Sample characteristics
A total of 83,695 adolescents aged 13–17 years were surveyed. Participants with missing values for any of the variables of interest were excluded from the analysis. This resulted in an analytical sample size of 51,597, representing a total adolescent population of 22,945,384 individuals, after applying the sampling weights. As shown in Table 1, 45% of the participants were boys, and the mean age was 14.8 years.
Country-specific scenarios of depression risk
As shown in Table 1, 26.4% of adolescents had risk of depression. The highest proportion of adolescents with a risk of depression was found in Afghanistan (50%), followed by Namibia (43%), Benin (40%), Kuwait (37%), Bahrain (37%), Yemen (37%), and Timor-Leste (32%). The lowest prevalence of depression risk was reported in Laos (12%) followed by Indonesia (13%) and Bangladesh (18%). Supplementary Fig. 1 shows that depression risk among girls was highest in Afghanistan (59%), followed by Kuwait (44%), Bahrain (43%), Benin (43%), Yemen (42%), the Bahamas (36%), Mongolia (35%), and the Philippines (34%). In the countries included in this study, the risk of depression was reported to be higher among girls than among boys, except in Timor-Leste (boys: 26%; girls: 12%).
Country-specific scenarios of lifestyle and social risk indexes
Figure 1 shows the percentage of lifestyle and social risk indexes among adolescents in each country. The prevalence of unhealthy (moderate and high risk) dietary habits ranged from 77% (the Cook Islands) to 94% (Timor-Leste, Mongolia, Nepal). The highest percentage of physical inactivity was observed in the Philippines (91%) followed by Thailand (87%), Timor-Leste (87%), Tuvalu (86%), Indonesia (85%), and Brunei (85%), whereas the lowest percentage was seen in Bangladesh (48%). The prevalence of unhealthy (moderate and high risk) sedentary behavior varied from 34% (Nepal) to 88% (Kuwait). The prevalence of high-risk unhealthy sedentary behavior (daily sedentary time ≥ 4 hours) was highest in Kuwait (42.0%), followed by Bahrain (37.2%), the Bahamas (34.4%), Thailand (31.1%), and Brunei (27.3%). Approximately 15% of adolescents in Kuwait, Bahrain, and Polynesia used tobacco. In the other countries in this study, less than 10% of adolescents reported using tobacco.
Overall, 27% of the participants were exposed to school violence (moderate and high risk). The prevalence of exposure to school violence (moderate and high risk) varied from 7% (Laos) to as high as 47% (Tuvalu). Around 25% adolescents in Nepal and Tuvalu, and around 20% adolescents in Namibia, Afghanistan, the Philippines, Timor-Leste, Yemen, and Bangladesh reported a high risk of exposure to school violence. Furthermore, 71% of participants had insufficient interactions with parents (moderate and high risk). The prevalence of insufficient interactions with parents (moderate and high risk) was lowest in Nepal (56%) and highest in Timor-Leste (93%). Around 20% adolescents in Bahrain, Kuwait, Timor-Leste, Laos, Tuvalu, Yemen, Brunei, and the Philippines had a high risk of insufficient interactions with parents. 62% of participants had insufficient interactions with friends (moderate and high risk). The highest proportions of adolescents with insufficient interactions with friends (moderate and high risk) were observed in Benin and Laos (82%) and the lowest proportion was observed in Bahrain (37%). Around 17% adolescents in Tuvalu, Namibia, Afghanistan, and Benin reported a high risk of insufficient interactions with friends.
Gender-based lifestyle factors and social risk index values
The prevalence of unhealthy dietary habits was similar for boys and girls, but in most examined countries, girls were 5–20% more physically inactive than boys. The largest gender differences in physical inactivity were observed in Laos (boys: 73% vs. girls: 91%), followed by Brunei (boys: 78% vs. girls: 93%), Polynesia (boys: 69% vs. girls: 84%), the Cook Islands (boys: 69% vs. girls: 84%), and Thailand (boys: 78% vs. girls: 94%). See supplementary table 2, available online, for gender-specific prevalence of unhealthy dietary habits and insufficient physical activity by country. No gender-based differences were found in sedentary behavior, except in Bahrain, where the percentage of unhealthy sedentary behavior was substantially higher among girls than among boys (boys: 77% vs. girls: 91%). The use of tobacco was higher among boys (10%) than that among girls (3%), except in Polynesia, where the percentage of tobacco use was much higher among girls than that among boys (boys: 11% vs. girls: 18%). Significant gender differences in tobacco use were found in Bahrain (boys: 25% vs. girls: 4%), Kuwait (boys: 22% vs. girls: 6%), and Timor-Leste (boys: 18% vs. girls: 4%). See supplementary table 3, available online, for gender-specific prevalence of unhealthy sedentary behavior and tobacco use by country.
The prevalence of exposure to school violence was 5–15% higher among boys than that among girls across countries. The largest gender differences in exposure to school violence were found in Mongolia (boys, 38.0% vs girls, 14.0%), followed by Tuvalu (boys, 62.0% vs girls, 34.0%), Thailand (boys, 30.0% vs girls, 13.0%), Kuwait (boys, 37.0% vs girls, 22.0%), Yemen (boys, 47.0% vs girls, 19.0%), and Bahrain (boys, 42.0% vs girls, 21.0%). The prevalence of insufficient interactions with parents was higher among boys than that among girls. Major gender differences for insufficient interactions with parents were observed in Bangladesh (boys, 67.7% vs girls, 57.2%), Benin (boys, 76.0% vs girls, 69.4%), Fiji (boys, 63.0% vs girls, 55.2%), Mongolia (boys, 78.7% vs girls, 71.6%), and Afghanistan (boys, 55.4% vs girls, 40.7%). There were no gender-specific differences in the prevalence of insufficient interactions with friends. Significant gender differences in insufficient interactions with friends were observed in Bangladesh (boys, 53.0% vs girls, 63.4%), the Cook Islands (boys, 50.3% vs girls, 57.4%), Indonesia (boys, 66.0% vs girls, 57.0%), Tuvalu (boys, 80.0% vs girls, 70.0%), Thailand (boys, 69.4% vs girls, 56.3%), Bahrain (boys, 41.4% vs girls, 32.0%), and Yemen (boys, 71.0% vs girls, 63.0%). See supplementary table 4, available online, for gender-specific prevalence of exposure to school violence and insufficient interactions with parents and friends by country.
Associations between the risk of depression and lifestyle and social risk index values
A multilevel logistic regression analysis was performed to determine the associations between the four lifestyle and three social risk indexes and the risk of depression among adolescents. Total and gender specific ORs with 95% CIs for overweightor obese status in relation to different lifestyle risk scores are presented in Table 4.
Regarding the dietary risk index, the odds of depression risk were 15% (OR = 1.15; 95% CI, 1.07–1.23) higher for participants with high risk index values compared with those with a low dietary risk index. However, for total samples and each subgroup, there were no significant associations between physical inactivity and the risk of depression. Regarding unhealthy sedentary behavior, participants with high risk index values had 53% (OR = 1.53; 95% CI, 1.43–1.63) higher odds of depression risk than those with a low risk score for unhealthy sedentary behavior. The association between unhealthy sedentary behavior and depression risk was stronger among girls (OR = 1.65; 95% CI, 1.51–1.80) than among boys (OR = 1.33; 95% CI, 1.21–1.47). Participants with moderate and high risk scores for tobacco use had 65% (OR = 1.65; 95% CI, 1.46–1.86) and 57% (OR = 1.57; 95% CI, 1.41–1.76) higher odds of depression risk, respectively, compared with those with a low risk score of tobacco use. The association between tobacco use and depression risk was stronger among girls (OR = 2.54; 95% CI, 2.01–3.22) than among boys (OR = 1.44; 95% CI, 1.26–1.64).
Compared with participants with a low risk score of exposure to school violence, those with moderate (OR = 1.79; 95% CI, 1.69–1.90) or high (OR = 3.06; 95% CI, 2.88–3.26) risk scores of exposure to school violence had higher odds of depression risk. Participants with high risk scores for insufficient interactions with parents (OR = 2.09; 95% CI: 1.92–2.28) and friends (OR = 1.82; 95% CI: 1.69–1.96) showed higher odds of depression risk than those with low risk scores of insufficient interactions with parents and friends, respectively. Associations between insufficient interactions with parents and depression risk were stronger among boys (OR = 2.33; 95% CI: 2.05–2.65) than among girls (OR = 1.95; 95% CI: 1.74–2.19), whereas associations between insufficient interactions with friends and depression risk were stronger among girls (OR = 2.24; 95% CI: 2.03–2.48) than among boys (OR = 1.39; 95% CI: 1.25–1.55).