A Framework for Determining the Physiological Benefits of Dietary Fiber

Transcription

1 A Framework for Determining the Physiological Benefits of Dietary Fiber Session 11, March th Vahouny Symposium Jonathan DeVries Sr. Technical Manager Medallion Laboratories TM Slides from today s presentation will be available online:

2 Keys to Today s Session Determining Physiological Benefits of DF Why? Without benefits-no need for DF labeling What? Defining adequate physiological effect-quality and Quantity Immediate Concern? Codex definition-parts 2 and 3 Proposed US regulations How? Harmonious approaches/tools to determining adequacy TM **Note: we will be recording the panel discussion for note taking purposes

3 To Address These Topics: Nicola McKeown-Dietary Fiber Database Julie Jones-Beneficial Effects-Variety Stuart Craig-Regulatory Perspective(s) Panel/Open Discussion TM

4 The Nine Vahouny Health Benefits: Launch of a New Database Nicola McKeown Nutritional Epidemiology Tufts University Boston, MA, USA TM

5 DATABASE-A Key Tool Developed based on objective criteria Public access encourages consistency Database benefits rulemaking Consumer education Research tool for current stakeholders Tool for guiding future research TM

6 The Nine Vahouny Health Benefits: The Launch of a New Database 10 th Vahouny Dietary Fiber Symposium 28 th March 2014 Nicola McKeown, PhD Nutritional Epidemiology Program

7 Funding Acknowledgment International Life Sciences Institute.

8 Fiber and Health Outcomes Fibers differ in structure, physical & chemical composition, and may confer different physiological health benefits Human studies demonstrate health benefits of fiber but vary considerably Different types of fiber administered Variation in the experimental design Dose and duration.

9 Defining Fiber Codex Alimentarius Commission adopted a definition in 2009 that provides the foundation for global agreement on the nature and identity of dietary fiber 3 fiber categories Codex 1: Edible carbohydrate polymers found natural in foods as consumed Codex 2: Carbohydrate polymers obtained from food raw material by physical, enzymatic, or chemical means Codex 3: Synthetic carbohydrate polymers Codex 2 & 3 must demonstrate physiological health benefit Codex Alimentarus Commission. Guidelines on Nutrition Labeling, CAC/GL Joint FAO/WHO Food Standards Programme, Secretariat of the CODEX Alimentarius Commission. Rome, Italy, 1985 (amended 2010).

10 What constitutes a physiological health effect? At the 9 th Vahouny Fiber Symposium, ILSI Europe & ILSI North America agreed upon a list of 9 physiological health effects: 1. Total and LDL cholesterol 2. Post-prandial glucose & insulin 3. Blood pressure 4. Increased fecal bulk and laxation 5. Transit time 6. Colonic fermentation & SCFA production 7. Modulation of colonic microflora 8. Weight loss, weight maintenance, and reduction in adiposity 9. Increased satiety Howlett JF, Betteridge VA, Champ M, Craig SA, Meheust A, Jones JM. The definition of dietary fiber - discussions at the Ninth Vahouny Fiber Symposium: building scientific agreement. Food Nutr Res 2010;54.

11 Fiber Database & Evidence Map If we want to link dietary fibers to health outcomes, we need to compile and synthesize the available body of literature 1. Create a database that systematically compile and provide access to the science linking dietary fibers, according to codex fiber type designation, to one or more of 9 potential health benefits 2. Develop an evidence map to describe the quantity, design and characteristics of research in this field.

12 Fiber Database: Objectives 1. Facilitate a better understanding of how different fibers are characterized in studies 2. Facilitate identifying gaps in the current research 3. Create database to serve as a foundation for conducting future evidence-based reviews and meta-analyses 4. Efficiently assist researchers in identifying fibers of interest 5. Long-term: provide annual, updated evidence on fiber literature

13 Free, publicly available data repository Web-based and serving two main purposes: Data repository (open and searchable archive ) Data extraction tool Developed by the Evidence-Based Practice Center at Brown University with support from the Agency for Healthcare Research and Quality Allows for ongoing critique, updating & revision Improves data access Provides an efficient way to conduct and update systematic reviews Promotes transparency and reliability in the systematic review process Used PICO paradigm to develop questions for fiber database Population, Intervention, Comparator, Outcomes Huang et al. Evaluation of PICO as a Knowledge Representation for Clinical Questions AMIA Symposium Proceedings.

14 Identifying Papers for Inclusion in Database Step 1: Create a search criteria Inclusion criteria English language Abstract/keywords must contain a fiber term and one of the Vahouney health outcomes Included designs from observational, intervention, and limited animal studies keywords for search: RCTs, clinical trials/trials, random allocation, double, single, triple blind methods, placebos, comparative studies, evaluation studies, follow-up studies, prospective studies, volunteer studies, cross over studies, intervention and dietary intervention studies; observational studies defined as case-control, matched pair, cohort, prospective, longitudinal, or follow-up Exclusion criteria Reviews/bibliographies Prevalence and cross-sectional designs Animal studies, unless the study was looking at outcomes related to colonic fermentation and SCFA production or modulation of colonic microflora

15 Identifying Papers for Inclusion in Database Create a search criteria Conduct search in OVID Searched published papers in Medline, CAB Abstracts, Cochrane Central Register of Controlled Trials

16 Identifying Papers for Inclusion in Database Create a search criteria Conduct search in OVID Screen abstracts in ABSTRACKR

17 Abstract screening: ABSTRACKR Free, open source platform, developed by EBP Center at Brown University Allows for double (or more) abstract screening (we performed 10% double screen) Facilitates organization and rapid classification of abstracts Ability to maintain a file of included, excluded, and questionable abstracts Tagging function

18 Identifying Papers for Inclusion in Database Create a search criteria Conduct search in OVID Screen abstracts in ABSTRACKR Conduct full text review of included abstracts

19 Review Process n=11,833 abstracts identified in OVID Further exclusions made: duplicate studies from search, in vitro studies and animal studies n=542 pending text request from library or further review n=7,257 abstracts screened n=5,210 excluded n=18 papers from hand search included n=2,047 moved to full text screen phase + n=735 included n=770 excluded but tracked Reasons for exclusion Review, bibliography, case report, etc Cross-sectional or prevalence study Population exclusions: infants (<3 yrs); pregnant or breastfeeding; has any type of cancer; has bowel disease; has renal failure; has other disease condition (i.e. ileostomy) Intervention has no control arm Fiber dose not reported Intervention not sufficiently controlled to measure effect of fiber Animal only study Observational study In vitro study

20 Identifying Papers for Inclusion in Database Create a search criteria Conduct search in OVID Screen abstracts in ABSTRACKR Conduct full text review of included abstracts Prioritize papers to be entered into database

21 Identifying Papers for Inclusion in Database Create a search criteria Conduct search in OVID Screen abstracts in ABSTRACKR Conduct full text review of included abstracts Prioritize papers to be entered into database Build fiber database in SRDR

22 Our Fiber Database in SRDR

23 SRDR List of Studies Project manager is able to upload studies to SRDR using PMID and assign studies to team members for data extraction Progress (% of the entry completed) is visible allowing for return to incomplete entries Progress: 50% Progress: 100% Progress: 0%

24 Data Capture: Publication Information

25 Data Capture: Design We are collecting the following study design information : Design type, blindness, diet type, level of feeding control (i.e. food provided), age, health status, anthropometry, gender, sample size, run-in and/or washout periods RCT crossover RCT parallel Non-randomized controlled trial Other Single blind Double blind Unspecified Other Weight loss Isocaloric/maintenance Hypercaloric Acute feeding study Unspecified Other Food recommended (nutrition counseling intervention) Food partially provided (partial metabolic study) All food provided (metabolic study) Unspecified Other

26 Data Capture: Design (Fiber Exposures) Capturing a maximum of 4 fiber exposures Natural food Food raw material Synthetic Diet Single food Powder Tablet Beverages Combination of foods Combination of beverage + foods Test meal Unspecified

27 Data Capture: Design (Comparators) Capturing a maximum of 4 comparators

28 Data Capture: Health Outcomes Capturing up to 8 health outcomes (prioritizing entry of Vahouny outcomes) and allowing for free text field to list additional outcomes V-Total & LDL cholesterol V-Postprandial glycemia/insulinemia V-Blood pressure V-Fecal bulk/laxation V-Transit time V-Modulation of colonic microfloria V-Colonic fermentation/scfa production V-Weight/Adiposity V-Satiety Other

29 Next Step: Create an Evidence Map Create a search criteria Conduct search in OVID Screen abstracts in ABSTRACKR Conduct full text review of included abstracts Prioritize papers to be entered into database Build fiber database in SRDR Create evidence map

30 What is an Evidence Map? Method of identifying, organizing, and summarizing scientific evidence on a broad topic Systematically examines extent and range of research activity (but does not summarize results) Replicable, potentially non-exhaustive Valuable in areas of fast-paced research where gold standard reviews and meta-analyses can become quickly outdated Useful for identifying: How published reviews fit into the field Existing mix of studies on a topic Research gaps & research questions

31 Fiber Evidence Map: PILOT DATA Study Characteristics n=174 papers with cholesterol as an outcome in the abstract were examined (10% had multiple fiber exposures) 66% of studies had isocaloric/maintence diet; 22% had weight loss diet Study Design n (%) RCT Parallel 95 (54.6) RCT Crossover 74 (42.5) Other 5 (2.9) Region of publication Australia/New Zealand South America Asia Study population characteristics Age (years) Gender (% male) BMI (kg/m 2 ) Mean ± SD 48.0 ± ± ± 4 Range Europe North America

32 List of Fiber Types Oat β-glucan Psyllium Combination/Mixture Gum, guar Psyllium hydrophilic mucilloid (Metamucil) Dietary fiber Chitosan Oat Bran Cereal fiber Inulin Legume fiber Pectin Wheat Bran Barley β -glucan Cellulose Fruit fiber Konjac mannan/glucomannan Rice Bran Arabinogalactan β -glucan (non-specific) Corn bran Flaxseed fiber Galactooligosaccharide Gum, locust bean Hydroxypropyl methylcellulose Rye Bran Soy fiber Wheat starch Agar Alginates Alphacyclodextrin Balsamodendron mukul Calcium polycarbophil Chitin-glucan Cocoa bran Corn starch/corn flour/maize starch Dextrin Fructan Fructooligosaccharide Gum, carboxymethyl cellulose Gum, karaya Gum, xanthan Highly viscous polysaccharide Lignin Lupin kernel fiber Lupin kernel flour Methylcellulose Oat Fiber Partially Hydrolyzed Guar Gum (PHGG) Pea fiber Polydextrose PolyGlycoplex (PGX) Potato fiber Resistant Starch Senna Soluble fiber Soy hulls Sugar beet fiber Tannin-rich fiber Whole grain oats Xylo-oligosaccharide

33 What about the combined fibers? Fiber Combinations n (%) Acacia gum & psyllium husk 1 (7.1) Arabic gum & pectin 1 (7.1) Flaxseed fiber & oat bran 1 (7.1) Guar gum & oat flour 1 (7.1) Guar gum and pectin with soy fiber, pea fiber, and corn bran 1 (7.1) High fiber mixed diet 2 (14.3) Inulin & oligofructose 2 (14.3) Mixture of guar gum and pectin with mixture of soy fiber, pea fiber, and corn bran 1 (7.1) Plantago ovata seed husks & glucomannan 1 (7.1) Psyllium & beta-glucans 1 (7.1) Psyllium & pectin 1 (7.1) Soy, other vegetable protein, and soluble-fiber foods 1 (7.1)

34 Fiber Types and Codex Classification Total and LDL Cholesterol Outcomes Synthetic Natural Food Food Raw Material Barley β-glucan Chitosan Combination/Mixture Guar Gum Inulin Oat β-glucan Type of Fiber Oat Bran Pectin Psyllium Metamucil* * Psyllium hydrophilic mucilloid

35 Fiber Type and Vahouny Outcomes Weight/ adiposity Other Vahouny Outcomes Satiety Postprandial glycemia/insulinemia Transit time Blood pressure Fecal bulk/laxation Colonic fermentation/ SCFA production Barley β-glucan Chitosan Combination/Mixture Guar Gum Inulin Oat β-glucan Type of Fiber Oat Bran Pectin Psyllium Metamucil* * Psyllium hydrophilic mucilloid

36 Acknowledgments Colleagues in Nutritional Epidemiology Department Paul Jacques, PhD Gail Rogers, MS Kara Livingston, MPH Micaela Karlsen, PhD student Jiantao Ma, PhD student Naglaa El-Abbadi, PhD student Nicole Baruch, MS Friedman School of Nutrition Science and Policy Nutritional Epidemiology students Caleigh Sawicki, MS Caitlin Blakeley, MS Carrie Brown, MS Yinan Jia, MS

37 DATABASE-A Key Tool Developed based on objective criteria Public access encourages consistency Database benefits rulemaking Consumer education Research tool for current stakeholders Tool for guiding future research TM