You say you want a nutrition data revolution? Well, you know, we’d all love to see the plan

One of my favourite book reviews (as long as it never applies to me) is: “This book fills a much needed gap” (think about it).  Those who advocate for more nutrition data need to ask themselves the same thing: do these data simply fill a much needed gap?  In other words, do we really need to fill any given data gap?

Yesterday I participated in a brainstorming meeting at WHO on data gaps that matter for scaling up nutrition interventions.  Data are needed at the subnational, national and global levels to guide action, to assess progress, to help advocacy and to strengthen accountability. 

We spent the whole day (a) identifying gaps that need filling, (b) mapping ways forward on data gaps that can be filled relatively straightforwardly and (c) brainstorming on the more difficult or “gnarly” data gap issues.

Some reflections:

The participants work mostly at the “global” level—and it showed.  It was important to keep in mind the constraints faced by people working with data in the domestic nutrition system.  There was a tension between framing things as “what do we need to do to get more domestic data into the international system?” and “what do countries want/need to move their nutrition agendas forward?”  Another tension was around framing that said “what additional data would be useful to countries?” and “what do countries need to make existing data more useful?”  All of the framings are valid, it seems to me, it is about listening to countries while also meeting global accountability and advocacy needs.  But how to “listen to countries” without burdening countries, while doing it in a demand driven way that is not patronising?  And then how to aggregate up?  Perhaps via country typologies in terms of the nutrition problems they face and the capacity they have to collect and use data.  

The parlous state of food consumption data. This was mentioned by nearly everyone as a problem.  Food intake is vital to address undernutrition, but also other forms of malnutrition driven by unbalanced diets.  How can countries formulate a strategy to address undernutrition in a balanced way if they do not know what their population eats? There are a number of initiatives in process to strengthen food intake data, but there was a general sense that data collection in this area needs to step up a gear.

Coverage data.  As the GNR 2014 pointed out, of 10 Lancet nutrition specific interventions, only 4 have internationally comparable data on coverage for more than a handful of data (vitamin A supplementation of under 5’s, iodized salt intake, zinc treatment of under 5’s with diarrhea and iron folate capsule intake during pregnancy). 

But what to do about nutrition specific intervention coverage?  Why is it not more routinely embedded in on going data collection activities such as DHS/MICS/SMART surveys?  One reason is that we do not have a set of agreed on and easily implemented coverage indicators that can be embedded in surveys or, indeed, in facility-based data collection.  We need such a set. 

And what to do about nutrition sensitive programmes? How do we know if they are nutrition sensitive and how do we figure out who they reach?  No answers I’m afraid, just prioritizing unknowns.

Cost data for nutrition interventions.  If we want to know how much reducing malnutrition costs—and try to make action more cost effective--cost information is essential.  What can we do to make sure these are always collected?  And how can these data be curated and made available to a wider audience?

So more questions than answers, but a list of follow up actions was drawn up, and different people committed to deliver on these actions.  Many were process actions, but that is OK it seems to me.  The costs of data collection are high and the costs of collecting the wrong data are even higher.  Doing data things in a deliberate and coordinated way is important.  

Surprises?  There were 2-3 strong advocates for better micronutrient surveys but not as many arguing for it as I would have thought given the large numbers of people affectedly them, the lack of specificity about who they are and the lack of progress in addressing the burdens.  We did not talk much about indicators for capacity (beyond the number of front line health workers) although this is an important issue.  We talked about the need for better data on preventing severe acute malnutrition and on the coverage of treatment, but not as much as the issue perhaps deserves.  As usual, the issues raised depend on who is in the room, no matter how objective they are trying to be. 

In the context of the broader dialogue, the group is aiming to write a more considered piece for public consumption to contribute to the wider discussion about which data to invest in.  And why that investment would fill a gap that is not much needed.