying their mechanism of action may help improve health outcomes among patients with NAFLD.

Rumination refers to repeated passive thinking, including obsessive thinking and introspective meditation, in which individuals think frequently about adverse life events, including suffering from an illness [19]. Harvey’s cognitive model of insomnia indicates that the arousal and distress caused by excessive negative cognition are crucial in causing and maintaining sleep problems [20]. The cognitive approach in this model provides a framework for understanding and treating sleep problems. The cognitive model suggests that rumination and worries about stressful life events disrupt one’s sleep. Numerous studies have demonstrated a positive relationship between rumination and poor sleep [21, 22]. NAFLD could be considered as such an adverse life event that can induce patients’ repeated passive thinking. Hence, reducing rumination could help cope with the sleep disturbances associated with NAFLD. In addition, as two sub-factors of rumination, brooding and reflection have been proposed by Treynor et al. in a psychometric analysis of rumination [23]. Brooding is a tendency to cling to negative situations or actions, while reflection refers to considering the reasons for these negative situations or actions. It is important to distinguish these two components of rumination and measure them separately because they have different relationships with emotional outcomes [23].

Although the initial concept of rumination focused on depression and its causes and effects, more and more studies have expanded the concept of rumination, including the attention to adverse life events experienced by individuals. Increasing evidences have demonstrated that rumination may also be associated with other negative emotional outcomes, such as anxiety [24, 25] and post-traumatic stress disorder [26, 27], following various adverse life events. Therefore, the impact of rumination on sleep quality might be accounted for by other emotional variables. Anxiety symptoms are common among patients with NAFLD, and multiple studies have suggested a potential association between NAFLD and anxiety disorder [28, 29]. Anxiety symptoms refer to a group of negative mood states that is characterized by excessive worry that interferes with a person’s ability to function normally. The psychological process involved in coping with adverse life events (e.g., NAFLD) starts from the cognitive process (e.g., rumination) and then enters into the emotion and affection process (e.g., anxiety symptoms). Thus, we postulated that anxiety symptoms could be a psychological outcome of rumination in this study.

According to previous studies [30, 31], anxiety could be a predictor of sleep insufficiency and poor sleep quality among NAFLD patients. Anxiety and sleep are intricately connected through a self-reinforcing feedback loop [32]. Anxiety symptoms can prevent individuals from falling asleep or staying asleep, and make it difficult to get restful sleep regularly. Therefore, anxiety symptoms could account for the impact of rumination on poor sleep quality among NAFLD patients. At present, whether rumination and anxiety symptoms act on the sleep quality of patients with NAFLD is uncertain, and the relevant mechanism also needs to be clarified.

Resilience is an ability that allows individuals to adjust successfully to challenging situations [33]. Resilience can alter a person’s risk assessment, cognition in reaction to sensed emotions, and coping strategy selection. It has been discovered to have a protective effect on preventing physical and mental health issues [34]. It may help patients with NAFLD reduce the potentially harmful effects of rumination. Research has also found that individuals may mitigate miserable experiences (e.g., anxiety, post-traumatic stress disorder) via resilience [35, 36]. Kumpfer’s theory of resilience framework states that positive resilience may help individuals resist negative emotions in their lives, thus achieving a higher level of resilience, which is a process of positive feedback [34]. Therefore, it can be inferred that resilience could counteract the impact of rumination on sleep quality in patients with NAFLD [21, 35]. In addition, rumination refers to a cognitive process focused on adverse life events, whereas both anxiety symptoms and sleep quality could be considered the negative emotional and behavioral outcomes of the passive cognitive process. In the psychological process of coping with NAFLD, resilience, as a positive psychological capability, would play a moderating role on the relationship between cognitive style and emotional and behavioral outcomes. However, to date, no research has discussed the resilience’s moderating role on the relationships of rumination with anxiety symptoms and sleep quality among NAFLD patients.

Given the theory and evidence concerned above, this present study aimed to verify these hypotheses among patients with NAFLD: (1) rumination (including brooding and reflection) has a positive relationship with poor sleep quality and anxiety symptoms; (2) anxiety symptoms would account for the link between rumination and sleep quality; (3) resilience plays a moderating role on the associations of rumination with anxiety symptoms and sleep quality. Specifically, we put anxiety symptoms as the mediator and resilience as the moderator in an integrated model explaining the mechanism of rumination on sleep quality (Fig. 1), which references Hayes’s study [37]. This research will help us better understand the relationship between rumination and sleep quality, and the roles of anxiety symptoms and resilience in the relationship among patients with NAFLD. In clinical practice, the results of this study will suggest potential intervention targets for coping with poor sleep quality in this population. In future research and practice, targeted intervention will be implemented to improve sleep quality and manage NAFLD well.

This is the first study to illustrate the roles of anxiety symptoms and resilience in the underlying mechanism of rumination (including brooding and reflection) with sleep quality in patients with NAFLD. The positive association between rumination and poor sleep quality could be accounted for by anxiety symptoms. Furthermore, the impact of rumination on anxiety symptoms could be moderated by resilience, and the positive association between rumination and anxiety symptoms was significant in NAFLD patients with low and medium resilience, but not those with high resilience. Although there was no significant moderating role of resilience on the direct effect of rumination on sleep quality, resilience could attenuate the role of anxiety symptoms in accounting for the positive association between rumination and sleep quality among patients with NAFLD.

Rumination was shown to be positively linked to poor sleep quality and anxiety symptoms in our study, which was consistent with the first hypothesis. This is in line with earlier studies, which have shown that the more ruminant a person is, the more anxious he/she is, and the opposite is true [24, 25]. As a passive cognitive process, rumination may be a mechanism that links stressful life events with adverse emotional outcomes, such as depressive and anxiety symptoms. Rumination reflects an individual’s passive and repetitive thinking about the present and past symptoms, losses, and failures. Individuals with ruminant response styles tend to regulate cognitive activity adversely, triggering autonomous arousal and emotional suffering, according to Harvey’s cognitive model of insomnia [20]. Another important finding was that there were basically consistent associations of brooding, reflection and rumination with anxiety symptoms and poor sleep quality among patients with NAFLD. The finding is contrary to previous findings suggested that brooding is cognitively maladaptive and more strongly associated with depressive and anxiety symptoms and sleep problems than reflection [23, 46‐48]. Reflection is adaptive in reducing negative affect because it can prompt effective problem-solving [23]. Therefore, the finding should be interpreted with caution and needs to be further verified as different samples.

In this study, NAFLD patients with anxiety symptoms could suffer from poor sleep quality, partially because of their overthinking or intrusive worries. Anxious state can make individuals preferentially concentrate on internal and external sleep-related hazard indicators, such as NAFLD’s symptoms and unsupportive environment. Unfortunately, excessive and escalating anxiety may eventually be related to poor sleep. Our findings also revealed that anxiety symptoms could account for the relationships of brooding, reflection and rumination with poor sleep quality among the NAFLD patients. Rumination was indirectly related to poor sleep quality through anxiety symptoms. Higher brooding, reflection and rumination were associated with higher level of anxiety symptoms, which was correlated with poor sleep quality. This finding deserves attention in terms of anxiety symptoms treatment for improving sleep quality in NAFLD patients [49].

Our research also discovered the moderating role of resilience on the indirect effects of brooding, reflection and rumination on sleep quality through anxiety symptoms. To be specific, as resilience increased, the role of anxiety symptoms in accounting for the relationship between rumination and poor sleep quality reduced. Previous research on resilience’s moderating effects suggested that it has a protective role [21, 35]. High resilience weakened the impacts of perceived stress, anxiety, stressful life events, and rumination on sleep quality. In our study, resilience was negatively associated with anxiety symptoms and could attenuate the positive association between rumination and anxiety symptoms. In brief, resilience might improve sleep quality in indirect and moderate ways among patients with NAFLD. However, resilience did not show a moderating role on the direct impacts of brooding, reflection and rumination on sleep quality. Li et al. found that compared with college students with higher resilience, those with lower resilience had a stronger positive relationship between rumination and poor sleep quality [21]. As a result, further study is required to corroborate present results across a wide range of groups, including NAFLD patients with different cultural backgrounds and disease severity, as well as patients with other liver diseases, even the general population.

These new findings of this study could help clarify the impact of rumination on sleep quality among patients with NAFLD and provide insights for managing sleep-related risk and protective factors. Rumination mainly involves repetitive negative thoughts. Cognitive behavioral therapy can change negative thinking and behavioral patterns. Rumination-focused and mindfulness-based cognitive behavioral therapy has been demonstrated to be effective in reducing rumination [50]. Thus, NAFLD patients with poor sleep quality could benefit from cognitive behavioral training and mindfulness. Tousignant et al. have confirmed that sleep quality could be improved through interventions that reduce negative thinking about unfortunate events or enhance positive encouragement (e.g., mindfulness practice, positive re-evaluation training) [51]. Moreover, healthcare providers should pay attention to patients’ anxiety levels and create a supportive environment to help them reduce anxiety and improve sleep quality. Finally, given that resilience plays a moderating role on the relationship between rumination and anxiety symptoms, it may be a feasible measure to improve the level of resilience to solve sleep problems among patients with NAFLD. Interventions for resilience enhancement and stress management could be used to improve NAFLD patients’ resilience, such as positive psychology group intervention and mindfulness-based stress reduction [52].

Several limitations should be illustrated. First, this study was only carried out in one digestive disease treatment center in Northeast China. Thus, the findings should be cautiously generalized to other NAFLD populations across cultural backgrounds. Moreover, the exclusion of patients with a history of serious psychiatric illness from this study could cause difficulties in exactly interpreting the relationship between variables because the study mainly examined psychiatric variables (e.g., rumination, anxiety). Second, due to the cross-sectional design, the causal associations between rumination, anxiety symptoms, resilience, and sleep quality cannot be determined. Anxiety and sleep are linked through a self-reinforcing feedback loop. It was postulated that anxiety symptoms could cause poor sleep quality in this study. Conversely, sleep problems can exacerbate anxiety. Additionally, conceptual distinct between rumination and anxiety symptoms could be enough to suggest that rumination causes anxiety symptoms in this study [53], it seems equally plausible to arrange the study variables in multiple other configurations. Third, the self-administered method was adopted to measure our study variables, and there may be a recall or reporting bias that could influence the association among the variables. This potential influence was minimized by the use of high reliability and validity scales, anonymity protection, voluntary participation, and survey data quality control. Fourth, the study measured subjective sleep quality using PSQI, and did not include an objective measure of sleep (e.g., polysomnography) or a prospective monitoring subjective measure (e.g., sleep diary). In fact, there may be differences between subjective and objective sleep quality measurements, which affects the validity of the research results. Polysomnography is usually only performed in clinics or hospitals by specialized professional, and interferes with the patients, thus affecting sleep and possibly causing measurement deviation. Sleep diary relies on participants’ sustained adherence, and there is no standardized diary for research or clinical purposes [54].

Rumination (including brooding and reflection) could be positively related to poor sleep quality, and anxiety symptoms had a significant role in accounting for the relationship among patients with NAFLD. Resilience showed a moderating role that could attenuate the positive associations of brooding, reflection and rumination with anxiety symptoms, which were gradually reduced as resilience increased. Thus, resilience could decline the roles of anxiety symptoms in accounting for the impacts of brooding, reflection and rumination on poor sleep quality. The findings suggested that interventions aimed at alleviating rumination, reducing anxiety symptoms, and enhancing resilience might help NAFLD patients sleep better.