The psychological burden of the COVID-19 pandemic and associated lockdown measures: Experience from 4000 participants

Highlights

• •The COVID-19 pandemic had significant mental health impacts on the population of Kuwait.
• •Unemployment, uncertainty, distress, increasing deaths and lockdown measures contribute to the burden.
• •From our experience risk factors include; female, smoker, past psychiatric history, increased social media use.
• •Availability of psychological support and mental health awareness reduces the burden.

Abstract

Background

In February 2020 the first cases of COVID-19 were identified in Kuwait. Inevitably as many countries worldwide, the general public were negatively affected by the pandemic. Unemployment, uncertainty, distress, increasing deaths, lockdown measures all of which are potential burdens on mental health.

Aim

To assess the impact of COVID19 outbreak on mental health in Kuwait, and to explore the potential influencing risk factors.

Methods

We conducted an online questionnaire-based study in Kuwait between 25^th May 2020 to 30^th May 2020. Questions were based on demographics, lifestyle during outbreak, depression and anxiety assessment. Total PHQ9 and GAD7 scores were calculated for each responder.

Results

We had 4132 responders. Most were females (69.31%), married (59.37%), between the age of 21-30 (23.84%) and non-smokers (81.46%). Only (7.96%) had a positive past psychiatric history, (32.04%) had a past history of a chronic medical disease. During the outbreak most of the responders lost their jobs (39.21%) and only (12.83%) were attending work regularly, only (6.82%) worked in the healthcare sector. 59.27% report increased social media use compared to before the lockdown. When asked about their daily time spent following COVID19-related news, most (37.8%) spend more than 2 hours and (7.74%) spend more than 4 hours. The overall prevalence of depressive symptoms was (30.13%) and the prevalence of anxiety symptoms was (25.28%).

Limitations

In the cross-sectional nature of the study.

Conclusion

The COVID-19 pandemic caused a burden on mental health. Psychological support and mental health awareness should be implemented and made accessible to all individuals during pandemics.

1. Introduction

The novel severe acute respiratory syndrome coronavirus 2 (COVID-19) emerged in December 2019, when cases of pneumonia of unknown cause were identified in people that visited a wholesale seafood market in Wuhan, China (Zhu et al., 2020). The COVID-19 virus is primarily transmitted through droplets, affecting those within close proximity of a host. The most common symptoms of COVID-19 are fever and cough (Guan et al., 2020). Since its outbreak in December of 2019, COVID-19 has affected several countries worldwide and has been declared a pandemic in early March 2020 by the World Health Organization (WHO) (World Health organization 2020). Countries all over the world implemented social distancing rules and lockdown measures, in an attempt to limit the spread of the virus. The pandemic has caused noticeable psychological distress and public panic, this is partly due to the false information being spread online (Bao et al., 2020) or the fear of contracting the virus. (Brooks et al., 2020)

In February 24 2020 the first few cases of COVID-19 were identified in Kuwait, all of which arrived from the Iranian city of Mashhad. The government of Kuwait rushed to set up field hospitals and quarantine facilities in locations such as schools, fairgrounds and gymnasiums in anticipation of an increase in the number of cases. Over the following month the number of positive cases increased and on March 23, Kuwait announced its first partial lockdown measures. Malls, restaurants, gyms, schools and universities were all to remain fully closed, resulting in many losing their jobs. Volunteers joined the front line forces working in supermarkets to ensure food and supplies remained in abundance and to provide a delivery service to households. Unfortunately even with the partial lockdown in place cases were on the rise, the Kuwait council of ministers decided to implement a total lockdown on all activities from May 8 2020 to May 30 2020 (Al Arabiya Network 2020). No persons were allowed outside of their homes. As of June 14 2020, in Kuwait there has been a total of 36,958 cases of COVID-19, while Worldwide there has been 7,690,708 cases and 427,630 confirmed deaths (World Health Organization 2020).

1.1. Aim

To assess the psychological burden of the COVID-19 pandemic and associated lockdown measures on the public in Kuwait, and to explore the potential influencing risk factors.

2. Methods

2.1. Study design and participants

We conducted an online questionnaire-based cross- sectional study between 25^th May 2020 to 30^th May 2020. This corresponds to the final week of the complete lockdown measures implemented by the government of Kuwait. A web-based survey was sent through various social media platforms, along with the reason behind the questionnaire and a disclosure that the questionnaire is anonymous and confidential. A consent was provided to all participants prior to proceeding to the questionnaire. All of the responders were over the age of 18 and incomplete questionnaires were omitted from analysis. The questionnaire included 36 questions involving demographics, lifestyle during the pandemic, depression and anxiety assessments.

Age, gender, marital status (single, married/in a relationship, divorce/separated, widowed), smoking history, height and weight (in which we calculated the BMI and classified patients into underweight, normal, overweight or obese based on the WHO classification), education obtained (high school, university bachelor's/diploma, postgraduate degree; which included master's degree, PhD or MD), medical history of chronic diseases and past psychiatric history.

We also collected information regarding daily life during the pandemic. Those questions included, current work status (working/studying from home, work suspended, attending work as usual, retired prior to lockdown or unemployed prior to lockdown), exercise frequency per week (0, 1-2, 3-4, 5-6, daily), volunteering in community service (yes, or no), work in the health care sector (yes or no); healthcare sector includes staff who are directly involved in daily patient care on the wards (doctors, nurses, physiotherapists, dieticians).

Perceived risk of acquiring the virus at work (no, yes working with positive cases, yes working with suspected cases). This includes individuals who are at risk of acquiring the virus at work but who are not working in the healthcare sector e.g janitors at hospitals and quarantine sites, catering services, receptionists at hospitals, security staff. If they were positive for COVID-19 (no, yes mild symptoms, yes severe symptoms), if any of their close friends or family members positive for COVID-19, if they were quarantined (either suspected or positive), daily social media use during the outbreak compared with before the outbreak (less than before, similar to before, more than before). The daily duration spent following COVID19-related news (less than 30 minutes, 30minutes to 1 hour, 1-2 hours, 2-4 hours and more than 4 hours). Responders were also asked to rate the government protocols on a scale of 1-10 (0-1, 2-3, 4-5, 6-7, 8-9, 10).

2.2. Assessment of anxiety/depression

Symptoms of depression were assessed using the 9-item Patient Health Questionnaire (PHQ-9) score and symptoms of anxiety were assessed using the Generalized Anxiety Disorder scale-7 (GAD7) score. The PHQ-9 is a self-report questionnaire that is commonly used as a screening tool for depression (Kroenke et al., 2001). It is composed of 9 questions assessing the frequency of depressive symptoms on a 4-point Liker-scale ranging from 0 (never) to 3 (nearly every day). The total score was calculated for each patient and was interpreted as follows: minimal (1-4), mild (5-9), moderate (10-14), moderately severe (15-21), severe (20-27). A standard cutoff of score of 10 was used, indicating a diagnosis of depression (Levis and Benedetti, 2019, Moriarty et al., 2015).

The GAD-7 was initially developed as a screening tool for generalized anxiety disorder (GAD) (Spitzer et al., 2006), but proved to have good sensitivity and specificity as a screening tool for panic disorder, social anxiety disorder, and post-traumatic stress disorder (Kroenke et al., 2007). Seven items are used to assess the frequency of anxiety symptoms over a period of two weeks on a 4-point liker-scale ranging from 0 (never) to 3 (nearly every day). The total score of GAD-7 ranges from 0 to 21, with increasing scores indicating more severe functional impairments as a result of anxiety. The total score was calculated for each patient and was interpreted as follows: normal (0-4), mild (5-9), moderate (10-14), and severe (15-21). A cut-off score of 8 was used, indicating a diagnosis of anxiety. (Kroenke et al., 2007)

2.3. Statistical analysis

Statistical analysis was performed on 4132 observations and 22 variables, with the aim to find out whether there is any relationship between anxiety, depression and the analyzed variables. The target sample size was determined using the formula $n=\frac{z_{\alpha }^{2}p\left(1-p\right)}{e^2}$, where $z_{\alpha }=1.96$ for a confidence level of 95% (α = 0.05). The estimated margin of error (e) in the survey was  ± 1.4%, for the proportion (p) of depression and anxiety was 0.3. For ordinal and nominal variables, we used the Chi-squared test of independence to determine whether there is a significant association between two variables. The null hypothesis states that the two tested variables are independent, i.e., there is no association in the underlying bivariate population. It is rejected, if the probability associated with the test statistic (p-value), that is of observing a sample statistic as extreme as the test statistic, is less than the significance level (α = 0.05). In case we reject the null hypothesis, we can conclude that there is a relationship between the two variables. Then we calculated the standardized residuals (SR) to show the dependence for the significant results (α = 0.05) from the χ ² test. Positive values of SR are interpreted as having greater observed frequency than would be found under independence assumption, and negative values indicate that they occur less often than under independence. The SR > |2| are considered significant at alpha = 0.05, and SR > |4| at α = 0.0001. With SR < |2| we are unable to say whether they are significant.

For continuous variables, we compared the means in the two groups (anxiety vs. no anxiety, depression vs. no depression) in relation to anxiety and to depression, by calculating the W statistics derived from the two-tailed Mann-Whitney-Wilcoxon tests, at α = 0.05. The significant results are listed in Table 3. We calculated the point-biserial correlation coefficient r_pb to measure the strength of association between a continuous and a binary variable (anxiety and depression). The coefficient ranges from -1 (negative association) to +1 (positive association), and zero indicates no association.

Table 3.

Significant results from the Mann – Whitney test

Anxiety	No	Yes	W	P-value
Height (cm)	164.85	163.87	1741897	0.004
BMI	28.17	27.61	1743026	0.003
Weight (kg)	76.79	74.42	1741897	< 0.01
Depression	No	Yes	W	P-value
Weight (kg)	76.42	75.56	1853205	0.046

Furthermore, we performed a univariate logistic regression analysis, where we calculated the odds ratios (OR) to find out whether the factor under examination has an impact on depression, and on anxiety, respectively. A low odds ratio (OR < 1) indicates that the potential risk factor is associated with a decrease of odds related to depression, and anxiety. A high odds ratio (OR > 1) indicates an increase of odds related to anxiety and depression. The odds ratio of 1 indicates no change of odds related to anxiety and depression. The statistically significant variables were considered for the multivariable logistic regression analysis, what we performed to determine potential risk factors for depression and anxiety. We adjusted for confounders and calculated the odds ratios, the 95% confidence intervals (CI) and the corresponding p-values.

3. Results

We had 4132 responders above the age of 18 who completed the questionnaire fully, and were hence included in the statistical analysis (Table 1 ). Most responders were females 69.31% (n = 2864), married 59.37% (n = 2453), between the age of 21-30 23.84% (n = 985) and non-smokers 81.46% (n = 3366). Most responders obtained a university bachelor's degree 71.59% (n = 2958). Only 7.96% (n = 329) had a positive past psychiatric history, all of which were on maintenance treatment with no relapse in the past year, 32.04% (n = 1324) had a past history of a chronic medical disease. Only some responders provided details regarding their psychiatric conditions, most common conditions were depression and generalized anxiety disorder (Table 5) and most common medical conditions were hypertension, diabetes mellitus, cardiovascular disease, hypothyroidism, inflammatory bowel disease (IBD) and rheumatoid arthritis.

Table 1.

Demographics of responders (N = 4132).

Variable	n (%)
Total	4132 (100.0)
Gender
Male	1268 (30.7)
Female	2864 (69.3)
Marital status
Single	1339 (32.4)
Married	2453 (59.4)
Widowed	105 (2.5)
Separated/Divorced	235 (5.7)
Smoker
No	3366 (81.5)
Yes	766 (18.5)
Job status during lockdown
Working/studying from home
Attending work as usual	530 (12.8)
Stopped working/studying	1620 (39.2)
Retired prior to lockdown	856 (20.7)
Unemployed prior to lockdown	310 (7.5)
Volunteering in community serviceA
No	3790 (91.7)
Yes	342 (8.3)
ExercisingB
No	1608 (38.9)
Yes, 1-2 times per week	1043 (25.2)
Yes, 3-4 times per week	649 (15.7)
Yes, 5-6 times per week	400 (9.7)
Yes, every day of the week	432 (10.5)
Working in the health sectorC
No	3850 (93.2)
Yes	282 (6.8)
Risk of acquiring the virus at work
No	3316 (80.3)
Yes, dealing with positive cases	107 (2.6)
Yes, dealing with suspected cases	709 (17.2)
Psychiatric history
No	3803 (92.0)
Yes	329 (8.0)
Medical history of chronic disease
No	2808 (68.0)
Yes	1324 (32.0)
Where you diagnosed with COVID-19
No	4110 (99.5)
Yes	22 (0.5)
COVID-19 diagnosis in friends/family
No	2819 (68.2)
Yes	1313 (31.8)
Were you quarantined
No	3431 (83.0)
Yes, Home	617 (14.9)
Yes, institutional	84 (2.0)
Depression
No	2887 (69.9)
Yes	1245 (30.1)
Anxiety
No	3046 (73.7)
Yes	1086 (26.3)
BMI
Overweight	1469 (35.6)
Normal	1294 (31.3)
Obese	1291 (31.2)
Underweight	61 (1.5)
Other (not disclosed)	17 (0.4)
Age (years)
<=20	276 (6.7)
21-30	985 (23.8)
31-40	874 (21.2)
41-50	756 (18.3)
51-60	839 (20.3)
61-70	363 (8.8)
>=71	39 (0.9)
Education Obtained
High School	592 (14.3)
University Bachelors/Diploma	2958 (71.6)
Post- graduate degree.X	582 (14.1)
Rating of government protocols
0-1	172 (4.2)
2-3	228 (5.5)
4-5	716 (17.3)
6-7	1356 (32.8)
8-9	1110 (26.9)
10	550 (13.3)
Time spent on social media per day during lockdown
Less than before	514 (12.4)
Same as before	1169 (28.3)
More than before	2449 (59.3)
Time spent following COVID19-related news per day
Less than half an hour	909 (22.0)
30 mins to 1 hour	641 (15.5)
1-2 hours	700 (16.9)
More than 2 hours	1562 (37.8)
More than 4 hours	320 (7.7)
Average height (cm)	164.6 (117-198)
Average weight (kg)	76.2 (63-191)
Average BMI	28.0 (13.8-71)

^XPostgraduate degree; Master's degree, PhD, MD.

^ACommunity service volunteering; includes individuals who volunteered in supermarkets.

^BExercise; walking, cardio, weights.

^CHealth care sector; individuals directly involved in patient care on the wards. Doctors, nurses, dieticians, physiotherapists.

Table 5.

Frequency of psychiatric disorders.

Psychiatric diroder	Frequency
Depression	150
GAD	71
OCD	24
Panic disorder	12
ADHD	12
Bipolar disorder	11
Post partum depression	4
PTSD	4
Anorexia Nervosa	3
Specific phobia (Claustropgobia, aerophobia)	3
Emotionally unstable personality disorder (EUPD)	2
Trichotillomania	2
Social phobia	2
Agoraphobia	1
Hypochondriasis	1
Body dismorphic disorder	1

During the pandemic most of the responders lost their jobs 39.21% (n = 1620) and only 12.83% (n = 530) were attending work regularly. Only 6.82% (n = 282) worked in the healthcare sector, this includes anyone who is directly involved in patient care on the wards (doctors, nurses, physiotherapists, dieticians). The prevalence of responders who perceive a risk of acquiring the virus was 19.75% (n = 816), 2.59% (n = 107) work with positive cases while 17.16% (n = 709) work with suspected cases. Despite the fact that most people were not working, when asked about volunteering only 8.28% (n = 342) answered yes. People did experience a lot of free time during the lockdown, even in those responders that still attended work. Therefore, responders were asked about how they spent their time during lockdown. Most 61.08% (n = 2524) exercise at least 1-2 times a week (Table 1). When asked about social media use during lockdown compared to before, 59.27% (n = 2449) report increased usage. When asked about their daily time spent following COVID-19 related news, 37.8% of responders (n = 1562) spend more than 2 hours and 7.74% of responders (n = 320) spend more than 4 hours. Most of our responders, 99.47% (n = 4110) were not affected by the virus themselves, but a higher percentage of 31.78% (n = 1313) had either close friends or family members who were positive for the virus.

3.1. Depression/ anxiety

The overall prevalence of depression was 30.13% (n = 1245), and the prevalence of anxiety symptoms was 25.28% (n = 1086). Significant positive and negative associations are summarized in (Table 2 , 4 , 6 and 7 ).

Table 2.

Significant results from the χ² test, Depression.

Variable	n (%)	Depression	No depression	χ2	P-value
Total	4132 (100.0)	2887 (69.9)	1245 (30.1)
Gender
Male	1268 (30.7)	952 (33.2)	316 (11.0)	23.2	< 0.0001
Female	2864 (69.3)	1935 (67.6)	929 (32.4)
Marital status
Single	1339 (32.4)	809 (28.2)	530 (18.5)	101.2	< 0.0001
Married	2453 (59.4)	1838 (64.2)	615 (21.5)
Widowed	105 (2.5)	89 (3.1)	16 (0.6)
Separated/Divorced	235 (5.7)	151 (5.3)	84 (2.9)
Smoker
No	3366 (81.5)	2403 (83.9)	963 (33.6)	19.6	< 0.0001
Yes	766 (18.5)	484 (16.9)	282 (9.8)
Job status
Working/studying from home	816 (19.8)	555 (19.4)	261 (9.1)	110.6	< 0.0001
Attending work as usual	530 (12.8)	371 (13.0)	159 (5.6)
Stopped working/studying	1620 (39.2)	1039 (36.3)	581 (20.3)
Retired prior to the lockdown	856 (20.7)	719 (25.1)	137 (4.8)
Unemployed prior to the lockdown	310 (7.5)	203 (7.1)	107 (3.7)
Exercise
No	1608 (38.9)	1049 (25.39)	559 (13.53)	79.0	< 0.0001
Yes, 1-2 times per week	1043 (25.2)	687 (16.63)	356 (8.62)
Yes, 3-4 times per week	649 (15.7)	475 (11.50)	174 (4.21)
Yes, 5-6 times per week	400 (9.7)	327 (7.91)	73 (1.77)
Yes, every day of the week	432 (10.5)	349 (8.45)	83 (2.01)
Working in the health sector
No	3850 (93.2)	2707 (65.51)	1143 (27.66)	4.9	0.0262
Yes	282 (6.8)	180 (4.36)	102 (2.47)
Risk of acquiring the virus at work
No	3316 (80.3)	2378 (57.55)	938 (22.70)	27.3	< 0.0001
Yes, dealing with positive cases	107 (2.6)	65 (1.57)	42 (1.02)
Yes, dealing with suspected cases	709 (17.2)	444 (10.75)	265 (6.41)
Psychiatric history
No	3803 (92.0)	2714 (65.68)	1089 (26.36)	49.8	< 0.0001
Yes	329 (8.0)	173 (4.19)	156 (3.78)
COVID-19 diagnosis in friends/family
No	2819 (68.2)	2009 (48.62)	810 (19.60)	8.0	0.0046
Yes	1313 (31.8)	878 (21.25)	435 (10.53)
Were you quarantined?
No	3431 (83.0)	2452 (59.34)	979 (23.69)	24.5	< 0.0001
Yes, Home	617 (14.9)	383 (9.27)	234 (5.66)
Yes, Institutional	84 (2.0)	52 (1.26)	32 (0.77)
Age (years)
<=20	276 (6.7)	147 (3.56)	129 (3.12)	224.3	< 0.0001
21-30	985 (23.8)	600 (14.52)	385 (9.32)
31-40	874 (21.2)	566 (13.70)	308 (7.45)
41-50	756 (18.3)	523 (12.66)	233 (5.64)
51-60	839 (20.3)	687 (16.63)	152 (3.68)
61-70	363 (8.8)	327 (7.91)	36 (0.87)
>=71	39 (0.9)	37 (0.90)	2 (0.05)
Education Obtained
High School	592 (14.3)	383 (9.27)	209 (5.06)	9.2	0.01018
University Bachelors/diploma	2958 (71.6)	2086 (50.48)	872 (21.10)
Post-graduate degree	582 (14.1)	418 (10.12)	164 (3.97)
Rating of government protocols
0-1	172 (4.2)	98 (2.37)	74 (1.79)	75.1	< 0.0001
2-3	228 (5.5)	140 (3.39)	88 (2.13)
4-5	716 (17.3)	450 (10.89)	266 (6.44)
6-7	1356 (32.8)	928 (22.46)	428 (10.36)
8-9	1110 (26.9)	845 (20.45)	265 (6.41)
10	550 (13.3)	426 (10.31)	124 (3.00)
Time spent on social media per day
Less than before	514 (12.4)	373 (9.03)	141 (3.41)	112.6	< 0.0001
Same as before	1169 (28.3)	948 (22.94)	221 (5.35)
More than before	2449 (59.3)	1566 (37.90)	883 (21.37)
Time spent following COVID-19 related news per day
Less than half an hour	909 (22.0)	674 (16.31)	235 (5.69)	73.4	< 0.0001
30 mins to 1 hour	641 (15.5)	499 (12.08)	142 (3.44)
1-2 hours	700 (16.9)	518 (12.54)	182 (4.40)
More than 2 hours	1562 (37.8)	1009 (12.42)	553 (13.38)
More than 4 hours	320 (7.7)	187 (4.53)	133 (3.22)

Table 4.

Significant results from the χ² test, anxiety.

Variable	n (%)	Anxiety	No anxiety	χ2	P-value
Total	4132 (100)	1086(26.28)	3046(73.72)
Gender
Male	1268(30.69)	240(5.81)	1028(24.88)	50.5	< 0.0001
Female	2864(69.31)	846(20.47)	2018(48.84)
Marital Status
Single	1339(32.41)	464(11.23)	875(21.18)	81.2	< 0.0001
Married	2453(59.37)	549(13.29)	1904(46.08)
Widowed	105(2.54)	11(0.27)	94(2.27)
Seperated/Divorced	235(5.69)	62(1.50)	173(4.19)
Job Status
Working/studying from home	816(19.75)	236(5.71)	580(14.04)	94.4	< 0.0001
Attending work as ususal	530(12.83)	132(3.19)	398(9.63)
Stopped working/studying	1620(39.21)	509(12.32)	1111(26.89)
Retired prior to the lockdown	856(20.72)	119(2.88)	737(17.84)
Unemployed prior to the lockdown	310(7.5)	90(2.18)	220(5.32)
Exercise
No	1608(38.92)	481(11.64)	1127(27.27)	17.6	< 0.0001
Yes	2524(61.08)	605(14.64)	1919(46.44)
Exercise
No	1608(38.92)	1127(27.27)	481(11.64)
Yes, 1-2 times per week	1043(25.24)	751(18.18)	292(7.07)
Yes, 3-4 times per week	649(15.71)	491(11.88)	158(3.82)	39.23	< 0.0001
Yes, 5-6 times per week	400(9.68)	322(7.79)	78(1.89)
Yes, everyday of the week	432(10.45)	355(8.59)	77(1.86)
Working in the healthcare sector
No	3850(93.18)	990(23.96)	2860(69.22)	9.0	0.0027
Yes	282(6.82)	96(2.32)	186(4.50)
Risk of acquiring the virus at work
No	3316(80.25)	828(20.04)	2488(60.21)	15.0	0.00055
Yes, dealing with positive cases	107(2.59)	35(0.85)	72(1.74)
Yes, dealing suspected cases	709(17.16)	223(5.40)	486(11.76)
Past psychiatric history
No	3803(92.04)	926(22.41)	2877(69.63)	90.9	< 0.0001
Yes	329(7.96)	160(3.86)	169(4.09)
COVID-19 diagnosis in close friends/family
No	2819(68.22)	678(16.41)	2141(51.82)	22.4	< 0.0001
Yes	1313(31.78)	408(9.87)	905(21.90)
Were you quarantined?
No	3431(83.03)	849(20.55)	2582(62.49)	24.8	< 0.0001
Yes, Home	617(14.93)	210(5.08)	407(9.85)
Yes, institutional	84(2.03)	27(0.65)	57(1.38)
BMI
Overweight	1469(35.55)	370(8.95)	1099(26.60)	9.2	0.027
Normal	1294(31.32)	376(9.10)	918(22.22)
Obese	1291(31.24)	318(7.70)	973(23.55)
Underweight	61(1.48)	20(0.48)	41(0.99)
Other (not disclosed)	17(0.41)	NA	NA
Age
<=20	276(6.68)	121(2.93)	155(3.75)	197.7	< 0.0001
21-30	985(23.84)	337(8.16)	648(15.68)
31-40	874(21.15)	261(6.32)	613(14.84)
41-50	756(18.3)	205(4.96)	551(13.33)
51-60	839(20.30)	123(2.98)	716(17.33)
61-70	363(8.79)	34(0.82)	329(7.96)
>=71	39(0.94)	5(0.12)	34(0.82)
Education obtained
High School	592(14.33)	186(4.50)	406(9.83)	9.5	0.009
University Bachelors/Diploma	2958(71.59)	749(18.13)	2209(53.46)
Post-graduate degree	582(14.09)	151(3.65)	431(10.43)
Rating of government protocols
0-1	172(4.16)	61(1.48)	111(2.69)	63.9	< 0.0001
2-3	228(5.52)	76(1.84)	152(3.68)
4-5	716(17.33)	235(5.69)	481(11.64)
6-7	1356(32.82)	379(9.17)	977(23.64)
8-9	1110(26.86)	216(5.23)	894(21.64)
10	550(13.31)	119(2.88)	431(10.43)
Time spent on social media per day
Less than before	514(12.44)	124(3)	390(9.44)	40.9	< 0.0001
Same as before	1169(28.29)	233(5.64)	936(22.65)
More than before	2449(59.27)	729(17.64)	1720(41.63)
Time spent following COVID-19-related news per day
Less than half an hour	909(22)	213(5.15)	696(16.84)	46.6	< 0.0001
30 mins to 1 hour	641(15.51)	132(3.19)	509(12.32)
1-2 hours	700(16.94)	154(3.73)	546(13.21)
More than 2 hours	1562(37.8)	477(11.54)	1085(26.26)
More than 4 hours	320(7.74)	110(2.66)	210(5.08)

Table 6.

Risk factors for anxiety, identified by the multivariable logistic regression.

Variable	n (%)	Adjusted OR (95% CI)	P-value
Total	4132 (100.0)
Gender
Female	2864 (69.3)	2.11 (1.71 - 2.52)	< 0.001
Job status
Stopped working/studying	1620 (39.2)	1.39 (1.04 - 1.86)	0.026
Working in the health sector
Yes	282 (6.8)	1.43 (1.03 - 1.98)	0.032
Risk of acquiring the virus at work
Yes, dealing with suspected cases	709 (17.2)	1.38 (1.11 - 1.72)	0.004
Past psychiatric history
Yes	329 (8.0)	2.63 (2.06 - 3.36)	< 0.001
Were you quarantined?
Home	617 (14.9)	1.31 (1.06 - 1.60)	0.01
Age (years)
51-60	839 (20.3)	1.61 (1.14 - 2.25)	0.005
Rating of government protocols
6-7	1356 (32.8)	1.42 (1.16 - 1.74)	0.001
Time spent on social media per day
more than before	2449 (59.3)	1.28 (1.09 - 1.50)	0.002
Time spent following COVID-19-related news per day
30 mins to 1 hour	641 (15.5)	1.35 (1.09 - 1.67)	0.006
< 30 min.	909 (22.0)	1.27 (1.04 - 1.55)	0.017

Table 7.

Risk factors for depression, identified by the multivariable logistic regression.

Variable	n (%)	Adjusted OR (95% CI)	P-value
Total	4132 (100.0)
Gender
Female	2864 (69.3)	2.06 (1.72 - 2.48)	< 0.001
Smoker
Yes	766 (18.5)	1.59 (1.30 - 1.95)	< 0.001
Job status
Stopped working/studying	1620 (39.2)	1.24 (1.10 - 1.84)	0.007
Risk of acquiring the virus at work
Yes, dealing with positive cases	107 (2.6)	1.68 (1.06 - 2.63)	0.025
Yes, dealing with suspected cases	709 (17.2)	1.52 (1.23 - 1.86)	< 0.001
Psychiatric history
Yes	329 (8.0)	1.96 (1.53 - 2.50)	< 0.001
Were you quarantined?
Yes, Home	617 (14.9)	1.49 (1.22 - 1.81)	< 0.001
Yes, Institutional	84 (2.0)	1.70 (1.04 - 2.73)	0.031
Rating of government protocols
6-7	1356 (32.8)	1.60 (1.31 - 1.95)	< 0.001
Time spent on social media per day
Same as before	1169 (28.3)	1.29 (1.10 - 1.51)	0.002
More than before	2449 (59.3)	1.60 (1.37 - 1.87)	< 0.001
Time spent following COVID-19-related news per day
30 mins to 1 hour	641 (15.5)	1.65 (1.35 - 2.02)	< 0.001
< 30 min.	909 (22.0)	1.37 (1.13 - 1.66)	0.001
Exercise
Yes, 5-6 times per week	400 (9.68)	1.33 (1.08 - 1.64)	0.007

Women have a higher tendency for depression (SR = 2.25) and for anxiety (SR = 3.40), while men have a lesser tendency for depression (SR = 3.38) and anxiety (SR = -5.11). Responders who were single show a very high tendency for depression (SR = 6.30) and anxiety (SR = 5.97). On the contrary, married individuals tend to have no depression (SR = -4.56), nor anxiety (SR = -3.77). Smokers tend to be diagnosed with depression more often (SR = 3.37). But the association with anxiety was not statistically significant. People who stopped working during the curfew times, seem to have very high tendency to feel depressed (SR = 4.20) and anxious (SR = 4.03). Retired people, on the contrary seem to be very unlikely to get seriously depressed (SR = -7.53) or be anxious (SR = -7.07). People who work at the health sector are more likely to experience anxiety, but no statistically significant results for higher rates of depression. Workers who perceive a risk of acquiring the virus at work due to suspected cases, are more likely to have anxiety and depression (SR = 3.51). Individuals with psychiatric history, have very high tendency to be diagnosed with depression (SR = 5.71), and with anxiety (SR = 7.91). People who have friends or family members diagnosed with COVID-19, tend to have anxiety, and also a slight tendency for depression (SR = 1.98). Individuals at home quarantine also tend to have the diagnosis of anxiety (SR = 3.76) and depression (SR = 3.53). People who are younger than 30 years old, show a very high tendency for depression (SR = 5.12) and anxiety (SR = 4.85). People between 31 and 40 years have a lesser tendency for depression (SR = 2.75) and anxiety (SR = 2.06). Those between the age of 51 and 70 displayed an even lower tendency for depression (SR = -7.02) and anxiety (SR = -6.29) as shown in Fig. 1 . Individuals with high school degree tend to be more depressed (SR = 2.29) and more anxious (SR = 2.44). There is a strong negative association between high ratings of government protocols and anxiety (SR = -4.43) as well as depression (SR = -3.80). Showing that people who were more satisfied with government protocols has less anxiety and depression. There is a very strong negative association between depression (SR = - 6.99) and anxiety (SR = - 4.24), respectively, and those spending the same amount of time on social media as they did before. People spending same amount of time on social media during the lockdown as they did prior to the lockdown have less anxiety and depression. But people who spend more time on the social media than before, have very strong tendency to be diagnosed with depression (SR = 5.34) and anxiety (SR = 3.36). Also, people who spend more than two hours per day following COVID-19 related news tend to have depression (SR = 3.80) and anxiety (SR = 3.28). Exercising more than 5 times per week seems to be very useful to avoid depression (SR = -4.33) and anxiety (SR = -2.65). There is a statistically significant difference between the average body mass index (BMI) of people who were diagnosed with anxiety and those who were not. We also found out that there is a statistically significant difference between the average weight of people diagnosed with depression and those without depression. We observed a slight positive association between anxiety and height (r_pb = 0.047), weight (r_pb = 0.057) and BMI (r_pb = 0.042). However, these results show no strong association between depression and height, weight or BMI.

Fig. 1.

Mosaic plot of age categories versus depression (0 – no, 1-yes), showing strong negative and positive associations.

The multivariable logistic regression analysis showed that, after controlling for confounding factors, being a woman (OR = 2.11, CI = (1.71 - 2.52), p-value < 0.001), at the age of 51-60 (OR = 1.61, CI = (1.14 - 2.25), p-value = 0.005), working at the health sector (OR = 1.43, CI = (1.03 - 1.98), p-value = 0.032), being at risk of acquiring the virus at work from suspected cases (OR = 1.38, CI = (1.11 - 1.72), p-value = 0.004), stopped working or studying (OR = 1.39, CI = (1.04 - 1.86), p-value = 0.026), with a psychiatric history (OR = 2.63, CI = (2.06 - 3.36), p-value < 0.001), in a quarantine at home (OR = 1.31, CI = (1.06 - 1.60), p-value = 0.01), rating the government protocols 6-7 (OR = 1.42, CI = (1.16 - 1.74), p-value = 0.001), using social media more than before the pandemic (OR =1.28, CI = (1.09 - 1.50), p-value = 0.002), and following the daily news related to COVID-19 for less than 1 hour per day (OR = 1.35 (1.09 - 1.67), p-value = 0.006) were significant risk factors for anxiety. (Shown in Table 6) Similarly, being a woman (OR = 2.06, CI = (1.72 - 2.48), p-value < 0.001), smoking (OR = 1.59, CI = (1.30 - 1.95), p-value < 0.001), being at risk of acquiring the virus at work from suspected (OR = 1.52, CI = (1.23 - 1.86), p-value < 0.001) or positive cases (OR = 1.68, CI = (1.06 - 2.63), p-value = 0.025), stopped working or studying (OR = 1.24, CI = (1.10 - 1.84), p-value = 0.007), with a psychiatric history (OR = 1.96, CI = (1.53 - 2.50), p-value < 0.001), in a home- (OR = 1.49, CI = (1.22 - 1.81), p-value < 0.001) or institutional quarantine (OR = 1.70, CI = (1.04 - 2.73), p-value = 0.031), rating the government protocols 6-7 (OR = 1.60, CI = (1.31 - 1.95), p-value < 0.001), using social media more than before the curfew (OR = 1.60, CI = (1.37 - 1.87), p-value < 0.001), and following the daily news related to COVID-19 for less than 1 hour per day (OR = 1.65, CI = (1.35 - 2.02), p-value < 0.001) were significant risk factors for depression, as shown in Table 7.

4. Discussion

The study suggests that the COVID-19 pandemic had a significant impact on the mental health of our population. This is similar to what was seen in several studies undertaken during to the pandemic (Ahmed et al., 2020, Cullen et al., 2020, González-Sanguino et al., 2020, Guo et al., 2020, Huang and Zhao, 2020, Wang et al., 2020, Mazza et al., 2020). A systematic review of 19 studies by Xiong, J et al showed a prevalence of anxiety symptoms reaching up to 50.9%, and a prevalence of depressive symptoms ranging between 14.6%-48.3% (Xiong et al., 2020). Many factors were found to be associated with an increased psychological burden including; female gender, age <40 years, chronic or psychiatric illnesses, unemployment, student status, frequent use of social media and prolonged time spent following news concerning COVID-19 (Xiong et al., 2020, Salari et al., 2020).

In comparison to studies undertaken in Saudi Arabia (Alkhamees et al., 2020), Algeria (Madani et al., 2020), Bosnia (Sljivo et al., 2020), Cyprus (Solomou and Constantinidou, 2020) and Turkey (Özdin and Bayrak Özdin, 2020), Kuwait showed a lower prevalence of anxiety symptoms, this is possibly due to the widespread awareness and psychological support provided by the authorities during the pandemic. Daily live press conferences were held on national television, detailing the number of new and recovered cases along with an update on any future plans being discussed by the government. Furthermore Kuwait launched an online website (coronacare.kw) providing free psychological counseling sessions along with educational videos focusing on anxiety and stress coping mechanisms. Other services provided by the website include simplified short videos containing information about the virus and methods to reduce transmission. This obviated the distress that can be brought about by some of the false information circulating online.

We identified a number of risk factors; those with family members who were confirmed positive for COVID-19 showed higher levels of anxiety. Healthcare workers are significantly at higher risk of psychological sequelae. This is supported by other studies (Huang and Zhao, 2020, Zhu et al., 2020, Chew et al., 2020, Lu et al., 2020). Potential reasons include; increased working hours, feeling helpless and having to self-isolate from families and partners. Smokers showed a higher level of depression. Women, subjects of single marital status, those who stopped working during the pandemic, having had a past psychiatric history, isolated to home quarantine, subjects younger than 30 and those who spent more time on social media than prior to the pandemic showed significantly increased levels of both depression and anxiety. Individuals with a history of psychiatric illness have a higher risk for depression and anxiety during the COVID-19 pandemic. Fear of infection, isolation, loss of jobs, financial stress and reduced social support during quarantine can trigger a relapse.

Interestingly we found a negative correlation between increased satisfaction rates with government protocols and levels of psychological burden. Feeling of safety and trust in the country's standards of care including decisions made by the government are likely contributory factors. Older populations of >50 years of age, as well as those retired seemed to be unlikely to experience psychological impacts of the pandemic compared to their younger counterparts. However, younger populations (<30) as well those with high school education degrees showed higher levels of anxiety and depression. Schools, parks, playground have been shut down in Kuwait, and adolescents were not able engage in social activities. Delays in examinations, applications to universities, pandemic-driven uncertainty regarding future decisions and opportunities are potential factors. Kuwait also did not have an online-based teaching system for either schools or universities, resulting in delays of academic progress.

Higher rates of depression were associated with a longer time spent following COVID-19 related news. Similar associations were noted by other studies (Ni et al., 2020, Gao et al., 2020). Increased exposure to news regarding deaths, casualties, increasing number of cases contributes to this finding. People who spent the same amount of time on social media as they did prior to the pandemic, showed negatively correlating anxiety and depression scores; while increased social media use resulted in higher scores. This highlights the power of social media and emphasizes the need for channeling such tool in raising awareness and promoting education.

There were several limitations to our study. The study was undertaken during the final 5 days of the lockdown period, which is when the psychological impact would be at it's highest. This could have lead to overestimation in the prevalence of the mental health burden, hence the need for longitudinal studies over an extended period of time is essential to minimize bias. Follow-up studies may also be needed to understand the long-term psychological sequelae of the pandemic. The number of participants may not be representative of the entire population; a larger sample size is needed to include more subgroups of our population e.g healthcare workers.

5. Conclusion

The COVID-19 pandemic caused a significant burden on mental health, with some people showing a significantly higher risk than others. The statistically significant relationships discovered are good predictors of anxiety and depression during pandemics, identifying high-risk groups. Psychological support and mental health awareness should be made accessible to all individuals during pandemics. This can be achieved through the use of social media or internet-based interventions. Further studies are required to understand the long-term mental health effects of COVID-19, especially in those infected by the virus.

Informed Consent

Written informed consent was obtained from the responders for the publication of their responses anonymously.

Financial Disclosure

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author statement contributors

Waleed Burhamah (Corresponding author)- study design, administering the questionnaire, literature review, writing of the first draft, final manuscript writing and editing.

Abdulaziz AlKhayyat- literatures review, writing the discussion.

Melinda Oroszlányová- statistical analysis, writing results, review and editing of the manuscript.

Ahmad AlKenane- writing the introduction, review and edit of the final draft.

Abdulrahman Almansouri- literature review, review and edit of the draft.

Mousa Behbehani- review and edit of the draft.

Naser Karimi- review and edit of the draft

Hana Jafar- review and edit of the draft.

Mohammad AlSuwaidan- research supervision, review and edit of the final draft.

Declaration of Competing Interest

No conflict of interest.