Nutrition Reviews

Iguacel,, Isabel;Huybrechts,, Inge;Moreno, Luis, A;Michels,, Nathalie

doi: 10.1093/nutrit/nuaa030pmid: 32483598

Abstract Context Vegetarian and vegan diets are increasing in popularity. Although they provide beneficial health effects, they may also lead to nutritional deficiencies. Cognitive impairment and mental health disorders have a high economic burden. Objective A meta-analysis was conducted to examine the relationship between vegan or vegetarian diets and cognitive and mental health. Data Sources PubMed, Scopus, ScienceDirect, and Proquest databases were examined from inception to July 2018. Study Selection Original observational or interventional human studies of vegan/vegetarian diets were selected independently by 2 authors. Data Extraction Raw means and standard deviations were used as continuous outcomes, while numbers of events were used as categorical outcomes. Results Of 1249 publications identified, 13 were included, with 17 809 individuals in total. No significant association was found between diet and the continuous depression score, stress, well-being, or cognitive impairment. Vegans/vegetarians were at increased risk for depression (odds ratio = 2.142; 95%CI, 1.105–4.148) and had lower anxiety scores (mean difference = −0.847; 95%CI, −1.677 to −0.018). Heterogeneity was large, and thus subgroup analyses showed numerous differences. Conclusions Vegan or vegetarian diets were related to a higher risk of depression and lower anxiety scores, but no differences for other outcomes were found. Subgroup analyses of anxiety showed a higher risk of anxiety, mainly in participants under 26 years of age and in studies with a higher quality. More studies with better overall quality are needed to make clear positive or negative associations. Systematic Review Registration PROSPERO registration no. CRD42018097204 anxiety, cognition, dementia, depression, memory, mental health, stress, vegan, vegetarian diet, well-being INTRODUCTION Cognitive and mental disorders are critical public health issues.1,2 Dementia is a worldwide problem with a current prevalence of 47.5 million people, a number predicted to double every 20 years.3 Depression affects more than 300 million people and is linked to cognitive dysfunction and a higher risk of anxiety, stress, and many other mental health issues. Poor mental health status not only negatively affects a person’s emotional and physical health but also leads to reduced productivity as a result of absenteeism or unemployment, which in turn can result in loss of income. Mental health disorders account for a worldwide economic burden of more than $1 trillion per year, with $818 billion in costs attributed to dementia alone.2 Diet has been hypothesized to play an important role not only in physical health but also in cognitive and mental health.4 The Mediterranean diet, for example, can reduce and even prevent cardiovascular disease, breast cancer, depression, and cognitive decline.5,6 Vegetarian and vegan diets have also been widely associated with improved physical health outcomes, including a lower incidence and mortality of ischemic heart disease and lower incidence of total cancer.7 These improved outcomes have been attributed to the high content of fiber, folic acid, vitamins C and E, potassium, magnesium, phytochemicals, and unsaturated fat in the Mediterranean diet.8 Nevertheless, vegetarian and vegan diets, especially the latter, may be deficient in vitamin B12, creatine, and omega-3 fatty acids.9–11 Deficiencies of these nutrients have been associated with neurodegenerative disease, cognitive impairment, and poor mental health.12–14 Furthermore, vegetarian diets may lead to deficiencies of iron and zinc, a result of both the absence of iron- and zinc-containing flesh foods and a higher content of absorption inhibitors, such as phytate and polyphenols, which reduce the bioavailability of these nutrients.15 Such deficiencies might lead to an increased risk of mental health disorders in vegetarians and vegans.16 Results in the literature are conflicting,17–29 with some investigations showing positive associations of vegetarian and vegan diets with different mental health and cognitive outcomes17,19–21 and others showing an inverse association.18,25,29,30 Equivocal results in studies can be partially due to the following: different definitions of vegetarian and vegan diets (eg, some studies define vegetarian diets as those including the consumption of fish or chicken); the particular characteristics of groups defined to have a healthier lifestyle (ie, Seventh Day Adventists)31; or variations in the length of time a vegetarian or vegan diet is consumed. Although the number of vegetarians and vegans worldwide is still low (except in India, where around one-third of the population is vegetarian),32 most recent surveys have shown an increasing number of vegetarians and vegans, mostly in high-income countries. In fact, the percentage of vegetarians or vegans represents more than 10% of the total population in countries such as Australia, New Zealand, Israel, and Sweden.33 In view of the increasing number of people adhering to vegetarian/vegan diets and the conflicting literature on related mental and cognitive outcomes, this systematic review and meta-analysis was conducted to investigate the possible associations of vegetarianism or veganism with mental and cognitive outcomes. METHODS Search strategy The present systematic review and meta-analysis was registered in the PROSPERO database (ID number: CRD42018097204) and followed the systematic review methodology proposed in the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) statement (see Appendix S1 in the Supporting Information online).34,35 A specific question was constructed according to the PICOS (Participants, Interventions, Control, Outcomes, Study Design) principle (Table 1).36 Table 1 PICOS criteria for inclusion and exclusion of studies Parameter . . Participants Children, adolescents, or adults of both sexes and all nationalities, without chronic diseases Interventions Vegetarianism or veganism Control/comparator group Omnivores Outcomes Mental health (depression, anxiety, stress, neuropsychiatric disorders, psychological health, and well-being) Cognitive health (attention, memory, orientation, executive functions, social cognition) Study design Observational studies with a comparison group (cohort studies, cross-sectional studies, case-control studies), randomized controlled trials, and nonrandomized controlled trials Parameter . . Participants Children, adolescents, or adults of both sexes and all nationalities, without chronic diseases Interventions Vegetarianism or veganism Control/comparator group Omnivores Outcomes Mental health (depression, anxiety, stress, neuropsychiatric disorders, psychological health, and well-being) Cognitive health (attention, memory, orientation, executive functions, social cognition) Study design Observational studies with a comparison group (cohort studies, cross-sectional studies, case-control studies), randomized controlled trials, and nonrandomized controlled trials Open in new tab Table 1 PICOS criteria for inclusion and exclusion of studies Parameter . . Participants Children, adolescents, or adults of both sexes and all nationalities, without chronic diseases Interventions Vegetarianism or veganism Control/comparator group Omnivores Outcomes Mental health (depression, anxiety, stress, neuropsychiatric disorders, psychological health, and well-being) Cognitive health (attention, memory, orientation, executive functions, social cognition) Study design Observational studies with a comparison group (cohort studies, cross-sectional studies, case-control studies), randomized controlled trials, and nonrandomized controlled trials Parameter . . Participants Children, adolescents, or adults of both sexes and all nationalities, without chronic diseases Interventions Vegetarianism or veganism Control/comparator group Omnivores Outcomes Mental health (depression, anxiety, stress, neuropsychiatric disorders, psychological health, and well-being) Cognitive health (attention, memory, orientation, executive functions, social cognition) Study design Observational studies with a comparison group (cohort studies, cross-sectional studies, case-control studies), randomized controlled trials, and nonrandomized controlled trials Open in new tab A systematic search of the literature was carried out using the PubMed, Scopus, ScienceDirect, and Proquest databases (from database inception to July 2018). When possible, the search included a vocabulary thesaurus (list of MeSH terms in PubMed). First, the diet terms were combined as follows: “vegetarian” OR “vegan” OR “vegetarians” OR “diet, vegetarian” OR “diet, vegan.” Secondly, the mental and cognitive outcome terms were combined as follows: “cognition” OR “cognitive” OR “depression” OR “executive function” OR “anxiety” OR “memory” OR “mental health” OR “psychological stress” OR “emotion.” Finally, both the diet and the mental and cognitive outcome terms were combined with “AND.” In Scopus, ScienceDirect, and Proquest, these terms had to appear in the title, abstract, or keywords. The filters “humans,” “articles,” and “in English, Spanish, French, Italian, and Portuguese” were applied when possible. Two authors (I.I and N.M) independently searched each database to obtain publications. Agreement between the authors was found for 90% of the publications, while remaining discrepancies were resolved by discussion. Relevant articles were obtained in full and assessed against the inclusion and exclusion criteria. Inclusion criteria The following inclusion criteria were applied: (1) original studies; (2) studies performed in humans; (3) studies written in English, Spanish, French, Italian, or Portuguese; (4) studies including vegetarian (lacto-ovo-vegetarian, ovo-vegetarian, or lacto-vegetarian) or vegan diets as exposures; and (5) studies including raw data on mental or cognitive outcomes (ie, means and standard deviations for continuous variables and cases and events for categorical outcomes). In the present study, vegetarians and, in particular, lacto-ovo-vegetarians were defined as those who excluded meat, fish, and seafood but not milk and dairy products from their diet. Vegans were defined as those who excluded any kind of animal product. Exclusion criteria The following exclusion criteria were applied: (1) articles that did not provide original data (eg, systematic reviews, meta-analyses, literature reviews); (2) case reports; (3) articles that did not present data about the control group (omnivores); and (4) studies in which several interventions were carried out (ie, vegetarian or vegan diet combined with an increase in physical activity levels), making it impossible to separate the individual effects of the diet; and (5) studies that included eating disorders as a causative mental health outcome. Data extraction After all relevant literature was examined, depression, anxiety, stress, mental health/well-being, and mood disturbances were identified as outcomes of mental health and dementia/memory impairment was identified as the outcome of cognitive health for the present meta-analysis. For each study that included a mental health or cognitive outcome, relevant data were extracted (Table 2),17–29 including number of participants, sex, mean age, type of diet (eg, vegetarian, vegan, or omnivore), instruments used to assess the outcomes, study design, and quality assessment score. Instruments used to assess the outcomes were reported by study participants themselves or by a specialist. The following tools were used to evaluate depression: the Depression Anxiety Stress Scales–Depression (DASS-D), the Center for Epidemiologic Studies Depression Scale, the Hamilton Rating Scale for Depression. To assess anxiety, the Depression Anxiety Stress Scale–Anxiety (DASS-A) and the State-Trait Anxiety Inventory (STAI) were used. To evaluate stress, the Depression Anxiety Stress Scale–Stress (DASS-S) was used. The 36-Item Short Form Health Survey was used to assess depression, anxiety, and mental health all together. The Profile of Mood States tool was used assess mood disturbances. To gauge mental health, the Positive Mental Health Scale was used. To assess dementia/memory impairment, the Mini-Mental Status Examination was used. Other information extracted included the following: (1) whether the studies reported any confounders or used raw data; (2) possible differences between groups for relevant confounders (ie, vegan and vegetarians were usually more physically active, had a lower body mass index, had a higher education level, and were less likely to be married than omnivores); (3) the length of time the vegetarian or vegan diet had been followed; and (4) the country or countries in which the study was conducted, along with the predominant ethnicity of study participants, if reported. Table 2 Characteristics of included studies Reference . Participants . Outcomes of mental health and cognitive health . Study design . Other information . Quality assessment . . No. of vegans (mean age ± SD) . No. of LOVs (mean age ± SD) . No. of OMVs (mean age ± SD) . Sex . Continuous outcome . Instrument used . Mean ± SD . Categorical outcome . Instrument used . No. of events . Agarwal et al (2015)17 91 (43.8 ± 10.6) 92 (43.8 ± 10.6) 91 (43.8 ± 10.6) 110 (45.4 ± 11.3) 110 (45.4 ± 11.3) 109 (45.4 ± 11.3) F/M F/M F/M Depression Anxiety Mental health SF-36 SF-36 SF-36 Vegansa (18.2 ± 3.9) OMVs (24.4 ± 5.9) Vegansa (27.4 ± 6.3) OMVs (33.2 ± 10.4) Vegansa (78.3 ± 16.3) OMVs (72.1 ± 18.3) RCT Raw data No significant differences between intervention and control groups for any demographic or clinical measurements, except for sex (more men in intervention group than in control group) Period the diet was followedb: 18 wk (intervention) Country (predominant ethnicity): USA (Caucasian) 50% Baines et al (2007)18 252 (22–27) 8034 (22–27) F Mental health SF-36 LOVsa (47.6 ± 10.5) OMVs (50.5 ± 9.1) Depression Anxiety Deliberate self-harm Panic attacks or palpitations Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis LOVs: 51a OMVs: 1074 LOVsa: 24 OMVs: 466 LOVsa: 25 OMVs: 249 LOVs: 46 OMVs: 249 CS Raw data Vegetarians were more physically active, had a lower BMI, were more likely to live in an urban area, be unmarried, and have a university degree but lower income. No significant differences in smoking or alcohol consumption Period the diet was followedb: NR Country (predominant ethnicity): Australia (NR) 43% Beezhold et al (2010)19 60 (45.0 ± 11.1) 78 (41.0 ± 12.3) F/M Depression Anxiety Stress Mood disturbances DASS-D DASS-A DASS-S POMS (total) LOVsa (1.7 ± 2.1) OMVs (4.8 ± 6.0) LOVsa (1.5 ± 1.8) OMVs (4.3 ± 4.7) LOVsa (5.1 ± 4.0) OMVs (8.4 ± 8.1) LOVsa (0.1 ± 15.4) OMVs (15.3 ± 27.3) CS Raw data Vegetarians were older, more physically active, and had a lower BMI than nonvegetarians. No differences by education or sex between vegetarians and omnivores Period the diet was followedb: NR Country (predominant ethnicity): USA (NR) 36% Beezhold et al (2012)20 13 (NR) 13 (NR) F/M Depression Anxiety Stress Mood disturbances DASS-D DASS-A DASS-S POMS (total) LOVs (1.0 ± 1.4) OMVs (1.0 ± 1.4) LOVs (15 ± 23.7) OMVs (10.0 ± 14.0) LOVsa (8.5 ± 10.3) OMVs (14.0 ± 11.8) LOVs (5.0 ± 17.0) OMVs (3.0 ± 16.3) Pilot RCT Raw data Control and intervention groups did not differ by age, sex, BMI, educational level, ethnicity, total PA level, or fatty acid intake Period the diet was followedb: 2 wk (intervention) Country (predominant ethnicity): USA (NR) 57% Beezhold et al (2015)21 283 (37.2 ± 10.3) 109 (32.7 ± 9.5) 228 (34.6 ± 10.8) F/M Depression Anxiety Stress DASS-D DASS-A DASS-S Vegans (5.4 ± 6.5) LOVs (5.4 ± 5.9) OMVs (5.4 ± 6.7) Vegansa (3.4 ± 4.2) LOVs (3.9 ± 4.5) OMVs (4.7 ± 5.3) Vegansa (7.5 ± 6.3) LOVs (9.3 ± 7.5) OMVs (10.0 ± 8.2) CS Raw data Vegetarians and vegans were less likely to be married and had better lifestyle (higher PA, higher intakes of fruits and vegetables, lower alcohol intake) but lower education level than omnivores. No significant differences by sex or smoking Period the diet was followedb: NR Country (predominant ethnicity): USA (NR) 50% Forestell & Nezlek. (2018)22 223 (18.9 ± 1.3) 4192 (18.9 ± 1.3) F/M Depression CESD LOVs/Vegansa (17.1 ± 10.6) OMVs (14.2 ± 9.5) CS Confounder differences: NR Period the diet was followedb: NR Country (predominant ethnicity): USA (Caucasian) 50% Giem (1993)23 68 (> 65) 68 (> 65) 136 (> 65) F Dementia/memory impairment Physical and neurological examination Vegansa: 5 LOVsa: 3 OMVs: 16 PS Raw data and adjusted models (sex, age, education, ZIP code, previous diseases, alcohol, smoking, and BMI) Cases were matched for age and sex. Differences in other confounders: NR Period the diet was followedb: ≥ 20 y Country (predominant ethnicity): USA (NR) 86% Katcher et al (2010)24 65 (23–65) 44 (21–62) F/M Mental health SF-36 Vegans (76.9 ± 18.5) OMVs (76.8 ± 17.9) NRCT Raw data Vegans (intervention) were older, but there were no other significant differences between control and intervention groups Period the diet was followedb: 22 wk (intervention) Country (predominant ethnicity): USA (NR) 43% Kapoor et al (2017)25 100 (27.7 ± 5.8) 100 (28.8 ± 5.7) F/M Dementia/memory impairment Depression Personality change Psychosis MMSE HRSD Diagnosed by a doctor Diagnosed by a doctor LVsa: 7 OMVs 2 LVsa: 31 OMVs: 12 LVsa: 5 OMVs: 1 LVsa: 11 OMVs: 3 CS Raw data Confounder differences: NR. Vegetarians had lower mean B12 levels Period the diet was followedb: Since childhood Country (predominant ethnicity): Pakistan (NR) 57% Lindeman (2002)26 42 (29.0 ± 10.8) 197 (29.0 ± 10.8) F Depression CESD LOVsa (12.5 ± 6.18) OMVs (9.6 ± 5.7) CS Raw data Confounder differences: NR Period the diet was followedb: NR Country (predominant ethnicity): Finland (NR) 50% Rodríguez Jiménez et al (1998)27 20 (23–70) 20 (23–70) 40 (23–70) F/M Depression Anxiety CESD IDARE-STAI Vegansa (6.1 ± 4.9) LOVsa (4.5 ± 4.2) OMVs (12.5 ± 10.7) Vegansa (26.4 ± 4.7) LOVsa (29.5 ± 6.3) OMVs (33.9 ± 9.6) CS Raw data Confounder differences: NR Period the diet was followedb: > 3 y Country (predominant ethnicity): Puerto Rico (NR) 50% Timko et al (2012)28 35 (26.9 ± 7.9) 111 (26.7 ± 9.1) 265 (23.4 ± 9.7) F/M Depression Anxiety Stress DASS-D DASS-A DASS-S Vegans (6.5 ± 7.6) LOVs (9.5 ± 10.4) OMVs (7.0 ± 7.6) Vegans (0.6 ± 0.9) LOVs (0.5 ± 0.9) OMVs (0.4 ± 0.8) Vegans (11.8 ± 10.3) LOVs 11.5 ± 8.8) OMVs (10.4 ± 7.7) CS Raw data Lower BMI and higher rates of smoking in vegans and vegetarians than in omnivores. Not significant differences by sex or age Period the diet was followedb: ≥ 6 y Country (predominant ethnicity): USA (Caucasian) 57% Velten et al (2018)29 52 (23.4 ± 3.8) 69 227 (21.4 ± 3.3) 259 2204 (21.6 ± 4.0) 2483 F/M Depression Anxiety Stress Mental health DASS-D DASS-A DASS-S PMHS Vegansa (6.5 ± 5.3) LOVsa (5.3 ± 4.8) OMVs (4.2 ± 4.3) Vegansa (5.3 ± 4.5) LOVsa (5.3 ± 4.8) OMVs (3.4 ± 3.6) Vegansa (8.5 ± 4.4) LOVsa (8.0 ± 4.9) OMVs (7.1 ± 4.6) Vegansa (16.2 ± 5.5) LOVsa (16.9 ± 5.9) OMVs (18.3 ± 5.3) PS Raw and adjusted data Models (for sex, age, and baseline mental health) Differences found in lifestyle factors between the Chinese and German samples Period the diet was followedb: NR Country (predominant ethnicity): Germany and China (NR) 57% Reference . Participants . Outcomes of mental health and cognitive health . Study design . Other information . Quality assessment . . No. of vegans (mean age ± SD) . No. of LOVs (mean age ± SD) . No. of OMVs (mean age ± SD) . Sex . Continuous outcome . Instrument used . Mean ± SD . Categorical outcome . Instrument used . No. of events . Agarwal et al (2015)17 91 (43.8 ± 10.6) 92 (43.8 ± 10.6) 91 (43.8 ± 10.6) 110 (45.4 ± 11.3) 110 (45.4 ± 11.3) 109 (45.4 ± 11.3) F/M F/M F/M Depression Anxiety Mental health SF-36 SF-36 SF-36 Vegansa (18.2 ± 3.9) OMVs (24.4 ± 5.9) Vegansa (27.4 ± 6.3) OMVs (33.2 ± 10.4) Vegansa (78.3 ± 16.3) OMVs (72.1 ± 18.3) RCT Raw data No significant differences between intervention and control groups for any demographic or clinical measurements, except for sex (more men in intervention group than in control group) Period the diet was followedb: 18 wk (intervention) Country (predominant ethnicity): USA (Caucasian) 50% Baines et al (2007)18 252 (22–27) 8034 (22–27) F Mental health SF-36 LOVsa (47.6 ± 10.5) OMVs (50.5 ± 9.1) Depression Anxiety Deliberate self-harm Panic attacks or palpitations Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis LOVs: 51a OMVs: 1074 LOVsa: 24 OMVs: 466 LOVsa: 25 OMVs: 249 LOVs: 46 OMVs: 249 CS Raw data Vegetarians were more physically active, had a lower BMI, were more likely to live in an urban area, be unmarried, and have a university degree but lower income. No significant differences in smoking or alcohol consumption Period the diet was followedb: NR Country (predominant ethnicity): Australia (NR) 43% Beezhold et al (2010)19 60 (45.0 ± 11.1) 78 (41.0 ± 12.3) F/M Depression Anxiety Stress Mood disturbances DASS-D DASS-A DASS-S POMS (total) LOVsa (1.7 ± 2.1) OMVs (4.8 ± 6.0) LOVsa (1.5 ± 1.8) OMVs (4.3 ± 4.7) LOVsa (5.1 ± 4.0) OMVs (8.4 ± 8.1) LOVsa (0.1 ± 15.4) OMVs (15.3 ± 27.3) CS Raw data Vegetarians were older, more physically active, and had a lower BMI than nonvegetarians. No differences by education or sex between vegetarians and omnivores Period the diet was followedb: NR Country (predominant ethnicity): USA (NR) 36% Beezhold et al (2012)20 13 (NR) 13 (NR) F/M Depression Anxiety Stress Mood disturbances DASS-D DASS-A DASS-S POMS (total) LOVs (1.0 ± 1.4) OMVs (1.0 ± 1.4) LOVs (15 ± 23.7) OMVs (10.0 ± 14.0) LOVsa (8.5 ± 10.3) OMVs (14.0 ± 11.8) LOVs (5.0 ± 17.0) OMVs (3.0 ± 16.3) Pilot RCT Raw data Control and intervention groups did not differ by age, sex, BMI, educational level, ethnicity, total PA level, or fatty acid intake Period the diet was followedb: 2 wk (intervention) Country (predominant ethnicity): USA (NR) 57% Beezhold et al (2015)21 283 (37.2 ± 10.3) 109 (32.7 ± 9.5) 228 (34.6 ± 10.8) F/M Depression Anxiety Stress DASS-D DASS-A DASS-S Vegans (5.4 ± 6.5) LOVs (5.4 ± 5.9) OMVs (5.4 ± 6.7) Vegansa (3.4 ± 4.2) LOVs (3.9 ± 4.5) OMVs (4.7 ± 5.3) Vegansa (7.5 ± 6.3) LOVs (9.3 ± 7.5) OMVs (10.0 ± 8.2) CS Raw data Vegetarians and vegans were less likely to be married and had better lifestyle (higher PA, higher intakes of fruits and vegetables, lower alcohol intake) but lower education level than omnivores. No significant differences by sex or smoking Period the diet was followedb: NR Country (predominant ethnicity): USA (NR) 50% Forestell & Nezlek. (2018)22 223 (18.9 ± 1.3) 4192 (18.9 ± 1.3) F/M Depression CESD LOVs/Vegansa (17.1 ± 10.6) OMVs (14.2 ± 9.5) CS Confounder differences: NR Period the diet was followedb: NR Country (predominant ethnicity): USA (Caucasian) 50% Giem (1993)23 68 (> 65) 68 (> 65) 136 (> 65) F Dementia/memory impairment Physical and neurological examination Vegansa: 5 LOVsa: 3 OMVs: 16 PS Raw data and adjusted models (sex, age, education, ZIP code, previous diseases, alcohol, smoking, and BMI) Cases were matched for age and sex. Differences in other confounders: NR Period the diet was followedb: ≥ 20 y Country (predominant ethnicity): USA (NR) 86% Katcher et al (2010)24 65 (23–65) 44 (21–62) F/M Mental health SF-36 Vegans (76.9 ± 18.5) OMVs (76.8 ± 17.9) NRCT Raw data Vegans (intervention) were older, but there were no other significant differences between control and intervention groups Period the diet was followedb: 22 wk (intervention) Country (predominant ethnicity): USA (NR) 43% Kapoor et al (2017)25 100 (27.7 ± 5.8) 100 (28.8 ± 5.7) F/M Dementia/memory impairment Depression Personality change Psychosis MMSE HRSD Diagnosed by a doctor Diagnosed by a doctor LVsa: 7 OMVs 2 LVsa: 31 OMVs: 12 LVsa: 5 OMVs: 1 LVsa: 11 OMVs: 3 CS Raw data Confounder differences: NR. Vegetarians had lower mean B12 levels Period the diet was followedb: Since childhood Country (predominant ethnicity): Pakistan (NR) 57% Lindeman (2002)26 42 (29.0 ± 10.8) 197 (29.0 ± 10.8) F Depression CESD LOVsa (12.5 ± 6.18) OMVs (9.6 ± 5.7) CS Raw data Confounder differences: NR Period the diet was followedb: NR Country (predominant ethnicity): Finland (NR) 50% Rodríguez Jiménez et al (1998)27 20 (23–70) 20 (23–70) 40 (23–70) F/M Depression Anxiety CESD IDARE-STAI Vegansa (6.1 ± 4.9) LOVsa (4.5 ± 4.2) OMVs (12.5 ± 10.7) Vegansa (26.4 ± 4.7) LOVsa (29.5 ± 6.3) OMVs (33.9 ± 9.6) CS Raw data Confounder differences: NR Period the diet was followedb: > 3 y Country (predominant ethnicity): Puerto Rico (NR) 50% Timko et al (2012)28 35 (26.9 ± 7.9) 111 (26.7 ± 9.1) 265 (23.4 ± 9.7) F/M Depression Anxiety Stress DASS-D DASS-A DASS-S Vegans (6.5 ± 7.6) LOVs (9.5 ± 10.4) OMVs (7.0 ± 7.6) Vegans (0.6 ± 0.9) LOVs (0.5 ± 0.9) OMVs (0.4 ± 0.8) Vegans (11.8 ± 10.3) LOVs 11.5 ± 8.8) OMVs (10.4 ± 7.7) CS Raw data Lower BMI and higher rates of smoking in vegans and vegetarians than in omnivores. Not significant differences by sex or age Period the diet was followedb: ≥ 6 y Country (predominant ethnicity): USA (Caucasian) 57% Velten et al (2018)29 52 (23.4 ± 3.8) 69 227 (21.4 ± 3.3) 259 2204 (21.6 ± 4.0) 2483 F/M Depression Anxiety Stress Mental health DASS-D DASS-A DASS-S PMHS Vegansa (6.5 ± 5.3) LOVsa (5.3 ± 4.8) OMVs (4.2 ± 4.3) Vegansa (5.3 ± 4.5) LOVsa (5.3 ± 4.8) OMVs (3.4 ± 3.6) Vegansa (8.5 ± 4.4) LOVsa (8.0 ± 4.9) OMVs (7.1 ± 4.6) Vegansa (16.2 ± 5.5) LOVsa (16.9 ± 5.9) OMVs (18.3 ± 5.3) PS Raw and adjusted data Models (for sex, age, and baseline mental health) Differences found in lifestyle factors between the Chinese and German samples Period the diet was followedb: NR Country (predominant ethnicity): Germany and China (NR) 57% Abbreviations: BMI, body mass index; CESD, Center for Epidemiologic Studies Depression Scale; CS, cross-sectional study; DASS-A, Depression Anxiety Stress Scale–Anxiety; DASS-D, Depression Anxiety Stress Scale–Depression; DASS-S, Depression Anxiety Stress Scale–Stress; CESD, Center for Epidemiologic Studies Depression Scale; HRSD, Hamilton Rating Scale for Depression; IDARE, Inventario de Ansiedad Ragso-Estado (Spanish version of State-Trait Anxiety Inventory); LOV, lacto-ovo-vegetarian; LV, lacto-vegetarian, MMSE, Mini-Mental Status Examination; NR, not reported; NRCT, nonrandomized controlled trial; OMV, omnivore; PMHS, Positive Mental Health Scale; POMS, Profile of Mood States; PS, prospective study; SD, standard deviation; SF-36, 36-Item Short Form Health Survey; STAI, State-Trait Anxiety Inventory. a Results statistically significant between vegetarians or vegans and omnivores. b Minimum number of years the vegetarian or vegan diet was followed. Open in new tab Table 2 Characteristics of included studies Reference . Participants . Outcomes of mental health and cognitive health . Study design . Other information . Quality assessment . . No. of vegans (mean age ± SD) . No. of LOVs (mean age ± SD) . No. of OMVs (mean age ± SD) . Sex . Continuous outcome . Instrument used . Mean ± SD . Categorical outcome . Instrument used . No. of events . Agarwal et al (2015)17 91 (43.8 ± 10.6) 92 (43.8 ± 10.6) 91 (43.8 ± 10.6) 110 (45.4 ± 11.3) 110 (45.4 ± 11.3) 109 (45.4 ± 11.3) F/M F/M F/M Depression Anxiety Mental health SF-36 SF-36 SF-36 Vegansa (18.2 ± 3.9) OMVs (24.4 ± 5.9) Vegansa (27.4 ± 6.3) OMVs (33.2 ± 10.4) Vegansa (78.3 ± 16.3) OMVs (72.1 ± 18.3) RCT Raw data No significant differences between intervention and control groups for any demographic or clinical measurements, except for sex (more men in intervention group than in control group) Period the diet was followedb: 18 wk (intervention) Country (predominant ethnicity): USA (Caucasian) 50% Baines et al (2007)18 252 (22–27) 8034 (22–27) F Mental health SF-36 LOVsa (47.6 ± 10.5) OMVs (50.5 ± 9.1) Depression Anxiety Deliberate self-harm Panic attacks or palpitations Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis LOVs: 51a OMVs: 1074 LOVsa: 24 OMVs: 466 LOVsa: 25 OMVs: 249 LOVs: 46 OMVs: 249 CS Raw data Vegetarians were more physically active, had a lower BMI, were more likely to live in an urban area, be unmarried, and have a university degree but lower income. No significant differences in smoking or alcohol consumption Period the diet was followedb: NR Country (predominant ethnicity): Australia (NR) 43% Beezhold et al (2010)19 60 (45.0 ± 11.1) 78 (41.0 ± 12.3) F/M Depression Anxiety Stress Mood disturbances DASS-D DASS-A DASS-S POMS (total) LOVsa (1.7 ± 2.1) OMVs (4.8 ± 6.0) LOVsa (1.5 ± 1.8) OMVs (4.3 ± 4.7) LOVsa (5.1 ± 4.0) OMVs (8.4 ± 8.1) LOVsa (0.1 ± 15.4) OMVs (15.3 ± 27.3) CS Raw data Vegetarians were older, more physically active, and had a lower BMI than nonvegetarians. No differences by education or sex between vegetarians and omnivores Period the diet was followedb: NR Country (predominant ethnicity): USA (NR) 36% Beezhold et al (2012)20 13 (NR) 13 (NR) F/M Depression Anxiety Stress Mood disturbances DASS-D DASS-A DASS-S POMS (total) LOVs (1.0 ± 1.4) OMVs (1.0 ± 1.4) LOVs (15 ± 23.7) OMVs (10.0 ± 14.0) LOVsa (8.5 ± 10.3) OMVs (14.0 ± 11.8) LOVs (5.0 ± 17.0) OMVs (3.0 ± 16.3) Pilot RCT Raw data Control and intervention groups did not differ by age, sex, BMI, educational level, ethnicity, total PA level, or fatty acid intake Period the diet was followedb: 2 wk (intervention) Country (predominant ethnicity): USA (NR) 57% Beezhold et al (2015)21 283 (37.2 ± 10.3) 109 (32.7 ± 9.5) 228 (34.6 ± 10.8) F/M Depression Anxiety Stress DASS-D DASS-A DASS-S Vegans (5.4 ± 6.5) LOVs (5.4 ± 5.9) OMVs (5.4 ± 6.7) Vegansa (3.4 ± 4.2) LOVs (3.9 ± 4.5) OMVs (4.7 ± 5.3) Vegansa (7.5 ± 6.3) LOVs (9.3 ± 7.5) OMVs (10.0 ± 8.2) CS Raw data Vegetarians and vegans were less likely to be married and had better lifestyle (higher PA, higher intakes of fruits and vegetables, lower alcohol intake) but lower education level than omnivores. No significant differences by sex or smoking Period the diet was followedb: NR Country (predominant ethnicity): USA (NR) 50% Forestell & Nezlek. (2018)22 223 (18.9 ± 1.3) 4192 (18.9 ± 1.3) F/M Depression CESD LOVs/Vegansa (17.1 ± 10.6) OMVs (14.2 ± 9.5) CS Confounder differences: NR Period the diet was followedb: NR Country (predominant ethnicity): USA (Caucasian) 50% Giem (1993)23 68 (> 65) 68 (> 65) 136 (> 65) F Dementia/memory impairment Physical and neurological examination Vegansa: 5 LOVsa: 3 OMVs: 16 PS Raw data and adjusted models (sex, age, education, ZIP code, previous diseases, alcohol, smoking, and BMI) Cases were matched for age and sex. Differences in other confounders: NR Period the diet was followedb: ≥ 20 y Country (predominant ethnicity): USA (NR) 86% Katcher et al (2010)24 65 (23–65) 44 (21–62) F/M Mental health SF-36 Vegans (76.9 ± 18.5) OMVs (76.8 ± 17.9) NRCT Raw data Vegans (intervention) were older, but there were no other significant differences between control and intervention groups Period the diet was followedb: 22 wk (intervention) Country (predominant ethnicity): USA (NR) 43% Kapoor et al (2017)25 100 (27.7 ± 5.8) 100 (28.8 ± 5.7) F/M Dementia/memory impairment Depression Personality change Psychosis MMSE HRSD Diagnosed by a doctor Diagnosed by a doctor LVsa: 7 OMVs 2 LVsa: 31 OMVs: 12 LVsa: 5 OMVs: 1 LVsa: 11 OMVs: 3 CS Raw data Confounder differences: NR. Vegetarians had lower mean B12 levels Period the diet was followedb: Since childhood Country (predominant ethnicity): Pakistan (NR) 57% Lindeman (2002)26 42 (29.0 ± 10.8) 197 (29.0 ± 10.8) F Depression CESD LOVsa (12.5 ± 6.18) OMVs (9.6 ± 5.7) CS Raw data Confounder differences: NR Period the diet was followedb: NR Country (predominant ethnicity): Finland (NR) 50% Rodríguez Jiménez et al (1998)27 20 (23–70) 20 (23–70) 40 (23–70) F/M Depression Anxiety CESD IDARE-STAI Vegansa (6.1 ± 4.9) LOVsa (4.5 ± 4.2) OMVs (12.5 ± 10.7) Vegansa (26.4 ± 4.7) LOVsa (29.5 ± 6.3) OMVs (33.9 ± 9.6) CS Raw data Confounder differences: NR Period the diet was followedb: > 3 y Country (predominant ethnicity): Puerto Rico (NR) 50% Timko et al (2012)28 35 (26.9 ± 7.9) 111 (26.7 ± 9.1) 265 (23.4 ± 9.7) F/M Depression Anxiety Stress DASS-D DASS-A DASS-S Vegans (6.5 ± 7.6) LOVs (9.5 ± 10.4) OMVs (7.0 ± 7.6) Vegans (0.6 ± 0.9) LOVs (0.5 ± 0.9) OMVs (0.4 ± 0.8) Vegans (11.8 ± 10.3) LOVs 11.5 ± 8.8) OMVs (10.4 ± 7.7) CS Raw data Lower BMI and higher rates of smoking in vegans and vegetarians than in omnivores. Not significant differences by sex or age Period the diet was followedb: ≥ 6 y Country (predominant ethnicity): USA (Caucasian) 57% Velten et al (2018)29 52 (23.4 ± 3.8) 69 227 (21.4 ± 3.3) 259 2204 (21.6 ± 4.0) 2483 F/M Depression Anxiety Stress Mental health DASS-D DASS-A DASS-S PMHS Vegansa (6.5 ± 5.3) LOVsa (5.3 ± 4.8) OMVs (4.2 ± 4.3) Vegansa (5.3 ± 4.5) LOVsa (5.3 ± 4.8) OMVs (3.4 ± 3.6) Vegansa (8.5 ± 4.4) LOVsa (8.0 ± 4.9) OMVs (7.1 ± 4.6) Vegansa (16.2 ± 5.5) LOVsa (16.9 ± 5.9) OMVs (18.3 ± 5.3) PS Raw and adjusted data Models (for sex, age, and baseline mental health) Differences found in lifestyle factors between the Chinese and German samples Period the diet was followedb: NR Country (predominant ethnicity): Germany and China (NR) 57% Reference . Participants . Outcomes of mental health and cognitive health . Study design . Other information . Quality assessment . . No. of vegans (mean age ± SD) . No. of LOVs (mean age ± SD) . No. of OMVs (mean age ± SD) . Sex . Continuous outcome . Instrument used . Mean ± SD . Categorical outcome . Instrument used . No. of events . Agarwal et al (2015)17 91 (43.8 ± 10.6) 92 (43.8 ± 10.6) 91 (43.8 ± 10.6) 110 (45.4 ± 11.3) 110 (45.4 ± 11.3) 109 (45.4 ± 11.3) F/M F/M F/M Depression Anxiety Mental health SF-36 SF-36 SF-36 Vegansa (18.2 ± 3.9) OMVs (24.4 ± 5.9) Vegansa (27.4 ± 6.3) OMVs (33.2 ± 10.4) Vegansa (78.3 ± 16.3) OMVs (72.1 ± 18.3) RCT Raw data No significant differences between intervention and control groups for any demographic or clinical measurements, except for sex (more men in intervention group than in control group) Period the diet was followedb: 18 wk (intervention) Country (predominant ethnicity): USA (Caucasian) 50% Baines et al (2007)18 252 (22–27) 8034 (22–27) F Mental health SF-36 LOVsa (47.6 ± 10.5) OMVs (50.5 ± 9.1) Depression Anxiety Deliberate self-harm Panic attacks or palpitations Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis Self-reported on basis of doctor’s diagnosis LOVs: 51a OMVs: 1074 LOVsa: 24 OMVs: 466 LOVsa: 25 OMVs: 249 LOVs: 46 OMVs: 249 CS Raw data Vegetarians were more physically active, had a lower BMI, were more likely to live in an urban area, be unmarried, and have a university degree but lower income. No significant differences in smoking or alcohol consumption Period the diet was followedb: NR Country (predominant ethnicity): Australia (NR) 43% Beezhold et al (2010)19 60 (45.0 ± 11.1) 78 (41.0 ± 12.3) F/M Depression Anxiety Stress Mood disturbances DASS-D DASS-A DASS-S POMS (total) LOVsa (1.7 ± 2.1) OMVs (4.8 ± 6.0) LOVsa (1.5 ± 1.8) OMVs (4.3 ± 4.7) LOVsa (5.1 ± 4.0) OMVs (8.4 ± 8.1) LOVsa (0.1 ± 15.4) OMVs (15.3 ± 27.3) CS Raw data Vegetarians were older, more physically active, and had a lower BMI than nonvegetarians. No differences by education or sex between vegetarians and omnivores Period the diet was followedb: NR Country (predominant ethnicity): USA (NR) 36% Beezhold et al (2012)20 13 (NR) 13 (NR) F/M Depression Anxiety Stress Mood disturbances DASS-D DASS-A DASS-S POMS (total) LOVs (1.0 ± 1.4) OMVs (1.0 ± 1.4) LOVs (15 ± 23.7) OMVs (10.0 ± 14.0) LOVsa (8.5 ± 10.3) OMVs (14.0 ± 11.8) LOVs (5.0 ± 17.0) OMVs (3.0 ± 16.3) Pilot RCT Raw data Control and intervention groups did not differ by age, sex, BMI, educational level, ethnicity, total PA level, or fatty acid intake Period the diet was followedb: 2 wk (intervention) Country (predominant ethnicity): USA (NR) 57% Beezhold et al (2015)21 283 (37.2 ± 10.3) 109 (32.7 ± 9.5) 228 (34.6 ± 10.8) F/M Depression Anxiety Stress DASS-D DASS-A DASS-S Vegans (5.4 ± 6.5) LOVs (5.4 ± 5.9) OMVs (5.4 ± 6.7) Vegansa (3.4 ± 4.2) LOVs (3.9 ± 4.5) OMVs (4.7 ± 5.3) Vegansa (7.5 ± 6.3) LOVs (9.3 ± 7.5) OMVs (10.0 ± 8.2) CS Raw data Vegetarians and vegans were less likely to be married and had better lifestyle (higher PA, higher intakes of fruits and vegetables, lower alcohol intake) but lower education level than omnivores. No significant differences by sex or smoking Period the diet was followedb: NR Country (predominant ethnicity): USA (NR) 50% Forestell & Nezlek. (2018)22 223 (18.9 ± 1.3) 4192 (18.9 ± 1.3) F/M Depression CESD LOVs/Vegansa (17.1 ± 10.6) OMVs (14.2 ± 9.5) CS Confounder differences: NR Period the diet was followedb: NR Country (predominant ethnicity): USA (Caucasian) 50% Giem (1993)23 68 (> 65) 68 (> 65) 136 (> 65) F Dementia/memory impairment Physical and neurological examination Vegansa: 5 LOVsa: 3 OMVs: 16 PS Raw data and adjusted models (sex, age, education, ZIP code, previous diseases, alcohol, smoking, and BMI) Cases were matched for age and sex. Differences in other confounders: NR Period the diet was followedb: ≥ 20 y Country (predominant ethnicity): USA (NR) 86% Katcher et al (2010)24 65 (23–65) 44 (21–62) F/M Mental health SF-36 Vegans (76.9 ± 18.5) OMVs (76.8 ± 17.9) NRCT Raw data Vegans (intervention) were older, but there were no other significant differences between control and intervention groups Period the diet was followedb: 22 wk (intervention) Country (predominant ethnicity): USA (NR) 43% Kapoor et al (2017)25 100 (27.7 ± 5.8) 100 (28.8 ± 5.7) F/M Dementia/memory impairment Depression Personality change Psychosis MMSE HRSD Diagnosed by a doctor Diagnosed by a doctor LVsa: 7 OMVs 2 LVsa: 31 OMVs: 12 LVsa: 5 OMVs: 1 LVsa: 11 OMVs: 3 CS Raw data Confounder differences: NR. Vegetarians had lower mean B12 levels Period the diet was followedb: Since childhood Country (predominant ethnicity): Pakistan (NR) 57% Lindeman (2002)26 42 (29.0 ± 10.8) 197 (29.0 ± 10.8) F Depression CESD LOVsa (12.5 ± 6.18) OMVs (9.6 ± 5.7) CS Raw data Confounder differences: NR Period the diet was followedb: NR Country (predominant ethnicity): Finland (NR) 50% Rodríguez Jiménez et al (1998)27 20 (23–70) 20 (23–70) 40 (23–70) F/M Depression Anxiety CESD IDARE-STAI Vegansa (6.1 ± 4.9) LOVsa (4.5 ± 4.2) OMVs (12.5 ± 10.7) Vegansa (26.4 ± 4.7) LOVsa (29.5 ± 6.3) OMVs (33.9 ± 9.6) CS Raw data Confounder differences: NR Period the diet was followedb: > 3 y Country (predominant ethnicity): Puerto Rico (NR) 50% Timko et al (2012)28 35 (26.9 ± 7.9) 111 (26.7 ± 9.1) 265 (23.4 ± 9.7) F/M Depression Anxiety Stress DASS-D DASS-A DASS-S Vegans (6.5 ± 7.6) LOVs (9.5 ± 10.4) OMVs (7.0 ± 7.6) Vegans (0.6 ± 0.9) LOVs (0.5 ± 0.9) OMVs (0.4 ± 0.8) Vegans (11.8 ± 10.3) LOVs 11.5 ± 8.8) OMVs (10.4 ± 7.7) CS Raw data Lower BMI and higher rates of smoking in vegans and vegetarians than in omnivores. Not significant differences by sex or age Period the diet was followedb: ≥ 6 y Country (predominant ethnicity): USA (Caucasian) 57% Velten et al (2018)29 52 (23.4 ± 3.8) 69 227 (21.4 ± 3.3) 259 2204 (21.6 ± 4.0) 2483 F/M Depression Anxiety Stress Mental health DASS-D DASS-A DASS-S PMHS Vegansa (6.5 ± 5.3) LOVsa (5.3 ± 4.8) OMVs (4.2 ± 4.3) Vegansa (5.3 ± 4.5) LOVsa (5.3 ± 4.8) OMVs (3.4 ± 3.6) Vegansa (8.5 ± 4.4) LOVsa (8.0 ± 4.9) OMVs (7.1 ± 4.6) Vegansa (16.2 ± 5.5) LOVsa (16.9 ± 5.9) OMVs (18.3 ± 5.3) PS Raw and adjusted data Models (for sex, age, and baseline mental health) Differences found in lifestyle factors between the Chinese and German samples Period the diet was followedb: NR Country (predominant ethnicity): Germany and China (NR) 57% Abbreviations: BMI, body mass index; CESD, Center for Epidemiologic Studies Depression Scale; CS, cross-sectional study; DASS-A, Depression Anxiety Stress Scale–Anxiety; DASS-D, Depression Anxiety Stress Scale–Depression; DASS-S, Depression Anxiety Stress Scale–Stress; CESD, Center for Epidemiologic Studies Depression Scale; HRSD, Hamilton Rating Scale for Depression; IDARE, Inventario de Ansiedad Ragso-Estado (Spanish version of State-Trait Anxiety Inventory); LOV, lacto-ovo-vegetarian; LV, lacto-vegetarian, MMSE, Mini-Mental Status Examination; NR, not reported; NRCT, nonrandomized controlled trial; OMV, omnivore; PMHS, Positive Mental Health Scale; POMS, Profile of Mood States; PS, prospective study; SD, standard deviation; SF-36, 36-Item Short Form Health Survey; STAI, State-Trait Anxiety Inventory. a Results statistically significant between vegetarians or vegans and omnivores. b Minimum number of years the vegetarian or vegan diet was followed. Open in new tab Means and SDs were used for continuous outcomes, while the number of events in each group (ie, number of cases of depression diagnosed in vegetarians and vegans vs omnivores) was used for categorical outcomes. Only raw data (unadjusted) were used to perform the meta-analyses, as only 2 publications in the present meta-analysis included adjusted data.23,29 Reporting unadjusted estimates also reduces the bias of selective reporting of adjusted estimates in primary studies and the risk of overadjustment with multiple confounders. When a study offered information about matched and nonmatched data, the matched data were used for analysis. Quality assessment Depending on the study design, the Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies37 or the Quality Assessment Tool of Controlled Intervention Studies38 provided by the National Heart, Lung, and Blood Institute (NHBLI) was used to assess the quality of the included studies. Two authors (I.I and N.M) independently assessed and thereafter discussed the quality of the studies. Agreement between authors was found for 90% of the publications, with discrepancies resolved by discussion. Search summary A total of 1249 articles were extracted from PubMed, Scopus, Proquest, and ScienceDirect. After eliminating all duplicates, 911 manuscripts were evaluated by reading titles and abstracts, which resulted in the exclusion of 857 records. Fifty-four full-text articles were assessed for eligibility, of which a further 41 were excluded (Figure 1): 13 because the studies did not include a purely vegetarian or vegan diet (ie, those that included flexitarians or self-reported vegetarians who reported fish consumption)13,30,39–49; 11 because the studies had a different objective than that of the present meta-analysis50–60; 6 because the studies did not include a control group (omnivorous group) to compare the mental or cognitive outcomes61–66; 4 because eating disorders were included as a mental health outcome67–70; 3 because the studies were conducted in unhealthy individuals (ie, studies in which participants reported having eating disorders before starting the diet)71–73; 1 because no raw data were provided60,74; 2 because the studies did not employ purely dietary interventions in which individual dietary effects could be separated75,76; and 1 because it was not an original article.77 The main author was contacted in cases in which the article did not provide raw or adjusted data and to clarify obscure definitions of diets. The main author was also contacted in cases in which mixed diets had been reported in the same group (eg, vegetarian, vegan, and pescatarian were reported in the same group). In such cases, separated data were requested, if available, which was possible for 1 study that was included.29 Figure 1 Open in new tabDownload slide Flow diagram of the literature search process. Figure 1 Open in new tabDownload slide Flow diagram of the literature search process. Statistical analyses All analyses were performed using OpenMeta[Analyst] software. For continuous outcomes, information on sample size (mean ± SD) was obtained, and for categorical outcomes, the number of events and observations in each group was extracted for meta-analysis. When the SD was not reported in the study, the authors were contacted. If no response was received, the following formula was applied: standard error =SD/√n; SD = interquartile range/1.35. When the mean was not reported in the studies, the median was used. The mean difference (MD) and the relative risk or the odds ratio (OR) with a 95%CI were calculated for continuous and categorical outcomes, respectively. DerSimonian and Laird estimators using random-effects models were applied for continuous and categorical data. Effect sizes were calculated for each outcome. When possible, subgroup analyses were conducted. In fact, several specific subgroup analyses were not performed because of a lack of studies (ie, subgroups for which data could be obtained from only 1 or 2 studies). Sources of heterogeneity were investigated by subgroup analyses that compared results on the basis of age (≤ 25 years, 26–45 years, or > 45 years, or not reported); sex (women, men, mainly women, men and women); length of time the vegetarian or vegan diet had been followed (short-term period: <1 year; long-term: ≥1 year; or not reported); type of diet (vegetarian or vegan), instrument used to assess the mental or cognitive outcome; and quality assessment score (< 50 vs ≥ 50), when information was available. The heterogeneity of the studies was tested using the I2 statistic,78 which describes the variance between studies as a proportion of the total variance. A value of less than 25% indicated low heterogeneity, 25% to 50% moderate heterogeneity, 50% to 75% high heterogeneity, and more than 75% very high heterogeneity. The P value of the heterogeneity of the studies was also calculated, with a nonsignificant result indicating an absence of heterogeneity. Funnel plots were created (RevMan, version 5.2) to investigate publication bias. Publication bias was assessed by the Egger linear regression test, following the guidelines provided by Peters et al.79 Thus, funnel plots were created and tests were performed when the meta-analysis had more than 10 studies, as a small number of studies lowers the test power to a point where it is too low to distinguish chance from actual asymmetry.80 RESULTS Included studies After applying the exclusion criteria, 13 articles17–29 were included in this review: 11 examined depression as an outcome (9 as continuous and 2 as categorical); 8 examined anxiety as an outcome (7 as continuous and 1 as categorical); 5 examined stress as an outcome (as continuous); 4 examined mental health/well-being as an outcome (continuous); 3 examined memory impairment/Alzheimer disease as an outcome (categorical); 2 examined mood disturbances as an outcome (continuous); 1 examined neuropsychiatric disorders as an outcome (categorical); and 1 examined personality change as an outcome (categorical) (Figure 1). The present meta-analysis includes 17 809 individuals. The mean age of the participants varied from 18 to 80 years, and most of the studies distributed the sample equally between omnivores and vegetarians/vegans. Most of the studies included both women and men (although with a higher participation of women than men) or only women. Of the 13 eligible studies, 6 included vegetarians only, 2 included vegans only, and 5 included both vegetarians and vegans. Eight studies were conducted in the United States, 1 in Australia, 1 in Pakistan, 1 in Finland, 1 in Puerto Rico, and 1 in both Germany and China. Eight of the 13 studies were designed as cross-sectional studies, 2 were prospective, 2 were randomized clinical trials, and 2 were nonrandomized trials (Table 2). Most of the studies were of medium quality, with quality assessment scores varying between 28 and 64 on a 100-point scale (see Tables S1 and S2 in the Supporting Information online). Mental health differences between vegetarians/vegans and omnivores Figures 2 through 717–29 show the results of individual studies as well as several subgroup analyses of the global effect of vegetarianism/veganism on mental health and cognitive outcomes. Several specific subgroup analyses were not performed because studies contained insufficient data (ie, subgroups for which data could be obtained from only 1 or 2 studies could not be analyzed). Figure 2 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on depression (continuous outcome). (A) Differences in depression (continuous outcome) between vegetarians or vegans and omnivores; (B) Subgroup analyses by age (≤ 25 y, 26–45 y, > 45 y, or not reported); (C) Subgroup analyses by sex (women, men, women/men, or not reported); (D) Subgroup analyses by length of time the diet was followed (short-term, long-term, or not reported); (E) Subgroup analyses by type of diet (vegetarians vs vegans); (F) Subgroup analyses by instrument used (DASS-D, others); (G) Subgroup analyses by quality assessment score (< 50 vs ≥ 50). Abbreviations: CESD, Center for Epidemiologic Studies Depression Scale; DASS-D, Depression Anxiety Stress Scale–Depression; NR, not reported. Figure 2 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on depression (continuous outcome). (A) Differences in depression (continuous outcome) between vegetarians or vegans and omnivores; (B) Subgroup analyses by age (≤ 25 y, 26–45 y, > 45 y, or not reported); (C) Subgroup analyses by sex (women, men, women/men, or not reported); (D) Subgroup analyses by length of time the diet was followed (short-term, long-term, or not reported); (E) Subgroup analyses by type of diet (vegetarians vs vegans); (F) Subgroup analyses by instrument used (DASS-D, others); (G) Subgroup analyses by quality assessment score (< 50 vs ≥ 50). Abbreviations: CESD, Center for Epidemiologic Studies Depression Scale; DASS-D, Depression Anxiety Stress Scale–Depression; NR, not reported. Depression (continuous outcome) As shown in Figure 2A,17,19–22,26–29 no statistically significant differences were found between vegetarian or vegan diets and omnivores regarding the incidence of depression (MD = −0.531 95%CI, −2.047 to 0.984). Heterogeneity among studies for depression was very high (I2 = 92.43; P = 0.040). Subgroup analysis showed a statistically significant higher incidence of depression in vegetarians/vegans under 26 years of age (MD = 1.739; 95%CI, 0.758–2.719; Figure 2B17,19,21,22,26–29), and in studies with higher quality assessment scores (MD = 1.418; 95%CI, 0.473–2.363; Figure 2F17,19–22,26–29). Subgroup analyses of sex (Figure 2C19,21,22,27–29), the length of time the diet was followed (Figure 2C19,21,22,26–29), the type of diet (vegan vs vegetarian; Figure 2D17,19–21,26–29), or the instrument used for assessment (Figure 2E19–22,27–29) did not show any significant differences between vegetarians/vegans and omnivores. Heterogeneity was more pronounced in those over 25 years of age (I2 = 93.22; P < 0.001), in studies that included mainly women (I2 = 94.29; P < 0.001), in short-term studies (I2 = 97.87; P < 0.001), in studies that included vegans (I2 = 95.09; P < 0.001), in studies that used instruments other than the Depression Anxiety Stress Scales tool (I2 = 96.35; P < 0.001), and in lower-quality studies (I2 = 93.62; P < 0.001). Depression (categorical outcome) As shown in Figure 3,18,25 vegetarians and vegans had a higher risk of depression when depression was assessed as a categorical variable (OR = 2.142; 95%CI, 1.105–4.148), but heterogeneity among studies was high (I2 = 65.4; P = 0.089). In addition to the significant heterogeneity between the 2 studies, conclusions are tentative because the analysis only included 2 studies. Figure 3 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on depression (categorical outcome). Differences in depression (categorical outcomes) between vegetarians or vegans and omnivores are shown. Abbreviation: HRSD, Hamilton Rating Scale for Depression. Figure 3 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on depression (categorical outcome). Differences in depression (categorical outcomes) between vegetarians or vegans and omnivores are shown. Abbreviation: HRSD, Hamilton Rating Scale for Depression. Anxiety (continuous outcome) Vegetarian and vegan diets were associated with lower levels of anxiety (MD = −0.847; 95%CI, −1.677 to −0.018), but heterogeneity among studies was very high (I2 = 92.08; P = 0.001) (Figure 4A17,19–21,27–29). Figure 4 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on anxiety (continuous outcome). (A) Differences in anxiety (continuous outcomes) between vegetarians or vegans and omnivores; (B) Subgroup analyses by age (≤ 25 y, 26–45 y, > 45 y, or not reported); (C) Subgroup analyses by sex (women, men, women/men, or not reported); (D) Subgroup analyses by length of time the diet was followed (short-term, long-term, or not reported); (E) Subgroup analyses by type of diet (vegetarian vs vegan); (F) Subgroup analyses by instrument used (DASS, others); (G) Subgroup analyses by quality assessment score (≤ 50 vs > 50). Abbreviations: DASS, Depression Anxiety Stress Scale; DASS-A, Depression Anxiety Stress Scale–Anxiety; IDARE, Inventario de Ansiedad Rasgo-Estado (Spanish version of State-Trait Anxiety Inventory); NR, not reported; SF-36, 36-Item Short Form Health Survey; PA, physical activity. Figure 4 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on anxiety (continuous outcome). (A) Differences in anxiety (continuous outcomes) between vegetarians or vegans and omnivores; (B) Subgroup analyses by age (≤ 25 y, 26–45 y, > 45 y, or not reported); (C) Subgroup analyses by sex (women, men, women/men, or not reported); (D) Subgroup analyses by length of time the diet was followed (short-term, long-term, or not reported); (E) Subgroup analyses by type of diet (vegetarian vs vegan); (F) Subgroup analyses by instrument used (DASS, others); (G) Subgroup analyses by quality assessment score (≤ 50 vs > 50). Abbreviations: DASS, Depression Anxiety Stress Scale; DASS-A, Depression Anxiety Stress Scale–Anxiety; IDARE, Inventario de Ansiedad Rasgo-Estado (Spanish version of State-Trait Anxiety Inventory); NR, not reported; SF-36, 36-Item Short Form Health Survey; PA, physical activity. Results of the subgroup analysis confirmed these lower levels of anxiety in vegetarians/vegans aged 26 to 45 years (MD = −3.144; 95%CI, −4.727 to −1.561; Figure 4B17,19,21,27–29), in studies that included predominantly women (MD = −0.744; 95%CI, −1.461 to −0.026; Figure 4C17,19–21,27–29), in studies that used an assessment tool other than DASS-A (MD = −5.940; 95%CI, −7.704 to −4.175; Figure 4F17,19–21,27–29), and in studies with lower quality assessment scores (MD = −3.144; 95%CI, −4.727 to −1.561; Figure 4G17,19–21,27–29). In contrast, higher levels of anxiety were detected in vegetarian/vegans under 26 years of age (MD = 0.901; 95%CI, 0.143–1.658; Figure 4B) and in studies with higher quality assessment scores (MD = 0.909; 95%CI, 0.158–1.660; Figure 4G). Subgroup analyses of the length of time the vegan or vegetarian diet was followed did not show any significant differences when compared with the control group (Figure 4D,19,21,27–29,Figure 4E17,19–21,27–29). Heterogeneity was more pronounced in those under 26 years of age (I2 = 92.27; P < 0.001), in studies that included women and men (I2 = 93.87; P < 0.001), in studies that did not report the length of time the diet was followed (I2 = 94.94; P < 0.001), in studies that included vegetarians (I2 = 93.35; P < 0.001), in studies that used DASS as an assessment instrument (I2 = 91.25; P < 0.001), and in studies with higher quality assessment scores (I2 = 89.80; P < 0.001). Stress (continuous outcome) Vegetarian and vegan diets did not show any statistically significant associations with stress (MD = −0.422; 95%CI, −1.823 to 0.979). Heterogeneity among studies that examined stress was very high (I2 = 82.71; P = 0.001) (Figure 5A19–21,28,29). Figure 5 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on stress (continuous outcome). (A) Differences in stress (continuous outcome) between vegetarians or vegans and omnivores; (B) Subgroup analyses by age (≤ 25 y, 25–45 y, >45 y, or not reported); (C) Subgroup analyses by sex (women, men, women/men, or not reported); (D) Subgroup analyses by type of diet (vegetarian vs vegan); (E) Subgroup analyses by quality assessment score (< 50 vs ≥ 50). Abbreviation: DASS-S, Depression Anxiety Stress Scale–Stress. Figure 5 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on stress (continuous outcome). (A) Differences in stress (continuous outcome) between vegetarians or vegans and omnivores; (B) Subgroup analyses by age (≤ 25 y, 25–45 y, >45 y, or not reported); (C) Subgroup analyses by sex (women, men, women/men, or not reported); (D) Subgroup analyses by type of diet (vegetarian vs vegan); (E) Subgroup analyses by quality assessment score (< 50 vs ≥ 50). Abbreviation: DASS-S, Depression Anxiety Stress Scale–Stress. Results of the subgroup analysis showed lower levels of stress in vegetarians/vegans aged 26 to 45 years than in omnivores aged 26 to 75 years (MD = −2.178; 95%CI, −3.538 to −0.818), and in studies with low quality assessment scores (MD = −2.178; 95%CI, −3.538 to −0.818; Figure 5E19–21,28,29). In contrast, higher levels of stress were found in younger vegetarians/vegans than in younger omnivores (MD = 1.033; 95%CI, 0.478–1.587; Figure 5B19,21,28,29) and in studies with higher quality assessment score (MD = 1.005; 95%CI, 0.452–1.559; Figure 5E). The results did not differ by f (Figure 5C19–21,28,29), or type of diet (vegan vs vegetarian) (Figure 5D19–21,28,29). Subgroup analyses by assessment tool were not conducted because all studies used the DASS-S. Heterogeneity was more pronounced in adults aged 26 to 45 years (I2 = 49.24; P = 0.139), in studies that included mainly women (I2 = 83.14; P < 0.001), in studies that included vegetarians (I2 = 78.17; P = 0.001), and in studies with low quality assessment scores (I2 = 49.24; P = 0.139). Mental health/well-being (continuous outcome) As shown in Figure 6,17,18,24,29 being a vegetarian or vegan was not statistically significantly associated with levels of well-being (MD = −1.319; 95%CI, −2.834 to 0.197), and heterogeneity among studies was high (I2 = 73.62; P = 0.004). Figure 6 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on mental health/well-being (continuous outcome). Differences in mental health/well-being (continuous outcome) between vegetarians or vegans and omnivores. Abbreviations: PMHS, Positive Mental Health Scale; SF-36, 36-Item Short Form Health Survey. Figure 6 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on mental health/well-being (continuous outcome). Differences in mental health/well-being (continuous outcome) between vegetarians or vegans and omnivores. Abbreviations: PMHS, Positive Mental Health Scale; SF-36, 36-Item Short Form Health Survey. Other mental health outcomes A meta-analysis was not conducted on other mental health outcomes because the number of studies was insufficient (n = 1 or 2). However, Beezhold et al19,20 concluded that vegetarians reported significantly fewer negative emotions than omnivores. On the contrary, Kapoor et al25 reported a greater number of neuropsychiatric problems (psychosis and personality change) in young vegetarians than in young omnivores, and Baines et al18 found that the incidence of panic attacks or palpitations, deliberate self-harm, and other psychosomatic problems was significantly higher in vegetarians. Memory impairment/dementia Vegetarian and vegan diets did not show any statistically significant associations with memory impairment when compared with an omnivorous diet (OR = 0.825; 95%CI, 0.242–2.809), and heterogeneity among studies was very high (I2 = 63.13; P = 0.066) (Figure 7).23,25 Figure 7 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on cognitive outcomes (memory impairment/dementia, categorical outcome). Differences in memory impairment/dementia (categorical outcome) between vegetarians or vegans and omnivores are shown. Abbreviations: Ctrl, control group (omnivores); Ev, events; MMSE, Mini-Mental State Examination; Trt, treatment group (vegetarians or vegans). Figure 7 Open in new tabDownload slide Random-effects meta-analysis of the effects of vegetarian and vegan diets on cognitive outcomes (memory impairment/dementia, categorical outcome). Differences in memory impairment/dementia (categorical outcome) between vegetarians or vegans and omnivores are shown. Abbreviations: Ctrl, control group (omnivores); Ev, events; MMSE, Mini-Mental State Examination; Trt, treatment group (vegetarians or vegans). Publication bias No indication of publication bias was found for studies that included depression (Egger test, P = 0.230) and anxiety (Egger test, P = 0.324) as outcomes. Moreover, visual inspection of the funnel plot did not suggest publication bias for either depression or anxiety, as the studies were distributed symmetrically (ie, inverted funnel shape) around the summary effect size (see Figures S1 and S2 in the Supporting Information online). DISCUSSION Overall findings To date, no previous systematic review or meta-analysis has been conducted on the associations between vegetarianism or veganism and mental and cognitive outcomes. In this meta-analysis, vegetarian/vegan diets had no effect on the continuous outcomes of depression (n = 9), stress (n = 5), well-being (n = 4), or cognitive impairment (n = 3), but vegans/vegetarians were at increased risk of depression when depression was assessed as a categorical variable (OR = 2.142; 95%CI, 1.105–4.148, n = 2). Moreover, vegetarian/vegans under 26 years of age showed higher levels of anxiety, while older vegetarian/vegans had lower anxiety scores (MD = −0.847; 95%CI, −1.677 to −0.018, n = 7). Heterogeneity was large, and thus subgroup analyses showed broad differences between subgroups. For all outcomes of mental health tested, higher risk was found in those under 26 years of age and in studies with higher quality assessment scores. The instruments used to assess mental health outcomes often revealed no differences between vegetarian/vegans and omnivores, and lower anxiety levels were found only when the specific DASS-A tool was not used. Sex-specific patterns were inconsistent. For example, more symptoms of depression were seen only in female vegetarians/vegans, while fewer anxiety symptoms were found only in studies that included predominantly women above the age of 26 years. A lower level of well-being was observed only in studies that included both men and women. Finally, no differences were detected depending on whether vegans or vegetarians were included or on the length of time the diet was followed. Vegetarian and vegan diets have been found to have a protective effect against many chronic diseases such as heart disease, hypertension, type 2 diabetes, obesity, and some cancers such as colorectal and prostate cancer.81 This might be attributable to a greater intake of dietary fiber, polyunsaturated fats, vitamin C, and bioactive molecules and a lower intake of saturated fats. All of these nutrients act directly or indirectly (via other diseases) to reduce inflammation.82 Moreover, vegetarians and vegans usually report higher levels of physical activity and lower levels of sedentarism, alcohol intake, and tobacco consumption, all of which indicate a healthier lifestyle.83 Nevertheless, the current meta-analysis indicates the potential for negative mental health outcomes in individuals that adopt a vegetarian/vegan diet. Despite the benefits of vegetarian/vegan diets, several investigations have found associations between nutrient deficiencies among vegetarians and vegans18,84,85 and poorer mental health outcomes.18,22,25,26 Subgroup analyses showed clear age-dependent patterns: negative effects of a vegetarian/vegan diet on mental health outcomes at a younger age were identified, while the opposite effects were observed in those older than 25 years. This might reflect the greater vulnerability of young people to nutritional deficiencies, since their brain and personality are still developin.86 While women have consistently higher prevalence rates of anxiety and depression than men, no such consistent sex-related patterns appeared in this review. Similarly, vegan individuals are often at greater risk for nutrient deficiencies, but again, no higher risk was detected in vegans compared with vegetarians in subgroup analyses. Since information on the duration of adherence to a specific type of diet was often imprecise or absent, the lack of differences in subgroup analysis of the duration of diet adherence is not surprising. Possible mechanisms involved in the associations between vegetarian/vegan diets and mental health outcomes Large heterogeneity was found among studies that examined the relationship between vegetarianism or veganism and mental health outcomes. Although several nonsignificant results were found, some significant findings were identified, such as a higher risk of depression and a lower risk of anxiety (although this effect was age dependent). In general, higher-quality studies and studies in populations younger than 26 years suggested a higher risk of all types of mental disorders in vegetarians and vegans compared with omnivores. One possible explanation for these associations is a reverse causal relationship between vegetarianism or veganism and mental health outcomes. Individuals who have psychosocial disorders may be more eager to follow a vegetarian or vegan diet in order to improve their mental health. In a German study, the age at which a diet is adopted showed that a vegetarian diet was adopted after the onset of mental disorders, thus confirming the possibility of reverse causation.42 Still, 1 intervention study reported an improvement in depression, anxiety, and mood after adoption of a vegan diet,17,20 while another found improvements in stress but no significant changes in depression, anxiety, or mood in omnivores who adopted a vegetarian diet for 2 weeks.19 These intervention studies, however, were conducted for only a short period of time (less than 4 months), and effects over a longer period have not yet been investigated. Since mental health disorders often develop at a young age, the subgroup analysis showing higher risk in those younger than 26 years of age might reflect this reverse causation, apart from the greater vulnerability of young people to nutritional deficiencies. Other possible explanations from the literature include nutrient deficiencies common in vegan diets (eg, certain amino acids, long-chain omega-3 fatty acids, vitamins B6, and B12, zinc, creatine, and even cholesterol) that could accelerate or worsen preexisting mental conditions.8 All these deficiencies have been linked to a higher risk of mental health disorders.87,88 Several amino acids, such as methionine, tryptophan, lysine, arginine, tyrosine, and β-alanine, can have a protective effect against depression and anxiety,89,90 as some neurotransmitters derived from these amino acids, eg, dopamine and serotonin, are important for mood regulation.91 Still, any effects are nuanced, since the levels of methionine, tryptophan, and tyrosine are highest in fish eaters and vegetarians, followed by meat eaters, but are lowest in vegans.92 Thus, only in vegans could negative health effects be attributed to amino acid deficiencies. Another possible explanation for negative mental health outcomes could be that, in many populations, vegetarian or vegan diets are consumed by a minority of individuals (at least until recently, since the prevalence of veganism/vegetarianism has skyrocketed in recent years), and being part of a minority group can induce feelings of reduced well-being.22 Possible mechanisms involved in changes in cognitive outcomes No statistically significant association between a vegetarian/vegan diet and cognitive outcomes was found in the 2 studies that examined this: Giem et al23 found a trend toward delayed onset of dementia in vegetarians, while Kapoor et al25 found a higher incidence of neuropsychiatric and neurological disorders such as memory impairment, personality change, and psychosis in vegetarians. Studies that were not included in this meta-analysis (because they did not include a purely vegetarian or vegan diet) also reported conflicting results: Xu et al48 found a higher risk of memory impairment and dementia in those who were defined as vegetarians, while Avgerinos et al14 reported that vegetarians performed better than meat eaters in memory tasks, although no other differences were observed for other cognitive domains,14 and Louwman et al13 found that adolescents who consumed macrobiotic diets (similar to vegan diets, except for occasional fish consumption) performed worse on most cognitive tests than omnivores.13 The apparently contradictory results might be explained by differences in vitamin B12 values. Vegetarians had lower vitamin B12 status (and 50% had a deficiency) compared with omnivores in a study that found higher risk of cognitive impairment for vegetarians. In the study of the macrobiotic diet,13 low vitamin B12 levels were indeed associated with lower performance in certain cognitive tests, but not all individuals who consumed the macrobiotic diet had vitamin B12 deficiency. The other studies did not report serum vitamin B12 profiles.23,48 A second possible mechanism of the effect of vegetarian/vegan diets on cognitive outcomes is related to the role of phytoestrogens in cognitive function. Phytoestrogens, the most bioactive components of soy (a product eaten mainly by vegetarians and vegans), seem to have a neuroprotective effect.93 Creatine is another nutrient that plays a critical role in brain development and function. Some investigators have hypothesized that it aids cognition by improving energy supply and neuroprotection. Creatine is a peptide found mostly in meat, fish, and other animal products, and the levels of creatine in muscle are known to be lower in vegetarians. After supplementation with creatine, the memory of vegetarians was better than that of meat eaters.51 However, memory at baseline did not differ depending by dietary pattern, so any hypothesized creatine deficiency in vegetarians did not influence memory; rather, vegetarians were found to be more sensitive to supplementation with creatine. Strengths and weaknesses To the best of knowledge, this is the first meta-analysis to examine the association between vegetarianism/veganism and mental health and cognitive outcomes. A strength of this meta-analysis is that these outcomes were interpreted quite broadly while considering both categorical and continuous variables. This study also included many subgroup analyses (by diet, age, sex, instrument of assessment, length of time the diet was followed, and quality assessment score) to detect heterogeneity that might reflect more vulnerable subgroups and methodological issues. This allowed the total effect of the vegan and vegetarian diets relative to mental health and cognitive outcomes to be estimated with a larger sample size, despite the large heterogeneity in the included studies. A first limitation is that no studies in this meta-analysis included children, and most of the studies included a higher percentage of women than men. Secondly, there were only a few interventional studies, some of which had a low quality score and a short intervention period (ie, 2 weeks), which hindered the ability to detect long-term consequences on cognition or mental health. Although many studies have described important differences in lifestyle among vegetarians and vegans compared with omnivores (lower body mass index, higher levels of physical activity, and lower levels of sedentarism, alcohol intake, and tobacco consumption), only 2 studies in the present meta-analysis accounted for key potential confounders.23,29 Consequently, the meta-analysis was conducted on raw data only. Nevertheless, adjustment for confounders did not drastically change results in these 2 studies. Moreover, many relevant studies were finally discarded because the term vegan or vegetarian was used inaccurately (ie, participants who ate fish or poultry were included) or because they included participants who self-reported to be vegetarians or vegans but who also reported occasional consumption of meat and fish in the food frequency questionnaire.30,40,42,48 The wide variety in the assessment instruments used and the approaches to define vegetarians and vegans limits study comparability as well as power to find significant relationships. Although it might be interesting to perform subgroup analyses in pesco-vegetarians (to see whether outcomes might be due to fish omission), only 3 studies identified pesco-vegetarians, and this population was always a minority combined with those who consumed a purely vegetarian diet. Another outcome that was discarded in the current systematic review was eating disorders as mental health outcome. Although some studies reported vegetarians and vegans to be more likely than omnivores to have eating disorders,39,68 these were discarded because vegetarian/vegan diets can camouflage an existing eating disorder.94 Finally, subgroup analyses were based on a small number of studies and might thus be biased or lack power. Implications for public health The necessary intakes of protein, fat, carbohydrate, vitamins, and minerals within vegetarian and vegan diets for optimal health is still under investigation. Vegetarians and, in particular, vegans may require supplementation, as some nutrients may not be adequately available from plant sources. Some supplements (eg, vitamin B12, zinc, and creatine) to improve short-term memory and intelligence/reasoning may help in very restricted diets. Otherwise, a well-chosen plant-based diet provides all the necessary protein, fats, carbohydrates, vitamins, and minerals for optimal health. Apart from dietary patterns that strictly omit animal-based protein, other dietary patterns can be adjusted on a continuous scale by just limiting the amount of animal-based protein consumed.95 The terms semivegetarianism, flexitarianism, and plant-based diets reflect dietary patterns in which consumption of animal-based proteins is limited,96 which might confer positive health effects without resulting in the deficiencies sometimes associated with a vegetarian diet. A 6% to 9% reduction in the risk of all-cause mortality and cardiovascular disease mortality was observed when 3% of energy from animal protein was replaced with plant protein.97 Implications for further research It is difficult to draw conclusions from a meta-analysis when the number of included studies is low, when different definitions or assessment tools are used, or when the quality of the included studies is low. A major flaw in current literature on this topic is the lack of adjustment for confounders. Future studies should adjust for sociodemographic factors, physical activity, alcohol intake, tobacco consumption, weight status, and medical history. Before causal conclusions can be drawn, additional well-designed intervention studies are needed, as only 3 interventional studies could be identified, and these studies used short-term interventions, described differences in participants at baseline, or reported a low rate of adherence to the intervention. To identify causal pathways, studies should measure the baseline nutrient status of participants, as deficiencies are frequent in vegetarians and vegans. Moreover, studies should examine different gradients of a plant-based diet by, for example, exploring the role of fish (with omega-3 fats), milk, and eggs and the quality of plant-based foods (since omission of meat or any other animal-product does not guarantee highly nutritious food choices). Finally, the duration of diet adherence, which is seldom examined, should be scrutinized in future studies. CONCLUSION A vegan or vegetarian diet was associated with a higher risk of depression and lower anxiety scores, but no differences for other outcomes were found. Subgroup analyses for anxiety showed numerous differences, with a greater risk of anxiety found mainly in individuals under 26 years of age who followed a vegan or vegetarian diet and in studies with higher quality assessment scores. The large heterogeneity among the studies included in this meta-analysis prevents definitive conclusions from being drawn. More studies on the effects of vegan or vegetarian diets on mental health, especially cognitive outcomes, with overall better quality are needed before clear positive or negative associations can be confirmed. Acknowledgments I.I. takes responsibility for all aspects of the reliability, freedom from bias, and interpretation of the data presented. Author contributions. I.I. and N.M conducted the literature search and agreed on inclusion and exclusion of studies. I.I. extracted data and performed the meta-analysis. I.I drafted the manuscript. I.I., N.M., I.H., and L.M revised and provided intellectual input into the final manuscript. Funding/support. N.M performed this systematic review while receiving financial support from the Alpro Foundation, Welvelgem, Belgium. The Alpro Foundation was not involved in the conception, design, performance, or approval of this work. Declaration of interest. The authors have no relevant interests to declare. Supporting Information The following Supporting Information is available through the online version of this article at the publisher’s website: Appendix S1 PRISMA checklist Table S1 Quality assessment for observational cohort and cross-sectional studies Table S2 Quality assessment for controlled intervention studies Figure S1 Funnel plot of vegetarian and vegan diets and depression. The effect estimate line is shown as a dashed line Figure S2 Funnel plot of vegetarian and vegan diets and anxiety. The effect estimate line is shown as a dashed line References 1 Medalia A , Erlich M. Why cognitive health matters . Am J Public Health. 2017 ; 107 : 45 – 47 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Wimo A , Guerchet M, Ali GC, et al. . The worldwide costs of dementia 2015 and comparisons with 2010 . Alzheimers Dement . 2017 ; 13 : 1 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Prince M , Bryce R, Albanese E, et al. . The global prevalence of dementia: a systematic review and metaanalysis . Alzheimers Dement . 2013 ; 9 : 63 – 75.e62 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Parletta N , Milte CM, Meyer BJ. Nutritional modulation of cognitive function and mental health . J Nutr Biochem . 2013 ; 24 : 725 – 743 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Widmer RJ , Flammer AJ, Lerman LO, et al. . The Mediterranean diet, its components, and cardiovascular disease . Am J Med . 2015 ; 128 : 229 – 238 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Petersson SD , Philippou E. Mediterranean diet, cognitive function, and dementia: a systematic review of the evidence123 . Adv Nutr . 2016 ; 7 : 889 – 904 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Dinu M , Abbate R, Gensini GF, et al. . Vegetarian, vegan diets and multiple health outcomes: a systematic review with meta-analysis of observational studies . Crit Rev Food Sci Nutr . 2017 ; 57 : 3640 – 3649 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Craig WJ. Health effects of vegan diets . Am J Clin Nutr . 2009 ; 89 : 1627S – 1633S . Google Scholar Crossref Search ADS PubMed WorldCat 9 Gilsing AM , Crowe FL, Lloyd-Wright Z, et al. . Serum concentrations of vitamin B12 and folate in British male omnivores, vegetarians and vegans: results from a cross-sectional analysis of the EPIC-Oxford cohort study . Eur J Clin Nutr. 2010 ; 64 : 933 – 939 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Kornsteiner M , Singer I, Elmadfa I. Very low n-3 long-chain polyunsaturated fatty acid status in Austrian vegetarians and vegans . Ann Nutr Metab. 2008 ; 52 : 37 – 47 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Delanghe J , De Slypere JP, De Buyzere M, et al. . Normal reference values for creatine, creatinine, and carnitine are lower in vegetarians . Clin Chem . 1989 ; 35 : 1802 – 1803 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Moore E , Mander A, Ames D, et al. . Cognitive impairment and vitamin B12: a review . Int Psychogeriatr. 2012 ; 24 : 541 – 556 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Louwman MW , van Dusseldorp M, van de Vijver FJ, et al. . Signs of impaired cognitive function in adolescents with marginal cobalamin status . Am J Clin Nutr . 2000 ; 72 : 762 – 769 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Avgerinos KI , Spyrou N, Bougioukas KI, et al. . Effects of creatine supplementation on cognitive function of healthy individuals: a systematic review of randomized controlled trials . Exp Gerontol . 2018 ; 108 : 166 – 173 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Gibson RS , Heath AL, Szymlek-Gay EA. Is iron and zinc nutrition a concern for vegetarian infants and young children in industrialized countries? Am J Clin Nutr . 2014 ; 100(suppl 1 ): 459S – 468S . Google Scholar Crossref Search ADS PubMed WorldCat 16 Li Z , Li B, Song X, et al. . Dietary zinc and iron intake and risk of depression: a meta-analysis . Psychiatry Res . 2017 ; 251 : 41 – 47 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Agarwal U , Mishra S, Xu J, et al. . A multicenter randomized controlled trial of a nutrition intervention program in a multiethnic adult population in the corporate setting reduces depression and anxiety and improves quality of life: the GEICO study . Am J Health Promot. 2015 ; 29 : 245 – 254 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Baines S , Powers J, Brown WJ. How does the health and well-being of young Australian vegetarian and semi-vegetarian women compare with non-vegetarians? Public Health Nutr. 2007 ; 10 : 436 – 442 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Beezhold BL , Johnston CS, Daigle DR. Vegetarian diets are associated with healthy mood states: a cross-sectional study in Seventh Day Adventist adults . Nutr J. 2010 ; 9 : 26 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Beezhold BL , Johnston CS. Restriction of meat, fish, and poultry in omnivores improves mood: a pilot randomized controlled trial . Nutr J. 2012 ; 11 : 9 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Beezhold B , Radnitz C, Rinne A, et al. . Vegans report less stress and anxiety than omnivores . Nutr Neurosci . 2015 ; 18 : 289 – 296 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Forestell CA , Nezlek JB. Vegetarianism, depression, and the five factor model of personality . Ecol Food Nutr . 2018 ; 57 : 246 – 259 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Giem P , Beeson WL, Fraser GE. The incidence of dementia and intake of animal products: preliminary findings from the Adventist Health Study . Neuroepidemiology . 1993 ; 12 : 28 – 36 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Katcher HI , Ferdowsian HR, Hoover VJ, et al. . A worksite vegan nutrition program is well-accepted and improves health-related quality of life and work productivity . Ann Nutr Metab. 2010 ; 56 : 245 – 252 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Kapoor A , Baig M, Tunio SA, et al. . Neuropsychiatric and neurological problems among Vitamin B12 deficient young vegetarians . Neurosciences (Riyadh). 2017 ; 22 : 228 – 232 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Lindeman M. The state of mind of vegetarians: psychological well-being or distress? Ecol Food Nutr . 2002 ; 41 : 75 – 86 . Google Scholar Crossref Search ADS WorldCat 27 Rodríguez Jiménez J , Rodríguez JR, González MJ. Indicators of anxiety and depression in subjects with different kinds of diet: vegetarians and omnivores [in Spanish] . Bol Asoc Med P R . 1998 ; 90 : 58 – 68 . PMID: 9866269 Google Scholar PubMed OpenURL Placeholder Text WorldCat 28 Timko CA , Hormes JM, Chubski J. Will the real vegetarian please stand up? An investigation of dietary restraint and eating disorder symptoms in vegetarians versus non-vegetarians . Appetite . 2012 ; 58 : 982 – 990 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Velten J , Bieda A, Scholten S, et al. . Lifestyle choices and mental health: a longitudinal survey with German and Chinese students . BMC Public Health . 2018 ; 18 : 632 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Hibbeln JR , Northstone K, Evans J, et al. . Vegetarian diets and depressive symptoms among men . J Affect Disord . 2018 ; 225 : 13 – 17 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Fonnebo V. The healthy Seventh-Day Adventist lifestyle: what is the Norwegian experience? Am J Clin Nutr . 1994 ; 59(suppl 5 ): 1124S – 1129S . doi:10.1093/ajcn/59.5.1124S Google Scholar Crossref Search ADS WorldCat 32 Leitzmann C. Vegetarian nutrition: past, present, future . Am J Clin Nutr . 2014 ; 100(suppl 1 ): 496S – 502S . Google Scholar Crossref Search ADS PubMed WorldCat 33 Edelstein S, ed. Food Science: an Ecological Approach . 1st ed. Burlington, MA : Jones & Bartlett Learning ; 2014 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 34 Liberati A , Altman DG, Tetzlaff J, et al. . The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration . BMJ . 2009 ; 339 : b2700 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Hutton B , Salanti G, Caldwell DM, et al. . The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations . Ann Intern Med. 2015 ; 162 : 777 – 784 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Boudin F , Nie JY, Bartlett JC, et al. . Combining classifiers for robust PICO element detection . BMC Med Inform Decis Mak. 2010 ; 10 : 29 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies. NIH National Heart, Lung, and Blood Institute website. https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools. Accessed September 30, 2018. 38 Quality Assessment Tool of Controlled Intervention Studies. NIH National Heart, Lung, and Blood Institute website. https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools. Accessed August 20, 2018. 39 Baş M , Karabudak E, Kiziltan G. Vegetarianism and eating disorders: association between eating attitudes and other psychological factors among Turkish adolescents . Appetite . 2005 ; 44 : 309 – 315 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Burkert NT , Muckenhuber J, Großschädl F, et al. . Nutrition and health—the association between eating behavior and various health parameters: a matched sample study . PLoS One . 2014 ; 9 : e88278 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Hosseinzadeh M , Vafa M, Esmaillzadeh A, et al. . Empirically derived dietary patterns in relation to psychological disorders . Public Health Nutr. 2016 ; 19 : 204 – 217 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Michalak J , Zhang XC, Jacobi F. Vegetarian diet and mental disorders: results from a representative community survey . Int J Behav Nutr Phys Act. 2012 ; 9 : 67 . Google Scholar Crossref Search ADS PubMed WorldCat 43 Thygesen LC , Gimsing LN, Bautz A, et al. . Chronic neurodegenerative illnesses and epilepsy in Danish Adventists and Baptists: a nationwide cohort study . J Alzheimers Dis . 2017 ; 56 : 1429 – 1435 . Google Scholar Crossref Search ADS PubMed WorldCat 44 Herens MC , Dagnelie PC, Kleber RJ, et al. . Nutrition and mental development of 4–5‐year‐old children on macrobiotic diets . J Hum Nutr Diet . 1992 ; 5 : 1 – 78 . Google Scholar Crossref Search ADS WorldCat 45 Northstone K , Joinson C, Emmett P. Dietary patterns and depressive symptoms in a UK cohort of men and women: a longitudinal study . Public Health Nutr. 2018 ; 21 : 831 – 837 . Google Scholar Crossref Search ADS PubMed WorldCat 46 Salehi-Abargouei A , Esmaillzadeh A, Azadbakht L, et al. . Do patterns of nutrient intake predict self-reported anxiety, depression and psychological distress in adults? SEPAHAN study . Clin Nutr . 2019 ; 38 : 940 – 947 . doi:10.1016/j.clnu.2018.02.002 Google Scholar Crossref Search ADS PubMed WorldCat 47 Sánchez-Villegas A , Henríquez-Sánchez P, Ruiz-Canela M, et al. . A longitudinal analysis of diet quality scores and the risk of incident depression in the SUN Project . BMC Med. 2015 ; 13 : 197 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Xu X , Xiao S, Rahardjo TB, et al. . Tofu intake is associated with poor cognitive performance among community-dwelling elderly in China . J Alzheimers Dis . 2014 ; 43 : 669 – 675 . Google Scholar Crossref Search ADS WorldCat 49 Burkert NT , Freidl W, Großschädel F, et al. . Nutrition and health: different forms of diet and their relationship with various health parameters among Austrian adults . Wien Klin Wochenschr. 2014 ; 126 : 113 – 118 . Google Scholar Crossref Search ADS PubMed WorldCat 50 Larsen PS , Nybo Andersen AM, Uldall P, et al. . Maternal vegetarianism and neurodevelopment of children enrolled in the Danish National Birth Cohort . Acta Paediatr. 2014 ; 103 : e507 – e509 . Google Scholar Crossref Search ADS PubMed WorldCat 51 Benton D , Donohoe R. The influence of creatine supplementation on the cognitive functioning of vegetarians and omnivores . Br J Nutr. 2011 ; 105 : 1100 – 1105 . Google Scholar Crossref Search ADS PubMed WorldCat 52 Cook R , Benton D. The relationship between diet and mental health . Pers Indiv Differ . 1993 ; 14 : 397 – 403 . Google Scholar Crossref Search ADS WorldCat 53 Crichton GE , Bryan J, Hodgson JM, et al. . Diet adherence and self-reported psychological functioning in an Australian sample . Appetite . 2013 ; 70 : 53 – 59 . Google Scholar Crossref Search ADS PubMed WorldCat 54 Leung B , Lauche R, Leach M, et al. . Special diets in modern America: analysis of the 2012 National Health Interview Survey data . Nutr Health. 2018 ; 24 : 11 – 18 . Google Scholar Crossref Search ADS PubMed WorldCat 55 McLean JA , Barr SI. Cognitive dietary restraint is associated with eating behaviors, lifestyle practices, personality characteristics and menstrual irregularity in college women . Appetite . 2003 ; 40 : 185 – 192 . Google Scholar Crossref Search ADS PubMed WorldCat 56 von Gunten A , Mosimann UP, Antonietti JP. A longitudinal study on delirium in nursing homes . Am J Geriatr Psychiatry . 2013 ; 21 : 963 – 972 . Google Scholar Crossref Search ADS PubMed WorldCat 57 Wills WJ. Food and eating practices during the transition from secondary school to new social contexts . J Youth Stud . 2005 ; 8 : 97 – 110 . Google Scholar Crossref Search ADS WorldCat 58 Yazigi Solis M , de Salles Painelli V, Giannini Artioli G, et al. . Brain creatine depletion in vegetarians? A cross-sectional (1)H-magnetic resonance spectroscopy 1H-magnetic resonance spectroscopy (1H-MRS) study . Br J Nutr. 2014 ; 111 : 1272 – 1274 . Google Scholar Crossref Search ADS PubMed WorldCat 59 Yackobovitch-Gavan M , Golan M, Valevski A, et al. . An integrative quantitative model of factors influencing the course of anorexia nervosa over time . Int J Eat Disord. 2009 ; 42 : 306 – 317 . Google Scholar Crossref Search ADS PubMed WorldCat 60 Craig BA , Morton DP, Morey PJ, et al. . The association between self-rated health and social environments, health behaviors and health outcomes: a structural equation analysis . BMC Public Health . 2018 ; 18 : 440 . Google Scholar Crossref Search ADS PubMed WorldCat 61 Bardone-Cone AM , Fitzsimmons-Craft EE, Harney MB, et al. . The inter-relationships between vegetarianism and eating disorders among females . J Acad Nutr Diet . 2012 ; 112 : 1247 – 1252 . Google Scholar Crossref Search ADS PubMed WorldCat 62 Bobic J , Cvijetic S, Baric IC, et al. . Personality traits, motivation and bone health in vegetarians . Coll Antropol . 2012 ; 36 : 795 – 800 . PMID: 23213935 Google Scholar PubMed OpenURL Placeholder Text WorldCat 63 Cooper CK , Wise TN, Mann L. Psychological and cognitive characteristics of vegetarians . Psychosomatics . 1985 ; 26 : 521 – 523 , 526 – 527 . Google Scholar Crossref Search ADS PubMed WorldCat 64 Dwyer JT , Miller LG, Arduino NL, et al. . Mental age and I.Q. of predominantly vegetarian children . J Am Diet Assoc. 1980 ; 76 : 142 – 147 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 65 Link LB , Hussaini NS, Jacobson JS. Change in quality of life and immune markers after a stay at a raw vegan institute: a pilot study . Complement Ther Med . 2008 ; 16 : 124 – 130 . Google Scholar Crossref Search ADS PubMed WorldCat 66 Behari M , Srivastava AK, Das RR, et al. . Risk factors of Parkinson’s disease in Indian patients . J Neurol Sci . 2001 ; 190 : 49 – 55 . Google Scholar Crossref Search ADS PubMed WorldCat 67 Zuromski KL , Witte TK, Smith AR, et al. . Increased prevalence of vegetarianism among women with eating pathology . Eat Behav . 2015 ; 19 : 24 – 27 . Google Scholar Crossref Search ADS PubMed WorldCat 68 Klopp SA , Heiss CJ, Smith HS. Self-reported vegetarianism may be a marker for college women at risk for disordered eating . J Acad Nutr Diet . 2003 ; 103 : 745 – 747 . OpenURL Placeholder Text WorldCat 69 Robinson-O’Brien R , Perry CL, Wall MM, et al. . Adolescent and young adult vegetarianism: better dietary intake and weight outcomes but increased risk of disordered eating behaviors . J Acad Nutr Diet . 2009 ; 109 : 648 – 655 . OpenURL Placeholder Text WorldCat 70 Neumark-Sztainer D , Story M, Resnick MD, et al. . Adolescent vegetarians. A behavioral profile of a school-based population in Minnesota . Arch Pediatr Adolesc Med. 1997 ; 151 : 833 – 838 . Google Scholar Crossref Search ADS PubMed WorldCat 71 Kahleova H , Hrachovinova T, Hill M, et al. . Vegetarian diet in type 2 diabetes—improvement in quality of life, mood and eating behaviour . Diabet Med . 2013 ; 30 : 127 – 129 . Google Scholar Crossref Search ADS PubMed WorldCat 72 Bakan R , Birmingham CL, Aeberhardt L, et al. . Dietary zinc intake of vegetarian and nonvegetarian patients with anorexia nervosa . Int J Eat Disord . 1993 ; 13 : 229 – 233 . Google Scholar Crossref Search ADS PubMed WorldCat 73 Stokes N , Gordon CM, DiVasta A. Vegetarian diets and mental health in adolescents with anorexia nervosa [abstract] . J Adolesc Health . 2011 ; 48(suppl ): S50 . Google Scholar Crossref Search ADS WorldCat 74 Worsley A , Skrzypiec G. Teenage vegetarianism: beauty or the beast? Nutr Res . 1997 ; 17 : 391 – 404 . Google Scholar Crossref Search ADS WorldCat 75 Chan AS , Sze SL, Woo J, et al. . A Chinese Chan-based lifestyle intervention improves memory of older adults . Front Aging Neurosci . 2014 ; 6 : 50 . doi:10.3389/fnagi.2014.00050 Google Scholar PubMed OpenURL Placeholder Text WorldCat 76 Yu R , Woo J, Chan AS, et al. . A Chinese Chan-based mind-body intervention improves psychological well-being and physical health of community-dwelling elderly: a pilot study . Clin Interv Aging . 2014 ; 9 : 727 – 736 . doi:10.2147/CIA.S59985 Google Scholar Crossref Search ADS PubMed WorldCat 77 Wolpert T. Nutritional psychology: is diet the key to psychological health? Vegetarian Times . May 1982 . https://books.google.es/books?id=eAgAAAAAMBAJ&pg=PA55&lpg=PA55&dq=nutritional+psychology:+is+diet+the+key+to+psychological+health?+wolpert&source=bl&ots=4k3QPjhnx0&sig=ACfU3U1l23i5iBN0G-KFUN4LoIuO8j2s6A&hl=es&sa=X&ved=2ahUKEwjw_86lzvLoAhVFBGMBHW_0CQcQ6AEwAXoECAgQLA#v=onepage&q=nutritional%20psychology%3A%20is%20diet%20the%20key%20to%20psychological%20health%3F%20wolpert&f=false. Accessed June 2018. OpenURL Placeholder Text WorldCat 78 Higgins JP , Altman DG, Gotzsche PC, et al. ; and Cochrane Bias Methods Group. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials . BMJ . 2011 ; 343 : d5928 . Google Scholar Crossref Search ADS PubMed WorldCat 79 Peters JL , Sutton AJ, Jones DR, et al. . Comparison of two methods to detect publication bias in meta-analysis . JAMA . 2006 ; 295 : 676 – 680 . Google Scholar Crossref Search ADS PubMed WorldCat 80 Sterne JA , Sutton AJ, Ioannidis JP, et al. . Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials . BMJ . 2011 ; 343 : d4002 . Google Scholar Crossref Search ADS PubMed WorldCat 81 Melina V , Craig W, Levin S. Position of the Academy of Nutrition and Dietetics: vegetarian diets . J Acad Nutr Diet . 2016 ; 116 : 1970 – 1980 . Google Scholar Crossref Search ADS PubMed WorldCat 82 Ambroszkiewicz J , Chełchowska M, Rowicka G, et al. . Anti-inflammatory and pro-inflammatory adipokine profiles in children on vegetarian and omnivorous diets . Nutrients . 2018 ; 10 : 1241 . Google Scholar Crossref Search ADS WorldCat 83 Alewaeters K , Clarys P, Hebbelinck M, et al. . Cross-sectional analysis of BMI and some lifestyle variables in Flemish vegetarians compared with non-vegetarians . Ergonomics . 2005 ; 48 : 1433 – 1444 . Google Scholar Crossref Search ADS PubMed WorldCat 84 Iguacel I , Miguel-Berges ML, Gómez-Bruton A, et al. . Veganism, vegetarianism, bone mineral density, and fracture risk: a systematic review and meta-analysis . Nutr Rev . 2019 ; 77 : 1 – 18 . Google Scholar Crossref Search ADS PubMed WorldCat 85 Davey GK , Spencer EA, Appleby PN, et al. . EPIC-Oxford: lifestyle characteristics and nutrient intakes in a cohort of 33 883 meat-eaters and 31 546 non meat-eaters in the UK . Public Health Nutr. 2003 ; 6 : 259 – 269 . Google Scholar Crossref Search ADS PubMed WorldCat 86 Arain M , Haque M, Johal L, et al. . Maturation of the adolescent brain . Neuropsychiatr Dis Treat. 2013 ; 9 : 449 – 461 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 87 Sathyanarayana Rao TS , Asha MR, Ramesh BN, et al. . Understanding nutrition, depression and mental illnesses . Indian J Psychiatry. 2008 ; 50 : 77 – 82 . Google Scholar Crossref Search ADS PubMed WorldCat 88 Hvas AM , Juul S, Bech P, et al. . Vitamin B6 level is associated with symptoms of depression . Psychother Psychosom. 2004 ; 73 : 340 – 343 . Google Scholar Crossref Search ADS PubMed WorldCat 89 De Berardis D , Orsolini L, Serroni N, et al. . A comprehensive review on the efficacy of S-adenosyl-L-methionine in major depressive disorder . CNS Neurol Disord Drug Targets . 2016 ; 15 : 35 – 44 . Google Scholar Crossref Search ADS PubMed WorldCat 90 Aucoin M , LaChance L, Cooley K, et al. . Diet and psychosis: a scoping review . Neuropsychobiology . 2020 ; 79 : 20 – 23 . Google Scholar Crossref Search ADS PubMed WorldCat 91 Fernstrom JD. Dietary amino acids and brain function . J Acad Nutr Diet . 1994 ; 94 : 71 – 77 . OpenURL Placeholder Text WorldCat 92 Schmidt JA , Rinaldi S, Scalbert A, et al. . Plasma concentrations and intakes of amino acids in male meat-eaters, fish-eaters, vegetarians and vegans: a cross-sectional analysis in the EPIC-Oxford cohort . Eur J Clin Nutr. 2016 ; 70 : 306 – 312 . Google Scholar Crossref Search ADS PubMed WorldCat 93 Soni M , Rahardjo TB, Soekardi R, et al. . Phytoestrogens and cognitive function: a review . Maturitas . 2014 ; 77 : 209 – 220 . Google Scholar Crossref Search ADS PubMed WorldCat 94 Perry CL , McGuire MT, Neumark-Sztainer D, et al. . Characteristics of vegetarian adolescents in a multiethnic urban population . J Adolesc Health . 2001 ; 29 : 406 – 416 . Google Scholar Crossref Search ADS PubMed WorldCat 95 Satija A , Hu FB. Plant-based diets and cardiovascular health . Trends Cardiovasc Med . 2018 ; 28 : 437 – 441 . Google Scholar Crossref Search ADS PubMed WorldCat 96 Corrin T , Papadopoulos A. Understanding the attitudes and perceptions of vegetarian and plant-based diets to shape future health promotion programs . Appetite . 2017 ; 109 : 40 – 47 . Google Scholar Crossref Search ADS PubMed WorldCat 97 Song M , Fung TT, Hu FB, et al. . Animal and plant protein intake and all-cause and cause-specific mortality: results from two prospective US cohort studies . JAMA Intern Med. 2016 ; 176 : 1453 – 1463 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2020. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For permissions, please e-mail: [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)