A Cluster Randomised Controlled Trial On The
Effect Of Plant Based Homestead Food
Production (HFP) With And Without Small-
Scale Aquaculture On Nutritional Outcomes In
Rural Cambodia: Rationale And Study Design
Tim Green, Judy McLean, Larry Lynd
Land and Food Systems
University of British Columbia
Zaman Talukder, Hou Kroeun
Helen Keller International, Cambodia
Effect of Agricultural Interventions on
Nutritional Outcomes
Participation in Program
Technology Adoption
Household Income
Food Expenditure
Energy, Protein, and Micronutrient Intake
Diet Composition
Nutritional Status (outcomes)
BMJ 2012;344:d8222
Effect of Agricultural Interventions on
Nutritional Outcomes
Participation in Program
Technology Adoption
Household Income
Food Expenditure
Energy, Protein, and Micronutrient Intake (outcome)
Diet Composition
Nutritional Status (outcome)
Masset et al BMJ 2012;344:d8222
What are these Outcomes?
• Income
• Diet composition/
diversity
• Dietary Intake
• Biochemical Indicators
– Anemia, serum retinol,
etc.
• Anthropometric
– Stunting, underweight,
wasting
Outcomes
• Short term (3-12 mo)
– Dietary Intake
• Medium term (9-24 mo?)
– Biochemical Indicators
• Longer term (5+ years)
– Anthropometric
– Key indicator (stunting)
Limitations of Agricultural
Interventions
• Lack of adequate control
• Lack of statistical power
• Lack of baseline surveys
• Lack of proper outcomes
– Lack of biochemical indicators beyond
hemoglobin and serum retinol
– Poor measures of dietary intake
• Insufficient duration
BMJ 2012;344:d8222
Control
• No Control
– Before and after intervention
Province A
Intervention
N=1000
Province B
Control
N=1000
Control
• No Control
– Before and after intervention
Province A
Intervention
N=1000
Province B
Control
N=1000
Statistician says n =2
Limitations
• Lack of adequate control
• Lack of statistical power
• Lack of baseline surveys
• Lack of proper outcomes
– Lack of biochemical indicators beyond
hemoglobin and serum retinol
– Poor measures of dietary intake
• Insufficient duration
Summary
• Studies examining the benefits of
agricultural interventions have been
limited
• Those that have been conducted
suffered from design problems
• Duration and lack of optimal outcomes
major problems
Cambodian Key Demographics
• Population: 14 million
• Life Expectancy:
64 (female), 61 (male)
• Provinces: 24
• Villages: 14,073
• Health Centres: 960
• Hospitals: 82
• 47% households with
access to safe water
Background: Food Security
Cambodia is RICE secure
• 5% of children <5 years die
– 30% due to undernutrition
• 40% of children <5 years are stunted
• 55% of children <5 years are anemic
• 44% of women are anemic
Study Site Prey Veng Province
• Agricultural Area
• Close to Phnom Penh
• Suitable for aquaculture
Research Objective
To improve household food security and
nutrition outcomes, livelihoods and women’s
empowerment through an integrated
homestead food production model.
Objective
To demonstrate the effectiveness of HFP
with and without aquaculture using a
cluster- randomized controlled trial
Emphasis on Women and
Young Children
Why Fish Ponds?
Study Design
VMF
F
Cluster:
1 Village Model Farm
10 Village Farmers
Study Design (N=990 households)
n=330
Comparison Arm
n=330
n=330
Study Design
• Represented by a woman
• Fall within the “poor”category
• Have access to land
• Have a child <5 years of
age
• Have suitable land for
pond
Household Inclusion Criteria:
Timeline
Apr Jun Aug Oct Dec Feb Apr Jun Aug Oct Dec Mar Apr Jun
2012 2014
Aug
BL
Survey
EL
Survey
Monitoring
Baseline Survey
• Identical to endline
survey
• Allows us to adjust
for differences that
existed at baseline
• Increases power
and reduces sample
size required
Outcomes
• Income
– All sources of
income
• Diet composition/
diversity
– Food production
– HDDS
Short term Outcome –
Dietary Intake• 24 hour recall on
mother and youngest
child under five
– Measurement of
nutritional composition
of fish
Medium term Outcome –
Biochemical assessment• Hemoglobin
• Ferritin/ CRP
• Transferrin saturation
• Hemoglobinopathies
• Vitamin B12
• Retinol Binding
Protein
• Thiamin and riboflavin
• Zinc
• Fatty acids
Only on 450 women
Hemocue on children
Long term Outcome –
Anthropometrics• Height
• Weight
• Stunting
• Wasting
• Study will not be
long enough to
assess changes
Food Security Status (HFIAS)
• Only 18% of households were food secure
• 83% were worried the family would not
have enough to eat (previous month)
• 43% only ate 2 meals the previous day
Baseline Survey
• Women
–Anemia 40%
• Microcytic 61%
– Hemoglobinopathy
44%
–Underweight 13%
–Consumed fish 73%
• Children
–Anemia 60%
–Stunting 30%
–Underweight
26%
–Wasted 7%
Other Components
• Environment
• BCC
• Hygiene and
Sanitation
• Fisheries
Technology (mixed
pond aquaculture)
• Gender
• Cost/Benefit
Thank you!
