1 INTRODUCTION

In December 2019, Wuhan, the capital of Hubei Province of China, was hit by the coronavirus 2019 (COVID-19; Bao et al., 2020). It was another infectious disease outbreak that has since posed a major threat to public health worldwide after severe acute respiratory syndrome (SARS), H1N1, and avian influenza. The sudden surge of this novel coronavirus occurred at the time of the Chinese New Year during which the heavy human traffic contributed to the rapid spread of the virus. A 76-day lockdown and quarantine measures were implemented in Wuhan from January 23 to 8 April 2020 in an attempt to check the large scale global spread of this pandemic (Lancet, 2020b). The quarantine of a city with a population of more than 10 million was unprecedented in the public health history. The outbreak of the pandemic undoubtedly impacted all walks of life in a society, especially the mental health and the quality of life (QoL) of the Wuhan's populace.

Sudden public health incidents can affect one's mental health, and adults are more likely to suffer from adverse mental health symptoms such as anxiety (Guan et al., 2020; Rao et al., 2020). The residents of Wuhan who went through a long period of quarantine during the COVID-19 may have experienced various levels of anxiety and powerlessness as well as other negative emotions that may even drive them towards suicide and other self-sabotaging behaviours. Besides, the aftermath might persist three years after lifting of the quarantine (Brooks et al., 2020; Sher, 2020). During the pandemic, there was a severe shortage of masks, sanitizers, food, and other materials in Wuhan. A lack of such badly needed primary life-sustaining commodities and medical supplies (Z. Zhang, Yao, et al., 2020) may pose a threat to the population's existence and mental health (Maslow, 1954). Those who lived outside Wuhan could take a glimpse into the home-isolated Wuhan populace only through the public media, which was flooded with rumours and false information that may arouse panic as well as exacerbate anxiety and fear among the general public (Ayittey et al., 2020; Olatunji et al., 2020; Zarocostas, 2020). Past studies (Hideki et al., 2008) have indicated that social support can reduce the negative impacts of public health emergencies on people's mental health. A previous report has demonstrated that improved social support during the COVID-19 pandemic may alleviate the detrimental beliefs eroding the mental health of the Wuhan population (Yu et al., 2020). Therefore, understanding the factors associated with people's anxiety in Wuhan would help the government tailor appropriate intervention measures targeting the emotional turmoil triggered by the pandemic.

QoL, which reflects the degree of fulfilment of the physical, psychological, social, and emotional needs of an individual in response to environmental requirements (WHOQOL Group, 1998), has not been adequately addressed for those living in a region heavily affected by COVID-19. The QoL of those with chronic diseases and the elderly in such a badly affected area (i.e., Wuhan) is further hampered because of their increased risk of contracting the disease. In addition, the physical and mental impair caused by the pandemic also adversely affects the QoL of those involved (Agarwal et al., 2020; Zomalheto et al., 2020). QoL in the midst of a growing epidemic can be influenced by several factors, including knowledge of the disease, information sources and material needs (H.-C. Wu et al., 2006; Y. Zhong et al., 2021). In this aspect, an all-round social-environmental support program may have a positive impact on their QoL.

However, studies consistently reported that the social-environment support could protect individuals from developing mental health problems when they experience difficult time (Xu & He, 2012). Social support is a multiconstruct with multiple dimensions such as subjective support, objective support, and seeking-social support (Xiao, 1994). It has been found to be a major way of improving the QoL and maintaining a healthy state of mind and body (Lan et al., 2015; Yilmaz, Piyal, & Akdur, 2017). Previous studies have demonstrated that the importance of social support in both emotional (e.g., from parents, friends, caregivers) and material aspects for protecting against anxiety that have been triggered by disasters, calamities, and outbreaks of infectious diseases (Bloom et al., 2017; Veenema et al., 2017). In addition, environmental support is just as significant as social support, which including the accessibility to accurate information about COVID-19 to avoid public consternation caused by the ‘infodemic’ (Veenema et al., 2017; Wang et al., 2020). Although there may be no effective way to prevent the spread of the COVID-19 pandemic in this era of globalization when physical distance is no longer a barrier, a proper understanding of the disease may help in suppressing rumours and the resulting panic (C.-Y. Lin, 2020). For instance, imprudent hoarding of commodities and medical supplies (e.g., hand sanitizer, medicines, protective masks, and even toilet paper) by those who over-reacted to the pandemic (Dubey et al., 2020) may result in social chaos.

Conceivably, inhabitants of Wuhan who were forced to be placed on prolonged home quarantine may be anxiety that negatively impacted their QoL. Therefore, this study aimed at elucidating the factors associated with the impacts on the anxiety and QoL of the Wuhan populace during the COVID-19 assault in an attempt to improve the mental health and QoL of those being affected by the pandemic.

2 METHODS 2.1 Study design and participants

This cross-sectional questionnaire-based study, which was conducted between July 6 and 10, 2020, mainly targeted Wuhan residents during the COVID-19 pandemic. Inclusion criteria were: (1) Inhabitants of Wuhan during the COVID-19 attack (23 January to 10 July 2020); (2) Individuals over 20 years of age; (3) Those who could understand the contents of the questionnaire. Participants who failed to complete the questionnaire were excluded. This study was approved by the Research Ethics Center of China Medical University and Affiliated Hospital (CRREC-109-077). A total of 226 responses were retrieved after screening.

This study adopted the method of snowball sampling that involved the recruitment of participants through the ‘WeChat’ application and those enrolled were encouraged to recruit more subjects for the study. The participants were then required to complete a questionnaire through an online survey platform (‘SurveyStar,’ Changsha Ranxing Science and Technology, Shanghai, China). During the process, the participants were honestly informed that the study would be beneficial to society, and their responses would remain anonymous. The participants signed informed consent before answering the questionnaire and were free to withdraw from the study at any time without any repercussions. At the end of the survey, the data were collected in the form of a structured questionnaire.

To ensure the quality and completeness of the collected data, all information needed to be uploaded through a single mobile phone or computer to avoid duplicated submissions. The responses were checked logically by the system with the invalid ones discarded. All valid responses were automatically entered into a data file and checked by one independent researcher. The information regarding the demographic characteristics of participants, factors reflecting the degree of social-environmental support, anxiety level, and QoL were also collected in the questionnaire.

2.2 Questionnaire

The questionnaire was divided into four main sections. The first section aimed at collecting the necessary demographic information. The participants were required to fill in their gender, age, body mass index (BMI), monthly income, health status, and whether they were infected with COVID-19 according to the results of official nucleic acid testing conducted in Wuhan on May 14. Their names and other personal information were not collected to ensure anonymity of their responses.

The second section of the questionnaire focused on the degree of social-environmental support that the participant received. The strength of social support was assessed using the Social Support Rating Scale (SSRS), while the degree of environment support was evaluated with four self-developed items. The SSRS, which is a 10-item self-reported scale that assesses the level of an individual's social support (Xiao, 1994), consists of three subscales: subjective support (four items), objective support (three items), and seeking-social support (three items). While subjective support reflects the perceived interpersonal network that an individual can count on, objective support signifies the degree of actual support an individual received in the past. Support-seeking behaviour refers to the pattern of behaviour that an individual utilizes when seeking social support. Each item was scored using a four-point Likert Scale. Item scores of the SSRS were computed by summation, generating a total support score ranging from 12 to 66, a subjective support score ranging from 8 to 32, an objective support score ranging from 1 to 22, and a support-seeking behaviour score ranging from 3 to 12. Higher scores indicate stronger social support. SSRS has been shown to have good reliability and validity (Xiao, 1994).

Regarding the environmental support, previous studies indicated that during times of pandemic many people exhibit fear and anxiety-related distress responses that include the following: fear of supplies are running low, and fear of information uncertainty (Baloran, 2020; Hobbs, 2020; Ma et al., 2020; Ranney et al., 2020; Taylor et al., 2020; Y. Zhong et al., 2021). The environment support questionnaire was developed to measure the aforementioned features as well as to assess COVID-19-related distress, thus targeted giving of support. Four self-developed items were used for assessment: ‘Do you have enough medical supplies?’; ‘Are your basic commodities adequate to sustain daily life?’; ‘Do you have accessibility to information about COVID-19?’ and ‘Do you possess sufficient knowledge to deal with COVID-19?’. The response for each item was scored with a five-point Likert scale from 1 (not at all) to 5 (completely). Higher scores indicate stronger environmental support. The total score of the social-environment support was the sum of SSRS and environmental support. The Cronbach's α coefficient of the social-environment support was 0.77 in this study (Social support α = 0.75 and the environment support α = 0.72).

The third section of the questionnaire assessed the degree of anxiety of participants by using the Generalized Anxiety Scale (GAD-7; Spitzer et al., 2006) in which a 7-item scale was used to estimate the incidence of anxiety disorder in the past two weeks. A four-point Likert scale (0: not at all; 1: several days; 2: over half the period; 3: nearly every day) was utilized to score the response to each item. The total score ranging between 0 and 21 was acquired by summation of the scores from item 1 to 7. The criteria for the interpretation of the degree of anxiety were: none/normal (0 to 5 points), mild (5 to 9 points), moderate (10 to 14 points) and severe (15 to 21 points). A previous study has validated the Chinese version of the scale as a clinical screening tool for primary medical care in China (He et al., 2010). The Cronbach's α coefficient of the GAD-7 was 0.93 in this study.

The fourth section of the questionnaire involved evaluation of QoL of participants using the Chinese version of the brief version of the World Health Organization QoL (WHOQOL-BREF), which is a self-assessment and cross-cultural instrument that has been translated into several languages (WHOQOL Group, 1998). It includes four domains, namely, physical, psychological, social relationships, and environment. Of the 28 items in the WHOQOL-BREF, two focussing on overall health and general QoL are not included in the four domains. There are 26 items in the Chinese version of the WHOQOL-BREF: physical health (seven items), psychological health (six items), social relationships (three items), and environment (eight items) as well as two additional local items: ‘Does family friction affect your life?’ and ‘How is your appetite?’ (Fang, 2000). A five-point Likert scale was used with minimum and maximum scores of 1 and 5, respectively, for each question, where a higher score indicated a higher QOL. The Cronbach's α coefficient of the WHOQOL-BREF was 0.94 in this study (physical health α = 0.77, psychological health α = 0.84, social relations α = 0.74, and the environment α = 0.87).

2.3 Statistical analysis

The statistical software, SPSS version 22.0, was used for the whole study. Descriptive statistics were used to present demographic data and social-environmental support (including SSRS total score and another four self-developed items). Independent-samples t-test and one-way analysis of variance (ANOVA) were used to evaluate whether there were any significant differences between demographic data as well as that between social-environmental support items and GAD-7 scores. Scheffe post hoc test was used to check the pairwise difference between the groups.

Multiple regression analysis was used to confirm the association between social-environmental support and GAD-7 affected by the COVID-19 pandemic. The total score on GAD-7 and the scores on social-environmental support served as the dependent and independent variables, respectively. Additionally, multiple regression analysis was used to finally confirm the association between social-environmental support factors and QoL affected by the COVID-19 pandemic. The total score on WHOQOL-BREF and the scores on social-environmental support served as the dependent and independent variables, respectively. Because gender (Campos et al., 2014; Furukawa et al., 2001; Özdin & Bayrak Özdin, 2020; J. Zhang, Li, et al., 2020), age (Asar & Hakeem, 2013; Bando et al., 2015; Yueqin Huang et al., 2019), BMI (Kelderman-Bolk et al., 2015; Kukreti, 2015), monthly income (Campos et al., 2014; Maria et al., 2010; Yoshitake et al., 2016), and health status (Campos et al., 2014; Dai et al., 2020) might affect anxiety and QoL, they were controlled during the analysis. In addition, variance expansion factors were used to diagnose collinearity in multiple regression analyses in this study. However, it was found that the variable inflation factor (VIF) of all independent variables was less than 10, indicating that the issue of collinearity can be ignored (Marquardt, 1980).

3 RESULTS 3.1 Demographic characteristics and social-environmental support of study participants

The demographic data of the participants are shown in Table 1. A total of 226 Wuhan residents were invited to participate in this study. Most participants were males (69.5%). The mean age was 32.58 ± 13.67 years with an average BMI of 22.25 ± 2.96. In addition, 30.5% had monthly incomes of 2000 or below. The total score of the GAD-7 was 4.90 ± 4.06. The total score of the QoL was 57.44 ± 9.03 (Physical health 15.09 ± 2.43, psychological health 14.58 ± 2.72, social relationships 13.94 ± 2.60, environment 13.82 ± 2.63). Moreover, 42.9% of participants had moderately adequate medical supplies, while 43.8% had moderately adequate supplies of basic commodities. None of the participant was infected with COVID-19. Analysis with t-test and one-way ANOVA revealed significant differences in GAD-7 scores with respect to gender, age, monthly income, health status, medical supplies, and basic commodities (p < 0.01–0.05). The results of the Scheffe post hoc analysis are also shown in Table 1.

TABLE 1. Variations in GAD-7 and SSRS scores with demographic characteristics of study participants (n = 226) GAD-7 Total M ± SD F/t p-value (Post hoc) Gendera (n, %) −2.505 0.01* Male (a) 157 (69.5) 4.46 ± 3.99 b > a* Female (b) 69 (30.5) 5.90 ± 3.95 Ageb, (n, %) 5.957 20–39 (a) 162 (71.7) 4.48 ± 3.80 <0.01 40–59 (b) 55 (24.3) 5.51 ± 4.27 c > a* ≥60 (c) 9 (4) 8.78 ± 4.35 BMI (mean ± SD) 22.8 ± 2.96 Monthly income (Chinese yuan)b (n, %) 4.142 <0.01 Below 2000 (a) 69 (30.5) 3.41 ± 3.73 d > a* 2001–4000 (b) 28 (12.4) 6.00 ± 3.26 4001–6000 (c) 47 (20.8) 4.87 ± 3.60 6001–8000 (d) 46 (20.4) 5.89 ± 4.47 Above 8001 (e) 36 (15.9) 5.67 ± 4.36 Health statusb,c (n, %) 5.764 <0.01 Very unhealthy (a) 0 0 c > e* Unhealthy (b) 1 (4.0) 7.00 ± 0.00 Normal (c) 10 (4.4) 9.30 ± 4.79 Healthy (d) 60 (26.5) 5.50 ± 3.78 Very healthy (e) 155 (68.6) 4.37 ± 3.89 GAD-7 (mean ± SD) 4.90 ± 4.06 WHOQOL-BREF (mean ± SD) 57.44 ± 9.03 Social-environment support SSRS total scores (mean ± SD) 39.61 ± 7.75 Enough medical suppliesb (n, %) 4.344 <0.01 Not at all (a) 15 (6.6) 4.53 ± 3.20 b > e* A little (b) 62 (27.4) 5.42 ± 4.32 c > e* Moderately (c) 97 (42.9) 5.57 ± 4.12 Mostly (d) 32 (14.2) 3.84 ± 3.36 Completely (e) 20 (8.8) 2.00 ± 2.51 Adequate basic commoditiesb (n, %) 5.842 <0.01 Not at all (a) 5 (2.2) 8.00 ± 7.48 c > d* A little (b) 30 (13.3) 5.97 ± 4.00 Moderately (c) 99 (43.8) 5.74 ± 4.00 Mostly (d) 55 (24.3) 3.76 ± 3.5 Completely (e) 37 (16.4) 3.05 ± 3.5 Access to information on COVID-19b (n, %) Not at all (a) 6 (2.7) 4.17 ± 3.19 1.847 0.12 A little (b) 14 (6.2) 5.43 ± 4.01 Moderately (c) 56 (24.8) 6.00 ± 4.21 Mostly (b) 78 (34.5) 4.77 ± 3.95 Completely (e) 72 (31.9) 4.14 ± 3.92 Sufficient knowledge to cope with COVID-19b (n, %) Not at all (a) 2 (0.9) 5.50 ± 2.12 1.190 0.32 A little (b) 37 (16.4) 5.43 ± 4.83 Moderately (c) 90 (39.8) 5.38 ± 3.90 Mostly (d) 61 (27) 4.36 ± 3.52 Completely (e) 36 (15.9) 4.03 ± 4.24 Abbreviations: BMI, Body mass index; Gad-7, generalized anxiety disorder; SSRS, The Social Support Rating Scale; WHOQOL-BREF, World Health Organization Quality of Life-BREF. a Independent-Samples t-test. b One-way ANOVA and post hoc analysis using Scheffe test. c Post hoc analysis using Kruskal–Wallis test. p < 0.05. 3.2 Multiple regression analysis

The multiple regression analysis on the scores of the GAD-7 is depicted in Table 2. The results demonstrated that social-environment support total score were the factors negative correlation with GAD-7 (β = −0.24, p < 0.01). In addition, the multiple regression analysis on the scores of the four domains of WHOQOL-BREF is depicted in Table 3, The results demonstrated that social-environment support were the factors positively correlated with QoL (including physical and emotional health, social relationships, and environment) (β = 0.09–0.14, p < 0.01).

TABLE 2. Multiple regression analysis on the scores of the GAD-7 Model R2 Adjusted R2 F B SE β p 0.18 0.16 7.89 Gender 1.80 0.59 0.21 <0.01 Age 0.00 0.03 0.00 0.99 BMI 0.25 0.09 0.19 0.01* Monthly income 0.49 0.21 0.18 0.02* Health state −0.98 0.50 0.14 0.05 Social-environment support −0.11 0.03 −0.24 <0.01 Abbreviations: BMI: body mass index; GAD-7, generalized anxiety disorder. p < 0.05. TABLE 3. Multiple regression analysis on the scores of the four domains of WHOQOL-BREF Model R2 Adjusted R2 F B SE β p Physical health 0.21 0.19 9.51 Gender −0.43 0.35 −0.08 0.22 Age