The primary objective of the proposed pilot clinical trial is to assess the feasibility and acceptability of delivering diet counselling plus supplementation intervention to adult women with generalized anxiety disorder (GAD) as measured by recruitment, retention, and intervention uptake. The secondary objectives are to evaluate changes in anxiety symptom severity, quality of life, diet quality, mindful eating behaviours, self-efficacy, and biomarkers, over the intervention period as compared to the waitlist control period. This trial will lay critical groundwork for a larger, adequately powered clinical trial, designed to assess the effectiveness of dietary counselling in the management of GAD.
Trial design and setting
This pilot intervention study will recruit 50 adult women with moderate to severe GAD and provide a dietary counselling intervention in combination with an omega-3 supplement that also contains vitamin D. A randomized, waitlist-controlled design will be used (Fig. 1). The waitlist-control design was selected to allow for random allocation to a non-treatment condition while still allowing all participants to complete the intervention. Waitlist-control designs have been suggested to improve rigour and preserve ethical considerations for studies where participant blinding is difficult . Twenty-five participants will be randomized to begin the intervention immediately and 25 participants will be randomized to observe a 12-week waitlist period before completing the intervention. This trial will take place at a single site. The study setting is the community clinic at the Canadian College of Naturopathic Medicine in Toronto, Canada. The study has been reviewed by the Research Ethics Board of the Canadian College of Naturopathic Medicine (CCNMREB040.Aucoin). Any amendments to this protocol will be communicated to the Research Ethics Board and Health Canada (the authority responsible for approval, conduct, oversight, and inspection of clinical trials in Canada) and modified on the trial registry. The design of this study involved participation from people with expertise in nutrition, psychiatry, research methods, laboratory testing, culinary education, behaviour changes, and lived experience with mental illness.
A person aged 18–65 years who currently identifies as a woman.
Primary diagnosis of generalized anxiety disorder based on a clinical interview with a psychiatrist using the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria.
Moderate to severe anxiety is defined as a Beck Anxiety Inventory score of 22 or higher.
Low quality diet is defined as a score of 8.5 or less on the MEDI-LITE tool.
All psychiatric medication, psychotherapy, and natural health products stable for the past 4 weeks.
Established therapeutic relationship with a family doctor, psychiatrist, or clinical psychologist as their primary source of mental health care.
Ability to swallow capsules.
Ability to read in English and provide informed consent.
Currently meets DSM-5 criteria for obsessive–compulsive disorder, bipolar disorder, a psychotic disorder, an eating disorder, or substance use disorder based on clinical interview with a psychiatrist.
Current high level of suicidality was assessed by a trial psychiatrist using the Columbia Suicide Severity Scale.
Starting or changing the dose of a psychiatric medication, psychotherapy, or natural health products in the past 4 weeks.
Severe food allergies, intolerances, or aversions that would prevent the individual from modifying their diet (includes aversion to bovine gelatin which is a component of the fish oil supplement).
Current participation in a program, research study, or treatment plan involving diet or lifestyle modification.
Baseline Omega Score of > 5% (suggesting adequacy of omega-3 fatty acids).
Supplementation of greater than 4000 IU of vitamin D daily for at least 1 month in the previous 6 months.
Allergy to fish or any of the non-medicinal ingredients in the study product.
Participants who are assigned to the waitlist control, and who do not meet one or more of the criteria at the time of starting the intervention will still be permitted to participate. This data will be included in the analysis. If a participant begins a new medication or psychotherapy, or adjusts the dosage, during the waitlist period, this new treatment must be stable for 4 weeks prior to beginning the treatment phase of the study, even if this requires extending the waitlist period by up to 4 weeks.
The inclusion and exclusion criteria were designed with the intention of recruiting a sample of participants who are likely to be good candidates for the proposed intervention. Assessment of baseline diet quality, and inclusion of individuals with low quality at baseline, was completed in a similar trial ; similarly, trials utilizing omega-3 supplementation have highlighted the importance of including participants with sub-optimal baseline levels . The participants are also similar with respect to non-modifiable factors that might influence the impact of the intervention. For example, there is evidence that emotional eating behaviours, which many influence the eating habits of people experiencing anxiety and depression, may be influenced by sex and/or gender and that women may be more affected . In the absence of clear evidence differentiating the respective roles of sex and gender, we have elected to include participants who presently identify as women. We recognize that this may introduce greater heterogeneity within the sample with respect to hormonal status; however, we acknowledge that the sample of adult women will include variation in hormonal status related to hormonal birth control, menstrual and hormonal cycle irregularities and disorders, and medically induced or natural menopause. We have included a question in the demographic questionnaire about sex assigned at birth. This will be used to characterize the study participants. If a significant number of participants respond with an answer other than female, sub-analysis may be undertaken. Future studies can be undertaken to assess the impact of diet on men or people of other genders who experience anxiety disorders.
Participants with co-morbid depression will be eligible to participate in the intervention because of the high level of comorbidity with GAD . The criterion related to recent high-dose vitamin D supplementation is included to identify potential participants who would exceed the maximum recommended daily allowance of vitamin D if they began taking the study product. Initially, this exclusion criterion was “Supplementation of greater than 1700 IU of vitamin D daily […]”; however, early results from screening identified a high number of interested individuals who were ineligible because of this criterion, and an amendment was made to make the study more inclusive without compromising on participant safety. In order to ensure that participants have access to adequate care, should they experience a serious worsening of their psychiatric illness during the study period, an established therapeutic relationship with a mental health professional is required.
Sample size and recruitment plan
Because of the pilot nature of the present trial, a formal sample size calculation was not completed. No intervention study has attempted to deliver a dietary intervention to participants with GAD and thus, the expected effect is unclear. The results of this pilot trial will inform the sample size calculation for a full-scale trial.
Participants will be recruited through multiple strategies. Letters will be mailed to local mental health service providers including psychiatrists, counsellors, psychotherapists, and psychologists. The clinicians will be provided with a poster to display in their office and fliers to distribute directly to clients. Additional recruitment strategies will include online advertising through Facebook and other social media sites. This will include posts in existing, local anxiety/mental health groups and paid advertisements. If needed, newspaper advertisements may be used.
Pre-study screening, randomization, and baseline evaluation
Interested participants who contact the study will complete a virtual screening visit. A member of the research team will administer the MEDI-LITE Score, and the Beck Anxiety Inventory and review the additional inclusion and exclusion criteria by phone or secure video call. If the participant meets these criteria, they will complete the clinical interview and Columbia Suicide Severity Scale with the trial psychiatrist via secure video platform. Following the completion of a verbal consent process, the participant will be mailed a kit for collecting a blood spot sample at home and will return the sample by mail for analysis of baseline omega-3 status.
Individuals who meet all study inclusion criteria will be contacted and invited to enrol in the study. There is no predetermined time between screening and baseline; as soon as all screening requirements are complete, enrollment and baseline data collection can begin. The baseline activities and data collection will take place through a combination of virtual and in-person activities. In an enrollment video call with the study coordinator, participants will complete the informed consent process by reviewing the informed consent form and asking questions. Following enrollment, baseline data collection will involve the completion of several questionnaires through a secure online platform (Supplemental file 2). If the participant does not have access to an internet-enabled device, they will be mailed a paper copy of the baseline questionnaires. The baseline assessment will also include the provision of a fasting blood sample. Following baseline data collection, participants will attend an in-person study visit. At this time, baseline height and weight will be measured, and the participants will be randomized. Randomization will be completed by the trial coordinator using a computer-generated internet-based randomization service provider using random permuted blocks. Following the assignment, the trial coordinator, intervention deliverer, and participants will be aware of allocation. Outcome assessors will be blind to allocation when measuring height and weight. Measurement of the omega-3 samples and interpretation of the data will be completed by a research team member who is blind to allocation. Data analysts will be blind to allocation; they will be provided with data from two groups of participants labelled groups 1 and 2.
For participants randomized to the waitlist control group, the baseline visit will conclude with instructions for the waitlist period. Participants randomized to the immediate start group will immediately begin the first dietary counselling session. There is no predetermined time between baseline data collection and commencement of the intervention; once all baseline electronic data are collected, participants may schedule the first in-person visit which will include the collection of the remaining baseline data and the first intervention session.
The intervention will consist of seven dietary counselling sessions delivered over 12 weeks (one session every 2 weeks). Three visits (the first, fourth, and seventh) will take place in person and the remaining will take place by secure video platform. If participants do not have access to a computer or the Internet, they may complete all sessions in person. In the event that COVID-19 restrictions change during study delivery, the number of virtual sessions may be altered in order to comply with local restrictions. The sessions will be 45 to 60 min in length.
The dietary recommendations are based on the Mediterranean diet  with minor modifications based on the results of a recent scoping review which identified and collated all of the existing research on diet factors and anxiety . The Mediterranean diet has been studied extensively; a meta-analysis has demonstrated an association between greater adherence to the Mediterranean diet and reduced risk of depression  and this diet has served as the intervention in other clinical trials where diet was used in the treatment of depression [10, 12, 13], with positive outcomes. The scoping review additionally suggested that the following dietary components were important for people with GAD: limitation of caffeine intake, the inclusion of culinary herbs and spices, green tea and herbal teas, dietary factors that support microbiome health, and an emphasis on adequate protein and lower glycemic index foods . The Canadian Food Guide (2019) was utilized as a reference for national dietary recommendations and culturally relevant foods . Because we expected the study participants to reflect the multicultural nature of Canada, we intentionally included foods and recipes reflecting a range of cultural backgrounds. Lastly, the dietary interventions and resources used in nutritional psychiatry studies that have previously reported benefits were reviewed in order to guide the design of the present intervention .
Use of theory
The behavioural change component of the intervention was guided by Social Cognitive Theory which posits a role of environmental, individual, and behavioural factors on behaviour . Environmental factors influencing eating habits will be addressed through the provision of recommended foods during the study period, prompts to consider individuals in the participants’ lives who could support them in their goals, and instruction to modify the physical environment. With respect to individual factors, we will increase knowledge about nutrition, food preparation, and the role of diet in mental health as well as increasing knowledge about participants’ own eating habits through self-monitoring. We will increase self-efficacy and confidence through motivational interviewing . Regarding behaviour, participants will be provided with educational videos in order to improve their skills and ability to perform the desired behaviours (i.e., preparing nutritious food for themselves). We will also address habitual behaviours that are known to influence food choices in this population such as comfort eating.
Educational material and resources
A set of educational handouts were developed for use in the EASe-GAD study (Table 1).
A collection of recipes was assembled as well as a collection of videos demonstrating recipe preparation, food preparation skills, and food safety videos. Participants will also receive shelf-stable food items at the first and fourth (in-person) visits in order to encourage adherence to the recommendations and decrease barriers to change.
The dietary counselling sessions will be conducted by a licensed Naturopathic Doctor with more than 10 years of experience providing dietary counselling to individuals with mental health concerns and a certificate in motivational interviewing. The counselling sessions will aim to focus on the benefits of the recommended foods, highlighting the advantages of following the recommendations rather than the risks of following alternative diet patterns in order to maintain a positive message. The emphasis of the intervention will be on changes to dietary patterns; participants will be advised to consume the diet ad libitum with no restrictions on caloric intake.
In addition to providing education about nutrition, there will be significant efforts to support behaviour change (Supplemental file 3: Summary of Behaviour Change Techniques). There will be an emphasis on providing encouragement, communicating that diet changes are possible among individuals experiencing mental illness, and evoking past behaviour change successes. Goal setting will include a focus on setting measurable, specific, and realistic goals that are generated through a guided discussion with the participants and specific action planning related to the goal. Additionally, participants will be guided in identifying barriers to the achievement of the goal and strategies to overcome barriers. Participants will be instructed to modify their physical environment so that the encouraged foods are easy to access and require limited preparation and discouraged foods are less accessible. Because comfort eating is recognized as a driver of diet choices among the study population, participants will complete exercises related to mindful eating (ex. the raisin exercise) and review strategies for managing non-hunger cravings to eat such as grounding, acknowledging feelings, and engaging in enjoyable or distracting activities. The dietary counselling activities to be completed at each study visit are listed in Table 2.
During the 12-week study period, participants will be supplied with an omega-3 capsule supplement (AquaOmega High EPA) and instructed to take a dose of four capsules, once daily (containing a total of 3456 mg of omega-3 fatty acids: 2659 mg of Eicosapentaenoic acid (EPA), 532 mg of docosahexaenoic acid (DHA), and 800 IU of vitamin D), with a meal. Significantly more trials have been completed using omega-3 supplementation in the treatment of depressive disorders than anxiety disorders . Recent guidelines recommend a daily dose of 1–2 g of EPA with a ratio of EPA:DHA of at least 2:1 for the treatment of depression . Due to the lack of specific recommendations on the dose of omega-3 used for anxiety, our intervention will draw inference from these guidelines on depression to employ a dose that has been shown to be beneficial in other psychiatric disorders. We have selected a high dose, high EPA formula in order to maximize the likelihood of detecting a signal of effect in the present pilot trial.
All concomitant medication, psychotherapy, or natural health products will be allowed during this trial. In order to minimize the risk of adverse events related to hypervitaminosis D, participants who are supplementing 1701 to 4000 IU of vitamin D daily at the time of screening will be eligible to participate if they agree to decrease their vitamin D supplementation to 1700 so that their total daily intake of vitamin D, including the study product, is 2500 IU. Participants will be encouraged to continue all other previous treatments at the same dose for the duration of the trial, if possible, but allowed to make changes if recommended by their mental health care provider. Participants will be queried at baseline and each follow-up visit about their use of medication, psychotherapy, or natural health products, and any changes will be recorded, analysed, and reported.
Safety and criteria for withdrawal
As rescue medicine, participants will be advised to take the over-the-counter product, diphenhydramine (Benadryl), in the event of an allergic reaction to the study product or any of the new foods that they consume, as recommended on the product label. Diphenhydramine will be provided to participants with the study product. Participants will be queried about the use of the rescue medication and asked to return any unused product at the end of the trial. The use of rescue medication for an allergic reaction will be recorded. Participants will be instructed to contact the study coordinator immediately if they experience a negative reaction to the study product or a serious worsening of their health. They will be provided with an email and phone number for this purpose. The qualified investigator will assess the patient and recommend a course of action. If the participant requires medical visits or medications that are not reimbursed by the Ontario Health Insurance Plan (OHIP), these costs will be reimbursed by the study sponsor, the Canadian College of Naturopathic Medicine. At each study visit, participants will be asked if they have experienced any negative effects since beginning the study product. Participants will be withdrawn from the study if they experience an allergic response, or other serious adverse event, as a result of the study product. These participants will be referred to their primary healthcare provider or a walk-in clinic for management. Participants will also be withdrawn if the trial psychiatrist deems the participant not adequately stable to continue to participate in the trial. This includes the presence of severe suicidal ideation or symptoms of psychosis. At each follow-up visit, participants will be asked by the intervention deliverer if they have experienced a significant worsening in their mental health since the previous visit. If they report symptoms related to suicidal ideation, the Ask Suicide-Screening Questionnaire (ASQ) will be completed . If the participant screens positive, they will be referred to the study psychiatrist for a complete assessment. If they screen acutely positive, they will be referred directly to a crisis centre or the hospital emergency room. If they report any symptoms related to psychosis (any report of hallucinations, delusions, or disordered thinking), the participant will be referred to the study psychiatrist for assessment. If they meet the criteria for withdrawal, they will be referred to their current mental healthcare provider for appropriate assessment and treatment. If a participant reports a significant worsening in symptoms of anxiety or depression, they will be permitted to remain enrolled in the study. They will be encouraged to contact their primary mental health care provider for assessment and management. If the mental health care provider suggests that psychiatric rescue medication is needed, this clinician will recommend the treatment to the participant. If rescue medication is used for mental health symptoms, participants will be allowed to continue participating in the study. The use of rescue medication will be queried at each study visit and recorded. Given that the primary objective is feasibility and acceptability, participants with poor compliance with the study intervention will not be withdrawn from the study.
Expected adverse events for this study product include a fishy aftertaste, bad breath, heartburn, nausea, or loose stools, atrial fibrillation, and increased bleeding. All adverse events will be recorded, and these records will be maintained for 15 years. The clinical trial sponsor will inform the Natural and Non-prescription Health Product Directorate of Health Canada of any serious expected or unexpected adverse events related to the study product immediately if possible and no later than 7 days after becoming aware of the information. Serious adverse events will also be reported to the Research Ethics Board which has reviewed the study protocol.
Participants will be withdrawn from the study if their baseline serum levels of 25-hydroxy-vitamin-D are greater than 120 nmol/L, as this level has been associated with safety concerns . These individuals will be instructed to contact their primary healthcare provider for assessment and management. In the event that the baseline or final blood work assessment suggests the presence of an undiagnosed medical condition, the participant will be notified immediately that there is an abnormal finding and told to contact their primary healthcare provider for appropriate assessment and treatment. With the participants’ permission, the lab reports will be sent to the healthcare provider. The participant will be allowed to continue to participate in the study.
As the intervention is a pilot study, process evaluation (feasibility and acceptability) will be the primary outcome, helping to inform a future full-scale trial. The time to recruit 50 participants will be evaluated as part of the assessment of feasibility. In order to assess acceptability, the number of sessions attended will be counted. All participants will be asked to complete a brief satisfaction survey concerning the participant’s level of satisfaction with the program, the aspects which were helpful or unhelpful, and opportunities for improvement (Supplemental file 4: Outcome assessment). A sub-sample will be recruited to participate in a brief focus group in order to obtain more detailed qualitative responses about the participant experience and participant satisfaction. We will aim to recruit 8–18 participants to run 2–4 focus groups, each having 4–6 participants. After the second focus group, we will complete a brief data analysis and run additional focus groups if data saturation has not been achieved. The focus groups will be conducted using a semi-structured qualitative interview methodology. The sessions will be audio recorded and transcribed. Transcripts will be analysed using thematic analysis . Additionally, we will assess the fidelity of the intervention by comparing the intervention protocol with the intervention delivery. The frequency of use of each questionnaire and resource will be counted. At baseline, a brief demographic questionnaire will capture information related to age, sex at birth, marital status, employment status, ethno-racial background, income, smoking status, and alcohol consumption (Supplemental file 4: Outcome assessment). Assessment of physical activity levels and food insecurity will also be completed using the International Physical Activity Questionnaire  and Canadian Community Health Survey Household Food Security Survey Module , respectively. Height and weight will be measured before and after the waitlist period and the intervention period. Body mass index and change in weight during the intervention and control period will be calculated. These data will characterize the study participants and allow for the assessment of confounding factors.
In order to assess changes in anxiety symptom severity, the Beck Anxiety Inventory, a validated, 21-item self-report inventory for measuring the severity of anxiety in participants with psychiatric illness, will be administered . The impact on quality of life will be assessed using PROMIS-29 . In order to assess the impact of the intervention on behaviour change and qualitatively assess the acceptability of the intervention, questionnaires related to mindful eating and diet quality will be administered before and after the intervention. Mindful eating behaviours will be measured using Mindful Eating Questionnaire (MEQ) which has been validated for use in a diverse adult population . Changes in diet quality will be measured using the MEDI-LITE score, a concise tool for measuring consistency with the Mediterranean diet, which has been validated for use in adult populations . Changes in self-efficacy will be assessed using the General Self-Efficacy Scale . Changes in diet quality will also be assessed objectively through blood tests. Omega-3 fatty acids levels will be assessed using the dried blood spot test, Omega Score , which represents summed EPA, DHA, and Docosapentaenoic acid as a percentage of the total fatty acids in whole blood. The analysis will be completed by Lipid Analytical Laboratories. Additional assessment will be completed to measure fasting serum cholesterol panel, markers of blood sugar levels and regulation (Haemoglobin, fasting insulin, fasting glucose, and calculated HOMA-IR [55, 56]), vitamin C, beta-carotene, and C-reactive protein. These analyses will be completed by LifeLabs. Completion of the questionnaires and blood tests will be done at baseline, at the end of the waitlist control period (for participants assigned to this group) and at the end of the intervention. Participants who do not attend all of the scheduled visit but have not withdrawn will be contacted to complete the questionnaires and blood tests. See Supplemental file 4 for a summary of assessments on participant demographics, satisfaction with the dietary intervention, the research program, and participant experience.
Compliance with the supplement will be assessed at the final visit. Participants will be instructed to bring back all study product bottles containing any remaining capsules. The remaining capsules will be counted and recorded by a member of the study team. Compliance with the supplement will be reported as a percentage based on calculating the amount of pills consumed in comparison to the amount expected according to instructions.
Since the pilot study is not powered to demonstrate efficacy, analysis of the pilot study results will be strictly descriptive. Baseline demographics and outcome measures will be summarized by group as means and confidence intervals, medians and interquartile ranges, or percentages and confidence intervals as appropriate. Process evaluation will be described both quantitatively (percent screened positive for inclusion, percent of eligible patients who consent to participate, percent with significant protocol violations) and qualitatively (e.g., reasons for failing inclusion criteria, reasons for refusing to participate, participant impressions of the intervention). All participants will be analysed in the group to which they were randomized. Missing data will be carefully documented to inform the design of the larger trial. Participants who complete the treatment but do not report a satisfaction level will be treated as ‘unsatisfied.’ No interim analysis will be undertaken. The study will be considered feasible if 50 participants are recruited within 2 years, and participants in the intervention group take at least 60% of the study product and attend at least 60% of the diet counselling sessions. The study will be considered acceptable if at least 70% of participants report that they were satisfied with the intervention. The statistical analysis plan for the full-scale trial is included in Supplemental file 5.
The collection and storage of data will be done according to Good Clinical Practice Guidelines. All study staff involved in the collection or entry of data will be trained by the trial coordinator and will complete ethical conduct of research training (Tri-Council Policy Statement: Ethical Conduct of Research Involving Humans 2). Data from potential and enrolled participants will be stored in the secure platform, Research Electronic Data Capture (REDCap), and a password-protected database on a secure server managed by the Canadian College of Naturopathic Medicine. Data collected at each study visit will be captured using an electronic case report form which will be entered into the database. Questionnaire completion will be done by the study participants through REDCap. Laboratory data will be entered into the database and double-checked for accuracy by a team member blind to participant allocation. An audit of the data will be completed by a senior member of the research team who is not involved in data collection after 10 participants are enrolled or 3 months of time, whichever comes first. All information obtained during the trial will be kept in a locked cabinet or on a password-protected computer and destroyed (deleted or shredded) after 15 years.
Access to data and dissemination
All personal health information will be kept confidential and only accessed by the research team, unless required by law. This includes an audit by the CCNM Research Ethics Board or Health Canada Natural and Non-prescription Health Products Directorate. All data will be maintained on password-protected, secure servers or locked filing cabinets for the required length of time. Following this time, data will be destroyed securely by deleting the digital files or shredding paper documents. No identifying information from any participant will be used in the dissemination of study results. Dissemination of the study findings will occur through publication. Study participants will have the option to be notified of the study findings.