Prevalence of Post-traumatic Stress Disorder Symptoms Among Psychiatric Patients during the COVID-19 Pandemic

Background: The outbreak of the COVID-19 pandemic has caused extensive public health concerns, posing signi�cant challenges to healthcare services. One particular area of concern is the mental health of psychiatric patients, who are often a neglected group. The aim of this study was to investigate the prevalence of, and associated factors for symptoms of post-traumatic stress disorder (PTSD) among psychiatric patients in China during the COVID-19 pandemic. Methods: Self-reported questionnaires were distributed to patients in four psychiatric hospitals in Beijing, China, between April 28 th and May 30 th , 2020. Information regarding sociodemographic characteristics, COVID-19 related factors, support, psychosomatic factors, and PTSD symptoms was collected data using a series of scales, such as the Impact of Event Scale-Revised, the 7-item Generalized Anxiety Disorder Scale, the 9-item Patient Health Questionnaire depression scale, and so on. Multivariate regression was used to identify factors related to PTSD symptoms. Results: A total of 1,055 psychiatric patients were included in the �nal sample. The prevalence of PTSD symptoms was 41.3%. Hierarchical linear regression demonstrated that fear of the pandemic and anxiety were shared associated factors for both symptoms of PTSD and its subscales. Additionally, age was an associated factor for the total PTSD (β = 0.12, p < 0.01), intrusion (β = 0.18, p < 0.001), and avoidance (β = 0.1, p < 0.05) symptoms; depression was an associated factor for the total PTSD s (β = 0.13, p < 0.001), intrusion (β = 0.11, p < 0.01), and hyperarousal (β = 0.19, p < 0.001) symptoms. Conclusions: The prevalence of PTSD symptoms was high among psychiatric patients during the COVID-19 pandemic in China. This study found that age, fear of the pandemic, anxiety and depression are signi�cant associated factors of PTSD symptoms in psychiatric patients during the pandemic. We call for higher awareness and introduction of PTSD interventions to relieve the psychological stress in these patients.


Background
The COVID-19 pandemic has had a substantial impact on many aspects of the physical and mental health of the population worldwide [1].Psychiatric symptoms have been increasing in both the general population and in patients with the infection during the pandemic [2][3][4].Psychiatric patients, who are often a neglected group, have also encountered mental health problems during the pandemic, even if not infected with COVID-19 [5].Psychiatric patients with severe mental illness, including affective and schizophrenia spectrum disorders, are at a higher risk of negative mental health outcomes related to the pandemic [6].There are indications of worsening psychiatric symptoms among patients with pre-existing psychiatric disorders [7,8].Some experts have speculated that the COVID-19 pandemic might be negatively effecting individuals with pre-existing mental disorders [9][10][11].With 16.6% lifetime prevalence of mental disorders among adults in China [12], millions of psychiatric patients need to be concerned, as they may face barriers when seeking help and timely management of their mental health condition during the pandemic [13].However, little appears to be known about the pandemic's impact on patients with preexisting psychiatric disorders [14].
Post-traumatic stress disorder (PTSD) is caused by exposure to actual or threatened death, serious injury or sexual violence [15].There are three main types of symptoms: intrusion symptoms associated with the traumatic events (such as intrusive memories, recurrent distressing dreams, intense or prolonged psychological distress, dissociative reactions, and marked physiological reactions), persistent avoidance symptoms (including avoidance of distressing memories, thoughts or feelings, and numbing of responsiveness), and hyperarousal symptoms (including irritable behavior, angery outbursts, problems with concentration, hypervigilance, and exaggeratedstartle response) [16].Individuals with PTSD are generally at higher risk of suicide [17].Long-term exposure to stress may worsen pre-existing chronic health conditions, accelerate the disease's progression, or increase nancial burden on patients [18].Some experts consider PTSD as a secondary effect of the pandemic [17], during which many people are reporting numbness, stiffness, high vigilance, and other psychiatric symptoms [18].Studies on COVID-19 revealed that PTSD can occur during and after the infectious diseases [19].The prevalence of PTSD symptoms ranged from 7-53.8% in the general population during the COVID-19 pandemic in China, Spain, Italy, Iran, the US, Turkey, Nepal, and Denmark [20].A meta-analysis including 68 independent samples and sub-samples indicate that the PTSD prevalence was 21.94% during the COVID-19 pandemic, and pandemic-affected groups have signi cantly higher PTSD prevalence compared to the general population under normal circumstances [21].A systematic review of the relationship between the COVID-19 pandemic and mental health consequences found that mental health issues in COVID-19 patients present a high level of post-traumatic stress symptoms (96.2%) [14].Previous psychiatric disorders displayed suggestive evidence of increasing the risk of PTSD [22].The onset of PTSD symptoms can make the psychiatric disorder itself more complex and di cult to treat, leading to a greater disease burden [23].Therefore, clinical doctors need to increase the awareness and importance of PTSD symptoms in psychiatric patients.However, until now there is no research on the prevalence of PTSD symptoms among psychiatric patients during the COVID-19 pandemic.
The causes of PTSD are not fully understood, and whether people who have experienced the same traumatic event develop PTSD is related to sociodemographic characteristics and pre -, peri-, and post-traumatic factors, which interact in complex ways [22].A systematic review of 54 studies on PTSD found that six pre-traumatic predictors of PTSD included: cognitive level, coping styles; personality characteristics, psychopathology, psychophysiological factors, and socio-ecological factors [24].Variables related to coping strategies and social/family support showed evidence as PTSD associated factors [22,25].All potential consequences of trauma (i.e., symptoms of anxiety, avoidance, or depression) had evidence as post-trauma risk factors [22].In previous literature, a number of risk and protective factors for PTSD have been identi ed, however, these ndings have not always been consistent [26], inconsistency may re ect unrecognized or unaccounted sources of genuine heterogeneity or biases.
The aim of this study was to examine the prevalence of PTSD symptoms among psychiatric patients during the COVID-19 pandemic and to identify associated factors for PTSD symptoms and its subscales.We assume that due to their susceptibility and vulnerability to crisis, the prevalence of PTSD symptoms among psychiatric patients might be higher than that of the general population during the pandemic [10,11].Based on previous study, in addition to sociodemographic characteristics and COVID-19-related factors, we used psychosomatic factors from the perspective of psychological factors (i.e., loneliness, anxiety, and depression), somatic factors (i.e., quality of life, sleep quality), and social ecological factors (i.e., social support) as possible associated factors for PTSD symptoms [24,27].Above all, we hypothesized as follows: (1) The prevalence of PTSD symptoms among psychiatric patients will be higher than that of the general population during the pandemic.(2) Demographic characteristics of the psychiatric patients, such as age and gender, will be signi cantly associated with the PTSD symptoms.(3) COVID-19 related factors, such as fear of the pandemic, increased pressure by pandemic, will be associated factors with PTSD symptoms.(4) Psychosomatic factors (i.e., loneliness, quality of life, sleep quality, anxiety, and depression) will be signi cant associated factors with PTSD symptoms among psychiatric patients.

Participants and Procedures
This cross-sectional survey was conducted from April 28 and May 30, 2020.Cluster sampling was used to construct the sample.A questionnaire was distributed by three psychiatrists to all patients in four psychiatric hospitals located in different districts of Beijing, China.There were 1,104 participants in total.The three psychiatrists came from the four different hospitals, and one of them worked in two hospitals.Of these, 550 (49.8%) participants were from a tertiary specialised hospital, 279 (25.3%) were from a second-level psychiatric hospital, and 275 (24.9%) were from two community psychiatric hospitals.The inclusion criteria were as follows: able to write; aged 18-60 years; diagnosed with anxiety disorder, major depressive disorder, bipolar disorder, or schizophrenia.All study participants were informed about the purpose of the study and its procedures before taking part.They were also informed that they could refuse to answer any item and withdraw at any time during the study.Each participant signed an informed consent form before completing the questionnaire.49 participants did not complete the questionnaire and were excluded, leaving data of 1,055 patients available for analysis, with a completion rate of 95.6%.The study protocol was performed in accordance with the Declaration of Helsinki [28].The Ethics Committee of the Beijing Anding Hospital a liated to the Medical Capital University has approved this investigation.

Sociodemographic characteristics
Sociodemographic data were collected, including sex, age, education level, marital status, employment status, annual family income, place of residence, medication status during the pandemic, substance use, and living circumstances during the pandemic.Information was also collected on psychiatric diagnoses, including anxiety disorder, major depressive disorder, bipolar disorder, and schizophrenia.
Medication status during the pandemic was recorded as one of three types: on antipsychotic medications, not on antipsychotic medications, and on nonpsychotropic medications.Substance use was assessed by two questions: "During last month, have you ever experienced symptoms of intoxication such as dizziness, headache and drowsiness due to drinking too much?" on a 5-point scale (1 = never; 2 = rarely; 3 = occasionally; 4 = often; 5 = almost every day); and "How many cigarettes did you smoke per day on average?" on a 5-point scale (1 = No smoking history; 2 = Smoking 1-5 cigarettes per day; 3 = 5-10 cigarettes; 4 =10-20 cigarettes; 5 = more than 20 cigarettes per day), with higher scores indicating more severe substance use.The nal score were obtained by the summing the items.

PTSD symptoms
PTSD symptoms were measured using the 22-item, self-reported Impact of Event Scale-Revised (IES-R) [16], which assesses the severity of subjective distress caused by traumatic events.Its items are categorised into the three symptom subdomains of intrusion, avoidance, and hyperarousal, and each item is rated on a 5-point scale (0 = not at all; 4 = extremely).The IES-R has been validated, including in China [29], for research regarding health-related trauma, such as that associated with severe acute respiratory syndrome [30,31].We considered the cut-off score was 24 in the present study based on previous research [16].We calculated the composite score for PTSD (Cronbach's α = 0.93) and its symptom subscales (Cronbach's α = 0.84-0.84).

COVID-19 related factors
COVID-19 related factors includeed fear of the pandemic, increased pressure by pandemic, clinical treatment during pandemic, mental health guidance during pandemic, and medication barriers due to pandemic.
Fear of the pandemic was assessed by two questions: "In the last week, were you worried about getting infected with COVID-19?" and "Do you wash your hands excessively for fear of getting infected with the virus?", both on a 5-point scale (1 = never; 2 = rarely; 3 = occasionally; 4 = often; 5 = almost every day), with higher scores indicating stronger concern.The nal values were obtained by the average score.Increased pressure by pandemic was assessed by two questions: "Is your medication status greatly affected by the outbreak?", and "Did your mental state uctuate during the outbreak?"both on a 5-point scale with higher scores indicating much less affection and uctuation.Clinical treatment during pandemic was assessed by one question: "Did you see a psychiatrist during the outbreak?",requiring a "yes" or "no" answer.Mental health guidance during pandemic was assessed by one question: "Did you receive mental health services during the outbreak?",again with a yes or no answer possibility.Medication barriers due to pandemic was also assessed by one question: "Has your access to medicines been affected during the outbreak?"on a 6-point scale, with answers ranging from 1 = never to 6 = very often, with higher scores representing more severe medication barriers.

Psychosomatic factors
Psychosomatic factors include loneliness, quality of life, sleep quality, anxiety, and depression in the present study.
Anxiety was evaluated using the 7-item Generalized Anxiety Disorder Scale (GAD-7) [32], which assesses the frequency of anxiety symptoms in the past two weeks on a 4-point scale (0 = not at all; 3 = nearly every day).The Chinese version of the GAD-7 has been validated (α = 0.90), and we considered 5 to be of the clinical cut-off score based on previous research.Cronbach's α was 0.95 in the present study.
past two weeks on a 4-point scale (0 = none; 1 = on a few days; 2 = on more than half of the days; 3 = almost every day).The Chinese version of the PHQ-9 has been validated (Cronbach's α = 0.94) [34], and we considered 5 to be of clinical cut-off score based on previous research [35].
Loneliness was assessed by the question "Do you feel lonely?" on a 5-point scale (1 = never; 2 = rarely; 3 = occasionally; 4 = often; 5 = almost every day), with higher scores representing more severe loneliness.Quality of life was assessed by two questions: "How do you feel about your quality of life?" and "Are you satis ed with your present state of health?" on a 5-point scale (1 = extremely unsatisfactory; 2 = rarely not satis ed; 3 = not satis ed or dissatis ed; 4 = very satis ed; 5 = extremely satisfactory), with higher scores indicating higher quality of life and greater satisfaction.We obtain the nal score by summing the items, with Cronbach's α of 0.76.Sleep quality was assessed by the following three items according to the Pittsburgh Sleep Questionnaire (PSQI-PT) [36]: "How many hours of sleep have you usually been getting per night during the pandemic?", "Do you take medication to help you sleep during the pandemic?", and "In general, what do you think of your ability to sleep during the pandemic?" The responses were evaluated using a 5-point scale, with higher scores indicating worse quality of sleep.We obtained the a composite score by adding the items, with Cronbach's α of 0.52.

Support
Support was assessed by the level of agreement with the following six statements: (1) "During the pandemic, certain people were always besides me when I encounter problems"; (2) "During the pandemic, I was able to share happiness and sadness with certain people"; (3) "During the pandemic, I was able to obtain help from my family "; (4) "During the pandemic, I was able to obtain emotional help and support from my family when needed"; (5) "I was able to obtain help from my friends during the pandemic"; and (6) "My friends shared happiness and sadness with me during the pandemic".The responses were rated on a 5point scale (1 = strongly disagree; 2 = disagree; 3 = not sure; 4 = agree; 5 = strongly agree) [37].The composite scores of social support, family support and friends supports were calculated by the sum scores of item ( 1), (2), item (3), (4) and item ( 5), (6), respectively (Cronbach's α = 0.75-0.87).

Data analysis
The sample's characteristics are presented as the mean ± standard deviation for continuous variables, and as the percentage for categorical variables.Hierarchical linear regression models were used to identify factors related to the PTSD symptom subscales.In Step 1, we entered the sociodemographic variables into the model, before adding COVID-19 related factors in Step 2. In Step 3, we added support and in Step 4, we included psychosomatic factors.An increasing R 2 value further con rmed the importance of the independent variables regarding the dependent ones.One-way variance analysis and post-hoc tests were used to evaluate PTSD symptoms according to the underlying diagnosis.The statistical analyses were performed using the IBM SPSS Statistics software version 20.0 and R software version 3.6.1.All tests were two-tailed.A p-value less than 0.05 was considered statistically signi cant.

Sociodemographic and clinical characteristics
A total of 1,055 psychiatric patients completed the survey.As shown in Table 1, the average age of the participants was 37.15 (SD = 13.21)years.The characteristics that represented a majority of the patients were as follows: female (65.5%), a lower education level (62.2%), unmarried/others (52.6%), living in an urban area (88.4%), annual family income lower than 150,000 CNY (71.9%), and not infected with COVID-19 (93.9%).All study participants had a preexisting diagnosis (35.4% with anxiety disorder, 26.7% with major depressive disorder, 17.6% with bipolar disorder, and 20.3% with schizophrenia).During the COVID-19 pandemic, almost half of these patients reported that they did not seek mental health services (47.7%).More than half of the sample were taking psychiatric medicine (57.3%) but reported signi cant COVID-19-related barriers to continuing treatment.

Discussion
The prevalence of PTSD symptoms (41.3%) in this study is much higher than the previously reported lifetime prevalence level in the general population (2-9%) [38].A systematic review showed the pooled prevalence of PTSD among participants during the COVID-19 pandemic to be 21.94% [21].A study conducted between April 4 and 6, 2020, among the medics working in Wuhan upon their return after work indicated overall prevalence of clinically concerned PTSD symptoms of 31.6% [39], by the same scale (IES-R).A survey-based cross-sectional study performed from January 29 to February 7, 2020, in Wuhan, China, showed an estimated PTSD prevalence of 9.8% [40].In our study, this prevalence of PTSD was even higher.Exact comparisons are di cult to make because some of the other research have used different measures.However, it is suggested that although Beijing is not an area with the highest risk of COVID-19, the prevalence of PTSD symptoms in psychiatric patients in Beijing is high, indicating the susceptibility of this group.Clinicians must be aware that these patients may experience higher rates and severity of post-traumatic stress disorder than the general population [20].
In this study, there was no signi cant difference between the severity of the total PTSD score among different patients with mental disorders.This may indicate that patients with different diseases share similar psychological characteristics, including vulnerability and susceptibility, causing similar effects during the COVID-19 pandemic.However, the scores of the PTSD hyperarousal symptoms were higher in patients with major depressive disorder than in those with anxiety disorder or schizophrenia.Evidence suggests that the associations between PTSD and depression are complex, involving bidirectional causality, common risk factors, and common vulnerabilities [23,41,42].Hyperarousal includes irritability, anger, di cult concentrating, hypervigilance, and a heightened startle response [16].The results of this study suggested that more attention should be paid to the characteristics of high arousal in patients with major depressive disorder.
The study found evidence for the second hypothesis that demographic characteristics were associated with PTSD symptoms.This study showed that age was an associated factor for the total PTSD score, intrusion, and avoidance.Since the COVID-19 virus is more serious and with higher mortality rate in older people [43], they may have more severe PTSD.Retirement was a shared associated factor for both the total PTSD score and intrusion in the study, indicating that retirement may be a protective factor for PTSD.A possible explanation is that retirees may need to travel less during the epidemic and have higher nancial security, therefore being less stressed by the epidemic [1].
The third hypothesis, COVID-19 related factors are associated with PTSD symptoms, was well supported by the data.Fear of the pandemic were shared associated factors for both PTSD symptoms and its subscales.There have been reports of anxiety and fear often co-existing and comorbid with PTSD [44,45].Mental health guidance during the pandemic was a unique associated factor, while clinical treatment during pandemic, or medication barriers due to pandemic were not signi cant associated factors for PTSD symptoms, which might indicate that mental health interventions and resources could help psychiatry patients reduce the stress caused by the epidemic and the incidence of PTSD.
The fourth hypothesis, psychosomatic factors are signi cant associated factors with PTSD symptoms, was supported by the data.Anxiety symptoms was shared associated factors for both PTSD symptoms and its subscales.Recent neuroscience research suggested that higher sensitivity to anxiety tends to increase the severity of PTSD [46].Individuals with higher stress/fear levels may become impatient, feel upset or agitated, and experience di culty relaxing, all of which have a negative impact on PTSD symptoms [47].Depression symptoms were associated factors for the total PTSD score, intrusion and hyperarousal.As depression is the disorder most commonly associated with PTSD [23,42], people with depressive symptoms may be more likely to develop PTSD, which should be particularly noticed.Quality of life was a unique associated factor for avoidance, implying that patients were more concerned about it.
During the epidemic, people's quality of life deteriorated [48].Self-rated poor health during an outbreak was signi cantly associated with greater psychological impact and higher levels of stress, according to a study in China [31].
Another prominent nding was that several unique factors were associated with sub -dimensions of PTSD.Most obviously, urban residence, increased pressure, loneliness, support from friends and sleep quality were all unique associated factor for hyperarousal but not associated with intrusion or avoidance.These results might indicate that there were differences among the related factors of the three dimensions of PTSD, and hyperarousal require unique attention [16].During an epidemic, isolation policies and inadequate social support can lead to feelings of loneliness [1].Previous studies showed that isolation can negatively affect mental health [7,8].Our ndings correlate to those of other studies on the general population.Social support plays a key role in mitigating the risk of mental health problems [49].The results also demonstrated that support from friends was associated with a lower incidence of hyperarousal symptoms, while support from family might increase patients' hyperarousal symptoms.This nding is a reminder that too much unnecessary care from family could increase patients' hyperarousal symptoms.Thus, "moderate" care from friends is necessary for psychiatric patients.These results have great implications for clinicians in predicting and treating patients with high hyperarousal symptoms.

Implications
To the best of our knowledge, this is the rst study to screen for PTSD symptoms in patients with a pre-existing psychiatric diagnosis during the COVID-19 pandemic in Beijing, China.Primarily, the prevalence of symptoms of PTSD among psychiatric patients was not encouraging, arousing attentions from medical staff, related psychologists and mental health centers.Next, this study explored some risk factors (e.g., old age, depressive disorder, fear) and protective factors (e.g., retirement, mental health guidance) for PTSD, providing speci c reference and guidance for the psychological prevention and intervention among psychiatric patients in face of the COVID-19 pandemic.Furthermore, this study examined PTSD as well as the three subscales, discriminating the difference in the relationship between PTSD subscales and related psychosomatic factors.The uniqueness of hyperarousal factor provided a theoretical reference for better understanding the structure of PTSD symptoms.

Limitations
This study has several limitations that should be considered when interpreting its ndings.First, it adopted a cross-sectional design, so the way PTSD symptoms in psychiatric patients may change over time is unclear. Figures

Figure 1 Total
Figure 1 Total scores of hyperarousal symptom of PTSD among patients with different psychiatric diagnosis Note: The width of the gures indicates the sample size of each group.

Table 1
Demographic and clinical characteristics of the study sample (N = 1,055) Note.The unit of annual income is CNY yuan.Categorical variables were presented in the form of mean and standardized deviation, and continuous variables were presented in the form of number and proportion.Note.The unit of annual income is CNY yuan.Categorical variables were presented in the form of mean and standardized deviation, and continuous variables were presented in the form of number and proportion.