Scaling up the treatment of acute childhood malnutrition in Niger
Milton Tectonidis
By Isabelle Defourny, Emmanuel Drouhin, Mego Terzian, Mercedes Tatay, Johanne Sekkenes and Milton Tectonidis
Emmanuel Drouhin is the Niger Desk Officer, Isabelle Defourney the deputy Desk Officer for Niger and Mego Terzian is the Emergency Desk Officer, all based in Paris with MSF France. Johanne Sekkenes is the Head of Mission of MSF Niger and Milton Tectonidis the Nutrition Consultant in the Medical Department of MSF France.
This article presents a strong case from Niger that managing severe malnutrition on a large scale through outpatient treatment is a real possibility.
For a short period during the summer of 2005, Niger - a country whose people are amongst the poorest in the world - had the dubious privilege of hitting prime time on international television, as officials and advisors attempted to explain why so many children were starving and why nothing had been done to help them. Despite a prevalence of wasting constantly hovering around or above 15% and the third highest under-5 mortality rate in the world (259 per 1000 live births1), health authorities and their bilateral, multilateral and international partners present in the country were almost exclusively focused on development programmes addressing 'underlying causes'. Food security early warning systems in Niger were essentially limited to rainfall data and agricultural production estimates. The absence of any sustained attempt to monitor, let alone treat, acute malnutrition, combined with a fatalistic complacency towards high 'structural' rates of wasting, led to unacceptable delays and errors in the response to the epidemic of malnutrition that affected Niger in 2005.
Dakoro
Stabilisation
Centre
The development of solid equivalents of therapeutic milks in the last five years is inspiring a major change in the treatment of acute malnutrition2,3,4. These nutrient-dense Ready to Use Therapeutic Foods (RUTF) have led to the development of an outpatient approach to treatment based on simplified medical and nutritional protocols for all but the most complicated cases. In rural settings, multiplication and decentralisation of programme entry points leads to rapid and thorough diagnosis of wasted cases in the community. Effects on programme capacity and coverage are dramatic and costs per patient treated reduced. Programme indicators such as cure, mortality and default rates have, to the surprise of many, consistently outperformed those obtained with the classic approach advocating predominant or exclusive inpatient management5,6.
Applying this new strategy to the crisis in Niger in 2005, Médecins Sans Frontières (MSF) expanded operations well beyond the programme established in Maradi region in 2001, going on to admit over 63,000 severely malnourished children in therapeutic feeding programmes during the year, by far the largest nutritional intervention in the organisation's history. Even outside periods of acute crisis, home consumption of RUTF has the, as yet, unexploited potential of making the effective management of acute malnutrition far more accessible in resource poor countries characterised by high malnutrition rates and numbers of wasted.
MSF in Niger and the response to the 2005 crisis
The outpatient treatment programme for severe malnutrition in Maradi was opened in July 2001 after a measles epidemic swept through the region. The continuing large numbers admitted (4,443 in five months) exceeded expectations and led MSF to maintain the project. The number of admissions rose every year reaching 9,524 in 2004, by which time direct admissions into outpatient care had increased to over 50% of all admissions. Reluctance amongst clinicians to discharge stabilised patients quickly from the inpatient centre had been overcome, cure rates had reached 83.5% and overall mortality rates had fallen to 6%.
In the first few weeks of 2005, the MSF team in Maradi noticed a distinct change in the pattern of admissions compared to previous years. By early February 2005, weekly admission rates were triple what they had been in 2004, without any additional deployment on the part of the existing programme. By mid March 2005, MSF launched evaluations in Maradi and Tahoua that confirmed high rates of acute malnutrition - months earlier than the usual hunger gap - and began to expand and extend activities in both regions. Weekly family food rations and a large discharge ration were added to the therapeutic package of RUTF and medical care offered to all admitted children. By early July 2005, 45 international staff and 660 national staff were running 27 outpatient centres (OC) and five stabilisation centres (SC) for severely wasted children. From August 2005 onwards, activities in Tahoua and the northern parts of Maradi were handed over to other international agencies, and subsequent MSF efforts were concentrated on the most affected areas of Maradi and Zinder, where the majority of admissions for severe malnutrition were being recorded. In late September 2005, the feeding programme in Maradi admitted 2,043 new patients and was monitoring 8,727 malnourished patients a week, 934 of them in the four inpatient facilities and the rest in outpatient care. In the three southernmost departments of Maradi, two decentralised paediatric units were also opened and medicines were provided to a dozen government health centres, to facilitate free outpatient and referral health care for all under 5 children. Between July and October 2005, MSF distributed over 4,000 tons (129,487 rations) of blended, enriched flour and cooking oil to families of 53,031 at risk or moderately malnourished children aged less than 5 years old. In Zinder region, MSF would go on to admit over 21,000 severely malnourished children in the last five months of the year, using the same simplified outpatient system as in Maradi.
Madarounfa
Outpatient
Centre
Programme design
Children between 65 and 110 cm of height were admitted on the basis of mid-upper arm circumference (MUAC) < 110 mm, weight-forheight (W/H) < -3 Z scores (ZS) of the NCHS standard or the presence of bilateral oedema. Children in the same height range with a weight for height between -2 and -3 ZS accompanied by severe pathology were also admitted. Children between 60 to 65 cm height and above 6 months of age were admitted on the basis of weight-for-height or oedema criteria only.
All admissions received systematic amoxicillin for 5 days, single doses of albendazole, folic acid and vitamin A according to weight, and measles immunisation. Those identified as positive for falciparum malaria by rapid blood test received artemesin-based combination therapy (ACT). Specific treatments were given for respiratory, gastrointestinal or cutaneous diseases according to standardised protocols. All complicated cases presenting with anorexia, severe pathology or moderate to severe bilateral oedema were immediately referred to a SC. Uncomplicated cases were consulted and weighed weekly and sent home with further specific treatments, 1000 kcal/day of a RUTF (two 92g packages of Plumpy'nut daily) and as of March 2005, a family protection ration of 5 kg of blended, enriched flour (Unimix) and 1 litre of cooking oil. Returning outpatients with anorexia, severe pathology, appearance of moderate to severe oedema, abrupt or progressive weight loss or failure to gain weight after 4 weeks in the programme, were referred to a SC. Children reaching exit criteria (W/H > -2 ZS for 2 consecutive weeks, mid upper arm circumference > 110 mm, no oedema and absence of ongoing infection) were sent home with a discharge ration of 50 kg of millet, 25 kg of cow peas and 10 litres of cooking oil. All six SCs had planned capacities of up to 250 children, with actual patient counts reaching over 300/day in some centres during the peak part of the year when 100 patients/day were being admitted. Along with standard feeding centre facilities, SCs had well staffed intensive care units of up to 50 beds to handle critical cases referred from outpatient care. These units were equipped with oxygen, a mini blood bank, broad spectrum parenteral and oral antibiotics and the ability to monitor a large number of children receiving tube feeding (F- 100 milk) or rehydration (Resomal solution). Stabilised patients were referred back to outpatient care unless they had already reached discharge criteria (W/H > -2 ZS for 3 consecutive days, mid upper arm circumference > 110 mm, no oedema and absence of ongoing infection) in which case they were discharged directly home with a one-month discharge ration.
Results
MSF admitted 43,529 malnourished children into its programmes in Maradi and Tahoua region between January 1st and December 31st 2005 (see figure 1). Almost 20,000 were admitted in a ten week period (weeks 30 to 40) in August and September 2005. As in previous years, 95% of admitted children were under 85 cm in height. Moderately malnourished children between -2 and -3 ZS with severe pathology represented 6.4% of admissions, and kwashiorkor only 2.8%. These results are typical for countries in the Sahel, with chronic high rates of wasting striking weaning age children less than 24 months old, reaching dramatic levels during hunger gap periods and epidemic proportions during bad years.
Detailed results are given here for all 37,483 patients treated and discharged from programmes in Maradi region, including the SCs in Maradi, Aguié, Tiberi and Dakoro and 18 associated OCs in the six rural departments (Madarounfa, Guidan Roumdji, Aguié, Tessaoua, Mayahi and Dakoro) opened for at least part of the year (see table 1). Results from 2005 are compared with results for the same region between the years 2001-2004 when there was one SC in Maradi and seven outpatient centres in the three heavily populated, agricultural departments of southern Maradi, which also accounted for 75% of the admissions in 2005 (see figure 2). Geographical expansion was not the determining factor explaining the dramatic increase in the number of admissions to MSF programmes in Maradi over previous years.
Out of the 39,353 admissions in Maradi region in 2005, only 0.8% was readmissions (relapse within 3 months of previous discharge). Atotal of 25,688 (65.3%) were admitted directly into outpatient care and of these, 1,996 (7.8%) were subsequently referred to a SC at some point during the course of treatment. OCs accounted for 31,246 (83.4%) of 37,483 total discharges (see table 2). Overall programme cure rate was 91.4%, mortality rate 3.2%, default rate 4.7% and 0.7% were transferred out of MSF programmes.
These results are superior to those obtained in 2004 when cure rates were 83.5%, default rates 10.3% and mortality rates 6.0%. Readmission rates also dropped from 1.6% in 2004 to 0.8% in 2005. The better results in 2005, despite much larger patient numbers, was probably due to early diagnosis and greater participation in the programme, encouraged by the introduction of protection and discharge rations in March 2005 amidst widespread household food insecurity. Although children discharged from outpatient care have longer durations of stay and lower daily weight gains than those treated as inpatients, they still spend less than a month (29.1 days) in the programme and their daily weight gain (10.5 g/kg/d) is well above recognised benchmarks. It could be argued that spreading the total weight gain over a longer period of time within the patient's usual family environment may be a factor explaining the low rate of relapse (0.8%) following cure.
Discussion
Niger faced an epidemic of acute malnutrition in 2005 primarily affecting young children less than 24 months of age in the southern areas of Maradi and Zinder provinces during the hunger gap period between June and October. This epidemic occurred on top of chronically high rates of wasting and mortality amongst young children.
In Niger, most rural families are highly dependent on market food purchases for a large part of their dietary intake . In 2005, millet prices in Maradi reached up to 28,000 CFA for a 100 kg bag in July compared to 8,000 CFA received by farmers at the time of the previous harvest in 2004. There is a striking correlation between the market price of millet in Maradi in 2005 and the number of admissions of wasted children into MSF programmes five weeks later (see figure 3).
MSF's experience in Niger has important implications for medical practice in countries with high endemic rates of childhood malnutrition and large numbers of children requiring treatment
- The results of MSF's outpatient programme described here, combined with results from an increasing number of similar pro grammes elsewhere, suggests that the vast majority of severely wasted children can be cured with simplified weekly surveillance and ready to use therapeutic foods for home consumption.
- In Niger since 2001, tens of thousands of mothers and fathers have been making sustained efforts to bring their wasted children to MSF feeding programmes where, in over whelming numbers, they persist in attending each week until complete cure of their child has been achieved. All reason and evidence suggests that faced with an effective, accessible and easy to use remedy, parents are willing to go to great lengths to save the lives of their children.
- The new paradigm of outpatient care using nutrient dense therapeutic foods makes nonsense of the argument that the treatment of malnutrition is a desirable but impossible objective for resource poor countries. Therapeutic foods can be produced and made widely available where the need is greatest. Acute childhood malnutrition, highly prevalent and frequently lethal, is also eminently treatable.
For further information, contact Milton Tectonidis, email: milton.tectonidis@paris.msf.org or Isabelle Defourny, email: isabelle.defourny@paris.msf.org
| Table 1 Programme indicators for Maradi programme, 2005 |
| |
Maradi |
Aguié |
Tibiri |
Dakoro |
Total Maradi region |
| Weeks SC open |
Weeks 01-52 |
Weeks 25-48 |
Weeks 36-52 |
Weeks 16-48 |
|
| Admissions OC (n) |
7,926 |
1,982 |
2,256 |
1,501 |
13,665 |
| Admissions SC (n) |
13,612 |
4,277 |
4,239 |
3,560 |
25,688 |
| Total admissions (n) |
21,538 |
6,259 |
6,495 |
5,061 |
39,353 |
| Proportion direct admissions to OC (%) |
63.2% |
68.3% |
65.3% |
70.3% |
65.3% |
| |
|
|
|
|
|
Cured n
(%) |
15,968 (90.8%) |
6,340 (90.7%) |
7,139 (91.7%) |
4,800 (93.6%) |
34,247 (91.4%) |
Died n
(%) |
707
(4%) |
190 (2.7%) |
140 (1.8%) |
181 (3.5%) |
1,218 (3.2%) |
Defaulted n
(%) |
875
(5%) |
279
(4%) |
494 (6.3%) |
114 (2.2%) |
1,762 (4.7%) |
Transferred n
(%) |
37
(0.2%) |
180 (2.6%) |
8
(0.1%) |
31
(0.6%) |
256
(0.7%) |
| Total Discharges (n) |
17,587 |
6,989 |
7,781 |
5,126 |
37,483 |
| |
|
|
|
|
|
| Deaths in SC (n) |
627 |
144 |
82 |
127 |
980 |
| Discharges from SC (n) |
3,379 |
1,175 |
733 |
950 |
6,237 |
| Movements from SC to OC (n) |
4,957 |
1,457 |
2,302 |
1,083 |
9,799 |
| In hospital mortality (%) |
7.5% |
5.5% |
2.7% |
6.2% |
6.1% |
| Movements OC to SC (n) |
625 |
645 |
364 |
362 |
1,996 |
| Direct admissions to OC (n) |
13,612 |
4,277 |
4,239 |
3,560 |
25,688 |
| OC to SC/Direct admission OC (%) |
4.6% |
15.1% |
8.6% |
10.2% |
7.8% |
| |
|
|
|
|
|
| Average length of stay OC (days) |
29.1 |
31.9 |
28.5 |
26.5 |
29.1 |
| Average length of stay SC (days) |
13 |
11.7 |
16.4 |
16.1 |
13.9 |
| Average weight gain OC (g/kg/d) |
10.6 |
10.4 |
10.2 |
10.4 |
10.5 |
| Average weight gain SC (g/kg/d) |
18.9 |
19.1 |
14.3 |
14.2 |
17.3 |
SC=Stabilisation centre OC=Outpatient centre
| Table 2 Discharge profile per year for Maradi region |
| |
Total discharge from SC and OC |
Discharge from OC only |
| |
n |
n |
| 2002 |
5,307 |
3,557 |
| 2003 |
6,355 |
3,871 |
| 2004 |
9,524 |
7,104 |
| 2005 |
37,483 |
31,246 |
1Unicef. State of the world's children 2006: Excluded and invisible. NY: Unicef 2005.
2Briend A, Lacsala R, Prudhon C, Mounier B, Grellety Y, Golden MH. Ready-to-use therapeutic food for treatment of marasmus. Lancet 1999; 353: 1767-1768.
3Collins S. Changing the way we address severe malnutrition during famine. Lancet 2001; 358: 498-501.
4Community Based Therapeutic Care (Khara T. Collins, S. ed). ENN Special Supplement Series, No. 2, Emergency Nutrition Network, November 2004.
5Collins S, Sadler K. Outpatient care for severely malnourished children in emergency relief programmes: a retrospective cohort study. Lancet 2002; 360: 1824-1830.
6Ciliberto MA, Sandige H, Ndekha MJ, Ashorn P, Briend A, Ciliberto HM, and Manary, MJ. Comparison of home-based therapy with ready-to-use therapeutic food with standard therapy in the treatment of malnourished Malawian children: a controlled, clinical effectiveness trial. Am J Clin Nutr 2005; 81: 864 -870.