{"paper_id":"4764be9e-da06-43fd-8965-eb232d87f9f5","body_text":"Optimal Allocation of Antenatal and Young Child Nutrition Interventions: An Individual-Based GBD-Calibrated Microsimulation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Optimal Allocation of Antenatal and Young Child Nutrition Interventions: An Individual-Based GBD-Calibrated Microsimulation Alison Bowman, Sylvia Lutze, James Albright, Nathaniel Blair-Stahn, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4651040/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract Background: Undernutrition remains a global crisis and is a focus of Sustain Development Goals. While there are multiple known, effective interventions, complex interactions between prevention and treatment and resource constraints can lead to difficulties in allocating funding. Simulation studies that use in silico computer simulation can help illuminate the interactions between interventions and provide insight into the cost-effectiveness of alternative packages of options. Methods: We developed an individual-based microsimulation model to test a range of antenatal and child nutrition interventions (iron and folic acid, multiple micronutrients, balanced energy protein supplementation, treatment for severe acute malnutrition, treatment for moderate acute malnutrition, and wasting prevention with small-quantity lipid-based nutrient supplements) and an analytic approach to processes the results of the microsimulation and provide the optimal intervention funding allocation for a given budget size. Results: The reallocation of the baseline budget to minimize DALYs resulted in first funding multiple micronutrients to their maximum coverage, and then funding treatment for severe acute malnutrition. Relative to the baseline allocation, the reallocation optimized to minimizing DALYs resulted in 592,000 fewer annual DALYs. For budgets larger than the baseline, our model recommended funding first targeted moderate acute malnutrition treatment, second universal moderate acute malnutrition treatment, third wasting prevention with small-quantity lipid-based nutrient supplements, and fourth balanced energy protein supplementation. Conclusions: Our simulation is a novel model for estimating optimal allocation of spending on antenatal and child health nutrition interventions. We utilized a detailed health model that accounts for the interaction between preventive and therapeutic approaches. Our results show that an optimized reallocation of current spending on multiple micronutrients and treatment for severe acute malnutrition can substantially improve pregnancy-related and child health without additional funding. Additionally, we provide direction for how to best allocate expanded budgets to maximize impact. Maternal and child health simulation model health policy optimal allocation antenatal care young child feeding Figures Figure 1 Figure 2 Figure 3 Figure 4 Background Undernutrition remains a global crisis, contributing significantly to the burden of disease and hindering progress toward the Sustainable Development Goal (SDG) to end hunger (1). This persistent challenge demands urgent action and investment in proven interventions to address undernutrition during pregnancy and childhood. Despite a range of known effective interventions, coverage remains insufficient and unevenly distributed, particularly in regions with the greatest need (2). Two key areas where current practices lag are antenatal supplementation and the prevention and treatment of childhood acute malnutrition (AM). While iron and folic acid (IFA) supplementation is the standard of care in many settings, the World Health Organization (WHO) advocates for research on IFA-containing multiple micronutrient (MMN) supplements (3) and balanced energy and protein (BEP) supplementation for undernourished pregnant people (World Health Organization, 2016). Notably, there is suggestion that BEP supplementation targeted to undernourished pregnancies (in contrast to undernourished populations) may be a cost-effective strategy (5). Furthermore, WHO recently issued new guidance on the prevention and treatment of AM among children under 5 (6). This guidance provides recommendations for the treatment of moderate acute malnutrition (MAM) in a targeted fashion in addition to the prior recommendations for treating severe acute malnutrition (SAM). It additionally issues new recommendations for the prevention of AM with strategies, including the consideration of medium- or small-quantity lipid-based nutrient supplementation (MQ-LNS or SQ-LNS), particularly in contexts of high food insecurity. Even with the recognized need for scaling up nutrition interventions, resource constraints and the complexity of maximizing health impact with limited budgets pose significant challenges (7). While prevention is generally more effective (8), resource allocation decisions must consider various factors, including intervention costs, population reach, and potential interactions between different interventions. In silico computer models offer a valuable tool for guiding decision-making by simulating intervention effectiveness and cost-effectiveness. Existing models like the MMS Cost Benefit tool (9), FACET4SNF (10), CMAM costing tool (11), and WBCi (12) provide insights into specific interventions, while MINIMOD focuses on micronutrient interventions (13). The Optima Nutrition model (14), utilizing LiST (15), offers a broader analysis but has limitations regarding MAM treatment and the inability to test interventions independently. The development of additional models that can incorporate targeted MAM treatment and evaluate optimal spending allocations could enhance decision-making by providing robust evidence across multiple platforms. In particular, conclusions that are robust across multiple models may reduce concerns regarding structural uncertainty (16). In this interest, we developed an individual-based microsimulation model using Global Burden of Disease (GBD) 2021 study data to estimate the health impact of several nutrition-related interventions and paired it with an allocative efficiency analysis to determine allocation of intervention spending to optimize impact for a specified budget size(s). In this paper, we provide an overview of our simulation and include results specific to Ethiopia as an illustrative example. A note on language in this paper: Our team acknowledges that not all people who get pregnant or give birth are women, and we therefore strive to use more inclusive language that encompasses all identities without othering individuals. We are also aware that women often face unique challenges and focused work on women and girls is essential to improving health for all. The accepted language in literature is often women-centric such as “maternal health” or “women of reproductive age”. Our team advocates for the use of language inclusive to all people that also honors the unique experiences and needs of women. Therefore, throughout this paper we try to use gender inclusive language such as “pregnancy-related health” or “women and birthing people of reproductive age”. More information and a full glossary can be found in appendix 1, additional file 1. Methods Our model consists of two main components: (1) an individual-based microsimulation model of health events, and (2) an allocative efficiency analytic model that processes the results of the microsimulation model and outputs the optimal intervention funding allocation for a given budget size. Baseline health model We utilize Vivarium (17), a mature, open-source, Python-based simulation framework for our baseline health model. Published examples of models utilizing Vivarium can be found elsewhere (5,18,19). For this application, our model consists of a closed cohort of simulated individuals that we track across discrete time steps. Simulated individuals are assigned various attributes (such as age, sex, risk factor exposures, disease status, and vital status) that evolve over time and influence their trajectory through the simulation. Specifically, at each time step, simulated individuals are subject to some probability of a disease event (non-infected individuals may become infected, infected individuals may recover or die), which is modified by their other assigned attributes such as age, sex, and risk factor exposures. The model is calibrated such that individual-level heterogeneity reflects the appropriate magnitude across specific attributes while averages across simulated individuals reflect appropriate population-level statistics. The primary data source for our model was the GBD 2021 study, which estimates mortality and disability at the location-, year-, sex-, and age-specific level across hundreds of diseases, injuries and risk factors (20–24). Each timestep spent affected by a morbidity-causing condition results in accumulation of years lived with disability (YLDs) in accordance with the disability weight (DW) of that condition utilized in the GBD study. Years of life lost (YLLs) are accumulated in accordance with the theoretical minimum risk life expectancy (TMRLE) specific to a simulated individual’s age at the moment of death as informed from the GBD study. Figure 1 represents all modeled components in our baseline health model and the interactions between them, which are discussed in more detail in the following sections. We separated our baseline health model into two distinct simulated population groups: pregnant women and birthing people, and children under 5. Our simulation utilizes Monte Carlo methods to propagate parameter uncertainty throughout the model, which is covered in more depth in appendix 7, additional file 1. Pregnancy simulation The population structure, divided into five-year age groups, of the simulated cohort of pregnancies is informed from demographic estimates of women and birthing people of reproductive age (10–54 years) paired with estimates of age-specific fertility rates from the GBD study at a location- and year-specific level. Each simulated individual begins the simulation on day zero of their pregnancy in our model. Possible pregnancy outcomes include live birth, stillbirth, and abortion/miscarriage, with relative frequencies informed from GBD estimates. Infant sex is determined according to estimates of live births by sex from the GBD study. Pregnancy duration is determined according to the location- and sex-specific gestational age distribution of live births in GBD for both live births and stillbirths. For pregnancies that result in abortion/miscarriage, duration is determined according to a uniform distribution between 6 and 24 weeks. Each simulated individual is also assigned a continuous hemoglobin concentration and a dichotomous exposure for pre-pregnancy BMI above or below 18.5, each informed from GBD risk exposure estimates. Hemoglobin and BMI exposures are correlated to one another as informed from the Woman First trial (25) (more details in appendix 2, additional file 1). We advanced the simulation clock in increments of seven-day time steps and adjusted simulant ages accordingly. We used pregnancy-specific hemoglobin threshold values for anemia and simulants’ hemoglobin values to assign severity-specific anemia exposures and YLDs due to anemia accrued according to the corresponding disability weight (more details in appendix 2, additional file 1). Simulated birth events occur when the simulation time clock reaches the end of a simulant’s assigned pregnancy duration. At the moment of birth, simulants experience incident non-fatal or fatal cases of pregnancy-related disorders according to the population-level age- and location-specific probability from GBD that is further modified by their hemoglobin concentration at birth. Pregnancy-related disorders in this model are defined as all conditions resulting in a loss of health tied to pregnancy, birth, or postpartum complications. A full list of the included health conditions can be found in appendix 2, additional file 1. Incident, non-fatal pregnancy-related disorder causes accumulate YLDs as estimated in the 2021 GBD study. In a similar manner to pregnancy-related disorders, simulants also may experience incident cases of postpartum hemorrhage at birth with a likelihood modified by their hemoglobin exposure. For simulants that experience incident cases of postpartum hemorrhage, we applied a corresponding decrease on their postpartum hemoglobin level. We assigned infant birthweight according to the joint distribution with gestational age at birth (equivalent to pregnancy duration) from GBD and further correlated with joint antenatal anemia and pre-pregnancy/first trimester BMI exposure. Details on the magnitude and data sources of these effects can be found in appendix 2, additional file 1. Simulants are followed for anemia morbidity in the pregnancy simulation for an additional six weeks following birth before exiting the simulation. No background mortality due to causes other than pregnancy-related disorders are considered in the pregnancy simulation. Childhood simulation We initialized the simulated population of children under 5 with the birth events in the pregnancy simulation. Each live birth that occurs in the pregnancy simulation is initialized into the simulation on day zero of their life, with values for sex, gestational age at birth, and birthweight as determined in the pregnancy simulation. Our model progresses in time steps of four days, and we track ages within the age groups of 0–6 days, 7–28 days, 1–5 months, 6–11 months, 12–23 months, and 2–4 years. Assigned gestational age and birthweight exposures affect mortality due to causes affected by the risk factor in the GBD 2021 study (21) for the first 28 days of life, in a manner adapted from the GBD risk effect model described in appendix 3, additional file 1. Starting at 28 days of life, we assigned four-category exposure values for wasting (based on weight-for-height z-scores; WHZ), stunting (based on height-for-age z-scores; HAZ), and underweight (based on weight-for-age z-scores; WAZ). For all three metrics, we included exposure categories of severe (z-score < -3), moderate (-3 < z-score < -2), mild (-2 < z-score < -1), and unaffected (z-score > -1) with exposure prevalence informed from 2021 GBD study estimates. We additionally subdivided the moderate wasting exposure category into two substates: WHZ between -3 and -2.5 and WHZ between -2.5 and -2, in order to support the targeted MAM treatment intervention in our model. The relative exposure of each of these substates is informed from the most recently available Demographic Health Survey (DHS) data for the modeled location pooled across age groups and sexes (26), and relative morbidity and mortality risk of each substate is derived from the GBD 2021 study estimates. We assigned each simulant a fixed stunting percentile throughout life such that a stunting exposure value may change as simulants age into the next age group with a different population-level exposure distribution, but the percentile within the age-specific population will not change. We model a dynamic transition model of child wasting that is calibrated to the GBD exposure distribution and wasting state-specific mortality rates, estimates of wasting incidence rates from longitudinal cohort studies that tracked child anthropometry in low- and low-middle-income countries (details in appendix 4, additional file 1), and observed recovery rates from MAM and SAM in the ComPAS trial (27) under the assumption of a steady state equilibrium (more details in appendix 4, additional file 1). This calibration allows us to estimate average recovery rates from MAM and SAM states among populations without access to treatment, which are generally not available in the literature (with some exceptions, such as James et al., 2016). Notably, wasting transition rates do not vary by moderate wasting substate exposure. We assume no correlation between stunting and wasting exposures. Underweight exposures are assigned according to observed location-, age-, and sex-specific correlation with four-category wasting and four-category stunting obtained from the DHS and are updated dynamically as simulant age, wasting exposure, and/or stunting exposure evolves throughout the simulation (more details in appendix 4, additional file 1). Infant birthweight influences population-level stunting exposures used to determine individual-level stunting exposures in accordance with evidence from McGovern (2019). Infant birthweight influences wasting state at 28 days of life in accordance with observed data from DHS but does not influence wasting transition rates thereafter. More details on the impact of birthweight on child growth failure can be found in appendix 5, additional file 1. We model incidence (and associated morbidity) and mortality due to diarrheal diseases, lower respiratory infections, malaria, and measles from ages 28 days to 5 years. Incidence and mortality rates due to these causes are modified by stunting, underweight, and wasting exposures (including moderate wasting substate exposures) in accordance with GBD estimates of child growth failure effects, described elsewhere (21). Briefly, GBD methods account for overlap between these indices of child growth failure to ensure impacts on morbidity and mortality are not overestimated. We additionally model morbidity and mortality due to protein energy malnutrition, which is entirely informed by wasting exposure among simulants aged 28 days to 5 years. Finally, we model background mortality due to all causes other than those directly modeled for all ages in our child simulation. Intervention models Our model considers three antenatal supplementation intervention products, including iron and folic acid (IFA), multiple micronutrients (MMN), and balanced energy protein (BEP) supplementation, in addition to three child nutrition interventions, including treatment for SAM, treatment for MAM, and wasting prevention with SQ-LNS. We assume that antenatal supplementation products are distributed at routine antenatal care visits and taken for a duration of six months. We assume that BEP is provided in addition to MMN for pregnancies with a pre-pregnancy/first trimester BMI of less than 18.5 and otherwise antenatal supplementation products are mutually exclusive at the individual level. We additionally assume that BEP + MMN are only provided to low BMI pregnancies if MMN is also provided to adequate BMI pregnancies. Antenatal supplementation products affect the probability of stillbirth, gestational age at birth, and birthweight, with effect sizes shown in Table 1. Intervention mean differences in antenatal hemoglobin are applied at the individual level at eight weeks of gestation. We assume that reduction in stillbirth outcomes associated with interventions results in increases in live birth outcomes, with no changes in other birth outcomes such as abortion, miscarriage, or ectopic pregnancy. For intervention effects on preterm birth, we calculated location- and year-specific population mean differences that resulted in the relative risks reported in table 1 and applied the mean differences to our continuous measure of gestational age at birth at the individual level in our simulation. Notably, for the effect of MMN on preterm birth, we estimated a population mean difference conditional on gestational age at birth +/- 32 weeks, such that there was a larger effect among those born at less than 32 weeks’ gestation than those born at greater than 32 weeks’ gestation so that both the effects on preterm birth (<37 weeks) and very preterm birth (<32 weeks) were replicated. Table 1. Antenatal intervention effects Outcome Iron folic acid (IFA), relative to no supplementation Multiple micronutrients (MMN), relative to IFA Balanced energy protein (BEP), relative to MMN Antenatal hemoglobin, grams per liter MD = +7.8 (4.08, 11.52) (30) MD = +0 MD = +0 Stillbirth RR = 1 RR = 0.91 (0.71, 0.93) (30) RR = 0.39 (0.19, 0.80) (31) Birthweight, grams MD = +57.73 (7.66, 107.79) (32) MD = +45.16 (32.31, 58.02) (5,33) MD = +66.96 (13.13, 120.78) (34)* Preterm birth (<37 weeks) RR = 0.90 (0.86, 0.95) (35) RR = 0.91 (0.84, 0.99) (36) RR = 1 (34) Very preterm birth (<32 weeks) - RR = 0.81 (0.71, 0.93) (33) Table 1 Footnote: MD: mean difference; RR: relative risk *Effect specific to subpopulation of undernourished women and birthing people In our model, the SAM treatment intervention is administered to incident cases of SAM, as defined by WHZ < -3, between 6 and 59 months of age. In universal protocol of the MAM treatment intervention, treatment is administered to incident cases of MAM, as defined by WHZ between -2 and -3, between 6 and 59 months of age. The targeted protocol of the MAM treatment intervention is administered to those eligible for the universal MAM treatment intervention who also satisfy at least one of the following criteria: (a) less than 24 months of age, (b) WHZ between -2.5 and -3, or (c) WAZ less than -3. Treatment protocol and time-to-recovery for both the MAM and SAM interventions are informed from the intervention arm of the ComPAS trial (27). We assume that a fraction of cases does not respond to treatment and recover according to the untreated recovery rates in our wasting model. In our model, SQ-LNS supplementation begins at six months of age for a duration of 12 months for all infants. SQ-LNS supplementation decreases the prevalence of moderate and severe stunting and increases the prevalence of no stunting in accordance with effects obtained from study authors of an individual participant meta-analysis on the intervention (37). We model effects of SQ-LNS supplementation on transition rates from no wasting to mild wasting, mild wasting to moderate wasting, and moderate wasting to severe wasting with no effects on wasting recovery rates. These effect sizes are calibrated to replicate the wasting prevalence ratios from the same source as stunting effect sizes, as described in appendix 6, additional file 1. We assume that the initial point of care for all child nutrition interventions is monthly community management of acute malnutrition (CMAM) screenings. We assume that children are assessed for acute malnutrition and referred for treatment as appropriate at these meetings and that SQ-LNS product counseling and distribution occurs at these meetings as well, as has been performed in trial settings (38). Notably, this assumption dictates that for a population of 100 children 6–59 months, five of whom are afflicted with SAM, all 100 children would be screened in order to reach and treat the five afflicted children for SAM. Optimization structure Scenario layout We ran the health model for every possible combination of our modeled interventions. For instance, if we only considered IFA and SAM treatment interventions, we would run the health model under the following conditions: 0% coverage IFA, 0% coverage SAM treatment 100% coverage IFA, 0% coverage SAM treatment 0% coverage IFA, 100% coverage SAM treatment 100% coverage IFA, 100% coverage SAM treatment We rely on the key assumption that individuals within our simulated population are independent of one another in that the health status of one individual does not impact any others. This allows us to assume that the population health status under 50% coverage of a given intervention can be equivalently represented as the average between the population health status under 100% coverage of the intervention and the population health status under 0% coverage. With all seven of our modeled interventions, after excluding illogical combinations (such as simultaneous presence of universal MAM treatment and targeted MAM treatment interventions), this amounts to a total of 48 scenarios in addition to one baseline scenario. In each of these scenarios, we record counts of deaths, stillbirths, YLLs, YLDs, incident wasting cases, person-time, and intervention administration counts that occurred in our simulated populations. Deaths, YLLs, and YLDs are stratified by pregnancy and child populations, and person-time counts are stratified by stunting state and age group for children. Across modeled scenarios, we utilize common random numbers (39) at the individual simulant level such that in the absence of intervention influence, each simulant would undergo the exact same trajectory in one scenario as the others, reducing noise across scenarios in our model. Optimization function and constraints We use the scipy.optimize package in Python (40) to perform our allocative efficiency analysis. Inputs to our optimization function include the recorded health outcomes for each modeled scenario and the calculated cost of each scenario (obtained by multiplying the recorded intervention administration counts for each scenario by the intervention costs). The optimization objective is to find the fractional combination of scenarios that maximizes/minimizes the specified health quantity (ex: minimize DALYs or maximize person-time) under the following constraints: Total cost cannot exceed user-defined budget envelope Total intervention-specific coverage cannot exceed specified saturation coverage limits The two intervention coverage rules described below are followed Intervention coverage rule one: all interventions share a “common care-seeking propensity” that dictates if population coverage of intervention A and B are both 50%, we assume the same 50% of the population is covered by interventions A and B and that the remaining 50% of the population has access to neither intervention A nor B. Intervention coverage rule two: we assume that ANC intervention products could not be “targeted” to a subpopulation based on their coverage of child nutrition interventions or vice versa. For instance, if saturation coverage of ANC products is greater than that of child nutrition interventions, a solution that allocates IFA to the portion of the population with access to child nutrition interventions and MMN to the portion of the population without access to child nutrition interventions is not allowed. Following these and the previously stated assumption of simulant independence, we calculate overall cost, health impact, and intervention coverage as outputs of our optimization function for a single budget size. Illustrative example parameters As an illustrative example, we ran the model for Ethiopia in 2021. We utilized a simulated population size of 1,600,000 pregnancies and informed the simulated child population from the resulting birth outcomes from a random sample of 400,000 of these pregnancies. These population sizes were selected as the minimum size that achieved stability in the difference between health outcomes across modeled scenarios among the antenatal and child populations, respectively. We scaled results from the pregnancy and child baseline health models by a factor of 3.5 and 14, respectively, to reflect the total estimated number of births in Ethiopia in 2021. We ran our simulation model for 20 Monte Carlo uncertainty draws (see appendix 7, additional file 1). Values for baseline intervention coverage, maximum/saturation intervention coverage, and intervention costs used for this model run are displayed in table 2. Intervention costs used in our model represent product costs only and were obtained from the UNICEF Supply Catalogue for IFA, MMN, BEP, and SQ-LNS (UNICEF Supply Catalogue, n.d.) interventions and the ComPAS trial (Bailey et al., 2020) for MAM and SAM treatment interventions. Costs from the UNICEF Supply Catalogue were obtained in November of 2023. We assumed there were no product costs associated with CMAM screenings. Table 2. Intervention coverage and cost assumptions Intervention Baseline coverage Saturation coverage Cost (2023 USD) IFA 60.2% (95% CI: 48.7, 72.6) (Personal communication of work by Nat Henry at CIFF) 75.7% (ANC1; GBD 2021) 2.27 per supplemented pregnancy MMN 0%, assumption 75.7% (ANC1; GBD 2021) 3.47 per supplemented pregnancy BEP 0%, assumption 75.7% (ANC1; GBD 2021) 40.28 per supplemented pregnancy SAM treatment 48.8% (95% CI: 37.4, 60.4) (41) 70%, assumption 41.84 per treated child MAM treatment 15% (95% CI: 10, 20), assumption. Baseline coverage assumed to be universal implementation of MAM treatment intervention 70%, assumption (applies to both universal and targeted implementations of MAM treatment intervention) 29.70 per treated child (applies to both universal and targeted implementations of MAM treatment intervention) SQ-LNS 0%, assumption 70%, assumption 35.75 for 12 months supplementation CMAM screening 48.8% (95% CI: 37.4, 60.4), assumed to be maximum of SAM treatment, MAM treatment, and SQ-LNS baseline coverage 70%, assumption 0 per child screened Table 2 Footnote: IFA: iron and folic acid; MMN: multiple micronutrients; BEP: balanced energy protein; SAM: severe acute malnutrition; MAM: moderate acute malnutrition; SQ-LNS: small quantity lipid-based nutrient supplementation; CMAM: community management of acute malnutrition; USD: United States dollars We performed our allocative efficiency analysis at the estimated baseline budget size in addition to 25 evenly spaced increments between zero and the spending required to reach saturation coverage of all interventions. We performed allocative efficiency analysis at the draw level as well as for the average result across all draws from the underlying health model. Results From our assumed intervention coverage values and intervention costs, we estimated total annual spending on our modeled interventions to be 34.4 (95% UI: 28.3, 42.4) million USD in Ethiopia, with 5.6 (95% UI: 4.8, 6.7) million allocated to IFA, 14.7 (95% UI: 11.2, 20.4) million allocated to SAM treatment, and 14.1 (95% UI: 8.8, 19.7) million allocated to universal MAM treatment (Figure 2). Our model estimated 18.0 (95% UI: 15.1, 21.7) million annual DALYs among pregnancies and children under 5 at baseline due to the modeled childhood and pregnancy-related health conditions. This was estimated to be 986,000 (95% UI: 535,000, 1,588,000) DALYs fewer than the 19.0 (95% UI: 15.8, 22.9) million annual DALYs under the counterfactual scenario of zero spending on our modeled interventions. We found that reallocation of the baseline budget to minimize DALYs resulted in funding MMN to its maximum coverage level of 75.6%, followed by investment of the remaining budget into treatment for SAM (which attained 62.7% population coverage). When the baseline budget was reallocated to maximize child time spent not stunted, the budget was likewise first spent maximizing MMN, but then the remaining funds were allocated to SQ-LNS supplementation alone (which attained 16.4% population coverage). We maximize child time spent not stunted rather than minimizing child time stunted because minimizing child time stunted can lead to optimizing for fewer children alive. These results fit with our modeling parameters since SQ-LNS impacts stunting and is a preventive treatment, while treatment for SAM directly impacts wasting, which is more deadly for young children. Figure 2 panel A displays the baseline and reallocated intervention-specific funding, and figure 2 panel B displays the baseline and reallocated intervention-specific coverage levels. Relative to the baseline allocation, the reallocation optimized to minimize DALYs resulted in 592,000 fewer annual DALYs and the reallocation optimized to maximize child time spent not stunted resulted in 187,000 fewer annual DALYs among pregnancies and children under 5. This amounted to a 60% and 19% increase in DALYs averted relative to zero spending, compared to baseline allocation, for the reallocation optimized to DALYs and the reallocation optimized to child time spent not stunted, respectively. Figure 3 displays annual DALYs under each of these scenarios. When we performed this allocative efficiency analysis for successively larger budget sizes and optimized to minimize DALYs, our model found that MMN supplementation alone should be prioritized until MMN reaches its maximum coverage level, followed by the SAM treatment intervention as budget size allows. Starting at annual budget sizes over 35 million USD, our model recommends investment in the targeted MAM treatment followed by universal MAM treatment, SQ-LNS, and finally targeted BEP supplementation at the largest budget sizes. The annual budget required to achieve maximum impact (reach saturation coverage of most impactful interventions) was 196 (95% UI: 165, 225) million USD (approximately 5.7 times more than we estimate is currently spent). Figure 4 panel A displays the baseline and reallocated intervention-specific expenditures, and figure 4 panel B displays the baseline and reallocated intervention-specific coverage levels across increasing budget sizes. Panel C in Figure 4 displays the annual DALYs averted relative to zero spending on any of the modeled interventions at each successive budget size, with a total of 2.2 (95% UI: 1.4, 3.0) million annual DALYs averted at the maximum impact level. Panel D of Figure 4 displays the incremental cost-effectiveness ratio (ICER) in USD per DALY averted relative to zero spending on any of our modeled interventions across increasing budget sizes; note that all ICERs represented in this figure are positive with positive health gains (DALYs averted) and positive incremental costs (the northeast quadrant in as described by (42)). Notably cost-effectiveness decreases (ICER increases) with increasing budget as the most cost-effective interventions are prioritized at lower budget sizes, with all the ICER at all budget sizes at or below 88.94 (95% UI: 65.10, 154.04) dollars per DALY averted. Notably, without simultaneous coverage of MAM treatment or SQ-LNS interventions, 23 (95% UI: 19, 27) million USD was required to maintain saturation coverage (70%) of the SAM treatment intervention. However, the full scale-up of MAM treatment and SQ-LNS interventions to saturation coverage (70%) reduced the operating cost of maintaining saturation coverage of the SAM treatment intervention by approximately half to 12 (95% UI: 8.3, 16) million USD annually. When analyzed at the draw-level, 100% of draws resulted in intervention priority ordering starting with MMN, followed by SAM treatment and targeted MAM treatment. However, the relative priority ordering of the universal MAM treatment, SQ-LNS, and BEP interventions had significant draw-level variation. Specifically, BEP was prioritized behind universal MAM treatment and SQ-LNS in only 65% of draws, and universal MAM treatment was prioritized ahead of SQ-LNS in only 70% of draws (see appendix 7, additional file 1 for more information on draws). Discussion Our simulation provides a new model to estimate optimal allocation of spending on antenatal and child health nutrition interventions utilizing a detailed health model informed from GBD study estimates that accounts for the interaction between preventive and therapeutic approaches. Our results show that an optimized reallocation of current spending on these products can substantially improve pregnancy-related and child health without additional funding and provides direction for how to best allocate expanded budgets to maximize impact. Our model provides additional benefit relative to existing options such as the MMS Cost Benefit Tool, the CMAM Costing Tool, the FACET4SNF model, WBCi, and MINIMOD in that it integrates both antenatal and child micro- and macronutrition interventions and performs allocative efficiency analysis of such interventions. Our underlying health model utilizes a similar framework as the LiST model that is utilized by Optima Nutrition, which, to the best of our knowledge, is the only other allocative efficiency model that evaluates similar interventions. However, our model differs from LiST in several ways. Firstly, LiST is a compartmental rather than microsimulation model and utilizes a variety of data sources described elsewhere ( 15 ) to inform baseline population and mortality dynamics, whereas our model primarily utilizes GBD study data. While our models generally utilize similar data sources for intervention effect estimations, they differ with respect to the derivation of MAM and SAM treatment intervention impacts on recovery rates, with our model ultimately utilizing a slightly lower effect of MAM treatment on MAM recovery and slightly higher effect of SAM treatment on SAM recovery. Further, unlike our model, LiST applies intervention effects to subsets of the population only; for instance, the effect of SQ-LNS is applied to the food-insecure portion of the population only ( 43 ). With regard to the impact of the BEP supplementation intervention, our model benefits from a correlation of the population eligible for BEP (low BMI pregnancies) with low birthweight outcomes, which is not the case in the LiST and Optima Nutrition models. Unlike our model that accounts for correlation between wasting, stunting, and underweight exposures in the attribution of their effects on morbidity and mortality, LiST does not report to account for this in the estimation of risk associated with wasting and stunting in their model. Notably, intervention protocols and definitions may slightly differ between our models. LiST also supports several interventions not included in our model. Our model benefits from high-quality and detailed age-, sex-, location-, and year-specific estimates from the 2021 GBD study and framework. A strength of this data source is how it accounts for correlation between wasting, stunting, and underweight exposures in its estimation of their impacts in morbidity and mortality; this avoids overestimation of the aggregate effects. Another strength is the continuous joint distribution of birthweight and gestational age at birth. Our modeling approach allows us to represent non-linear interactions between intervention combinations in that the presence of prevention interventions may remove the need for treatment interventions at the individual level. Finally, our model supports MAM treatment intervention as an independent program from SAM treatment that allows for testing of optimal coverage levels of these programs and additionally supports a targeted implementation of the MAM treatment intervention in a manner intended to reflect the recent WHO guidelines on the matter. Our model additionally supports BEP supplementation targeting to undernourished pregnancies rather than undernourished populations, which is expected to reach a greater proportion of those who may benefit from the intervention. Our model is limited by the intensive computational resources required to run the microsimulation for our underlying health model, which we have performed on a high-performance computing cluster. Additionally, our model is limited in that it relies on assumptions and/or uncertain estimates of baseline intervention coverage and therefore baseline spending allocation. Furthermore, the intervention product costs used in this analysis are not necessarily reflective of the true costs of implementing intervention programs at scale. Our baseline health model is limited in that it does not consider seasonality in wasting burden and relies on classification of acute malnutrition using WHZ alone rather than WHZ, mid-upper arm circumference (MUAC), and the presence of edema combined. Furthermore, our targeted MAM treatment intervention is inspired by the recent WHO guidelines but does not consider all recommended criteria for determining which MAM cases should receive treatment. Likewise, our model of BEP supplementation targeting is not exactly aligned with recommendations in the WHO guidelines. We do not model SQ-LNS intervention effects on anemia or vitamin A deficiency ( 44 ) in our model, nor do we consider a targeted implementation of the intervention such as that suggested in the WHO guideline. Finally, while our model may not represent all causal pathways by which our modeled interventions affect morbidity and mortality, as the absence of evidence between a given risk/outcome pair does not imply the absence of a causal association. Notably, we also do not consider feedback between wasting and stunting exposures (wasting leading to future stunting or vice versa) nor do we consider any causal impacts of infectious disease episodes leading to future wasting and/or stunting exposures, despite some evidence for such associations ( 45 , 46 ). Our next steps for this model include the development of an interactive, online tool similar to the other models referenced here. This will allow user customization of the optimization and a more in-depth review of our results. Additionally, we will integrate a targeted SQ-LNS intervention implementation and perform sensitivity analyses around intervention cost assumptions, optimization measure, and location. We will also explore the robustness of intervention priority and evaluate the “next best” options. We also plan to extend this work for use in capacity planning for CMAM programs as they are adapted and/or expanded to align with updated guidelines on acute malnutrition prevention and treatment. Future extensions of this model may also include adaptation to support non-linear intervention cost and coverage functions and/or to support subnational-level allocative efficiency analysis within a given country. Our model enables estimation of optimal spending allocations across several antenatal and child nutrition interventions for a specified budget size using high-quality GBD study estimates and may be compared to results from the existing LiST and/or Optima Nutrition models as an exploration of structural uncertainty in the pregnancy-related and child nutrition intervention optimization space. We hope our model can aid policy makers in decisions of how to best allocate future spending on pregnancy-related and child health nutrition interventions, especially as they may be planning for integration of the recent WHO guidelines on the treatment and prevention of childhood acute malnutrition. Conclusion Our simulation offers a novel approach to optimizing spending on antenatal and child health nutrition interventions, leveraging the GBD study to create a detailed health model that can more accurately account for the interaction between preventative and therapeutic approaches. By reallocating funding to MMN and SAM treatment, 592,000 DALYs could be averted annual without further financial investment. The addition of this new simulation allows for greater confidence in modeling results overall, and new information for key decision makers to consider. Abbreviations Abbreviation Definition AM Acute malnutrition BEP Balanced energy protein CMAM Community management of acute malnutrition DALY Disability-adjusted life year DHS Demographic health survey DW Disability weight GBD Global burden of disease study HAZ Height-for-age z-score ICER Incremental cost effectiveness ratio IFA Iron folic acid MAM Moderate acute malnutrition MMN Multiple micronutrient MQ-LNS Medium-quantity lipid-based nutrient supplement MUAC Mid-upper arm circumference SAM Severe acute malnutrition SDG Sustainable development goal SQ-LNS Small-quantity lipid-based nutrient supplement TMREL Theoretical minimum risk exposure level TMRLE Theoretical minimum risk life expectancy WAZ Weight-for-age z-score WHO World health organization WHZ Weight-for-height z-score YLD Years lives with disability YLL Years of life lost Declarations Ethics approval and consent to participate Not applicable Consent for publication Not applicable Availability of data and materials The datasets generated for use during the current study are available in the ihmeuw/vivarium_gates_nutrition_optimization and ihmeuw/vivarium_gates_nutrition_optimization_child repositories, the link for the pregnancy simulation data is: https://doi.org/10.5281/zenodo.11661268 and the link for the child simulation data is: https://doi.org/10.5281/zenodo.11661095 Competing interests The authors declare that they have no competing interests. Funding This research was funded by the Bill and Melinda Gates Foundation. The funders had input in the high-level design of the model and specifically in the selection of interventions used here but were not involved in decision to publish or manuscript preparation. Authors' contributions AB, AF, and SL led the design of the simulation model and analysis of results. AB led the writing of the manuscript and figure generation. NB, AP, and SK also assisted in data analysis, figure generation, and design questions. CK led the project management. JA, HJ, and PN all contributed to the software development of the simulation model code. RM led the software development team and contributed to the model design. All authors read and approved the final manuscript. Acknowledgements Our team would like to acknowledge the following groups: The Maternal, Newborn, Child Health & Nutrition team at the Bill and Melinda Gates Foundation, specifically Laura Lamberti, Rahul Rawat, and Chris Troeger, for their support in design decisions, and data access to the KI database. The Institute for Global Nutrition at UC Davis, specifically Kathryn Dewey, Christine Stewart, Christine McDonald, and Charles Arnold who shared feedback and expert opinions on model design, and data for the development of the SQ-LNS intervention. Indi Trehan for his expert opinion in the design of the SQ-LNS intervention. Navideh Noori for her expert opinion on model design, specifically for the in the design of the SQ-LNS intervention. All of the GBD team’s at IHME that created the robust data used here. In particular, we relied on data and expertise from the Neonatal and Child Health team and the Reproductive, Genitourinary, & Digestive Diseases team. References The State of Food Security and Nutrition in the World 2022 [Internet]. FAO; 2022 [cited 2022 Jul 29]. Available from: http://www.fao.org/documents/card/en/c/cc0639en Hong Nguyen P, Singh N, Scott S, Neupane S, Jangid M, Walia M, et al. Unequal coverage of nutrition and health interventions for women and children in seven countries. Bull World Health Organ. 2022 Jan 1;100(1):20–9. World Health Organization. WHO antenatal care recommendations for a positive pregnancy experience Nutrition interventions update: Multiple micronutrient supplements during pregnancy [Internet]. 2020. Available from: https://iris.who.int/bitstream/handle/10665/333561/9789240007789-eng.pdf?sequence=1 World Health Organization. WHO recommendations on antenatal care for a positive pregnancy experience [Internet]. 2016. Available from: https://iris.who.int/bitstream/handle/10665/250796/9789241549912-eng.pdf?sequence=1 Young N, Bowman A, Swedin K, Collins J, Blair-Stahn ND, Lindstedt PA, et al. Cost-effectiveness of antenatal multiple micronutrients and balanced energy protein supplementation compared to iron and folic acid supplementation in India, Pakistan, Mali, and Tanzania: A dynamic microsimulation study. Myers JE, editor. PLOS Med. 2022 Feb 22;19(2):e1003902. World Health Organization. WHO guideline on the prevention and management of wasting and nutritional oedema (acute malnutrition) in infants and children under 5 years [Internet]. 2023. Available from: https://app.magicapp.org/#/guideline/noPQkE Heidkamp RA, Piwoz E, Gillespie S, Keats EC, D’Alimonte MR, Menon P, et al. Mobilising evidence, data, and resources to achieve global maternal and child undernutrition targets and the Sustainable Development Goals: an agenda for action. The Lancet. 2021 Apr;397(10282):1400–18. Ruel MT, Menon P, Habicht JP, Loechl C, Bergeron G, Pelto G, et al. Age-based preventive targeting of food assistance and behaviour change and communication for reduction of childhood undernutrition in Haiti: a cluster randomised trial. The Lancet. 2008 Feb;371(9612):588–95. Verney AMJ, Busch‐Hallen JF, Walters DD, Rowe SN, Kurzawa ZA, Arabi M. Multiple micronutrient supplementation cost–benefit tool for informing maternal nutrition policy and investment decisions. Matern Child Nutr. 2023 Oct;19(4):e13523. The Food Assistance Cost-Effectiveness Tool for Specialized Nutritious Foods [Internet]. Available from: https://foodaidquality.nutrition.tufts.edu/focus/cost-effectiveness-tools CMAM Costing Tool [Internet]. Available from: https://www.fantaproject.org/tools/cmam-costing-tool Holla-Bhar R, Iellamo A, Gupta A, Smith JP, Dadhich JP. Investing in breastfeeding – the world breastfeeding costing initiative. Int Breastfeed J. 2015 Dec;10(1):8. Brown KH, Engle-Stone R, Kagin J, Rettig E, Vosti SA. Use of Optimization Modeling for Selecting National Micronutrient Intervention Strategies: An Example Based on Potential Programs for Control of Vitamin A Deficiency in Cameroon. Food Nutr Bull. 2015 Sep;36(3_suppl):S141–8. Pearson R, Killedar M, Petravic J, Kakietek JJ, Scott N, Grantham KL, et al. Optima Nutrition: an allocative efficiency tool to reduce childhood stunting by better targeting of nutrition-related interventions. BMC Public Health. 2018 Dec;18(1):384. Walker N, Tam Y, Friberg IK. Overview of the Lives Saved Tool (LiST). BMC Public Health. 2013;13(Suppl 3):S1. Baldissera Pacchetti M. Structural uncertainty through the lens of model building. Synthese. 2021 Nov;198(11):10377–93. Vivarium [Internet]. Available from: https://vivarium.readthedocs.io/en/latest/ Haddock B, Pletcher A, Blair-Stahn N, Keyes O, Kappel M, Bachmeier S, et al. Simulated data for census-scale entity resolution research without privacy restrictions: a large-scale dataset generated by individual-based modeling. Gates Open Res. 2024 May 3;8:36. Kannan A, Tsoi D, Xie Y, Horst C, Collins J, Flaxman A. Cost-effectiveness of Vitamin A supplementation among children in three sub-Saharan African countries: An individual-based simulation model using estimates from Global Burden of Disease 2019. Horton S, editor. PLOS ONE. 2022 Apr 7;17(4):e0266495. Bhattacharjee NV, Schumacher AE, Aali A, Abate YH, Abbasgholizadeh R, Abbasian M, et al. Global fertility in 204 countries and territories, 1950–2021, with forecasts to 2100: a comprehensive demographic analysis for the Global Burden of Disease Study 2021. The Lancet. 2024 Mar;S0140673624005506. Brauer M, Roth GA, Aravkin AY, Zheng P, Abate KH, Abate YH, et al. Global burden and strength of evidence for 88 risk factors in 204 countries and 811 subnational locations, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021. The Lancet. 2024 May 18;403(10440):2162–203. Ferrari AJ, Santomauro DF, Aali A, Abate YH, Abbafati C, Abbastabar H, et al. Global incidence, prevalence, years lived with disability (YLDs), disability-adjusted life-years (DALYs), and healthy life expectancy (HALE) for 371 diseases and injuries in 204 countries and territories and 811 subnational locations, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021. The Lancet. 2024 Apr;S0140673624007578. Naghavi M, Ong KL, Aali A, Ababneh HS, Abate YH, Abbafati C, et al. Global burden of 288 causes of death and life expectancy decomposition in 204 countries and territories and 811 subnational locations, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021. The Lancet. 2024 Apr;S0140673624003672. Schumacher AE, Kyu HH, Aali A, Abbafati C, Abbas J, Abbasgholizadeh R, et al. Global age-sex-specific mortality, life expectancy, and population estimates in 204 countries and territories and 811 subnational locations, 1950–2021, and the impact of the COVID-19 pandemic: a comprehensive demographic analysis for the Global Burden of Disease Study 2021. The Lancet. 2024 Mar;S0140673624004768. Hambidge KM, Westcott JE, Garcés A, Figueroa L, Goudar SS, Dhaded SM, et al. A multicountry randomized controlled trial of comprehensive maternal nutrition supplementation initiated before conception: the Women First trial. Am J Clin Nutr. 2019 Feb;109(2):457–69. Ethiopian Public Health Institute (EPHI) [Ethiopia] and ICF. Ethiopia Mini Demographic and Health Survey 2019 [Dataset]. ETKR81DT.ZIP. Rockville, Maryland, USA: EPHI and ICF [Producers]. ICF [Distributor], 2021. 2021. Bailey J, Opondo C, Lelijveld N, Marron B, Onyo P, Musyoki EN, et al. A simplified, combined protocol versus standard treatment for acute malnutrition in children 6–59 months (ComPAS trial): A cluster-randomized controlled non-inferiority trial in Kenya and South Sudan. Tumwine JK, editor. PLOS Med. 2020 Jul 9;17(7):e1003192. James P, Sadler K, Wondafrash M, Argaw A, Luo H, Geleta B, et al. Children with moderate acute malnutrition with no access to supplementary feeding programmes experience high rates of deterioration and no improvement: Results from a prospective cohort study in rural ethiopia. PLOS ONE. 2016 Apr 21;11(4):e0153530. McGovern ME. How much does birth weight matter for child health in developing countries? Estimates from siblings and twins. Health Econ. 2019 Jan;28(1):3–22. Oh C, Keats E, Bhutta Z. Vitamin and Mineral Supplementation During Pregnancy on Maternal, Birth, Child Health and Development Outcomes in Low- and Middle-Income Countries: A Systematic Review and Meta-Analysis. Nutrients. 2020 Feb 14;12(2):491. Lassi ZS, Padhani ZA, Rabbani A, Rind F, Salam RA, Das JK, et al. Impact of Dietary Interventions during Pregnancy on Maternal, Neonatal, and Child Outcomes in Low- and Middle-Income Countries. Nutrients. 2020 Feb 19;12(2):531. Peña-Rosas JP, De-Regil LM, Gomez Malave H, Flores-Urrutia MC, Dowswell T. Intermittent oral iron supplementation during pregnancy. Cochrane Pregnancy and Childbirth Group, editor. Cochrane Database Syst Rev [Internet]. 2015 Oct 19 [cited 2024 Jan 19];2015(10). Available from: http://doi.wiley.com/10.1002/14651858.CD009997.pub2 Keats EC, Haider BA, Tam E, Bhutta ZA. Multiple-micronutrient supplementation for women during pregnancy. Cochrane Pregnancy and Childbirth Group, editor. Cochrane Database Syst Rev [Internet]. 2019 Mar 15 [cited 2024 Jan 19]; Available from: https://doi.wiley.com/10.1002/14651858.CD004905.pub6 Ota E, Hori H, Mori R, Tobe-Gai R, Farrar D. Antenatal dietary education and supplementation to increase energy and protein intake. Cochrane Pregnancy and Childbirth Group, editor. Cochrane Database Syst Rev [Internet]. 2015 Jun 2 [cited 2024 Jan 19];2015(6). Available from: http://doi.wiley.com/10.1002/14651858.CD000032.pub3 Li B, Zhang X, Peng X, Zhang S, Wang X, Zhu C. Folic Acid and Risk of Preterm Birth: A Meta-Analysis. Front Neurosci. 2019 Nov 28;13:1284. Gomes F, Askari S, Black RE, Christian P, Dewey KG, Mwangi MN, et al. Antenatal multiple micronutrient supplements versus iron‐folic acid supplements and birth outcomes: Analysis by gestational age assessment method. Matern Child Nutr. 2023 Jul;19(3):e13509. Dewey KG, Stewart CP, Wessells KR, Prado EL, Arnold CD. Small-quantity lipid-based nutrient supplements for the prevention of child malnutrition and promotion of healthy development: overview of individual participant data meta-analysis and programmatic implications. Am J Clin Nutr. 2021 Nov;114:3S-14S. Huybregts L, Le Port A, Becquey E, Zongrone A, Barba FM, Rawat R, et al. Impact on child acute malnutrition of integrating small-quantity lipid-based nutrient supplements into community-level screening for acute malnutrition: A cluster-randomized controlled trial in Mali. Persson LÅ, editor. PLOS Med. 2019 Aug 27;16(8):e1002892. Flaxman AD, Deason AW, Dolgert AJ, Mumford JE, Sorensen RJD, Eldrenkamp E, et al. Untangling uncertainty with common random numbers: a simulation study. Summer Simul Multi-Conf. 2017 Jul;(31):1–12. Virtanen P, Gommers R, Oliphant TE, Haberland M, Reddy T, Cournapeau D, et al. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods. 2020 Mar 2;17(3):261–72. Isanaka S, Andersen CT, Cousens S, Myatt M, Briend A, Krasevec J, et al. Improving estimates of the burden of severe wasting: analysis of secondary prevalence and incidence data from 352 sites. BMJ Glob Health. 2021 Mar;6(3):e004342. Klok RM, Postma MJ. Four quadrants of the cost-effectiveness plane: some considerations on the south-west quadrant. Expert Rev Pharmacoecon Outcomes Res. 2004 Dec;4(6):599–601. Tong H, Piwoz E, Ruel MT, Brown KH, Black RE, Walker N. Maternal and child nutrition in the Lives Saved Tool: Results of a recent update. J Glob Health. 2022 Dec 30;12:08005. Wessells KR, Arnold CD, Stewart CP, Prado EL, Abbeddou S, Adu-Afarwuah S, et al. Characteristics that modify the effect of small-quantity lipid-based nutrient supplementation on child anemia and micronutrient status: an individual participant data meta-analysis of randomized controlled trials. Am J Clin Nutr. 2021 Nov;114:68S-94S. Thurstans S, Sessions N, Dolan C, Sadler K, Cichon B, Isanaka S, et al. The relationship between wasting and stunting in young children: A systematic review. Matern Child Nutr [Internet]. 2022 Jan [cited 2022 Sep 15];18(1). Available from: https://onlinelibrary.wiley.com/doi/10.1111/mcn.13246 Troeger C, Colombara DV, Rao PC, Khalil IA, Brown A, Brewer TG, et al. Global disability-adjusted life-year estimates of long-term health burden and undernutrition attributable to diarrhoeal diseases in children younger than 5 years. Lancet Glob Health. 2018 Mar;6(3):e255–69. Additional Declarations No competing interests reported. Supplementary Files AdditionalFile1.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 14 Aug, 2024 Reviews received at journal 12 Aug, 2024 Reviews received at journal 29 Jul, 2024 Reviews received at journal 24 Jul, 2024 Reviewers agreed at journal 16 Jul, 2024 Reviewers agreed at journal 15 Jul, 2024 Reviewers agreed at journal 10 Jul, 2024 Reviewers invited by journal 05 Jul, 2024 Editor assigned by journal 01 Jul, 2024 Submission checks completed at journal 28 Jun, 2024 First submitted to journal 27 Jun, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-4651040\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":329437041,\"identity\":\"4e410100-3427-488e-a12f-ad67d7d899b1\",\"order_by\":0,\"name\":\"Alison Bowman\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Institute of Health Metrics and Evaluation\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Alison\",\"middleName\":\"\",\"lastName\":\"Bowman\",\"suffix\":\"\"},{\"id\":329437042,\"identity\":\"bf468e26-bd79-4512-ac22-a29cb3370a2d\",\"order_by\":1,\"name\":\"Sylvia Lutze\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAl0lEQVRIiWNgGAWjYBACPghlAyYliNLCxsAMotJI13KYFC0S+Yc/fPhzPnHDAeaDt3mI05LMYDiz7TZQC1uyNdFaknkbbiduO8BjJk20lsN//pwDauH/RrQWxmYGtgMgW9iI1MLz2Jixty3ZeP9hNmPLOcRo4WdPfPzhxx872ZntzQ9vvCFGCwIwk6Z8FIyCUTAKRgE+AADdIS0se8wn2QAAAABJRU5ErkJggg==\",\"orcid\":\"\",\"institution\":\"Institute of Health Metrics and Evaluation\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Sylvia\",\"middleName\":\"\",\"lastName\":\"Lutze\",\"suffix\":\"\"},{\"id\":329437043,\"identity\":\"9c698e36-2999-4252-9836-a0768a622614\",\"order_by\":2,\"name\":\"James Albright\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Institute of Health Metrics and Evaluation\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"James\",\"middleName\":\"\",\"lastName\":\"Albright\",\"suffix\":\"\"},{\"id\":329437044,\"identity\":\"0442041c-6b0d-4371-bb02-7b24ef352877\",\"order_by\":3,\"name\":\"Nathaniel Blair-Stahn\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Institute of Health Metrics and Evaluation\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Nathaniel\",\"middleName\":\"\",\"lastName\":\"Blair-Stahn\",\"suffix\":\"\"},{\"id\":329437045,\"identity\":\"af7d4191-a1ee-42c0-9ea8-7801d37b5dde\",\"order_by\":4,\"name\":\"Hussain Jafari\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Institute of Health Metrics and Evaluation\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Hussain\",\"middleName\":\"\",\"lastName\":\"Jafari\",\"suffix\":\"\"},{\"id\":329437046,\"identity\":\"9c05f579-66b9-4144-a99e-e56c9d6d62e1\",\"order_by\":5,\"name\":\"Simar Kaur\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Institute of Health Metrics and Evaluation\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Simar\",\"middleName\":\"\",\"lastName\":\"Kaur\",\"suffix\":\"\"},{\"id\":329437047,\"identity\":\"bce56e16-78eb-48ad-ba85-25fbf22048b3\",\"order_by\":6,\"name\":\"Caroline Kinuthia\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Institute of Health Metrics and Evaluation\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Caroline\",\"middleName\":\"\",\"lastName\":\"Kinuthia\",\"suffix\":\"\"},{\"id\":329437048,\"identity\":\"4861cd3a-ccda-4ac2-9c02-ebbdf5fa32ed\",\"order_by\":7,\"name\":\"Rajan Mudambi\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Institute of Health Metrics and Evaluation\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Rajan\",\"middleName\":\"\",\"lastName\":\"Mudambi\",\"suffix\":\"\"},{\"id\":329437049,\"identity\":\"45c97724-531c-47ad-a339-3ebfb9133fa9\",\"order_by\":8,\"name\":\"Patrick Nast\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Institute of Health Metrics and Evaluation\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Patrick\",\"middleName\":\"\",\"lastName\":\"Nast\",\"suffix\":\"\"},{\"id\":329437050,\"identity\":\"9027d027-1a92-48c5-b0c9-4ad02b758ab2\",\"order_by\":9,\"name\":\"Alix Pletcher\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Institute of Health Metrics and Evaluation\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Alix\",\"middleName\":\"\",\"lastName\":\"Pletcher\",\"suffix\":\"\"},{\"id\":329437051,\"identity\":\"8c56dad0-45a2-45f4-98cd-1c212ce7a829\",\"order_by\":10,\"name\":\"Abraham Flaxman\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Institute of Health Metrics and Evaluation\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Abraham\",\"middleName\":\"\",\"lastName\":\"Flaxman\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2024-06-27 22:14:20\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-4651040/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-4651040/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":60851000,\"identity\":\"908a5def-987c-4186-a2ed-3efb2640d3ea\",\"added_by\":\"auto\",\"created_at\":\"2024-07-22 20:57:26\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":238425,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eDiagram of all modeled components and the causal relationships between them in the microsimulation health model.\\u003c/p\\u003e\\n\\u003cp\\u003eFootnote: Colored components represent interventions and bolded components represent model outputs. Arrows represent causal impacts and dashed lines represent non-causal correlations.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4651040/v1/c48f6deadb12756012c1002a.png\"},{\"id\":60852877,\"identity\":\"08e02dad-4571-4751-94dd-986ec6af4dd5\",\"added_by\":\"auto\",\"created_at\":\"2024-07-22 21:13:26\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":354551,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eA: Spending allocation by intervention at baseline and under optimized reallocations of the baseline budgetB: Intervention coverage at baseline and under optimized reallocations of the baseline budget\\u003c/p\\u003e\\n\\u003cp\\u003eFigure 2 Footnote: IFA: iron and folic acid; MMN: multiple micronutrients; BEP: balanced energy protein; SAM: severe acute malnutrition; MAM: moderate acute malnutrition; SQ-LNS: small-quantity lipid-based nutrient supplementation; Tx: treatment; USD: United States dollars; DALY: disability-adjusted life year.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4651040/v1/25d691afd61c7c5665dad209.png\"},{\"id\":60851905,\"identity\":\"d7f0a20c-db42-414a-b0b1-ddeecfec1175\",\"added_by\":\"auto\",\"created_at\":\"2024-07-22 21:05:26\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":367300,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eAnnual disability-adjusted life years (DALYs) among pregnancies and children under 5 by scenario\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4651040/v1/5f430a000bc82b4c34e37d69.png\"},{\"id\":60853405,\"identity\":\"37086c4e-e84b-485e-9454-600d6a956e47\",\"added_by\":\"auto\",\"created_at\":\"2024-07-22 21:21:26\",\"extension\":\"png\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":186591,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eA: Annual intervention spending allocation optimized to DALYs among pregnancies and children under 5 by increasing budget size\\u003c/p\\u003e\\n\\u003cp\\u003eB: Intervention coverage optimized to DALYs among pregnancies and children under 5 by increasing budget size\\u003c/p\\u003e\\n\\u003cp\\u003eC: Disability-adjusted life years (DALYs) averted among pregnancies and children under 5 relative to zero spending by increasing budget size with intervention spending allocation optimized to DALYs averted\\u003c/p\\u003e\\n\\u003cp\\u003eD: Incremental cost-effectiveness ratio (ICER) relative to zero spending by increasing budget size with intervention spending allocation optimized to DALYs averted\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;footnote: IFA: iron and folic acid; MMN: multiple micronutrients; BEP: balanced energy protein; SAM: severe acute malnutrition; MAM: moderate acute malnutrition; SQ-LNS: small-quantity lipid-based nutrient supplementation; USD: United States dollars; DALY: disability-adjusted life year.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4651040/v1/ff849246ca52f7b1033b39c8.png\"},{\"id\":60853408,\"identity\":\"a6e45c18-62e7-4c76-8899-50ee41d2a77f\",\"added_by\":\"auto\",\"created_at\":\"2024-07-22 21:21:31\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":1959966,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4651040/v1/930e349a-d9e0-430a-98b0-b6fe1958ac7e.pdf\"},{\"id\":60851004,\"identity\":\"40787649-3099-4df8-9a24-8481ae409fc3\",\"added_by\":\"auto\",\"created_at\":\"2024-07-22 20:57:26\",\"extension\":\"docx\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":756840,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"AdditionalFile1.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4651040/v1/8e356664fff3575fc2b9a0fc.docx\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Optimal Allocation of Antenatal and Young Child Nutrition Interventions: An Individual-Based GBD-Calibrated Microsimulation\",\"fulltext\":[{\"header\":\"Background\",\"content\":\"\\u003cp\\u003eUndernutrition remains a global crisis, contributing significantly to the burden of disease and hindering progress toward the Sustainable Development Goal (SDG) to end hunger\\u0026nbsp;(1). This persistent challenge demands urgent action and investment in proven interventions to address undernutrition during pregnancy and childhood.\\u003c/p\\u003e\\n\\u003cp\\u003eDespite a range of known effective interventions, coverage remains insufficient and unevenly distributed, particularly in regions with the greatest need\\u0026nbsp;(2). Two key areas where current practices lag are antenatal supplementation and the prevention and treatment of childhood acute malnutrition (AM). While iron and folic acid (IFA) supplementation is the standard of care in many settings, the World Health Organization (WHO) advocates for research on IFA-containing multiple micronutrient (MMN) supplements\\u0026nbsp;(3)\\u0026nbsp;and balanced energy and protein (BEP) supplementation for undernourished pregnant people\\u0026nbsp;(World Health Organization, 2016). Notably, there is suggestion that BEP supplementation targeted to undernourished pregnancies (in contrast to undernourished populations) may be a cost-effective strategy\\u0026nbsp;(5). Furthermore, WHO recently issued new guidance on the prevention and treatment of AM among children under 5\\u0026nbsp;(6). This guidance provides recommendations for the treatment of moderate acute malnutrition (MAM) in a targeted fashion in addition to the prior recommendations for treating severe acute malnutrition (SAM). It additionally issues new recommendations for the prevention of AM with strategies, including the consideration of medium- or small-quantity lipid-based nutrient supplementation (MQ-LNS or SQ-LNS), particularly in contexts of high food insecurity.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eEven with the recognized need for scaling up nutrition interventions, resource constraints and the complexity of maximizing health impact with limited budgets pose significant challenges\\u0026nbsp;(7). While prevention is generally more effective\\u0026nbsp;(8), resource allocation decisions must consider various factors, including intervention costs, population reach, and potential interactions between different interventions. In silico computer models offer a valuable tool for guiding decision-making by simulating intervention effectiveness and cost-effectiveness. Existing models like the MMS Cost Benefit tool\\u0026nbsp;(9), FACET4SNF\\u0026nbsp;(10), CMAM costing tool\\u0026nbsp;(11), and WBCi\\u0026nbsp;(12)\\u0026nbsp;provide insights into specific interventions, while MINIMOD focuses on micronutrient interventions\\u0026nbsp;(13). The Optima Nutrition model\\u0026nbsp;(14), utilizing LiST\\u0026nbsp;(15), offers a broader analysis but has limitations regarding MAM treatment and the inability to test interventions independently. The development of additional models that can incorporate targeted MAM treatment and evaluate optimal spending allocations could enhance decision-making by providing robust evidence across multiple platforms. In particular, conclusions that are robust across multiple models may reduce concerns regarding structural uncertainty\\u0026nbsp;(16).\\u003c/p\\u003e\\n\\u003cp\\u003eIn this interest, we developed an individual-based microsimulation model using Global Burden of Disease (GBD) 2021 study data to estimate the health impact of several nutrition-related interventions and paired it with an allocative efficiency analysis to determine allocation of intervention spending to optimize impact for a specified budget size(s). In this paper, we provide an overview of our simulation and include results specific to Ethiopia as an illustrative example.\\u003c/p\\u003e\\n\\u003cp\\u003eA note on language in this paper: Our team acknowledges that not all people who get pregnant or give birth are women, and we therefore strive to use more inclusive language that encompasses all identities without othering individuals. We are also aware that women often face unique challenges and focused work on women and girls is essential to improving health for all. The accepted language in literature is often women-centric such as \\u0026ldquo;maternal health\\u0026rdquo; or \\u0026ldquo;women of reproductive age\\u0026rdquo;. Our team advocates for the use of language inclusive to all people that also honors the unique experiences and needs of women. Therefore, throughout this paper we try to use gender inclusive language such as \\u0026ldquo;pregnancy-related health\\u0026rdquo; or \\u0026ldquo;women and birthing people of reproductive age\\u0026rdquo;. More information and a full glossary can be found in appendix 1, additional file 1.\\u0026nbsp;\\u003c/p\\u003e\"},{\"header\":\"Methods\",\"content\":\"\\u003cp\\u003eOur model consists of two main components: (1) an individual-based microsimulation model of health events, and (2) an allocative efficiency analytic model that processes the results of the microsimulation model and outputs the optimal intervention funding allocation for a given budget size.\\u003c/p\\u003e\\n\\u003ch2\\u003eBaseline health model\\u003c/h2\\u003e\\n\\u003cp\\u003eWe utilize Vivarium\\u0026nbsp;(17), a mature, open-source, Python-based simulation framework for our baseline health model. Published examples of models utilizing Vivarium can be found elsewhere\\u0026nbsp;(5,18,19). For this application, our model consists of a closed cohort of simulated individuals that we track across discrete time steps. Simulated individuals are assigned various attributes (such as age, sex, risk factor exposures, disease status, and vital status) that evolve over time and influence their trajectory through the simulation. Specifically, at each time step, simulated individuals are subject to some probability of a disease event (non-infected individuals may become infected, infected individuals may recover or die), which is modified by their other assigned attributes such as age, sex, and risk factor exposures. The model is calibrated such that individual-level heterogeneity reflects the appropriate magnitude across specific attributes while averages across simulated individuals reflect appropriate population-level statistics.\\u003c/p\\u003e\\n\\u003cp\\u003eThe primary data source for our model was the GBD 2021 study, which estimates mortality and disability at the location-, year-, sex-, and age-specific level across hundreds of diseases, injuries and risk factors (20\\u0026ndash;24). Each timestep spent affected by a morbidity-causing condition results in accumulation of years lived with disability (YLDs) in accordance with the disability weight (DW) of that condition utilized in the GBD study. Years of life lost (YLLs) are accumulated in accordance with the theoretical minimum risk life expectancy (TMRLE) specific to a simulated individual\\u0026rsquo;s age at the moment of death as informed from the GBD study.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eFigure 1 represents all modeled components in our baseline health model and the interactions between them, which are discussed in more detail in the following sections. We separated our baseline health model into two distinct simulated population groups: pregnant women and birthing people, and children under 5. Our simulation utilizes Monte Carlo methods to propagate parameter uncertainty throughout the model, which is covered in more depth in appendix 7, additional file 1.\\u003c/p\\u003e\\n\\u003ch3\\u003ePregnancy simulation\\u003c/h3\\u003e\\n\\u003cp\\u003eThe population structure, divided into five-year age groups, of the simulated cohort of pregnancies is informed from demographic estimates of women and birthing people of reproductive age (10\\u0026ndash;54 years) paired with estimates of age-specific fertility rates from the GBD study at a location- and year-specific level. Each simulated individual begins the simulation on day zero of their pregnancy in our model. Possible pregnancy outcomes include live birth, stillbirth, and abortion/miscarriage, with relative frequencies informed from GBD estimates. Infant sex is determined according to estimates of live births by sex from the GBD study. Pregnancy duration is determined according to the location- and sex-specific gestational age distribution of live births in GBD for both live births and stillbirths. For pregnancies that result in abortion/miscarriage, duration is determined according to a uniform distribution between 6 and 24 weeks. Each simulated individual is also assigned a continuous hemoglobin concentration and a dichotomous exposure for pre-pregnancy BMI above or below 18.5, each informed from GBD risk exposure estimates. Hemoglobin and BMI exposures are correlated to one another as informed from the Woman First trial (25) (more details in appendix 2, additional file 1).\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eWe advanced the simulation clock in increments of seven-day time steps and adjusted simulant ages accordingly. We used pregnancy-specific hemoglobin threshold values for anemia and simulants\\u0026rsquo; hemoglobin values to assign severity-specific anemia exposures and YLDs due to anemia accrued according to the corresponding disability weight (more details in appendix 2, additional file 1).\\u003c/p\\u003e\\n\\u003cp\\u003eSimulated birth events occur when the simulation time clock reaches the end of a simulant\\u0026rsquo;s assigned pregnancy duration. At the moment of birth, simulants experience incident non-fatal or fatal cases of pregnancy-related disorders according to the population-level age- and location-specific probability from GBD that is further modified by their hemoglobin concentration at birth. Pregnancy-related disorders in this model are defined as all conditions resulting in a loss of health tied to pregnancy, birth, or postpartum complications. A full list of the included health conditions can be found in appendix 2, additional file 1. Incident, non-fatal pregnancy-related disorder causes accumulate YLDs as estimated in the 2021 GBD study. In a similar manner to pregnancy-related disorders, simulants also may experience incident cases of postpartum hemorrhage at birth with a likelihood modified by their hemoglobin exposure. For simulants that experience incident cases of postpartum hemorrhage, we applied a corresponding decrease on their postpartum hemoglobin level. We assigned infant birthweight according to the joint distribution with gestational age at birth (equivalent to pregnancy duration) from GBD and further correlated with joint antenatal anemia and pre-pregnancy/first trimester BMI exposure. Details on the magnitude and data sources of these effects can be found in appendix 2, additional file 1.\\u003c/p\\u003e\\n\\u003cp\\u003eSimulants are followed for anemia morbidity in the pregnancy simulation for an additional six weeks following birth before exiting the simulation. No background mortality due to causes other than pregnancy-related disorders are considered in the pregnancy simulation.\\u003c/p\\u003e\\n\\u003ch3\\u003eChildhood simulation\\u003c/h3\\u003e\\n\\u003cp\\u003eWe initialized the simulated population of children under 5 with the birth events in the pregnancy simulation. Each live birth that occurs in the pregnancy simulation is initialized into the simulation on day zero of their life, with values for sex, gestational age at birth, and birthweight as determined in the pregnancy simulation. Our model progresses in time steps of four days, and we track ages within the age groups of 0\\u0026ndash;6 days, 7\\u0026ndash;28 days, 1\\u0026ndash;5 months, 6\\u0026ndash;11 months, 12\\u0026ndash;23 months, and 2\\u0026ndash;4 years. Assigned gestational age and birthweight exposures affect mortality due to causes affected by the risk factor in the GBD 2021 study (21) for the first 28 days of life, in a manner adapted from the GBD risk effect model described in appendix 3, additional file 1.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eStarting at 28 days of life, we assigned four-category exposure values for wasting (based on weight-for-height z-scores; WHZ), stunting (based on height-for-age z-scores; HAZ), and underweight (based on weight-for-age z-scores; WAZ). For all three metrics, we included exposure categories of severe (z-score \\u0026lt; -3), moderate (-3 \\u0026lt; z-score \\u0026lt; -2), mild (-2 \\u0026lt; z-score \\u0026lt; -1), and unaffected (z-score \\u0026gt; -1) with exposure prevalence informed from 2021 GBD study estimates. We additionally subdivided the moderate wasting exposure category into two substates: WHZ between -3 and -2.5 and WHZ between -2.5 and -2, in order to support the targeted MAM treatment intervention in our model. The relative exposure of each of these substates is informed from the most recently available Demographic Health Survey (DHS) data for the modeled location pooled across age groups and sexes\\u0026nbsp;(26), and relative morbidity and mortality risk of each substate is derived from the GBD 2021 study estimates.\\u003c/p\\u003e\\n\\u003cp\\u003eWe assigned each simulant a fixed stunting percentile throughout life such that a stunting exposure value may change as simulants age into the next age group with a different population-level exposure distribution, but the percentile within the age-specific population will not change. We model a dynamic transition model of child wasting that is calibrated to the GBD exposure distribution and wasting state-specific mortality rates, estimates of wasting incidence rates from longitudinal cohort studies that tracked child anthropometry in low- and low-middle-income countries (details in appendix 4, additional file 1), and observed recovery rates from MAM and SAM in the ComPAS trial\\u0026nbsp;(27)\\u0026nbsp;under the assumption of a steady state equilibrium (more details in appendix 4, additional file 1). This calibration allows us to estimate average recovery rates from MAM and SAM states among populations without access to treatment, which are generally not available in the literature (with some exceptions, such as\\u0026nbsp;James et al., 2016). Notably, wasting transition rates do not vary by moderate wasting substate exposure.\\u003c/p\\u003e\\n\\u003cp\\u003eWe assume no correlation between stunting and wasting exposures. Underweight exposures are assigned according to observed location-, age-, and sex-specific correlation with four-category wasting and four-category stunting obtained from the DHS and are updated dynamically as simulant age, wasting exposure, and/or stunting exposure evolves throughout the simulation (more details in appendix 4, additional file 1).\\u003c/p\\u003e\\n\\u003cp\\u003eInfant birthweight influences population-level stunting exposures used to determine individual-level stunting exposures in accordance with evidence from\\u0026nbsp;McGovern (2019). Infant birthweight influences wasting state at 28 days of life in accordance with observed data from DHS but does not influence wasting transition rates thereafter. More details on the impact of birthweight on child growth failure can be found in appendix 5, additional file 1.\\u003c/p\\u003e\\n\\u003cp\\u003eWe model incidence (and associated morbidity) and mortality due to diarrheal diseases, lower respiratory infections, malaria, and measles from ages 28 days to 5 years. Incidence and mortality rates due to these causes are modified by stunting, underweight, and wasting exposures (including moderate wasting substate exposures) in accordance with GBD estimates of child growth failure effects, described elsewhere\\u0026nbsp;(21). Briefly, GBD methods account for overlap between these indices of child growth failure to ensure impacts on morbidity and mortality are not overestimated. We additionally model morbidity and mortality due to protein energy malnutrition, which is entirely informed by wasting exposure among simulants aged 28 days to 5 years. Finally, we model background mortality due to all causes other than those directly modeled for all ages in our child simulation.\\u003c/p\\u003e\\n\\u003ch2\\u003eIntervention models\\u003c/h2\\u003e\\n\\u003cp\\u003eOur model considers three antenatal supplementation intervention products, including iron and folic acid (IFA), multiple micronutrients (MMN), and balanced energy protein (BEP) supplementation, in addition to three child nutrition interventions, including treatment for SAM, treatment for MAM, and wasting prevention with SQ-LNS.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eWe assume that antenatal supplementation products are distributed at routine antenatal care visits and taken for a duration of six months. We assume that BEP is provided in addition to MMN for pregnancies with a pre-pregnancy/first trimester BMI of less than 18.5 and otherwise antenatal supplementation products are mutually exclusive at the individual level. We additionally assume that BEP + MMN are only provided to low BMI pregnancies if MMN is also provided to adequate BMI pregnancies.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eAntenatal supplementation products affect the probability of stillbirth, gestational age at birth, and birthweight, with effect sizes shown in Table 1.\\u0026nbsp;Intervention mean differences in antenatal hemoglobin are applied at the individual level at eight weeks of gestation. We assume that reduction in stillbirth outcomes associated with interventions results in increases in live birth outcomes, with no changes in other birth outcomes such as abortion, miscarriage, or ectopic pregnancy. For intervention effects on preterm birth, we calculated location- and year-specific population mean differences that resulted in the relative risks reported in table 1 and applied the mean differences to our continuous measure of gestational age at birth at the individual level in our simulation. Notably, for the effect of MMN on preterm birth, we estimated a population mean difference conditional on gestational age at birth +/- 32 weeks, such that there was a larger effect among those born at less than 32 weeks\\u0026rsquo; gestation than those born at greater than 32 weeks\\u0026rsquo; gestation so that both the effects on preterm birth (\\u0026lt;37 weeks) and very preterm birth (\\u0026lt;32 weeks) were replicated.\\u003c/p\\u003e\\n\\u003cp\\u003eTable 1. Antenatal intervention effects\\u0026nbsp;\\u003c/p\\u003e\\n\\u003ctable border=\\\"1\\\" cellspacing=\\\"0\\\" cellpadding=\\\"0\\\"\\u003e\\n \\u003ctbody\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eOutcome\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eIron folic acid (IFA), relative to no supplementation\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eMultiple micronutrients (MMN), relative to IFA\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eBalanced energy protein (BEP), relative to MMN\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eAntenatal hemoglobin, grams per liter\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eMD = +7.8 (4.08, 11.52)\\u0026nbsp;(30)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eMD = +0\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eMD = +0\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eStillbirth\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eRR = 1\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eRR = 0.91 (0.71, 0.93)\\u0026nbsp;(30)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eRR = 0.39 (0.19, 0.80)\\u0026nbsp;(31)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eBirthweight, grams\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eMD = +57.73 (7.66, 107.79)\\u0026nbsp;(32)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eMD = +45.16 (32.31, 58.02)\\u0026nbsp;(5,33)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eMD = +66.96 (13.13, 120.78)\\u0026nbsp;(34)*\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003ePreterm birth (\\u0026lt;37 weeks)\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eRR = 0.90 (0.86, 0.95)\\u0026nbsp;(35)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eRR = 0.91 (0.84, 0.99)\\u0026nbsp;(36)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eRR = 1\\u0026nbsp;(34)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eVery preterm birth (\\u0026lt;32 weeks)\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e-\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eRR = 0.81 (0.71, 0.93)\\u0026nbsp;(33)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cul\\u003e\\n \\u003cli\\u003e\\u0026nbsp;\\u003c/li\\u003e\\n \\u003c/ul\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/tbody\\u003e\\n\\u003c/table\\u003e\\n\\u003cp\\u003eTable 1 Footnote: MD: mean difference; RR: relative risk\\u003c/p\\u003e\\n\\u003cp\\u003e*Effect specific to subpopulation of undernourished women and birthing people\\u003c/p\\u003e\\n\\u003cp\\u003eIn our model, the SAM treatment intervention is administered to incident cases of SAM, as defined by WHZ \\u0026lt; -3, between 6 and 59 months of age. In universal protocol of the MAM treatment intervention, treatment is administered to incident cases of MAM, as defined by WHZ between -2 and -3, between 6 and 59 months of age. The targeted protocol of the MAM treatment intervention is administered to those eligible for the universal MAM treatment intervention who also satisfy at least one of the following criteria: (a) less than 24 months of age, (b) WHZ between -2.5 and -3, or (c) WAZ less than -3. Treatment protocol and time-to-recovery for both the MAM and SAM interventions are informed from the intervention arm of the ComPAS trial\\u0026nbsp;(27). We assume that a fraction of cases does not respond to treatment and recover according to the untreated recovery rates in our wasting model.\\u003c/p\\u003e\\n\\u003cp\\u003eIn our model, SQ-LNS supplementation begins at six months of age for a duration of 12 months for all infants. SQ-LNS supplementation decreases the prevalence of moderate and severe stunting and increases the prevalence of no stunting in accordance with effects obtained from study authors of an individual participant meta-analysis on the intervention\\u0026nbsp;(37). We model effects of SQ-LNS supplementation on transition rates from no wasting to mild wasting, mild wasting to moderate wasting, and moderate wasting to severe wasting with no effects on wasting recovery rates. These effect sizes are calibrated to replicate the wasting prevalence ratios from the same source as stunting effect sizes, as described in appendix 6, additional file 1.\\u003c/p\\u003e\\n\\u003cp\\u003eWe assume that the initial point of care for all child nutrition interventions is monthly community management of acute malnutrition (CMAM) screenings. We assume that children are assessed for acute malnutrition and referred for treatment as appropriate at these meetings and that SQ-LNS product counseling and distribution occurs at these meetings as well, as has been performed in trial settings\\u0026nbsp;(38). Notably, this assumption dictates that for a population of 100 children 6\\u0026ndash;59 months, five of whom are afflicted with SAM, all 100 children would be screened in order to reach and treat the five afflicted children for SAM.\\u003c/p\\u003e\\n\\u003ch2\\u003eOptimization structure\\u003c/h2\\u003e\\n\\u003ch3\\u003eScenario layout\\u003c/h3\\u003e\\n\\u003cp\\u003eWe ran the health model for every possible combination of our modeled interventions. For instance, if we only considered IFA and SAM treatment interventions, we would run the health model under the following conditions:\\u003c/p\\u003e\\n\\u003col class=\\\"decimal_type\\\"\\u003e\\n \\u003cli\\u003e0% coverage IFA, 0% coverage SAM treatment\\u003c/li\\u003e\\n \\u003cli\\u003e100% coverage IFA, 0% coverage SAM treatment\\u003c/li\\u003e\\n \\u003cli\\u003e0% coverage IFA, 100% coverage SAM treatment\\u003c/li\\u003e\\n \\u003cli\\u003e100% coverage IFA, 100% coverage SAM treatment\\u003c/li\\u003e\\n\\u003c/ol\\u003e\\n\\u003cp\\u003eWe rely on the key assumption that individuals within our simulated population are independent of one another in that the health status of one individual does not impact any others. This allows us to assume that the population health status under 50% coverage of a given intervention can be equivalently represented as the average between the population health status under 100% coverage of the intervention and the population health status under 0% coverage.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eWith all seven of our modeled interventions, after excluding illogical combinations (such as simultaneous presence of universal MAM treatment and targeted MAM treatment interventions), this amounts to a total of 48 scenarios in addition to one baseline scenario. In each of these scenarios, we record counts of deaths, stillbirths, YLLs, YLDs, incident wasting cases, person-time, and intervention administration counts that occurred in our simulated populations. Deaths, YLLs, and YLDs are stratified by pregnancy and child populations, and person-time counts are stratified by stunting state and age group for children.\\u003c/p\\u003e\\n\\u003cp\\u003eAcross modeled scenarios, we utilize common random numbers\\u0026nbsp;(39)\\u0026nbsp;at the individual simulant level such that in the absence of intervention influence, each simulant would undergo the exact same trajectory in one scenario as the others, reducing noise across scenarios in our model.\\u003c/p\\u003e\\n\\u003ch3\\u003eOptimization function and constraints\\u003c/h3\\u003e\\n\\u003cp\\u003eWe use the \\u003cem\\u003escipy.optimize\\u0026nbsp;\\u003c/em\\u003epackage in Python\\u0026nbsp;(40)\\u0026nbsp;to perform our allocative efficiency analysis. Inputs to our optimization function include the recorded health outcomes for each modeled scenario and the calculated cost of each scenario (obtained by multiplying the recorded intervention administration counts for each scenario by the intervention costs). The optimization objective is to find the fractional combination of scenarios that maximizes/minimizes the specified health quantity (ex: minimize DALYs or maximize person-time) under the following constraints:\\u003c/p\\u003e\\n\\u003cul\\u003e\\n \\u003cli\\u003eTotal cost cannot exceed user-defined budget envelope\\u003c/li\\u003e\\n \\u003cli\\u003eTotal intervention-specific coverage cannot exceed specified saturation coverage limits\\u003c/li\\u003e\\n \\u003cli\\u003eThe two intervention coverage rules described below are followed\\u003cul style=\\\"list-style-type: circle;\\\"\\u003e\\n \\u003cli\\u003eIntervention coverage rule one: all interventions share a \\u0026ldquo;common care-seeking propensity\\u0026rdquo; that dictates if population coverage of intervention A and B are both 50%, we assume the same 50% of the population is covered by interventions A and B and that the remaining 50% of the population has access to neither intervention A nor B.\\u003c/li\\u003e\\n \\u003cli\\u003eIntervention coverage rule two: we assume that ANC intervention products could not be \\u0026ldquo;targeted\\u0026rdquo; to a subpopulation based on their coverage of child nutrition interventions or vice versa. For instance, if saturation coverage of ANC products is greater than that of child nutrition interventions, a solution that allocates IFA to the portion of the population with access to child nutrition interventions and MMN to the portion of the population without access to child nutrition interventions is not allowed. \\u003c/li\\u003e\\n \\u003c/ul\\u003e\\n \\u003c/li\\u003e\\n\\u003c/ul\\u003e\\n\\u003cp\\u003eFollowing these and the previously stated assumption of simulant independence, we calculate overall cost, health impact, and intervention coverage as outputs of our optimization function for a single budget size.\\u003c/p\\u003e\\n\\u003ch2\\u003eIllustrative example parameters\\u003c/h2\\u003e\\n\\u003cp\\u003eAs an illustrative example, we ran the model for Ethiopia in 2021. We utilized a simulated population size of 1,600,000 pregnancies and informed the simulated child population from the resulting birth outcomes from a random sample of 400,000 of these pregnancies. These population sizes were selected as the minimum size that achieved stability in the difference between health outcomes across modeled scenarios among the antenatal and child populations, respectively. We scaled results from the pregnancy and child baseline health models by a factor of 3.5 and 14, respectively, to reflect the total estimated number of births in Ethiopia in 2021. We ran our simulation model for 20 Monte Carlo uncertainty draws (see appendix 7, additional file 1).\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eValues for baseline intervention coverage, maximum/saturation intervention coverage, and intervention costs used for this model run are displayed in table 2. Intervention costs used in our model represent product costs only and were obtained from the UNICEF Supply Catalogue for IFA, MMN, BEP, and SQ-LNS (UNICEF Supply Catalogue, n.d.) interventions and the ComPAS trial (Bailey et al., 2020) for MAM and SAM treatment interventions. Costs from the UNICEF Supply Catalogue were obtained in November of 2023. We assumed there were no product costs associated with CMAM screenings.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cu\\u003eTable 2. Intervention coverage and cost assumptions\\u003c/u\\u003e\\u003c/p\\u003e\\n\\u003ctable border=\\\"1\\\" cellspacing=\\\"0\\\" cellpadding=\\\"0\\\"\\u003e\\n \\u003ctbody\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"19.23076923076923%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eIntervention\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eBaseline coverage\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"21.153846153846153%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eSaturation coverage\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"34.61538461538461%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eCost (2023 USD)\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"19.23076923076923%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eIFA\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e60.2% (95% CI: 48.7, 72.6) (Personal communication of work by Nat Henry at CIFF)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"21.153846153846153%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e75.7% (ANC1; GBD 2021)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"34.61538461538461%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e2.27 per supplemented pregnancy\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"19.23076923076923%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eMMN\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e0%, assumption\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"21.153846153846153%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e75.7% (ANC1; GBD 2021)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"34.61538461538461%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e3.47 per supplemented pregnancy\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"19.23076923076923%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eBEP\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e0%, assumption\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"21.153846153846153%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e75.7% (ANC1; GBD 2021)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"34.61538461538461%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e40.28 per supplemented pregnancy\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"19.23076923076923%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eSAM treatment\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e48.8% (95% CI: 37.4, 60.4)\\u0026nbsp;(41)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"21.153846153846153%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e70%, assumption\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"34.61538461538461%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e41.84 per treated child\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"19.23076923076923%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eMAM treatment\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e15% (95% CI: 10, 20), assumption. Baseline coverage assumed to be universal implementation of MAM treatment intervention\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"21.153846153846153%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e70%, assumption (applies to both universal and targeted implementations of MAM treatment intervention)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"34.61538461538461%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e29.70 per treated child\\u003c/p\\u003e\\n \\u003cp\\u003e(applies to both universal and targeted implementations of MAM treatment intervention)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"19.23076923076923%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eSQ-LNS\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e0%, assumption\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"21.153846153846153%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e70%, assumption\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"34.61538461538461%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e35.75 for 12 months supplementation\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"19.23076923076923%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003eCMAM screening\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"25%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e48.8% (95% CI: 37.4, 60.4), assumed to be maximum of SAM treatment, MAM treatment, and SQ-LNS baseline coverage\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"21.153846153846153%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e70%, assumption\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"34.61538461538461%\\\" valign=\\\"top\\\"\\u003e\\n \\u003cp\\u003e0 per child screened\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/tbody\\u003e\\n\\u003c/table\\u003e\\n\\u003cp\\u003eTable 2 Footnote: IFA: iron and folic acid; MMN: multiple micronutrients; BEP: balanced energy protein; SAM: severe acute malnutrition; MAM: moderate acute malnutrition; SQ-LNS: small quantity lipid-based nutrient supplementation; CMAM: community management of acute malnutrition; USD: United States dollars\\u003c/p\\u003e\\n\\u003cp\\u003eWe performed our allocative efficiency analysis at the estimated baseline budget size in addition to 25 evenly spaced increments between zero and the spending required to reach saturation coverage of all interventions. We performed allocative efficiency analysis at the draw level as well as for the average result across all draws from the underlying health model.\\u003c/p\\u003e\"},{\"header\":\"Results\",\"content\":\"\\u003cp\\u003eFrom our assumed intervention coverage values and intervention costs, we estimated total annual spending on our modeled interventions to be 34.4 (95% UI: 28.3, 42.4) million USD in Ethiopia, with 5.6 (95% UI: 4.8, 6.7) million allocated to IFA, 14.7 (95% UI: 11.2, 20.4) million allocated to SAM treatment, and 14.1 (95% UI: 8.8, 19.7) million allocated to universal MAM treatment (Figure 2). Our model estimated 18.0 (95% UI: 15.1, 21.7) million annual DALYs among pregnancies and children under 5 at baseline due to the modeled childhood and pregnancy-related health conditions. This was estimated to be 986,000 (95% UI: 535,000, 1,588,000) DALYs fewer than the 19.0 (95% UI: 15.8, 22.9) million annual DALYs under the counterfactual scenario of zero spending on our modeled interventions.\\u003c/p\\u003e\\n\\u003cp\\u003eWe found that reallocation of the baseline budget to minimize DALYs resulted in funding MMN to its maximum coverage level of 75.6%, followed by investment of the remaining budget into treatment for SAM (which attained 62.7% population coverage). When the baseline budget was reallocated to maximize child time spent not stunted, the budget was likewise first spent maximizing MMN, but then the remaining funds were allocated to SQ-LNS supplementation alone (which attained 16.4% population coverage). We maximize child time spent not stunted rather than minimizing child time stunted because minimizing child time stunted can lead to optimizing for fewer children alive. These results fit with our modeling parameters since SQ-LNS impacts stunting and is a preventive treatment, while treatment for SAM directly impacts wasting, which is more deadly for young children. Figure 2 panel A displays the baseline and reallocated intervention-specific funding, and figure 2 panel B displays the baseline and reallocated intervention-specific coverage levels.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eRelative to the baseline allocation, the reallocation optimized to minimize DALYs resulted in 592,000 fewer annual DALYs and the reallocation optimized to maximize child time spent not stunted resulted in 187,000 fewer annual DALYs among pregnancies and children under 5. This amounted to a 60% and 19% increase in DALYs averted relative to zero spending, compared to baseline allocation, for the reallocation optimized to DALYs and the reallocation optimized to child time spent not stunted, respectively. Figure 3 displays annual DALYs under each of these scenarios.\\u003c/p\\u003e\\n\\u003cp\\u003eWhen we performed this allocative efficiency analysis for successively larger budget sizes and optimized to minimize DALYs, our model found that MMN supplementation alone should be prioritized until MMN reaches its maximum coverage level, followed by the SAM treatment intervention as budget size allows. Starting at annual budget sizes over 35 million USD, our model recommends investment in the targeted MAM treatment followed by universal MAM treatment, SQ-LNS, and finally targeted BEP supplementation at the largest budget sizes. The annual budget required to achieve maximum impact (reach saturation coverage of most impactful interventions) was 196 (95% UI: 165, 225) million USD (approximately 5.7 times more than we estimate is currently spent). Figure 4 panel A displays the baseline and reallocated intervention-specific expenditures, and figure 4 panel B displays the baseline and reallocated intervention-specific coverage levels across increasing budget sizes.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003ePanel C in Figure 4 displays the annual DALYs averted relative to zero spending on any of the modeled interventions at each successive budget size, with a total of 2.2 (95% UI: 1.4, 3.0) million annual DALYs averted at the maximum impact level. Panel D of Figure 4 displays the incremental cost-effectiveness ratio (ICER) in USD per DALY averted relative to zero spending on any of our modeled interventions across increasing budget sizes; note that all ICERs represented in this figure are positive with positive health gains (DALYs averted) and positive incremental costs (the northeast quadrant in as described by\\u0026nbsp;(42)). Notably cost-effectiveness decreases (ICER increases) with increasing budget as the most cost-effective interventions are prioritized at lower budget sizes, with all the ICER at all budget sizes at or below 88.94 (95% UI: 65.10, 154.04) dollars per DALY averted. Notably, without simultaneous coverage of MAM treatment or SQ-LNS interventions, 23 (95% UI: 19, 27) million USD was required to maintain saturation coverage (70%) of the SAM treatment intervention. However, the full scale-up of MAM treatment and SQ-LNS interventions to saturation coverage (70%) reduced the operating cost of maintaining saturation coverage of the SAM treatment intervention by approximately half to 12 (95% UI: 8.3, 16) million USD annually.\\u003c/p\\u003e\\n\\u003cp\\u003eWhen analyzed at the draw-level, 100% of draws resulted in intervention priority ordering starting with MMN, followed by SAM treatment and targeted MAM treatment. However, the relative priority ordering of the universal MAM treatment, SQ-LNS, and BEP interventions had significant draw-level variation. Specifically, BEP was prioritized behind universal MAM treatment and SQ-LNS in only 65% of draws, and universal MAM treatment was prioritized ahead of SQ-LNS in only 70% of draws (see appendix 7, additional file 1 for more information on draws).\\u003c/p\\u003e\"},{\"header\":\"Discussion\",\"content\":\"\\u003cp\\u003eOur simulation provides a new model to estimate optimal allocation of spending on antenatal and child health nutrition interventions utilizing a detailed health model informed from GBD study estimates that accounts for the interaction between preventive and therapeutic approaches. Our results show that an optimized reallocation of current spending on these products can substantially improve pregnancy-related and child health without additional funding and provides direction for how to best allocate expanded budgets to maximize impact.\\u003c/p\\u003e \\u003cp\\u003eOur model provides additional benefit relative to existing options such as the MMS Cost Benefit Tool, the CMAM Costing Tool, the FACET4SNF model, WBCi, and MINIMOD in that it integrates both antenatal and child micro- and macronutrition interventions and performs allocative efficiency analysis of such interventions. Our underlying health model utilizes a similar framework as the LiST model that is utilized by Optima Nutrition, which, to the best of our knowledge, is the only other allocative efficiency model that evaluates similar interventions. However, our model differs from LiST in several ways. Firstly, LiST is a compartmental rather than microsimulation model and utilizes a variety of data sources described elsewhere (\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e) to inform baseline population and mortality dynamics, whereas our model primarily utilizes GBD study data. While our models generally utilize similar data sources for intervention effect estimations, they differ with respect to the derivation of MAM and SAM treatment intervention impacts on recovery rates, with our model ultimately utilizing a slightly lower effect of MAM treatment on MAM recovery and slightly higher effect of SAM treatment on SAM recovery. Further, unlike our model, LiST applies intervention effects to subsets of the population only; for instance, the effect of SQ-LNS is applied to the food-insecure portion of the population only (\\u003cspan citationid=\\\"CR43\\\" class=\\\"CitationRef\\\"\\u003e43\\u003c/span\\u003e). With regard to the impact of the BEP supplementation intervention, our model benefits from a correlation of the population eligible for BEP (low BMI pregnancies) with low birthweight outcomes, which is not the case in the LiST and Optima Nutrition models. Unlike our model that accounts for correlation between wasting, stunting, and underweight exposures in the attribution of their effects on morbidity and mortality, LiST does not report to account for this in the estimation of risk associated with wasting and stunting in their model. Notably, intervention protocols and definitions may slightly differ between our models. LiST also supports several interventions not included in our model.\\u003c/p\\u003e \\u003cp\\u003eOur model benefits from high-quality and detailed age-, sex-, location-, and year-specific estimates from the 2021 GBD study and framework. A strength of this data source is how it accounts for correlation between wasting, stunting, and underweight exposures in its estimation of their impacts in morbidity and mortality; this avoids overestimation of the aggregate effects. Another strength is the continuous joint distribution of birthweight and gestational age at birth. Our modeling approach allows us to represent non-linear interactions between intervention combinations in that the presence of prevention interventions may remove the need for treatment interventions at the individual level. Finally, our model supports MAM treatment intervention as an independent program from SAM treatment that allows for testing of optimal coverage levels of these programs and additionally supports a targeted implementation of the MAM treatment intervention in a manner intended to reflect the recent WHO guidelines on the matter. Our model additionally supports BEP supplementation targeting to undernourished pregnancies rather than undernourished populations, which is expected to reach a greater proportion of those who may benefit from the intervention.\\u003c/p\\u003e \\u003cp\\u003eOur model is limited by the intensive computational resources required to run the microsimulation for our underlying health model, which we have performed on a high-performance computing cluster. Additionally, our model is limited in that it relies on assumptions and/or uncertain estimates of baseline intervention coverage and therefore baseline spending allocation. Furthermore, the intervention product costs used in this analysis are not necessarily reflective of the true costs of implementing intervention programs at scale.\\u003c/p\\u003e \\u003cp\\u003eOur baseline health model is limited in that it does not consider seasonality in wasting burden and relies on classification of acute malnutrition using WHZ alone rather than WHZ, mid-upper arm circumference (MUAC), and the presence of edema combined. Furthermore, our targeted MAM treatment intervention is inspired by the recent WHO guidelines but does not consider all recommended criteria for determining which MAM cases should receive treatment. Likewise, our model of BEP supplementation targeting is not exactly aligned with recommendations in the WHO guidelines. We do not model SQ-LNS intervention effects on anemia or vitamin A deficiency (\\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e44\\u003c/span\\u003e) in our model, nor do we consider a targeted implementation of the intervention such as that suggested in the WHO guideline. Finally, while our model may not represent all causal pathways by which our modeled interventions affect morbidity and mortality, as the absence of evidence between a given risk/outcome pair does not imply the absence of a causal association. Notably, we also do not consider feedback between wasting and stunting exposures (wasting leading to future stunting or vice versa) nor do we consider any causal impacts of infectious disease episodes leading to future wasting and/or stunting exposures, despite some evidence for such associations (\\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR46\\\" class=\\\"CitationRef\\\"\\u003e46\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eOur next steps for this model include the development of an interactive, online tool similar to the other models referenced here. This will allow user customization of the optimization and a more in-depth review of our results. Additionally, we will integrate a targeted SQ-LNS intervention implementation and perform sensitivity analyses around intervention cost assumptions, optimization measure, and location. We will also explore the robustness of intervention priority and evaluate the \\u0026ldquo;next best\\u0026rdquo; options. We also plan to extend this work for use in capacity planning for CMAM programs as they are adapted and/or expanded to align with updated guidelines on acute malnutrition prevention and treatment. Future extensions of this model may also include adaptation to support non-linear intervention cost and coverage functions and/or to support subnational-level allocative efficiency analysis within a given country.\\u003c/p\\u003e \\u003cp\\u003eOur model enables estimation of optimal spending allocations across several antenatal and child nutrition interventions for a specified budget size using high-quality GBD study estimates and may be compared to results from the existing LiST and/or Optima Nutrition models as an exploration of structural uncertainty in the pregnancy-related and child nutrition intervention optimization space. We hope our model can aid policy makers in decisions of how to best allocate future spending on pregnancy-related and child health nutrition interventions, especially as they may be planning for integration of the recent WHO guidelines on the treatment and prevention of childhood acute malnutrition.\\u003c/p\\u003e\"},{\"header\":\"Conclusion\",\"content\":\"\\u003cp\\u003eOur simulation offers a novel approach to optimizing spending on antenatal and child health nutrition interventions, leveraging the GBD study to create a detailed health model that can more accurately account for the interaction between preventative and therapeutic approaches. By reallocating funding to MMN and SAM treatment, 592,000 DALYs could be averted annual without further financial investment. The addition of this new simulation allows for greater confidence in modeling results overall, and new information for key decision makers to consider.\\u003c/p\\u003e\"},{\"header\":\"Abbreviations\",\"content\":\"\\u003ctable border=\\\"1\\\" cellspacing=\\\"0\\\" cellpadding=\\\"0\\\"\\u003e\\n \\u003ctbody\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eAbbreviation\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eDefinition\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eAM\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eAcute malnutrition\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eBEP\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eBalanced energy protein\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eCMAM\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eCommunity management of acute malnutrition\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eDALY\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eDisability-adjusted life year\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eDHS\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eDemographic health survey\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eDW\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eDisability weight\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eGBD\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eGlobal burden of disease study\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eHAZ\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eHeight-for-age z-score\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eICER\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eIncremental cost effectiveness ratio\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eIFA\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eIron folic acid\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eMAM\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eModerate acute malnutrition\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eMMN\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eMultiple micronutrient\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eMQ-LNS\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eMedium-quantity lipid-based nutrient supplement\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eMUAC\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eMid-upper arm circumference\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eSAM\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eSevere acute malnutrition\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eSDG\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eSustainable development goal\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eSQ-LNS\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eSmall-quantity lipid-based nutrient supplement\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eTMREL\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eTheoretical minimum risk exposure level\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eTMRLE\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eTheoretical minimum risk life expectancy\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eWAZ\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eWeight-for-age z-score\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eWHO\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eWorld health organization\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eWHZ\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eWeight-for-height z-score\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eYLD\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eYears lives with disability\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"25%\\\"\\u003e\\n \\u003cp\\u003eYLL\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd width=\\\"75%\\\"\\u003e\\n \\u003cp\\u003eYears of life lost\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/tbody\\u003e\\n\\u003c/table\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eEthics approval and consent to participate\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eNot applicable\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConsent for publication\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eNot applicable\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAvailability of data and materials\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe datasets generated for use during the current study are available in the ihmeuw/vivarium_gates_nutrition_optimization and ihmeuw/vivarium_gates_nutrition_optimization_child repositories, the link for the pregnancy simulation data is: https://doi.org/10.5281/zenodo.11661268 and the link for the child simulation data is: https://doi.org/10.5281/zenodo.11661095\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCompeting interests\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe authors declare that they have no competing interests.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eFunding\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis research was funded by the Bill and Melinda Gates Foundation. The funders had input in the high-level design of the model and specifically in the selection of interventions used here but were not involved in decision to publish or manuscript preparation.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAuthors\\u0026apos; contributions\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eAB, AF, and SL led the design of the simulation model and analysis of results. AB led the writing of the manuscript and figure generation. NB, AP, and SK also assisted in data analysis, figure generation, and design questions. CK led the project management. JA, HJ, and PN all contributed to the software development of the simulation model code. RM led the software development team and contributed to the model design.\\u0026nbsp;All authors read and approved the final manuscript.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAcknowledgements\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eOur team would like to acknowledge the following groups:\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cul\\u003e\\n \\u003cli\\u003eThe Maternal, Newborn, Child Health \\u0026amp; Nutrition team at the Bill and Melinda Gates Foundation, specifically Laura Lamberti, Rahul Rawat, and Chris Troeger, for their support in design decisions, and data access to the KI database.\\u0026nbsp;\\u003c/li\\u003e\\n \\u003cli\\u003eThe Institute for Global Nutrition at UC Davis, specifically Kathryn Dewey, Christine Stewart, Christine McDonald, and Charles Arnold who shared feedback and expert opinions on model design, and data for the development of the SQ-LNS intervention.\\u0026nbsp;\\u003c/li\\u003e\\n \\u003cli\\u003eIndi Trehan for his expert opinion in the design of the SQ-LNS intervention.\\u0026nbsp;\\u003c/li\\u003e\\n \\u003cli\\u003eNavideh Noori for her expert opinion on model design, specifically for the in the design of the SQ-LNS intervention.\\u0026nbsp;\\u003c/li\\u003e\\n \\u003cli\\u003eAll of the GBD team\\u0026rsquo;s at IHME that created the robust data used here. In particular, we relied on data and expertise from the Neonatal and Child Health team and the Reproductive, Genitourinary, \\u0026amp; Digestive Diseases team.\\u0026nbsp;\\u003c/li\\u003e\\n\\u003c/ul\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\n\\u003cli\\u003eThe State of Food Security and Nutrition in the World 2022 [Internet]. FAO; 2022 [cited 2022 Jul 29]. Available from: http://www.fao.org/documents/card/en/c/cc0639en\\u003c/li\\u003e\\n\\u003cli\\u003eHong Nguyen P, Singh N, Scott S, Neupane S, Jangid M, Walia M, et al. Unequal coverage of nutrition and health interventions for women and children in seven countries. Bull World Health Organ. 2022 Jan 1;100(1):20\\u0026ndash;9. \\u003c/li\\u003e\\n\\u003cli\\u003eWorld Health Organization. WHO antenatal care recommendations for a positive pregnancy experience Nutrition interventions update: Multiple micronutrient supplements during pregnancy [Internet]. 2020. Available from: https://iris.who.int/bitstream/handle/10665/333561/9789240007789-eng.pdf?sequence=1\\u003c/li\\u003e\\n\\u003cli\\u003eWorld Health Organization. WHO recommendations on antenatal care for a positive pregnancy experience [Internet]. 2016. Available from: https://iris.who.int/bitstream/handle/10665/250796/9789241549912-eng.pdf?sequence=1\\u003c/li\\u003e\\n\\u003cli\\u003eYoung N, Bowman A, Swedin K, Collins J, Blair-Stahn ND, Lindstedt PA, et al. Cost-effectiveness of antenatal multiple micronutrients and balanced energy protein supplementation compared to iron and folic acid supplementation in India, Pakistan, Mali, and Tanzania: A dynamic microsimulation study. Myers JE, editor. PLOS Med. 2022 Feb 22;19(2):e1003902. \\u003c/li\\u003e\\n\\u003cli\\u003eWorld Health Organization. WHO guideline on the prevention and management of wasting and nutritional oedema (acute malnutrition) in infants and children under 5 years [Internet]. 2023. Available from: https://app.magicapp.org/#/guideline/noPQkE\\u003c/li\\u003e\\n\\u003cli\\u003eHeidkamp RA, Piwoz E, Gillespie S, Keats EC, D\\u0026rsquo;Alimonte MR, Menon P, et al. Mobilising evidence, data, and resources to achieve global maternal and child undernutrition targets and the Sustainable Development Goals: an agenda for action. The Lancet. 2021 Apr;397(10282):1400\\u0026ndash;18. \\u003c/li\\u003e\\n\\u003cli\\u003eRuel MT, Menon P, Habicht JP, Loechl C, Bergeron G, Pelto G, et al. Age-based preventive targeting of food assistance and behaviour change and communication for reduction of childhood undernutrition in Haiti: a cluster randomised trial. The Lancet. 2008 Feb;371(9612):588\\u0026ndash;95. \\u003c/li\\u003e\\n\\u003cli\\u003eVerney AMJ, Busch‐Hallen JF, Walters DD, Rowe SN, Kurzawa ZA, Arabi M. Multiple micronutrient supplementation cost\\u0026ndash;benefit tool for informing maternal nutrition policy and investment decisions. Matern Child Nutr. 2023 Oct;19(4):e13523. \\u003c/li\\u003e\\n\\u003cli\\u003eThe Food Assistance Cost-Effectiveness Tool for Specialized Nutritious Foods [Internet]. Available from: https://foodaidquality.nutrition.tufts.edu/focus/cost-effectiveness-tools\\u003c/li\\u003e\\n\\u003cli\\u003eCMAM Costing Tool [Internet]. Available from: https://www.fantaproject.org/tools/cmam-costing-tool\\u003c/li\\u003e\\n\\u003cli\\u003eHolla-Bhar R, Iellamo A, Gupta A, Smith JP, Dadhich JP. Investing in breastfeeding \\u0026ndash; the world breastfeeding costing initiative. Int Breastfeed J. 2015 Dec;10(1):8. \\u003c/li\\u003e\\n\\u003cli\\u003eBrown KH, Engle-Stone R, Kagin J, Rettig E, Vosti SA. Use of Optimization Modeling for Selecting National Micronutrient Intervention Strategies: An Example Based on Potential Programs for Control of Vitamin A Deficiency in Cameroon. Food Nutr Bull. 2015 Sep;36(3_suppl):S141\\u0026ndash;8. \\u003c/li\\u003e\\n\\u003cli\\u003ePearson R, Killedar M, Petravic J, Kakietek JJ, Scott N, Grantham KL, et al. Optima Nutrition: an allocative efficiency tool to reduce childhood stunting by better targeting of nutrition-related interventions. BMC Public Health. 2018 Dec;18(1):384. \\u003c/li\\u003e\\n\\u003cli\\u003eWalker N, Tam Y, Friberg IK. Overview of the Lives Saved Tool (LiST). BMC Public Health. 2013;13(Suppl 3):S1. \\u003c/li\\u003e\\n\\u003cli\\u003eBaldissera Pacchetti M. Structural uncertainty through the lens of model building. Synthese. 2021 Nov;198(11):10377\\u0026ndash;93. \\u003c/li\\u003e\\n\\u003cli\\u003eVivarium [Internet]. Available from: https://vivarium.readthedocs.io/en/latest/\\u003c/li\\u003e\\n\\u003cli\\u003eHaddock B, Pletcher A, Blair-Stahn N, Keyes O, Kappel M, Bachmeier S, et al. Simulated data for census-scale entity resolution research without privacy restrictions: a large-scale dataset generated by individual-based modeling. Gates Open Res. 2024 May 3;8:36. \\u003c/li\\u003e\\n\\u003cli\\u003eKannan A, Tsoi D, Xie Y, Horst C, Collins J, Flaxman A. Cost-effectiveness of Vitamin A supplementation among children in three sub-Saharan African countries: An individual-based simulation model using estimates from Global Burden of Disease 2019. Horton S, editor. PLOS ONE. 2022 Apr 7;17(4):e0266495. \\u003c/li\\u003e\\n\\u003cli\\u003eBhattacharjee NV, Schumacher AE, Aali A, Abate YH, Abbasgholizadeh R, Abbasian M, et al. Global fertility in 204 countries and territories, 1950\\u0026ndash;2021, with forecasts to 2100: a comprehensive demographic analysis for the Global Burden of Disease Study 2021. The Lancet. 2024 Mar;S0140673624005506. \\u003c/li\\u003e\\n\\u003cli\\u003eBrauer M, Roth GA, Aravkin AY, Zheng P, Abate KH, Abate YH, et al. Global burden and strength of evidence for 88 risk factors in 204 countries and 811 subnational locations, 1990\\u0026ndash;2021: a systematic analysis for the Global Burden of Disease Study 2021. The Lancet. 2024 May 18;403(10440):2162\\u0026ndash;203. \\u003c/li\\u003e\\n\\u003cli\\u003eFerrari AJ, Santomauro DF, Aali A, Abate YH, Abbafati C, Abbastabar H, et al. Global incidence, prevalence, years lived with disability (YLDs), disability-adjusted life-years (DALYs), and healthy life expectancy (HALE) for 371 diseases and injuries in 204 countries and territories and 811 subnational locations, 1990\\u0026ndash;2021: a systematic analysis for the Global Burden of Disease Study 2021. The Lancet. 2024 Apr;S0140673624007578. \\u003c/li\\u003e\\n\\u003cli\\u003eNaghavi M, Ong KL, Aali A, Ababneh HS, Abate YH, Abbafati C, et al. Global burden of 288 causes of death and life expectancy decomposition in 204 countries and territories and 811 subnational locations, 1990\\u0026ndash;2021: a systematic analysis for the Global Burden of Disease Study 2021. The Lancet. 2024 Apr;S0140673624003672. \\u003c/li\\u003e\\n\\u003cli\\u003eSchumacher AE, Kyu HH, Aali A, Abbafati C, Abbas J, Abbasgholizadeh R, et al. Global age-sex-specific mortality, life expectancy, and population estimates in 204 countries and territories and 811 subnational locations, 1950\\u0026ndash;2021, and the impact of the COVID-19 pandemic: a comprehensive demographic analysis for the Global Burden of Disease Study 2021. The Lancet. 2024 Mar;S0140673624004768. \\u003c/li\\u003e\\n\\u003cli\\u003eHambidge KM, Westcott JE, Garc\\u0026eacute;s A, Figueroa L, Goudar SS, Dhaded SM, et al. A multicountry randomized controlled trial of comprehensive maternal nutrition supplementation initiated before conception: the Women First trial. Am J Clin Nutr. 2019 Feb;109(2):457\\u0026ndash;69. \\u003c/li\\u003e\\n\\u003cli\\u003eEthiopian Public Health Institute (EPHI) [Ethiopia] and ICF. Ethiopia Mini Demographic and Health Survey 2019 [Dataset]. ETKR81DT.ZIP. Rockville, Maryland, USA: EPHI and ICF [Producers]. ICF [Distributor], 2021. 2021. \\u003c/li\\u003e\\n\\u003cli\\u003eBailey J, Opondo C, Lelijveld N, Marron B, Onyo P, Musyoki EN, et al. A simplified, combined protocol versus standard treatment for acute malnutrition in children 6\\u0026ndash;59 months (ComPAS trial): A cluster-randomized controlled non-inferiority trial in Kenya and South Sudan. Tumwine JK, editor. PLOS Med. 2020 Jul 9;17(7):e1003192. \\u003c/li\\u003e\\n\\u003cli\\u003eJames P, Sadler K, Wondafrash M, Argaw A, Luo H, Geleta B, et al. Children with moderate acute malnutrition with no access to supplementary feeding programmes experience high rates of deterioration and no improvement: Results from a prospective cohort study in rural ethiopia. PLOS ONE. 2016 Apr 21;11(4):e0153530. \\u003c/li\\u003e\\n\\u003cli\\u003eMcGovern ME. How much does birth weight matter for child health in developing countries? Estimates from siblings and twins. Health Econ. 2019 Jan;28(1):3\\u0026ndash;22. \\u003c/li\\u003e\\n\\u003cli\\u003eOh C, Keats E, Bhutta Z. Vitamin and Mineral Supplementation During Pregnancy on Maternal, Birth, Child Health and Development Outcomes in Low- and Middle-Income Countries: A Systematic Review and Meta-Analysis. Nutrients. 2020 Feb 14;12(2):491. \\u003c/li\\u003e\\n\\u003cli\\u003eLassi ZS, Padhani ZA, Rabbani A, Rind F, Salam RA, Das JK, et al. Impact of Dietary Interventions during Pregnancy on Maternal, Neonatal, and Child Outcomes in Low- and Middle-Income Countries. Nutrients. 2020 Feb 19;12(2):531. \\u003c/li\\u003e\\n\\u003cli\\u003ePe\\u0026ntilde;a-Rosas JP, De-Regil LM, Gomez Malave H, Flores-Urrutia MC, Dowswell T. Intermittent oral iron supplementation during pregnancy. Cochrane Pregnancy and Childbirth Group, editor. Cochrane Database Syst Rev [Internet]. 2015 Oct 19 [cited 2024 Jan 19];2015(10). Available from: http://doi.wiley.com/10.1002/14651858.CD009997.pub2\\u003c/li\\u003e\\n\\u003cli\\u003eKeats EC, Haider BA, Tam E, Bhutta ZA. Multiple-micronutrient supplementation for women during pregnancy. Cochrane Pregnancy and Childbirth Group, editor. Cochrane Database Syst Rev [Internet]. 2019 Mar 15 [cited 2024 Jan 19]; Available from: https://doi.wiley.com/10.1002/14651858.CD004905.pub6\\u003c/li\\u003e\\n\\u003cli\\u003eOta E, Hori H, Mori R, Tobe-Gai R, Farrar D. Antenatal dietary education and supplementation to increase energy and protein intake. Cochrane Pregnancy and Childbirth Group, editor. Cochrane Database Syst Rev [Internet]. 2015 Jun 2 [cited 2024 Jan 19];2015(6). Available from: http://doi.wiley.com/10.1002/14651858.CD000032.pub3\\u003c/li\\u003e\\n\\u003cli\\u003eLi B, Zhang X, Peng X, Zhang S, Wang X, Zhu C. Folic Acid and Risk of Preterm Birth: A Meta-Analysis. Front Neurosci. 2019 Nov 28;13:1284. \\u003c/li\\u003e\\n\\u003cli\\u003eGomes F, Askari S, Black RE, Christian P, Dewey KG, Mwangi MN, et al. Antenatal multiple micronutrient supplements versus iron‐folic acid supplements and birth outcomes: Analysis by gestational age assessment method. Matern Child Nutr. 2023 Jul;19(3):e13509. \\u003c/li\\u003e\\n\\u003cli\\u003eDewey KG, Stewart CP, Wessells KR, Prado EL, Arnold CD. Small-quantity lipid-based nutrient supplements for the prevention of child malnutrition and promotion of healthy development: overview of individual participant data meta-analysis and programmatic implications. Am J Clin Nutr. 2021 Nov;114:3S-14S. \\u003c/li\\u003e\\n\\u003cli\\u003eHuybregts L, Le Port A, Becquey E, Zongrone A, Barba FM, Rawat R, et al. Impact on child acute malnutrition of integrating small-quantity lipid-based nutrient supplements into community-level screening for acute malnutrition: A cluster-randomized controlled trial in Mali. Persson L\\u0026Aring;, editor. PLOS Med. 2019 Aug 27;16(8):e1002892. \\u003c/li\\u003e\\n\\u003cli\\u003eFlaxman AD, Deason AW, Dolgert AJ, Mumford JE, Sorensen RJD, Eldrenkamp E, et al. Untangling uncertainty with common random numbers: a simulation study. Summer Simul Multi-Conf. 2017 Jul;(31):1\\u0026ndash;12. \\u003c/li\\u003e\\n\\u003cli\\u003eVirtanen P, Gommers R, Oliphant TE, Haberland M, Reddy T, Cournapeau D, et al. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods. 2020 Mar 2;17(3):261\\u0026ndash;72. \\u003c/li\\u003e\\n\\u003cli\\u003eIsanaka S, Andersen CT, Cousens S, Myatt M, Briend A, Krasevec J, et al. Improving estimates of the burden of severe wasting: analysis of secondary prevalence and incidence data from 352 sites. BMJ Glob Health. 2021 Mar;6(3):e004342. \\u003c/li\\u003e\\n\\u003cli\\u003eKlok RM, Postma MJ. Four quadrants of the cost-effectiveness plane: some considerations on the south-west quadrant. Expert Rev Pharmacoecon Outcomes Res. 2004 Dec;4(6):599\\u0026ndash;601. \\u003c/li\\u003e\\n\\u003cli\\u003eTong H, Piwoz E, Ruel MT, Brown KH, Black RE, Walker N. Maternal and child nutrition in the Lives Saved Tool: Results of a recent update. J Glob Health. 2022 Dec 30;12:08005. \\u003c/li\\u003e\\n\\u003cli\\u003eWessells KR, Arnold CD, Stewart CP, Prado EL, Abbeddou S, Adu-Afarwuah S, et al. Characteristics that modify the effect of small-quantity lipid-based nutrient supplementation on child anemia and micronutrient status: an individual participant data meta-analysis of randomized controlled trials. Am J Clin Nutr. 2021 Nov;114:68S-94S. \\u003c/li\\u003e\\n\\u003cli\\u003eThurstans S, Sessions N, Dolan C, Sadler K, Cichon B, Isanaka S, et al. The relationship between wasting and stunting in young children: A systematic review. Matern Child Nutr [Internet]. 2022 Jan [cited 2022 Sep 15];18(1). Available from: https://onlinelibrary.wiley.com/doi/10.1111/mcn.13246\\u003c/li\\u003e\\n\\u003cli\\u003eTroeger C, Colombara DV, Rao PC, Khalil IA, Brown A, Brewer TG, et al. Global disability-adjusted life-year estimates of long-term health burden and undernutrition attributable to diarrhoeal diseases in children younger than 5 years. Lancet Glob Health. 2018 Mar;6(3):e255\\u0026ndash;69. \\u003c/li\\u003e\\n\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":false,\"hideJournal\":false,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":true,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"bmc-global-and-public-health\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"\",\"sideBox\":\"Learn more about [BMC Global and Public Health](https://bmcglobalpublichealth.biomedcentral.com/)\",\"snPcode\":\"44263\",\"submissionUrl\":\"https://submission.springernature.com/new-submission/44263/3\",\"title\":\"BMC Global and Public Health\",\"twitterHandle\":\"@BMC_GPH\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"stoa\",\"reportingPortfolio\":\"BMC Series\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true},\"keywords\":\"Maternal and child health, simulation model, health policy, optimal allocation, antenatal care, young child feeding \",\"lastPublishedDoi\":\"10.21203/rs.3.rs-4651040/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-4651040/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003e\\u003cstrong\\u003eBackground:\\u003c/strong\\u003e Undernutrition remains a global crisis and is a focus of Sustain Development Goals. While there are multiple known, effective interventions, complex interactions between prevention and treatment and resource constraints can lead to difficulties in allocating funding. Simulation studies that use \\u003cem\\u003ein silico\\u003c/em\\u003e computer simulation can help illuminate the interactions between interventions and provide insight into the cost-effectiveness of alternative packages of options.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eMethods:\\u003c/strong\\u003e We developed an individual-based microsimulation model to test a range of antenatal and child nutrition interventions (iron and folic acid, multiple micronutrients, balanced energy protein supplementation, treatment for severe acute malnutrition, treatment for moderate acute malnutrition, and wasting prevention with small-quantity lipid-based nutrient supplements) and an analytic approach to processes the results of the microsimulation and provide the optimal intervention funding allocation for a given budget size.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eResults:\\u003c/strong\\u003e The reallocation of the baseline budget to minimize DALYs resulted in first funding multiple micronutrients to their maximum coverage, and then funding treatment for severe acute malnutrition. Relative to the baseline allocation, the reallocation optimized to minimizing DALYs resulted in 592,000 fewer annual DALYs.\\u003c/p\\u003e\\n\\u003cp\\u003eFor budgets larger than the baseline, our model recommended funding first targeted moderate acute malnutrition treatment, second universal moderate acute malnutrition treatment, third wasting prevention with small-quantity lipid-based nutrient supplements, and fourth balanced energy protein supplementation.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConclusions:\\u003c/strong\\u003e Our simulation is a novel model for estimating optimal allocation of spending on antenatal and child health nutrition interventions. We utilized a detailed health model that accounts for the interaction between preventive and therapeutic approaches. Our results show that an optimized reallocation of current spending on multiple micronutrients and treatment for severe acute malnutrition can substantially improve pregnancy-related and child health without additional funding. Additionally, we provide direction for how to best allocate expanded budgets to maximize impact.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Optimal Allocation of Antenatal and Young Child Nutrition Interventions: An Individual-Based GBD-Calibrated Microsimulation\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2024-07-22 20:57:21\",\"doi\":\"10.21203/rs.3.rs-4651040/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0},{\"type\":\"decision\",\"content\":\"Revision requested\",\"date\":\"2024-08-14T08:13:59+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2024-08-12T22:05:47+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2024-07-30T00:52:11+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2024-07-24T15:03:15+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"46706863672275119733891613986805588941\",\"date\":\"2024-07-16T15:32:56+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"158891823289770745713212149952720932596\",\"date\":\"2024-07-15T17:39:43+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"202557796314989243047488513953846027718\",\"date\":\"2024-07-10T16:04:26+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewersInvited\",\"content\":\"\",\"date\":\"2024-07-05T10:22:05+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorAssigned\",\"content\":\"\",\"date\":\"2024-07-01T09:06:58+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"checksComplete\",\"content\":\"\",\"date\":\"2024-06-28T09:19:19+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"submitted\",\"content\":\"BMC Global and Public Health\",\"date\":\"2024-06-27T22:11:43+00:00\",\"index\":\"\",\"fulltext\":\"\"}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"bmc-global-and-public-health\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"\",\"sideBox\":\"Learn more about [BMC Global and Public Health](https://bmcglobalpublichealth.biomedcentral.com/)\",\"snPcode\":\"44263\",\"submissionUrl\":\"https://submission.springernature.com/new-submission/44263/3\",\"title\":\"BMC Global and Public Health\",\"twitterHandle\":\"@BMC_GPH\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"stoa\",\"reportingPortfolio\":\"BMC Series\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"e0459050-6bc5-4754-89d0-79af2f18cc61\",\"owner\":[],\"postedDate\":\"July 22nd, 2024\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"under-review\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2024-12-18T10:23:06+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2024-07-22 20:57:21\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-4651040\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-4651040\",\"identity\":\"rs-4651040\",\"version\":[\"v1\"]},\"buildId\":\"qtupq5eGEP_6zYnWcrvyt\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}