Does Diet Play a Role in Dementia Prevention? Data From the Nurses’ Health Study : Does Diet Play a Role in Dementia Prevention? Data From the Nurses’ Health Study Jae Hee Kang, ScD
Instructor in Medicine
Channing Laboratory
Department of Medicine
Brigham and Women’s Hospital, Boston MA
May 21, 2007
Objectives : Objectives Cognitive function cohort studies:
Purpose: to determine relations between key epidemiologic risk factors and cognitive aging
Particular interest: to identify risk factors for earliest stages of cognitive decline
Why nutrition is a good target:
It can be changed
It is under control of the individual
Objectives : Objectives Four topic areas:
Fruits, vegetables, and antioxidants
Fat consumption
Insulin and diets that promote insulin resistance
Moderate alcohol intake
Impact of onset delay on US AD population projections : Impact of onset delay on US AD population projections Intervention (1998)
delay of 6 months
delay of 1 year
delay of 2 years
delay of 5 years
Effect (by 2007)
100,000
210,000
570,000
1.15 million
Brookmeyer et al, Am J Pub Health, 1998
Slide5 : FRUIT, VEGETABLE AND
ANTIOXIDANT INTAKE
Antioxidants: Background : Antioxidants: Background Oxidants
Products of normal aerobic metabolism and the inflammatory response
Oxidative damage in Alzheimer’s disease
One of the earliest pathophysiologic events
How might antioxidants prevent Alzheimer’s? : How might antioxidants prevent Alzheimer’s?
Reduce free radicals that harm brain cells
Decrease the toxicity of the amyloid protein
Lower cardiovascular disease risk
Where do we find antioxidants? : Where do we find antioxidants? Beta Carotene: in foods that are orange in color (carrots, cantaloupe, squash, pumpkin) and some green leafy vegetables
Lutein: in green leafy vegetables (spinach, collard greens, kale)
Lycopene: in tomatoes, watermelon, papaya, apricots, and pink grapefruit
Selenium: in rice, wheat and brazil nuts
Vitamin C: in fruits and vegetables
Vitamin E: in almonds, wheat germ, safflower, corn, and soybean oils, nuts, and broccoli
Antioxidants: Epidemiological studies : Antioxidants: Epidemiological studies Several studies have shown that higher intake of antioxidants, is associated with a lower risk of cognitive decline and Alzheimer disease
Fruits and vegetables and cognitive function in NHS : Fruits and vegetables and cognitive function in NHS Nurses’ Health Study (NHS): prospective cohort of 121,700 RNs, began in 1976; followed biennially
Cognitive study: ~ 20,000 NHS participants free of stroke and aged 70-81 years at baseline (1995+) Kang et al, 2005 Ann Neurology
NHS: Telephone Cognitive Assessment : NHS: Telephone Cognitive Assessment Brief telephone interview cognitive tests:
Telephone Interview of Cognitive Status – TICS
Verbal Memory: immediate and delayed recall of TICS 10-word list, East Boston Memory Test- immed. and delayed
Category Fluency: Animal naming test
Working Memory: Digit span backwards
Global score of all of the above tests
Participation rate: 93% follow-up
NHS: Measurement of Fruit & Vegetable Intake and Cognitive Function : NHS: Measurement of Fruit & Vegetable Intake and Cognitive Function Start ‘80 ‘84 ‘86 ‘90 ‘94 ‘98 ‘03
of NHS Dietary Assessment using Food Frequency Questionnaires 1st cognitive
assessment 2nd cognitive
assessment
Slide13 : FFQ: food frequency questionnaire
Validation: 4 1-week weighed diet records vs. FFQ correlation (r) was 0.9
Long-term average intake:
Fruits + vegetables: Median 5.6 servings/day
Green leafy vegetables: Median 0.8 servings/day
Cruciferous vegetables: Median 0.4 servings/day NHS: Measurement of Fruit & Vegetable Intake
NHS: Results : NHS: Results
No association with all fruits
Modest inverse associations between high intake of total vegetables and specific types of vegetables and cognitive decline
Green Leafy Vegetables: consumption at highest vs. lowest quintile was cognitively equivalent to being 1.7 years younger
Cruciferous Vegetables: consumption at highest vs. lowest quintile was cognitively equivalent to being 1.3 years younger
Chicago Health and Aging Project confirms finding with vegetables : Chicago Health and Aging Project confirms finding with vegetables Morris et al 2006 Neurology
ANTIOXIDANTS: ROTTERDAM STUDY : ANTIOXIDANTS: ROTTERDAM STUDY TERTILE OF VITAMIN E INTAKE Engelhart et al., JAMA, 2000 RATE
RATIO
OF ALZHEIMER’S DISEASE
ANTIOXIDANTS: ROTTERDAM STUDY : ANTIOXIDANTS: ROTTERDAM STUDY TERTILE OF VITAMIN C INTAKE RATE
RATIO
OF ALZHEIMER’S DISEASE Engelhart et al., JAMA, 2000
Slide18 : Grodstein et al, Am J Clin Nutr, 2003 TICS 10-word immed 10-word
delayed EBMT
immed EBMT
delayed * * Fluency Digit span Global score NHS: Vitamin E and C users, compared with non-users * *
Slide19 : Petersen et al, N Engl J Med, 2005
Summary: Fruits, Vegetables and Antioxidants : Limitation of studies
Duration of follow-up
Later food frequency questionnaires with possible cognitive dysfunction (responses require sustained motivation, attention and memory)
Unmeasured confounders
Antioxidant intake from food vs. supplement sources – why different results? Summary: Fruits, Vegetables and Antioxidants
INSULIN AND PROMOTERS OF INSULIN RESISTANCE : INSULIN AND PROMOTERS OF INSULIN RESISTANCE
Insulin and cognitive function : Epidemiologic evidence links diabetes to dementia
Elevated insulin secretion and insulin resistance may also be risk factors for cognitive decline
Factors that promote insulin resistance and elevated blood insulin levels:
Obesity, especially “spare tire” distribution
Dietary: high glycemic index pattern and excess intake of sugary foods Insulin and cognitive function
Insulin secretion and cognitive function in NHS : Insulin secretion and cognitive function in NHS Study population:
Women who had blood drawn between 1989-1990 and had plasma c-peptide measured
C-peptide is an excellent measure of insulin secretion
Exclusions: anyone with diabetes
Total of 718 women
Average age at blood draw: 64 years
Average interval between blood draw and cognitive testing: 10 years
Mean Differences in Cognitive Function, by C-peptide Quartile : Mean Differences in Cognitive Function, by C-peptide Quartile Q1 Q2 Q3 Q4 0
-0.5
-1.0
-1.5
VERBAL MEMORY GLOBAL SCORE Q1 Q2 Q3 Q4
0
-0.5
-1.0
-1.5
-2.0
-2.5
-3.0
Okereke et al., Arch Intern Med, 2005
Relative Risk of Poor Cognitive Performance, by C-peptide Quartile : Relative Risk of Poor Cognitive Performance, by C-peptide Quartile QUARTILE OF PLASMA C-PEPTIDE * * * Okereke et al., Arch Intern Med, 2005
Slide26 : FAT INTAKE
Fat: mechanisms : Fat: mechanisms Different fats may raise or reduce risk of cardiovascular disease – for example, by impact on balance of good/bad cholesterol
Possible interactions with APOE gene status
High intake of saturated fats may promote insulin resistance and high insulin concentrations
High intake of some fats may induce oxidative stress – implicated in Alzheimer and coronary artery diseases
Slide28 : Columbia Aging Project Luchsinger et al.,Arch Neurol,2002
Slide29 : Chicago Health and Aging Project Morris et al., Arch Neurol, 2003
Consumption of Fish and Omega-3 Fatty Acids: : Consumption of Fish and Omega-3 Fatty Acids: Chicago Health and Aging Project: 815 people, aged 65-94 years
Highest intake of fish (once or more weekly) vs. rare/never: RR=0.4 (CI 0.2-0.9)
Highest intake of omega-3 fatty acids vs. lowest: RR=0.4 (CI 0.1-0.9) Morris et al, Arch Neurol 2003
Slide31 : Fat Intake: Rotterdam Study Engelhart et al., Neurology, 2002
Nurses’ Health Study: bad fats are associated with worse performance among diabetics (n=1486) : Nurses’ Health Study: bad fats are associated with worse performance among diabetics (n=1486) Difference in Performance (Global Score)
Sat. fat -0.12 (-0.22, -0.01)
P:S ratio 0.08 (0.01, 0.16)
Trans fat -0.15 (-0.24, -0.06)
Summary: Fat intake : Summary: Fat intake Still limited data re: cognition and dementia
Cardiovascular health :
low in saturated and trans fatty acids and high in monounsaturated and polyunsaturated (fish) fats, especially omega-3
Randomized trial data are not available
Slide34 : ALCOHOL INTAKE
Alcohol: Mechanism : High intake of alcohol adverse
Neurotoxic at higher doses to brain structures
Light to moderate drinking protective
Increases “good” cholesterol
Decreases platelet adhesiveness
Protects against heart disease, stroke
Wine: flavonoids act as antioxidants Alcohol: Mechanism
Alcohol and Dementia: Rotterdam Study : Alcohol and Dementia: Rotterdam Study Ruitenberg et al., Lancet, 2002 Hazard
Ratio
Other Evidence : Other Evidence Honolulu Asia Aging Study:
3,556 men age 71-93 years who had been followed for 18 years
Drinkers up to 1/day had better scores than non-drinkers, but 4+/day was associated with cognitive impairment
Columbia Aging Study:
980 people without dementia studied for 4 yrs
Light to moderate (1-3 /day) pattern: reduced risk by 40%
Nurses’ Health Study : Nurses’ Health Study Participants
12,480 subjects at baseline
11,102 with follow-up cognitive data
Measurement of alcohol intake
Administered 1980, 84, 86, 90, 94, 98
Frequency of wine, liquor, beer consumption
Slide39 : Risk of Substantial Cognitive Decline by Alcohol Stampfer et al., NEJM, 2005
Summary: Alcohol intake : Summary: Alcohol intake
Consistent associations with moderate alcohol intake in observational studies
Randomized trials unlikely
Pay attention to health risks (may overshoot with excessive intake)
However, if current consumption is 1-2 drinks a day, there should be no need to reduce (if no other medical reason to reduce)
Summary : Summary To date, preliminary data show nutrition is a promising research focus in cognitive aging
Future research directions
Mediterranean diet pattern
Carbohydrate balance and the glycemic index
Gene-diet interactions (e.g., types of fat and apolipoprotein E genotype)
Summary : Summary Overall, we need more studies of preventive approaches, including diet
UNTIL THEN
- get 5 servings of fruits & vegetables a day
- lay off sweets and eat more whole grains
- consume heart-healthy fats
- don’t stop light to moderate drinking
Acknowledgements : Acknowledgements
Advisors, Mentors, Colleagues
Francine Grodstein
Olivia Okereke
Elizabeth Devore
Meir Stampfer