I’m certain that all readers on the site (Global Health Hub) have heard of the Thousand Days concept. This is the notion that children are especially vulnerable to the effects of malnutrition and illness from conception up until they reach two years of age and that interventions targeted at this window have the best chance of improving child health worldwide. Over the last few years, the concept has served as a rallying point for major donors and has helped to shape the programs of numerous governments and nongovernmental and aid organizations. The primary success of the Thousand Days concept is that it provides a tangible, concrete tool for talking about nutrition programming, and as a result it has been crucial in galvanizing political will for combating child nutrition (1).
Equally importantly, it has shifted the focus of child nutrition work back towards very early childhood, a major step forward, especially in some countries (like Guatemala, where I primarily work), where most nutrition programming had, for decades, languished in the form of primary school-based programs (2). Furthermore, it has revived interest, research, and programming in primary prevention of malnutrition and, especially, the form of malnutrition known as stunting. Stunting is the most prevalent form of malnutrition in the world, with more than 150 million cases in children under 5 years of age globally (3). Stunting manifests as short height with relatively normal weight, and it presumed to be a proxy for brain development. Children who are stunted have delayed attainment of milestones and decreased cognitive potential. They also experience longer term effects, such as decreased rates of school completion and lower economic earning potential as adults (4). I don’t mean to reduce the entire focus of the Thousand Days approach to a focus on stunting, since it certainly covers many other things as well related to early child health. However, in high stunting areas like Guatemala the practical impact of the Thousand Days approach has been to reignite a the conversation about stunting.
The Thousand Days approach asserts that investments in the health and nutrition of child at the earliest possible stages of life are most likely to be effective (both in terms of improving growth and from a cost effectiveness stand point). As a corollary, it asserts that prevention of malnutrition is more likely to be effective than treatment of malnutrition. All of this is true, and validated by the scientific data. However, there is a danger that these assertions can be exaggerated or misinterpreted, leading to the crowding out of other important nutrition work, and this is what I’d like to discuss here. Increasingly in the field, I’m interacting with agencies and development workers who significantly over interpret the Thousand Days concepts. For example, rather than stating “Nutritional approaches in the first two years of life are the most effective,” I’m increasingly hearing “No nutritional interventions are effective after the first two years.” This is not true. Secondly, although the Thousand Days’ concept that prevention is more effective than treatment of stunting is both intuitive and supported by clinical data, I’m increasingly hearing the following, “Stunting can only be prevented, it can’t be treated.” This is also not true.
Again, this is not meant in any way to criticize the important step forward in global nutrition policy that the Thousand Days concept represents. However, let’s walk through the data a bit, and think through some of the potential unintended consequences of focusing just on the Thousand Days or just on prevention. Throughout what follows, I include several insights from a remarkable perspective piece on this issue by Prentice and colleagues (5), which I recommend reading in detail.
1. The scientific evidence that stunting sets in with a vengeance early in life, usually in the first year, is very clear. What happens after two years of age, however, is less clear.
As Prentice et al (5) eloquently outline in their perspective piece, much of the contemporary discussion about stunting emerged in the wake of the publication of the new WHO international child growth standards in 2006 (6). Analyses of national-level child growth data using these new standards highlighted several points. First, in under-resourced settings, children are born “behind the eight ball” with heights that are already well below the international standard right at birth. This finding is the key piece of data underpinning the Month -9 to Month 0 (conception to birth) component of the Thousand Days framework; improving the health of mothers is vital to guaranteeing the health and growth of infants. Second, over the first two years of life, and especially after 6 months, height velocity continues to drop steadily and then seems to plateau with little catch up (Figure 1, Ref 7). This leads to the second element of the Thousand Days framework – you have to prevent stunting early, or possibly treat it early, because after about 2 years it is harder to be effective.
From: Victora et al. 2010. Worldwide Timing of Growth Faltering: Revisiting Implications for Interventions. Pediatrics DOI: 10.1542/peds.2009-1519
However, as Prentice and colleagues (5) point out, the global growth data that has been used to generalize that height loss is fixed and irreversible by 2 years of age is heterogeneous and brought together from various national programs – i.e., it is not “scientific-grade” growth data collected in a rigorous fashion. As one illustration of this point, Prentice et al point out the “humps” at year boundaries in Figure 1, which suggest that the data has all been “rounded up” by local health workers and therefore might be making growth seem worse than it is. In several rigorous, clinical investigations in single sites, there is countervailing evidence that plenty of height catch-up can happen after two years of age, which we’ll get to towards the end of this piece.
2. The studied interventions that fall into a Thousand Days approach are not game-changing when it comes to stunting.
There are numerous modern interventions that have been studied for attempting to prevent or to treat stunting. These usually include education about breastfeeding and complementary foods, provision or supplementation of complementary foods, micronutrient supplementation, and so on. The problem is, however, that stunting is a complex multifactorial outcome, a final common pathway which integrates the innumerable medical, genetic, environmental, and political factors that determine child health and growth. Therefore, there is no “magic bullet.” In a metaanalysis of interventions that included complementary feeding interventions, for example, Dewey and Adu-Afarwauah estimated the average positive effect of these interventions at a shift in the population mean height-for-age Z-score of around 0.25 (7). A separate group of authors (8) performed a similar analysis, arriving at similar conclusions. The effect on height-for-age Z-score of complementary food provision and education to food insecure populations was a change of around 0.40, without actually changing the prevalence of stunting (defined as a height for age Z-score of less than -2). These authors also looked at the use of multiple micronutrient supplements, which are a major component of several countries’ responses to the Thousand Days’ concepts (including, for example Guatemala, 8) and found no effect on either height-for-age or stunting.
To put these effect sizes into perspective, I’ll give an example, using Dewey and Adu-Afarwauah’s estimate of around 0.25 change in height-for-age Z score. Let’s take a population in which 50% of children are stunted, which means that the population bell curve would be centered right at -2 in terms of height-for-age Z score. Now, let’s assume a positive change of 0.25, so that the curve is now centered at -1.75. This would represent roughly a 10% change in the prevalence of stunting for this population. From my perspective as a pediatrician and clinician, 10% is awesome. But, let’s face it: it is an incremental change, not a “change in the game.” So, if we view the Thousand Days perspective as a useful and important framework for spurring critical thinking, encouraging innovation, and engaging political will to fight malnutrition, then I think that is very useful and very exciting. However, if we somehow conflate the utility of the Thousand Days as a framework with the actual effectiveness of the known interventions that we are implementing under the rubric of the Thousand Days, then I think we are in trouble, and this is what I am hearing more and more in my conversations with rank and file development organizations. The Thousand Days should be a call to action. But fighting stunting is a long, slow grind–trench warfare if you will–and we’ll be at it for a really long time yet.
3. Prevention of stunting is an intuitive and important approach, but the effect size is small and we have limited data.
It has been very interesting to watch the evolution of the prevention discussion over the last few years, up through and including the elaboration of the Thousand Days approach. As a front line implementer, I had struggled for years with working to get our programs to efficiently identify undernourished children and target them for treatment. This was extremely difficult, and unacceptable delays between identification by a community health worker and formulation of a treatment plan were inevitable. As a result, in our Guatemala programs we had switched to blanket (“preventative”) complementary feeding and micronutrient supplementation in 2006, with excellent results, in some cases more than 25% reductions in prevalence of stunting over 1-2 years. Nevertheless, none of our programs were controlled, and firm analysis was difficult due to the fact that our programs and targets frequently changed as we collaborated with communities to develop shared priorities.
Therefore, we were very excited when in 2008 Ruel et al published the results of their cluster randomized trial evaluating the difference between an intervention package and a preventative package in two groups of communities in Haiti served by World Vision (9). This trial received a great deal of press, deservedly so. It showed, unequivocally, that the blanket approach was more effective on all three main metrics of wasting, stunting, and underweight, which reinforced the experiences of front line workers like ourselves in Guatemala and elsewhere. The ripple effect from this study was marked. For one, it anchored the articulation of USAID’s PM2A (Preventing Malnutrition in Children Under 2) strategy, which was influential in the emerging policy consensus that was to form around the Thousand Days. In Guatemala, immediate effects from the study were a major shift in the funding activities of big donors, as well as the emergence of a new national discussion about stunting, which was also facilitated by the latest (terrible) statistics on national stunting prevalence (10). At the same time a bolus of small clinical trials, which were strictly prevention trials (e.g., only enrolling children at 6 months of age and following them prospectively), emerged, many of them from the International Lipid-Based Nutrient Supplements (iLiNS) Project consortium (www.ilins.org), further adding to the momentum (although most of these trials were very mixed in terms of outcomes).
Nevertheless, several persistent questions undermine the definitiveness of the prevention approach. First, the Haiti study occurred in a relatively food secure population. For example, a majority of children in the trial had regular access to animal proteins and consumed the recommended number of meals. This is very different from the situation in more food insecure populations, including the population of rural Guatemala, where animal protein consumption is virtually nonexistent and meal frequency is suboptimal (11, 12). Second, the “treatment” arm of the study targeted children for treatment based on underweight not underheight; it is surprising, therefore, how much this study has been used to justify prevention approaches to under-height/stunting; indeed the effect size for stunting in the study was only a change in height for age-z-score of +0.14 (around a 5% change in stunting). Finally, Haiti has only a moderate prevalence of stunting, and so it is uncertain whether the results are applicable to higher-prevalence countries, like Bangladesh or Guatemala (3).
Taken together, I would conclude that we should be very excited about the push to prevention of stunting that has emerged under the Thousand Days consensus. However, we should be quite realistic about both the quantity of the evidence for effective prevention interventions and also the magnitude of the effect of these interventions on our populations, especially in more highly-stunted and more food-insecure areas. Practically, this means we have to fight hard against how articulation of the Thousand Days consensus is morphing into discourse by front line workers and organizations that “prevention is the only effective approach.”
In particular, in the most highly-stunted populations – such as Guatemala, whose poorest 20% are probably the most stunted population in the world (3) – we should realize that, from the standpoint of community buy-in, prevention-only programs are destined, in the short term, to symbolic failure. I work in some communities where stunting prevalence is 90%. In this setting, a decrease of 5% in prevalence is not perceptible to the population, and will be interpreted locally as a failure. Furthermore, we’ve discovered in our work in rural indigenous communities that an exclusive focus on children under 2 is a nonstarter at all levels. And this should not surprise us, when ALL children of all ages are severely malnourished. Therefore, recuperative programming, and programming targeted at children older than 2, is an ethical and pragmatic mandate in highly stunted communities, and it will remain so for the foreseeable future. In addition to wholeheartedly endorsing the energy and innovation of the Thousand Days approach, we must also be wiling and able, at the community level, to innovate on nutrition for children of other ages. This is the incipient danger I perceive when the Thousand Days are inappropriately glossed as “interventions after 2 years of age are ineffective.”
4. Good news: Catch-up growth after 2 years happens!
I’ve already discussed above under point 1 how the interpretation of large combined child growth data sets need to be interpreted with caution, and Prentice et al (5) go into this in much more detail. To contrast, let’s look quickly at some “trial quality” data sets from single sites. For example, there’s an amazing paper by Stein and colleagues (13) writing for the Cohorts Group, who show quite significant catch-up growth from 2-5 years in several independent cohorts, from Brazil, Guatemala, the Philippines, and South Africa. In Guatemala during the classic INCAP cohort trial, which compared growth in children supplemented with a high quality supplement vs. a low quality supplement, a difference in linear height attainment of 0.4 cm/yr could be observed up to 36 months of age (14). Granted, in line with the Thousand Days principal that interventions are more effective earlier in life, this was not as large an effect as the 0.9 cm/yr observed in children under 12 months of age, but it shows us that there is still considerable work that can be done outside the Thousand Days window. Similar findings are available from other cohorts as well. For example, supplemented Columbian children aged 24-36 months had a height difference of 0.47 cm over their unsupplemented peers, while children aged 9-12 months had a similar advantage of 0.45 cm; in this cohort, the effect of intervention in children outside the Thousand Days window was of the same magnitude as for younger children!
In conclusion, the core principles of the Thousand Days initiative represent a remarkable and important consensus, and a “change in the wind” with respect to political will around early child nutrition and, particularly, stunting. At the same time however, we must resist any overstating of the evidence base for or efficacy of the core Thousand Days initiatives. Most importantly, we must not forget that, especially in food-insecure and highly stunted populations, we are ethically obligated also to continue to innovate and design programs to meet the nutritional and health needs of older children.
1. Gillespie et al. 2013. The politics of reducing malnutrition: building commitment and accelerating progress. Lancet 382:552
2. Kraemer and Rohloff. 2013. Guatemala. In Central America and the Caribbean: A Practical Guide for Global Health Workers (M Krasnoff, Ed). Dartmouth: University Press of New England.
3. Black et al. 2013. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 382:427
4. Dewey and Begum. 2011. Long-term consequences of stunting in early life. Matern Child Nutr 7 S3:5
5. Prentice et al. 2013. Critical windows for nutritional interventions against stunting.Am J Clin Nutr 97:911
6. WHO. 2006. WHO child growth standards based on length/height, weight and age. Acta Paediatr S450:76
7. Victora et al. 2010. Worldwide Timing of Growth Faltering: Revisiting Implications for Interventions. Pediatrics DOI: 10.1542/peds.2009-1519
8. Dewey and Adu-Afarwuah. 2008. Systematic review of the efficacy and effectiveness of complementary feeding interventions in developing countries. Maternal and Child Nutrition 4: 24
9. Bhutta et al. 2013. Evidence-based interventions for improvement of maternal and child nutrition: what can be done and at what cost? Lancet 382:452
10. Secretaria de Seguridad Alimentaria y Nutricional. 2012. Plan “Hambre Cero”. SESAN: Guatemala City, Guatemala.
11. Ruel et al. 2008. Age-based preventive targeting of food assistance and behavior change and communication for reduction of childhood undernutrition in Haiti: a cluster randomized trial. Lancet 371:588
12. Ministerio de Salud Pública y Asistencia Social. 2009. V Encuesta Nacional de Salud Materno Infantil 2008-2009. MSPAS: Guatemala City.
13. Chary et al. (2013). Formative assessment of infant and young child nutrition in two indigenous communities in Guatemala. Wuqu’ Kawoq/Edesia.
14. Food and Nutrition Technical Assistance III Project. 2013. Summary Report: Development of Evidence-Based Dietary Recommendations for Children, Pregnant Women, and Lactating Women Living in the Western Highlands in Guatemala. FHI360/FANTA: Washington DC.
15. Stein et al. 2010. Growth patterns in early childhood and final attained stature: data from five birth cohorts from low- and middle-income countries. Am J Hum Biol 22:353
16. Schroeder et al. 1995. Age Differences and the Impact of Nutritional Supplementation on Growth. J Nutr 125:1051S
17. Lutter et al. 1990. Age specific responsiveness of weight and length to nutritional supplementation. Am J Clin Nutr 51:359