In India, khichri (rice and green gram gruel) is often used for the dietary management of moderately malnourished children as a practical and culturally acceptable food for young children. However, khichri is required to be fed several times and this is difficult for caregivers because the recipe requires cooking and reheating before it is offered each time. Besides, green gram is costly for poor communities, where most malnourished children reside. Also, khichri is a good medium for bacterial growth.

An alternative solid ready-to-use therapeutic food (RUTF) has been developed(2-4). RUTF has a 5 times greater energy density than F100 but a similar nutrient to energy ratio. Part of the dried skim milk in F100 is replaced with a mixture of peanut butter and lacto serum, and oil by vegetable fat mixture, including essential fatty acids with adequate viscosity and melting point to produce the RUTF(2,3). RUTF has a high concentration of micronutrients and vitamins whose unpleasant taste is well masked(2-4). The child can consume RUTF without adding water, unlike the F100. Thus, the risk of bacterial growth and proliferation is reduced. These characteristics make RUTF a safer intervention in communities. Besides, RUTF does not require cooking.

In Africa, RUTF has shown excellent acceptability in cross over or comparative trials(2,3). The acceptability of this new high energy dense food is important to assess in India before its practical use can be addressed in the country. We conducted a preliminary study to test the acceptability and energy intake of RUTF as compared to commonly used cereal legume gruel (khichri). We anticipate that the study findings would provide a basis for planning large scale evaluations of RUTF.

The setting was the urban low to middle socioeconomic neighborhoods of Tigri in South Delhi. The population characteristics have been described elsewhere(5,6). Childhood malnutrition is common in this setting(7,8). The study was approved by the ethics committee of the Society for Applied Studies.

Sample size: Two studies comparing the intake from RUTF and F100 showed that the difference in intake to be expected between the two feeds was about 1 SD(2,4). To obtain a difference significant at the 0.05 percent level with 95% power, a minimum of 26 children were needed. To allow for dropouts and missing data, we planned enrolment of 30 children. Forty children were identified to account for dropouts during the ‘pretesting’ phase, due to illness and refusals.

The acceptability trial involved feeding of two meals a day, one of each type (RUTF and khichri) for two days. Prior to initiating the trial, the 36 potentially eligible children underwent a ‘pretesting’ phase. During this pretesting phase, khichri and RUTF were given for two days to the caregiver to be fed at 2 meals each (one meal of each on both days) to familiarize the mother and child with the diet.

Of the 36 children who participated in the pre-testing, 31 children were enrolled in the main trial (Fig.1). The trial children were brought to the study clinic set up in Tigri on two consecutive days. The 31 children when assessed using the Integrated Management of Neonatal and Childhood Illness protocol(11) did not reveal presence of local or systemic bacterial infection. After obtaining consent again, they were enrolled into the study.

Randomization: The study had a cross over design. The randomization list for the sequence in which meals would be fed to a particular child on the two days was generated offsite by a statistician not otherwise involved with the study. Using a simple randomization scheme, half the participants were assigned to receive khichri and the other half RUTF as their first meal on the first day. The other food i.e. khichri or RUTF was given as the second meal on the first day. The food received by a child at the second meal on the first day was administered as the first meal on the second day (i.e. if khichri was the second meal on day 1, khichri was given as first meal on the second day of the observation). Each enrolled child received four meals; one meal of khichri and one of RUTF on each of the two days.

Intervention: The energy and nutrient content of the two foods i.e. RUTF and khichri is given in Table I. Khichri was prepared using rice (100 g), green gram (60 g), groundnut oil (20 mL), spinach (100 g) and water (640 mL). 760 g of khichri so prepared yielded 760 kcal i.e. 1 kcal/g of cooked khichri (Table I). The energy content for khichri was calculated using national food composition tables(12).

TABLE I



Nutritive Value of RUTF and Khichri per 100 Grams

Children were offered the food that they were scheduled at two meals at 10 am and 2 pm on each of the two days. The meal time lasted an hour. Caregivers were instructed to feed their children and to encourage them to eat as much as they could of the pre-weighed food in the one hour period. If children ate the preweighed amount, the initial amount was reoffered. Measured volumes of plain water was also given in between the meals. Caregivers were asked not to feed the child any other foods during the interval between the two feeds. Once the meal time was over, the foods were weighed again and the amount of food consumed in grams was estimated. The leftover water was also measured to estimate the volume of water consumed.

Prior to commencing meals on the second day, each child was again examined for local and systemic infections and none of the children had either of these.

Observation of feeds and interviews with caregivers: Observations were conducted on each child on the second day for both the meals. An observation checklist was used to observe the child for a minute after every 15 minutes i.e. a total of 4 observations per child were obtained for khichri and RUTF. Caregivers of all enrolled children were also interviewed by nutritionists using a semi-structured questionnaire to enquire about their perceptions about the RUTF.

Analysis: Acceptability was categorized as ‘accepted eagerly’ if they ate food readily, did not make a fuss, spit out, vomit or cry during the observed meal. Children were categorized as ‘accepted but not eagerly’ if they ate the offered food but either made fuss, spit out, vomited or cried during the observed meal. The third category was children who did not consume the offered food at all.

A summary variable was created from the 4 observations for each food observed i.e. RUTF and khichri. The higher category of acceptability was assigned if the defined behavior was observed in at least two observations. The categories of acceptability across RUTF and khichri were compared by Pearson chi-square test.

Energy intakes were estimated from the amount (in grams) of each food consumed by a child at each meal. The difference in median energy intakes for both foods on the two days was estimated between the two groups and compared by a Wilcoxon signed rank test.

Among the 31 trial subjects, 16 (51.6%) were male. Eight, 19 and 4 children were aged 6 to 12, 13 to 24 and 25 to 36 months, respectively. Table II depicts the acceptability of RUTF and khichri. Table III provides the amount of RUTF and khichri consumed and the energy intakes in the two groups.

TABLE II



Acceptability of RUTF and Khichri

During the interviews with caregivers, 22 (71%) of the 31 caregivers reported that their children had liked the RUTF. The common reasons stated for liking RUTF were that children ate most of the portion served to them (22/22), child asked for more (3/22), did not make a fuss about eating (2/22), smacked his lips (1/22), never used to eat any food but ate RUTF (1/22) and that the child started passing a single stool per day (1/22). Nine (29%) caregivers reported that RUTF was not liked by their children; the most common reasons were children did not eat enough of the food (8/9), grimaced while being fed (1/9), spat out the food (1/9) and refused to eat it (3/9).

Twenty eight of the 31 caregivers thought the consistency of RUTF was appropriate for children, the appearance (packet and color) was good (10/28), the food looked like chocolate (3/28), their child ate it (1/28) and it gave more energy than home foods (2/28). Three caregivers felt the consistency was thick and therefore inappropriate for children. However, two of them had reported that their children had liked the food when asked so. Caregivers were also asked whether they would feed RUTF to their children if it were available to them in the future. Of the 31 interviewed, 29 said that they would use it; the reasons cited were that it gave energy to their children (22/29), their children liked it (11/29) and it increased the child’s appetite (3/29). Less common reasons were that it will protect the child from disease (1/29), the child’s weight will increase (1/29), it is like chocolate (1/29), it will improve child’s health (1/29) and that the government always gives products that are beneficial for children (1/29). Only 2 (6.4%) caregivers said they would not give it because their children did not like it (1/2) or eat it (1/2).

In this preliminary trial, while both khichri and RUTF were accepted well by young children, acceptability of khichri was somewhat better. Khichri is a culturally acceptable and palatable meal and widely consumed by families and children in this setting. Under conditions of unlimited access at each of the test meals, the intake of RUTF was lower but the energy intake from it was significantly higher. The energy density for RUTF was five fold higher and this would explain the lower consumption as well as the substantially higher energy intake.

The findings of this study are consistent with previous reports; improved energy intakes were reported in severely malnourished children in two other studies that compared RUTF with the F100 diet(2,4). In a therapeutic feeding centre in, Chad, in severely malnourished children, Briend, et al.(2) observed a two fold higher mean energy intake from RUTF as compared to F100 diet. In another study from Senegal, also in severely malnourished children, the mean daily energy intake was 70% higher in the RUTF group(4).

An important question to address is whether a product like RUTF has a potential use under special circumstances and whether a more decisive safety or efficacy evaluation is merited. RUTF has been proposed as potentially useful in situations of war and disaster, in home treatment of severe malnutrition when hospitalization is not feasible or in centres without a kitchen, or as an add-on in health facilities where diets can only be offered 2 or 3 times a day. This would require assessment of its use in less supervised circumstances. Further RUTF has been less well studied in infants and in malnourished children with infections. Clearly RUTF does not have a role in settings where standard treatment of malnutrition is feasible. As it avoids problems of quality control and microbial contamination, RUTF may be considered for evaluation in difficult circumstances where standard protocols are not feasible to implement.

We are grateful to Dr Andre Briend for his constructive inputs and for reviewing the manuscript. We thank Ms Baljeet Kaur for help in statistical analysis.

Contributors: BD, TR, NB and MKB were involved in concept and protocol design, oversight of all phases of the trial, and interpretation, preparation and finalization of manuscript. SM, ST and FR were involved in the review of literature, analysis and interpretation of manuscript. BD and FR collected the data under supervision of TR and NB.

Funding: World Health Organization, Geneva.

Competing Interests: None stated.