1 INTRODUCTION

COVID-19 is a disease related to a novel coronavirus (SARS-CoV-2) that has led to a global pandemic. Given the features of rapid spread and the potential for severe respiratory symptoms, COVID-19 may significantly impact physical health (Mohapatra et al., 2020; Wiwanitkit, 2020). This global public health issue also impacts health systems, economies, and social interactions. Many jurisdictions have implemented preventive measures such as lockdowns, spatial distancing, hand washing, and wearing of face masks (Kamenidou et al., 2020). Major efforts have emphasized limiting transmission and decreasing deaths through such public health interventions (Lahiri et al., 2020). At the time of this writing (August 12, 2020), more than 20 million people had been infected by SARS-CoV-2 globally (Worldometer, 2020). In Iran, a country that shortly after China began experiencing COVID-19, between 2000 and 2500 people have been infected daily with a mortality rate of 1% to 3% (Sharifi et al., 2020).

Given the novelty of COVID-19, precise information regarding the SARS-CoV-2 virus (e.g., virus transmission) is scarce, and the possibility of reinfection is not well known and under investigation (Lim, 2020). Uncertainties regarding SARS-CoV-2 and COVID-19 may generate fear and insecurity, especially in vulnerable populations (Fofana et al., 2020; Reznik et al., 2020). Other factors (e.g., growing numbers of affected people, quarantines, and lockdowns) may impact psychological health, and such effects may be greater in vulnerable groups including people with preexisting conditions, older adults, children, pregnant women, and healthcare workers (Amerio et al., 2020). A recent review with a focus of COVID-19-related post-traumatic stress disorder (PTSD) in healthcare workers reported that self-blame and stigma, apart from previous mental health status, emerged as possible risk factors for PTSD (Carmassi, Foghi, et al., 2020).

During the pandemic, people (especially those with COVID-19) may experience stress, anxiety, depression, irritability, frustration, and hopelessness (Guo et al., 2020; Serafini et al., 2020). Such mental health concerns may lead to both shorter- and longer-term problems, particularly when experienced in combination with other factors such as poverty and insufficient healthcare services (Rangel et al., 2020). Data from disasters suggest emotional distress as a prevalent phenomenon in directly affected and vulnerable populations (North & Pfefferbaum, 2013).

Self-stigma is an important factor related to mental distress during the COVID-19 pandemic. Affected individuals may perceive themselves as being different from normal, unaffected people in terms of physical or mental abilities (Lucksted & Drapalski, 2015). People may thus classify themselves as being undesirable because of having had COVID-19 (Bruns et al., 2020). With elevated likelihoods of self-stigma, having had COVID-19 may lead to social isolation, temporary unemployment, and unknown shorter- and longer-term effects on health (Bruns et al., 2020; Turner-Musa et al., 2020). Behaviors of family members, relatives, and close friends to protect individuals with COVID-19 may exacerbate disease-related self-stigma (Turner-Musa et al., 2020). People who travel from infected regions to relatively clear zones may also encounter xenophobia that may intensify self-stigma (Cheng, 2020). Degrees of stigmatization may vary based on context and culture (Bruns et al., 2020).

Mental health concerns such as PTSD and sleep disorders are reported complications of COVID-19 (Liu, Stevens, et al., 2020). For example, one-fifth of 1,300 Italian participants had significant symptoms of PTSD (Castelli et al., 2020). Individuals with lower health-related quality of life (HRQoL) were more likely to experience PTSD (Castelli et al., 2020; Liu, Stevens, et al., 2020). PTSD symptoms may relate to concerns about COVID-19, such as infection with a potentially deadly disease and possible transmission of the disease to others (e.g., parents and children; Fekih-Romdhane et al., 2020).

Traumatic events may negatively impact sleep quality. A recent review on sleep disorders among people locked down due to COVID-19 in Europe indicated that individuals went to bed later and spent more time on smart phones and digital devices with poorer quality of sleep (Altena et al., 2020). Associations between stress, anxiety, and depression with sleep problems have been reported (Stanton et al., 2020). Such changes in mental health and related domains may link to reduced HRQoL, especially psychological HRQoL (Liu, Stevens, et al., 2020).

Relationships between stigmatization and PTSD with lower HRQoL and sleep problems have been identified in patients with different conditions (McCarthy et al., 2019; Wang et al., 2020). However, it is unclear how improving mental health may be effective in reducing potential negative effects of self-stigma and PTSD on HRQoL and sleep. A mediation model can provide insight into how independent variables (self-stigma and PTSD) may operate through a mediating factor (mental health) to influence dependent variables (HRQoL and insomnia).

Mediating roles of mental health in relationships between psychological factors and sleep problems/HRQoL have been reported (Galletta et al., 2019; Holtge et al., 2019; Pavlova et al., 2017). However, to the best of our knowledge, such investigations have never been studied among patients with COVID-19. Therefore, the aim of the current study was to investigate whether poor mental health may mediate concerns related to infection with COVID-19 (i.e., self-stigma and PTSD) and outcomes such as poor sleep and HRQoL in people having recently recovered from COVID-19.

2 MATERIALS AND METHODS

3 RESULTS

Among 844 participants with a mean age of 45.91 (SD = 13.40) years, approximately two-thirds (n = 579; 68.6%) were male. Over three-fourths (n = 660; 78.2%) were currently married and most had comorbidities (n = 704; 83.4%). The average time from discharge was 37.2 (SD = 29.4) days (Table 1).

All study variables were correlated with one another with moderate or greater magnitudes (Table 2). Correlations were in anticipated directions: MHI-5, PCS, and MCS were positively correlated with each other (r = .388 to 0.769; p < .01); ISI, PCL-5, and COVID-19-related SSS-S were positively correlated with each other (r = .334 to 0.454; p < .01); and MHI-5, PCS, and MCS were negatively correlated with ISI, PCL-5, and COVID-19-related SSS-S (r = −.688 to −0.376; p < .01).

The mediation model showed a good fit as reflected by satisfactory fit indices (CFI = 0.968, TLI = 0.950, SRMR = 0.071, RMSEA = 0.068), except for the significant χ² test (p < .001). PTSD was directly associated with physical QoL (unstandardized coefficient [B] = −24.61; SE = 2.26; p = .002), mental QoL (B = −17.87; SE = 1.92; p = .003), and insomnia (B = 0.51; SE = 0.09; p = .003) and indirectly associated with physical QoL (B = −11.20; bootstrapping SE = 6.53; 95% CI = −26.19, −1.16), mental QoL (B = −1.44; bootstrapping SE = 0.69; 95% CI = −2.81, −0.12), and insomnia (B = 0.24; bootstrapping SE = 0.05; 95% CI = 0.14, 0.35) via mental health. COVID-19-related self-stigma was directly associated with physical QoL (B = −3.17; SE = 1.00; p = .002), mental QoL (B = −2.33; SE = 0.68; p = .001), and insomnia (B = 0.12; SE = 0.05; p = .011) and indirectly associated with physical QoL (B = −2.50; bootstrapping SE = 1.31; 95% CI = −6.41, −0.75), mental QoL (B = −0.55; bootstrapping SE = 0.30; 95% CI = −1.27, −0.06), and insomnia (B = 0.09; bootstrapping SE = 0.02; 95% CI = 0.04, 0.14) via mental health (Table 3).

4 DISCUSSION

This study examined the hypothesis that mental health would mediate relationships between COVID-19-related self-stigma and PTSD and HRQoL and insomnia in patients who recovered from COVID-19. Findings supported mediation effects. The direct and significant relationships between the independent variables and outcome measures were confirmed, and SEM results showed that a model in which mental health is considered as mediator between study variables had good fit indices. Moreover, both components of HRQoL (i.e., physical and mental components) as well as insomnia were associated with the mediator. Therefore, addressing mental health as an entity through which associations between COVID-19-related self-stigma/PTSD link to HRQoL/insomnia may help design effective intervention strategies to prevent negative psychological effects of COVID-19 on HRQoL and sleep.

Prior studies have examined the extent to which mental health may have a mediating role. For example, Galletta et al.( 2019) examined the mediation effect of mental health between sense of coherence and physical HRQoL in 209 people with chronic diseases. They found that despite a nonsignificant direct association between sense of coherence and physical HRQoL, the indirect effect of the independent variable on HRQoL through mental health was significant. They concluded that a sense of coherence as a marker of HRQoL may be considered as a psychological process influencing mental health, which in turn may affect physical HRQoL as well. Choi et al.( 2016) identified a mediating role of mental health linking symptom severity and HRQoL in 519 patients with urinary tract infections. In this study, all three subscales of the Depression, Anxiety and Stress Scale-21 displayed similar mediation effects in relationships between symptom severity and HRQoL. Thus, they suggested that when improving HRQoL among such patients, the role of mental health in designating effective programs should be considered along with physical relief of symptoms. Yuan et al.( 2020) used a similar design to ours in examining and identifying a partially mediating role of mental health between sleep quality and HRQoL in older adults in China. While these studies used different scales to assess mental health, the findings resonate with the current ones in suggesting a mediating role of mental health.

A point that differentiates our study from those mentioned above is that we used a brief and complete scale (i.e., MHI-5) to assess mental health. We believe that although the MCS of the HRQoL may be considered as a measure to assess mental health, it is a part of the measure specifically designed to measure HRQoL. We thus believe that our use of an independent scale to assess mental health may be better to assess independent relationships.

As hypothesized, mental health, and HRQoL measures were negatively associated with insomnia and PTSD. Therefore, patients with COVID-19 who had sleep problems and significant PTSD symptomatology experienced poor HRQoL and mental health. Poor sleep quality has been observed in people with COVID-19. Among hospitalized patients in intensive care wards of the Wuhan Union Hospital, COVID-19 patients with poor sleep quality compared to those with better sleep quality had lower recovery rates, poorer immune system functioning and longer hospital stays, negatively impacting their HRQoL (Zhang et al., 2020). Among patients who had recovered from COVID-19 and were followed for 8 weeks after discharge, more than 85% had sleep problems including changes in subjective sleep quality, sleep latency, sleep duration, and daytime function. More than half of participants were also identified with significant generalized anxiety (Yang et al., 2020). As good sleep may promote better immune system function in COVID-19 patients, early interventions to improve sleep may be helpful.

Among 675 people who were discharged following COVID-19 treatment, PTSD and self-stigma were significant concerns (Liu, Baumeister, et al., 2020). Patients also reported considerable perceived discrimination that generated stress and anxiety for them. Similarly, we found that the association between COVID-19-related self-stigma and PTSD may lead to poorer mental health in general and poorer HRQoL generally, although these measures should be evaluated longitudinally to confirm such relationships. Therefore, finding solutions to target these relationships by decreasing self-stigma and improving the mental health of patients by reducing stress and anxiety may help improve HRQoL and limit sleep problems among this population. This may be especially important for people with psychiatric problems. Specifically, these patients have been reported to experience more severe sequelae after a COVID-19 infection than people in the general population. Therefore, specific attention should be paid to this vulnerable population. Indeed, recent research has found that patients with psychiatric problems had high levels of mental health burden (including PTSD and insomnia) when they were exposed to lockdowns and spatial distancing (Carmassi, Bertelloni, et al., 2020).

The current study has limitations. First, we collected data from only four hospitals that admitted patients with COVID-19. Therefore, our findings may not be generalizable to those who received care in their homes or those with milder symptoms. Second, the study examined a convenient sample using telephone interviews. Third, the survey was cross-sectional and causality cannot be inferred. Fourth, the survey was subject to recall and other biases. Fifth, discharged patients were contacted about 1 month after discharge and COVID-19-related effects on outcomes (HRQoL, insomnia) may change over time. Hence, following such patients for longer periods may help to better understand associations between study variables over time.

In conclusion, our results supported the hypothesis that mental health is a mediator between COVID-19-related self-stigma/PTSD and poor HRQoL/insomnia. Therefore, considering mental health concerns in intervention or prevention measures when helping patients with COVID-19 may provide more effective approaches to decrease complications and potential negative consequences of the disease, especially on their HRQoL and sleep. Future studies conducted within other jurisdictions may help examine the generalizability of the findings and understand mechanisms underlying associations between physical and psychological health.

CONFLICT OF INTEREST

Dr. Potenza has consulted for and advised Opiant Pharmaceuticals, Idorsia Pharmaceuticals, AXA, Game Day Data, and the Addiction Policy Forum; has received research support from the Mohegan Sun Casino, the Connecticut Council on Problem Gambling, and the National Center for Responsible Gaming; has participated in surveys, mailings or telephone consultations related to drug addiction, impulse control disorders, or other health topics; and has consulted for law offices and gambling entities on issues related to impulse control or addictive disorders. The other authors report no financial relationships with commercial interests.

AUTHOR CONTRIBUTIONS

HM, MS, HS, HRJ, and AHP designed the study and contributed to prepare the initial draft of the manuscript. CYL and MNP critically reviewed the manuscript. Others contributed to collect data and entering data in the analysis software. All authors studied the manuscript and provided their comments on it.