Premium
A Program Impact Pathways Mediation Analysis of a Multi‐sectoral Nutrition Program in Nepal
Author(s) -
Choufani Jowel,
Kennedy Andrew,
Pandey Pooja,
Cunningham Kenda
Publication year - 2017
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.31.1_supplement.786.34
Subject(s) - sanitation , environmental health , malnutrition , hygiene , program design language , program evaluation , behavior change communication , agriculture , geography , medicine , engineering , population , political science , health services , software engineering , public administration , pathology , archaeology
Background The complexity of nutrition programming globally is increasing with the growing importance of simultaneously addressing the diverse determinants of malnutrition by linking nutrition‐specific and nutrition‐sensitive approaches. In Nepal, for example, Suaahara was designed to improve the nutritional status of 1000‐day women and children by integrating health services, family planning, nutrition, water, sanitation and hygiene (WASH), and agriculture/homestead food production activities across 41 of Nepal's 75 districts in the country's three agro‐ecological zones. Evaluation of such a complex program requires researchers to carefully unpack the potential pathways to impact. This can be done using program impact pathways (PIPs) which map out program phases from inputs via process to outputs and outcomes to assess each step in the intended program pathways to impact. Methods A PIP analysis was conducted to understand whether program inputs of Suaahara translated into intended outcomes, and identify gaps along the theorized pathways to impact. Using data from a cross‐sectional process evaluation household‐level survey (n=472) of 80 communities split evenly throughout 4 Suaahara and 4 comparison districts, conducted two years after the start of the program, regression analysis was used to test associations between residing in a Suaahara intervention area and six ideal household health and nutrition behaviors. Diet during pregnancy was constructed to reflect whether mothers ate less than, the same, or more than usual during pregnancy. For WASH facilities and practices, a scale comprised of 4 observed and 1 reported household WASH facilities and practices was constructed. For positive, significant associations, structural equation modelling was used to empirically test hypothesized mediators along the pathways to impact. Results Belonging to the intervention arm was positively associated with optimal household WASH facilities and practices (b=0.69, p<0.001) and maternal diet during pregnancy (b=0.41, p<0.001). Exposure to specific WASH messages was found to be a positive mediator for how belonging to the intervention arm translated into ideal household WASH facilities and practices (b=0.18, p<0.001). There was no indirect pathway between belonging to the intervention arm and optimal dietary practices during pregnancy: although being in the intervention arm influenced both maternal knowledge and practices related to diet during pregnancy, there was no association between knowledge and practices. Conclusions PIP analyses are useful for monitoring and evaluating complex nutrition programs to assess program fidelity and explore pathways of intervention effects. The evidence for a link between the intervention and WASH facilities and practices through direct and indirect pathways is encouraging and can be used to support continuation of related intervention components. Further investigation into the reasons for a lack of translation of knowledge from exposure to Suaahara integrated messages to diet during pregnancy should be considered. Support or Funding Information Suaahara was funded by United Stated Agency for International Development (USAID).