Plant power? A systematic review of the effects of plant-based diets on people with mental illness

Objective There is increasing interest in plant-based diets in the general population and an increasing evidence base for the positive impact of plant-based diets on health outcomes for many chronic diseases. This systematic review aims to identify the effects of plant-based diets on people with mental health conditions. Methods A systematic review of intervention and observational studies. We conducted a systematic electronic search of MEDLINE (Ovid), EMBASE (Ovid), PsycINFO (ProQuest), British Nursing Index (ProQuest), CINAHL (EBSCO) and the Cochrane library to April 2019, with no date limits. We extracted data on outcomes and assessed the studies for bias using validated tools. Results We retrieved 588 studies. One study met the inclusion criteria with high risk of bias. The intervention was a plant-based diet for people with moderate to severe depression, without a control group. The study recruited 500 people, but recorded 66.8% attrition. Of the completers, 62% reported improvements in depressive symptoms, and 59% in anxiety symptoms. Completers lost 5.7lbs (2.6kg) during the trial and 15lb (6.8kg) at six month follow up. Conclusion There is not enough research to make conclusions about the effects of plant-based diets on people with mental health conditions. Given the evidence for positive effects of plant-based diets on physical health, further research is urgently required to understand the effects on people with mental health conditions. This will support the provision of advice and guidance for patients with mental illness who want to optimise their diet to improve their mental and physical health. PROSPERO registration: CRD42019133440


Introduction
Plant-based diets and vegetarian diets are terms used interchangeably for a group of diets which exclude or minimise the consumption of meat, including vegan diets which also exclude eggs and dairy. There is evidence that suggests that plant-based and vegetarian diets have beneficial effects on physical health, which include reduced risk of developing diabetes and cardiovascular disease, improved diabetic control and weight loss in patients with type 2 diabetes, and improved survival outcomes for people with colorectal cancer [1][2][3][4][5][6][7]. A recent systematic review found evidence for the positive impact The World Health Organisation predicted in 2001 that 450 million people were affected by a mental health condition globally [14]. A more recent review suggests that mental disorders and addictive disorders combined affect about one billion people globally [15]. Each year about 20% of the English population will experience a common mental disorder, yet perhaps only one in three will access help [16,17].
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.28.20203026 doi: medRxiv preprint improve depressive symptoms for people with major depressive disorders, who were previously consuming an unhealthy diet [35]. This suggests that dietary interventions may support improved mental health outcomes for the population with a mental illness. Dietary interventions could be a low cost, public health intervention with potential benefits for mental health but also on physical health co-morbidities. As such, this is an important topic for consideration from both a clinical psychology and public health perspective.
Therefore, we conducted a systematic review of the literature to determine the benefits and risks of plant-based diets to people with mental illness and mental health conditions.

Systematic Search
We registered a review protocol with the International Prospective Register of Systematic Reviews (PROSPERO) database (CRD42016027656 http://www.crd.york.ac.uk/PROSPERO).
We conducted a systematic electronic search of MEDLINE (Ovid), EMBASE (Ovid), PsycINFO (ProQuest), British Nursing Index (ProQuest), CINAHL (EBSCO) and the Cochrane library to April 2019, with no date limits. The search strategy included terms for plant-based diets and mental illness. Search words were plant-based diets or vegetarian or vegan and diet, and mental . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . illness or mental health or mentally ill or mental problem or psychological disorder or psychiatric disorder or anxiety or depression or psychosis or schizophrenia or bipolar or personality disorder or mood or emotion. The Medline search strategy is shown in Table 1. This was amended for other databases and supplementary searches are shown in Supplementary Tables 1 to 5. Grey literature searches were conducted using keywords "vegetarian", "vegan", "plant-based diet" and "mental illness" and "mental health condition".

Eligibility Criteria
RCTs, meta-analyses and systematic reviews to assess potential benefit of plant-based diets in people with mental illness were eligible. Observational studies were included to assess the benefits and harmful effects due to the likelihood of a limited number of RCTs. Studies would be included where they involved adults with neurotic and psychotic disorders (as diagnosed using recognised diagnostic criteria). Studies involving adolescents, and adults with learning disability, dementia, organic disorders and eating disorders were excluded. Studies had to be published as full text in English. Interventions of interest were plant-based diets, vegetarian diets and vegan diets. Comparator interventions of interest were usual diet, including omnivore diet, control diets and traditional calorie control diets. A broad range of outcomes were of interest including improvement in mental health symptoms, change in weight, metabolic . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.28.20203026 doi: medRxiv preprint status, disease markers, medicine doses and quality of life markers. Adherence outcomes were not included.

Study Selection
Duplicates were removed after a complete list of studies was generated. An initial title screen was conducted by the librarian at Lancashire and South Cumbria NHS Foundation Trust. Two reviewers (JB and HC) independently assessed the remaining sample against eligibility criteria at the abstract screening stage and resolved discrepancies by discussion. Full copies were obtained of all potentially eligible studies and assessed against eligibility criteria by all three researchers (JB, HC and AJ). Discrepancies were resolved through discussion.
During the screening process, the decision was taken to widen the eligibility criteria to include interventions where plant-based diet was one component.

Data Collection
A standardised form was developed to extract data from the included studies for assessment of study quality and evidence synthesis. Information was extracted on: study setting; study population, participant demographics and baseline characteristics; details of the intervention and control; study methodology; retention and completion rates; outcomes and times of measurement; . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.28.20203026 doi: medRxiv preprint suggested mechanisms of intervention action; information for the assessment of risk of bias. Two authors (JB and HC) extracted data independently and any discrepancies identified were resolved by discussion.

Quality assessment
Risk of bias was to be assessed using a range of tools: the Cochrane risk of bias tool, the GRADE tool, the Newcastle-Ottawa scale and the ROBINS-I tool.
Two reviewers (HC and AJ) independently assessed the risk of bias in included studies with disagreement resolved through discussion with a third reviewer (JB).

Synthesis of Results and Analysis
Narrative synthesis of the results of the systematic review were planned due to the likely small number of diverse studies. The review was reported in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (see Supplementary Table 6 for the PRISMA checklist) [37].

Systematic search results
The database searches identified 850 records, 588 after duplicates were removed ( Figure 1). No further results were identified from grey literature searches. Following title and abstract screen, 41 articles were included in the . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.28.20203026 doi: medRxiv preprint full review. All but one were ineligible, even with the widened inclusion criteria for the intervention.
The included study was of a 12 week, non-randomised, non-controlled diet and lifestyle intervention [38]. The extracted data is shown in Table 2. The study included 500 men and women with a diagnosis of chronic moderate to severe depression and anxiety. The diet component of the intervention was described as an anti-inflammatory plant-based diet with 70% raw foods, and no animal products. This was combined with a number of other components including juicing, exercise, mindfulness and environmental hygiene. Many outcomes were measured, but the key mental health outcomes were that 62% reported large improvement or complete remission of depression symptoms, 59% of anxiety symptoms. In terms of physical health, the key outcomes were average weight loss of 5.7lbs during the trial, increasing to 15.0lbs at six month follow up. However, 66.8% of the participations dropped out of the study, largely within the first two weeks and mostly because they found the diet and behaviour modification too rigorous. The results were reported as per protocol, rather than by intention to treat. The study was assessed to be of serious risk of bias.

Discussion
Given the increasing evidence base and guidance which advocates for plantbased diets to improve health outcomes in the general population [1][2][3][4][5][6][7][8][9][10][11][12][13], there . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.28.20203026 doi: medRxiv preprint is a concerning lack of evidence about the potential impact of plant-based diets on outcomes in people with mental health conditions. Our systematic review found just one study. While the results of this study found a plant-based diet (in a combined lifestyle intervention) had a positive impact on symptoms of people with major depressive disorder, the study was assessed as having serious risk of bias and the singular study means there is no data to synthesise.
There is some evidence of the impact of diet on the outcomes of people with mental health conditions. A randomised controlled trial found improvement in symptoms for people with major depressive disorder eating poor diets, who were subsequently helped to eat diets of whole, unprocessed, mostly plantbased foods, which included meat and dairy [39]. However, it is hard to know whether or not it was the whole plant foods which had the positive impact, or if including animal products had any bearing on this result. The diet recommended in this study aligns with the EAT-Lancet commission [11].
Although there is almost no research looking at the impact of exclusively plantbased diets on outcomes in people with mental ill health, there is some evidence for the impact of such diets on mental health and psychological wellbeing outcomes in the general population. Trials have suggested that different forms of plant-based diets (vegan and low fat, whole food diets) can lead to improvements in psychological wellbeing or positive impacts on mental . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. While it is generally accepted that diet does affect mental health [43], the current evidence is unable to provide clarity. Despite the lack of evidence, it remains plausible that diets of whole foods, which contain little or no processed foods, such as the Mediterranean diet, are likely to have a beneficial impact to reduce the risk of developing depressive symptoms, and contribute to recovering from depressive disorders. However, this needs further research, given the increasing evidence of the benefit of whole food plant-based diets on . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. There is significant variety in dietary patterns within groups such as 'vegan', 'vegetarian' and 'omnivores', making like-for-like comparisons difficult. It is possible to follow vegan and vegetarian diets which are composed of entirely processed foods with no fruit and vegetables, which would not be as healthy as a whole food plant based diet. Similarly, a diet which is composed of mostly whole plant-based foods with small amounts of lean meat is likely to be more beneficial for health than one without whole foods. These variations make comparisons of diets under broad headings difficult. There may be confounding effects from multiple other unknown sources also, such as pesticide use. We . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . must also be cautious using the current evidence base which focuses on depressive and anxiety symptoms of the general population, rather than those people suffering with mood disorders and anxiety disorders. There also seems to be a significant lack of research on the effects of dietary patterns on other mental disorders.
Given that there is increasing evidence for inflammatory processes in the aetiology of mental health conditions, [47,48] and that plant-based foods are considered as having a low dietary inflammatory index,[49,50] the interplay between plant-based diets, inflammation and mental illness requires further exploration. It is imperative to explore if intervening on a pro-inflammatory Western diet will have a positive impact on mental health.

Conclusion
We have conducted a comprehensive systematic review to explore the impact of plant-based diets on mental illness, which is the first of its kind to our knowledge. A key strength of the study is the rigour of the methods. In addition, the results align with a recently published systematic review on the effects of plant-based diets on the body and the brain in the general population.
[8] The authors of this review were able to demonstrate the impact of a plant based diet on metabolic measures in health and disease, but were unable to locate evidence for cognitive and mental effects. Our study differs as we were . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. .

DATA ACCESS
All data is supplied within the document.

ACKNOWLEDGEMENTS
Cath Harris, University of Central Lancashire for conducting the literature searches and the initial title screening.

CONFLICTS OF INTEREST
The authors declared no conflict of interest with respect to the authorship or publication of this article.

FUNDING
No funding was available for this study which was conducted by the researchers within their roles at Lancashire and South Cumbria Foundation trust.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.28.20203026 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review) preprint
The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.28.20203026 doi: medRxiv preprint TABLES  . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review) preprint
The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.28.20203026 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review) preprint
The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.28.20203026 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review) preprint
The copyright holder for this this version posted September 29, 2020. .

29
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.28.20203026 doi: medRxiv preprint ( 1 9 4 9 1 9 ) . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. .
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. .

Supplementary table 2: PsycINFO (ProQuest platform) search strategy
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. .

Supplementary table 3: British Nursing Index (ProQuest platform) search strategy
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. .

S34
S7 AND S33 128 S33 S8 OR S9 OR S10 OR S11 OR S12 OR S13 OR S14 OR S15 OR S16 OR S17 OR S18 OR S19 OR S20  . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. .
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. .

Section/topic # Checklist item Reported on page #
Data collection process 10 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators. 8 Data items 11 List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made. 8

Risk of bias in individual studies
12 Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level), and how this information is to be used in any data synthesis.

9
Summary measures 13 State the principal summary measures (e.g., risk ratio, difference in means).

9
Synthesis of results 14 Describe the methods of handling data and combining results of studies, if done, including measures of consistency (e.g., I 2 ) for each meta-analysis.

9
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. .