Access the full text.
Sign up today, get DeepDyve free for 14 days.
Z. Vickers, E. Holton, Jian Wang (1998)
EFFECT OF YOGURT SWEETNESS ON SENSORY SPECIFIC SATIETYJournal of Sensory Studies, 13
Bernadette Marriott, N. Cole, Ellie Lee (2009)
National estimates of dietary fructose intake increased from 1977 to 2004 in the United States.The Journal of nutrition, 139 6
G. Guyatt, A. Oxman, R. Kunz, J. Woodcock, J. Brożek, M. Helfand, P. Alonso-Coello, P. Glasziou, Roman Jaeschke, E. Akl, S. Norris, G. Vist, P. Dahm, Vijay Shukla, J. Higgins, Y. Falck-Ytter, H. Schünemann (2011)
GRADE guidelines: 7. Rating the quality of evidence--inconsistency.Journal of clinical epidemiology, 64 12
Yang Pan, C. Pratt (2008)
Metabolic syndrome and its association with diet and physical activity in US adolescents.Journal of the American Dietetic Association, 108 2
V. Azaı̈s-Braesco, D. Sluik, M. Maillot, F. Kok, L. Moreno (2017)
A review of total & added sugar intakes and dietary sources in EuropeNutrition Journal, 16
Almiron-Roig (2013)
Factors that determine energy compensation: a systematic review of preload studies.Nutr Rev, 71
G. Bekkering, Ross Harris, Steve Thomas, A. Mayer, R. Beynon, A. Ness, R. Harbord, C. Bain, G. Smith, J. Sterne (2008)
How much of the data published in observational studies of the association between diet and prostate or bladder cancer is usable for meta-analysis?American journal of epidemiology, 167 9
S. Duval, R. Tweedie (2000)
Trim and Fill: A Simple Funnel‐Plot–Based Method of Testing and Adjusting for Publication Bias in Meta‐AnalysisBiometrics, 56
Guo-Chong Chen, I. Szeto, Li-Hua Chen, Shufen Han, Yan-Jie Li, R. Hekezen, L. Qin (2015)
Dairy products consumption and metabolic syndrome in adults: systematic review and meta-analysis of observational studiesScientific Reports, 5
D. Moher, A. Liberati, J. Tetzlaff, D. Altman (2009)
Preferred reporting items for systematic reviews and meta-analyses: the PRISMA StatementOpen Medicine, 3
S. Greenland, M. Longnecker (1992)
Methods for trend estimation from summarized dose-response data, with applications to meta-analysis.American journal of epidemiology, 135 11
D. Coyle, Rhoda Ndanuko, Sarinda Singh, Polly Huang, Jason Wu (2019)
Variations in Sugar Content of Flavored Milks and Yogurts: A Cross-Sectional Study across 3 CountriesCurrent Developments in Nutrition, 3
Kiyah Duffey, P. Gordon-Larsen, L. Steffen, D. Jacobs, B. Popkin (2010)
Drinking caloric beverages increases the risk of adverse cardiometabolic outcomes in the Coronary Artery Risk Development in Young Adults (CARDIA) Study.The American journal of clinical nutrition, 92 4
C. Sayón-Orea, M. Bes-Rastrollo, A. Martí, A. Pimenta, N. Martín-Calvo, M. Martínez-González (2015)
Association between yogurt consumption and the risk of Metabolic Syndrome over 6 years in the SUN studyBMC Public Health, 15
Julian Higgins, Sally Green (2010)
Cochrane Handbook for Systematic Reviews of InterventionsInternational Coaching Psychology Review
Elly Fletcher, Sarah McNaughton, Kathleen Lacy, David Dunstan, Valerie Carson, Jo Salmon (2016)
Mediating effects of dietary intake on associations of TV viewing, body mass index and metabolic syndrome in adolescentsObesity Science & Practice, 2
K. Alberti, P. Zimmet, J. Shaw (2005)
The metabolic syndrome—a new worldwide definitionThe Lancet, 366
Sylvie St‐Pierre, Jo‐Anne Gilbert, Lisa‐Anne Elvidge (2015)
Estimating the Sugars Content of Simulated Diets that Follow Eating Well with Canada's Food GuideThe FASEB Journal, 29
Canada's dietary guidelines. In: Canada's Food Guide. Health Canada
C. Tufanaru, Z. Munn, M. Stephenson, E. Aromataris (2015)
Fixed or random effects meta-analysis? Common methodological issues in systematic reviews of effectivenessInternational Journal of Evidence-Based Healthcare, 13
C. Ferreira-Pêgo, N. Babio, M. Bes-Rastrollo, D. Corella, R. Estruch, E. Ros, M. Fitó, L. Serra-Majem, F. Arós, M. Fiol, J. Santos-Lozano, C. Muñoz-Bravo, X. Pintó, M. Ruíz-Canela, J. Salas‐Salvadó (2016)
Frequent Consumption of Sugar- and Artificially Sweetened Beverages and Natural and Bottled Fruit Juices Is Associated with an Increased Risk of Metabolic Syndrome in a Mediterranean Population at High Cardiovascular Disease Risk.The Journal of nutrition, 146 8
M. Egger, G. Smith, Martin Schneider, C. Minder (1997)
Bias in meta-analysis detected by a simple, graphical testBMJ, 315
Yang-Im Hur, Hyesook Park, J. Kang, H. Lee, H. Song, Hae-Jeung Lee, Okjun Kim (2015)
Associations between Sugar Intake from Different Food Sources and Adiposity or Cardio-Metabolic Risk in Childhood and Adolescence: The Korean Child–Adolescent Cohort StudyNutrients, 8
Stroup (2000)
Meta-analysis of observational studies in epidemiology: a proposal for reporting.JAMA, 283
(2006)
Cochrane Handbook for Systematic Reviews of Interventions Edited by Julian P. T. Higgins &
S. Grundy, H. Brewer, J. Cleeman, Sidney Smith, C. Lenfant, R. Bonow, R. Eckel, D. Einhorn, S. Haffner, R. Holman, E. Horton, R. Krauss, Daniel Levy, X. Pi-Sunyer, D. Porte, G. Reaven, F. Sacks, N. Stone, P. Wilson (2004)
Definition of Metabolic Syndrome: Report of the National Heart, Lung, and Blood Institute/American Heart Association Conference on Scientific Issues Related to DefinitionCirculation: Journal of the American Heart Association, 109
G. Assmann, R. Guerra, G. Fox, P. Cullen, H. Schulte, D. Willett, S. Grundy (2007)
Harmonizing the definition of the metabolic syndrome: comparison of the criteria of the Adult Treatment Panel III and the International Diabetes Federation in United States American and European populations.The American journal of cardiology, 99 4
T. Khan, J. Sievenpiper (2016)
Controversies about sugars: results from systematic reviews and meta-analyses on obesity, cardiometabolic disease and diabetesEuropean Journal of Nutrition, 55
K. Stanhope (2016)
Sugar consumption, metabolic disease and obesity: The state of the controversyCritical Reviews in Clinical Laboratory Sciences, 53
N. Babio, N. Becerra-Tomás, Miguel Martínez-González, D. Corella, R. Estruch, E. Ros, C. Sayón-Orea, M. Fitó, L. Serra-Majem, F. Arós, R. Lamuela-Raventós, J. Lapetra, E. Gómez-Gracia, M. Fiol, A. Díaz-López, J. Sorlí, J. Martínez, J. Salas‐Salvadó (2015)
Consumption of Yogurt, Low-Fat Milk, and Other Low-Fat Dairy Products Is Associated with Lower Risk of Metabolic Syndrome Incidence in an Elderly Mediterranean Population.The Journal of nutrition, 145 10
T. Brisbois, Sandra Marsden, G. Anderson, J. Sievenpiper (2014)
Estimated Intakes and Sources of Total and Added Sugars in the Canadian DietNutrients, 6
T. Khan, L. Chiavaroli, A. Zurbau, J. Sievenpiper (2019)
A lack of consideration of a dose–response relationship can lead to erroneous conclusions regarding 100% fruit juice and the risk of cardiometabolic diseaseEuropean Journal of Clinical Nutrition, 73
K. Stanhope (2012)
Role of fructose-containing sugars in the epidemics of obesity and metabolic syndrome.Annual review of medicine, 63
Yunjin Kang, Jihye Kim (2017)
Soft drink consumption is associated with increased incidence of the metabolic syndrome only in womenBritish Journal of Nutrition, 117
S. Parvez, K. Malik, S. Kang, Hong-Yeoul Kim (2006)
Probiotics and their fermented food products are beneficial for healthJournal of Applied Microbiology, 100
C. Lortie (2022)
Doing Meta-Analysis with R - A Hands-On GuideJ. Stat. Softw., 102
Vickers (1998)
Effect of yogurt sweetness on sensory specific satiety.J Sens Stud, 13
K. Alberti, R. Eckel, S. Grundy, P. Zimmet, J. Cleeman, K. Donato, J. Fruchart, W. James, C. Loria, Sidney Smith (2009)
Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International AsCirculation, 120 16
G. Wells, G. Wells, B. Shea, B. Shea, Dianne O'Connell, J. Peterson, Welch, M. Losos, P. Tugwell, Sb Ga, G. Zello, J. Petersen (2014)
The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Nonrandomised Studies in Meta-Analyses
P. Mirmiran, Z. Bahadoran, H. Delshad, F. Azizi (2014)
Effects of energy-dense nutrient-poor snacks on the incidence of metabolic syndrome: a prospective approach in Tehran Lipid and Glucose Study.Nutrition, 30 5
R. Clemens, A. Drewnowski, M. Ferruzzi, C. Toner, D. Welland (2015)
Squeezing Fact from Fiction about 100% Fruit Juice123Advances in Nutrition, 6
P. Wilson, R. D'Agostino, H. Parise, L. Sullivan, J. Meigs (2005)
Metabolic Syndrome as a Precursor of Cardiovascular Disease and Type 2 Diabetes MellitusCirculation, 112
J. Lorenzen, A. Astrup (2011)
Dairy calcium intake modifies responsiveness of fat metabolism and blood lipids to a high-fat dietBritish Journal of Nutrition, 105
J. John, S. Ziebland, P. Yudkin, L. Roe, H. Neil (2002)
Effects of fruit and vegetable consumption on plasma antioxidant concentrations and blood pressure: a randomised controlled trialThe Lancet, 359
M. Chollet, D. Gille, A. Schmid, B. Walther, P. Piccinali (2013)
Acceptance of sugar reduction in flavored yogurt.Journal of dairy science, 96 9
(2018)
DRMETA:StataModuleforDose-ResponseMeta-analysisStatisticalSoftwareComponents
B. Appelhans, A. Baylin, Mei-hua Huang, Hong Li, I. Janssen, R. Kazlauskaite, Elizabeth Avery, H. Kravitz (2017)
Beverage Intake and Metabolic Syndrome Risk Over 14 Years: The Study of Women's Health Across the Nation.Journal of the Academy of Nutrition and Dietetics, 117 4
C. Montagnese, L. Santarpia, Margherita Buonifacio, Arturo Nardelli, A. Caldara, E. Silvestri, F. Contaldo, F. Pasanisi (2015)
European food-based dietary guidelines: a comparison and update.Nutrition, 31 7-8
Vasanti Malik, B. Popkin, G. Bray, J. Despres, W. Willett, F. Hu (2010)
Sugar-Sweetened Beverages and Risk of Metabolic Syndrome and Type 2 DiabetesDiabetes Care, 33
Z. Vickers, E. Holton, Jian Wang (2001)
Effect of ideal–relative sweetness on yogurt consumptionFood Quality and Preference, 12
D. Moher, A. Liberati, J. Tetzlaff, D. Altman (2009)
METHODS OF SYSTEMATIC REVIEWS AND META-ANALYSIS Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement
Ingrid Santaren, S. Watkins, A. Liese, L. Wagenknecht, M. Rewers, S. Haffner, C. Lorenzo, A. Festa, R. Bazinet, A. Hanley (2017)
Individual serum saturated fatty acids and markers of chronic subclinical inflammation: the Insulin Resistance Atherosclerosis Study[S]Journal of Lipid Research, 58
A. Discacciati, A. Crippa, N. Orsini (2015)
Goodness of fit tools for dose–response meta‐analysis of binary outcomesResearch Synthesis Methods, 8
E. Almiron-Roig, L. Palla, K. Guest, C. Ricchiuti, N. Vint, S. Jebb, A. Drewnowski (2013)
Factors that determine energy compensation: a systematic review of preload studiesNutrition Reviews, 71
N. Orsini, Ruifeng Li, A. Wolk, P. Khudyakov, D. Spiegelman (2012)
Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software.American journal of epidemiology, 175 1
Yue Tian, Li-Jung Su, Jiantao Wang, Xiaolin Duan, Xiubo Jiang (2017)
Fruit and vegetable consumption and risk of the metabolic syndrome: a meta-analysisPublic Health Nutrition, 21
C. Begg, Madhuchhanda Mazumdar (1994)
Operating characteristics of a rank correlation test for publication bias.Biometrics, 50 4
Scientific Advisory Committee on Nutrition. Carbohydrates and Health. Scientific Advisory Committee on Nutrition
L. Schwingshackl, G. Hoffmann, Tamara Kalle-Uhlmann, María Arregui, B. Buijsse, H. Boeing (2015)
Fruit and Vegetable Consumption and Changes in Anthropometric Variables in Adult Populations: A Systematic Review and Meta-Analysis of Prospective Cohort StudiesPLoS ONE, 10
Jun Zhang, K. Yu (1998)
What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes.JAMA, 280 19
R. Stevens, D. Dowdy, R. Michaels, P. Mendez-Tellez, P. Pronovost, D. Needham (2007)
Neuromuscular dysfunction acquired in critical illness: a systematic reviewIntensive Care Medicine, 33
A. Crippa, A. Discacciati, M. Bottai, D. Spiegelman, N. Orsini (2019)
One-stage dose–response meta-analysis for aggregated dataStatistical Methods in Medical Research, 28
Miso Lim, Jihye Kim (2019)
Association between fruit and vegetable consumption and risk of metabolic syndrome determined using the Korean Genome and Epidemiology Study (KoGES)European Journal of Nutrition, 59
G. Guyatt, A. Oxman, E. Akl, R. Kunz, G. Vist, J. Brożek, S. Norris, Y. Falck-Ytter, P. Glasziou, H. DeBeer, Roman Jaeschke, D. Rind, J. Meerpohl, P. Dahm, H. Schünemann (2011)
GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables.Journal of clinical epidemiology, 64 4
(2013)
Handbook for Grading the Quality of Evidence and the Strength of Recommendations Using the GRADE approach
Discacciati (2017)
Goodness of fit tools for dose-response meta-analysis of binary outcomes.Res Synth Methods, 8
Audrey Galvin (1980)
Consensus StatementSpine Deformity, 3
P. Huth, V. Fulgoni, D. Keast, Keigan Park, N. Auestad (2013)
Major food sources of calories, added sugars, and saturated fat and their contribution to essential nutrient intakes in the U.S. diet: data from the national health and nutrition examination survey (2003–2006)Nutrition Journal, 12
P. Lutsey, L. Steffen, J. Stevens (2008)
Dietary Intake and the Development of the Metabolic Syndrome: The Atherosclerosis Risk in Communities StudyCirculation, 117
Nur Ramli, K. Chin, Khairul Zarkasi, Fairus Ahmad (2018)
A Review on the Protective Effects of Honey against Metabolic SyndromeNutrients, 10
P. Mirmiran, E. Yuzbashian, G. Asghari, S. Hosseinpour-Niazi, F. Azizi (2015)
Consumption of sugar sweetened beverage is associated with incidence of metabolic syndrome in Tehranian children and adolescentsNutrition & Metabolism, 12
(2018)
International Association for the Study of ObesityThe Grants Register 2019
Dasom Kim, Jihye Kim (2017)
Dairy consumption is associated with a lower incidence of the metabolic syndrome in middle-aged and older Korean adults: the Korean Genome and Epidemiology Study (KoGES).The British journal of nutrition, 117 1
R. Dersimonian, N. Laird (2015)
Meta-analysis in clinical trials revisited.Contemporary clinical trials, 45 Pt A
Aditya Narain, C. Kwok, M. Mamas (2017)
Soft drink intake and the risk of metabolic syndrome: A systematic review and meta‐analysisInternational Journal of Clinical Practice, 71
Moher (2009)
Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement.Ann Intern Med, 151
Zahra Cheraghi, P. Mirmiran, M. Mansournia, N. Moslehi, D. Khalili, S. Nedjat (2016)
The association between nutritional exposures and metabolic syndrome in the Tehran Lipid and Glucose Study (TLGS): a cohort study.Public health, 140
Parvez (2006)
Probiotics and their fermented food products are beneficial for health.J Appl Microbiol, 100
Key Points Question What is the association of IMPORTANCE Sugar-sweetened beverages (SSBs) are associated with increased risk of metabolic major food sources of fructose- syndrome (MetS). However, the role of other important food sources of fructose-containing sugars in containing sugars with incident the development of MetS remains unclear. metabolic syndrome? Findings In this systematic review and OBJECTIVE To examine the association of major food sources of fructose-containing sugars with meta-analysis of 13 prospective studies incident MetS. including 49 591 participants, the adverse association of sugar-sweetened DATA SOURCES MEDLINE, Embase, and Cochrane Library were searched from database inception beverages with incident metabolic to March 24, 2020, in addition to manual searches of reference lists from included studies using the syndrome did not extend to other major following search terms: sugar-sweetened beverages, fruit drink, yogurt, metabolic syndrome, and food sources of fructose-containing prospective study. sugars. Yogurt, fruit, 100% fruit juice, and mixed fruit juice all had a protective STUDY SELECTION Inclusion criteria included prospective cohort studies of 1 year or longer that association with incident metabolic investigated the association of important food sources of fructose-containing sugars with incident syndrome. MetS in participants free of MetS at the start of the study. Meaning Generalized statements on DATA EXTRACTION AND SYNTHESIS Study quality was assessed using the Newcastle-Ottawa the adverse effects of fructose- Scale. Extreme quantile risk estimates for each food source with MetS incidence were pooled using a containing sugars cannot be random-effects meta-analysis. Interstudy heterogeneity was assessed (Cochran Q statistic) and extrapolated from sugar-sweetened quantified (I statistic). Dose-response analyses were performed using a 1-stage linear mixed-effects beverage results, as assessment of other model. The certainty of the evidence was assessed using GRADE (Grading of Recommendations, important food sources of fructose- Assessment, Development, and Evaluation). Results were reported according to the Meta-analysis of containing sugars show protective Observational Studies in Epidemiology (MOOSE) and Preferred Reporting Items for Systematic associations with metabolic syndrome Reviews and Meta-analyses (PRISMA) reporting guidelines. incidence. MAIN OUTCOMES AND MEASURES Pooled risk ratio (RR) of incident MetS (pairwise and dose Invited Commentary response). Supplemental content RESULTS Thirteen prospective cohort studies (49 591 participants [median age, 51 years; range, Author affiliations and article information are 6-90 years]; 14 205 with MetS) that assessed 8 fructose-containing foods and MetS were included. listed at the end of this article. An adverse linear dose-response association for SSBs (RR for 355 mL/d, 1.14; 95% CI, 1.05-1.23) and an L-shaped protective dose-response association for yogurt (RR for 85 g/d, 0.66; 95% CI, 0.58-0.76) and fruit (RR for 80 g/d, 0.82; 95% CI, 0.78-0.86) was found. Fruit juices (mixed and 100%) had a U-shaped dose-response association with protection at moderate doses (mixed fruit juice: RR for 125 mL/d, 0.58; 95% CI, 0.42-0.79; 100% fruit juice: RR for 125 mL/d, 0.77; 95% CI, 0.61-0.97). Honey, ice cream, and confectionary had no association with MetS incidence. The certainty of the evidence (continued) Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 1/15 JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome Abstract (continued) was moderate for SSBs, yogurt, fruit, mixed fruit juice, and 100% fruit juice and very low for all other food sources. CONCLUSIONS AND RELEVANCE The findings of this meta-analysis suggest that the adverse association of SSBs with MetS does not extend to other food sources of fructose-containing sugars, with a protective association for yogurt and fruit throughout the dose range and for 100% fruit juice and mixed fruit juices at moderate doses. Therefore, current policies and guidelines on the need to limit sources of free sugars may need to be reexamined. JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 Introduction Metabolic syndrome (MetS) is a cluster of major health risk factors associated with an increased incidence of type 2 diabetes and cardiovascular disease. Although the definition and criteria for 2,3 identifying MetS can vary, all definitions consider important risk factors, including large waist circumference, elevated blood pressure, low high-density lipoprotein level, elevated levels of triglycerides, and hyperglycemia. Fructose-containing sugars (eg, sucrose and high-fructose corn syrup) in the diet have been 4,5 implicated as potential contributing factors to increased MetS risk. There is strong evidence that sugar-sweetened beverages (SSBs), a major source of fructose in the North American diet, are associated with increased incident MetS. The role of other important food sources of fructose- containing sugars in the development of MetS, however, has yet to be fully elucidated. This systematic review and dose-response meta-analysis of prospective cohort studies examines the association of food sources of fructose-containing sugars and incident MetS and evaluates the strength and quality of the evidence using GRADE (Grading of Recommendations, Assessment, Development, and Evaluation). Methods Data Sources and Searches This meta-analysis followed the Cochrane Handbook for Systematic Reviews of Interventions. Results were reported according to the Meta-analysis of Observational Studies in Epidemiology (MOOSE) and 9,10 Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guidelines. The study protocol was registered at ClinicalTrials.gov. Data sources included MEDLINE, Embase, and the Cochrane Library from database inception to March 24, 2020. Manual search of the reference lists from included studies supplemented the database search. Search terms reflected the most consumed food sources of fructose-containing sugars (based on national surveys that outlined the leading 12-14 fructose-containing added or free sugar foods ) (eg, sugar-sweetened beverages, fruit drink, and yogurt), the outcome of interest (eg, metabolic syndrome), and the study design (eg, prospective study) (eTable 1 in the Supplement). Study Selection Prospective cohort studies of 1 year or longer that investigated the association of major food sources of fructose-containing sugars with incident MetS in participants free of MetS at baseline were included (eTable 1 in the Supplement). If multiple publications of the same cohort provided results on the same outcome with overlapping groups of individuals, the longest follow-up study was included. Abstracts and unpublished studies were excluded. JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 2/15 JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome Data Extraction and Quality Assessment Two independent reviewers (Z.S.A. and T.A.K) extracted relevant data, including sample size, participant characteristics, food source of fructose-containing sugars, exposure levels, follow-up duration, number of MetS cases, covariates in fully adjusted models, and the relative risk (RR) with 95% CIs of incident MetS per category of intake, median dose in each category, and funding source, dual-sequentially. Studies were assessed for risk of bias using the Newcastle-Ottawa Scale. Newcastle-Ottawa Scale points were awarded based on cohort selection, adequacy of outcome measures, and comparability of cohorts regarding design or analysis. A maximum of 9 points could be awarded, with 0 points indicating lowest study quality and 9 points indicating highest study quality. A score of 6 points was the minimum threshold for the study to be considered higher quality. Disagreements were resolved by consensus or by involving a third person (J.L.S.). The GRADE approach was used to assess the overall certainty and strength of the evidence, ranging from high to very low certainty (eAppendix 1 in the Supplement). Statistical Analysis Pairwise meta-analyses and sensitivity analyses were conducted in R software, version 3.6.1 (R Foundation for Statistical Computing) using dmetar. Dose-response analyses were conducted in Stata software, version 16 (StataCorp) using drmeta. Each food source of fructose-containing sugar was considered as an independent exposure. Risk ratios (RRs) of extreme quantiles from the most adjusted models were used for pairwise analyses. When studies used continuous RRs per dose, we imputed the extreme quantiles from other publications of the same or similar cohort. Hazard ratios and odds ratios were converted to RRs based on the recommended method by Zhang and Yu (eAppendix 2 in the Supplement). Summary estimates were determined by natural log transforming and pooling the RRs using the DerSimonian and Laird random-effects model. A fixed-effects model was used if the number of studies was 5 or fewer. Unit-of-analysis error (for studies that appeared more than once in the same food source analysis) was addressed by dividing participants equally among the multiple comparisons and readjusting the log SEs. Interstudy heterogeneity was assessed using the Cochran 2 2 2 Q (χ ) statistic and quantified by the I statistic, where I of 50% or greater and P < .10 determined by the Q statistic represented evidence of substantial heterogeneity. Sources of heterogeneity were assessed by sensitivity analyses that involved the systematic removal of each study for food sources with more than 2 cohorts. If 10 or more cohort comparisons were available, a priori subgroup analyses were performed. If 10 or more cohort comparisons were available, studies were assessed for publication bias by 22,23 visual inspection of funnel plots and formal testing using the Begg and Egger tests, with significance set at P < .10. In the presence of publication bias, the Duval and Tweedie trim and fill method was used. Dose responses were modeled using RRs (95% CIs) from dose categories to determine the shape of the association between the dose of the fructose-containing foods and the risk of MetS 25,26 (eAppendix 3 in the Supplement). Doses were defined as the mean consumption in each reported category or quantile. We reported nonlinear associations for a study if results of the Wald test for departure from linearity were significant at P < .10 (2-sided). The significance for the main pooled RR for the pairwise analyses was based on P < .05. Results Search Results 28-40 Thirteen reports (49 591 participants and 14 205 cases) with data from 8 unique prospective cohorts met the inclusion criteria (Figure 1). Eight major food sources of fructose-containing sugars were identified, including SSBs (7 cohort comparisons; 20 480 participants and 7406 28,32,34,36-38 32,33,38 cases ), mixed fruit juice (3 cohort comparisons; 3062 participants and 1322 cases ), JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 3/15 JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome 31,32,40 100% fruit juice (2 cohort comparisons; 5464 participants and 1389 cases ), fruit (4 cohort 30,33,40 comparisons; 10 074 participants and 3002 cases ), yogurt (5 cohort comparisons; 19 057 29,30,35,39 30 participants and 3877 cases ), honey (1 cohort; 3616 participants and 590 cases ), ice cream (1 cohort; 3616 participants and 590 cases ), and confectionary (2 cohort comparisons; 1476 participants and 250 cases ). Prospective cohort studies that assessed grain and grain-based products or other fruit- or dairy-based products with incident MetS were not identified. Study Characteristics 28-40 The Table gives the characteristics of the 13 prospective cohort studies. Studies included data 28,31,36 29,32,39 30,37,38 33-35,40 from the US, Spain, Iran, and South Korea. Participants ranged from adolescents to older adults (median age, 51 years; range, 6-90 years). Appelhans et al exclusively studied a female cohort. The mean (SD) duration of follow-up was 5.7 (3.3) years (range, 2.0-14.0 years). Fruit juice was considered to be mixed fruit juice if the study combined fruit drinks and fruit juice or did not specify the kind of fruit juice (100% fruit juice or fruit drink). Yogurt was considered a source of fructose given that more than 70% of the yogurt products are flavored and consumers prefer yogurt products with a moderate (approximately 7%-10%) concentration of added 42-44 31,34-38,40 sucrose. MetS was defined using the Adult Treatment Panel III, harmonized 28-30,32,39 33 criteria, or a continuous scale (eAppendix 4 in the Supplement). All studies were agency funded. 28,29,31-40 30 All studies, except for the study by Cheraghi et al, adjusted for age and multiple prespecified primary confounding variables, including sex, markers of obesity, smoking, family history of MetS, energy or calorie intake, diabetes, physical activity, and alcohol intake (eTable 2 in the Supplement). Between 4 and 26 variables were adjusted for in fully adjusted models of the 12 28,29,31-40 studies that detailed their statistical process. Risk of Bias None of the studies were rated as high risk of bias (eTable 3 in the Supplement). Statistical tests for publication bias could not be assessed for any food source because of 10 or fewer cohort comparisons. Figure 1. Diagram of Study Selection 675 Reports identified 163 MEDLINE (through March 24, 2020) 448 Embase (through March 24, 2020) 614 Excluded based on title and/or abstract 56 Cochrane (through March 24, 2020) 254 Wrong exposure 8 Manual search (through March 24, 2020) 174 Wrong end point 36 Reviews 35 Intervention studies 32 Cross-sectional studies 26 Meta-analyses 16 Nonhumans 15 Conferences or meetings 11 No abstract available 7 Casecontrol studies 3 Retrospective studies 3 Abstracts 2 Surveys 61 Read in full 48 Excluded 29 Wrong exposure 7 Cross-sectional studies 7 Wrong end point 4 Abstracts 1 Conference or meeting 13 Included in the analysis 49 591 Participants 14 205 MetS cases 13 Cohorts MetS indicates metabolic syndrome. JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 4/15 JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 5/15 Table. Characteristics of Prospective Cohort Studies Investigating Dietary Intake of Food Sources of Fructose-Containing Sugars and MetS No. of Follow-up MetS Baseline age Funding Source Cohort name Country duration, y Sex No. of participants cases range, y Dietary assessment Food source MetS assessment source 28 a c d Appelhans et al, 2017 SWAN US 14 Female 1401 268 42-52 FFQ (interviewer SSB Harmonized criteria Agency administered) 29 b Babio et al, 2015 PREDIMED Spain 3.2 Both 1868 930 Male: 55-80; SFFQ Yogurt Harmonized criteria Agency female: 60-80, 30 b Cheraghi et al, 2016 TLGS Iran 2.05 Both 3616 590 ≥20 FFQ (interviewer Fruit, yogurt, ice Harmonized criteria Agency administered) cream, honey 31 a Duffey et al, 2010 CARDIA US 7 Both 3596 459 18-30 SFFQ (interviewer 100% fruit juice ATP III Agency administered) 32 b Ferreira-Pêgo et al, 2016 PREDIMED Spain 3.24 Both 1868 930 55-80 SFFQ SSB, mixed fruit Harmonized criteria Agency juice,100% fruit juice 33 a Hur et al, 2016 KoCAS South Korea 4 Both 770 345 9-10 3-d FR Fruit sugar, cMET Agency beverage sugar 34 a Kang and Kim, 2017 KoGES South Korea 5.7 Both 5797 2129 40-69 SFFQ SSB ATP III Agency 35 a Kim and Kim, 2017 KoGES South Korea 5.7 Both 5510 2103 40-69 SFFQ Yogurt ATP III Agency 40 a Lim and Kim, 2019 KoGES South Korea 8 Both 5688 2067 40-69 SFFQ Fruit ATP III Agency 36 a Lutsey et al, 2008 ARIC US 9 Both 9514 3782 45-64 FFQ SSB ATP III Agency 37 a Mirmiran et al, 2014 TLGS Iran 3 Both 1476 249 19-70 SFFQ Biscuits and ATP III with specific Agency cakes, candies waist circumference and chocolate, cutoffs for Iranian SSB adults 38 a Mirmiran et al, 2015 TLGS Iran 3.6 Both 424 47 6-18 SFFQ SSB, mixed fruit ATP III adapted Agency juice definition for adolescents 39 a Sayón-Orea et al, 2015 SUN Spain 6 Both 8063 306 20-90 SFFQ Yogurt Harmonized criteria Agency Abbreviations: ARIC, Atherosclerosis Risk in Communities Study; ATP III, Adult Treatment Panel III; CARDIA, Mean value. Coronary Artery Risk Development in Young Adults; cMET, continuous MetS score; FFQ, Food Frequency Median value. Questionnaire; FR, food records; KoCAS, Korean Child-Adolescent Cohort Study; KoGES, Korean Genome and Harmonized criteria of the American Heart Association/National Heart, Lung, and Blood Institute, and the Epidemiology Study; MetS, metabolic syndrome; PREDIMED, Prevención con Dieta Mediterránea; TLGS, Tehran International Diabetes Federation definitions for metabolic syndrome. Lipid and Glucose Study; SFFQ: Semiquantitative Food Frequency Questionnaire; SSB, sugar-sweetened beverage; Agency funding is that from government, university, or not-for-profit health agency. SUN, Sequimiento University of Navarra; SWAN, Study of Women’s Health Across the Nation. JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome Important Food Sources of Fructose-Containing Sugars and Incident MetS Figure 2 and eFigures 1 through 8 in the Supplement illustrate the association between food sources of fructose-containing sugars and incident MetS. Intake of SSBs was associated with an increased risk of incident MetS (RR, 1.20; 95% CI, 1.06-1.36), with evidence of significant heterogeneity (I =68%; P = .005 determined by the Q statistic). Fruit and yogurt intake had an inverse association with incident MetS (fruit: RR, 0.91; 95% CI, 0.89-0.93; I = 0%; P = .78 determined by the Q statistic; yogurt: RR, 0.83; 95% CI, 0.77-0.90; I =65%; P = .02 determined by the Q statistic). No association was found between mixed fruit juice, 100% fruit juice, honey, ice cream, or confectionary with MetS incidence. Sensitivity Analyses eTable 5 in the Supplement details the sensitivity analysis after systematic removal of each cohort study for food sources with more than 2 studies. Results for SSBs did not alter in direction and significance of association (eg, maintained an adverse association with MetS incidence) or the evidence of heterogeneity. Similar results were found for mixed fruit juice and fruit, where removal of each study maintained no association for mixed fruit juice and a significant protective association for fruit. Heterogeneity in both mixed fruit juice and fruit remained nonsignificant. Removal of the study by Cheraghi et al resulted in nonsignificant evidence of interstudy heterogeneity for yogurt; however, it did not significantly affect the pooled estimate. Because none of the comparisons had 10 or more cohorts, subgroup analyses were not performed. Dose Response Figure 2 and Figure 3 show the dose-response association of each food source and incident MetS. 32,34,36-38 Data from 6 cohorts, with a dose range of 0 to 680 mL/d, demonstrated an adverse linear dose-response association between SSB intake and MetS (RR for 355 mL/d, 1.14; 95% CI, 1.05-1.23), with no evidence for departure from linearity (P = .27) (Figure 3). 32,38 32 Data for mixed fruit juice (2 cohorts ) and 100% fruit juice (1 cohort ) indicate a U-shaped, significant, nonlinear dose-response association with incident MetS, with the curve suggesting a maximum protection between 75 and 150 mL. There was no protective association after 200 mL/d for mixed fruit juice intake and after 175 mL/d for 100% fruit juice. The estimated RR for 125 mL/d was 0.58 (95% CI, 0.42-0.79) for mixed fruit juice and 0.77 (95% CI, 0.61-0.97) for 100% fruit juice. Data from 2 cohorts with a dose range of 0 to 600 g/d found a significant L-shaped, protective, nonlinear dose response for fruit intake and incident MetS, suggesting a sharp reduction of RR until 450 g/d. The estimated RR for 240 g (3 servings) was 0.61 (95% CI, 0.55-0.68). 29,35 Data from 3 cohorts with a dose range of 0 to 129 g/d of yogurt intake found an L-shaped, protective, nonlinear dose-response association with incident MetS, with the curve suggesting a sharp reduction of RR until 80 g/d. The estimated RR for 85 g (one-third cup serving) was 0.66 (95% CI, 0.58-0.76). Confectionary data from 2 cohorts with a dose range of 8 to 84 g/d found no evidence of a dose-response association with incident MetS (RR per 50 g, 1.18; 95% CI, 0.98-1.42). Relevant data were not available to assess the dose-response association for honey and ice cream. GRADE Assessment The GRADE certainty of evidence was moderate for adverse association for SSBs and protective association for mixed fruit juice, 100% fruit juice, fruit, and yogurt with MetS risk attributable to upgrades for dose-response gradient (Figure 2 and eTable 4 in the Supplement). Although both SSBs and 100% fruit juice had substantial interstudy heterogeneity (I = 68% for SSBs and 73% for fruit juice), the RR estimates for SSB studies were all in the same direction with considerable overlap. In addition, the heterogeneity observed with 100% fruit juice was explained by the nonlinear dose- response model. Therefore, these 2 food sources were not downgraded for inconsistency. The certainty of evidence of no association was very low for honey, ice cream, and confectionary because JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 6/15 JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 7/15 Figure 2. Summary Superplot for the Association Between Important Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome Grade Downgrade Upgrade No. of cases/ Favors Favors Heterogeneity No. No. of Food source and Risk ratio positive adverse Certainty of 2 Q of CC participants comparison (95% CI) P l , % P association association the evidence Sugar-sweetened beverage Moderate Extreme quantiles 7 7406/20 480 1.21 (1.06-1.31) .005 68 .005 Linear DRM per 355 mL 1.14 (1.05-1.23) .001 Mixed fruit juice Moderate Extreme quantiles 3 1322/3062 1.13 (0.91-1.41) .27 0 .87 Nonlinear DRM at 50 mL 0.67 (0.54-0.83) <.001 Nonlinear DRM at 125 mL 0.58 (0.42-0.79) Nonlinear DRM at 200 mL 0.87 (0.69-1.09) 100% Fruit juice Moderate Extreme quantiles 2 1389/5464 1.02 (0.92-1.13) .66 73 .052 Nonlinear DRM at 50 mL 0.82 (0.72-0.94) <.001 Nonlinear DRM at 125 mL 0.77 (0.61-0.97) Nonlinear DRM at 200 mL 1.14 (0.90-1.45) Fruit Moderate Extreme quantiles 4 3002/10 074 0.91 (0.89-0.93) <.001 0 .78 Nonlinear DRM at 80 g 0.82 (0.78-0.86) <.001 Nonlinear DRM at 240 g 0.61 (0.55-0.68) Nonlinear DRM at 480 g 0.53 (0.47-0.60) Yogurt Moderate Extreme quantiles 5 3877/19 057 0.83 (0.77-0.90) <.001 .65 .02 Nonlinear DRM at 35 g 0.76 (0.69-0.84) <.001 Nonlinear DRM at 60 g 0.68 (0.60-0.78) 0.66 (0.58-0.76) Nonlinear DRM at 85 g Honey Very low Extreme quantiles 1 590/3616 1.00 (0.50-2.00) .99 NA NA Ice cream Very low Extreme quantiles 1 590/3616 0.94 (0.84-1.06) .31 NA NA Confectionary Very low Extreme quantiles 2 250/1476 1.21 (0.92-1.60) .17 0 .60 Linear DRM per 50 g 1.18 (0.98-1.42) .07 0.5 0.6 0.8 1 1.2 1.6 2 Risk ratio (95% CI) 2 45 Pooled risk estimate is represented by the data markers. I values of 50% or greater (P < .10 determined by the Q statistic) indicate substantial heterogeneity, and risk ratios greater than 1.00 indicate an adverse association. The Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) of prospective cohort studies are rated as low certainty of evidence and can be downgraded by 5 domains and upgraded by 3 domains. The filled black squares indicate downgrade and/or upgrades for each outcome. DRM indicate dose response meta-analysis; NA, not applicable. Risk of bias Inconsistency Indirectness Imprecision Publication bias Dose response Large effect size Attenuation JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome of downgrades for serious imprecision, indirectness for honey, ice cream, and confectionary with no upgrades. Discussion In our systematic review and meta-analysis, 13 prospective cohort studies (including 49 591 participants and 14 205 MetS cases) found that SSB intake was associated with an increased risk for MetS incidence, whereas yogurt and fruit were associated with a reduced risk. Mixed fruit juice and 100% fruit juice had a U-shaped association with MetS, presenting a protective association between 75 and 150 mL/d and an adverse association for more than 175 to 200 mL/d. No association was found between honey, ice cream, and confectionary and MetS incidence. The adverse association of SSB intake and MetS incidence in our study is consistent with the 6 6,46 current literature. Previous meta-analyses found a 20% and 46% increased MetS risk with higher SSB consumption from 3 prospective and 8 cross-sectional studies, respectively. Our findings expand on current findings by the inclusion of 7 prospective cohorts and the assessment of dose response, which found a 14% increased risk of MetS incidence per 355-mL daily serving of SSBs. The association between SSB and incident MetS may reflect a general unhealthy lifestyle whereby individuals with greater SSB intake are likely to have a poorer diet quality, higher caloric intake, and a sedentary lifestyle. Furthermore, SSBs are a source of liquid calories, which can have a lower effect on satiety compared with solid foods, resulting in increased energy intake, weight gain, 48 28-40 and downstream complications related to MetS. Although the prospective studies included Figure 3. Dose-Response Association of Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome A Sugar-sweetened beverage B Mixed fruit juice C 100% Fruit juice 1.5 1.8 1.3 1.4 1.4 1.3 1.0 0.8 1.2 1.0 1.1 0.6 0.8 1.0 0.4 0.6 0.9 0 100 200 300 400 500 600 700 0 50 100 150 200 250 0 50 100 150 200 250 Consumption, mL/d Consumption, mL/d Consumption, mL/d D Fruit E Yogurt F Confectionary 1.0 1.0 1.8 0.9 0.9 1.6 0.8 0.8 1.3 0.7 0.7 1.0 0.6 0.6 0.8 0.5 0.5 0.6 0 100 200 300 400 500 600 0 20 40 60 80 100 120 140 0 20 40 60 80 100 Consumption, g/d Consumption, g/d Consumption, g/d Dose-response association between intake of sugar-sweetened beverages (linearity: risk from linearity: RR per 80 g, 0.82; 95% CI, 0.78-0.85; P < .001) (D), yogurt (linearity: RR ratio [RR] per 355 mL, 1.14; 95% CI, 1.05-1.23; P = .001; departure from linearity: RR per per 85 g, 0.92; 95% CI, 0.91-0.94; P < .001; departure from linearity: RR per 85 g, 0.66; 355 mL, 1.16; 95% CI, 1.07-1.26; P = .27) (A), mixed fruit juice (linearity: RR per 125 mL, 95% CI, 0.58-0.76; P < .001) (E), and confectionary (linearity: RR per 50 g, 1.18; 95% CI, 1.00; 95% CI, 0.87-1.14; P = .96; departure from linearity: RR per 125 mL, 0.58; 95% CI, 0.98-1.42; P = .07; departure from linearity: RR per 50 g, 0.96; 95% CI, 0.71-1.30; P = .17) 0.42-0.79; P < .001) (B), 100% fruit juice (linearity: RR per 125 mL, 1.09; 95% CI, (F) with the risk of metabolic syndrome. The solid orange line represents the linear model 0.93-1.27; P = .31; departure from linearity: RR per 125 mL, 0.77; 95% CI, 0.51-0.97; and the blue line the nonlinear model. Dotted lines represent 95% CIs for the P < .001) (C), fruit (linearity: RR per 80 g, 0.92; 95% CI, 0.91-0.94; P < .001; departure nonlinear model. JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 8/15 Relative risk Relative risk Relative risk Relative risk Relative risk Relative risk JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome in our SSB analysis controlled for potential confounding factors, all except 1 study controlled for 28,36 total energy intake, and 2 studies did not adjust for adiposity, an important risk factor and component of MetS. Thus, residual and unmeasured confounding could have contributed to the observed adverse association. Conversely, yogurt had a protective association against MetS incidence, with a dose-dependent benefit with intakes of 60 to 80 g/d. The nonlinear findings indicate that the association above 85 g/d plateaus, and data are lacking to suggest any benefit associated with increasing intake beyond this dose. The role of yogurt, or more broadly dairy intake, and MetS has gained attention during the past decade. A meta-analysis highlighted that higher dairy consumption was inversely associated with MetS incidence by 14% among 7 prospective cohorts with a dose-response reduction with incremental intake. Our findings broadly concur with these results. This protective association of yogurt may be attributable to its rich micronutrient composition. Calcium, a major nutrient in yogurt, decreases fat absorption, lowers triglyceride concentration, improves the overall ratio of high- density lipoprotein to low-density lipoprotein. In addition, dairy-derived saturated fats have anti- inflammatory properties and potentially improve insulin sensitivity and glycemic response. Furthermore, the probiotic bacteria found in yogurt products have been linked to modulating gut microbiota through the reduction of pathogenic bacteria while increasing metabolite production and modulating various inflammatory reactions, all of which can aid in reducing the risk of MetS. Similarly, fruit consumption presented a protective association against MetS incidence, with the greatest dose benefit at 300 to 450 g/d (equivalent to 3-5 servings). Fruit intake is protective for 53 54 some components of MetS, including waist circumference and blood pressure ; however, 40 36,55,56 evidence on the dose range with MetS is limited. Most fruit intake and MetS studies are cross-sectional in design or are assessed in combination with vegetable intake, making it difficult to determine the association of specific fructose-containing fruits with MetS. One such meta-analysis of cross-sectional studies found that fruit intake had a protective association with MetS risk. We identified a U-shaped dose-dependent association with mixed fruit juice and 100% fruit juice intake, showing protective associations against MetS with intakes less than 200 mL. The benefit of 100% fruit juice seen at moderate doses may be attributable to the range of fruit-derived nutrients and bioactive compounds in fruit juice, and the potential for harm at higher doses may be attributable to the consumption of excess calories outweighing any benefit of these bioactive nutrients. Mixed fruit juices are a combination of fruit drinks (which are similar to SSBs because they are sugary drinks without the accompanying nutrients) and 100% fruit juice. The observed moderate doses of intake may represent the beneficial nutrients from natural fruit within the mixed fruit juice, thus indicating an association similar to that of 100% fruit juices rather than SSBs. The lack of linear association in 100% fruit juice and mixed fruit juice underscores that without consideration of the dose-response association, a naive analysis of extreme intakes assumes a false-linear association and fails to detect important dose ranges for protection or harm. Furthermore, honey, ice cream, and confectionary intake was not associated with MetS incidence. Although animal models suggest potential protective effects of honey in MetS, to our knowledge, only 1 prospective cohort study assessed honey with MetS incidence and found no significant association. Similarly, the current limited evidence indicates that ice cream and confectionary were not significantly associated with MetS incidence. Future data might clarify our association, particularly for confectionaries, for which CIs did not eliminate significant harm. The protective and neutral association in our results highlight 2 important considerations. First, the small beneficial effects of some foods might be driven by catalytic doses of fructose intake. Second, the food composition is important. SSBs are without beneficial nutrients and thus offer an unchecked source of fructose-containing sugar, whereas in other foods (eg, yogurt), nutrients other than sugars (eg, polyphenols, minerals, and fiber) may offer protection that might overcome harms from added sugars. More data are needed to enable a complete dose-response assessment and JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 9/15 JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome reveal dose ranges for increased or reduced risk, depending on the balance between nutrient matrixes vs excess sugars. Strengths and Limitations There are numerous strengths associated with our study. To our knowledge, this study is the first meta-analysis to comprehensively compare major food sources of fructose-containing sugars with incident MetS in prospective cohort studies. We conducted a thorough literature search, performed quantitative synthesis, and assessed the certainty of the evidence using GRADE. Selected studies included a large sample size, long follow-up durations, and adjustment for multiple lifestyle factors. We also assessed dose responses for all food sources and identified ranges and cutoffs for benefit and harm. This study also had some limitations. The observational nature of prospective cohort studies may result in unmeasured and residual confounding and may suffer from reverse causality. Thus, GRADE evaluation for observational studies is low certainty of evidence. Although SSBs, yogurt, and 100% fruit juice had substantial interstudy heterogeneity, we did not consider this as a serious inconsistency. The estimates were all in the same direction, and there was considerable overlap for SSB and yogurt. The nonlinear dose-response model explained the heterogeneity for yogurt and 100% fruit juice. Honey, ice cream, and confectionary were downgraded for serious indirectness for the inability to assess inconsistency because only 1 study was available for each exposure. Furthermore, they were downgraded for serious imprecision, indicating no association with MetS incidence in the extreme quantile analysis. The CIs were wide and could not conclude clinically important harm for confectionary or clinically important benefit or harm for honey and ice cream. In our dose-response analysis, we found a significant linear dose response of harm for SSBs and a nonlinear dose response of benefit for mixed fruit juice, 100% fruit juice, fruit, and yogurt, leading to an upgrade for the certainty of evidence. Data were not available for grain and grain-based products, a leading source of sugar. Conclusions Our study provides supporting evidence that increased SSB consumption is associated with MetS incidence. Generalizing statements on the adverse effects of fructose-containing sugars, however, cannot be extrapolated to other major food sources of fructose-containing sugars. Furthermore, our dose-response assessment found that mixed and 100% fruit juice presented consistent dosage for benefit that align with some national nutrition guidelines, suggesting that a 150-mL intake may 63,64 contribute toward the recommended daily fruit consumption. Thus, well-intentioned policies and guidelines to limit sources of free sugars, such as fruit juice or sweetened yogurts, based on evidence from SSBs may need to be reexamined with a food-based lens, such as those of the new 65 66 Canada’s Food Guide or Scientific Advisory Committee on Nutrition. Additional prospective studies are needed to improve our estimates and better understand the dose-response association between important food sources of fructose-containing sugars and MetS. Moreover, high-quality, large randomized clinical trials are needed on other fructose-containing foods. Furthermore, studies of whole diets and dietary patterns that consist of various food sources of fructose-containing sugars with cardiometabolic-related health outcomes can also contribute to the evidence of the association of these diets with MetS. ARTICLE INFORMATION Accepted for Publication: April 29, 2020. Published: July 9, 2020. doi:10.1001/jamanetworkopen.2020.9993 JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 10/15 JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Semnani-Azad Zetal. JAMA Network Open. Corresponding Author: John L. Sievenpiper MD, PhD, Toronto 3D Knowledge Synthesis and Clinical Trials Unit, Risk Factor Modification Centre, St Michael’s Hospital, 6137-61 Queen St E, Toronto, ON, M5C 2T2, Canada (john. sievenpiper@utoronto.ca). Author Affiliations: Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada (Semnani-Azad, Khan, Blanco Mejia, de Souza, Leiter, Kendall, Hanley, Sievenpiper); Toronto 3D Knowledge Synthesis and Clinical Trials Unit, Risk Factor Modification Centre, St Michael’s Hospital, Toronto, Ontario, Canada (Khan, Blanco Mejia, de Souza, Leiter, Kendall, Sievenpiper); Department of Health Research Methods, Evidence, and Impact, Faculty of Health Sciences, McMaster University, Hamilton, Ontario, Canada (de Souza); Population Health Research Institute, Hamilton, Ontario, Canada (de Souza); Division of Endocrinology and Metabolism, University of Toronto, Toronto, Ontario, Canada (Leiter, Hanley, Sievenpiper); Li Ka Shing Knowledge Institute, St Michael’s Hospital, Toronto, Ontario, Canada (Leiter, Sievenpiper); Division of Nutrition and Dietetics, College of Pharmacy and Nutrition, University of Saskatchewan, Saskatoon, Saskatchewan, Canada (Kendall); Leadership Sinai Centre for Diabetes, Mount Sinai Hospital, Toronto, Ontario, Canada (Hanley). Author Contributions: Dr Sievenpiper had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Semnani-Azad, Khan, Sievenpiper. Acquisition, analysis, or interpretation of data: All authors. Drafting of the manuscript: Semnani-Azad, Khan. Critical revision of the manuscript for important intellectual content: All authors. Statistical analysis: Semnani-Azad, Khan, Blanco Mejia. Obtained funding: Sievenpiper. Administrative, technical, or material support: Blanco Mejia, Kendall, Hanley. Supervision: Sievenpiper. Conflict of Interest Disclosures: Dr Khan reported receiving grants from the Canadian Institutes of Health Research and the International Life Science Institute and nonfinancial support from the Calorie Control Council outside the submitted work. Dr de Souza reported receiving grants from the Canadian Foundation for Dietetic Research, the Canadian Institutes for Health Research, the Hamilton Health Sciences Corporation, and the Hamilton Health Sciences Corporation/Population Health Research Institute; personal fees and nonfinancial support from the World Health Organization; personal fees from McMaster Children's Hospital; and speaker fees from the University of Toronto, the College of Family Physicians of Canada, Royal College, and McMaster Children’s Hospital. Dr de Souza also served as an external resource person to the World Health Organization’s Nutrition Guidelines Advisory Group on trans fats, saturated fats, and polyunsaturated fats. The World Health Organization paid for his travel and accommodation to attend meetings from 2012-2017 to present and discuss this work. He also serves as an independent director of the Helderleigh Foundation (Canada). Dr Kendall reported receiving grants from the Advanced Food Materials Network, the Agriculture and Agri-Foods Canada, the Canadian Institutes of Health Research, the Canola Council of Canada, and the National Dried Fruit Trade Association; grants and nonfinancial support from Almond Board of California, Barilla, the International Tree Nut Council Research and Education Foundation, Loblaw Brands Ltd, Pulse Canada, and Unilever; grants, nonfinancial support, and travel support from the International Nut and Dried Fruit Council; nonfinancial support and travel support from the American Peanut Council; nonfinancial support from the California Walnut Commission, Danone, Kellogg Canada, Nutrartis, Primo, Quaker, Unico, and Upfield; travel support from the International Pasta Organization, the Oldways Preservation Trust, and The Peanut Institute; and personal fees from the McCormick Science Institute outside the submitted work and serving on the scientific advisory board for the International Pasta Organization, McCormick Science Institute, and Oldways Preservation Trust. Dr Kendall is a member of the International Carbohydrate Quality Consortium, Executive Board of the Diabetes and Nutrition Study Group of the European Association for the Study of Diabetes, and the Clinical Practice Guidelines Expert Committee for Nutrition Therapy of the European Association for the Study of Diabetes and is a director of the Toronto 3D Knowledge Synthesis and Clinical Trials Foundation. Dr Hanley reported receiving grants from Dairy Farmers of Canada outside the submitted work. Dr Sievenpiper reported receiving grants from the Canadian Institutes of Health Research, Diabetes Canada, Canadian Institutes of Health Research, Canadian Foundation for Innovation, Ontario Research Fund, Province of Ontario Ministry of Research and Innovation and Science, PSI Foundation, Banting and Best Diabetes Centre, American Society for Nutrition; grants from the International Nut and Dried Fruit Council Foundation, the National Dried Fruit Trade Association, The Tate and Lyle Nutritional Research Fund at the University of Toronto, The Glycemic Control and Cardiovascular Disease in Type 2 Diabetes Fund at the University of Toronto (a fund established by the Alberta Pulse Growers), and the Nutrition Trialists Fund at the University of JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 11/15 JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome Toronto (a fund established by an inaugural donation from the Calorie Control Council) outside the submitted work; nonfinancial support from the Almond Board of California, the California Walnut Commission, the American Peanut Council, Barilla, Unilever, Upfield, Nutrartis, Unico/Primo, Loblaw Companies, Quaker, Kellogg Canada, and WhiteWave Foods; and personal fees from Dairy Farmers of Canada, FoodMinds LLC, International Sweeteners Association, Nestlé, Pulse Canada, Canadian Society for Endocrinology and Metabolism, GI Foundation, Abbott, Biofortis, American Society for Nutrition, Northern Ontario School of Medicine, Nutrition Research & Education Foundation, European Food Safety Authority, Comité Européen des Fabricants de Sucre, Physicians Committee for Responsible Medicine, Perkins Coie LLP, Tate & Lyle, Wirtschaftliche Vereinigung Zucker e.V, European Fruit Juice Association, and Soy Nutrition Institute. He is on the Clinical Practice Guidelines Expert Committees of Diabetes Canada, European Association for the Study of Diabetes, Canadian Cardiovascular Society, and Obesity Canada. He serves or has served as an unpaid scientific adviser for the Food, Nutrition, and Safety Program and the Technical Committee on Carbohydrates of the International Life Science Institute North America. He is a member of the International Carbohydrate Quality Consortium, Executive Board of the Diabetes and Nutrition Study Group of the European Association for the Study of Diabetes, and a director of the Toronto 3D Knowledge Synthesis and Clinical Trials Foundation. His wife is an employee of AB InBev. No other disclosures were reported. Funding/Support: This work was funded by grant 129920 from the Canadian Institutes of Health Research. The Diet, Digestive Tract, and Disease (3-D) Centre, funded through the Canada Foundation for Innovation and the Ministry of Research and Innovation's Ontario Research Fund, provided the infrastructure for the conduct of this project. Ms Semnani-Azad has received funding by the Canadian Institutes of Health Research Graduate Scholarships, Ontario Graduate Scholarship, and the University of Toronto Banting and Best Scholarship. Dr Khan has received the Toronto 3D Knowledge Synthesis and Clinical Trials Post-Doctoral Fellowship. Dr Sievenpiper was funded by a PSI Graham Farquharson Knowledge Translation Fellowship, Canadian Diabetes Association Clinician Scientist Award, Canadian Institutes of Health Research Institute of Nutrition, Metabolism and Diabetes/Canadian Nutrition Society New Investigator Partnership Prize, and Banting & Best Diabetes Centre Sun Life Financial New Investigator Award. Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. REFERENCES 1. Wilson PWDA, D’Agostino RB, Parise H, Sullivan L, Meigs JB. Metabolic syndrome as a precursor of cardiovascular disease and type 2 diabetes mellitus. Circulation. 2005;112(20):3066-3072. doi:10.1161/ CIRCULATIONAHA.105.539528 2. Grundy SMB, Brewer HB Jr, Cleeman JI, Smith SC Jr, Lenfant C; National Heart, Lung, and Blood Institute; American Heart Association. Definition of metabolic syndrome: report of the National Heart, Lung, and Blood Institute/American Heart Association conference on scientific issues related to definition. Arterioscler Thromb Vasc Biol. 2004;24(2):e13-e18. 3. Alberti KG, Zimmet P, Shaw J; IDF Epidemiology Task Force Consensus Group. The metabolic syndrome—a new worldwide definition. Lancet. 2005;366(9491):1059-1062. doi:10.1016/S0140-6736(05)67402-8 4. Stanhope KL. Role of fructose-containing sugars in the epidemics of obesity and metabolic syndrome. Annu Rev Med. 2012;63:329-343. doi:10.1146/annurev-med-042010-113026 5. Stanhope KL. Sugar consumption, metabolic disease and obesity: The state of the controversy. Crit Rev Clin Lab Sci. 2016;53(1):52-67. doi:10.3109/10408363.2015.1084990 6. Malik VS, Popkin BM, Bray GA, Després JP, Willett WC, Hu FB. Sugar-sweetened beverages and risk of metabolic syndrome and type 2 diabetes: a meta-analysis. Diabetes Care. 2010;33(11):2477-2483. doi:10.2337/dc10-1079 7. Schünemann H, Brożek J, Oxman G. Handbook for Grading the Quality of Evidence and the Strength of Recommendations Using the GRADE approach. 2013. GRADE Working Group; 2015. Accessed May 27, 2020. https:// gdt.gradepro.org/app/handbook/handbook.html 8. Higgins JPT, Thomas J, Chandler J, et al. Cochrane Handbook for Systematic Reviews of Interventions. 2nd ed. Cochrane Collaboration; 2019. Accessed May 27, 2020. https://training.cochrane.org/handbook 9. Stroup DF, Berlin JA, Morton SC, et al; Meta-analysis Of Observational Studies in Epidemiology (MOOSE) Group. Meta-analysis of observational studies in epidemiology: a proposal for reporting. JAMA. 2000;283(15): 2008-2012. doi:10.1001/jama.283.15.2008 10. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151(4):264-269, W64. doi:10.7326/0003- 4819-151-4-200908180-00135 JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 12/15 JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome 11. Relation of important food sources of sugars with incident cardiometabolic disease: a series of systematic reviews and meta-analyses to inform guidelines, public health policy, and future research design. ClinicalTrials.gov identifier: NCT02702375. Accessed May 27, 2020. https://clinicaltrials.gov/ct2/show/NCT02702375 12. Brisbois TD, Marsden SL, Anderson GH, Sievenpiper JL. Estimated intakes and sources of total and added sugars in the Canadian diet. Nutrients. 2014;6(5):1899-1912. doi:10.3390/nu6051899 13. Huth PJ, Fulgoni VL, Keast DR, Park K, Auestad N. Major food sources of calories, added sugars, and saturated fat and their contribution to essential nutrient intakes in the U.S. diet: data from the National Health and Nutrition Examination Survey (2003-2006). Nutr J. 2013;12:116. doi:10.1186/1475-2891-12-116 14. Azaïs-Braesco V, Sluik D, Maillot M, Kok F, Moreno LA. A review of total & added sugar intakes and dietary sources in Europe. Nutr J. 2017;16(1):6. doi:10.1186/s12937-016-0225-2 15. Wells G, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Accessed May 27, 2020. http://www.ohri.ca/programs/clinical_ epidemiology/oxford.asp 16. Stevens RD, Dowdy DW, Michaels RK, Mendez-Tellez PA, Pronovost PJ, Needham DM. Neuromuscular dysfunction acquired in critical illness: a systematic review. Intensive Care Med. 2007;33(11):1876-1891. doi:10. 1007/s00134-007-0772-2 17. Harrer M, Cuijpers P, Furukawa TA, Ebert DD. Doing Meta-Analysis in R: A Hands-on Guide. PROTECT Lab; 2019. doi:10.5281/zenodo.2551803 18. Orsini N. DRMETA: Stata Module for Dose-Response Meta-analysis Statistical Software Components. Dept of Economics, Boston College; 2018. 19. Zhang J, Yu KF. What’s the relative risk? a method of correcting the odds ratio in cohort studies of common outcomes. JAMA. 1998;280(19):1690-1691. doi:10.1001/jama.280.19.1690 20. DerSimonian R, Laird N. Meta-analysis in clinical trials revisited. Contemp Clin Trials. 2015;45(Pt A):139-145. doi:10.1016/j.cct.2015.09.002 21. Tufanaru C, Munn Z, Stephenson M, Aromataris E. Fixed or random effects meta-analysis? common methodological issues in systematic reviews of effectiveness. Int J Evid Based Healthc. 2015;13(3):196-207. doi:10. 1097/XEB.0000000000000065 22. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics. 1994;50(4):1088-1101. doi:10.2307/2533446 23. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629-634. doi:10.1136/bmj.315.7109.629 24. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics. 2000;56(2):455-463. doi:10.1111/j.0006-341X.2000.00455.x 25. Crippa A, Discacciati A, Bottai M, Spiegelman D, Orsini N. One-stage dose-response meta-analysis for aggregated data. Stat Methods Med Res. 2019;28(5):1579-1596. doi:10.1177/0962280218773122 26. Orsini N, Li R, Wolk A, Khudyakov P, Spiegelman D. Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software. Am J Epidemiol. 2012;175(1):66-73. doi:10. 1093/aje/kwr265 27. Discacciati A, Crippa A, Orsini N. Goodness of fit tools for dose-response meta-analysis of binary outcomes. Res Synth Methods. 2017;8(2):149-160. doi:10.1002/jrsm.1194 28. Appelhans BMB, Baylin A, Huang MH, et al. Beverage intake and metabolic syndrome risk over 14 years: the Study of Women’s Health Across the Nation. J Acad Nutr Diet. 2017;117(4):554-562. doi:10.1016/j.jand.2016.10.011 29. Babio N, Becerra-Tomás N, Martínez-González MÁ, et al; PREDIMED Investigators. Consumption of yogurt, low-fat milk, and other low-fat dairy products is associated with lower risk of metabolic syndrome incidence in an elderly mediterranean population. J Nutr. 2015;145(10):2308-2316. doi:10.3945/jn.115.214593 30. Cheraghi Z, Mirmiran P, Mansournia MA, Moslehi N, Khalili D, Nedjat S. The association between nutritional exposures and metabolic syndrome in the Tehran Lipid and Glucose Study (TLGS): a cohort study. Public Health. 2016;140:163-171. doi:10.1016/j.puhe.2016.07.003 31. Duffey KJG-L, Gordon-Larsen P, Steffen LM, Jacobs DR Jr, Popkin BM. Drinking caloric beverages increases the risk of adverse cardiometabolic outcomes in the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Am J Clin Nutr. 2010;92(4):954-959. doi:10.3945/ajcn.2010.29478 JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 13/15 JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome 32. Ferreira-Pêgo C, Babio N, Bes-Rastrollo M, et al; PREDIMED Investigators. Frequent consumption of sugar- and artificially sweetened beverages and natural and bottled fruit juices is associated with an increased risk of metabolic syndrome in a Mediterranean population at high cardiovascular disease risk. J Nutr. 2016;146(8): 1528-1536. doi:10.3945/jn.116.230367 33. Hur YIP, Park H, Kang JH, et al. Associations between sugar intake from different food sources and adiposity or cardio-metabolic risk in childhood and adolescence: the Korean Child-Adolescent Cohort Study. Nutrients. 2015;8(1):31. doi:10.3390/nu8010020 34. Kang Y, Kim J. Soft drink consumption is associated with increased incidence of the metabolic syndrome only in women. Br J Nutr. 2017;117(2):315-324. doi:10.1017/S0007114517000046 35. Kim D, Kim J. Dairy consumption is associated with a lower incidence of the metabolic syndrome in middle- aged and older Korean adults: the Korean Genome and Epidemiology Study (KoGES). Br J Nutr. 2017;117(1): 148-160. doi:10.1017/S000711451600444X 36. Lutsey PLS, Steffen LM, Stevens J. Dietary intake and the development of the metabolic syndrome: the Atherosclerosis Risk in Communities study. Circulation. 2008;117(6):754-761. doi:10.1161/CIRCULATIONAHA.107. 37. Mirmiran P, Bahadoran Z, Delshad H, Azizi F. Effects of energy-dense nutrient-poor snacks on the incidence of metabolic syndrome: a prospective approach in Tehran Lipid and Glucose Study. Nutrition. 2014;30(5):538-543. doi:10.1016/j.nut.2013.09.014 38. Mirmiran P, Yuzbashian E, Asghari G, Hosseinpour-Niazi S, Azizi F. Consumption of sugar sweetened beverage is associated with incidence of metabolic syndrome in Tehranian children and adolescents. Nutr Metab (Lond). 2015;12:25. doi:10.1186/s12986-015-0021-6 39. Sayón-Orea C, Bes-Rastrollo M, Martí A, Pimenta AM, Martín-Calvo N, Martínez-González MA. Association between yogurt consumption and the risk of metabolic syndrome over 6 years in the SUN study. BMC Public Health. 2015;15:170. doi:10.1186/s12889-015-1518-7 40. Lim M, Kim J. Association between fruit and vegetable consumption and risk of metabolic syndrome determined using the Korean Genome and Epidemiology Study (KoGES). Eur J Nutr. 2020;59(4):1667‐1678. doi: 10.1007/s00394-019-02021-5 41. Coyle DH, Ndanuko R, Singh S, Huang P, Wu JH. Variations in sugar content of flavored milks and yogurts: a cross-sectional study across 3 countries. Curr Dev Nutr. 2019;3(6):nzz060. doi:10.1093/cdn/nzz060 42. Chollet M, Gille D, Schmid A, Walther B, Piccinali P. Acceptance of sugar reduction in flavored yogurt. J Dairy Sci. 2013;96(9):5501-5511. doi:10.3168/jds.2013-6610 43. Vickers Z, Holton E, Wang J. Effect of yogurt sweetness on sensory specific satiety. J Sens Stud. 1998;13(4): 377-388. doi:10.1111/j.1745-459X.1998.tb00096.x 44. Vickers Z, Holton E, Wang J. Effect of ideal–relative sweetness on yogurt consumption. Food Qual Prefer. 2001;12(8):521-526. doi:10.1016/S0950-3293(01)00047-7 45. Guyatt G, Oxman AD, Akl EA, et al. GRADE guidelines 1: introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol. 2011;64(4):383-394. doi:10.1016/j.jclinepi.2010.04.026 46. Narain A, Kwok CS, Mamas MA. Soft drink intake and the risk of metabolic syndrome: a systematic review and meta-analysis. Int J Clin Pract. 2017;71(2). doi:10.1111/ijcp.12927 47. Khan TA, Sievenpiper JL. Controversies about sugars: results from systematic reviews and meta-analyses on obesity, cardiometabolic disease and diabetes. Eur J Nutr. 2016;55(suppl 2):25-43. doi:10.1007/s00394-016- 1345-3 48. Almiron-Roig E, Palla L, Guest K, et al. Factors that determine energy compensation: a systematic review of preload studies. Nutr Rev. 2013;71(7):458-473. doi:10.1111/nure.12048 49. Chen GC, Szeto IM, Chen LH, et al. Dairy products consumption and metabolic syndrome in adults: systematic review and meta-analysis of observational studies. Sci Rep. 2015;5:14606. doi:10.1038/srep14606 50. Lorenzen JK, Astrup A. Dairy calcium intake modifies responsiveness of fat metabolism and blood lipids to a high-fat diet. Br J Nutr. 2011;105(12):1823-1831. doi:10.1017/S0007114510005581 51. Santaren ID, Watkins SM, Liese AD, et al. Individual serum saturated fatty acids and markers of chronic subclinical inflammation: the Insulin Resistance Atherosclerosis Study. J Lipid Res. 2017;58(11):2171-2179. doi:10. 1194/jlr.P076836 52. Parvez S, Malik KA, Ah Kang S, Kim HY. Probiotics and their fermented food products are beneficial for health. J Appl Microbiol. 2006;100(6):1171-1185. doi:10.1111/j.1365-2672.2006.02963.x JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 14/15 JAMA Network Open | Nutrition, Obesity, and Exercise Major Food Sources of Fructose-Containing Sugars and Incident Metabolic Syndrome 53. Schwingshackl L, Hoffmann G, Kalle-Uhlmann T, Arregui M, Buijsse B, Boeing H. Fruit and vegetable consumption and changes in anthropometric variables in adult populations: a systematic review and meta-analysis of prospective cohort studies. PLoS One. 2015;10(10):e0140846. doi:10.1371/journal.pone.0140846 54. John JH, Ziebland S, Yudkin P, Roe LS, Neil HA; Oxford Fruit and Vegetable Study Group. Effects of fruit and vegetable consumption on plasma antioxidant concentrations and blood pressure: a randomised controlled trial. Lancet. 2002;359(9322):1969-1974. doi:10.1016/S0140-6736(02)98858-6 55. Pan Y, Pratt CA. Metabolic syndrome and its association with diet and physical activity in US adolescents. JAm Diet Assoc. 2008;108(2):276-286. doi:10.1016/j.jada.2007.10.049 56. Fletcher EA, McNaughton SA, Lacy KE, Dunstan DW, Carson V, Salmon J. Mediating effects of dietary intake on associations of TV viewing, body mass index and metabolic syndrome in adolescents. Obes Sci Pract. 2016;2(3): 232-240. doi:10.1002/osp4.60 57. Tian Y, Su L, Wang J, Duan X, Jiang X. Fruit and vegetable consumption and risk of the metabolic syndrome: a meta-analysis. Public Health Nutr. 2018;21(4):756-765. doi:10.1017/S136898001700310X 58. Clemens R, Drewnowski A, Ferruzzi MG, Toner CD, Welland D. Squeezing fact from fiction about 100% fruit juice. Adv Nutr. 2015;6(2):236S-243S. doi:10.3945/an.114.007328 59. Khan TA, Chiavaroli L, Zurbau A, Sievenpiper JL. A lack of consideration of a dose-response relationship can lead to erroneous conclusions regarding 100% fruit juice and the risk of cardiometabolic disease. Eur J Clin Nutr. 2019;73(12):1556-1560. doi:10.1038/s41430-019-0514-x 60. Ramli NZ, Chin KY, Zarkasi KA, Ahmad F. A review on the protective effects of honey against metabolic syndrome. Nutrients. 2018;10(8):E1009. doi:10.3390/nu10081009 61. Guyatt GH, Oxman AD, Kunz R, et al; GRADE Working Group. GRADE guidelines, 7: rating the quality of evidence—inconsistency. J Clin Epidemiol. 2011;64(12):1294-1302. doi:10.1016/j.jclinepi.2011.03.017 62. Marriott BP, Cole N, Lee E. National estimates of dietary fructose intake increased from 1977 to 2004 in the United States. J Nutr. 2009;139(6):1228S-1235S. doi:10.3945/jn.108.098277 63. Montagnese C, Santarpia L, Buonifacio M, et al. European food-based dietary guidelines: a comparison and update. Nutrition. 2015;31(7-8):908-915. doi:10.1016/j.nut.2015.01.002 64. Health Canada. Eating well with Canada’s food guide. Accessed March 2, 2020. https://www.hc-sc.gc.ca/fn-an/ alt_formats/hpfb-dgpsa/pdf/food-guide-aliment/print_eatwell_bienmang-eng.pdf 65. Health Canada. Canada's dietary guidelines. In: Canada's Food Guide. Health Canada; 2019. Accessed May 27, 2020. https://food-guide.canada.ca/en/guidelines/ 66. Scientific Advisory Committee on Nutrition. Carbohydrates and Health. Scientific Advisory Committee on Nutrition; 2015. Accessed May 27, 2020. https://assets.publishing.service.gov.uk/government/uploads/system/ uploads/attachment_data/file/445503/SACN_Carbohydrates_and_Health.pdf SUPPLEMENT. eAppendix 1. Details on GRADE eAppendix 2. Conversion of OR to RR eAppendix 3. Method for Dose-Response Analysis eAppendix 4. Definition of MetS eTable 1. Search Strategy eTable 2. Analysis of Confounding Variables Among 13 Studies of Food Sources of Fructose-Containing Sugars and Incident MetS eTable 3. Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Cohort Studies eTable 4. GRADE Assessment eTable 5. GRADE Assessment (Continued) eFigure 1. Relationship Between SSB Intake and Incident MetS eFigure 2. Relationship Between Mixed Fruit Juice Intake and Incident MetS eFigure 3. Relationship Between 100% Fruit Juice Intake and Incident MetS eFigure 4. Relationship Between Fruit Intake and Incident MetS eFigure 5. Relationship Between Yogurt Intake and Incident MetS eFigure 6. Relationship Between Honey Intake and Incident MetS eFigure 7. Relationship Between Ice-Cream Intake and Incident MetS eFigure 8. Relationship Between Confectionary Intake (Including Cakes, Biscuits, Chocolate and Candies) and Incident MetS eReferences JAMA Network Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 (Reprinted) July 9, 2020 15/15 Supplementary Online Content Semnani-Azad Z, Khan TA, Blanco Mejia S, et al. Association of major food sources of fructose-containing sugars with incident metabolic syndrome: a systematic review and meta- analysis. JAMA Netw Open. 2020;3(7):e209993. doi:10.1001/jamanetworkopen.2020.9993 eAppendix 1. Details on GRADE eAppendix 2. Conversion of OR to RR eAppendix 3. Method for Dose-Response Analysis eAppendix 4. Definition of MetS eTable 1. Search Strategy eTable 2. Analysis of Confounding Variables Among 13 Studies of Food Sources of Fructose- Containing Sugars and Incident MetS eTable 3. Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Cohort Studies eTable 4. GRADE Assessment eTable 5. GRADE Assessment (Continued) eFigure 1. Relationship Between SSB Intake and Incident MetS eFigure 2. Relationship Between Mixed Fruit Juice Intake and Incident MetS eFigure 3. Relationship Between 100% Fruit Juice Intake and Incident MetS eFigure 4. Relationship Between Fruit Intake and Incident MetS eFigure 5. Relationship Between Yogurt Intake and Incident MetS eFigure 6. Relationship Between Honey Intake and Incident MetS eFigure 7. Relationship Between Ice-Cream Intake and Incident MetS eFigure 8. Relationship Between Confectionary Intake (Including Cakes, Biscuits, Chocolate and Candies) and Incident MetS eReferences This supplementary material has been provided by the authors to give readers additional information about their work. © 2020 Semnani-Azad et al. JAMA Network Open. eAppendix 1. Details on GRADE GRADE (Grading of Recommendations Assessment, Development and Evaluation) was used to evaluate the quality of each study. Evidence extracted from observational studies were defa - certainty and were downgraded or upgraded based on pre-specified criteria. Criteria to downgrade included risk of bias (weight of studies show risk of bias as assessed by NOS<6), inconsistency (substantial unexplained inter-study heterogeneity I >50%, P <0.10), indirectness (presence of factors that limit the generalizability of the results), imprecision in the pooled risk estimate (the 95% CI for risk estimates that cross a minimally important difference of 5% for benefit or harm [RR 0.95 1.05]), and publication bias (evidence of small-study effects). Upgraded criteria included a large magnitude of effect (RR>2 or RR<0.5 in the absence of plausible confounders), dose response gradient, and attenuation of the pooled effect estimate by plausible confounders. © 2020 Semnani-Azad et al. JAMA Network Open. eAppendix 2. Conversion of OR to RR For studies with reported hazard ratios, low incidence of MetS (<10%) or odds ratios (OR) between 0.5 and 2.5, values were treated as RRs. OR were converted to RR if OR was less than 0.5 or greater than 2.5 with an incident of MetS greater than 10%. As outlined by Zhang et al. , the following formulae and logic were applied: Thus, RR = , therefore: P = incidence of the outcome of interest in the non-exposed group; P = incidence of the outcome of interest in the 0 1 exposed group. © 2020 Semnani-Azad et al. JAMA Network Open. eAppendix 3. Method for Dose-Response Analysis We modelled dose-response model using RR and 95% CIs from dose categories to understand the shape of the association between the dose of the food source of fructose-containing sugar and the risk of MetS. Data on the dose, distribution of cases and person-years, RRs and 95% CIs were extracted from each study. We defined the assigned dose as the mean consumption in each reported category or quantile. If the assigned were not reported, we approximated the mean dose for each category by using the midpoint of its lower and upper bounds. If the lowest category of a study was open ended, we defined the lowest dose as zero. For open-ended upper categories, we took half of the adjacent category range to estimate the assigned dose. When cohort size or person-year per category were not available, categories were regarded equal in size and follow-up and the case number per category was obtained by the method of Bekkering . We excluded studies from the dose response meta-analysis that did not report any dose category cut points for the particular food source and studies that provided only RR estimates based on 1-unit increment in dose based on a linear model because these studies were unable to contribute to the assessment of departure from linearity. We fitted a dose-response relationship using restricted cubic splines with 3 knots at 15th, th 50th and 85 percentiles of distribution taking into account the correlation within each category of published RRs and combining the study- -stage linear mixed-effects meta-analysis . This method estimates the study specific slope lines and combines them to obtain an overall average slope based upon the work of Greenland and Orsini . If restricted cubic splines could not be calculated due to limited number of observations, we fitted a second order fractional polynomial curve to the data and tested for goodness-of-fit of the model using Akaike information criterion (AIC), deviance test (D) and the coefficient of determination (R2) to select the best-fitting model . We reported non-linear associations for a study if Wald test for departure from linearity was significant at p<0.10. RRs below 1 were considered as protective and above 1 as adverse association. © 2020 Semnani-Azad et al. JAMA Network Open. eAppendix 4. Definition of MetS The harmonized criteria classification for MetS takes into account definitions set by the International Diabetes Federation (IDF) and the American Heart Association/ National Heart, Lung, and Blood Institute ATPIII. Both IDF and ATP low HDL The ATP III identifies MetS as the presence of any 3 of 5 risk factors. The IDF defines MetS as the presence of abdominal obesity measured through waist circumference, with the addition of any 2 of 4 risk factors. The harmonized criteria defines MetS as the presence of any 3 of 5 risk factors, with specific waist circumference cut-points depending on ethnicity. © 2020 Semnani-Azad et al. JAMA Network Open. eTable 1. Search Strategy. MEDLINE EMBASE Cochrane 1 sugar*.mp. 1 sugar*.mp. 1 sugar*.mp. 2 exp fructose/ 2 exp sugar/ 2 exp fructose/ 3 fructose.mp. 3 exp fructose/ 3 fructose.mp. 4 HFCS.mp. 4 fructose.mp. 4 HFCS.mp. 5 exp High Fructose Corn Syrup/ 5 HFCS.mp. 5 exp Nutritive Sweeteners/ 6 sucrose.mp. 6 exp high fructose corn syrup/ 6 sucrose.mp. 7 exp Dietary Sucrose/ 7 sucrose.mp. 7 exp dietary sucrose/ 8 sugar sweetened beverage*.mp. 8 exp dietary sucrose/ 8 sugar sweetened beverage*.mp. 9 SSB.mp. 9 sugar sweetened beverage*.mp. 9 ssb.mp. 10 soda.mp. 10 SSB.mp. 10 soda.mp. 11 soft drink*.mp. 11 soda.mp. 11 soft drink*.mp. 12 exp Carbonated Beverages/ 12 soft drink*.mp. 12 exp carbonated beverages/ 13 carbonated beverages.mp. 13 exp soft drink/ 13 non alcoholic beverage*.mp. 14 non alcoholic beverage*.mp. 14 exp Carbonated Beverages/ 14 nonalcoholic beverage*.mp. 15 nonalcoholic beverage*.mp. 15 carbonated beverages.mp. 15 exp energy drinks/ 16 exp Energy Drinks/ 16 non alcoholic beverage*.mp. 16 energy drink*.mp. 17 energy drink*.mp. 17 nonalcoholic beverage*.mp. 17 smoothie*.mp. 18 smoothie*.mp. 18 exp energy drink/ 18 ((fruit or vegetable) and juice*).mp. 19 exp "Fruit and Vegetable Juices"/ 19 energy drink*.mp. 19 fruit.mp. 20 fruit.mp. 20 smoothie*.mp. 20 exp fruit/ 21 exp Fruit/ 21 exp "fruit and vegetable juice"/ 21 exp honey/ 22 exp Honey/ 22 fruit.mp. 22 y*g*rt.mp. 23 y*g*rt.mp. 23 exp fruit/ 23 exp yogurt/ 24 exp Yogurt/ 24 exp honey/ 24 ice cream*.mp. 25 ice cream*.mp. 25 y*g*rt.mp. 25 icecream*.mp. 26 icecream*.mp. 26 exp yoghurt/ 26 exp ice cream/ 27 exp Ice Cream/ 27 exp ice cream/ 27 cereal*.mp. 28 cereal*.mp. 28 ice cream*.mp. 28 dessert*.mp. 29 exp edible grain/ 29 icecream*.mp. 29 sweets.mp. 30 dessert*.mp. 30 cereal*.mp. 30 confection*.mp. 31 sweets.mp. 31 dessert*.mp. 31 pastries.mp. 32 confection*.mp. 32 sweets.mp. 32 biscuit*.mp. 33 pastries.mp. 33 confection*.mp. 33 cookie*.mp. 34 biscuit*.mp. 34 exp bakery product/ 34 cake*.mp. 35 cookie*.mp. 35 pastries.mp. 35 candy.mp. © 2020 Semnani-Azad et al. JAMA Network Open. 36 cake*.mp. 36 biscuit*.mp. 36 candies.mp. 37 candy.mp. 37 cookie*.mp. 37 exp candy/ 38 candies.mp. 38 cake*.mp. 38 (chocolate adj2 milk).mp. 39 exp Candy/ 39 candy.mp. 39 exp chocolate/ 40 (chocolate adj2 milk).mp. 40 candies.mp. 40 Chocolate.mp 41 exp chocolate/ 41 (chocolate adj2 milk).mp. 41 exp cacao/ 42 Chocolate.mp 42 exp chocolate/ 42 cacao.mp. 43 exp cacao/ 43 Chocolate.mp 43 or/1-42 44 cacao.mp. 44 exp cacao/ 44 cohort.mp. 45 or/1-44 45 cacao.mp. 45 exp Prospective Studies/ 46 cohort.mp. 46 or/1-45 46 (prospective adj2 (cohort or study)).mp. 47 exp prospective study/ 47 cohort.mp. 47 exp follow-up studies/ 48 (prospective adj2 (cohort or 48 exp prospective study/ 48 exp multivariate analysis/ study)).mp. 49 exp Follow-Up Studies/ 49 (prospective adj2 (cohort or 49 exp proportional hazards models/ study)).mp. 50 exp Multivariate Analysis/ 50 exp multivariate analysis/ 50 follow up study.mp. 51 exp Proportional Hazards Models/ 51 exp proportional hazards model/ 51 (longitudinal adj2 study).mp. 52 follow up study.mp. 52 follow up study.mp. 52 or/44-51 53 (longitudinal adj2 study).mp. 53 (longitudinal adj2 study).mp. 53 metabolic syndrome.mp. 54 or/46-53 54 or/47-53 54 syndrome x.mp. 55 metabolic syndrome.mp. 55 metabolic syndrome.mp. 55 cardio-metabolic syndrome.mp. 56 syndrome x.mp. 56 syndrome x.mp. 56 MetS.mp. 57 cardio-metabolic syndrome.mp. 57 cardio-metabolic syndrome.mp. 57 or/53-56 58 MetS.mp. 58 MetS.mp. 58 43 and 52 and 57 59 or/55-58 59 or/55-58 60 45 and 54 and 59 60 46 and 55 and 59 Database Total MEDLINE: March week 3, 2020 402 EMBASE: March week 3, 2020 584 Cochrane: March week 3, 2020 76 Manual search 8 Total 1071 Duplicates 396 Final Total 675 © 2020 Semnani-Azad et al. JAMA Network Open. eTable 2. Analysis of confounding variables among 13 studies of food sources of fructose-containing sugars and incident MetS. Sayon- Hur et Kang Kim and Lim and Lutsey Mirmiran Mirmiran Appelhans Babio et al., Cheraghi et Duffey et Ferreira-Pego Orea al., 2016 and Kim, Kim, Kim , et al., et al., et al., Study et al., 2017 2015 - al., 2016 al., 2010 et al., 2016 et al., 2017 2017 2019 - 2008 - 2014 2015 9 10 11 12 13 SWAN PREDIMED TLGS CARDIA PREDIMED 2015 14 15 16 17 18 19 20 KoCAS KoGES KoGES KoGES ARIC TLGS TLGS SUN Number of variables in 14 10 23 0 10 26 4 14 12 14 15 12 15 fully adjusted model PRESPECIFIED VARIABLES Age Sex Markers of overweight/ obesity (body mass index, weight, waist circumference, waist to hip ratio) Smoking Family history of MetS Energy or caloric intake Diabetes Physical activity Alcohol OTHER COVARIATE VARIABLES Ethnicity Education Hypertension/SBP HDL cholesterol Vegetables Fruit Whole grain Fibre Saturated fat Unsaturated fat Red meat Coffee Meat and fish Anti-hypertensive medication HRT Menopause © 2020 Semnani-Azad et al. JAMA Network Open. Insulin OTHER Hormone therapy use Depressive symptoms Income Study site Hypoglycemia and hypolipidemic drug Legume Cereal Baked goods Nuts Olive oil High fasting plasma glucose Hypertriglycermidemia Trans fat Glycemic index Magnesium Dairy products Percentage of fat Presence of disease Income Residential location Calcium Weight change Phytochemical index Dietary total antioxidant capacity Tea French fries Fast food Mediterranean diet Sedentary behaviour Hours sitting Snacking between meals Special diet Refined grains Means variable adjusted for in the most adjusted model. © 2020 Semnani-Azad et al. JAMA Network Open. eTable 3. Newcastle-Ottawa Scale (NOS) for assessing the quality of cohort studies. Study, year Selection* Outcome Comparability Total§ Appelhans et al. 2017 4 2 2 8 Babio et al. 2015 3 2 2 7 Cheraghi et al. 2016 4 2 0 6 Duffey et al. 2010 4 2 2 8 Ferreira-Pego et al. 2016 3 3 2 8 Hur et al. 2016 4 2 1 6 Kang and Kim 2017 4 2 2 8 Kim and Kim 2017 4 2 2 8 Lim and Kim 2019 4 2 2 8 Lutsey et al. 2008 3 3 2 8 Mirmiran et al. 2014 3 2 2 7 Mirmiran et al. 2015 3 2 2 7 Sayon-Orea et al. 2015 3 3 2 8 *Maximum 4 points awarded for cohort representativeness, selection of non-exposed cohort, exposure assessment, and demonstration outcome not present at baseline -up length, adequacy of follow-up, and outcome assessment Maximum 2 points awarded for controlling for the pre-specified primary confounding variable (age) and additional confounding variables. § A maximum of 9 points could be awarded © 2020 Semnani-Azad et al. JAMA Network Open. eTable 4. GRADE Assessment. Study event Quality assessment Estimate Quality Importance rates (%) Other No. of Risk of Publication Relative Risk Design Inconsistency Indirectness Imprecision consideration studies bias bias (95% CI) SSB intake on incident MetS (follow-up mean 7.5 years) 1.21 [1.06- 1.37] 9,13,15,18- % 7 Observational Not Dose-response 7,406/20,480 MODERATE* Not serious* Not serious Not serious Undetected Linear DRM 20 % Studies serious association (36%) Upgrade due to dose- RR 355-ml/day response association. 1.14 [1.05, 1.23] Mixed fruit juice intake on incident MetS (follow-up mean 3.4 years) 1.13 [0.91- 1.41] ±, %% Observational Not Not Dose-response 1,322/3,062 Non-linear MODERATE 13,14,20 ± 3 Not serious Not serious Undetected %% Studies serious serious association (43%) DRM Upgrade for dose- RR response association. 125-ml/day 0.58 [0.42, 0.79] 100% Fruit juice intake on incident MetS (follow-up mean 5.1 years) 1.10 [0.84- 1.44] MODERATE** Observational Not Not Dose-response 1,389/5,464 Non-linear 12,13 ± %%% 2 Not serious** Not serious Undetected %%% Studies serious serious association (25%) DRM Upgrade for dose- RR 125-ml/day response association. 0.77 [0.61, 0.97] Fruit on incident MetS (follow-up mean 4.7 years) 0.91 [0.89, 0.93] MODERATE Dose-response ±,%%%% Observational Not Not 3,002/10,074 Non-linear 11,14,17 ± 4 Not serious Not serious Undetected association Studies serious serious (30%) DRM Upgrade for dose- %%% % RR response association. 80gl/day 0.82 [0.78, 0.86] Yogurt intake on incident MetS (follow-up mean 3.4 years) © 2020 Semnani-Azad et al. JAMA Network Open. 0.83 [0.77, 0.90] MODERATE*** Dose-response ±, %%%%% Observational Not Not Not 3,877/19,057 Non-linear 10,11,16,21 ± 5 Not serious Undetected association Studies serious serious*** serious (20%) DRM Due to an upgrade for %%%% % RR dose-response -85-g/day 0.66 [0.58, association. 0.76] © 2020 Semnani-Azad et al. JAMA Network Open. eTable 4. GRADE Assessment (Continued). Study Quality Quality assessment event Estimate Importance rates (%) Other No. of Risk of Publication Relative Risk Design Inconsistency Indirectness Imprecision consideration studies bias bias (95% CI) Honey intake on incident MetS (follow-up 2.05 years) VERY LOW**** Observational Not 590/3,616 1.00 [0.5, 2.00] Due to 11 ± 1 Undetected**** Serious Serious Undetected None Studies serious (16%) downgrade for serious indirectness and serious imprecision. Ice-cream intake on incident MetS (follow-up 2.05 years) VERY LOW**** Observational Not 590/3,616 0.94 [0.84, 1.06] Due to 11 ± 1 Undetected**** Serious Serious Undetected None Studies serious (16%) downgrade for serious indirectness and serious imprecision. Confectionary intake on incident MetS (follow-up 3 years) VERY LOW Due to Observational Not 250/1,476 19 ± 2 Not serious Serious Serious Undetected None 1.21 [0.92, 1.60] downgrade for Studies serious (17%) serious indirectness and serious imprecision. * Although there was evidence of substantial inter-study heterogeneity (I = 68%), the estimates were all in the same direction and there was considerable overlap. Therefore, we did not consider this as serious inconsistency. ** There was substantial heterogeneity (I = 73%, P = 0.05) in the pairwise analysis. This was explained by the non-linear dose-response model. Therefore, we did not downgrade for serious inconsistency. © 2020 Semnani-Azad et al. JAMA Network Open. *** Although there was evidence of substantial inter-study heterogeneity (I = 65%), the estimates were all in the same direction and there was considerable overlap. Therefore, we did not consider this as serious inconsistency. **** Not able to assess inconsistency due to only one study included. Downgrade for serious indirectness due to only one cohort available, therefore affecting the generalizability to the general population. No downgrade for serious imprecision as lower bound of 95% CI does not cross the clinically unimportant effects (RR 0.95 1.05). The potential imprecision from pairwise meta-analysis was explained by the non-linear dose-response model. Therefore, we did not downgrade for serious imprecision for mixed fruit juice. The potential imprecision from pairwise meta-analysis was explained by the non-linear dose-response model. Therefore, we did not downgrade for serious imprecision for 100% fruit juice. No downgrade for serious imprecision as lower bound of 95% CI does not cross the clinically unimportant effects (RR 0.95 1.05). No downgrade for serious imprecision as the upper bound of 95% CI (RR 0.90) does not include the threshold for clinically unimportant effects (RR 0.95 1.05). Downgrade for serious imprecision as the lower bound of 95% CI (RR 0.50) includes clinically important benefit (RR <0.90) while the upper bound of the 95% CI (RR 2.00) includes the clinically important harm (RR >1.05). Downgrade for serious imprecision as the lower bound of 95% CI (RR 0.84) includes clinically important benefit (RR <0.90) while the upper bound of the 95% CI (RR 1.06) includes the unimportant effects (RR 0.95 1.05). Downgrade for serious imprecision as the lower bound of 95% CI (RR, 0.92) includes clinically unimportant effects (RR 0.95 1.05) while the upper bound of the 95% CI (RR, 1.60) includes the clinically important harm (RR >1.05). No downgrade for publication bias, as publication bias could not be assessed due to lack of power for assessing funnel plot asymmetry and small study effects (<10 cohorts included in our meta-analysis). Linear dose response relationship with suggestion of positive association with risk SSB (P=0.001). %% Non-linear dose response relationship with suggestion of inverse association with risk for mixed fruit juice (P<0.001). %%% Non-linear dose response relationship with suggestion of inverse association with risk for 100% fruit juice (P<0.01). %%%% Linear and non-linear dose response relationship with suggestion of inverse association with risk for fruit (both P<0.001). %%%%% Linear and non-linear dose response relationship with suggestion of inverse association with risk for yogurt (both P<0.001). DRM: Dose-response meta-analysis. © 2020 Semnani-Azad et al. JAMA Network Open. eTable 5. Sensitivity analysis for all food sources with more than 2 studies. Heterogeneity Removed Study RR [95% CI] P-value I P SSB All Studies Included 1.21 [1.06, 1.37] 0.005 68% 0.005 Appelhans, J Acad Nutr Diet, 2017 SWAN 1.37 [1.09, 1.71] 0.006 64% 0.016 1.18 [1.04, 1.35] 0.013 69% 0.006 Ferreira-Pêgo, J Nutri, 2016 PREDIMED 1.13 [1.02, 1.26] 0.025 55% 0.048 Kang, Br J Nutr, 2017 KoGES (Female) 1.23 [1.07, 1.42] 0.005 73% 0.002 Kang, Br J Nutr, 2017 KoGES (Male) 1.36 [1.07, 1.74] 0.012 73% 0.002 Lutsey, Circulation, 2008 ARIC 1.16 [1.03, 1.30] 0.011 62% 0.022 Mirmiran, Nutr Metab, 2015 TLGS 1.18 [1.04, 1.35] 0.013 69% 0.006 Mirmiran, Nutrition, 2014 TLGS Mixed Fruit Juice 1.13 [0.91, 1.43] 0.270 0% 0.867 All Studies Included Ferreira-Pêgo, J Nutri, 2016 PREDIMED 1.04 [0.45, 2.40] 0.927 0% 0.622 Hur, Nutrients, 2015 - KoCAS 1.14 [0.91, 1.43] 0.243 0% 0.934 Mirmiran, Nutr Metab, 2015 TLGS 1.13 [0.90, 1.42] 0.292 0% 0.599 FRUIT 0.91 [0.89, 0.93] < 0.001 0% 0.778 All Studies Included 0.91 [0.89, 0.93] <0.001 0% 0.919 Cheraghi, Public Health, 2016 TLGS 0.91 [0.89, 0.93] <0.001 0% 0.628 Hur, Nutrients, 2015 - KoCAS 0.91 [0.88, 0.95] <0.001 0% 0.583 Lim, Eur J Nutr, 2019 KoGES (Female) 0.91 [0.88, 0.94] <0.001 0% 0.580 Lim, Eur J Nutr, 2019 KoGES (Male) YOGURT All Studies Included 0.83 [0.77, 0.90] < 0.001 65% 0.021 0.85 [0.78, 0.92] <0.001 72% 0.014 Babio, J Nutr, 2015 PREDIMED Cheraghi, Public Health, 2016 TLGS 0.74 [0.66, 0.82] <0.001 0% 0.654 0.85 [0.79, 0.93] <0.001 66% 0.032 Kim, Br J Nutr, 2017 KoGES (Female) Kim, Br J Nutr, 2017 KoGES (Male) 0.85 [0.79, 0.93] <0.001 61% 0.051 0.83 [0.77, 0.90] <0.001 74% 0.009 Sayon-Orea, BMC Public Health, 2015 SUN © 2020 Semnani-Azad et al. JAMA Network Open. eFigure 1. Relationship between SSB intake and incident MetS. The black diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q 2 2 2 statistic (Chi ) at a significance level of P < 0.10, and quantified by the I statistic. An I indicate substantial heterogeneity. All results are presented as Relative Risks (RR) with 95% Confidence Intervals where estimable. © 2020 Semnani-Azad et al. JAMA Network Open. eFigure 2. Relationship between mixed fruit juice intake and incident MetS. The black diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q 2 2 2 statistic (Chi ) at a significance level of P < 0.10, and quantified by the I statistic. An I to indicate substantial heterogeneity. All results are presented as Relative Risks (RR) with 95% Confidence Intervals where estimable. © 2020 Semnani-Azad et al. JAMA Network Open. eFigure 3. Relationship between 100% fruit juice intake and incident MetS. The black diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q 2 2 2 statistic (Chi ) at a significance level of P < 0.10, and quantified by the I statistic. An I red to indicate substantial heterogeneity. All results are presented as Relative Risks (RR) with 95% Confidence Intervals where estimable. © 2020 Semnani-Azad et al. JAMA Network Open. eFigure 4. Relationship between fruit intake and incident MetS. The black diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q 2 2 2 statistic (Chi ) at a significance level of P < 0.10, and quantified by the I statistic. An I to indicate substantial heterogeneity. All results are presented as Relative Risks (RR) with 95% Confidence Intervals where estimable. © 2020 Semnani-Azad et al. JAMA Network Open. eFigure 5. Relationship between yogurt intake and incident MetS. The black diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q 2 2 2 statistic (Chi ) at a significance level of P < 0.10, and quantified by the I statistic. An I indicate substantial heterogeneity. All results are presented as Relative Risks (RR) with 95% Confidence Intervals where estimable. © 2020 Semnani-Azad et al. JAMA Network Open. eFigure 6. Relationship between honey intake and incident MetS. The black diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q 2 2 2 statistic (Chi ) at a significance level of P < 0.10, and quantified by the I statistic. An I to indicate substantial heterogeneity. All results are presented as Relative Risks (RR) with 95% Confidence Intervals where estimable. © 2020 Semnani-Azad et al. JAMA Network Open. eFigure 7. Relationship between ice-cream intake and incident MetS. The black diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q 2 2 2 statistic (Chi ) at a significance level of P < 0.10, and quantified by the I statistic. An I to indicate substantial heterogeneity. All results are presented as Relative Risks (RR) with 95% Confidence Intervals where estimable. © 2020 Semnani-Azad et al. JAMA Network Open. eFigure 8. Relationship between confectionary intake (including cakes, biscuits, chocolate and candies) and incident MetS. The black diamond represents the pooled risk estimate. Inter-study heterogeneity was tested using the Cochran Q 2 2 2 statistic (Chi ) at a significance level of P < 0.10, and quantified by the I statistic. An I to indicate substantial heterogeneity. All results are presented as Relative Risks (RR) with 95% Confidence Intervals where estimable. © 2020 Semnani-Azad et al. JAMA Network Open. eReferences 1. Zhang JY, K. F. What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. Jama. 1998;280(19):1690-1691. 2. Bekkering GE, Harris RJ, Thomas S, et al. How much of the data published in observational studies of the association between diet and prostate or bladder cancer is usable for meta- analysis? Am J Epidemiol. 2008;167(9):1017-1026. 3. Crippa A, Discacciati A, Bottai M, Spiegelman D, Orsini N. One-stage dose-response meta- analysis for aggregated data. Stat Methods Med Res. 2019;28(5):1579-1596. 4. Greenland S, Longnecker MP. Methods for trend estimation from summarized dose-response data, with applications to meta-analysis. Am J Epidemiol. 1992;135(11):1301-1309. 5. Orsini N, Li R, Wolk A, Khudyakov P, Spiegelman D. Meta-analysis for linear and nonlinear dose- response relations: examples, an evaluation of approximations, and software. Am J Epidemiol. 2012;175(1):66-73. 6. Discacciati A, Crippa A, Orsini N. Goodness of fit tools for dose nary outcomes. Research synthesis methods. 2017;8(2):149-160. 7. Assmann G, Guerra R, Fox G, et al. Harmonizing the definition of the metabolic syndrome: comparison of the criteria of the Adult Treatment Panel III and the International Diabetes Federation in United States American and European populations. Am J Cardiol. 2007;99(4):541- 8. Alberti KG, Eckel RH, Grundy SM, et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009;120(16):1640-1645. 9. Appelhans BMB, A. Huang, M. H. Li, H. Janssen, I. Kazlauskaite, R. Avery, E. F. Kravitz, H. M. Beverage Intake and Metabolic Syndrome Risk Over 14 Years: The Study of Women's Health Across the Nation. Journal of the Academy of Nutrition & Dietetics. 2017;117(4):554-562. 10. Babio NB-T, N.Martinez-Gonzalez, M. A.Corella, D.Estruch, R.Ros, E.Sayon-Orea, C.Fito, M.Serra- Majem, L.Aros, F.Lamuela-Raventos, R. M.Lapetra, J.Gomez-Gracia, E.Fiol, M.Diaz-Lopez, A.Sorli, J. V.Martinez, J. A.Salas-Salvado, J.Predimed Investigators. Consumption of Yogurt, Low-Fat Milk, and Other Low-Fat Dairy Products Is Associated with Lower Risk of Metabolic Syndrome Incidence in an Elderly Mediterranean Population. Journal of Nutrition. 2015;145(10):2308-2316. 11. Cheraghi ZM, P. Mansournia, M. A. Moslehi, N. Khalili, D. Nedjat, S. The association between nutritional exposures and metabolic syndrome in the Tehran Lipid and Glucose Study (TLGS): a cohort study. Public Health. 2016;140:163-171. 12. Duffey KJG-L, P. Steffen, L. M. Jacobs Jr, D. R. Popkin, B. M. Drinking caloric beverages increases the risk of adverse cardiometabolic outcomes in the coronary artery risk development in young adults (CARDIA) study. American Journal of Clinical Nutrition. 2010;92(4):954-959. 13. Ferreira-Pego CB, N. Bes-Rastrollo, M. Corella, D. Estruch, R. Ros, E. Fito, M. Lluis, S. M. Aros, F. Fiol, M. Santos-Lozano, J. M. Munoz-Bravo, C. Pinto, X. Ruiz-Canela, M. Salas-Salvado, J. Frequent consumption of sugar- and artificially sweetened beverages and natural and bottled fruit juices is associated with an increased risk of metabolic syndrome in a mediterranean population at high cardiovascular disease risk. Journal of Nutrition. 2016;146(8):1528-1536. 14. Hur YIP, H.Kang, J. H.Lee, H. A.Song, H. J.Lee, H. J.Kim, O. H. Associations between Sugar Intake from Different Food Sources and Adiposity or Cardio-Metabolic Risk in Childhood and Adolescence: The Korean Child-Adolescent Cohort Study. Nutrients. 2015;8(1):31. © 2020 Semnani-Azad et al. JAMA Network Open. 15. Kang YK, J. Soft drink consumption is associated with increased incidence of the metabolic syndrome only in women. British Journal of Nutrition. 2017;117(2):315-324. 16. Kim DK, J. Dairy consumption is associated with a lower incidence of the metabolic syndrome in middle-aged and older Korean adults: the Korean Genome and Epidemiology Study (KoGES). British Journal of Nutrition. 2017;117(1):148-160. 17. Lim MK, J. Association between fruit and vegetable consumption and risk of metabolic syndrome determined using the Korean Genome and Epidemiology Study (KoGES). Eur J Nutr. 18. Lutsey PLS, L. M.Stevens, J. Dietary intake and the development of the metabolic syndrome: the Atherosclerosis Risk in Communities study. Circulation. 2008;117(6):754-761. 19. Mirmiran PB, Z.Delshad, H.Azizi, F. Effects of energy-dense nutrient-poor snacks on the incidence of metabolic syndrome: a prospective approach in Tehran Lipid and Glucose Study. Nutrition. 2014;30(5):538-543. 20. Mirmiran PY, E.Asghari, G.Hosseinpour-Niazi, S.Azizi, F. Consumption of sugar sweetened beverage is associated with incidence of metabolic syndrome in Tehranian children and adolescents. Nutr Metab (Lond). 2015;12:25. 21. Sayon-Orea CB-R, M.Marti, A.Pimenta, A. M.Martin-Calvo, N.Martinez-Gonzalez, M. A. Association between yogurt consumption and the risk of metabolic syndrome over 6 years in the SUN study. BMC Public Health. 2015;15:170. © 2020 Semnani-Azad et al. JAMA Network Open.
JAMA Network Open – American Medical Association
Published: Jul 9, 2020
You can share this free article with as many people as you like with the url below! We hope you enjoy this feature!
Read and print from thousands of top scholarly journals.
Already have an account? Log in
Bookmark this article. You can see your Bookmarks on your DeepDyve Library.
To save an article, log in first, or sign up for a DeepDyve account if you don’t already have one.
Copy and paste the desired citation format or use the link below to download a file formatted for EndNote
Access the full text.
Sign up today, get DeepDyve free for 14 days.
All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.