Impact of mental health, fear, and social support on quality of life among patients with severe mental illness during the COVID-19 pandemic: a questionnaire survey study

For nursing students

Study summary

This study looked at how mental health, fear related to COVID-19, and social support affected quality of life for people with severe mental illness who were living in a forensic psychiatric hospital in South Korea. Forensic psychiatric hospitals treat patients who are receiving mandatory psychiatric treatment under court order following criminal proceedings, which makes their situation different from patients in a typical psychiatric ward. The researchers wanted to understand how psychological distress and support networks were connected to each other and to patients' overall quality of life during a public health emergency.

The study surveyed 136 inpatients, about 89% of whom were male, ranging in age from 20 to 65. All participants had a diagnosis of schizophrenia, schizoaffective disorder, bipolar disorder, or major depressive disorder. The survey was conducted in late 2021, several months after COVID-19 vaccines became available, using a cross-sectional, self-report questionnaire design, meaning data was collected from every participant at one single point in time rather than being tracked over a longer period. Almost all eligible patients (97.9%) took part, an unusually high response rate for this kind of research.

Participants completed several well-established, validated assessment tools: the Beck Anxiety Inventory, the Beck Depression Inventory, the Perceived Stress Scale, the Fear of COVID-19 Scale, the Multidimensional Scale of Perceived Social Support, and the EQ-5D-5L quality-of-life measure. Roughly 1 in 8 participants (13.2%) reported moderate-to-severe anxiety, and about 1 in 5 (22.1%) reported moderate-to-severe depressive symptoms. About 15.4% scored above the clinical threshold for pandemic-related fear. Social support scores were moderate overall, with support from family (mean 18.79) and healthcare providers (mean 18.72) rated slightly higher than support from friends (mean 16.26).

When researchers examined how these factors related to quality of life, anxiety stood out as the strongest link: patients with higher anxiety scores reported meaningfully lower quality of life (a strong negative correlation, r = -0.55). Depression and perceived stress were also negatively linked to quality of life, though less strongly than anxiety. On the positive side, support from friends, and social support overall, were linked to somewhat better quality of life, though these connections were weaker.

Using a statistical technique called hierarchical regression, the researchers tested how much of the variation in quality of life could be explained by demographic factors (like age and sex) compared with psychological factors (anxiety, depression, stress, fear, and social support). Demographic factors alone explained almost none of the variation (1.9%, not statistically significant). Once psychological factors were added, the model explained about 31% of the differences in quality of life between patients, and anxiety was, by a wide margin, the single strongest predictor. The study also found that male patients reported significantly higher quality of life than female patients.

Based on these results, the study authors concluded that institutionalized patients with severe mental illness face a compounded psychological burden during public health emergencies, and that anxiety in particular should be a priority target for nursing and psychiatric intervention. They recommended routine anxiety screening, stress-coping training, and stronger efforts to maintain family contact and staff-organized social activities, especially in settings like forensic hospitals where patients may already face isolation due to legal restrictions.

As a nursing student, it is worth remembering that this was a single-site study conducted in one country's forensic psychiatric system, with a strongly male-skewed sample, so the findings may not transfer directly to other psychiatric populations, general hospitals, or other cultural contexts. The cross-sectional design also means the researchers can describe associations, not prove that anxiety causes lower quality of life. Still, the consistency of the anxiety finding, combined with a very high response rate, makes this a useful signal for why mental health screening matters in institutional psychiatric care, particularly during periods of societal stress like a pandemic.

For educators

Teaching and appraisal notes

This cross-sectional survey study (N = 136) examined associations among anxiety, depression, stress, fear of COVID-19, perceived social support, and quality of life (QoL) among inpatients with severe mental illness (SMI) at a national forensic psychiatric hospital in South Korea. The sample consisted of patients receiving mandatory treatment under court order for schizophrenia, schizoaffective disorder, bipolar disorder, or major depressive disorder, surveyed in late 2021 using validated instruments: the Beck Anxiety Inventory (BAI), Beck Depression Inventory (BDI), Perceived Stress Scale (PSS), Fear of COVID-19 Scale (FCV-19S), Multidimensional Scale of Perceived Social Support (MSPSS), and the EQ-5D-5L for QoL. Internal consistency was strong across measures (Cronbach's alpha ranging from .76 to .94), and the response rate was notably high (97.9%).

Descriptively, 13.2% of participants met criteria for moderate-to-severe anxiety and 22.1% for moderate-to-severe depression; 15.4% scored above the clinical cutoff on the fear scale. Bivariate correlations with the EQ-5D-5L index showed anxiety as the strongest correlate of poorer QoL (r = -0.55, p < .001), followed by depression (r = -0.36, p < .001) and stress (r = -0.27, p = .002). Social support demonstrated smaller positive associations with QoL: friend support (r = 0.21, p = .016) and overall perceived social support (r = 0.19, p = .028).

The hierarchical regression is the analytic centerpiece: demographic variables alone (Model 1) explained a negligible 1.9% of QoL variance and were nonsignificant, whereas adding the psychological variables (Model 2) increased explained variance to 30.8% (F = 9.60, p < .001), with anxiety as the dominant predictor (beta = -0.50, p < .001). This pattern -- anxiety's outsized contribution relative to depression, stress, and support variables -- is the study's most teachable finding and a useful anchor for discussing how shared variance among correlated psychological constructs can obscure which factor truly drives an outcome in cross-sectional data. Students should be prompted to consider why anxiety might dominate over depression in this population, and whether measurement characteristics of the BAI could partly explain this.

An additional finding -- men reporting significantly higher QoL than women (t = 2.97, p = .01) -- is worth flagging for critical appraisal discussion given the sample's 89% male composition; subgroup comparisons in a small, imbalanced sample warrant caution before generalizing gender effects.

For appraisal purposes, this is a single-site, cross-sectional, self-report survey with real design constraints: it cannot establish causality, it was conducted in a heavily male-dominated forensic hospital in one country, and the timing (late 2021, post-vaccine rollout) likely shaped fear and anxiety levels differently than earlier pandemic phases would have. The 30.8% explained variance also signals that substantial unmeasured factors -- resilience, coping style, legal and institutional stressors specific to forensic status, length of confinement -- are probably at play and were not captured.

Pedagogically, this article is well suited to teaching critical appraisal of survey design, correlation versus causation, hierarchical regression interpretation, and the ethical and psychosocial complexity of forensic psychiatric nursing -- a context many students rarely encounter in clinical placements but that raises important social-justice and human-rights considerations around institutionalized care during public health emergencies.