The coronavirus disease 2019 (COVID-19) pandemic forced governments in various countries to take strict protective measures as a step to mitigate the spread of the infection. Such measures included imposing travel restrictions, closing shopping malls, businesses, schools, and universities, and canceling public events. Some countries adopted even more strict measures such as imposing total lockdowns or even curfews.1,2 Unfortunately, with such mandated restrictions, some people experienced significant distress in the form of anxiety, anger, and other symptoms of stress.3,4 Recent evidence has suggested that individuals who are kept in isolation and quarantine during infectious outbreaks experience several stressors, such as fear about their own health or fear of infecting others,5–7 boredom, frustration, and a sense of isolation from the rest of the world resulting from confinement, loss of usual routine, and reduced social and physical contact with others.5,6,8,9

The stress surrounding the COVID-19 pandemic led to sleep disturbances10 which in turn worsened stress levels, giving rise to a dangerous vicious cycle.11 Stress and sleep disturbances can stimulate tobacco cravings and decrease the ability to resist smoking and potentiate smoking intensity.12,13 They can also result in increased alcohol consumption.14 In March 2020, Qatar implemented several preventive measures to contain the spread of COVID-19, such as limiting the number of people going to work and switching them to working from home, closure of educational institutions, shopping malls, gyms, and restaurants, discouraging social gatherings, and strongly emphasizing the importance of home confinement through various media channels.

Easing and lifting of restrictions occurred in four distinct phases.15 Commencing on June 15, 2020, the first phase marked the beginning of this gradual process, which culminated in the fourth phase commencing on September 15, 2020. Regrettably, Qatar encountered delays and was unable to attain phase 4 within the intended timeframe to fully resume normal activities.

To provide further context, this study was conducted between January 2021 and February 2021, during which the majority of restrictions had been relaxed. However, despite these efforts, the nation had not achieved a “normal state” with the complete restoration of operational services as initially planned. In fact, many people continued working from home; restaurants, gyms, and malls were open with very limited capacities; and the government continued emphasizing and encouraging the importance of staying at home and strongly discouraged social gatherings.

This study is part of a bigger national study that assessed the impact of COVID-19 on several aspects of the lifestyle of Qatar’s population. The first part was published previously and assessed the impact on diet, physical activity, and body weight.16 In this study, our goal was to assess the impact of the stress of home confinement during the COVID-19 pandemic on other aspects of lifestyle we referred to as triple “S,” which includes substance use particularly smoking and alcohol use, sleep, and social connectedness among the public in Qatar. To our knowledge, studies exploring these aspects of lifestyle during home confinement or lockdown measures in the Middle East are limited. Our objectives were to explore changes in tobacco and alcohol use, sleep duration and quality, and social connectedness and their associated factors. We believe that evidence derived from this study will guide the implementation of effective lifestyle-related interventions targeting such areas both currently and during any potential future pandemics that mandate periods of movement restriction and home confinement.

METHODS Study Design and Target Population

A national cross-sectional survey was conducted between January 2021 and February 2021. The target population included adults ≥18 years of age who stayed in Qatar for at least 2 months between March and August 2020, which is the time when the State of Qatar announced strict public health measures and emphasized the importance of home confinement.

Study Procedure

We developed a web-based, self-administered survey using SurveyMonkey software. The link to the survey was posted on the social media platforms of Hamad Medical Corporation (eg, Instagram, Facebook, and Twitter), which are generally accessible by the public. The link to the survey was also circulated through emails and WhatsApp groups. Reminders and reposting of the links on social media were done on a regular basis. A letter describing the objectives of the study and assuring the confidentiality and anonymity of the collected data was attached to the survey. After potential participants read the letter attached to the survey, agreement to participate was sought by asking respondents to click a button. Clicking the button then took the participants to the survey questions. Ethical approval was obtained from the Institutional Review Board (IRB) of Hamad Medical Corporation.

Study Questionnaire

The questionnaire was adapted from other validated and reliable questionnaires.17,18 It was initially developed in English and then translated into 3 other languages (Arabic, Malayalam, and Urdu) by an accredited translator. Face and content validities of the questionnaire were assured by experts in the field. The questionnaire consisted of 3 sections. The first section explored the sociodemographic characteristics of the participants (age, sex, nationality, marital status, highest level of education, employment status, and whether they worked from home or not during home confinement, and the presence of chronic diseases). Other sections explored substance use, including status and changes in tobacco and alcohol use and associated underlying factors, and changes in sleep duration, sleep quality, and social connectedness since the start of the restrictive measures.

Outcome Measures

We focused on measuring the perceived changes in tobacco use and factors associated with such changes by asking participants to report whether they perceived an overall increase, decrease, or no change in the amount of tobacco used, to report the average amount used before and during the restrictive measures by selecting an amount category (<10, 10 to 20, 21 to 40, and >40 cigarettes/day) for cigarettes, and (≤ 1, 2 to 3, 4 to 6, and ≥7 times/week) for shisha (molasses-sweetened tobacco smoked in a hookah pipe), and to report the underlying causes for such changes in use. Participants also reported their overall perception of alcohol intake (increased, decreased, or remained the same). Sleep duration was assessed by asking participants to report the average hours of sleep per day before and during the restrictive measures. We assessed sleep quality by asking participants to rate their overall subjective sleep quality on a 5-point Likert scale: 1 (very good), 2 (good), 3 (average), 4 (poor), and 5 (very poor). Participants were also asked to indicate their degree of agreement with 6 statements concerning sleep latency, disturbances, and daytime dysfunction using a 4-point Likert scale: 1 (strongly disagree), 2 (disagree), 3 (agree), and 4 (strongly agree). A total score was calculated by summing the scores on the 6 statements, with higher scores indicating poorer sleep quality with a maximum score of 24. Social connectedness during restrictive measures was assessed using a similar agreement scale using a different set of statements. These statements were: “I feel disconnected from the world around me,” “I feel so distant from people,” and “I catch myself losing all sense of connectedness with society.”

Statistical Analysis

Data analysis was performed using the Statistical Product and Service Solutions (SPSS) Version 26.0 (IBM Corp, Armonk, NY). Descriptive statistics were presented as frequencies and percentages for categorical variables. The χ2 test was used to assess differences in the overall perceived changes in tobacco and alcohol use between different groups. After testing for normality using the Shapiro-Wilk test, we used the nonparametric Wilcoxon Signed Rank test to detect differences in the amount of tobacco use, sleep duration, and subjective sleep quality before and during the restrictive measures. Rank biserial correlation was calculated to measure the effect size for these comparisons (small 0.10 to <0.30, medium 0.30 to <0.50, and large ≥0.50). The Mann-Whitney U test and the Kruskal-Wallis test were applied to compare the scores for sleep quality and social connectedness between groups, as the scores were not normally distributed. Multivariable logistic regression analysis was conducted to explore predictors of adverse changes in outcome variables. The Hosmer-Lemeshow test was used to assess the goodness of fit of the model. P-values <0.05 were considered statistically significant.

RESULTS Sociodemographic Characteristics

As shown in Table 1, the survey was completed by 1408 participants. Over half (825, 58.6%) of the questionnaires were completed in English, followed by 23.2%, 17.3%, and 0.9% in Malayalam, Arabic, and Urdu, respectively. Males accounted for 58.8% of the total sample, with a male-to-female ratio of 1.4:1. The majority of participants (n=1084, 77%) were 25 to 44 years old, 1132 (80.4%) were married, and 1107 (78.6%) had completed a college degree or higher. Over 50 nationalities were reported, with the top 3 being Indians (53.4%), Filipino (5.2%), and Qatari (4.3%). Of the total participants, 1070 (76%) were employed, with 53.1% having shifted to work from home during the restrictive measures. About one-fifth (306, 21.7%) had a history of one or more chronic diseases.

TABLE 1 - Sociodemographic Profiles and Background Information of the Participants (N=1408) Variables N (%) Age, y  18-24 57 (4.0)  25-34 540 (38.4)  35-44 544 (38.6)  45-54 203 (14.4)  55-64 59 (4.2)  65+ 5 (0.4) Sex  Male 828 (58.8)  Female 580 (41.2) Nationality (classification by regions)*  Americas 45 (3.2)  Sub-Saharan Africa 54 (3.8)  Europe 117 (8.3)  Middle East—North Africa 273 (19.4)  Asia—Pacific 919 (65.3) Highest level of education  No formal education 18 (1.3)  High school diploma 228 (16.2)  College or higher 1107 (78.6)  Vocational training 55 (3.9) Marital status  Married 1132 (80.4)  Not married 276 (19.6) Presence of chronic disease/s†  Yes 306 (21.7)  No 1102 (78.3) Employment related information Employment status  Employed 1070 (76.0)  Not employed 338 (24.0) Nature of work‡  Mostly office work 746 (69.7)  Mostly field work 324 (30.3) Working from home as part of “staying at home” measures‡  Yes 568 (53.1)  No 502 (46.9)

*More than 50 different nationalities were reported.

†Most commonly reported chronic diseases were: diabetes, hypertension, asthma, and cardiovascular diseases.

‡Denominator is the number of employed participants (n=1070).

Substance Use During Restrictive Measures

Of the 1408 participants, 200 (14.2%) reported regular use of tobacco products (including smokeless tobacco). Of these, 72 (36%) perceived an increase in tobacco use, 53 (26.5%) perceived a decrease, and the remaining perceived no change in their use since the start of the restrictive measures. In assessing the change in the amount of tobacco used before and during the restrictive measures, about half (48.6%) of the 72 who perceived an increase in their use, reported a significant increase in the number of cigarettes (jumped into a higher amount category compared with before the restrictive measures), with P=0.035 but with a small effect size (r=0.28). While 22.2% reported an increase in shisha use (jumped into a higher amount category), the reported increase in shisha use was not significant P=0.843. On the other hand, among those who reported decreasing their smoking, one-third (32.1%) reported decreasing their cigarette amount to a lower category, and 22.6% reported decreasing their shisha use. For the remaining participants who perceived an increase or a decrease in their tobacco use, the change was either within the same amount category or in the use of other tobacco products. We detected a significant difference in reports of increased use between males and females (P=0.008), between those who perceived more stress during restrictive measures and those who did not (P=0.002), and between different nationalities (P=0.002). However, multivariable logistic regression showed that perceived stress was the only predictor for increased tobacco use during the restrictive measures [adjusted odds ratio (OR): 2.88, 95% CI: 1.32-6.31, P=0.008] (Table 2). Participants attributed the change in tobacco use to different factors. Many related their increase in use to feelings of boredom during restrictive measures (73.6%), worries about their employment and financial status during the pandemic (62.5%), and increased arguments with other family members during the restrictive measures (50%). Few reported stress about the current pandemic, homeschooling of children, and separation from family as triggers for their increase in tobacco use. Among those who decreased their tobacco use, 45.3% attributed this change to decreased social gatherings and less time spent with friends who smoke, 43.4% to fear of catching COVID-19 infection as smoking increases the risk, and 32.1% to pressure by family members to quit smoking. Few related their decrease in tobacco use to limited accessibility to tobacco products, and limited ability to buy such products in light of the financial consequences of COVID-19.

TABLE 2 - Differences in the Use of Tobacco Products During COVID-19-related Home Confinement Measures Among Different Sociodemographic Subgroups (N=200) Tobacco use during home confinement measures Multivariable logistic regression Variable No increase, N (%) Increase, N (%) P* AOR† (95%CI) P Age, y  18-24 1 (33.3) 2 (66.7) 0.770 1 [reference]  25-34 53 (64.6) 29 (35.4) 0.24 (0.01-3.97) 0.318  35-44 52 (65.0) 28 (35.0) 0.23 (0.1-3.88) 0.309  45-54 18 (66.7) 9 (33.3) 0.19 (0.01-3.63) 0.269  55+‡ 4 (50.0) 4 (50.0) 0.43 (0.02-9.67) 0.597 Sex  Male 109 (68.6) 50 (31.4) 0.008 0.62 (0.25-1.58) 0.320  Female 19 (46.3) 22 (53.7) 1 [reference] Nationality (classification by regions)  Americas 3 (30.0) 7 (70.0) 0.002 5.03 (0.98-25.78) 0.053  Sub-Saharan Africa 4 (57.1) 3 (42.9) 1.88 (0.33-1.68) 0.475  Europe 15 (55.6) 12 (44.4) 2.09 (0.68-6.43) 0.200  Middle East—North Africa 32 (53.3) 28 (46.7) 2.16 (0.99-4.70) 0.051  Asia—Pacific 74 (77.1) 22 (22.9) 1 [reference] Highest level of education  No formal education 1 (50.0) 1 (50.0) 0.428 0.54 (0.22-13.04) 0.703  High school diploma 25 (67.6) 12 (32.4) 0.31 (0.63-1.55) 0.154  College or higher 98 (64.9) 53 (35.1) 0.30 (0.07-1.27) 0.102  Vocational training 4 (40.0) 6 (60.0) 1 [reference] Employment status during confinement measures  Employed   Switched to working from home 52 (65.8) 27 (34.2) 0.139 0.82 (0.30-2.27) 0.706   Continued working regularly 61 (67.8) 29 (32.2) 0.82 (0.28-2.40) 0.711  Not employed 15 (48.4) 16 (51.6) 1 [reference] Marital status  Married 102 (65.4) 54 (34.6) 0.442 0.97 (0.43-2.17) 0.934  Not married 26 (59.1) 18 (40.9) 1 [reference] Presence of chronic disease/s  Yes 30 (68.2) 14 (31.8) 0.513 0.94 (0.40-2.22) 0.890  No 98 (62.8) 58 (37.2) 1 [reference] Perceived stress  Yes 80 (57.1) 60 (42.9) 0.002 2.88 (1.32-6.31) 0.008  No 48 (80.0) 12 (20.0) 1 [reference]

P values in bold indicate statistically significant results.

*Using χ2 test or Fisher exact test.

†Adjusted for all variables in the table.

‡We combined 55 to 64 and 65+ categories into one category 55+, because of the very low number of smokers in the 65+ age group.

AOR indicates adjusted odds ratio.

Of the 1408 participants, 255 (18.1%) reported regular use of alcohol. Of these, 105 (41.2%) reported decreased alcohol use during the restrictive measures, while 14.5% reported an increase. Males were significantly more likely to report a decrease in their alcohol use compared with females (P=0.022). Multivariable logistic regression analysis showed that sex, nationality, and marital status were predictors of decreased use of alcohol during the restrictive measures (Table 3).

TABLE 3 - Differences in the Use of Alcohol During COVID-19-related Home Confinement Measures Among Different Sociodemographic Subgroups (N=255) Alcohol use during home confinement measures Multivariable logistic regression Variables No decrease, N (%) Decrease, N (%) P* AOR† (95% CI) P Age  18-24 1 (33.3) 2 (66.7) 0.223 1 [reference]  25-34 36 (53.7) 31 (46.3) 1.22 (0.25-5.98) 0.803  35-44 63 (55.3) 51 (44.7) 2.04 (0.4-10.15) 0.383  45-54 36 (70.6) 15 (29.4) 1.20 (0.22-6.56) 0.831  55+‡ 14 (70.0) 6 (30.0) 1.81 (0.30-11.11) 0.521 Sex  Male 86 (53.4) 75 (46.6) 0.022 1.95 (1.15-3.30) 0.013  Female 64 (68.1) 30 (31.9) 1 [reference] Nationality (classification by regions)  Americas 15 (68.2) 7 (31.8) 0.102 2.61 (1.04-6.55) 0.040  Sub-Saharan Africa 4 (50.0) 4 (50.0) 1.05 (0.35-3.10) 0.933  Europe 45 (71.4) 18 (28.6) 2.50 (1.32-4.74) 0.005  Middle East—North Africa 9 (60.0) 6 (40.0) 0.27 (0.11-0.63) 0.003  Asia—Pacific 77 (52.4) 70 (47.6) 1 [reference] Highest degree of education  No formal education 2 (66.7) 1 (33.3) 0.708 0.76 (0.8-7.15) 0.809  High school diploma 11 (47.8) 12 (52.2) 0.64 (0.21-1.92) 0.423  College or higher 131 (60.1) 87 (39.9) 1.012 (0.39-2.66) 0.981  Vocational training 6 (54.5) 5 (45.5) 1 [reference] Employment status during confinement measures  Employed   Switched to working from home 77 (60.2) 51 (39.8) 0.036 1.57 (0.76-3.22) 0.220   Continued working regularly 41 (49.4) 42 (50.6) 1.65 (0.77-3.53) 0.198  Not employed 32 (72.7) 12 (27.3) 1 [reference] Marital status  Married 123 (61.2) 78 (38.8) 0.138 0.52 (0.31-0.89) 0.016  Not married 27 (50.0) 27 (50.0) 1 [reference] Presence of chronic disease/s  Yes 35 (54.7) 29 (45.3) 0.437 1.54 (0.95-2.52) 0.082  No 115 (60.2) 76 (39.8) 1 [reference] Perceived stress  Yes 100 (61.0) 64 (39.0) 0.349 1.20 (0.79-1.84) 0.394  No 50 (54.9) 41 (45.1) 1 [reference]

P values in bold indicate statistically significant results.

*Using χ2 test or Fisher exact test.

†Adjusted for all variables in the table.

‡We combined 55 to 64 and 65+ categories into one category 55+, because of the very low number of alcohol users in the 65+ age group.

AOR indicates adjusted odds ratio.

Sleep and Social Connectedness During Restrictive Measures

Of the 1408 participants, 649 (46.1%) reported an increase in average sleep duration per day since the start of the restrictive measures and 218 (15.5%) reported a decrease. The mean sleep duration per day significantly increased from 6.95 hours/day before to 7.72 hours/day during the restrictive measures (mean increase of 0.77 h, 95% CI: 0.66-0.88, P<0.001) with a large effect size (r=0.54). With regard to overall subjective sleep quality, 416 (29.5%) reported a significant decrease in sleep quality since the start of the restrictive measures (P<0.001) with a small effect size (r=0.21). Disturbance in the sleep-wake cycle, with more daytime than nighttime sleep, was reported by 38.1% of the participants. Indeed, 18.9% even reported daytime dysfunction and having trouble staying awake while driving, eating meals, or engaging in social activities. Participants also reported an increase in sleep onset latency, with difficulty falling asleep (44.3%). Group analysis with pairwise comparisons showed higher percentages of participants with poorer sleep quality (higher sleep quality scores) among those aged 18 to 34 years of age compared with those 35 to 55 years of age, among nationalities of Middle Eastern-North African origin compared with those of Asia-Pacific origin, and among those not employed, unmarried, and those experiencing more stress during the pandemic compared with the others. The last 2 categories were also more likely to include higher percentage