Megan Lee – PhD Candidate - PowerPoint Presentation

1. Megan Lee – PhD CandidateSupervisory Team – Dr Joanne Bradbury,, Assoc Prof Jacqui Yoxall, Assoc Prof Sally SargeantThe Role of Dietary Patterns in Depression: an Overview of the Evidence.

2. DepressionGlobally over 350 million people experience symptoms 1.5 million people die from suicide annuallyWHO estimate that depression will be the leading cause for DALY by 2030High individual and global economic costsTalking therapy, psychopharmacology, ECTCurrent treatment options are not suitable for everyoneNutritional psychiatry – dietary patterns and depression(Institute for Health Metrics and Evaluation, 2018) ( Owen and Corfe, 2017) (Rehm and Shield, 2019) (WHO, 2008)

3. Umbrella ReviewAn overview of all systematic reviews and meta-analyses on dietary patterns and depressionRegistered with PROSPEROPreferred Reporting Items for Systematic Reviews/Meta-Analyses (PRISMA)A MeaSurement Tool to Assess systematic Reviews (AMSTAR2)Grading of Recommendations Assessment, Development and Evaluation (GRADE)(Grant & Booth, 2009; Moher et al., 2009; Shea et al., 2017; Balshem et al., 2011)

4. Healthy Dietary Patterns(Jacka, 2019) (Radd-Vagenas et al., 2018) (Miki et al., 2018) (Sánchez-Villegas et al., 2019)MediterraneanJapaneseNorwegianTraditional Anti-Inflammatory

5. Unhealthy Dietary Patterns(Dipnall et al., 2015) (Jacka et al., 2012) (Sarris et al., 2015) (Parletta et al., 2017)WesternPro-InflammatoryHigh Glycaemic

6. Sample Characteristics16 publications incl 213 original articles10 meta-analyses, 6 systematic reviewsEach with 4000 - 310,000 participants aged between 15 and 103 years10 published between 2017 and 20196 published between 2013 and 2015

7. AMSTAR2 Quality RatingsAMSTAR2 – Rates the overall quality of each included paperCritical domains –comprehensive literature search, assessing and reporting risk of bias, appropriate meta-analysis technique and assessment of publication bias, list of excluded studies, registering review methods a priori, High quality (n = 9)Moderate quality (n = 4)Low to critically low quality (n =3)(Shea et al. (2017))

8. Preliminary FindingsOf the 16 meta-analyses and systematic review9 found an association between dietary patterns and depressive symptoms 2 found an association for healthy but not unhealthy dietary patterns5 reported inconclusive findings 3 older (2013/2014) narrative systematic reviews 2 evidence for cross-sectional but not cohort or vice versa

9. GRADE AssessmentObservational papers begin low, RCT’s begin highDeduct for inconsistency (heterogeneity), risk of bias, imprecision (CI and sample size), indirectness to PICO, publication biasIncrease for large effect sizes, dose response gradient, residual confounding (sensitivity analysis)

10. GRADE assessment – Mediterranean Dietary Pattern (n = 8)Strong recommendations for using intervention5 found association of Mediterranean dietary pattern and decreased odds of depressive symptoms – high GRADE, high/moderate AMSTAR2 (Lassale et al., 2018; Frith et al., 2019; Psaltopolou et al., 2013; Molendijk et al., 2017; Rahe et al., 2014)1 found evidence of Mediterranean dietary pattern in cross-sectional but not cohort studies – low GRADE, moderate AMSTAR2 (Shafei et al., 2019)2 found inconclusive evidence– low GRADE, moderate/low AMSTAR2 (Quirk et al., 2013; Sanhueza et al., 2013)

11. GRADE assessment – Healthy (n = 7)Strong recommendations for using intervention6 found association of healthy dietary patterns and decreased odds of depression –moderate/high GRADE, moderate/high AMSTAR2 (Lai et al., 2014; Li et al., 2017; Firth et al., 2019; Molendijk et al., 2017; Rahe et al., 2014; Lassale et al., 2018)2 found Inconclusive evidence – low GRADE, High AMSTAR2 (Quirk et al., 2013, Opie et al., 2015)

12. GRADE assessment – Western (n = 5)Weak recommendations for using intervention2 found an association of Western dietary patterns and increased odds of depression (low/very low GRADE, high AMSTAR2 (Li et al., 2017; Rahe et al., 2014)2 found no association of Western dietary patterns and increased odds of depression– low GRADE, High AMSTAR2 (Lai et al., 2014; Molendijk et al., 2017)1 found inconclusive evidence– very low GRADE, high AMSTAR2 (Quirk et al., 2013)

13. GRADE assessment – High Glycaemic Index (n = 2)Weak recommendations for using intervention1 found an association of high glycaemic index and increased odds of depression – low GRADE, moderate AMSTAR2 (Rahimlou et al., 2018)1 found an association of high glycaemic index and increased odds of depression across cohorts. No association across cross-sectional studies – very low GRADE, low AMSTAR2 (Salari-Moghadam et al., 2019)

14. GRADE assessment – Pro-Inflammatory (n = 3)Strong recommendations for using intervention3 found an association of pro-inflammatory dietary pattern and increased odds of depression 2 high GRADE, moderate/high AMSTAR2 (Wang et al., 2019; Tolkein et al., 2017)1 low GRADE, critically low AMSTAR2 (Kheirouri et al., 2019)

15. ConclusionsBody of evidence stronger for healthy dietary patterns than unhealthy dietary patterns GRADE assessment disadvantages – 14 of 16 papers observational. Systematic reviews do not meet all criteriaInconclusive results across older papers (2013/2015) SMILES trial and HELFIMED trials excluded from analysesOlder papers also unfairly disadvantaged by AMSTAR2 critical domain– registering review methods a priori

16. Answering the initial research problemEmail – megan.lee@scu.edu.auTwitter - @MeganLeePhDInstagram - @meganlovingmeagain