Secondary Logo

Journal Logo

Home-Based Treatment of Malnourished Malawian Children with Locally Produced or Imported Ready-to-Use Food

Sandige, Heidi*; Ndekha, MacDonald J.; Briend, André; Ashorn, Per§; Manary, Mark J.*

Journal of Pediatric Gastroenterology and Nutrition: August 2004 - Volume 39 - Issue 2 - p 141-146
Original Articles—Hepatology and Nutrition

Objectives: To determine the efficacy of home-based therapy with ready-to-use food (RTUF) in producing catch-up growth in malnourished children and to compare locally produced RTUF with imported RTUF for this purpose.

Methods: After a brief inpatient stabilization, 260 children with severe malnutrition were enrolled and systematically allocated to receive home therapy with either imported, commercially produced RTUF or locally produced RTUF. Each child received 730 kJ/kg/day and was followed up fortnightly. Children completed the study when they reached a weight-for-height Z score > −0.5 (WHZ), relapsed, died, or failed to achieve WHZ > −0.5 after 16 weeks. Analyses were stratified by human immunodeficiency virus (HIV) status.

Results: 78% of all children reached WHZ > −0.5, 95% of those with HIV-negative status and 59% of those with HIV-positive status. Eighty percent of those receiving locally produced RTUF and 75% of those receiving imported RTUF reached WHZ > −0.5. The difference between recovery rates was 5% (95% confidence interval [CI], −5–15%). The rate of weight gain was 0.4 g/kg/day (95% CI, −0.6, 1.4) greater among children receiving locally produced RTUF. The prevalence of diarrhea reported by mothers was 3.7% for locally produced RTUF and 4.3% for imported RTUF. After completion of home therapy and resumption of habitual diet for 6 months, 91% of all children maintained a normal WHZ.

Conclusions: Home-based therapy with RTUF was successful in affecting complete catch-up growth. In this study, locally produced and imported RTUF were similar in efficacy in treating of severe childhood malnutrition.

*Department of Pediatrics, Washington University School of Medicine, St. Louis, Missouri, U.S.A.; †College of Medicine, University of Malawi, Blantyre, Malawi; ‡Institut de Recherche pour le Développement, Paris, France; §Paediatric Research Centre, Tampere University Hospital, Tampere, Finland

Received October 3, 2003; accepted January 7, 2004.

Supported by gifts from Roger and Fran Koch, the Georgia Peanut Commission, and Frank and Mary Hellwig. Nutriset donated the imported RTUF for the study.

Address correspondence and reprint requests to Mark Manary, Department of Pediatrics, St. Louis Children’s Hospital, One Children’s Place, St. Louis, MO 63110, U.S.A. (e-mail:

Childhood malnutrition is common and severe in Malawi; 13% of all children younger than 5 years die as a result of malnutrition and tens of thousands seek care for this condition annually (1). Treatment is administered in nutritional rehabilitation units (NRUs), where children stay as inpatients for a few weeks. In NRUs, children initially receive medical care and feedings providing modest amounts of energy and protein. As their condition improves, their diet is advanced to provide more protein and energy in an effort to bring about catch-up growth. Therapeutic strategies used in these units are in accordance with standard World Health Organization recommendations (2). The outcomes from NRUs are disappointing. Only 25% of wasted children actually reach 80% of the weight-for-height standard, whereas 10% of children who are discharged as recovered go on to die, and 20% experience relapse (3–5). Severe childhood malnutrition in Malawi is exacerbated by the human immunodeficiency virus (HIV) epidemic of southern Africa. A significant fraction of malnourished children in Malawi is HIV infected, whereas others suffer because their caregivers have HIV.

Recently, home-based therapy with ready-to-use food (RTUF) has been shown to be efficacious in effecting complete recovery from childhood malnutrition in Malawi (6). The solid, peanut-butter based RTUF is an energy-dense lipid paste that resists bacterial contamination because of its low water content and does not require cooking before consumption (7). RTUF is available as a commercial product packed in oxygen-free foil sachets (Nutriset, Malaunay, France). RTUF has been used successfully to treat malnourished children in hospitals in Chad and Senegal and in the community in Ethiopia and Malawi (6,8–10).

To make this therapy more widely available in the developing world where childhood malnutrition is rampant, local production is needed. In this study, a local formulation of RTUF was devised and small-scale production implemented. This study compares the outcome of malnourished Malawian children treated with either locally produced or imported RTUF.

Back to Top | Article Outline



All children aged 1 to 5 years discharged from the NRU at Queen Elizabeth Central Hospital in Blantyre, Malawi, were eligible. While in hospital, before enrollment in this study, children received medical treatment for infections and metabolic complications and milk-based feedings providing 420 kJ/kg/day and 1.2 g protein/kg/day. Upon resolution of complications and return of appetite, children were eligible for the home-based therapy study. Children who had not been hospitalized with malnutrition were eligible if they were referred from the outpatient department for treatment of malnutrition with either edema or a weight-for-height Z score < −2 (WHZ). The mothers or caregivers of the participants gave their informed consent. The study was approved by the College of Medicine Research and Ethics Committee of the University of Malawi in Blantyre, Malawi, and the Human Studies Committee of Washington University in St. Louis, Missouri, U.S.A.

Back to Top | Article Outline

Diets and RTUF Production

Children received either imported or locally produced RTUF in a quantity that provided 733 kJ/kg/day (175 kcal/kg/day) and 5.3 g protein/kg/day with a complete compliment of minerals and vitamins (Table 1). Imported RTUF was the commercial product Plumpy-Nut (Nutriset). Locally produced RTUF was a mixture of 30% full-fat milk powder, 28% icing sugar, 15% cottonseed oil, 25% salt-free peanut butter, and 1.6% of a mineral and vitamin mixture (CMV, Nutriset). The quantity of RTUF provided a complete diet for catch-up growth, replacing the children’s habitual diet. Mothers were encouraged to feed the RTUF to the malnourished child in small quantities (a few spoonfuls) 7–10 times per day.



Ingredients were purchased locally, with the exception of the mineral and vitamin mixture, and stored in large plastic or metal drums. A battery-operated laboratory scale was used to measure quantities to the nearest 0.1 kg. A 40-L planetary bakery mixer (MacAdams SM 401; MacAdams Manufacturing Ltd., Capetown, South Africa) was used to prepare the RTUF. Oil and peanut butter were added directly into the mixing bowl and combined until homogenous using a mixing speed of 105 rpm. The z-shaped kneader blade was used to minimize the amount of air impregnated into the mixture. Sugar, milk powder, and the mineral and vitamin mixture were first hand mixed as dry powders in a dedicated plastic drum and then emptied into the electric mixing bowl. The RTUF was mixed at 105 rpm for 6 minutes, 210 rpm for 6 minutes, and 323 for 6 minutes. This mixing time was necessary to ensure homogeneity of the RTUF and to prevent separation during storage. The bakery mixer bowl and kneading blade, as well as all buckets, spatulas, scoops, and spoons, were washed thoroughly after each production day, but not during the production, to prevent the accidental addition of water to the product. RTUF was stored temporarily in large plastic drums holding 60 kg, then hand-packed into 275-g bottles with screw tops for distribution to mothers. Mothers returned the bottles, which were cleaned and reused.

The consistency and color of the locally made RTUF was indistinguishable from the imported version. The local RTUF was considerably sweeter in taste because of the use of sucrose, rather than dextromaltose.

An independent food technologist assessed the safety of locally produced RTUF (Peter Fellows, Valid International, and Oxford Brooks University: Results of Analysis of RTUF produced at Tambala Food Products, Malawi; March 2003). Bacterial cultures taken from samples of the locally produced RTUF before and after contamination with a dirty child’s hands revealed no coliforms, Salmonella species, S. aureus, or yeasts, and less than 1,000 colonies/g of mesophilic aerobes. Aflatoxin levels were below the most stringent safety standards (< 2 ppb).

Back to Top | Article Outline

Experimental Design

Upon patient enrollment, information about the history of illness, family demographics, and socioeconomic status was acquired from an interview with the child’s caregiver and the inpatient hospital chart. Children were given a test feeding of the RTUF to screen for food allergy and ensure acceptability. Each child’s length, weight, and midupper arm circumference (MUAC) were measured. After appropriate counseling, HIV status was determined by testing sera obtained from the child using an enzyme-linked immunoabsorbent assay (Vironostika HIV, Organon Teknika, Durham, NC, U.S.A.), equivocal and positive tests were confirmed using a second test. Each child was assigned a target graduation weight corresponding to a WHZ > −0.5, based on the child’s enrollment stature. The value of WHZ > −0.5 was chosen because it corresponds to the standard designated as full recovery by the Ministry of Health, government of Malawi.

Children were assigned to one of the RTUF groups by systematic allocation according to order of entry into the project, with even numbered entries receiving locally produced RTUF and odd numbered entries receiving imported RTUF. Because the packaging of the food was not identical, the trial was not blinded. A two-week supply of RTUF was distributed for each child. The imported RTUF was given in 92-g disposable foil sachets, whereas the locally produced RTUF was given in 275-g clear plastic jars with screw lids. A sample size of 125 children in each dietary group was chosen so we could determine if the recovery rate for the locally produced RTUF group was at least 90% of that in the imported RTUF group, assuming that the recovery in those receiving imported RTUF would be 80%.

Mothers and children returned to the clinic for reassessment every 2 weeks. The child’s weight was measured fortnightly, whereas the length and MUAC measurements were done every 4 weeks. At each 2-week visit, the caregivers was asked the number of days of fever, cough, and diarrhea in the previous fortnight. An additional 2-week supply of RTUF was distributed at each visit. Mothers who failed to return for any follow-up visits were considered to be refusing to participate. Children were discharged from the study when they reached their target graduation weight, experienced clinical relapse (recurrence of edema or systemic infection) requiring readmission to the NRU, failed to reach WHZ > −0.5 after 16 weeks, or died. Children who reached graduation weight or completed 16 weeks of therapy were requested to return for a follow-up visit after 6 months to assess their nutritional status.

The primary outcome for the study was successful recovery, children reaching WHZ > −0.5 by 14 weeks of therapy. Dropouts were considered to be failures. Secondary outcomes were rate of weight gain, statural growth, growth in MUAC, anthropometric status 6 months after discharge from home-based therapy, and the prevalence of fever, cough, and diarrhea during the first 2 weeks of therapy.

Back to Top | Article Outline

Data Analyses

Anthropometric indices were calculated using Epi 2002 (version 1.1.2, Centers for Disease Control, Atlanta, GA, U.S.A.), which uses the World Health Organization’s NCHS reference population. Weight gain, statural growth, and the growth in MUAC were determined by calculating the change per day during the first 4 weeks of the study. Four weeks was chosen because many children required only 4 weeks of therapy, and the greatest rates of growth are likely to be in this initial time interval. The differences in weight gain, statural growth, and the growth in MUAC between the two dietary groups were determined with 95% confidence intervals [Cis]. Because children with HIV are known to have a significantly worse outcome and inclusions of even small numbers of such children might affect the dietary group outcome parameters, comparisons made between the locally produced and imported RTUF were made using all children and the children without HIV. Comparison of continuous parameters was made using Student t test, and between dichotomous parameters using χ2 analyses (InStat version 3.05, GraphPad Software Inc, San Diego, CA, U.S.A.). Multiple linear regression was used to model the rate of weight gain with respect to fever, cough, and diarrhea. A probability of < 0.05 was considered to be statistically significant.

Back to Top | Article Outline


Two hundred sixty children were recruited between April and August 2002; 227 were enrolled from the in-patient unit and 33 as outpatients (Fig. 1). There were no group differences in the baseline demographic or nutritional characteristics between the children that received locally produced or imported RTUF (Table 2). There were no differences in primary or secondary outcome parameters when boys were compared with girls, so outcomes were not stratified by sex. Thirty percent of the children enrolled had positive HIV status.



FIG. 1.

FIG. 1.

Considering all study children, 202 (78%) reached WHZ > −0.5; 108 (80%) in the locally produced RTUF group and 94 (75%) in the imported RTUF group. Twenty-eight (11%) children died, experienced relapse, or did not reach target weight, and 13 (5%) dropped out. The difference in recovery rates for children receiving locally produced RTUF compared with imported RTUF was 5% (95% CI, −5–15%). The rates of growth were similar between the two groups (Table 3).



Considering only children with negative HIV status, there were no differences in demographic or anthropometric characteristics between those who received imported RTUF and those who received locally produced RTUF. Successful outcome for children with negative HIV status was 95% for both the imported and locally produced RTUF dietary groups (Fig. 1), the 95% CI for the difference between these recovery rates was −5%, 5%. Children with negative HIV status reached WHZ > −0.5 in a median of 43 days for locally produced RTUF (25th percentile–75th percentile, 29–74 days) and 43 days for imported RTUF (25th percentile–75th percentile, 29–71 days). The mean (95% CI) graduation WHZ was −0.3 (−0.4 to −0.2) for locally produced food and −0.2 (−0.2 to −0.0) for imported food. Rates of weight, length, and MUAC gain were similar for children with HIV negative status receiving the two diets (Table 4).



Diarrhea was reported on 198 of 5,371 (3.7%) days from mothers receiving locally produced RTUF, and on 180 of 4,223 days (4.3%) from mothers receiving imported RTUF.

Fever was reported on 487 of 9,594 (5.1%) days by caregivers, and 54 of 260 (21%) children were treated for malaria during the course of their home therapy. Linear regression modeling of rate of weight gain during the first 2 weeks with the number of days of fever, cough, and diarrhea reported by mothers showed that only fever is a significant predictor of weight gain (P = 0.003), and that fever accounts for 9% of the variation in weight gain.

Seventy-eight children with HIV participated in the study. These children had a lower mean WHZ than did children without HIV (−2.4 ± 0.8 v −2.0 ± 0.8, P < 0.001), but no other significant demographic or anthropometric differences between the children with HIV and those without were seen. Forty-six children with HIV reached WHZ > −0.5, 9 dropped out of the study, 9 experienced relapse and were readmitted to the hospital, 7 died, and 7 simply did not gain enough weight to achieve WHZ > −0.5. Having HIV represented a relative risk of 8.5 (95% CI, 4.3–17) for failing to reach WHZ >−0.5. The rate of weight gain for children with HIV was less than that for children without HIV (3.6 ± 4.7 g/kg/day v 5.6 ± 4.0 g/kg/d, P < 0.001). For the children with HIV who did recover, their average time to reach WHZ > −0.5 was significantly longer than that of the children without HIV (70 days; 95% CI, 59–81). Children with HIV reported cough on 211 of 1,050 days in the first 2 weeks (20%), substantially more than children without HIV (273/2,548; 11%) but reported no more fever or diarrhea.

One hundred sixty of the children who reached WHZ > −0.5 (73%) returned for follow-up after an average of 5.8 months, and 14 (9%) were wasted, indicating they had experienced relapse. The mean (95% CI) WHZ on 6-month follow-up was −0.6 (−0.7 to −0.5) compared with −0.2 (−0.3 to −0.1) on completion of home therapy. No risk factors for relapse after 6 months were identified, including HIV status, presence of mother or father in the home, or season during which child returned for follow-up.

Back to Top | Article Outline


Home-based therapy with RTUF was successful in effecting catch-up growth during recovery from malnutrition, providing a successful alternative to supervised, inpatient, high-energy feedings. Locally produced and imported RTUF was similar in efficacy in the treatment of severe childhood malnutrition in the current study. The lower limit of the 95% CI indicates that locally produced RTUF is likely to be at least 96% as efficacious as imported RTUF. Children receiving locally produced and imported RTUF achieved similar rates of anthropometric growth. The modest differences between the two groups are not of clinical significance.

The limitations of this study design were that subjects were systematically assigned to one diet or the other, rather than randomly assigned, and participants and study staff were not blinded to the dietary assignment. The locally produced RTUF was hand packed in recycled plastic containers that did not seal tightly, and the imported RTUF was shipped from France in packaging that did not readily leak. It was not practical to repackage the imported RTUF by hand in Malawi for the purposes of blinding the participants and staff. Because the primary outcome measure of the study was determined by nude body weight determined on a calibrated electronic scale rather than by more subjective assessment, the investigators do not believe this introduced bias. Systematic allocation is likely to be as reliable as random allocation in a therapeutic trial, provided the investigator enrolling the subjects does not play a role in determining the allocation of subjects (11). This was achieved in this study. No peanut allergies were observed in the study population, which might not be the case in other settings.

It seems likely that the high recovery rate in this study is primarily a consequence of the use of abundant amounts of the energy- and protein-dense RTUF. In addition, the home setting of therapy may have contributed contribute to its success. Standard inpatient care for nutritional rehabilitation usually cohorts 15 to 100 children in a single facility, and these children are at increased risk of infection from the communicable pathogens (12). There is evidence that antibiotic-resistant bacteria are readily transmitted in conditions of close contact such as these nutrition rehabilitation centers (13). These risks are reduced when RTUF is used at home.

The results of this study are in agreement with the results of other studies using RTUF. The 95% recovery for children without HIV receiving RTUF was similar to that found in Blantyre, Malawi, in 2001 (6). Collins and Sadler (10) documented recovery rates of 85% using home-based therapy in Ethiopia among a population that was more severely malnourished but less likely to have HIV.

Locally produced RTUF carries a larger solute load than does imported RTUF because it is made with the disaccharide sucrose rather than the polysaccharide dextromaltose. The greater osmotic load might be associated with more malabsorption and diarrhea, as it has been in other clinical scenarios (14,15). However, diarrhea was not a clinical problem in this study. Mothers did not report more diarrhea when receiving the locally produced RTUF, and growth rates were similar. This may be because the lipid paste dissolves slowly in the gut, releasing the osmotic load over a period of a few hours, as has been observed with RTUF in vitro (16).

Coincident infection during treatment for malnutrition may interfere with recovery (17). Fever and treatment for malaria were common among study participants. In the current study, fever was associated with lower rates of weight gain, and one might speculate that prevention of malaria, the most common febrile illness in Malawi, would improve recovery times. Perhaps prophylaxis for malaria would be a useful adjunct to home therapy?

It is understandable that the malnourished children with HIV had more severe wasting and poorer outcomes than did children without HIV because they, like most children with HIV in Malawi, did not receive antiretroviral chemotherapy. It is remarkable, however, that so many of these children achieved full catch-up growth, even if the period of nutritional therapy was prolonged. This observation demonstrates that outpatient, therapeutic feeding of malnourished children with HIV can indeed improve nutritional status. If antiretroviral chemotherapy becomes available to more children with HIV in sub-Saharan Africa, outcomes may be improved by offering RTUF in conjunction with chemotherapy. This is an important topic for further research.

Once children completed home therapy, they presumably resumed a diet of maize and beans at home. After 6 months, only 9% of children had experienced relapse. This suggests that normal growth can be maintained on the habitual Malawian diet once catch-up growth has been achieved.

The cost of locally produced RTUF was about $2 (US)/kg. The imported RTUF cost $3/kg. Duty and shipping added another $2/kg. Use of recycled packaging materials and donation of oil and sugar by the World Food Program reduced the cost of locally produced RTUF to $1.25/kg. In this study, a child used on average about 11 kg of RTUF at a cost of $14 per child.

In sub-Saharan Africa, recovery rates from severe childhood malnutrition in the poorest nations remain unacceptably low. This study of home-based therapy with RTUF suggests that substantially higher recovery rates can be attained while lessening the burden on poorly resourced inpatient facilities. Such an advance could be an important step forward in the care of acute childhood malnutrition. Local production of RTUF at a reduced cost makes it more widely available. Additional feasibility studies with home-based therapy using RTUF are needed to determine whether it can offer a superior alternative to standard care in the variety of settings (rural health centers, district hospitals, and mission hospitals) in which malnourished children are often treated.

Back to Top | Article Outline


1. Pelletier DL, Low JW, Johnson FC, et al. Child anthropometry and mortality in Malawi: testing for the effect modification by age and length of follow-up and confounding socioeconomic factors. J Nutr 1994;12(Suppl 10):2082S–105S.
2. World Health Organization (WHO). Management of Severe Malnutrition: A Manual for Physicians and Other Senior Health Workers. Geneva: WHO; 1999.
3. Perra A, de L Costello AM. Efficacy of outreach nutrition rehabilitation centers in reducing mortality and improving nutritional outcome of severely malnourished children in Guinea Bissau. Eur J Clin Nutr 1995;49:353–9.
4. Pécoul B, Soutif C, Hounkpevi M, et al. Efficacy of a therapeutic feeding center evaluated during hospitalization and follow-up period, Tahoua, Niger 1987–1988. Ann Trop Paeds 1992;12:47–54.
5. Brewster DR, Manary MJ, Graham SM. Case management of kwashiorkor: an intervention project at seven nutrition rehabilitation centers in Malawi. Eur J Clin Nutr 1997;51:139–47.
6. Manary MJ, Ndekha MJ, Ashorn P, et al. Home-based therapy for severe malnutrition with ready-to-use food. Arch Dis Child 2004;89:557–61.
7. Briend A. Highly nutrient-dense spreads: a new approach to delivering multiple micronutrients to high-risk groups. Br J Nutr 2001;85(Suppl 2):S175–9.
8. Briend A, Lacsala R, Kamnadji J, et al. Ready-to-use therapeutic food for treatment of marasmus. Lancet 1999;353:1767–8.
9. Diop EHI, Dossou NI, Ndour MM, Briend A, Wade S. Comparison of the efficacy of a solid ready to use food and a liquid milk-based diet for the rehabilitation of severely malnourished children: a randomized trial. Am J Clin Nutr 2003;78:302–7.
10. Collins S, Sadler K. Outpatient care for severely malnourished children in emergency relief programmes: a retrospective cohort study. Lance t 2002;360:1824–30.
11. Armitage P, Berry G. Statistical Methods in Medical Research. 3rd ed. Oxford: Blackwell Science; 1994.
12. Raymond J, Le Thomas I, Moulin F, et al. Sequential colonization by Streptococcus pneumoniae of healthy children living in an orphanage. J Infect Dis 2000;181:1983–8.
13. Reichler MR, Allphin AA, Breiman RF, et al. The spread of multiply resistant Streptococcus pneumoniae at a day care center in Ohio. J Infect Dis 1992;166:1346–53.
14. Kukuruzovic RH, Brewster DR. Milk formulas in acute gastroen-teritis and malnutrition: a randomized trial. J Paediatr Child Health 2002;38:571–7.
15. Hahn S, Kim Y, Garner P. Reduced osmolarity oral rehydration solution for treating dehydration due to diarrhea in children: systematic review. Br Med J 2001;323:81–5.
16. Briend A. Possible use of spreads as a FOODlet for improving the diets of infants and young children. Food Nutr Bull 2002;23: 239–43.
17. Scrimshaw NS, SanGiovanni J. Synergism of nutrition, infection and immunity: an overview. Am J Clin Nutr 1997;66:465S–77S.

Malnutrition; Ready-to-use food; Treatment

© 2004 Lippincott Williams & Wilkins, Inc.