Caloric Sweeteners and Health: What is the Truth? G. Harvey - PowerPoint PPT Presentation

Caloric Sweeteners and Health: What is the Truth? G. Harvey Anderson Professor Departments of Nutritional Sciences and Physiology Faculty of Medicine, University of Toronto. November 2 , 2016

Conflicts o of Interest • Consultant to many food and drug companies and associations. • Kelloggs, CocaCola, Mead Johnson, Nestle, Baxter- Travenol, EliLily, PepsiCo, Kraft, Alliance for Potato Research and Education, American Beverage Association, Winston Strawn LLP • Served on many industry science advisory committees • McCain Foods, Maple Leaf Foods, General Mills, McDonalds, Hillshire Brands, Unilever, Heinz, ADM, Healthy Grains Institute, Canadian Sugar Institute. • Own a farm in Ontario • Direct the UofT Program in Food Safety, Nutrition and Regulatory Affairs (15 Food Industry Members)

Caloric Sugars Under Siege • Institute of Medicine (2002) • Insufficient evidence to set a UL for either total or added sugars • Suggested that added sugars should not exceed 25% of total calories based on preventing the displacement of foods that are major sources of essential micronutrients • 2015 Dietary Guidelines Advisory Committee (DGAC) • Added sugars should be limited to less than 10% of calories • Based on analysis of USDA food patterns and a review of the evidence that added sugars negatively impacts the health risks for obesity, type II diabetes, cardiovascular disease and dental carries • World Health Organization (2015) • In both adults and children, intake of free sugars not to exceed 10% of total energy (strong recommendation) Based on moderate quality evidence from observational studies of dental caries • Further reduction to less than 5% of total energy (conditional recommendation) Based on very low quality evidence from ecological studies

Obesity Results from a failure to achieve energy balance Unclear whether obesity develops in susceptible individuals because physiological mechanisms of food intake control are compromised first or if these are simply overridden by the environment and become compromised (Ebbeling et al, JAMA 2004)

The Environment

The WHO's nutrition director, Dr Francesco Branca, said "nutritionally, people don't need any sugar in their diet". October 2016

Food Policy? Environment vs. Physiology • Epidemiology-We have the evidence! Associations enough. • Regulate the environment (food policies) • Tax SSB • Experimental-We don’t have the evidence? • Fat causes obesity? • High GI foods cause obesity • French fries cause obesity • SSB cause obesity • Sweetness causes obesity • Energy intake adjusts for exercise • Food TV ads increase food intake

The Sweet Environment • Exposure of sweet taste occurs from in utero to death • Frequency of exposure to sweetness has increased • Quantity of caloric sweetener consumption has declined in 20 years • Sweet foods and beverages are: • Safe • easily stored and transported, • need no preparation • inexpensive • Policy makers and government see no harm in reducing caloric sweetener consumption and see public support in taking action.

Dietary Sugars and Health: What does the evidence say? John L Sievenpiper, MD, PhD, FRCPC 1,2,3,4 1 Associate Professor, Department of Nutritional Sciences, University of Toronto 2 Consultant Physician, Division of Endocrinology & Metabolism, St. Michael’s Hospital 3 Scientist, Li Ka Shing Knowledge Institute, St. Michael’s Hospital 4 Knowledge Synthesis Lead, Toronto 3D Knowledge Synthesis Unit, St. Michael’s Hospital October 21, 2016

Hierarchy of evidence in evidence based medicine Systematic Reviews & meta-analyses RCTs Decreasing bias Non-randomized controlled trials (NRCT) Cohorts studies Case-control studies Cross-sectional studies Case series/time series Expert opinion http://www.sign.ac.uk/guidelines/fulltext/50/annexb.html http://www.cnpp.usda.gov/Publications/NutritionInsights/Insight38.pdf http://www.nice.org.uk/niceMedia/pdf/GDM_Chapter7_0305.pdf

Lack of relation of total sugars with diabetes: 13 cohort comparisons, n=108,170 (14,752 cases) Risk ratio IV random, 95% CI Benefit harm Relative Risk: 0.88 [0.74, 1.06] p = 0.17 Tsilas et al., CMAJ, under review

Lack of relation of fructose with diabetes: 6 cohort comparisons, n=107,972 (3,833 cases) Risk ratio IV random, 95% CI Benefit harm Relative Risk: 1.04 (0.84, 1.29) p = 0.72 Tsilas et al., CMAJ, under review

Lack of relation of sucrose with diabetes: 8 cohort comparisons, n=192,332 (4,535 cases) Risk ratio IV random, 95% CI Benefit harm Relative Risk: 0.89 (0.80, 0.98) p = 0.02 High BMI = Body Mass Index ≥ 29 kg/m 2 Low BMI = Body Mass Index < 29 kg/m 2 Tsilas et al., CMAJ, under review

SSBs and incident diabetes: Meta-analysis of 17 cohorts, N=464,937 (38,253 cases), FU=12y (3-21y) Adjusted for adiposity & Unadjusted for Adjusted for within person variation adiposity adiposity RR= 1.18 (1.09-1.28) RR= 1.13 (1.06-1.21) RR= 1.28 (1.12-1.46) RR=1.27 (1.10-1.46)* *Calibrated for publication bias Imamura et al. BMJ. 2015 Jul 21;351:h3576.

How do SSBs compare with other risk factors?

Increased servings of different foods contribute to weight change over 4 year intervals: NHS I (1986-2006), NHS II (1991-2003) and HPFS (1986-2006), N=120 877 +1.69lb +3.35lb +0.57lb +1.00lb +0.95 lb +0.93 lb +0.65lb +0.28 to 0.36lb **Multivariate adjustment for age, BMI, sleep, physical activity, alcohol, television watching, smoking, and all dietary factors** Mozaffarian et al. NEJM 2011;364:2392-2404

Increased servings of different foods contribute to weight change over 4 year intervals: NHS I (1986-2006), NHS II (1991-2003) and HPFS (1986-2006), N=120 877 -0.49lb -0.22lb -0.57lb -0.82lb -0.37lb -0.11lb **Multivariate adjustment for age, BMI, sleep, physical activity, alcohol, television watching, smoking, and all dietary factors** Mozaffarian et al. NEJM 2011;364:2392-2404

Relation of different dietary factors and incident type 2 diabetes in adults: Summary of 5 systematic reviews and meta-analyses or individual prospective cohort studies I 2 Food source of sugars Cohort Participants Cases Median Risk ratios (95% CIs) comparisons Follow-up Fried foods [104] 2 111,631 10,323 25y 1.55 (1.32 to 1.83) - Processed meat [105] 9 371,492 26,256 14y 1.51 (1.25 to 1.83) 94%* SSBs [14] 17 464,937 38,253 12y 1.28 (1.12 to 1.46) 73%* French fries [106] 1 84,555 4,496 20y 1.21 (1.09 to 1.33) - High GI diet [107] 20 394,039 35,715 11y 1.19 (1.14 to 1.24) 69%* Red meat [105] 10 442,101 28,228 14y 1.19 (1.04 to 1.37) 93%* Potatoes [106] 1 84,555 4,496 20y 1.16 (1.05 to 1.29) - High GL diet [107] 30 710,314 46,115 11y 1.13 (1.08 to 1.17) 26% 0 0.5 1 1.5 2 14. Imamura Fet al. BMJ. 2015 Jul 21;351:h3576. Benefit Harm 104. Cahill LE, et al. Am J Clin Nutr 2014;100(2):667-675 105. Pan A, et al. Am J Clin Nutr. 2011 Oct;94(4):1088-96. 106. Halton Tlet al. Am J Clin Nutr. 2006 Feb;83(2):284-90. 107. Bhupathiraju SN et al. Am J Clin Nutr. 2014 Jul;100(1):218-32. Sievenpiper et al. Can J Diabetes. 2016 Aug;40(4):287-95

So why do SSBs appear to be the special case? 1. Is it because liquid calories are poorly compensated? 2. Is it because they are easier to measure? 3. Is it because SSBs are a marker of an unhealthy lifestyle?

4 trial designs: To interpret results, follow the energy… “Substitution trials”= Energy from sugars substituted for other sources of energy in the diet “Addition trials”= Energy from sugars “added” to the diet “Subtraction trials” = Energy from sugars “subtracted” from the diet Ad libitum trials = Energy from sugars is freely replaced with other macronutrients

Sugars mediate weight change through excess calories Te Morenga et al. BMJ. 2012;345:e7492 Kaiser et al. Obes Rev. 2013 Aug;14(8):620-33. Malik et al. AJCN. 2013 Oct;98(4):1084-102.

Take away messages 1. It is difficult to separate the contribution of fructose-containing sugars from that of other factors in the epidemics of obesity, diabetes, and their complications, owing to the interaction with excess energy. 2. Any threshold for the effect of sugars on body weight and cardiometabolic risk is highly dependent on energy balance, nutrient adequacy (food sources). 3. There are many pathways to overconsumption leading to weight gain and diabetes. Dietary patterns that bring these pathways together have the greatest influence on weight gain and cardiometabolic risk and represent the best opportunity for successful interventions. 4. Targeting sugars as a source of excess calories remains a prudent strategy, as sugary foods and beverages can be a proximate pathway to overconsumption. 5. One cannot choose a healthy diet by sugars alone! A little sugar helps the wholegrains, fibre, fruits, and dairy/non-dairy alternatives to go down. Is the evidence enough for setting policy and regulations?

Recommendation: Not enough evidence to make policy Need to know more about sugars intakes from solid foods.