#1USAID GEVEN USAID FROM THE AMERICAN PEOPLE Food Aid Quality Review Burkina Faso Effectiveness and Cost-Effectiveness Study: First Round of Results PRELIMINARY RESULTS January 31, 2018 Washington, D.C.#2Background: Improve Efficiency of Food Aid We need to "identify strategic opportunities to make global humanitarian assistance more effective and efficient, including prioritizing needs and reducing duplication and costs." Gregory C. Gottlieb, Acting Assistant Administrator, Bureau For Democracy, Conflict and Humanitarian Assistance, March 22, 2017 to Senate Committee on Foreign Relations USA Photo courtesy of USAID 2#3FY2016 - Title II Food Aid ~$2 billion spent (emergency + non-emergency) 60,000 metric tons of nutrition value-added products 73,000 metric tons of vegetable oil FY 2016 USAID TITLE II USE OF FUNDS 31% 19% 31% Commodity Value Ocean Freight Inland Freight 12% 7% DITSH 202(e) USAID FY 2016 FOOD ASSISTANCE TABLES 3#4. . . Big Questions: Are we spending public money wisely when it comes to selecting and delivering nutritionally enhanced products? Are we programming products optimally for desired outcomes? What roles do non-food factors play in driving outcomes? Can we manage wasting by preventing stunting (or visa versa)? Are stunting and wasting episodes related in ways that food aid can address? 4#5Institutional Roles and Research Project Responsibilities. USAID Study Activities Program Activities Title II Funds Donor Tufts University Study Management Data management and analysis ACDI/VOCA Ensure Commodity Reception/Treatment in Ouaga IRSS Data Collection and Entry: Field surveys and focus groups Anthro data collected at FDPS Save the Children Ensure delivery to caregivers & BCC programming ViM Program Food procurement and storage Food distribution at FDPS 5#6Today's Presentation . . . Introduction -Background and overview of study design Results -Effectiveness -Cost-effectiveness -Factors potentially influencing effectiveness Discussion -What have we learned? -What questions remain? -Implications 6#7Overview of study design USAID USAID FROM THE AMERICAN PEOPLE PRELIMINARY RESULTS 7#8Aims of the Burkina Faso Field Study Evaluate the comparative effectiveness and cost- effectiveness of 4 food aid products in preventing stunting and wasting in children 6-24 months of age in Burkina Faso, and describe potential influencing factors. FINED VEGETABLES TAMINA&D FORT USAID 8#9Study Design | Conceptual framework & Objectives Factors Potentially Influencing Effectiveness: Utilization of study food Study Arms: Effectiveness: CSB+ with oil Prevention of CSWB with oil, SC+ Stunting and Wasting RUSF Controlling for Potential Confounders: Household & Community level characteristics Cost- Effectiveness: Cost of stunting and wasting averted compared to CSB+ with oil 9#10Study Design | The 4 supplementary study foods Study Food Contents CSB Plus + Oil* (CSB+) Corn Soy Whey Blend + Oil* (CSWB) Cornmeal, whole soybeans, vitamin/mineral premix USAID USASS L-VEGETAL ALBOR Cornmeal, soy flour, whey protein concentrate, vitamin/mineral premix SUPER CEREAL plus Super Cereal Plus Corn, dehulled soybeans, dried (SC+) skim milk powder, vitamin/mineral premix USAID Ready-to-Use Oilseeds, tree nuts, pulses, Supplementary Food (RUSF) cereals, sugar, dairy protein, vegetable oil, vitamin/mineral premix *Fortified with Vitamin A & D Monthly ration ~500 kcal/day (as delivered). USAID RUSF 10#11Study Setting . . Existing Title II USAID supplementary feeding program (VIM) targeting all pregnant and lactating mothers and children 6-23 months in Sanmatenga Province Four geographic regions randomly assigned one of four foods distributed at 48 distribution sites YILLARE DINTERVENTION OU PROVI SC+ CSWB RUSF CSB+#12Study Sampling ● . Target of 6,000 children, enrolled at age ~6 months (1,500 per study food arm) Each child measured monthly for ~18 months during the ViM program plus 3 consecutive monthly post-intervention follow-ups . Randomly selected sub- samples for interviews and in- home observations YILLARE DINTERVENTION OU PROVI CSWB RUSF SC+ CSB+ 12#13Data Collection Methods ☐ Effectiveness Anthropometry: monthly ■ Morbidity & treatment survey: monthly ■ Socio-economic survey: enrollment & exit ■ Community questionnaire: enrollment Cost-Effectiveness ■ Distribution observations ■ Warehouse observations ■ Individual surveys with beneficiaries & lead mothers ■ In-home observations with beneficiaries Influencing Factors ■ Distribution observations ■ Individual surveys with beneficiaries, lead mothers, health & nutrition promoters ■ In-home observations with beneficiaries ■ Focus Group Discussions with beneficiaries & distribution committees 13#14USAID GEVEN USAID FROM THE AMERICAN PEOPLE Effectiveness Results PRELIMINARY RESULTS 14#15Conceptual Framework Factors Potentially Influencing Effectiveness: Utilization of study food Study Arms: CSB+ with oil CSWB with oil, SC+ RUSF Effectiveness: Prevention of Stunting and Wasting Cost- Effectiveness: Cost of stunting and wasting averted compared to CSB+ with oil Controlling for Potential Confounders: Household & Community level characteristics 15#16Objectives SPOR . Compare the effectiveness of four foods as used in a real- world programmatic setting in preventing: Stunting (length-for-age Z- score < -2) - Wasting (weight-for-length Z-score < -2) Evaluate 3-month post- intervention follow-up. ― Measurement where child's age was between 22.9 and 23.9 months 16#17Methods | Longitudinal Data Collection Enrollment Household, SES, and community level data collected at enrollment ~6 months of age Followed for 18 months Monthly anthropometric data collected at each food distribution ~18 distributions Post-intervention follow-up Anthropometric data collected at 1, 2, and 3 months post food distribution Aug. 2014 - Jul. 2015 - Aug. 2014 Sept. 2016 Feb. 2016 – Dec. 2016 - 17#18Methods | Data Analysis Selection of covariates based on conceptual framework Statistical models → Y = ẞo + ẞstudy food + Bindividual characteristics ... ẞcommunity characteristics + ε 1 Prevalence of stunting at endline → Logistic Regression - Total number of months wasted → Negative Binomial Regression Mean LAZ and WLZ throughout study period → Mixed- effects Regression - Time to first stunted and wasted measures → Cox Model Models fit with and without those who had an endline measurement ● Multiple imputation for missing data . Models evaluated for collinearity and influential outliers 18#19Study Arms Similar with Regard to Programmatic Exposure and Loss-to-follow-up Total Study Sample n=6,112 Lost to follow-up No endline measurement n=1,503 SC+ n=1,564 RUSF n=1,526 CSB+ n=1,519 CSWB Total n=908 (15%) Distributions received 17 ± 4 Distributions received 17 ± 4 Distributions received 17 ± 3 Distributions received 17 ± 3 LTFU n=207 (14%) LTFU n=248 (17%) LTFU n=240 (15%) LTFU n=213 (14%) 19#20Enrollment Characteristics CSB+ (n=1519) CSWB (n=1503) SC+ (n=1564) RUSF (n=1526) Age (months) 5.9 ± 0.8 6.6 ± 0.9 6.3 ± 0.7 6.2 ± 1.2 Male sex 774 (51%) 779 (52%) 802 (51%) 755 (49%) Severely food 230 (15%) 201 (14%) 204 (13%) 209 (14%) insecure Poorest SES 375 (25%) 267 (18%) 294 (19%) 260 (17%) LAZ -0.72 ± 1.07 -0.56 ± 1.16 -0.53 ± 1.04 -0.58 ± 1.11 WLZ -0.54 ± 1.05 -0.58 ± 1.04 -0.44 ± 1.04 -0.59 ± 1.04 MUAC Z -0.46 ± 0.97 -0.43 ± 0.96 -0.31 ± 0.96 -0.45 ± 0.96 Stunted ~6 mos. 148 (10%) 135 (9%) 110 (7%) 138 (9%) Wasted ~6 mos. 121 (8%) 125 (8%) 111 (7%) 121 (8%) 20 20#21USAID Results on Stunting Prevention USAID FROM THE AMERICAN PEOPLE PRELIMINARY RESULTS 21#22The odds of stunting at endline were similar in SC+ and RUSF; twice as high in the CSWB arm 20% 28% 22% 20% Adjusted Ref. 2.07* 1.02 1.21 ORS → *p<0.05 CSB+ CSWB SC+ RUSF 22#23LAZ declined in all arms, with greatest decline in the CSWB arm Adjusted Predictions 9- -.8 ड- LAZ -1.4 -1.2 -1 HH 6 7 8 墨玉 H HH 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Age in months CSB+ SC+ CSWB RUSF 23#24Non-stunted time was similar among the arms, but slightly lower in the CSWB arm Proportion not stunted .25 .5 .75 0 2 4 6 8 10 12 14 16 18 Analysis Time (months) CSB+ SC+ CSWB RUSF 24#25- Results on Wasting Prevention USAID USAID FROM THE AMERICAN PEOPLE PRELIMINARY RESULTS 25#26Number of monthly measures wasted were similar in the SC+ and RUSF arms, but higher in the CSWB arm 2.62 ADJUSTED NUMBER OF MONTHLY MEASURES WASTED 3.29 Adjusted Ref. 1.25* IRRS → *p<0.05 2.51 2.42 I 0.96 0.92 CSB+ CSWB SC+ RUSF 26#27WLZ -1 -.9 WLZ trajectories are similar, with the RUSF arm showing a slower rate of decline Adjusted Predictions -.8 9- L- 6 7 8 HH HTH HHHH HIGH HAAHH HINA 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Age in months CSB+ SC+ CSWB RUSF 27#28Non-wasted time was similar among the arms, but slightly lower in the CSWB arm Proportion not wasted .25 .5 .75 0 2 6 8 10 12 14 16 18 Analysis Time (months) CSB+ SC+ CSWB RUSF 28#29Relationships between stunting and wasting USAID USAID FROM THE AMERICAN PEOPLE PRELIMINARY RESULTS 29 29#30Coexistence of stunting and wasting • At Baseline (6 months) . . . - 13% of stunted children were also wasted – 15% of wasted children were also stunted 64% of those who were stunted at endline were also wasted at least once throughout the study period Of those who were stunted at endline, the percentage never wasted ranged from 32-44% by study arm Of those who were ever stunted, 30% had their first stunted measure preceded by wasting 30#31Percentages never wasted among those who were stunted at endline 44% 34% 36% 32% CSB+ CSWB SC+ RUSF 31#32Summary points None of the foods prevented linear growth faltering • Children in the CSWB arm experienced higher rates of stunting and wasting, while rates in the SC+ and RUSF arms were similar to CSB+ • Understanding these effectiveness results requires understanding factors that may influence effectiveness 32#33USAID GEVEN USAID FROM THE AMERICAN PEOPLE Factors Influencing Effectiveness PRELIMINARY RESULTS 33#34Conceptual Framework Factors Potentially Influencing Effectiveness: Utilization of study food Study Arms: CSB+ with oil CSWB with oil, SC+ RUSF Effectiveness: Prevention of Stunting and Wasting Cost- Effectiveness: Cost of stunting and wasting averted compared to CSB+ with oil Controlling for Potential Confounders: Household & Community level characteristics 34#35Need to understand influencing factors Behaviors explored: Sharing • Preparation Consumption 2015 269 35#36Qualitative Data Collection Methods - In-home observations (n=209) - Individual interviews with beneficiary mothers (n=1,463) - Focus Group Discussions with beneficiary mothers and distribution committees (n=48) - Lead mother and promoter interviews (n=308) 36#37USAID Results - Diversion (sharing, giving away or selling ration) USAID FROM THE AMERICAN PEOPLE PRELIMINARY RESULTS 37#38Ration diversion: Caregivers report diverting the ration from the intended beneficiary most often in the CSWB arm 75% 74% 68% 69% 54% OVERALL CSB+ CSWB SC+ RUSF (N=1,632) (N=415) (N=406) (N=409) (N=401) No selling reported in any study arm Giving away any of the ration was rare (8-13%) Giving away oil was also rare (7-8%) 18-21% of people report using oil for other household cooking 38#39Observed presence of food in households: Study foods were observed to be present in the households most often in CSB+ and least often in CSWB ■Study food present ■Study food not present I Not documented 92% 60% 53% 55% 45% 25%, 25% °20% 15% 4% 4% 2% CSB+ (N=50) CSWB (N=55) SC+ (N=51) RUSF (N=53) Average over 4 days of observation 39#40How often the ration reportedly lasts the entire month: The ration lasts the entire month least often in the CSWB arm 35% 33% 25% 57% CSB+ CSWB SC+ RUSF 40 40#41Someone other than beneficiary observed eating the food: Others were observed eating the food most often in the CSWB arm 36% 24% 25% 23% CSB+ (N=46) CSWB SC+ RUSF (N=29) (N=28) (N=32) When study food was present, HH average over 4 days of observation 41#42Adjusted percentages of diversion: Caregivers are more likely to report diverting CSWB than the other foods 71% 57% 68% 44% CSB+ CSWB SC+ RUSF Adjusted ORs → Ref. 1.9* 1.63* 0.60* *p<0.05 42#43Ration sharing quotes from focus groups: "They told us that if the mother eats some herself at night, the next morning she will have plenty of milk for her child." "It's true that it was said that it is for the child only, but often we give some to his older brothers, that's how it is. If we do not give it to them they will cry, so we give them a bit to calm them down and we save the rest." "Often the mother tastes a bit and if the big brother of the child is nearby we also give him some." "Even when we are getting ready to make porridge, the big brother can approach and say 'can I have some?'. We cannot refuse so we give him a bit. Everywhere, it's the same.” 43#44USAID Results - Preparation USAID FROM THE AMERICAN PEOPLE PRELIMINARY RESULTS 44#45Fat content in household porridge samples: Samples indicate lower fat content than would be expected if porridge prepared according to recommendations 20 fat g/100 g 40 09 09 Intrinsic Oil Required fat content range for RUSF Target quantity of added oil (FAQR) CSB+ CSWB SC+ Average qty added: 7.3 g/100g 6.6 g/100g 45#46Contamination with E.coli: In all study arms, the majority of household water samples showed unsafe or high-risk levels of contamination with E.coli 13 9 20 20 58 Low risk Intermediate Risk 1210 20 58 High Risk Unsafe 13 17 27 45 19 12 10 58 CSB+ (N=400) CSWB (N=393) SC+ (N=400) RUSF (N=387) 46#47USAID Results - Consumption USAID FROM THE AMERICAN PEOPLE PRELIMINARY RESULTS 47#48Form in which ration is consumed: Those in RUSF and CSB+ consume the ration most often in the recommended form CSB+ (n=402) Other Raw 1% 1% Couscous 14% Raw 20% Porridge 84% CSWB (n=397) Other 2% Couscous Raw 1% 18% Porridge 79% SC+ (n=405) RUSF (n=395) Other 1% Mixed with other foods Other 2% 5% Porridge 79% Directly from packet 93% 48#49Form in which the ration is consumed from focus group "Well, if I make porridge today for him and he doesn't want it, and tomorrow again it's the same thing, I can try to make couscous with a bit of oil to see if he'll eat that. If he eats it once or twice, you know that the child prefers couscous to the porridge and I make that from now on all the time for him. And vice versa." "They told us that if the child does not like the porridge, we can simply give him the flour for him to eat raw." 49#50Adjusted percentages for recommended form: Caregivers report consuming CSWB in its recommended form significantly less often than the other foods, and RUSF significantly more often 72% 61% 74% 82% CSB+ CSWB SC+ RUSF Adjusted ORS → Ref. 0.60* 1.14 1.80* *p<0.05 50#51Observed consumption: Children were most often observed consuming the study food in the RUSF arm, and least often in the CSWB arm 38% 18% 60% 52% CSB+ (N=46) CSWB (N=29) SC+ (N=28) RUSF (N=32) When study food was present, HH average over 4 days of observation 110 ODLI 51#52Anecdotal Report: Bitter taste of CSWB ● ● CSWB reported to be bitter by beneficiaries at the end of IDIs and to ViM staff Tests showed that the bitterness did not impact the safety of the food ● ● ViM recalled and destroyed 14 MT of bitter CSWB after organoleptic tests confirmed bitter taste From organoleptic tests at NC State, June 2016: - Aftertaste: "Stale, cardboard, heat exposure, dirty" - Aroma quality: "Cardboard, fishy" - Higher bitterness scores than new batch of CSWB 52#53USAID Summary Points USAID FROM THE AMERICAN PEOPLE PRELIMINARY RESULTS 53#54Summary Points ● ● Study foods were generally not used as intended Patterns of use/sharing of the study foods within the households varied by study arm CSWB was shared more and eaten less frequently by beneficiary child CSWB reported to be bitter after 8-9 months of storage time 54#55USAID GEVEN USAID FROM THE AMERICAN PEOPLE Cost and Cost-Effectiveness PRELIMINARY RESULTS 55#56Conceptual Framework Factors Potentially Influencing Effectiveness: Utilization of study food 4 Study Arms: CSB+ with oil CSWB with oil, SC+ RUSF Effectiveness: Prevention of Stunting and Wasting Cost- Effectiveness: Cost of stunting and wasting averted compared to CSB+ with oil Controlling for Potential Confounders: Household & Community level characteristics 56#57Costing Methods • Two Costing Perspectives: • - Program Perspective (e.g. Product, Supply Chain, Distribution, etc) - Caregiver Perspective (Opportunity Cost of Caregiver Time) Activity-based Costing Approach Ongoing Cost (existing program, no start-up cost) Two Cost Lenses (for procurement and international freight): - Study-Incurred - Realistic 57#58Costing Matrix International Freight Inland Transportation Food Product/ Commodity Transport Loss- adjusted Storage Repacking (CSB flours) Administrative and Fixed Costs Cost by Component Reconditioning (Oil) Caregiver Opportunity Cost Summary Statistics Distribution Cost/MT Average Cost per Metric Ton Cost/Ration Average Cost per Monthly Ration Cost/Child Average Cost per Child Reached 58#59Assessing Cost-Effectiveness Across Study Arms Δ Cost Estimates Δ Measured Effect Difference in Average Cost per Child Reached 00 Primary Outcome for Stunting • Difference in % Stunted at Endline • Primary Outcome for Wasting Difference in # of Monthly Measurements Showing Wasting Comparative Cost- Effectiveness Each Arm Compared to Reference Arm (CSB+ with oil) 59#60| Results on Costs: Program Perspective Image Source: UNICEF PRELIMINARY RESULTS USAID USAID FROM THE AMERICAN PEOPLE W 60 60#614000 3500 3000 2500 Cost/MT ($USD) Breakdown by Food Product and Supply Chain (Realistic Procurement & Freight) 3707 2000 1811 1500 1000 500 1169 1532 Storage Reconditioning 2704 ■Repacking 0 Oil CSB+ CSWB RUSF SC+ ■Inland transportation International Freight Food Product/Commodity 61#62300 250 200 150 100 50 50 122 Cost per Child Reached ($USD) Breakdown by Food and Other Program Cost (Realistic Procurement & Freight) 253 Administrative & 135 0 CSB+ w/oil CSWB w/ oil RUSF SC+ 223 Fixed Cost Distribution ■Storage ■Reconditioning ■Repacking Inland transportation ■International Freight Food Product/Commodity 62#63Cost- Effectiveness USA Results USAID RUSF PRELIMINARY RESULTS USAID USAID FROM THE AMERICAN PEOPLE USAID O THE AMERICAN PROREK RUTF GH ENERGY BISCUITS Photo courtesy of USAID SUPER CEREAL plus Corn Soya Blend Special Formula for Infants and ung des 6 months and above 1.5 KG Net PO 4210001619 19 11-33-14 Beat Used By 00 2016 Lote 0115 USAID TROM THE AMERICAN SORE 63#64Linking Cost to Effectiveness: Stunting ― Average Cost per Child Reached ($USD) - % Stunted at Endline CSB+ CSWB Analysis: Excluding Lost-To-Follow-Up RUSF SC+ w/ oil w/ oil Program Cost per Child Reached 122.4 135.2 252.4 223.0 *Realistic Procurement & Freight Predicted % Stunted at Endline (95%CI) 20.1% 27.5% 20.3% 21.9% (18.0%, (25.0%, (18.3%, (20.0%, *Adjusted for baseline stunting and other confounders 22.2%) 30.0%) 22.4%) 23.9%) 64 44#65Incremental Cost-Effectiveness Plane for Stunting Prevention Perspective: Program Only; Realistic Procurement Cost 125 (-1.8%, $130.6) Difference in Program Cost per Child Reached Compared to CSB+ w/ oil($) -75 -100 0 25 75 100 150 175 200 More Costly (-0.2% $100.6) 50 p<0.001 (-7.4%, $12.8) -10% -5% -25 (0% $0) 0% 5% 10% Less Effective More Effective -50 Less Costly Predicted Difference in Averting Stunting at Endline Compared to CSB+ w/ oil(%) SC+, RUSF, and CSWB w/ oil Compared to CSB+ w/ oil(Reference Arm) Arm SC+ RUSF CSWB w/ oil CSB+ w/ oil 65#66Linking Cost to Effectiveness: Wasting - Average Cost per Child Reached ($USD) – # Monthly Measurements Showing Wasting per Child Analysis: Excluding Lost-To-Follow-Up Program Cost per Child Reached *Realistic Procurement & Freight CSB+ w/ CSWB RUSF SC+ oil w/ oil 122.41 135.2 252.4 223.0 Predicted # Monthly Measurements Showing Wasting per Child (95%CI) *Adjusted for total number of monthly measurements received and other confounders 2.6 3.3 2.4 2.5 (2.3, 3.0) (2.8, 3.7) (2.1, 2.7) (2.2, 2.8) 66#67Incremental Cost-Effectiveness Plane for Wasting Prevention Perspective: Program Only; Realistic Procurement Cost More Costly 200 175 150 125 100 555 75 50 50 Difference in Program Cost per Child Reached Compared to CSB+ w/ oil($) 25 25 (-0.7, $12.8) p=0.011 (√2 $130.6) (0.1, $100.6) (0, $0) -1.25 -1.00 -0.75 -0.50 -0.25 0.00 0.25 0.50 0.75 1.00 -25 Less Effective More Effective -50 -75 -100 Arm SC+ RUSF CSWB w/ oil CSB+ w/ oil Less Costly Predicted Difference in Monthly Measurements Showing Wasting per Child Compared to CSB+ w/oil SC+, RUSF, and CSWB w/ oil Compared to CSB+ w/ oil(Reference Arm) 67#68Caregiver Perspective PRELIMINARY RESULTS USAID USAID FROM THE AMERICAN PEOPLE SOFALLS 00#69Opportunity Cost of Caregiver Time Opportunity cost of time = hours of time x estimated hourly wage Self-reported & observed time (Mother IDI and IHO) $0.21/hr Calculated by: Mandated minimum monthly wage in the formal sector: 30684 CFA • Standard workweek of 60 hours for household workers 69#70Opportunity Cost of Caregiver Time per Monthly Ration Breakdown by Caregivers' Activity Types $12.00 $10.00 $8.00 $6.00 $4.00 $2.00 $0.00 CSB+ w/oil CSWB w/oil RUSF SC+ Study Food Feeding ■Study Food Preparation ■Waiting at Distribution Site ■Traveling to Distribution Site 70 70#71USAID The Impact of Adding Caregiver Opportunity Cost on Comparative Total Cost/Child by Arm USAID FROM THE AMERICAN PEOPLE PRELIMINARY RESULTS 71#72450 $Cost/Child Reached Program & Caregiver Perspective (Realistic Procurement & Freight) 400 350 300 250 200 174 152 150 100 50 179 99 0 CSB+ w/oil CSWB w/ RUSF SC+ oil ■Caregiver Opportunity Cost @$0.21/hr ■Total Cost to Program 72 12#73200 Incremental Cost-Effectiveness Plane for Stunting Prevention Perspective: Program Only: Realistic Procurement Cost More Costly 125 (-1.8%, $130.6) Difference in Program Cost per Child Reached Compared to CSB+ w/ oil($) -100 -75 50 0 25 75 100 150 175 50 p<0.001 (-7.4%, $12.8) -10% -25 Less Effective -50 H (-0.2% $100.6) -5% (0% $0) 0% 5% 10% More Effective Less Costly Predicted Difference in Averting Stunting at Endline Compared to CSB+ w/ oil(%) SC+, RUSF, and CSWB w/ oil Compared to CSB+ w/ oil(Reference Arm) Arm SC+ RUSF CSWB w/ oil CSB+ w/ oil 73#74200 175 Incremental Cost-Effectiveness Plane for Stunting Prevention Perspective: Program & Caregiver Realistic Procurement Cost More Costly Diff. in Program & Caregiver Cost per Child Reached Compared to CSB+ w/ oil($) -75 -100 0 25 50 75 100 125 150 p<0.001 (-0.2%, $105.8) (-1.8% $56) -15% -25 -10% (-7.4%, $-9.7)-5% (0% $0) 0% 5% 10% 15% Less Effective More Effective -50 Less Costly Predicted Difference in Averting Stunting at Endline Compared to CSB+ w/ oil(%) SC+, RUSF, and CSWB w/ oil Compared to CSB+ w/ oil(Reference Arm) Arm SC+ RUSF CSWB w/ oil CSB+ w/ oil 74#75Incremental Cost-Effectiveness Plane for Wasting Prevention Perspective: Program Only; Realistic Procurement Cost More Costly 200 175 150 125 100 555 75 50 50 Difference in Program Cost per Child Reached Compared to CSB+ w/ oil($) 25 25 (-0.7, $12.8) p=0.011 (√2 $130.6) (0.1, $100.6) (0, $0) -1.25 -1.00 -0.75 -0.50 -0.25 0.00 0.25 0.50 0.75 1.00 -25 Less Effective More Effective -50 -75 -100 Arm SC+ RUSF CSWB w/ oil CSB+ w/ oil Less Costly Predicted Difference in Monthly Measurements Showing Wasting per Child Compared to CSB+ w/oil SC+, RUSF, and CSWB w/ oil Compared to CSB+ w/ oil(Reference Arm) 75#76200 175 Incremental Cost-Effectiveness Plane for Wasting Prevention Perspective. Program & Caregiver: Realistic Procurement Cost More Costly 150 125 100 15 75 50 50 Diff. in Program & Caregiver Cost per Child Reached Compared to CSB+ w/ oil($) 25 25 (0.1, $105.8) SC+ RUSF (0.2, $56) Arm (0, $0) (-0.7, $-9.7) 0 -1.25 -25 -1.00 Less Effective -0.75 -0.50 -0.25 0.00 0.25 0.50 0.75 1.00 p=0.011 More Effective -50 CSWB w/ oil CSB+ w/ oil -75 -100 Less Costly Predicted Difference in Monthly Measurements Showing Wasting per Child Compared to CSB+ w/oil SC+, RUSF, and CSWB Compared to CSB+ w/ oil(Reference Arm) 76#77Summary Points • Cost and cost-effectiveness analysis provides information needed in the selection of supplementary foods. • • Cost-Effectiveness Results for stunting and wasting are similar: - CSB+ with oil was most cost-effective - RUSF was the most expensive from the program perspective; SC+ was most expensive from the program & caregiver perspective - CSWB had the second lowest cost, but also lower effectiveness Valuing caregiver time affects relative rankings of cost-effectiveness, but not conclusion that CSB+/oil was most cost-effective in this study. 77#78USAID GEVEN USAID FROM THE AMERICAN PEOPLE What have we learned? PRELIMINARY RESULTS 78#79Recent Systematic Review on targeted treatment of MAM shows similar findings . "RUTF used as part of the WHO 2013 protocol is effective in promoting recovery from SAM and reducing mortality." "Use of RUSF to treat MAM.. has the potential to improve nutritional recovery.” [treatment] HACH SISTEMATIC REVIEW RECOVERY, RELAPSE AND EPISODES OF DEFAULT IN THE MANAGEMENT OF ACUTE MALNUTRITION IN CHILDREN IN HUMANITARIAN EMERGENCIES "Weight gain reported by studies for children given RUSF or CSB was small. There were no substantial differences in mortality rates reported for children who received RUSF compared with children in CSB groups.” H 79#80Fig 5. Forest plot: Lipid-based nutrient supplement versus Specially formulated micronutrient fortified foods, outcome: 1.3 Recovery from moderate acute malnutrition. Study or Subgroup 1.3.2 RUSF Ackatia-Armah 2015 (C Delchevalerie 2015 (C) LNS Control Risk Ratio Events Total Events Total Weight M-H, Random, 95% CI 328 335 020 74.30 14.00 1201 Risk Ratio M-H, Random, 95% CI Karakochuk 2012 (C) LaGrone 2012 Matilsky 2009 Medoua 2016 Nikiema 2014 (C) Subtotal (95% CI) Total events Heterogeneity: Tau² = Test for overall effect: 2 1.3.3 RUTF Nackers 2010 Subtotal (95% CI) Total events Heterogeneity: Not applicable "Subset analyses suggested higher recovery rates with greater amount of calories provided." Test for overall effect: Z = 3.43 (P = 0.0006) Total (95% CI) 4683 Total events 3812 4251 100.0% 1.08 [1.02, 1.14] 3066 Heterogeneity: Tau² = 0.00; Chi² = 34.74, df = 7 (P < 0.0001); 1² = 80% Test for overall effect: Z = 2.69 (P=0.007) Test for subgroup differences: Chi 5.17, df = 1 (P = 0.02), 12 80.7% 0.01 0.1 1 10 100 Favours [LNS] Favours (control] Gera T, Pena-Rosas JP, Boy-Mena E, Sachdev HS (2017) Lipid based nutrient supplements (LNS) for treatment of children (6 months to 59 months) with moderate acute malnutrition (MAM): A systematic review. PLOS ONE 12(9): e0182096. https://doi.org/10.1371/journal.pone.0182096 http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0182096 PLOS ONE TENTH ANNIVERSARY#81Fig 12. Forest plot: Lipid-based nutrient supplement versus specially formulated micronutrient fortified foods, outcome: 1.23 Relapse after discharge. LNS Control Study or Subgroup Events Total Events Total Weight Risk Ratio M-H, Random, 95% CI LaGrone 2012 Nackers 2010 224 1314 33 170 110 653 81.0% 33 152 19.0% 1.01 [0.82, 1.25] 0.89 (0.58, 1.37] Total (95% CI) 1484 805 100.0% 0.99 [0.82, 1.19] Total events 257 143 Heterogeneity: Tau2 0.00; Chi 0.26, df = 1 (P = 0.61); 12=0% Test for overall effect: Z = 0.12 (P=0.90) Risk Ratio M-H, Random, 95% CI 0.1 0.2 0.5 1 2 5 10 Favours [LNS] Favours [control] Gera T, Pena-Rosas JP, Boy-Mena E, Sachdev HS (2017) Lipid based nutrient supplements (LNS) for treatment of children (6 months to 59 months) with moderate acute malnutrition (MAM): A systematic review. PLOS ONE 12(9): e0182096. https://doi.org/10.1371/journal.pone.0182096 http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0182096 PLOS ONE TENTH ANNIVERSARY#82. ● Some things were confirmed by this study Foods used in blanket supplementation in this context had limited effect in preventing stunting and wasting. Food products have to be consumed by intended index individual to have any chance of achieving intended outcome. Mostly this is not happening, regardless of product. Some ingredients may matter more (protein source, ASF, potassium, zinc, amino acids, etc.). But all foods performed in similar ways, so beware too much tinkering a) absent evidence of real physiological significance b) more findings on dosing. 82#83Height for age Z-scores (WHO) 1.5 1,25 Growth Faltering: What, when, and can we disrupt the curve? 1 0,75 0.5 0,25 0 -0,25 -0,5 -0,75 -1 -1,25 -1.5 -1,75 -2 -2,25 -2,5 1 Victora et al. 2010 10 13 16 19 22 25 31 Age (months) 37 40 It 49 - EURO EMRO AFRO PAHO SEARO 52 55 58 མཚ་ 83#84WHO growth curves Length-for-age GIRLS Birth to 2 years (percentiles) Length (am) Birth 10 Age (comp " Length(cm) 76 26 65 " 11 1 97 854 50th World Health Organization 3rd " Burkina study children 71 Length-for-age BOYS Birth to 2 years (percentiles) World Health Organization 97th 85th 50th 3rd 74 4 Mortha Birth 1 year 2 years Age (completed months and years) WHO Child Growth Standards 84#85What have we learned? (i) 1. None of the foods prevented decline in LAZ over time. All foods performed comparably, but carry very different costs. 2. Most cost-effective re: defined outcome was CSB+. 3. Least cost-effective: CSWB (least often observed in-home; most shared; most often reported 'not lasting entire month'). 4. We can't ascribe performance to product composition. Relatively poor performance of children in the CSWB arm does NOT mean that adding WPC80 resulted in stunting. 85#86Other PM2A evidence: food security matters . South Sudan (Doocy et al 2013) - "Household food security remained poor despite ration receipt." Burundi (Leroy et al 2017) - "Positive effect on household access to food... [but minimal impact] on household and mothers' dietary diversity."" • Guatemala (Ruel et al 2017) - "Larger [family] ration led to greater participation and impacts." • Haiti (Donegan et al 2013) - "Rigorous evaluations of food-assisted... programs are stymied by the ethics of randomizing recipients to a control." 86#87What have we learned (from Burkina Faso)? (ii) 5. Children in CSWB arm experiencing stunting earlier (than in other arms) may suggest insufficient food intake not ameliorated by supplementary food (which was shared more). 6. Sharing/diversion of foods typical for all foods (incl. >50% RUSF); children in CSWB arm seem least likely to be consuming adequate product. [We are not seeing reports of 'bitter taste' of CSWB in ongoing Sierra Leone study.] 7. Quality of programming matters at least as much as quality of products being programmed. We need much more evidence of effective programming actions surrounding delivery of food aid. 87#88What have we learned? (iii) 8. To better understand programming requires much more attention to drivers of community participation, compliance, substitution, links among programming components, etc. 88#89"While WASH alone will not eliminate stunting, it does have the potential to accelerate progress on eliminating stunting as a critical component of comprehensive strategies." Maternal & Child Maternal & Child Nutrition Nutrition Review Article DOK: 103311/12258 Can water, sanitation and hygiene help eliminate stunting? Current evidence and policy implications Oliver Cumming and Sandy Cairncross Dva of Abstract Corty of feat Dan andand make a significant contribution to edixing the p Stunting is a complex and enduring challenge with far-reaching consequences for those affected and society as a whole. To accelerate progress in eliminating stunting, becoder efforts are needed that reach beyond the nutrition sector to tackle the underlying determinants of undermatition. There is growing interest in how water, sanitation and hygiene (WASH) interventions might s port strategies to reduce o stunting in high-burden settings, such as South Asia and sub-Saharan A n Africa. This review article considers two broad questione (1) can WASH intervention and (2) can WASH in- be delivered to conditions have a significant detrimental elect on child p d growth and development resulting from sustained exposure to enteric pathogen but also due to wider social and economic mechar Realing the potential of WASH to reduce stunting requires a redoubling of efforts to achieve unmersal access to these ser vices in envisaged under the Sustainable Development Goals. It may also require new or modified WASH strategies that po beyond the scope of traditional interventions to spectically address exposure pathways in the first 2 years of life when the process of stunting is concentrated. gests that poor WASH their effect dull red terventions be Keyword sanitation, water, stunting, child nutrition, child public bealth, early growth Deportment of Deco Facilty of dion Tropical case London School of Higene and Dapical Medicine, London UK. Ek divecomm Introduction This article was inspired by the Stop Standing Conder ence held in Delhi last year to comene actors from mid tiple countries and sectors to address a shared concenc the enduring and seemingly intractaNe challenge of childhood stunting in South Asia. Hspe progress has been made in much of the South Asia region in extend ing healthcare, education and economic opportunity, und these investments have brought dramatic improve mends in maternal and child mortality, in school retention rates and in overall economic output. Despite this able progress, the prevalence of childhood stunting in South Asia remains high with profound aequences for those children affected increasing their suscepity to infectious disease morbidity and mortality, diminishing their future educational achievements and reducing their economic productivity in later life. The failure to address stunting in South Asia, and other high burden regions stands to undermine progress in other sectors and trap- ping future generations is poverty and ill health Stunting is a complex problem as depicted by various conceptual frameworks, focused on 'child malnutrition (UNKEF 1990 matemal and child undernutrition (Back in 2013) and food and nutrition security (Grower al 2000). The causes of stunting are multifacto tial and inter-linked spanning biological, social and cavi ronmental spheres Water, sanitation and hypicne (WASH) the focus of this paper, feature at various level in these frameworks with varying degrees of proximity to the outcome of stunting, as immediate or proximate risk factors but also as more distant causes or determinants of stunting For example, different aspects of WASH hone been plisbly linked to all four pillars' of the food and nutrition security framework (Cumming et al in press food availability, through water as a resource 0306 The Aun Mama&C Non poblebed by Way & Se1ad Mam & Chit Now (36, 12 Sep 11- This in an open access artide under the terms of the Chrative Conanens Anibution Lime, which permits debut and reprod sion in any mod provided the original sopper 9 68#90"There was no benefit to the integration of water, sanitation, and handwashing with nutrition. Adherence was high in all groups and diarrhoea prevalence was reduced in all intervention groups except water treatment. Combined water, sanitation, and handwashing interventions provided no additive benefit over single interventions." Effects of water quality, sanitation, handwashing, and nutritional interventions on diarrhoea and child growth in rural Bangladesh: a cluster randomised controlled trial Stephen Pady Mahbubenej American Sonic ArtStream Farge Hain jade Benjamin-Chong Erotiin Abu Mt Nesel Serie Ste Albuddetin feil A Name Karjamet Arrynal Abden, Thomas Desen Ketty Dry Lord Cha My Summary Articles oa Background Diarrhoea and growth faltering in early childhood are associated with subsequent adverse outcomes. We aimed to assess whether water quality, sanitation, and landwashing interventions alone or combined with mutrition eve interventions reduced diarrhoea or growth faltering Methods The WASH Benefits Bangladesh cluster-randomised trial enrolled pregnant women from villages in rural Bangladesh and evaluated outcomes at 1-year and 2-years follow-up. Pregnant women in geographically adjacent clusters were block-randomised to one of seven clusters chlorinated drinking water (water opgraded som sanitation (sanitation): promotion of handwashing with soap thandwashing: combined water, sanitation, and s handwashing counseling en appropriate child mitrition plus lipid-based nutrient supplements (nutrition); combined water, sanitation, handwashing, and mutrition; and control (data collection only). Primary outcomes were caregiver-reported diarrhoea in the past 7 days among children who were in utero or younger than 3 years a enrolment and length-decage Z score among children born to enrolled pregnant women. Masking was not possible for data collection, but analyses were masked. Analysis was by intention to treat. This trial is registered at ClinicalTrials.gov, number NCC01590095 School of of Findings Between May 31, 2012, and July 7, 2013, 5551 pregnant women in 720 clusters were randomly allocated to one of seven groups. 1382 women were assigned to the control group: 699 to water; 696 to sanitation 685 to SA handwashing: 702 to water, sanitation, and handwashing 699 to nutrition: and 656 to water, sanitation, handwashing, and nutrition. 331 (6%) women were lost to follow-up. Data on diarrhoea at year 1 or year 2 A (combined) were available for 14425 children (7331 in year 1, 7094 in year 2 and data on length-forage score in year 2 were available for 4554 children (92% of living children were measured at year 2. All interventions had high c adherence. Compared with a prevalence of 5.7% (200 of 3517 child weeks) in the control group. 7-day diarrhoea prevalence was lower among index children and children under 3 years at enrolment who received sanitation 1613-5%) of 176 prevalence ratio 0.61, 95% CI 0-46-0-81), handwashing (6213-5% of 1795; 0.60, 0-45-0.80) combined water, sanitation, and handwashing (7413-9%) of 1902, 06, 0-53-0-90), nutrition (62 [3-5% of 1766; 064,0 49-0-85), and combined water sanitation, Sandwashing, and nutrition (6613-5% of 1861 0.62, 0.47-0.53 diarrhoea prevalence was not significantly lower in children receiving water treatment (90 14-9%) of 1824: 0-89, A 0-70-1-13). Compared with control (mean length-far-age Z score -1-79), children were taller by year 2 in the nutrition group (mean difference 0-25 195% CI 0-15-0-363 and in the combined water, sanitation, handwashing, pati and matrition group (0-13 0 92-9-24 The individual water, sanitation, and handwashing groups, and combined water, sanitation, and handwashing group had no effect on linear growth. Interpretation Natrient supplementation and counselling modestly improved linear growth, but there was no benefit so the integration of water, sanitation, and handwashing with nutrition. Adhercoce was high in all group and diarrhoea prevalence was reduced in all intervention groups except water treatment. Combined water, sanitation, and handwashing interventions provided so additive benefit over single interventions Funding Bill & Melinda Gates Foundation. Copyright © The Author). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license Introduction Over 200 million children bon in low-income countries are at risk of not reaching their development potential Poor linear growth in early childhood is a marker for donc deprivation that is associated with increased mortality, impaired cognitive development, and reduced adult income Nutrition-specific intes ventions have been shown to improve child growth www.day 29 2008 20/52714-10- HA p C Pion Unity of Unity an 0 96#91What have we learned? (iii) 8. To better understand programming requires much more attention to drivers of community participation, compliance, substitution, links among programming components, etc. 9. Broader range of outcomes need to be considered in assessing 'growth': cognitive function, lean mass, inflammation markets? 10. CSB+ has no dairy ingredient; the other foods studied do. We're not able to conclude anything about the role of dairy per se in this study, but the option of continued programming of food aid (in blanket supplementation) without dairy should be discussed. II. The role of such foods in targeted (recuperative) programming needs to be determined separately. But, the interacting etiology of stunting with wasting needs to be better understood. 91#92i. (Many) remaining questions ii. Is blanket supplementation to prevent wasting and stunting cost- effective where child intake suboptimal (food shared/diverted)? What extra/new programming options can improve performance of blanket food supplementation, regardless of product choice? iii. No studies of blanket supplementation (including ours) show significant effectiveness in preventing wasting. iv. Are targeted supplementary foods more effective? Separate study in Sierra Leone is asking that question. V. What are the optimal elements of comprehensive programming? 92#93• Important implications • Huge $$$ invested in food assistance; ensuring more cost- effective programming is essential going forward. Investment in such research is critical to offering value-for- money to US taxpayers, and to intended beneficiaries. Greater impact of food aid interventions demands quality programming (design, implementation fidelity, coverage, duration, compliance), not just choice of food product. USG support for operations-relevant studies is key to making all-of-government actions on nutrition effective and sustained. 93#94Upcoming Analytical Steps ... • . • • Cost per healthy child (i.e. never stunted nor wasted) Cost per incremental Icm growth/length (% point improvements) Multiple interactions between wasting and stunting Analysis of detailed 24hr food/calorie/nutrient intake of child Analysis of post-intervention follow-up period (outcomes) • Analysis of in-home observations and self-reported behaviors. So please join Evidence Summit June 2018 for more... 94#95USAID GEVEN USAID FROM THE AMERICAN PEOPLE This work was made possible through support provided by the Office of Food For Peace, Bureau for Democracy, Conflict, and Humanitarian Assistance, U.S.Agency for International Development, under the terms of Contract No.AID-OAA-C-16-00020. The opinions expressed herein are those of the author(s) and do not necessarily reflect the views of the U.S.Agency for International Development. 95 95#96Thank you for the collaboration! ■ USAID: Funder USAID USAID FROM THE AMERICAN PEOPLE ■ Tufts University: Beneficiary of Food For Peace/USAID grant for food aid research ■ Institut de Recherche en Sciences de la Santé (IRSS): Research implementation partner ■ ACDI/VOCA: Prime awardee implementing Title II program, Victoire sur la Malnutrition (VIM). ■ Save the Children: Sub-prime in consortium implementing the ViM program Study participants from Sanmatenga Province, Burkina Faso ■ Industry partners who produce the foods. INSTITUT E RECHERCHE EN SCIENCES DE L ES DE LA SANT Tufts UNIVERSITY Gerald J. and Dorothy R. Friedman School of Nutrition Science and Policy 7192 0 ACDI VOCA Expanding Opportunities Worldwide Save the Children. 96