The Leading Voices in Food
E138: Inside the International Dietary Data Expansion Project
Researchers and policy makers in agriculture, food security and nutrition share a common need for accurate and timely information on the what, when, where, and why people eat and what they eat, of course, this is particularly true in low and middle income countries where the data infrastructure is less well developed. To put this challenge in perspective, in 2015 the Global Panel on Agriculture and Food Systems for Nutrition reported that, and I quote “more than half of the countries in the world do not collect the statistics, which are needed to assess whether or not they are making progress toward their nutrition goals.” So today we’re talking with two researchers who are working to solve this very data challenge. Our guests are food policy and applied nutrition researcher, Jennifer Coates, Associate Professor at the Tufts Friedman School of Nutrition, Science, and Policy and senior researcher Winnie Bell. Jennifer and Winnie are leading development of the International Dietary Data Expansion Project
Dr. Jennifer Coates is the Principal Investigator of the INDDEX Project and is an Associate Professor at the Friedman School of Nutrition Science and Policy and a Senior Researcher at the Tufts Feinstein International Center. Dr. Coates’ research focuses on the development of methods and metrics for improving the design, implementation, and evaluation of international nutrition and food security programs in both development and humanitarian emergency contexts. In addition to her research and policy engagement, Dr. Coates teaches a range of graduate courses at the Friedman School.
Winnie Bell is a Senior Researcher with the INDDEX Project. Winnie was responsible for leading the INDDEX24 research activities in Viet Nam, including conducting the feasibility and validation studies. In addition, she is providing technical expertise and managing the development of the INDDEX Project’s Data4Diets Platform. Prior to joining the INDDEX Project, Winnie had worked for over five years as food security analyst and researcher with the Food and Agriculture Organization, the Food Security Information Network (FAO/IFPRI/WFP), and Tufts University. Winnie is currently a PhD candidate at Tufts University, where she also earned an MS in Food Policy and Applied Nutrition and an MPH in Epidemiology and Biostatistics.
Jennifer, let’s start with you. Your research focus at Tufts is on the development of methods for improving the design implementation and evaluation of international nutrition and food security programs. And I know in particular, in the context of development in humanitarian emergencies. Could you help our listeners understand the issue of this INDDEX Project that you’re working on?
Jennifer – Just to give a little bit of history. The international community has probably over the past five or so years really begun to embrace the need to think about nutrition’s role at the nexus of agriculture and the environment and health. So not just a standalone sector, but rather a culmination of whole bunch of different forces within the broader food system that in combination have this remarkable impact on people’s food choices and ultimately their diets. So we also know that in turn, the food choices really drive health outcomes and in a very systemic fashion affect agriculture, the environment, food security, poverty, and so on. So the problem though, for those who are interested in the interplay of these forces in much of the world, is that we know shockingly little about what people eat and we’re not able to easily distinguish the dietary patterns of adult men, for instance, from school age girls or from those living in one region of the country to another. And this information gap has really hindered policies in all of the different sectors that I mentioned and others, because policy makers are often driving blind without having more frequent and more granular dietary information. This data gap is particularly a challenge across low and middle income countries.
So let me follow up on this. It seems like this information is pretty darn important because it could drive nutrition guidelines for countries that could drive bio-fortification programs. They could certainly have an impact on agriculture policies.
Jennifer – There are a number of different reasons, in fact, for one thing, we’re globally really lacking a kind of cohesive dietary data infrastructure, I will say, and this is made up of several different missing pieces of information. So we lack the kind of information that can help us actually carry out dietary surveys. So information on what are the most commonly consumed foods across many of the countries where people live and also what are the nutrient composition of that food in the US for instance, we have a very rich repository of data on the foods consumed by people all over this country, including many, many different kinds of processed foods. And we also have information on the nutrient content of those foods. And so it makes it very easy to take that information and go out and conduct a survey on what people are consuming, but we don’t have that kind of information for most low and middle income countries. And another reason is that partly because we lack this dietary infrastructure, people have over time, come to view dietary data as being very challenging to collect and have shied away from it because of that. So now we’re reaching a point, I would say, where we’re seeing a groundswell of recognition for the need for dietary data and then that is driving a lot of efforts, not just our project, but other related projects that are really trying to begin to close this gap.
I appreciate that hopeful note. So Winnie, you’re leading key aspects of the second phase of this project, which is the rollout of the INDDEX24 dietary assessment platform. Could you describe the platform in more detail and why you believe it is needed.
Winnie – The INDDEX project, the International Dietary Data Expansion Project has been around for about six years. We started in 2015 with funding from the Bill and Melinda Gates Foundation, and our main objective has been to increase the availability access and use of dietary data in low and middle income countries. The platform is designed to address some of these underlying challenges that Jenny just spoke about.
INDDEX was conceived to begin to address the lack of dietary infrastructure for collecting individual level quantitative dietary data. One part is the Global Food Matters Database where we house and fold all of the dietary reference data that are required for individual level food consumption surveys. So that includes food composition information, standard recipes, necessary conversion factors, and the probes that are used during the interview process. So that’s the first kind of big component of the INDDEX24 platform.
The next piece is the INDDEX24 mobile app, which is actually what’s used in the field to collect the dietary data during the survey, both of these pieces are connected through a survey management hub where the data can be monitored in real time during data collection, as well as used for easing the processing and analysis of the 24-hour recall data on the backend. So the 24-hour dietary recall method is commonly used to collect detailed quantitative information on what individuals consume. Individuals are asked to recall everything that they consumed in the past 24 hours followed by a more detailed information on each of the food items, getting into that type of food consumed and the brand, for example, the third step is to then collect the quantity of each food consumed and then finally reviewing that information with the respondent.
All this might sound a bit abstract, so I would encourage listeners to check out our website for a full visual and a demo of the different aspects of the platform and in terms of why it’s needed. And this, as Jenny highlighted, we have this critical gap in information and this platform has been designed to increase a collection of 24 hour dietary recall data in low and middle-income countries.
This could be enormously powerful to have this tool available and it’s really nice that you’re doing it around the world with a special emphasis on countries where the data are most lacking. So Jennifer, let me go back to you in your opinion, how has the international nutrition and dietary assessment landscape changed over the, say the past decade and looking into your crystal ball, how do you think it will change?
Jennifer – I mentioned earlier, one of the reasons there has traditionally been so little dietary data available is that it was often written off as being too expensive, too time consuming and too difficult to collect. So traditional dietary assessment methods, such as those that rely on respondent recall can also produce inaccurate results, which people recognize and were discouraged by this narrative. This ridiculous challenge of dietary data assessment really became a mantra, it became almost paralyzing I’d say, one of the biggest changes that I’ve noticed over the past five to seven years is an acceleration on so many fronts. There’s been an acceleration in terms of the demand for dietary data, absolutely. And then in response, but also out there leading, I’d say those involved in generating dietary data have developed a new can-do attitude, a real rolling up the sleeves and just getting it done.
Partly this has been catalyzed by very generous funding by the Bill and Melinda Gates Foundation, not just of our project, but of other similar projects with INDDEX24 we’re trying to make dietary data, I guess, increasingly available. We’re hoping that it will be collected more frequently and therefore be used by policy makers more often, but there are also other aspects of the dietary data value chain that other projects are addressing. For instance, the WHO/FAO GIFT Project has been working to make dietary data more readily accessible in a centralized repository and much more usable by policymakers, through simple visualizations that communicate very readily to those who aren’t up to their eyeballs and dietary data, and might not know otherwise how to interpret it. So this kind of information has improved on the way that people both access and use dietary data.
And then looking forward, one of the things that we’re seeing are a huge number of efforts going on to develop innovative technology. Much of it is still nascent, not quite ready for prime time, but to improve the possibility of capturing information on diets in real time, like while people are eating, researchers are using imaging approaches coupled with artificial intelligence in order to upend the traditional field of dietary assessment. So, no, we’re not there yet, we’ve already made leaps and bounds on this front, over the last, over the past 10 years and I think that 10 years from now this whole challenge will look completely different, in fact, I think we will probably have done away with many of the things that we now find to be very difficult about dietary assessment.
What you just mentioned about technology being used to more accurately assess what people are consuming sound very promising. And I remember back when I was first paying attention to work in this area, and there were a lot of concerns about the accuracy of things. Researchers were reporting that people may not be able to accurately judge portion sizes. They may intentionally under or over report consuming such things. And there was talk in those early years about having other information available where you could correct for things to actually establish a more realistic number. So say people underestimate by an average of X %, then you could correct for that, are people doing that at all? Or will this technology help with that issue?
Jennifer – There’s work going on for both of those fronts, so you’re absolutely right, that those types of corrections are possible and are actually very useful. One of the things that is very challenging right now in our efforts in low and middle income countries is that we don’t actually have enough information yet to really be able to establish appropriate correction factors. One of the reasons for that is that, for instance, in the US people most commonly under report, what they’ve consumed the previous day and that’s partly because they forget, and it might also be because they think it might be more socially acceptable to under-report. But we really don’t know all that much about how people from various cultures report their diets, when they’re trying to recall, there is documented under-reporting.
There’s also documented over-reporting and there’s a lot of variability establishing the correction factor is a whole other level of work that would be very helpful and remains to be done. One of the benefits of the imaging technology is that you’re not relying on people’s memories. And in fact, the camera is present, even though that sounds a little creepy, big brother watching everything that you’re eating, but the camera is present and often in an unobtrusive way, as people are eating their meals. And so of course you might be concerned about people modifying their behavior, if they’re aware of a camera recording, what it is that they’re eating, but at the same time, some of this research has suggested that having the camera be situated in a very, very small way on, the special glasses that people be asked to wear or as a button that they wear on their shirt, for instance, has really helped reduce some of that kind of potential bias too. So there’s a lot of promise from the imaging and there are still challenges to be worked out for instance, how to take the images that the camera’s capturing and in a very automated way, convert that into something that can be recognized as food, especially in complex dishes, like stew with lots of different kinds of ingredients. So that’s kind of the excitement, but also the challenge of this new frontier.
That’s really interesting. So let me close by asking you both this question moving forward, what do you see as the biggest challenges as well as the opportunities for increasing collection and use of dietary data in low and middle income countries? And Winnie let’s begin with you.
Winnie – So, as we’ve both mentioned, the dietary data infrastructure has been historically one of the largest barriers to successful completion of dietary surveys in low middle-income countries. The INDDEX Project has been focused on trying to build up this research infrastructure over the last six years of our project. But the challenge remains is it an enormous undertaking that really will require political will from a variety of key players and champions from both governments as well as on the donor and investment side, as well as on the research end of things. We’re absolutely moving in the right direction, but I think the challenge remains, how do we develop this robust and necessary dietary data infrastructure to ensure that we can collect more easily and use more frequently the dietary data from low middle-income countries.
We have successful examples of this being developed where political will has converged. So in the health sector, for example, we’ve seen huge movement over the last several decades to creating health systems in low middle-income countries. So we’re hoping that the momentum continues to grow, to overcome this challenge in relation to nutrition. And as far as opportunities, I think there are several, but a very concrete one is on the technology side of things as Jenny mentioned with artificial intelligence and other efforts. And I think the other overall opportunity is general growing awareness of the importance of dietary data. Everybody is interested, everybody wants to see the results, and historically people are happy to have the data available, but few people have been willing to put in the effort to invest and collect those data.
Thank you, Winnie. So Jennifer, what are your thoughts on this issue?
Jennifer – I’d say that with dietary assessment in low and middle income countries, having historically been such a neglected area relative to say the many advances that we’ve seen in terms of information systems in the health sector, there has also been a lack of opportunity for very vibrant knowledge exchange. There are countries that have developed a lot of expertise and a lot of capacity in the area of dietary assessment and yet the kind of infrastructure that would enable country to country learning as well as training and mentoring between those who have already amassed a deal of experience in dietary data assessment, and those who are trying to improve their skills in this area has really been lacking. I’d say that the flip side opportunity to that challenge would be to scale up a global community of practice that can serve as a platform for exchange, for training, for learning and for maintaining this positive momentum that we’ve seen over the last five years. So we need to be thinking about the next generation and equipping them with the tools and the skills and the capacity to be able to carry this forward.