DEBATE: Weight as a Measure of Health vs. Health at Every Size - - PowerPoint PPT Presentation

Christopher Gardner, PhD Professor of Medicine Nutrition Scientist

Stanford Prevention Research Center Stanford University, Department of Medicine

DEBATE:

Weight as a Measure of Health vs. Health at Every Size Concepts

No Conflicts of Interest to Disclose

Antonella Dewell Rise Cherin Valerie Alaimo Josephine Hau Susan Kirkpatrick Sarah Farzinkhou Dalia Perlman Jae Berman Mollie Shimer Diane Demis Lisa Offringa PhD Katherine Dotter Jennifer Hartle PhD Erin Avery Alana Koehler Abby King John Ioannidis Tom Robinson Manisha Desai Julie Parsonnet Robert Haile Tracey McLaughlin Justin Sonnenberg Kari Nadeau Jennifer Robinson Mandy Murphy Michelle Hauser MD Lucia Aronica PhD John Trepanowski PhD Liana Del Gobbo PhD Kenji Nagao, PhD Ben Chrisinger, PhD Katarina Balter, PhD Cindy Shih, MS

Summer Camp for Underserved Kids Full Circle Farm at Peterson Middle School

Outline

Ø My research & lessons learned

Ø Evidence for Obesity links to Morbidity/Mortality Ø Health at Every Weight Ø Take Home / Actionable Conclusions

Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults

The Evidence Report. National Institutes of Health. Obes Res 1998;6(Suppl)2:51S-209S.

≥55% energy from carbohydrate ≤30% energy from fat and approximately 15% energy from protein

Carb Fat Protein

>55% <30% ~15%

Low carb High carb

A Weight Loss Diet Study

Gardner, JAMA 2007;297:969-77

100 80 60 40 20 Protein Fat Carbohydrate

% Energy

From Low-Carb to Low-Fat

Gardner, JAMA 2007;297:969-77

LEARN Atkins Zone Ornish

6 months

30% 47% 23% 53% 29% 18%

1 year

32% 46% 22% 52% 29% 19% 17% 55% 28% 63% 21% 16%

8 weeks Carb Fat Protein

Gardner, JAMA 2007;297:969-77

A TO Z Study Diet Data NDS

3-day unannounced 24-hr recalls (3,137 recalls)

Data not presented

• 9
• 8
• 7
• 6
• 5
• 4
• 3
• 2
• 1

Base- line 8 weeks 6 months 1 year Atkins Zone Ornish

Percent weight change across time, by group

Atkins Zone LEARN Ornish 77 79 79 76 72 72 72 71 71 66 64 65 68 61 60 58

% Retention 1-year

88% 77% 76% 78%

Participants with available data

LEARN

Weight change as % of baseline

p<0.03

(Tukey’s studentized range test)

A vs. Z

Gardner, JAMA 2007;297:969-77

1 year

32% 46% 22% 52% 29% 19%

Carb Fat Protein

Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults

The Evidence Report. National Institutes of Health. Obes Res 1998;6(Suppl)2:51S-209S.

>55% <30% ~15%

Low-Carbohydrate National Guidelines

Gardner, JAMA 2007;297:969-77

Favored Group assigned to Atkins WEIGHT p=0.03 HDL-C p=0.0004 SBP p=0.001 DBP p=0.004

(not adjusted for multiple testing)

12-month net weight change (kg): Individual results

• 30
• 25
• 20
• 15
• 10
• 5

5 10 15 20

• 30
• 25
• 20
• 15
• 10
• 5

5 10 15 20

• 30
• 25
• 20
• 15
• 10
• 5

5 10 15 20

• 30
• 25
• 20
• 15
• 10
• 5

5 10 15 20

Gardner, JAMA 2007;297:969-77

12-month net weight change (kg): Individual results

• 30
• 25
• 20
• 15
• 10
• 5

5 10 15 20

• 30
• 25
• 20
• 15
• 10
• 5

5 10 15 20

• 30
• 25
• 20
• 15
• 10
• 5

5 10 15 20

• 30
• 25
• 20
• 15
• 10
• 5

5 10 15 20

Gardner, JAMA 2007;297:969-77

~30 kg RANGE of weight change WITHIN each diet group.

From losing 20-25 kg to gaining 5-10 kg

Percentage Change in Weight Percentage Change in Weight

• 30
• 30

Yancy et al., Arch Int Med, 2010;170:143

Dansinger et al., Comparison of the Atkins, Ornish, Weight Watchers, and Zone Diets for Weight Loss and Heart Disease Risk Reduction: A Randomized Trial. JAMA, 2005; 293:43-53

• 9
• 8
• 7
• 6
• 5
• 4
• 3
• 2
• 1

Base- line 8 weeks 6 months 1 year Atkins Ornish

Percent weight change across time, by group

Atkins Ornish 77 76 72 71 71 65 68 58

% Retention 1-year

88% 78%

Participants with available data

Weight change as % of baseline

Weight loss (kg)

Most Insulin Sensitive (<7 µIU/mL) Most Insulin Resistant (>10 µIU/mL)

Fasting Insulin Tertiles

0 2 6 12

• 10
• 8
• 6
• 4
• 2
• 10
• 8
• 6
• 4
• 2

Months

Ornish Diet (very low fat, high carb) Atkins (very low carb, unrestricted fat and protein)

A TO Z Study: Exploratory analyses

Mean, SEM n=19 n=24 n=23 n=21

Weight loss (kg)

Most Insulin Sensitive (<7 µIU/mL) Most Insulin Resistant (>10 µIU/mL)

Fasting Insulin Tertiles

0 2 6 12

• 10
• 8
• 6
• 4
• 2
• 10
• 8
• 6
• 4
• 2

Months

Ornish Diet (very low fat, high carb) Atkins (very low carb, unrestricted fat and protein)

A TO Z Study: Exploratory analyses

Mean, SEM n=19 n=24 n=23 n=21 Success with either diet for those who are relatively insulin sensitive For those who are insulin resistant, low-fat diet ineffective compared to low-carb diet

12-Month Weight Change (kg)

p=0.0006 p=0.06 p=0.01

Atkins Zone Ornish

• 10
• 8
• 6
• 4
• 2

2 Alhassan, Intl J Obesity, 2008; 57:49-56

Adherence tertiles

Highest Lowest

n= 23 n= 19 n= 19 n= 23 n= 19 n= 19

Weight Loss by Adherence Tertile (A TO Z Study)

+20% +10%

• 10%
• 30%
• 20%
• 40% 0

2 6 12 Months

Change in % Carbs

Fasting Insulin Tertiles Lowest (most Ins Sens) Highest (most Ins Res) Assigned to Atkins (Lowest Carb) Assigned to Ornish (Lowest Fat)

+10% +5%

• 5%
• 20%

+20% +15% +25%

• 10%
• 15%

2 6 12 Months

Change in % Fats

McClain et al., Diabetes Obes Metab 2013;15:87-90

Differential Adherence by Insulin Resistance Status

Insulin resistant individuals may find it inherently more difficult to adhere to a lower-fat/higher-carb diet

Ongoing Study: NIH R01 DK091831 + NuSI

Ø Study Population: Women & men, BMI 28-40,

age 18-50, non-diabetic, general good health

Ø Sample size: n=609 (enrollment complete) Ø Intervention: Healthy Low-Fat vs. Healthy Low-Carb

Weight loss diets Delivered in 22 instructional sessions (~17/class)

Ø Primary outcome: 12-month weight loss Ø Possible mediators/moderators:

Genome, metabolome, microbiome Insomnia, food addiction, psychosocial, many others

~20 grams/day (carbs or fat)?

Titr itrat ate up t e up to a

• a

le level y el you can

• u can

maint maintain….. ain…..

FOREVER OREVER How Low can you go?

Healthy Low-Fat vs. Healthy Low-Carb Healthy Low-Fat vs. Healthy Low-Carb

ID 14: Low Fat

BREAKFAST

2 slices whole wheat bread w/mustard Multigrain cereal w/skim milk Water

LUNCH

4 c salad mix w/ fat-free dressing Spinach spaghetti w/ marinara sauce Mid-afternoon snack Coffee

DINNER

Stir fried veggies w/ kung pao sauce, soy sauce and garlic on brown rice Water Evening Snack Pita bread w/ low-fat red pepper hummus

ID 36: Low Fat

BREAKFAST

Low Fat Latte Scone

LUNCH

Vegetable lasagna Soda Mid-afternoon snack Water Martini w/ olives

DINNER

Minestrone soup Linguini w/ shrimp, alfredo & marinara sauce Caesar Salad Evening Snack Red wine Chocolate cake

Low Fat ID 14 ID 36

ID 14 ID 36

1,950 36% 37% 17% 10% 15 g 1 g 36 g 39 g 1,700 13% 73% 14% 0% 45 g 1 g 4 g 20 g Kcal Fat Carbohydrate Protein Alcohol Fiber Omega-3 Saturated fat Added Sugars

Low Fat

ID 10: Low Carb

BREAKFAST

Tuna salad w/ tomatoes, olives & lettuce Fat free dressing Water

LUNCH

Deli ham Laughing cow cheese Afternoon Snack Coffee w/ half and half

DINNER

Chicken w/o skin Zucchini & Broccoli sauteéd in butter Evening Snack Strawberries & sparkling water

ID 51: Low Carb

BREAKFAST

Omelette w/ cheese, ham, spinach Coffee with half & half Water

LUNCH

Steak w/ cheese Pork ribs Bratwurst Broccoli salad Water

DINNER

Cheeseburger Sausage Avocado, tomato, spinach Red wine

ID 51 ID 10

Low Carb

ID 10 ID 51

2,150 66% 5% 23% 6% 8 g 2 g 61 g 4 g 1,200 48% 13% 39% 0% 13 g 2 g 21 g 5 g Kcal Fat Carbohydrate Protein Alcohol Fiber Omega-3 Saturated fat Added Sugars

Low Carb

Do Genotype Patterns Predict Weight Loss Success for Low Carb vs. Low Fat Diets? R01 DK091831 (2013-17) + NuSI

n=609

BMI 28-40 kg/m2 non-diabetic generally healthy adults 18-50 yrs ~55% women

Low Carbohydrate (n=305) . Low Fat (n=304)

Months 0

12 3 6

INTERVENTION:

22 group classes, 15-22 participants/group

XXXXXXXX X X X X X X X X X X X X X X

WEIGHT (1° outcome)

X X X X

Blood (DNA, lipids, glucose,

insulin, OGTT, cytokines)

X X (no OGTT) X X X X X X

Diet Assessment (NDS-R) Psychosocial (Questionnaires) X

X X X

DEXA, REE (Metabolic cart) Other: Microbiome (fecal samples) Adipocytes (fat biopsies), Other – Various times points, specific cohorts

X X X

90 100 110 120 130 140 150

>125 mg/dL cut off for diabetes Blood Glucose (mg/dL)

5 10 15 20 25 30 40 30 60 120

75 grams glucose Serum Insulin uU/mL

Non-Diabetic, Insulin Resistant

Minutes

JAMA November 27, 2013 Volume 310, Number 20 Alison E. Field, ScD, Carlos A. Camargo Jr, MD, DrPH, Shuji Ogino, MD, PhD

Obesity is a heterogeneous and complex disease influenced by exogenous and endogenous exposures. Stratifying obesity into meaningful subtypes could provide a better understanding of its causes and enable the design and delivery of more effective prevention and treatment interventions.

Outline

Ø My research & lessons learned

Ø Evidence for Obesity links to Morbidity/Mortality

Ø Health at Every Weight Ø Take Home / Actionable Conclusions

Age-Adjusted Prevalence of Obesity and Diagnosed Diabetes Among U.S. Adults Aged 18 Years or older

Obesity (BMI ≥30 kg/m2) Diabetes 1994 1994 2000 2000

No Data <14.0% 14.0-17.9% 18.0-21.9% 22.0-25.9% >26.0% No Data <4.5% 4.5-5.9% 6.0-7.4% 7.5-8.9% >9.0%

CDC’s Division of Diabetes Translation. National Diabetes Surveillance System available at http://www.cdc.gov/ diabetes/statistics

2010 2010

Obesity raises the risk of MORBIDITY from hypertension, dyslipidemia, type 2 diabetes, heart disease, stroke, gallbladder disease,

• steoarthritis, sleep apnea, respiratory

problems, and some cancers. Obesity is also associated with increased risk of all-cause and CVD MORTALITY.

…biomedical, psychosocial, and economic consequences…

More than 78 million adults in the US were

• bese in 2009 & 2010.

All of the following are associated with weight loss

AHA-style Step 1 Higher protein Higher protein Zone-type Lacto–ovo–vegetarian–style Low calorie Low carbohydrate Low fat Low fat vegan-style Lower fat, high-dairy Low–glycemic–load Macronutrient targeted diets Mediterranean style Moderate protein

With prescribed energy restriction, or Without formal prescribed energy restriction, but with a realized energy deficit. …if reduction in dietary energy intake is achieved:

Food, Nutrition and the Prevention of Cancer: a global perspective

World Cancer Research Fund American Institute for Cancer Research 1997

Second Expert Report Food, Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective 2007

American Institute for Cancer Research World Cancer Research Fund

Body Fatness

1.0

Random-Effects Hazard Ratios of All-Cause Mortality for Overweight and Obesity Relative to Normal Weight 1.10 1.20 1.40 1.30 0.80 0.90

Hazard Ratio

Body Mass Index

Flegal K, et al.

• JAMA. 2013;309:71-82

97 Prospective Studies

1.13

95% CI: 1.06 – 1.19

>30

Obese

< 25

“Normal”

Reference 0.93 25 to <30

Overweight

95% CI: 0.89 – 0.95

1.0

Random-Effects Hazard Ratios of All-Cause Mortality for Overweight and Obesity Relative to Normal Weight 1.10 1.20 1.40 1.30 0.80 0.90

Hazard Ratio

Body Mass Index

Flegal K, et al.

• JAMA. 2013;309:71-82

97 Prospective Studies

1.13

95% CI: 1.06 – 1.19

>30

Obese

< 25

“Normal”

Reference 0.93 25 to <30

Overweight

95% CI: 0.89 – 0.95

Relative to those who are “normal” weight (BMI <25), those who are overweight (BMI 25 to <30) have a 7% LOWER risk of all-cause mortality, while those who have obesity (BMI >30) have a 13% higher risk.

1.0

Random-Effects Hazard Ratios of All-Cause Mortality for Overweight and Obesity Relative to Normal Weight 1.10 1.20 1.40 1.30 0.80 0.90

Hazard Ratio

Body Mass Index

Flegal K, et al.

• JAMA. 2013;309:71-82

97 Prospective Studies

0.94 30 to <35

Obese I

95% CI: 0.86 – 1.03

1.25

95% CI: 1.13 – 1.39

>35

Obese II, III

< 25

“Normal”

Reference 0.93 25 to <30

Overweight

95% CI: 0.89 – 0.95

1.0

Random-Effects Hazard Ratios of All-Cause Mortality for Overweight and Obesity Relative to Normal Weight 1.10 1.20 1.40 1.30 0.80 0.90

Hazard Ratio

Body Mass Index

Flegal K, et al.

• JAMA. 2013;309:71-82

97 Prospective Studies

0.94 30 to <35

Obese I

95% CI: 0.86 – 1.03

1.25

95% CI: 1.13 – 1.39

>35

Obese II, III

< 25

“Normal”

Reference 0.93 25 to <30

Overweight

95% CI: 0.89 – 0.95

The HIGHER risk of all-cause mortality, is

• bserved in those who have obesity at the

higher levels of stage 2 and 3 and morbid

• besity (BMI >35).

Even those with stage 1 obesity (BMI 30 to <35) have a LOWER risk of mortality (not significant) than those with BMI <25

1.0

Body-Mass Index and Mortality among Adults with Incident Type 2 Diabetes

1.30 1.40 1.60 1.50 0.90

Hazard Ratio

Body Mass Index

Tobias & Hu, et al.

• NEJM. 2014;370:233-44

NHS & HPFS, >12,000, ~16 years, >3,000 deaths

18.5 – 22.4

Reference

22.5 – 24.9 25.0 – 27.4 27.5 – 29.9 30.0 – 34.9 >35.0 1.20 1.10

95% CI: 1.05 to 1.59

1.29

95% CI: 0.98 to 1.29

1.12

95% CI: 0.94 to 1.26

1.09

95% CI: 1.08 to 1.42

1.24

95% CI: 1.14 to 1.55

1.33

n=426 n=1,074 n=2,095 n=2,010 n=3,360 n=2,462

1.0

Body-Mass Index and Mortality among Adults with Incident Type 2 Diabetes

1.30 1.40 1.60 1.50 0.90

Hazard Ratio

Body Mass Index

Tobias & Hu, et al.

• NEJM. 2014;370:233-44

NHS & HPFS, >12,000, ~16 years, >3,000 deaths

18.5 – 22.4

Reference

22.5 – 24.9 25.0 – 27.4 27.5 – 29.9 30.0 – 34.9 >35.0 1.20 1.10

95% CI: 1.05 to 1.59

1.29

95% CI: 0.98 to 1.29

1.12

95% CI: 0.94 to 1.26

1.09

95% CI: 1.08 to 1.42

1.24

95% CI: 1.14 to 1.55

1.33

We found no evidence of lower mortality among patients with diabetes who were

• verweight or obese at diagnosis, as

compared with their normal-weight counterparts, or of an obesity paradox.

n=426 n=2,462

Outline

Ø My research & lessons learned Ø Evidence for Obesity links to Morbidity/Mortality

Ø Health at Every Weight

Ø Take Home / Actionable Conclusions

Lee, Blair, Jackson, Am J Clin Nutr. 1999;69:373-80 0.5

1.0

1.5 2.0 2.5 3.0 Fit Unfit Fit Unfit

Lean

Relative Risk all-cause mortality

5 10 15 20 25 30 35

Lean Overweight

All-cause deaths per 10,000 men per year

Conclusion: Fit and obese healthier than lean and unfit

Overweight

Fit and Fat?

50 100 150 200 250

LDL-C HDL-C TG Glucose INS-AUC DBP SBP Gender Age BMI % Body Fat Woman 42y 32 41% Woman 42y 32 30%

Same Gender, Age, BMI Different % Body Fat

mg/dL mmHg

50 100 150 200 250

LDL-C HDL-C TG Glucose INS-AUC DBP SBP Gender Age BMI % Body Fat Woman 39y 31 42% Woman 39y 30 41%

Same Gender, Age, BMI, and % Body Fat

mg/dL mmHg

Gender Age BMI % Body Fat LDL-C HDL-C Triglycerides Glucose INS-AUC SBP DBP Man 25y 34 34% 114 44 101 103 119 118 78 Man 48y 30 25% 88 27 429 103 224 125 82

Healthier at a Higher % Body Fat

Outline

Ø My research & lessons learned Ø Evidence for Obesity links to Morbidity/Mortality Ø Health at Every Weight

Ø Take Home / Actionable Conclusions

Reframe the Question

What is the “best diet”?

Take Home Point: #1

Reframe the Question

What is the “best diet”?

Which diet is best for whom? Take Home Point: #1

Stop using the term “diet”

For many people a “diet” is something you go on and off.

Which meal plan(?) is best for whom? Take Home Point: #2

Evidence for long-term weight loss maintenance?

INADEQUATE / LACKING

Take Home Point: #3

Individually Embrace the Variabiilty

Societally Food System Food Environment Social Justice

Take Home Point: #4