Use Of RUTF in Maradi, Niger2001 – 2007

From treatment to prevention of acute severe malnutrition

Context

• The programme strategies have evolved over time

– 2001 – 2005 : ambulatory care for severely malnourished children

– 2006 and 2007 : prevention (and treatment) of severe malnutrition

Guidan Roumji, Niger

Algeria

Mali

Burkina Faso

BeninNigeria

Tchad

Libya

Niger

Population in Guidan Roumji about 400 000

Guidan Roumji, Maradi region

• High prevalence and high incidence of malnutrition( Maradi region : Stunting : 62.2%, wasting 11.6%,

underweight : 54%, MICS III, February 2006)

• High mortality among children less than 5 years old ( under 5 mortality rate :198 %o, MICS III, 2006 )

Admissions of children with severe malnutrition (NCHS) in MSF program,

Guidan Roumji, Niger 2003 2007

0

50

100

150

200

1 27 1 27 1 27 1 27 1 27

2003 2004 2005 2006 2007

Seasonality severe malnutrition

Admissions of children with W/H < 70% (NCHS) weekly proportion of total inclusions in the year

Guidan Roumji, Niger, 2003-2005

0%

1%

2%

3%

4%

5%

6%

1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52

Week number

% o

f all

year

adm

issi

ons 2003

2004

2005

3-year Average

Young children are affected

Age classes among severely malnourished (NCHS) in MSF program

Guidan Roumji Niger, 2003-2007

0%

20%

40%

60%

80%

100%

2003 2004 2005 2006 2007

% o

f to

tal

yea

r

>36 monthsold

24-36months old

13-24months old

<= 12months old

• 2002-2004: growing knowledge on the use of RUTF

in the treatment of severely malnourished

• 2005: major crisis year, validates use on large

population

Management of severe malnutrition

Outcomes of children with Weight for Height <70% (NCHS), Guidan Roumji 2001-2005

56,2%66,1%

74,9% 74,7%

89,6%

0%

25%

50%

75%

100%

2001 2002 2003 2004 2005

% Lost to Fup

% Deaths

% Cured

Admissions: 1114 1803 1423 2471 7039

Management of SAM with RUTF

– Allows much more patients to be treated

– Significantly reduces lethality in program

– Only co-morbid patients are admitted in hospital: better use of medical resources

Management of SAM with RUTF

Remaining problems

– High number of severe malnourished children to take in charge every year

– Still a significant number of deaths in the program

– Many deaths not taken in charge within moderately malnourished?

Better strategy?

• Treatment of moderate malnourished • Prevention of SAM

• Passive recruitment

• Admission criteria : NCHS : W/H < 80%

• All children are managed by medical personnel

– Clinical screening (+ syst. RT for malaria)– Vitamin supplementation– No systematic antibiotics

• All malnourished children receive RUTF (184 g Plumpynut per day )

• All sick children hospitalized

• Others are sent home

Treatment of moderately affected

Moderate malnutrition Niger 2006

• High incidence • 32 262 admissions all over the year ( estimation of

65 000 under 3 children in Guidan Roumji)

• Results

• 32,254 discharged, all screened by medical personnel

• Exit criteria : 2 consecutive weeks, W/H > 80%• Cured rate : 95.5%• Weight gain ( g/kg/day) : 5.3 ( 5.25 ;5.32)• Mean length of stay ( days) : 31.4 (31.3;31.6)

Moderate malnutrition Niger 2006

• Too much for a health system to run– Lacking medical HR– Difficult supervision

• No obvious need for individual medical follow-up for all children

• Not the solution for a region with so high incidence

New strategy to give low dose RUF to non-malnourished children to prevent malnutrition

Systematic distribution in 2007

• Screening of all under 3 children organized before hunger gap period (over 62,000)

• Systematic distribution of RUF (46 g of Plumpydoz / day) every month for 6 months

• All severely malnourished treated as previous years

• Admission criteria WHO 2005 W/H < -3Zscore

Preventive distribution in 2007

• 62 878 children 6 months to 3 years, not severly malnourished received RUF

• 7258 admited to therapeutic program (severe WHO 2005)– Cure rate : 90%– 1532 were severly malnourished according

to NCHS

Preventive distribution

• The beneficiaries had to walk about 7km each way (14km) with their children in order to access the RUSF each month.

• The best way to evaluate the value placed upon this intervention by the population is to examine the drop-out rate from month to month as the program proceeded.

Beneficiaries Drop out Drop out

Month # # %

May 62,922 0 0.0

June 62,902 20 0.0

July 62,865 57 0.1

August 62,756 166 0.3

September 62,680 242 0.4

October 61,961 961 1.6

Defaulter rate in the distribution

Cases and new cases of SAM

Month TOTAL Normal Severe New severe

2007 # # % # % # %

May 49,248 33,737 69 200 0.4 200 0.41

June 56,172 40,784 73 204 0.4 136 0.24

July 60,605 47,089 78 125 0.2 30 0.05

Aug 61,506 47,646 77 108 0.2 12 0.02

Sept 61,858 45,472 74 129 0.2 18 0.03

OCT 60,556 42,529 70 147 0.2 37 0.06

• Cluster randomised trial• RUSF vs no prevention• Looking objectively into the reduction of incidence

of wasting ( + effect on stunting)• Mike Golden’s initial analysis

• Observational datas from the program :

– Reduction in number of severely malnourished children admitted with increased operational capacity

– Reduction in need for hospitalisation

Does prevention work?

Does prevention work?

2001 2002 2003 2004 2005 2006 2007

Admissions 1114 1803 1423 2471 7039 1599 1532

MUAC <110mm 700 1006 762 1267 3130 586 491

W/H >60% 27 75 51 49 139 26 29

Oedema 52 156 186 323 423 139 135

Hospitalisations 1091 1774 1264 1231 2893 756 680

W/H<60%

Relative severity of SAM (% complicated cases)

Percent of SAM requiring Hospital 2007

0%

5%

10%

15%

20%

25%

30%

35%

40%

Nigeria Madarounfa GR total GR with card

Hospitalised + hometreatment

Hospitalised based

Analysis

Sources of data to evaluate the impact

1. The village study

2. The MUAC values of the intervention group

3. The Admissions for treatment

The village study

• 12 villages were selected in 2006

• 6 were intervention villages and 6 control

• 3 of the intervention and 3 of the control were in Guidan Roumdji, the remainder were in Madarounfa Zones of Maradi region.

The village study

• The children in the intervention villages received one sachet of RUTF (Plumpy’nut) each day over the latter part of the “hungry season”: August, September and October.

• The teams took, inter alia, anthropometric data each month from all the children in each village

• Any child with malnutrition was referred for treatment.

• In April 2007 the 3 “control” villages in the Guidan Roumdji received another intervention (RUSF)

The village study

• The children of 6 to 36 months from each group were selected for analysis

• The weight-for-height, height-for-age and weight-for-age were computed using the NCHS standards. Less than <-2 Zscore was defined as global malnutrition and <-3 Zscore as severe malnutrition for this analysis.

• The mean and SD of the monthly data for each village and group used to compute the global and severe wasting prevalence, after checking that the data were normally distributed (Wilks Shapiro test).

Longitudinal prevalence of Global wasting in

Wasting RUTF vs No RUTF

0

2

4

6

8

10

12

14

16

18

20

Aug-06 Sep-06 Oct-06 Nov-06 Dec-06 Jan-07 Feb-07 Mar-07 Apr-07 May-07 Jun-07 Jul-07

<-2

WH

Z (

NC

HS

) ±

95

%C

I

NO-RUTF

RUTF group

RUTF Taken

Change in Severe Wasting RUTF vs No RUTF

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Aug-06 Sep-06 Oct-06 Nov-06 Dec-06 Jan-07 Feb-07 Mar-07 Apr-07 May-07 Jun-07 Jul-07

Δ%

<-3Z

WH

Z (N

CH

S)

±95%

CI

SAM non PPN

SAM PPN

RUTF Taken

• The groups were comparable at the start of the intervention

• The expected peak in GAM and SAM were ameliorated by intervention

• The prevention of malnutrition during the “hunger season” resulted in a prolonged improvement of the status of the intervened population

Stunting prevalence was not the same at baseline in the two groups of villages

Stunting RUTF vs No RUTF

40

45

50

55

60

65

70

Aug-06 Sep-06 Oct-06 Nov-06 Dec-06 Jan-07 Feb-07 Mar-07 Apr-07 May-07 Jun-07 Jul-07

%<

-2 H

AZ

(NC

HS

) ±

95

%C

I

No RUTF groupRUTF groupRUTF Taken

The effect was more pronounced for severe stunting which improved significantly. The lag in the peak of stuting vs the peak in wasting is frequently seen and is clearly illustrated here

Change in Severe Stunting RUTF vs No RUTF

-4

-3

-2

-1

0

1

2

3

4

5

6

7

Aug-06 Sep-06 Oct-06 Nov-06 Dec-06 Jan-07 Feb-07 Mar-07 Apr-07 May-07 Jun-07

Δ%

<-3

HA

Z(N

CH

S)

±9

5%

CI

No RUTF groupRUTF groupRUTF Taken

The Community Study

Guidan Roumdji

Data from Guidan Roumdji

• In Guidan Roumdji all the children of over 60 cm who were less than 3 years old were enrolled in a mass-intervention.

• They were each given 4 pots of RUSF (Plumpy’Doz) each month for 4 months over the hungry season.

• This amounted to an intake of about 42g/d

Relative nutrient intake from RUTF and RUSF

0

20

40

60

80

100

120

140

Village study data - GAMWasting RUSF vs no RUSF

0

5

10

15

20

25

Mar-07 Apr-07 May-07 Jun-07 Jul-07

%<

2 W

HZ

(N

CH

S)

±9

5%

CI

No RUSF

RUSF

Change in Severe Wasting RUSF vs No RUSF

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

Mar-07 Apr-07 May-07 Jun-07 Jul-07

Δ%

<-3

WH

Z(N

CH

S) ±

95%

CI

No RUSF

RUSF

• There was a marked reduction in wasting and severe wasting in the villages that received the RUSF distribution each month.

• There was no change in the prevalence of stunting or severe stunting.

• The level of underweight mirrored that of low weight-for-height.

What happened in the whole of the Intervention area

(of which the study villages were a part)?

Guidan Roumdji

0

1

2

3

4

5

6

7

8

9

May June July August September October

0

2

4

6

8

10

12

14

16new cases

old cases under treatment

seasonality

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

May June Jul Aug Sept Oct

Sev

ere

was

ting

(MU

AC

< 1

10m

m) %

pop

ulat

ion

0

2

4

6

8

10

12

14

16

Mon

thy

perc

ent a

nnua

l cas

es a

dmitt

ed fo

r tr

eatm

ent 2

002-

2006

New casesCases under treatmentExpected change in prevalence

Costs per patient

Direct cost : exit ration, coordination and support ( national storage) not included in the price

2004 2006 2007Cost of malnourrished patient 81€ 66€ 69€Cost of ambulatory patient 60€Cost of hospitalised patient 126€

Malnourrished child Distribution

2006 2007 2007Cost 66€ 69€ 35€Proportion of the costRUF 36% 33% 89%Human ressources 37% 54% 6%Logistic 13% 5% 5%Medical 7% 7%Others 7% 1%

Costs and feasibility

Guidan Roumji : estimation of 65 000 under 3 years old children

2006 2007Moderate NCHS 30663 RUF distribution 62878Severe NCHS 1599 Severe WHO 7258

Nb children hospitalised 3512 1722

Total children medically treated 32 262 7 258

2006 2007Cost of the program 2 100 000 € 2 700 000 €

Cost of RUF 766 000 € 2 081 000 €Cost of human ressources 790 000 € 400 000 €

Cost of medical 150 000 € 35 000 €

Cost 30% price reduction of RUF 1 870 200 € 2 075 700 €Cost 50% price reduction of RUF 1 717 000 € 1 669 000 €

Conclusion

• RUTF is very efficient in treating SAM

– Increased number of patients – High cure rate

• RUF prevent the appearance of SAM– Reduction of total number of severe cases – Reduction of the complicated cases

( admissions that require hospital stay)

• RUF preventive distribution could have an impact on the under 5 mortality

Conclusion

• In our opinion : enough evidence today to increase the use of fortified spread

• Especially in high burden areas

• Need more research :

– Comparison of products and dosage

– Demonstrate the impact on the under 5 mortality

• Cost reduction of RUTF? New cheaper products with same effectiveness! Increased funding !