The Effects of Vegetarian Diets on Mental Health

The last decade has seen portions of the global population choosing to go vegetarian for health reasons, among them cited mental wellbeing (Velten et al., 2018). Though these choices are validly founded on the benefits of a vegetarian diet on physical health, there is conflicting data on whether a vegetarian diet is indeed healthy for the mind. Michalak, Zhang & Jacobi (2012) demonstrated that vegetarians showed a higher prevalence of mental health disorders, and Matta et al. (2018) showed that the probability of mental disorder symptoms increased with every food group excluded from the diet. On the other hand, Beezhold et al. (2015) showed that vegetarians and vegans showed lower levels of stress and anxiety than meat-eaters, and these results were backed by Beezhold, Johnston & Daigle (2010) who found that vegetarian diets were associated with healthier mood states than omnivore diets.

With so many people choosing vegetarian diets for health reasons, it is paramount that the scientific community come to a consensus on whether or not vegetarian diets are healthy for the mind. The present study will seek to elucidate, using correlational and experimental research designs, whether a) there is a correlation between vegetarian diets and incidence of mental health disorders and b) whether mental disorder symptoms increase with the exclusion of animal-product food groups.

Vegetarian diets cause an increase in mental health disorders.

The purpose of this study is to find out the correlation between vegetarian diets and the incidence of mental health disorders.

For ease of study, the population will be the entirety of the student body at James Cook University, which is about 15,197 students. The sample for this correlational study will be randomly selected from among the student body, with an open recruitment policy based only on the participant’s diet (vegetarian or non-vegetarian). Volunteers will be invited from every college and academic year, with an incentive for participation: either course credits or gift cards. The target sample size is 250 students, which will represent about 1.6% of the student population. The sample will be divided according to diet: totally vegetarian, mostly vegetarian and non-vegetarian.

All participants will be asked to complete a form detailing their demographic data: age, gender and college of study. Their diet will be assessed using the question: “Do you currently follow a vegetarian diet, or have you followed a vegetarian diet in the past?” Respondents can answer either “yes, totally”, “yes, mostly”, or “no”. They will also be asked to detail when, if at all, the vegetarian diet was adopted or abandoned.

Psychiatric diagnoses will be conducted using the Diagnostic and Statistical Manual of Mental Disorders (DSM), fifth edition (DSM-V) (American Psychiatric Association, 2013), of which an abridged version will be developed for this study. However, it is note-worthy that recent studies have called attention to low test-retest reliability in DSM-V field trials (Spitzer, Williams & Endicott, 2012; Regier et al., 2013). In consideration of these concerns, the Munich-Composite International Diagnostic Interview (M-CIDI), which has shown good to excellent kappa values of reliability (Chmielewski et al., 2015): 0.83 for alcohol abuse and 0.55 for drug abuse (Lachner et al., 1998), will be used as the guideline for the DSM-V interview to be used in this study.

The assistance of clinically-trained psychologists will be recruited for the diagnostics, and the interviews will be kept to a maximum of one hour. The following diagnostic groups will constitute the focus of the study: anxiety, depressive, feeding and eating disorders, alcohol, somatic symptoms and related disorders.

The onset data for the above-listed disorders will be assessed, including the age and the dietary status at the time of onset. Prevalence rates will be assessed for three months, twelve months, and the respondents’ lifetime.

All participants will be asked to provide written, informed consent. It will also be made clear that the results of the psychological diagnostics are purely for research purposes, and participants would have to see their physicians for actionable mental health diagnoses and treatment.

The sample will be characterised by diet type using descriptive statistics and represented as percentages or means ± standard deviations. A bivariate (Pearson) correlation analysis will be conducted using the Statistical Package for the Social Sciences (SPSS) software. A scatter plot will be created to visualise the relationship between vegetarian diet and prevalence of mental health disorders, to verify that they have a linear relationship. Since this method runs the risk of creating a type II error in which the null hypothesis (that there is no correlation between the two variables), outliers will be removed to clean up the data (Osborne & Waters, 2002). The bivariate correlation coefficient will then be calculated with the two-tailed test of significance.

The Pearson correlation coefficient quantitatively assesses the relationship between two variables, and its value ranges between -1 and +1. A positive correlation coefficient indicates a positive relationship between the two variables, a negative value indicates a negative relationship, and a value of zero indicates that the two variables are independent and show no correlation in the population.

In this study, the hypothesis is that vegetarian diets cause an increase in mental health disorders. A correlation coefficient of zero would confirm the null hypothesis (N0), that there is no correlation. However, a positive correlation coefficient value would confirm our hypothesis, showing that the incidence of mental health disorders is higher in vegetarian populations.

It is emphasised that the results of the correlation study in no way indicate causation. They would purely indicate any correlation between the two variables, and as such should not be interpreted as whether or not vegetarian diets cause mental health disorders.

The aim of this study is to determine whether mental disorder symptoms increase with the exclusion of animal-product food groups.

This study will take only one cohort, and the population will be the student body of James Cook University. Volunteers will be sought from all colleges and academic years. However, as this will be a brief experimental study, the small population will eliminate other variables such as region and occupation. 250 participants will be selected from the student population of about 15,197 (about 1.64%), and incentives for participation offered in the form of gift cards or course credits. Volunteers will complete a questionnaire on health status, medical history, lifestyle and diet. All participants will be asked to complete a form detailing their demographic data: age, gender and college of study. Their diet will be assessed using the question: “Do you currently follow a vegetarian diet or have you followed a vegetarian diet in the past?” Respondents can answer either “yes, totally”, “yes, mostly”, or “no”. The sample group will be divided by diet: totally vegetarian, mostly vegetarian and non-vegetarian.

Over the duration of six months, a subset of the non-vegetarian group will be asked to progressively give up one food group, moving toward a vegetarian diet. The foods given up will be as follows: red meat, poultry, fish, eggs, milk and other dairy products. Since non-vegetarians traditionally represent a majority of the population – vegetarians were only 7% of the Singapore population in 2020 (Hirschmann, 2020) – we expect to see this trend represented in the sample group.

Considering the time commitment that will be required for this study and the need for repeated testing, it may prove impractical to use the DSM-V and M-CIDI diagnostic tools for the assessment of mental disorders. Instead, this study will employ the Primary Care Evaluation of Mental Disorders (PRIME-MD) Patient Health Questionnaire (PHQ) (Spitzer, Kroenke & Williams, 1999) for anxiety, somatic symptom and related disorders, depressive, alcohol and eating disorders.

As the study will take six months, the participants will maintain each of the above diets for one month, and there will be two PHQs per diet, which means mental health monitoring every two weeks for six months. The onset data for the above-listed disorders will be assessed, including the age and the dietary status at the time of onset, and prevalence rates assessed for six months. This study will have three control demographics: those who were total vegetarians from the start will remain total vegetarians throughout the study, those who were mostly vegetarian will also remain mostly vegetarian, and a subset of those who were non-vegetarian at the beginning will remain non-vegetarian throughout the study. The numbers for each subset will be determined at the point of recruitment for the study, depending on the demographics of the volunteers.

All participants will be asked to provide written, informed consent. It will also be made clear that the results of the psychological diagnostics are purely for research purposes, and participants would have to see their physicians for actionable mental health diagnoses and treatment.

The sample will be characterised by diet type using descriptive statistics and represented as percentages or means ± standard deviations.

Using SPSS, odds ratios and 95% confidence intervals will be used to estimate the association between diet and mental health disorders, computed using logistic regression analysis models, such as to answer the question, “How does the risk of mental health disorders change (yes or no) for every food group drop toward a vegetarian diet?” In this case, a change in incidence/prevalence of mental health disorders is the dependent variable, presented in a dichotomous nature (change vs. no change). The models will be adjusted for specific sociodemographic variables such as the college of study, age, sex and socioeconomic status, as well as other health behaviours like physical activity. Regression analysis will also be conducted for mental health disorders and the exclusion of each food group as a binary variable. Post hoc analysis will factor in the number of excluded food groups.

The hypothesis of this study is that vegetarian diets cause an increase in mental health disorders. Therefore, a significant association (p < 0.05) between mental health disorder symptoms and diet type would indicate that vegetarian diets do affect mental health. A change in the odds ratio of mental disorder symptoms with a change in the number of excluded food groups would indicate, more quantitatively, the degree to which a vegetarian diet increases the prevalence of mental health disorders.

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (DSM-5®). American Psychiatric Pub. Link

Beezhold, B. L., Johnston, C. S., & Daigle, D. R. (2010). Vegetarian diets are associated with healthy mood states: a cross-sectional study in seventh day adventist adults. Nutrition journal, 9(1), 1-7. Link

Beezhold, B., Radnitz, C., Rinne, A., & DiMatteo, J. (2015). Vegans report less stress and anxiety than omnivores. Nutritional neuroscience, 18(7), 289-296. Link

Chmielewski, M., Clark, L. A., Bagby, R. M., & Watson, D. (2015). Method matters: Understanding diagnostic reliability in DSM-IV and DSM-5. Journal of Abnormal Psychology, 124(3), 764. Link

Hirschmann, R. (2020, March 19). Dietary preferences Singapore 2020. Retrieved from link

Lachner, G., Wittchen, H. U., Perkonigg, A., Holly, A., Schuster, P., Wunderlich, U., ... & Pfister, H. (1998). Structure, content and reliability of the Munich-Composite International Diagnostic Interview (M-CIDI) substance use sections. European Addiction Research, 4(1-2), 28-41. Link

Matta, J., Czernichow, S., Kesse-Guyot, E., Hoertel, N., Limosin, F., Goldberg, M., ... & Lemogne, C. (2018). Depressive symptoms and vegetarian diets: results from the constances cohort. Nutrients, 10(11), 1695. Link

Michalak, J., Zhang, X. C., & Jacobi, F. (2012). Vegetarian diet and mental disorders: results from a representative community survey. International Journal of Behavioral Nutrition and Physical Activity, 9(1), 67. Link

Osborne, J. W., & Waters, E. (2002). Four assumptions of multiple regression that researchers should always test. Practical assessment, research, and evaluation, 8(1), 2. Link

Regier, D. A., Narrow, W. E., Clarke, D. E., Kraemer, H. C., Kuramoto, S. J., Kuhl, E. A., & Kupfer, D. J. (2013). DSM-5 field trials in the United States and Canada, Part II: test-retest reliability of selected categorical diagnoses. American journal of psychiatry, 170(1), 59-70. Link

Spitzer, R. L., Kroenke, K., Williams, J. B., & Patient Health Questionnaire Primary Care Study Group. (1999). Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. Jama, 282(18), 1737-1744. Link

Spitzer, R. L., Williams, J. B., & Endicott, J. (2012). Standards for DSM-5 reliability. American Journal of Psychiatry, 169(5), 537-537. Link

Velten, J., Bieda, A., Scholten, S., Wannemüller, A., & Margraf, J. (2018). Lifestyle choices and mental health: a longitudinal survey with German and Chinese students. BMC Public Health, 18(1), 632. Link