Calculating the global burden of disease due to low fruit and vegetable intake
Karen Lock, Joceline Pomerleau and Martin McKee
London School of Hygiene and Tropical Medicine

WHO Global Burden of Disease Study 1990
Previous study (1990) only considered 10 risk factors for disease
Malnutrition was only nutrition risk factor
In discussions for the update -
important to get more nutritional risk factors for non-communicable disease included

Comparative Risk Assessment, Global Burden of Disease study 2000
New approach - comparative risk assessment
Estimate contributions of 26 risk factors to global burden of disease using unified framework
Physiological, lifestyle, environmental and occupational factors
‘Nutrition-related’ risk factors included:
low fruit and vegetables, obesity, high serum cholesterol
Published in WHO World Health Report 2002:
Reducing Risks, Promoting Healthy Life

Risk factors included
Underweight
Iron, Vitamin A, Zinc deficiency
Tobacco
Alcohol
Illicit drugs
Unsafe sex
Lack of contraception
High Blood Pressure
Low fruit and vegetables
Obesity (high BMI)
High serum cholesterol
Unsafe water, sanitation
Outdoor air pollution
Indoor smoke
Climate change
Occupational risks
Childhood sexual abuse

General methods
For 14 WHO regions, by age and sex
Comprehensive reviews
Prevalence of risk factor exposure
Size of the ‘hazard’ or association (relative risk)
Estimate contribution of RF to disease (attributable fraction)
Applying potential impact fraction to burden of disease estimates in WHO GBD database
Counterfactual analysis: using theoretical minimum exposure

Fruit and Vegetable intake
Calculated average intake for each region (g/person/day)
Combined national intake survey data in regions
Where no surveys available: systematic extrapolation from FAO food balance sheets combined with survey data from other regions
Assumptions about whether population distribution of intake similar in countries

 Included intake surveys
Data available for 26 countries within 9 WHO Regions
Amr A: USA
Amr B:  Argentina, Mexico
Emr B:  Kuwait
Eur A:  Belgium, Denmark, Finland, France, Germany, Ireland, Israel, Italy, Norway, UK
Eur B:   Bulgaria
Eur C:   Estonia, Kazakhstan, Latvia, Lithuania, Russian Fed.
Sear D: Bangladesh, India
Wpr A:  Australia, Japan, Singapore
Wpr B:  China
No survey data: Afr D, Afr E, Amr D, Emr D, Sear B

Disease outcomes included
Systematic reviews of 6 diseases
Ischaemic heart disease
Stroke
Stomach, lung, oesophageal, colorectal cancer
Fruit and vegetables treated as continuous variable
Summary relative risks estimated using meta-analysis

Counterfactual distribution
Theoretical ‘minimum exposure’
= Maximum fruit and vegetable intake
600g/ day
Based on high intake populations, and level to which potential gains may continue given scientific evidence
Provides a ‘vision’ potential health gains

Global attributable mortality and DALY in year 2000 due to low fruit and vegetable intake

Attributable mortality (000) due to leading global risk factors (Ezzati et al 2002)

Attributable DALY (000) due to leading global risk factors (Ezzati et al 2002)

Burden of disease due to leading risk factors  in developed regions

Similar to previous estimates of the disease burden due to low fruit and vegetable intake
DALY’s lost due to inadequate fruit and vegetable consumption :
E.U (1997).: 3.5% (c.f. 1.1% for high saturated fat intake)
Australia (1999): 2.8% (10% of all cancer deaths)
New Zealand (2001): 2.4%

Key role of Nutrition in Global Burden of Disease
15% GBD due to under nutrition
BUT shows that significant burden of disease due to risk factors with dietary determinants including high cholesterol, obesity, F&V
Patterns not uniform across regions - represents nutrition transition

Implications for Policy
Puts fruit and vegetables as important nutritional risk factor on national and global policy agendas
Confirms high public health cost of low fruit and vegetable intake for non-communicable disease
1st direct comparison F&V with other leading risk factors in comparable way
Unified framework allows comparison of risk factors from a variety of PH disciplines (with caveats)
Both years of life and quality of life lost