Table 2. Table of evidence .
|
Author (year)
Country Study design |
Age, years [range; mean (SD)] | Sample size |
Gender
(% female) |
Race/Ethnicity | Depressive symptoms measurement tool | Dietary intake and diet quality measurement tool | Major findings (diet-depression relationship) |
|---|---|---|---|---|---|---|---|
| Xu et al (2020)1 China Cross-sectional |
10-20 14.9 (1.8) |
n = 14 500 | 49.3% | NR | Self-report: CDI | Discretionary food intake: FFQ (self-report) | Consumption of fast foods and sugar-sweetened beverages was significantly associated with depressive symptoms in Chinese adolescents. |
| Gonoodi et al (2018)15 Iran Cross-sectional |
12-18 15.2 (1.5) |
n = 408 | 100% | NR | Self-report: BDI | Micronutrient intake: three-day food record (self-report) | Intake of dietary zinc was significantly higher in individuals without or with minimal depressive symptoms when compared to those with mild to severe depressive symptoms. Dietary zinc intake, but not serum zinc levels, was inversely associated with depressive symptoms. |
| Mrug et al (2021)23 USA Cross-sectional |
11.1 at wave 1 13.1 at wave 2 16.1 at wave 3 |
Wave 1 n = 5147 Wave 2 n = 4773 Wave 3 n = 4521 |
51% | Black (34%) Hispanic (35%) Other minority (6%) White (24%) |
Self-report: Major depressive disorder subscale from DPS | Food group and discretionary food intake: dietitian interview | No significant association between soft drink consumption and depressive symptoms among adolescents. More frequent intake of soft drinks at age 13 was associated with fewer depressive symptoms at age 16. |
| Liu et al (2020)24 21 Low- and middle-income countries Cross-sectional (secondary data analysis) |
12-15 | n = 65 267 | 40.5% - 57.9% | NR | Self-report: GSHS | Food group intake: GSHS(self-report) | Inadequate intake of fruits and vegetables was significantly associated with a higher risk of depressive symptoms in Seychelles, Ecuador, Jordan, and Kenya, but only among males. |
| Khayyatzadeh et al (2020)25 Iran Cross-sectional |
12-18 14.5 (1.52) |
n = 988 | 100% | NR | Self-report: BDI | Micronutrient and macronutrient intake: FFQ(self-report) | Higher intakes of α-carotene, β-carotene, lutein, and vitamin C were significantly associated with a lower risk of depressive symptoms. Both soluble and insoluble dietary fiber intake was significantly higher in adolescents without depressive symptoms than those with depressive symptoms. No significant relationship between dietary intake of vitamin A and vitamin E and depressive symptoms. |
| Bahrami et al (2019)26 Iran Cross-sectional |
12-18 14.5 (1.5) |
n = 563 | 100% | NR | Self-report: BDI | Serum micronutrient levels | No significant differences between serum levels of vitamin A and E and depressive symptoms. |
| Khayyatzadeh et al (2019)27 Iran Cross-sectional |
10-14 14.5 (1.5) |
n = 750 | 100% | NR | Self-report: BDI-II | Food group intake: FFQ (self-report) | No significant associations between a traditional dietary pattern (high intake of potatoes, snacks, hydrogenated fats, vegetable oil, sugar, soft drinks, sweets, desserts, tea, salt, and spices) and a Western dietary pattern (high intake of refined grains, snacks, red meats, poultry, fish, organ meats, pizza, fruit juices, industrial juices and compotes, mayonnaise, nuts, soft drinks, sweets and desserts, coffee, and pickles) and depressive symptoms. |
| Tanaka and Hashimoto (2019)28 Japan Cross-sectional |
Junior high school; 14 (0.86) Senior high school; 17.1 (0.88) |
Junior high school n = 441 Senior high school n = 417 |
54.7% | NR | Self-report: CES-D | Food group intake: self-report questionnaire (breakfast and dietary intake) | Consumption of green and yellow vegetables once or more times per day was significantly inversely related to depressive symptoms among adolescents. |
| Ferrer-Cascales et al (2018)29 Spain Cross-sectional |
12-17 14.3 (1.52) |
n = 527 | 54.5% | NR | Self-report: CES-D | Diet quality: KIDMED (self-report) | Breakfast skippers had significantly lower levels of depression than those who ate a poor or very poor-quality breakfast (e.g., consuming commercially baked food). Adolescents who consumed a high-quality breakfast (e.g., bread, toast, cereal, and dairy products) had significantly lower levels of depression than those who consumed a poor-quality breakfast. |
| Khayyatzadeh et al (2018)30 Iran Cross-sectional |
12-18 14.8 (1.5) |
n = 535 | 100% | NR | Self-report: BDI | Micronutrient and food group intake: dietitian interview | High adherence to a DASH-style diet, including a higher intake of carbohydrates, dietary fiber, vitamin A, vitamin C, vitamin D, folate, calcium, magnesium, and potassium, was significantly associated with a lower risk of having depressive symptoms. |
| Neshatbini Tehrani et al (2018)31 Iran Cross-sectional |
15-18 16.2 (0.97) |
n = 263 | 100% | NR | Self-report: DASS-21 | Food group intake: FFQ (self-report) | Higher adherence to the Mediterranean dietary pattern was significantly negatively associated with depressive symptoms. |
| Yu et al (2018)32 China Cross-sectional |
9-11 9.8 (0.7) |
n = 188 | 22.3% | NR | Self-report: DSRSC | Food group intake: self-administered questionnaire (derived from the Global School-Based Student Health Survey) | Adequate intake of fruits and vegetables was associated with a significant reduction in the risk of depressive symptoms. |
| Singh et al (2017)33 USA Cross-sectional |
14-19 18 (1.2) |
n = 114 | 100% | NR | Self-report: RADS |
Micronutrient and macronutrient intake: ASA (self-report) | Higher intake of carbohydrates and fat was significantly associated with more depressive symptoms. A significant negative association was found between magnesium intake and depressive symptoms when controlling for total energy intake. |
| Sinclair et al (2016)34 Fiji Cross-sectional |
13-18 15.6 (1.4) at baseline 17.4 (0.9) at follow-up |
Baseline n = 7237 Follow- up n = 2948 |
Baseline 52.6% Follow- up 56% |
Indigenous Fijian (42.5%) Indo-Fijian (52.4%) Other (5.1%) |
Self-report: PedsQLTM 4.0 | Diet quality: ABAKQ | A higher diet quality, including fruits available at home, daily servings of fruits and vegetables, eating fruits after school, and consuming fruit drinks on school days, was significantly associated with lower scores of depressive symptoms at ages 15.6 and 17.4. |
| Rubio-López et al (2016)35 Spain Cross-sectional |
6-9 8.21 (1.32) |
n = 710 | 49.73% | NR | Self-report: CES-D | Micronutrient and macronutrient intake: 3-day food records (parent report) | Significant inverse relationships were found between higher dietary intakes of protein, carbohydrates, pantothenic acid, biotin, vitamin B12, vitamin E, zinc, manganese, cobalt, aluminum, thiamin, vitamin K, vitamin C, magnesium, iron, and bromine and depressive symptoms in children. A higher intake of fiber was significantly positively associated with depressive symptoms in children. |
| Richards and Smith (2015)36 UK Cross-sectional |
11-17 13.6 (1.49) |
n = 2307 | 51.5% | White (97.2%) | Self-report: Wellbeing Process Questionnaire | Food group and discretionary food intake: DABS (self-report) | A significant inverse relationship between caffeine intake and depressive symptoms in females but not in males. |
| Modan-Moses et al (2015)37 Israel Cross-sectional |
14-18 16 (2) |
n = 87 | 93.1% | NR | Clinical interview for DSM-IV: BDI | Micronutrient intake: 72-hour dietary recall (dietitian report) Serum micronutrient levels |
No significant differences in vitamin D levels of patients with and without eating disorders and depressive symptoms. |
| Smith et al (2014)38 USA Cross-sectional |
7-17 12.1 (3.1) |
n = 38 | 52.6% | White (94.7%) | Self-report: CDI | Serum micronutrient levels | Lower serum vitamin D levels were significantly associated with increased depressive symptoms in youth with cystic fibrosis. |
| Weng et al (2012)39 China Cross-sectional |
11-16 13.21 (0.99) |
n = 5003 | 47.9% | NR | Self-report: DSRS | Food group and discretionary food intake: FFQ (self-report) | High intake of animal foods and snacks was significantly associated with depressive symptoms (without anxiety) among adolescents. |
| Jacka et al (2010)40 Australia Cross-sectional |
10-14 11.6 (0.81) |
n = 7114 | 57.3% | NR | Self-report: SMFQ | Diet quality: 14-item dietary questionnaire (dietitian graded) | A significant inverse relationship between a healthy diet score (i.e., intake of breakfast every day, low-fat dairy food intake, consumption of fruits and vegetables every day) and depressive symptoms was shown both before and after adjustment for confounding variables, such as age, gender, physical activity, and fathers' employment status. |
| Murakami et al (2010)41 Japan Cross-sectional |
12-15 12 (6.5) |
n = 3067 | 68.8% | NR | Self-report and parent report: CES-D | Micronutrient intake: BDHQ (self-report and parent report) | Higher intake of dietary B vitamins, especially folate and vitamin B6, was associated with a lower risk of depressive symptoms. |
| Murakami et al (2010)42 Japan Cross-sectional |
12-15 Boys 12 + 6.5 Girls 13.5 + 8 |
n = 6517 | 52.9% | NR | Self-report: CES-D | Macronutrient and food group intake: BDHQ (self-report and parent report) | A higher intake of fish, EPA, and DHA was significantly and independently associated with a lower prevalence of depressive symptoms among males but not females. |
| Swann et al (2021)43 Australia Longitudinal |
Baseline 14 Follow up 17 |
Baseline n = 1260 Follow- up n = 653 |
48% at 14 54% at 17 |
NR | Self-report: BDI-Y | Macronutrient intake: FFQ (self-report) | A high intake of dietary fiber was significantly inversely associated with depressive symptoms among adolescents. |
| Oddy et al (2018)44 Australia Longitudinal |
Baseline 14 Follow up 17 |
Baseline n = 843 Baseline n = 838 |
51% | White (88%) | Self-report: BDI-Y | Food group and discretionary food intake: FFQ (parent report) | A high intake of a Western-style diet, which included red meat, take-out foods, refined foods, and confectionary foods at age 14 was significantly inversely associated with depressive symptoms at age 17. |
| Winpenny et al (2018)45 UK Longitudinal (Cohort study) |
Baseline 14.5 (3.5 months) Follow up 17.5 (4.1 months) |
Baseline n = 1238 Follow- up n = 932 |
60% | NR | Self-report: MFQ | Diet quality and food group intake: MDS and four-day diet diary (self-report) | No significant relationship between diet quality, intake of fruits and vegetables, intake of fish at age 14 years, and depressive symptoms at age 17 years. No significant cross-sectional association between diet quality or intake of fish and depressive symptoms. |
| Black et al (2015)46 Australia Longitudinal |
Baseline 14 Follow up 17 |
Baseline n = 667 Follow-up n = 607 |
53.4% | NR | Self-report: YSR | Micronutrient intake: FFQ (self-report and parent report) | A significant inverse association was found between magnesium intake and depressive symptoms. No significant relationship between dietary intake of zinc and depressive symptoms. |
| Tolppanen et al (2012)47 USA Longitudinal (Cohort study) |
Baseline 10.6 (NA) Follow up 13.8 (NA) |
Baseline n = 2759 Follow-up n = 2752 |
NR | NR | Self-report: MFQ | Serum micronutrient levels | Higher serum levels of vitamin D3 were associated with a lower risk of depressive symptoms at age 13.8 but not at age 10.6. No significant association between vitamin D2 and depressive symptoms at either age 10.6 or 13.8. |
| McMartin et al (2012)48 Canada Longitudinal |
10-11 | n = 3757 | 52% | NR | Self-report: ICD-9 and ICD-10 | Diet quality: FFQ (self-report) and DQI-I | Diet quality was not significantly associated with depressive symptoms. Greater dietary variety was associated with a significantly lower rate of receiving a diagnosis of depression. No significant association between dietary intake of vegetables, fruits, vitamin B6, folate, vitamin B12, and n-3 fatty acids and depressive symptoms. |
| Oddy et al (2011)49 Australia Longitudinal |
Baseline 14 Follow up 17 |
Baseline n = 1376 Follow-up n = 977 |
50.9% | NR | Self-report: BDI-Y | Macronutrient intake: FFQ(parent report) | A significant negative association between dietary intake of EPA, VLC-PUFA, n-6 dihomogamma-linoleic acid (dGLA), AA, and adrenic acid and depressive symptoms at age 14. Significant inverse correlation between dietary intake of ALA, EPA, VLC-PUFA, total omega-3 polyunsaturated fatty acid (total n-3 PUFA), n-6 eicosadienoic acid, dGLA, AA, and adrenic acid and depressive symptoms at age 17. Dietary intake of total saturated and monounsaturated fat was significantly inversely associated with depressive symptoms at both ages 14 and 17. |
| Kim et al (2015)50 Korea Case-control |
12-18 15 (1.5) |
n = 116 | 100% | NR | Self-report: K-BDI | Micronutrient and food group intake: FFQ (self-report) | Higher consumption of fast foods including ramen noodles, hamburger, pizza, and fried foods and processed foods including ham, fish paste, and snacks were significantly associated with a higher risk of depressive symptoms. Dietary intakes of green vegetables and 1 to 3 servings of fruits per day were significantly associated with decreased risk of depression. Depressive symptoms were significantly negatively associated with intakes of fiber, b-carotene, vitamin B6, vitamin E, vitamin C, potassium, zinc, folate, iron, and copper. |
| Tsuchimine et al (2015)51 Japan Case-control |
11-19 16 (2.2) |
n = 24 | 100% | NR | Self-report: BDI-II, DSRSC | Serum micronutrient levels | A significant inverse relationship between serum levels of AA, DHA, and folate and depressive symptoms. |
| Benko et al (2011)52 Brazil Case-control |
9-12 9.9 (1.0) |
n = 51 | 11.8% | NR | Self-report: CDI | Food group and discretionary food intake: NBI (self-report) | Caffeine intake is significantly associated with a higher risk of having depressive symptoms after adjusting for sugar intake. |
Abbreviation: NR, not reported; DPS, Diagnostic Interview Schedule for Children Predictive Scales; GSHS, Global School-based Health Survey; BDI, Beck Depression Inventory; FFQ, Food Frequency Questionnaires; CDI, Children’s Depression Inventory; BDI-II, Beck Depression Inventory, second edition; CES-D, Center for Epidemiologic Studies Depression scale; KIDMED, Mediterranean Diet Quality Index for children and teenagers; DASS-21, Depression, Anxiety, Stress Scale-2; DSRSC, Depression Self Rating Scale for Children; RADS, Reynolds Adolescent Depression Scales; ASA24, Automated Self-Administered 24-hour Dietary Recall; PedsQLTM, Pediatric Quality of life Inventory 4.0 Generic Core Scales; ABAKQ, Adolescent Behaviors, Attitudes and Knowledge Questionnaire; DABS, Diet and Behaviour Scale; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, fourth edition; DSRS, Depression Self Rating Scale; SMFQ, Short Mood and Feelings Questionnaire; BDHQ, Brief Self-administered Diet History Questionnaire; BDI-Y, Beck Depression Inventory, youth; MFQ, Mood and Feelings Questionnaire; MDS, Mediterranean Diet Score; YSR, Youth Self-Report; ICD-9, International Classification of Diseases, ninth edition; ICD-10, International Classification of Diseases, 10th edition; DQI-I, Diet Quality Index–International; K-BDI, Korean version of the Beck Depression Inventory; NBI, Nutrition-Behavior-Inventor; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; AA, arachidonic acid; VLC-PUFA, very long-chain omega-3 fatty acids; ALA, alpha-lipoic acid; DHA, docosahexanoic acid.