This was the first study that examined the prevalence of fatigue in older psychiatric patients and its impact on QOL during the COVID-19 outbreak. We found that around half of clinically stable older patients with psychiatric disorders reported fatigue during the COVID-19 outbreak, which is higher than the figure (31.2%) observed via standardized questions in the general older adult population of China (Meng et al., 2010), and similar to the figure (53.6%) in patients with COVID-19 assessed by the Fatigue Scale-14 (FS-14) (Qi et al., 2020). Although potentially informative, direct comparisons between studies using different measures of fatigue should be made with caution.
The high prevalence of fatigue in clinically stable older psychiatric patients could be due to several reasons. First, fatigue is a comorbid symptom of many psychiatric diagnoses. For instance, a study found that 60% of psychiatric inpatients reported significant level of fatigue (Waters et al., 2013), particularly those with depressive or anxiety disorders (Torossian et al., 2020). Although patients were clinically stable during the study period, the COVID-19 outbreak and related problems such as limited access to psychiatric services and fear of transmission could have triggered certain comorbid symptoms such as fatigue, in these patients. Second, many older adults suffered from severe and/or chronic physical illnesses. Due to the suspension of public transportation and mass quarantine measures, this group may have had more difficulty attending follow-up appointments and receiving maintenance treatments, in turn, exacerbating their physical illnesses (e.g., cancer, hypertension, chronic pain disorder, chronic obstructive pulmonary disease) and experience of fatigue (Torossian et al., 2020; Tralongo et al., 2003). Third, a lack of access to outdoor activities and physical exercises due to quarantine measures during the COVID-19 pandemic could cause fatigue (Marin et al., 2004). Finally, the high mortality rate and poor prognosis of COVID-19 in older adults could lead to perceptions of health risks and psychological distress, which may have resulted in increased loneliness and fatigue (Griffin et al., 2019).
As expected, patients with more severe depression (OR = 1.15, P < 0.001), insomnia symptoms (OR = 1.08, P < 0.001) and more severe pain (OR = 1.43, P < 0.001) were more likely to report fatigue. Fatigue is common in depressed patients, particularly in those with comorbid somatic symptoms. Previous studies have found that most patients with major depression complain of fatigue (Ghanean et al., 2018), which may be partly due to common underlying genetic factors (Corfield et al., 2016). In addition, fatigue could be induced by antidepressant medications (Marin et al., 2004). Some studies have found insomnia is a contributing factor to fatigue (Lavidor et al., 2003) and improvements in sleep problems as a correlate of decreases in fatigue complaints among older adults (Cooke et al., 1998). In this study, patients reporting more severe pain were at higher risk for fatigue, a finding that is consistent with previous findings in patients with primary Sjögren's syndrome (Conrad et al., 2018), cancer (Romero et al., 2018), or multiple sclerosis (Kratz et al., 2017).
We found that patients with fatigue had a lower overall QOL than those without, consistent with previous findings (Kratz et al., 2017; Nunes et al., 2017). QOL is determined by interactions between protective factors (e.g., higher economic status) and risk factors (e.g., poor physical health) (Hatoum et al., 1998). Fatigue is associated with sleep disturbances as well as reduced energy, motivation, and cognitive functions, all of which could negatively affect individuals’ daily functioning, increase physical and mental distress, and eventually lower their QOL (Cha, 2013; Yoo et al., 2018; Galland-Decker et al., 2019).
The strengths of this study included its focus on an understudied, at-risk population, multi-center design, use of validated measures, and large sample size. However, several methodological limitations should also be acknowledged. First, because we focused on clinically stable older patients, findings cannot be generalized to patients in other illness stages. Second, causal relations between fatigue and other variables could not be established due to the cross-sectional study design. Third, for logistical reasons (i.e., concerns about high response burdens in a vulnerable group), some factors associated with fatigue (e.g., treatment of physical illnesses and economic status), were not examined in this study.