Enhancing the evaluation of integrated, nutrition-sensitive strategies to improve maternal and child nutrition and health Restricted; Files Only

Self, Julie Lynn (2015)

Permanent URL: https://etd.library.emory.edu/concern/etds/08612p07q?locale=en
Published

Abstract

There is a critical gap in the evidence base for the estimated impact of nutrition-sensitive strategies. The objective of this dissertation is to demonstrate how path analysis and economic evaluation of integrated, nutrition-sensitive programs can enhance understanding about their role in improving maternal and child nutrition and health. We use path analysis to assess the mechanisms of effect for the Action Against Malnutrition through Agriculture (AAMA) Project in Nepal and assesses the cost-effectiveness of the Mama-SASHA Project in Kenya. The AAMA Project combined a nutrition-sensitive agriculture intervention with behavior change communication. Cross-sectional endline survey data were used to measure variables along the hypothesized Program Impact Pathway, from inputs through outcomes of interest, including child nutritional status, maternal underweight, child diarrhea, maternal night blindness, and maternal and child hemoglobin. Survey respondents included mothers with children aged 12-48 months. We used path analysis to assess the relative contribution of specific mechanisms of effect on nutrition and health. The hypothesized model for height-for-age z-score fit the data well (RMSEA=0.027; CFI=0.945), while the model for child hemoglobin (RMSEA=0.073, CFI=0.913) was a moderate fit, and some models did not achieve adequate fit. Areas of weak association and poor model fit may be attributed to weaknesses in measurement, intervention fidelity, and contextual factors unaccounted for in the models. The Mama-SASHA project aims to improve the health and nutrition of pregnant/lactating women and children <2 years through an integrated orange-fleshed sweetpotato (OFSP) and health service strategy in Western Kenya. We estimated the incremental cost-effectiveness ratio (ICER) of the intervention, which includes OFSP vouchers provided at antenatal care (ANC) visits, nutrition education, and pregnant women's clubs, compared to status quo ANC services. Effectiveness data from a quasi-experimental study and estimates from the literature were used to estimate DALYs for a range of benefits. We used ingredients-based micro-costing to estimate economic costs of agriculture, health and community interventions, including opportunity costs of labor. Annual net economic cost was USD $145,589. 72 DALYs were averted per year, mostly attributable to improvements in stunting and anemia. The ICER was USD $2,015 per DALY averted, which meets cost-effectiveness criteria set by WHO.

Table of Contents

CHAPTER 1: INTRODUCTION...................................................................................................1
RESEARCH AIM 1: .......................................................................................................................2
RESEARCH AIM 2:.......................................................................................................................2
REFERENCES ...............................................................................................................................3
CHAPTER 2: BACKGROUND.....................................................................................................4
THE BURDEN OF MALNUTRITION ..............................................................................................4
Global trends and consequences of malnutrition ...........................................................4
Micronutrient malnutrition ...............................................................................................5
Causes of malnutrition and conditions for optimal nutrition ......................................7
STRATEGIES FOR IMPROVING NUTRITION AND FOOD SECURITY...............................................9
Nutrition-specific strategies...............................................................................................9
Nutrition-sensitive strategies ..........................................................................................11
Impact pathways of nutrition-sensitive strategies ......................................................12
Evidence for the effectiveness of household food production programs...................14
Factors influencing the effect of household food production interventions .............16
EVALUATION OF NUTRITION-SENSITIVE PROGRAMS...............................................................17
Theory-driven evaluation of complex programs..........................................................17
Path analysis of nutrition-sensitive programs .............................................................18
Economic evaluation of nutrition-sensitive programs................................................20
Nutrition-sensitive interventions included in this research........................................25
REFERENCES .............................................................................................................................29
CHAPTER 3: ASSESSING PATHWAYS OF EFFECT FOR HOUSEHOLD FOOD PRODUCTION
INTERVENTIONS TO IMPROVE MATERNAL AND CHILD NUTRITION AND HEALTH: APPLYING
PATH ANALYSIS TO THE PROGRAM IMPACT PATHWAY FOR THE HOMESTEAD FOOD
PRODUCTION PROGRAM IN NEPAL.......................................................................................40
INTRODUCTION .........................................................................................................................40
METHODS..................................................................................................................................44
Intervention design and implementation ......................................................................44
Evaluation design and implementation.........................................................................45
Measurement and indicators...........................................................................................46
Theoretical models, simple mediation and path analyses ..........................................49
RESULTS ....................................................................................................................................51
Refining the path model and variable selection............................................................52
Simple mediation analyses...............................................................................................53
Path analyses......................................................................................................................53
DISCUSSION...............................................................................................................................54
CHAPTER 3 TABLES AND FIGURES ...........................................................................................61
REFERENCES .............................................................................................................................67
CHAPTER 4: HOW DOES HOMESTEAD FOOD PRODUCTION IMPROVE MATERNAL AND CHILD
NUTRITION? PATH ANALYSIS OF THE AAMA PROJECT IN NEPAL. ...........................................71
INTRODUCTION .........................................................................................................................71
METHODS..................................................................................................................................74
AAMA Project description ................................................................................................74
Evaluation design and data collection ...........................................................................75
Measurement......................................................................................................................76

Model specification and testing.......................................................................................78
RESULTS ....................................................................................................................................80
Child nutrition and health status ....................................................................................80
Maternal nutrition and health status .............................................................................82
DISCUSSION...............................................................................................................................83
CHAPTER 4 TABLES AND FIGURES ...........................................................................................91
REFERENCES ...........................................................................................................................101
CHAPTER 5: COST-EFFECTIVENESS OF MAMA-SASHA: A PROJECT TO IMPROVE HEALTH AND
NUTRITION THROUGH AN INTEGRATED ORANGE-FLESH SWEETPOTATO PRODUCTION AND
HEALTH SERVICE DELIVERY MODEL.....................................................................................104
INTRODUCTION .......................................................................................................................104
Description of the intervention......................................................................................106
METHODS................................................................................................................................107
Costs ...................................................................................................................................107
Health outcomes...............................................................................................................110
Economic benefits ............................................................................................................112
DALYs averted and incremental cost-effectiveness ratio..........................................112
Sensitivity analysis ..........................................................................................................114
RESULTS ..................................................................................................................................115
Economic costs and benefits...........................................................................................115
DALYs averted and ICER ...............................................................................................116
Multiple variable sensitivity and scenario analyses..................................................116
DISCUSSION.............................................................................................................................117
CHAPTER 5 TABLES AND FIGURES .........................................................................................121
REFERENCES ...........................................................................................................................129
CHAPTER 6: SUMMARY AND CONCLUSIONS........................................................................135
SUMMARY OF FINDINGS .........................................................................................................135
LIMITATIONS...........................................................................................................................137
STRENGTHS AND INNOVATIONS.............................................................................................138
CONCLUSIONS AND FUTURE RESEARCH................................................................................139
REFERENCES ...........................................................................................................................142


List of Tables
TABLE 1: SUMMARY STATISTICS FOR PATH MODEL VARIABLES AND PREVALENCE OF STUNTING...............63
TABLE 2: SUMMARY STATISTICS FOR PATH MODEL VARIABLES. CONTINUOUS OUTCOME VARIABLES ARE
REPORTED WITH THE CORRESPONDING PREVALENCE OF RELEVANT CONDITION...............................93
TABLE 3: SUMMARY OF BEST PATH MODEL AND FIT INDICES FOR EACH OUTCOME..................................100
TABLE 4: DALY INPUT VALUES, SOURCE DATA AND ASSUMPTIONS FOR ALL PROJECT-RELATED OUTCOMES
INCLUDED IN DALY ESTIMATES........................................................................................................121
TABLE 5: FINANCIAL COSTS, VALUE OF LABOR AND TOTAL ECONOMIC COSTS (AND PERCENTAGE OF TOTAL
COST) BY PROGRAM ACTIVITY AND CATEGORY OF ACTOR FOR 3-YEAR PROJECT (2011-2013), 2013
USD...................................................................................................................................................122
TABLE 6: TOTAL ECONOMIC BENEFITS ATTRIBUTED TO MAMA SAHSA (2011-2013), 2013 USD.........123
TABLE 7: ANNUAL CASES AND DALYS AVERTED FOR ALL OUTCOMES INCLUDED IN TOTAL DALY
ESTIMATES. ........................................................................................................................................124
TABLE 8: SUMMARY OF ANNUAL EFFECTIVENESS, ANNUAL ECONOMIC COSTS AND BENEFITS, AND COSTEFFECTIVENESS
RATIO, 2013 USD....................................................................................................125
TABLE 9:ICER SENSITIVITY ANALYSIS FOR CEA AND DALYS (PARAMETER VALUES - BASE CASE, HIGH
AND LOW, ETC), 2013 USD. ..............................................................................................................127


List of Figures
FIGURE 1: FRAMEWORK FOR ACTIONS TO ACHIEVE OPTIMUM FETAL AND CHILD NUTRITION AND
DEVELOPMENT. FROM BLACK 2013. .....................................................................................................8
FIGURE 2: HYPOTHESIZED MODEL FOR HOUSEHOLD FOOD PRODUCTION INTERVENTIONS [66]..............13
FIGURE 3: PROGRAM IMPACT PATHWAY FOR HELEN KELLER INTERNATIONAL'S HOMESTEAD FOOD
PRODUCTION INTERVENTION..............................................................................................................61
FIGURE 4: THEORETICAL PATH MODEL OF HOMESTEAD FOOD PRODUCTION'S EFFECT ON HEIGHT-FORAGE
Z-SCORE........................................................................................................................................62
FIGURE 5: EXAMPLE MEDIATION MODEL. ..................................................................................................64
FIGURE 6: SIMPLE MEDIATION MODELS WITHIN THE HEIGHT-FOR-AGE PATH MODEL. A) THE
RELATIONSHIP BETWEEN AGRICULTURE TRAINING AND HEIGHT-FOR-AGE Z-SCORE IS MEDIATED BY
THE NUMBER OF VEGETABLE VARIETIES GROWN IN THE HOME GARDEN. THE INDIRECT EFFECT
EXPLAINS 59.2% OF THE TOTAL EFFECT. B) THE RELATIONSHIP BETWEEN NUTRITION EDUCATION
AND DIETARY DIVERSITY IS PARTIALLY MEDIATED BY NUTRITION KNOWLEDGE. THE INDIRECT
EFFECT EXPLAINS 48.5% OF THE TOTAL EFFECT.................................................................................65
FIGURE 7: PATH ANALYTIC MODEL OF HOMESTEAD FOOD PRODUCTION'S EFFECT ON HEIGHT-FOR-AGE ZSCORE.
UNSTANDARDIZED PATH COEFFICIENTS ARE PROVIDED FOR ALL PATHS. *SIGNIFICANT AT
P<0.05.................................................................................................................................................66
FIGURE 8: PROGRAM IMPACT PATHWAY FOR HKI'S HOMESTEAD FOOD PRODUCTION INTERVENTION..91
FIGURE 9:INITIAL PATH MODELS: A) FULL MODEL TESTED FOR CHILD NUTRITIONAL STATUS OUTCOMES.
THE INITIAL MODEL FOR CHILD DIARRHEA REVERSES THE DIARRHEA AND HAZ VARIABLES. B) FULL
MODEL TESTED FOR CHILD HEMOGLOBIN AND ALL MATERNAL OUTCOMES. ......................................92
FIGURE 10: PATH MODEL FOR HEIGHT-FOR-AGE Z-SCORE (UNSTANDARDIZED PATH COEFFICIENTS).....94
FIGURE 11: PATH MODEL FOR CHILD HEMOGLOBIN, SHOWING UNSTANDARDIZED PATH COEFFICIENTS
FOLLOWED BY STANDARDIZED PATH COEFFICIENTS. ..........................................................................95
FIGURE 12: PATH MODEL FOR CHILD DIARRHEA........................................................................................96
FIGURE 13: PATH MODEL FOR MATERNAL BMI.........................................................................................97
FIGURE 14: PATH MODEL FOR MATERNAL HEMOGLOBIN...........................................................................98
FIGURE 15: PATH MODEL FOR MATERNAL NIGHT BLINDNESS....................................................................99
FIGURE 16: TORNADO DIAGRAM FOR ICER SENSITIVITY ANALYSES........................................................128

About this Dissertation

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
School
Department
Degree
Submission
Language
  • English
Research field
Keyword
Committee Chair / Thesis Advisor
Committee Members
Last modified No preview

Primary PDF

Supplemental Files