Project aim: The Agency on Statistics under the President of the Republic of Tajikistan (AoS)
and the Ministry of Health (MoH) called upon FAO to technically assist the implementation
of regular collection of nutrition related data at the national level in the framework of the
EU-FAO project Support to the strengthening of the National Food Security Information
System in Tajikistan. Tajikistan already conducts a Household Budget Survey (HBS) at the
national level which has not, until now, included any module to collect nutritional data
either at individual or household level. Integrating the Women’s Dietary Diversity Score
(WDDS) into the regular collection of HBS data is very useful to assess and monitor the
country’s food security and nutrition situation in a quick and inexpensive way both at a
national as well as at a decentralized level, and represents a validated proxy indicator for
micronutrient adequacy of the target population.
Project design: The pilot project was executed in three phases: i) setting the scene for
implementation, including the mapping of the existing technical capacity for carrying out
pilot data collection, ii) capacity development and adaptation of the WDDS tool to the Tajik
context, and iii) design and implementation of the pilot study in the Khatlon region for
testing the tool.
Results and discussion: Twenty-one enumerators and staff of the Agency on Statistics (AoS)
under the Presidency of Tajikistan were trained in the administration of WDDS, from tool
adaptation to data collection and analysis, during a six-day training that took place at AoS in
Kurgan Tube, Khatlon region. After the training, the fully adapted WDDS tool was used for
pilot data collection in four urban and 11 rural settlements in Khatlon region, in June 2014.
A total of 331 women of reproductive age (WRA), aged 15-49 (34.5 ± 9.9), were interviewed.
Results of the pilot study show that WDDS values in Khatlon region are relatively high as
compared with previous studies conducted in different countries: the results indicate a
higher probability of micronutrient adequacy in the surveyed women. However, high values
might be partly due to the timing of data collection (June is the harvest season) and the
confined geographical location of the survey. The median WDDS was 6, ranging from 1 to 9.
High values of WDDS indicate a more diversified diet compared with low values of WDDS.
The food groups that accounted for the differences between low and high WDDS (low and
high dietary diversity) were: dark green leafy vegetables; other vitamin A rich fruits and
vegetables; meat and fish; eggs; legumes, nuts and seeds; milk and milk products. They
were consumed by ≥ 50% of women in the group with a highly diversified diet (WDDS ≥ 7)
but by less than 50% of women in the lowest WDDS category (WDDS ≤ 4). The consumption
of vitamin A rich foods was relatively high in all WDDS groups while the consumption of
haem-iron rich foods was only found in women with high WDDS. Almost all women
consumed starchy staples and other fruits and vegetables (non vitamin rich). Legumes, nuts
and seeds were consumed mostly by women with high WDDS, which revealed a potential
for promotion to other women in order to improve the overall dietary quality of Tajik
women.
Two socioeconomic determinants of WDDS were identified: living area (urban or rural) and
education level. Women living in the urban areas and women with a higher education level
showed a significantly higher dietary diversity. No significant associations were found for
age, physiological state (pregnancy or lactation), marital status, number of children,
responsibility for food preparation, total household income as well as income determined
by applying the household equivalent scale. The mean household income was significantly
higher in those women consuming flesh meat and sweets compared with those not
consuming those food groups, and it was significantly lower in women consuming dark
green leafy vegetables. However, the association of food group consumption with
household income reveals that higher income is not necessarily associated with a better
quality of the diet and it also suggests that a nutrition transition (persistence of
undernutrition with concurrent increase in overweight and obesity, even within the same
family) might be taking place in Tajikistan, as observed in other low-income countries.
Conclusion: The Nutrition module with WDDS was successfully integrated in Tajikistan’s
current HBS system, including the training of enumerators, adaptation of materials as well
as conduction of a pilot study and data analysis. A nationwide implementation of the
nutrition module covering different seasons, different geographic regions and socioeconomic
groups would provide invaluable national representative nutritional data for
decision making by the government. It would also help to identify regions and population
groups at high risks of malnutrition and micronutrient inadequacy.