**1. Introduction**

The term anxiety describes the experience of worry, apprehension, or nervousness in association with physical, cognitive, and behavioral symptoms. Anxiety may be experienced occasionally as part of normal life and may be adaptive if it increases preparedness for novel situations. If anxiety symptoms are persistent, excessive, or interfere with functioning, they can become pathological [1].

Several anxiety disorders have been defined. Generalized anxiety disorder involves excessive worry in multiple domains and associated physical symptoms that are present for at least six months leading to clinically significant distress or impairment in functioning [1].

**Citation:** Aucoin, M.; LaChance, L.; Naidoo, U.; Remy, D.; Shekdar, T.; Sayar, N.; Cardozo, V.; Rawana, T.; Chan, I.; Cooley, K. Diet and Anxiety: A Scoping Review. *Nutrients* **2021**, *13*, 4418. https://doi.org/10.3390/ nu13124418

Academic Editors: Roser Granero, Diego Redolar Ripoll and Ina Bergheim

Received: 22 September 2021 Accepted: 4 December 2021 Published: 10 December 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Panic disorder is characterized by unexpected and recurrent panic attacks and at least one month of persistent worry about having a subsequent panic attack or significant behavior changes related to the attack [2]. Agoraphobia involves feelings of intense fear of situations or spaces where escape may be difficult or help may not be available in the event or panic or other incapacitating symptoms [3]. Social anxiety disorder involves marked anxiety and fear of a social situation where an individual is exposed to possible scrutiny by others [4]. Specific phobia is an excessive fear of specific object or situation [5].

Anxiety disorders exert a significant burden at both an individual and societal level. Individuals with anxiety disorders report a high degree of psychological distress, significant disability [6] and a reduction in quality of life [7]. The presence of an anxiety disorder is associated with higher use of both primary care, emergency room visits, and specialist healthcare services [8]. These disorders are also highly prevalent. The national comorbidities study established the lifetime prevalence of any anxiety disorder at 31.2%, the highest of any category of psychiatric illnesses [9].

The treatment approaches most frequently used in the management of anxiety disorders are psychotherapy and psychopharmacology [10]. While many patients find these therapies beneficial, a significant number of individuals report that these treatment options are not accessible, tolerable, or effective in providing adequate relief of anxiety symptoms [11]. For these reasons, there is interest in the evaluation of adjunctive or alternative therapeutic approaches.

Nutritional psychiatry is an emerging field of study related to the use of nutritional interventions in the prevention and treatment of mental health disorders. Despite increasing evidence of beneficial effects, nutritional recommendations are provided to psychiatric patients infrequently in clinical practice. Recently, high quality intervention studies have demonstrated an antidepressant effect of nutritional interventions [12,13]. However, the amount of research on anxiety disorders lags behind that of mood disorders [14,15]. There is a clear lack of studies delivering diet counselling, education, or food as an intervention to individuals with diagnosed anxiety disorders as well as systematic synthesis of the existing literature on the relationship between dietary factors and anxiety symptoms or disorders. The objective of the present review was to systematically map out the body of existing literature on anxiety symptoms/disorders and nutrition in order to identify nutritional factors associated with higher or lower levels of anxiety and to identify knowledge gaps and opportunities for further research.

### **2. Materials and Methods**

The review followed established methodological approaches for scoping reviews using the framework presented by Arskey and O'Malley for the conduct of scoping reviews [16]. Scoping reviews aim to identify and describe the breadth of literature on a topic when it is either highly complex, involves a broad array of study designs, or when a comprehensive review is being completed for the first time; all of these factors apply to the present review. Scoping reviews aim to map key concepts in a field of study and the available types of evidence. The review is completed in a way that is systematic, highly rigorous, and transparent in order to minimize bias. The protocol used in the present study was adapted from a similar project completed by the authors on the topic of diet and psychosis [17].

An extensive *a priori* search strategy was developed and executed with the guidance of an experienced medical librarian. Using the Ovid platform, we searched Ovid MEDLINE®, including Epub Ahead of Print, In-Process & Other Non-Indexed Citations, and Embase Classic + Embase. We used controlled vocabulary (e.g., "Anxiety Disorders", "Nutritional Physiological Phenomena", "Food") and keywords (e.g., anxiety, nutrition, diet). We adjusted vocabulary and syntax as necessary across the databases. There were no language or date restrictions on any of the searches, but we removed opinion pieces (e.g., editorials) from the results. We performed the searches on 25 March 2020. The full search, as executed, is available in Supplemental File S1.

Screening of abstracts and titles was completed using the online open-source program Abstrackr [18] which allowed for concurrent and blind duplicate screening as well as tagging by dietary constituent. Manual screening was completed until the program's artificial intelligence predicted the presence of additional relevant studies as unlikely. Previous testing of this program has demonstrated that the likelihood of missing relevant studies is very low [18–20]. As an extra precaution, once screening had reached the point where Abstrackr's prediction score reported 'zero additional studies likely to be relevant', and no studies were being identified as being relevant, an additional 100 articles were screened in each section before screening was stopped. Screening of abstracts and titles was completed in duplicate. Disagreement was resolved by consensus.

Studies were eligible for inclusion if they involved the evaluation of changes in the level of anxiety symptoms or the presence/absence of anxiety disorders in humans or animal models as well as assessing or modifying a component of participant diet. This included assessment or modification of dietary patterns, individual foods, supplements, or natural health products that provide an active constituent naturally occurring in the general North American diet. Studies were ineligible if they assessed or administered herbal medicines (apart from those used for culinary purposes in the general North American diet) or constituents which are not typically found in significant quantities in the human diet (i.e., St. John's Wort, GABA, or S-adenosylmethionine) or if they assessed levels of endogenously produced dietary components (i.e., cholesterol, vitamin D, or non-essential amino acids) in the absence of supplementation or measurement of intake. Eligible study designs included human observational and experimental studies, animal studies, and metaanalyses. Studies were excluded if they assessed the impact of maternal diet on offspring anxiety levels. Review articles, opinion papers, letters, and systematic reviews (without meta-analysis) were excluded, as were non-English language papers or inaccessible papers in cases where the abstract contained insufficient information for data extraction.

Full text screening was completed concurrently with data extraction. Data extraction was completed using piloted extraction templates developed for a similar scoping review conducted by the study authors and double checked by MA for accuracy [17]. Analysis was completed by sorting the studies with common interventions and methodology types and counting the number of studies reporting an association with increased or decreased anxiety symptoms/disorders, or no association. These data were used to create figures that communicate an overview of the evidence on each topic. Studies reporting a statistically significant improvement in at least one subpopulation or measure of anxiety symptomatology were categorized as "associated with decreased anxiety". Studies reporting an increase in anxiety symptoms or prevalence in at least one subpopulation or measure of anxiety symptomatology were categorized as "associated with increased anxiety". Studies that reported no significant change in anxiety symptoms or prevalence were categorized as "no association with anxiety". A small number of studies which reported mixed findings such as a combination of increased and decreased anxiety symptoms were not included in the figures. In order to allow concise display and the comparison of all studies, the number of studies reporting an association between improved symptoms with higher intake of a nutrient were combined with studies reporting and association between worse symptoms with lower intake of the nutrient. Counts are depicted in figures. The figures are oriented so that they report the relationship between higher intake of the diet constituent with anxiety. Within each section, a narrative summary was completed to highlight trends, gaps, and areas that warrant further study. When available, narrative summaries also reported on proposed mechanisms and safety. Finally, we created a list of dietary factors that, based on the review findings, may to be associated with less anxiety and more anxiety symptoms/disorders. This process of categorization was done based on the following criteria related to the volume and consistency of evidence. Dietary factors were included in these two categories when there were at least five studies reporting on the relationship with anxiety, and the majority of the data points (>60%) showed a consistent association. These criteria were developed post hoc as the volume, and consistency of the evidence was unknown at the time of protocol development.

#### **3. Results**

#### *3.1. Search Results*

The search identified 55,914 unique results that were screened in two phases: by title/abstract and by full text. The study authors manually screened 13,286 results while Abstrackr's artificial intelligence screened the remaining results. Following title and abstract screening, 2213 articles were included.

Seventeen articles could not be retrieved in full text. During full text screening, an additional 655 studies were excluded (Supplemental File S2). 1541 studies were included in the final data analysis (Figure 1, Supplemental File S3).

**Figure 1.** PRISMA flow diagram. AI: artificial intelligence.

#### *3.2. Study Characteristics*

More than half of the studies in our analysis were conducted using animal models (*n* = 859) (Figure 2). The animal studies were primarily conducted in rodents (97%) with the remaining 3% of studies conducted in zebrafish, pigs, lemurs, monkeys, cats, horses, and tilapia.

**Figure 2.** Distribution of included studies by methodology (count, percent).

The observational studies included 14 case reports; in 11 reports, subjects had anxiety disorders or elevated anxiety symptoms at baseline (Figure 3). In total, the reports described 44 individual cases. An additional 255 publications described cross-sectional, prospective, or retrospective observational studies. One meta-analysis of observational studies was conducted [21]. Of the observational studies, 88% were cross-sectional in design and 13 studies (5%) specifically included individuals with anxiety disorders or elevated anxiety symptoms. Nutrient intake was assessed in 201 studies while nutrient levels, in various body tissues, were measured in 40 studies. Sample size varied widely, from 14 to 296,121 participants (Mean: 6315.7, SD: 28,423.9).

**Figure 3.** Types of observational studies (count, percent).

Regarding experimental studies, 395 met criteria for inclusion as well as an additional 18 meta-analyses of experimental studies. Of the individual studies, the number of participants ranged from 3 to 2730 (Mean: 99.3, SD: 198.3). Of the 395 trials, 335 (85%) included a comparison arm, 312 (79%) utilized randomization, and 23 (61%) utilized blinding. Thirty-nine trials (10%) included participants with anxiety disorders or elevated anxiety

symptoms. An additional 57 trials assessed anxiety in participants with other psychiatric illnesses while the remaining studies included participants with medical illnesses or healthy participants (Figure 4). Excluding the studies that assessed the immediate impact of food on anxiety symptoms (*n* = 72), the average duration of the experimental studies was 15.8 weeks (SD: 18.3 weeks). Most of the studies (*n* = 331) provided the dietary intervention without co-interventions. An exercise co-intervention was delivered in 32 studies, while 20 included a psychosocial component and ten co-administered a medication. Sixty-nine percent of experimental studies identified a primary outcome related to mental health.

**Figure 4.** Participant populations in human experimental studies (count, percent).
