19 October 2017

Micronutrient Powders: Getting a Grip on Effective Implementation

I had heard of micronutrient powders (MNPs) before I joined GAIN, and, I suspect like many non-experts, I assumed it was a pretty simple intervention.
How wrong I was.
For those of you who are less familiar with the nutrition world, MNPs are sachets of powdered vitamins and minerals that can be added to complementary foods and given to children on a daily basis. They were introduced by Dr. Stanley Zlotkin (who is also a member of GAIN’s Board) in the late 1990s. Easy to use, MNPs do not change children’s dietary habits but can be added to semi-solid foods that are already part of a child’s diet. More importantly, MNPs boost children’s micronutrient intake and reduce their risk of iron deficiency and anaemia.
A just published Supplement of Maternal and Child Nutrition contains a series of papers emerging from a Consultation that summarise the lessons learned from operationalizing MNPs. A subsequent paper which took advantage of the Consultation’s extensive systematic literature search published in Current Developments in Nutrition summarises what is known about factors affecting adherence to MNP recommendations.
The Consultation drew on the reflections of 49 MNP implementers and experts and a review of published and grey literature.
Here are the key points from the overview article by Dhillon et al. and adherence paper by Tumilowicz et al.
1. Iron deficiency anemia (IDA) is the leading cause of years lived with disability among children (I did not know that!)
2. Peak prevalence of IDA occurs at 18 months after which iron requirements decline and iron intake increases through complementary foods (I did not know that either)
3. MNPs have generally replaced iron drops/syrups because they show similar efficacy in reducing IDA (by 51 percent) and anemia (by 31 percent) in children under two years of age, but with higher acceptability and fewer side effects
4. Global scale up has been helped by the fact that MNPs are easy to use and do not require dietary change and, on average, cost USD 0.02 per sachet to produce.
5. Scale up of MNPs has been rapid: from 36 interventions in 22 countries in 2011 to 59 interventions in 50 countries in 2014.
6. Of the 50 countries, nine were implementing national programmes and 20 subnational programmes (I guess the remaining 21 were doing pilots).
7. Despite rapid adoption the extent to which the quality and scalability of MNPs can be maintained is yet to be established. Hence this Consultation.
8. The Consultation looks at 3 dimensions of MNP implementation (1) planning and supply, (2) delivery, social and behaviour change communication, and (3) continuous program improvement
9. Planning and Supply
  • High quality research and data analysis to justify MNP intervention is not always conducted fully
  • Leadership is an important driver of intervention uptake, but so too is the need to find more effective approaches to address childhood anemia
  • Sustainable funding remains a challenge despite evidence of high cost effectiveness – it is good to discuss long term funding early on in implementation. Adding to existing national programmes improves sustainability (e.g. social protection in Mexico and Dominican Republic)
  • Securing reliable and regular supply of MNPs is a challenge for many countries. This means that countries often rely on global suppliers, but this uses valuable foreign exchange, and can lead to long procurement lead in times and problems around local languages and packaging
10. Delivery, Social and Behaviour Change Communication and Training
  • Here the models vary by design feature: (price: free, subsidized or full cost) x (point of distribution: facility health workers, community members, pharmacists) x (sector: health, social protection, agriculture
  • Free distribution through the non-health sector – for example, social protection and early childhood development programs – has shown higher coverage rates than free distribution via health systems
  • Subsidized distribution has achieved a wide variety of coverage rate
  • Whatever the delivery strategy, more and more MNP programmes have been linked to broader infant and young child nutrition objectives
  • Programs are increasingly measuring appropriate use and intake adherence in addition to just coverage
  • Social and behaviour change (SBCC) need to be applied throughout program cycle
  • Regular refresher training of MNP distributors is essential to ensure high quality counselling and messaging
  • Two thirds of all MNP interventions are funded entirely by development partners and this is not really sustainable
11. Continuous Program Improvement
  • Lack of documented MNP experiences, especially those implemented at scale. There are 15 peer reviewed papers and most come from pilots and were externally funded with few examples of them being used to improve implementation
  • Most programmes do not apply a mapping of programme theory to track progress or make course corrections efficiently
12. Factors Affecting Adherence
  • From the perspective of caregivers, positive changes in their children (for example, improved health, increased appetite, increased energy) and acceptance of food mixed with MNP are the main reasons for continuing to use it
  • Caregivers are less likely to stop feeding MNP if they are informed of potential negative side effects (such as changes in stool)
  • In addition to SBCC strategies, administration regimen (fixed or flexible dosage schedule), which may be related to caregivers’ capacity to remember to give MNPs, is frequently cited as a program design feature affecting adherence.
As the papers note, “..implementing MNP programmes effectively remains a complex challenge” and “preparing food with MNP correctly and succeeding in getting a child to eat it depends on a complementary feeding process that requires a complex set of caregiver behaviors and caregiver-child interactions”.
As someone who has spent time outside as well as inside the nutrition world, I’m wondering how complex MNP programs are to implement effectively compared to the program it replaced, namely iron drops/syrups. I’m also wondering how MNPs stack up against other nutrition interventions (e.g. promotion of exclusive breastfeeding, Vitamin A supplementation, biofortification) and against other development interventions (e.g. social protection, WASH, public works programs).  My experience suggests they all require careful planning, linking of supply and demand elements, and learning feedback loops.  They all require sustainable funding.  The impacts of none are resilient to poor design and implementation.
The state of MNP implementation reminds me of the state of conditional cash transfers in the 90’s.  Everyone was jumping on the bandwagon, and many of the transfer programs did not have an impact on nutrition status because of poor design, poor implementation, poor evaluation, or some combination of the three.  But as the good evaluations rolled in, meta-analyses were able to identify the rules of thumb that made positive impact more likely and sustainable at scale, and this was the trigger for governments, such as Ethiopia, to invest their own resources in them.
In sum, there are two things I have learned from the past year at GAIN that speak to these issues: (1) it is harder to design and implement effective programmes than it is to design and implement effective evaluations of them and (2) if you don’t evaluate programmes it is even harder to design and implement them effectively!
The papers in this supplement are a valuable reflection on the MNP experience to date, and I am happy that GAIN was a contributing partner in the endeavour.

No comments: