Dietary indicators need overhauling, food diary study finds

A recent study assessing dietary quality in Indonesia showed there is an urgent need to develop and validate assessment methods that will address the triple burden of malnutrition in an integrated way.

Anne-Marie Mayer's picture
Anne-Marie Mayer is a nutrition research consultant for IIED on the SD4ALL programme
15 October 2019
An Indonesia market trader stands at his stall while women customers look at the food on sale

Women in Jember, Indonesia, purchasing food from local vendors (Photo: Ian Rahmadi/Hivos)

As well as being home to more than half the world’s undernourished children, the Asia and Pacific region has the fastest-growing prevalence of childhood overweight and obesity.

Between 2000 and 2017, the prevalence of overweight children under five in Southeast Asia increased by a huge 128%.

Indonesia, where more than 30% of children are chronically undernourished and 11% are overweight, is one of seven countries in the region that have public health concerns around both under-nutrition and overweight. Together with micronutrient deficiencies, these form the triple burden of malnutrition.

Assessing dietary quality in Indonesia

As part of the Sustainable Diets for All programme (SD4ALL), IIED and Hivos are working with local partner Tanoker to promote healthy diets in East Java’s Jember Regency, Indonesia.

A soon-to-be-published SD4ALL study conducted with Tanoker and the University of Jember found that almost four in 10 children aged from 5-18 have insufficiently diverse diets. Their diets also include frequent consumption of ultra-processed foods that are high in fat, sugar and salt, convenient, cheap and lacking in essential nutrients.

The study used seven-day, self-administered food diaries to assess the diets of 97 households, measuring dietary diversity by counting the number of food groups individuals consumed.

The target was to eat something from at least five of the following groups in each 24-hour period: grains, white roots, tubers and plantains; pulses; nuts and seeds; dairy; meat, poultry and fish; eggs; dark green leafy vegetables; other vitamin A-rich fruits and vegetables; other vegetables; other fruits. But 35% of adults and 39% of five to 18-year-olds did not meet this recommendation.

However, not all respondents had poor diets. The food diaries revealed a wide range of foods available locally, including 18 types of dark green leafy vegetable and 22 types of fish. But individuals ate most of these rarely.

There was also little diversity within the food groups – for example, 67% of cereal consumption was rice, 61% of nuts and seeds was peanuts, 84% of legumes was soybeans, 61% of flesh meat was chicken, 80% of vitamin A-rich fruits was papaya and 98% of vitamin A-rich vegetables was carrots.

Ultra-processed food

Brazilian nutrition researcher Carlos Monteiro coined the term ultra-processed food for baked, fried, extruded, moulded, re-shaped, hydrogenated and hydrolysed substances that are derived from food. Although they can be fortified with micronutrients, they generally include preservatives, sweeteners, sensory enhancers, colourants, flavours and/or processing aids, contain little or no whole food and are considered to contribute to obesity.

The SD4ALL research in Indonesia showed that 51% of adults, 71% of children aged 5-18, and 73% of children under five consume ultra-processed foods at least three times a week. For sweet foods, these percentages were higher for adults (87%) and lower for children aged 5-18 (69%) and under-fives (54%).

Limitations of current nutrition survey methods

The study highlighted several limitations related to the usual methods and indicators for assessing diets. There is little published data on diets in Indonesia and most of this pertains to young children and women of reproductive age, so we know little about other demographic groups.

Further, dietary diversity scores (DDS) have only been validated for young children or women of reproductive age; there are (to my knowledge) no valid indicators for older children, men and older women.

The SD4ALL study used a DDS designed for women aged from 15-49 for everyone who was over five years old because no other validated indicators were available. Although there is another DDS for younger children, there were not enough children aged 6-24 months in the study to analyse their diets.

The DDS does not currently take into account the consumption of unhealthy and ultra-processed foods, making it difficult to compare surveys or account for the negative influence of such foods on nutrition adequacy. And because the score counts food groups rather than individual foods, it is difficult to understand the range of foods consumed.

Getting better data

There is an urgent need to develop and validate indicators and assessment methods that are ready to map, understand and address the triple burden of malnutrition in an integrated way.

Researchers need straightforward, ready-to use participatory dietary assessment methods that use negative counting for unhealthy foods, cover all age groups and genders, and gather information on within-food group diversity. This would make it easier to understand and compare the diets of all demographic groups and help us link diets with agro-biodiversity and sustainable food systems.

Finally, people are not used to recording their own dietary data, as most studies rely on trained enumerators, which can be both costly and time-consuming.

SD4ALL embraces a citizen agency perspective that challenges this expert-driven approach to research. Self-administered food diaries are a step forward, offering a method that allows people to collect their own dietary data, while saving some time and effort. But data analysis remains a skilled and laborious endeavour.

Combining food diaries with accessible electronic recording methods would ensure they are still participatory, while sophisticated software would enable semi-automated analysis, saving time and money encouraging more data collection.

This would help to build a fuller picture of diets in different contexts for use in policies and programmes to tackle the urgent problem of the triple burden of malnutrition.

About the author

Anne-Marie Mayer ( is a nutrition research consultant for IIED on the SD4ALL programme

Anne-Marie Mayer's picture