Know Your Deficiencies,Know Your Response,Know Your Costs

In 2007, the HIV community embraced a major shift in approach under the banner of “know your epidemic, know your response.” ¹ This was a major pivot in combatting HIV that continues to this day, from generalized approaches to more tailor-made programs based on increasingly finer understanding of the nature of the epidemic and its drivers. Tackling the challenge of undernutrition, including micronutrient deficiencies, has tended to rely on blanket approaches, due in large part to the lack of adequate data and tools to design and implement programs adapted to disaggregated geographic and socioeconomic settings. This series of articles, using vitamin A deficiency as a starting point, sets out a path forward for nutrition to start a similar pivot that HIV started almost a decade ago. It could be entitled “know your deficiencies, know your response, know your costs.” It also recognizes that we need to manage the tension between getting immediate results and building for long-term solutions. One of the recurrent criticisms of vitamin A supplementation programs, for example, is that they have somehow diverted attention away from ensuring “frequent intakes of vitamin A in physiological doses—eg, through food-based approaches.” ^2(p.1) Lack of data and tools to understand and design for the optimal mix of interventions, across groups and over time, is a key limitation in managing this inherent tension. This series of articles proposes a way forward to address this gap.

This commentary summarizes the responses of 5 leading public health nutrition practitioners who were invited to review and comment on the articles, with a particular focus on (1) the gaps in important information available to micronutrient program decision makers that would be filled by the suite of models we have developed, with particular attention paid to the economic optimization model, (2) the major challenges we can expect to face as we begin to use the models to engage with decision makers in designing and managing micronutrient intervention programs, and (3) changes to any of the models that would make them more relevant and useful. The commentary consolidates these points of view to point out areas of agreement and areas of differing opinions.

Overall, there was a strong consensus that this approach was an important step forward in evidence-based tailoring of intervention mixes and that this approach is very timely, as nutrition needs to “catch up” with other sectors. One commentator concluded that “These models contribute to addressing the current absence of a unified framework to address micronutrient deficiencies. They therefore have the potential to address major gaps faced by micronutrient deficiency control programs by improving the framing of programs, strengthening the use of data to quantify burden of micronutrient deficiencies and proposed solutions, and supporting international advocacy on micronutrient deficiency control.” Another contributor stated, “This set of papers provides the first comprehensive approach to priority setting that is soundly based on cost and potential for impact—a much overdue and urgently needed decision-making tool for nutrition.” A third stressed that “Nutrition has been a latecomer to this discussion and needs to play catch-up if it is to remain relevant and competitive in the wider development playing field. The focus of these articles on costs and cost-effectiveness as well as improving allocative and technical efficiencies moves the discussion in the right direction as countries and the development community become more and more focused on what results we can buy with our development dollars.” It was noted that further development of this approach is particularly relevant in the context of the Scaling Up Nutrition (SUN) movement, which is promoting better coordination of nutrition interventions and development of national plans and related budgets.

While there is enthusiasm for the path forward that these articles illustrate, all commentators also recognized that “Prioritization of interventions is both a technically and a politically negotiated exercise.” One contributor noted that the geographic targeting of interventions may be particularly politically unacceptable, as it would be perceived as inequitable. One proposed solution is, “An optimization model that could permit specific equity considerations to be incorporated would be aligned with need and priority-setting criteria for nutrition problems in many countries.” Furthermore, it is possible that “for a decision maker, efficiency and funding might be less important than maintaining institutional and working relations.” It was noted that the optimization modeling will be challenged by the realities of short-term financial planning. Without a predictable financing forecast, government decision makers might be reluctant to embrace the “all or nothing” choices that come with the economic optimization model. One contributor noted, “the model is currently estimated over a 10-year timeframe, but most policy and program decisions are made on shorter time frames, often 5 years, dictated by government and donor decision-making and funding cycles.” The trade-offs to be made over a 5-year horizon may be significantly different than those made over a 10-year horizon.

All contributors agreed that to bridge the technical and political divide, it will be essential to continue to generate better data on deficiencies, effective coverage, and program costs and perhaps, even more importantly, to build national capacity and processes to own the analyses and therefore own and implement the implications of the analyses. Improving data, capacity, and processes will require significant up-front investments. There is consensus that the authors make a strong case that these investments will result in savings and more effective programs, however, to realize these savings will require a longer term view of investment by national governments and donors. As one contributor summarized “The difficult decisions that would likely need to be made to adjust existing policies and programs and to adopt and implement new ones would require substantial buy-in and firm commitment from the highest, and across all levels of decision making in countries. It will also require substantial changes to the way many donor and international organizations work, which often go to countries with preconceived ideas of which interventions should be prioritized.” It is therefore critical to reach beyond technical experts to include policy makers, donors, and other stakeholders in countries to ensure that further development of the tool responds to their needs.

Contributors recognize that vitamin A is a good starting point but that for the tool to be applicable, it needs to address a broader array of nutrition interventions. Adding the full suite of micronutrient interventions will already be an improvement, but there is a strong call to bring together all nutrition interventions into the model, particularly in light of the move for more comprehensive nutrition programs. As one commentator stated, “Given the complexity of nutrition programming, we need tools that handle the entire gamut of nutrition interventions—not one micronutrient at a time.” One contributor noted that the model will have increased value if it can address risks of excessive intake (particularly of vitamin A) as well as optimizing for adequacy.

It is recommended that future refinements of the model draw on lessons learned from other optimization tools, such as HIV Optimization & Analysis Tool (OPTIMA) for HIV, and OneHealth for child mortality. Further alignment with other tools will likely increase uptake of this model.

The inaugural Global Nutrition Report published in 2014 notes that “In addition to data gaps on progress [toward achieving World Health Assembly nutrition targets], intervention coverage, and financial tracking, there are important gaps in data on food consumption, program costs, low birth weight, micronutrient status, capacity to scale up interventions, and impact” and declares that “Nutrition needs a data revolution.” ³ But such a “data revolution” is only useful if it will lead to better design, implementation, and quality improvement in nutrition programs. The authors and commentators point out the limitations and the challenges of the model that is presented here. However, it is an important step forward in demonstrating how investing in a data revolution for nutrition can lead us to an “optimization revolution” and match our programming to best fit the needs of populations at risk of malnutrition.