Assessing progress in implementing Uganda’s Nutrition Action Plan: District-level insights

Location: Uganda

What we know: There is no specific monitoring and evaluation system to assess progress on implementation of Uganda’s Nutrition Action Plan (UNAP).

What this article adds: Two serial cross-sectional panel surveys were conducted in 2011/12 and 2013/14 to determine progress on UNAP-related indicators. Two districts were randomly selected for panel analyses. Household, child and mother/caregiver data were collected. In Kisoro district, prevalence of underweight significantly fell (P<0.01) and dietary diversity significantly increased (p<0.01). Improvements in wasting, iron deficiency anaemia and exclusive breastfeeding were not significant. In Lira district, there was no significant change in measured indicators; while some improved, others deteriorated (e.g. anaemia, exclusive breastfeeding). The authors suggest a strong District Nutrition Coordination Committee (DNCC) role in nutrition planning and implementation, particularly in Kisoro, may explain the results. Empowering DNCCs to use data in planning is critical; incorporating UNAP indicators into existing systems is recommended.

There is no established monitoring and evaluation system that is specific to the implementation of Uganda’s Nutrition Action Plan (UNAP). Hence household data collected by USAID’s Feed the Future Nutrition Innovation Lab (NIL) was used to inform policy makers on the progress of UNAP-related indicators. Two serial, cross-sectional, panel surveys were conducted by NIL from November 2011 to January 2012 and again in 2013/14 (same time, same households). A district sample size of 600 households was determined as feasible within the existing budget and sufficient to represent the district size. A total of four sub-counties were selected. At the sub-county level, 25% of the parishes were selected using a list from the Uganda Bureau of Statistics (UBOS) and households were randomly selected with probability proportional to the population of the village.

One district was randomly selected from the southwest (Kisoro) and one from the north (Lira). The survey focused on the household, the mother of the household (and/or primary care giver), one index child aged 6 to 23 months, and all other children between the ages of 0 and 59 months. Data domains included household demographics (including education) and socioeconomic factors, as well as the following topics: diet and health status of mothers/primary caregivers; household food security; water, sanitation and hygiene; infant and young child feeding of the index child; and anthropometric and haemoglobin (Hb) measurements of the index child, his/her primary caregiver/mother and all other children in the household who were younger than five years. Quantitative, household-level panel data was analysed for the two study districts. Data on individual maternal and child outcomes were computed for n=600 households in each study district, and SPSS version 22 was used for analysis. The indicators selected for this analysis come from the UNAP table of key target indicators that have also been prioritised by the Government of Uganda. The key variables that included cut-offs were height-for-age, weight-for-age and weight-for-height (stunting, underweight and wasting, respectively). The prevalence of anaemia among mothers and children, as defined by Hb below 12 g/dL in non-pregnant women and 11 g/ dL for pregnant women and children, was also calculated.

Findings

A total of 598 households in Kisoro and 603 households in Lira were surveyed in panel one and 544 households in Kisoro and 552 households in Lira were followed up in panel two. The data show that the prevalence of underweight reduced significantly among children younger than five years old in Kisoro district (from 18.6% to 13.5%, P <0.01). There was also a significant increase in children aged 6 to 23 months achieving minimum dietary diversity (from 63.2% to 79.1%, P <0.01). Other child nutrition indicators improved between the two rounds of the panel survey (wasting, iron deficiency – both in women and children – and exclusive breastfeeding) but were not statistically significant. The prevalence of stunting and underweight among children younger than five years old is still above the UNAP targets. To achieve these targets, the prevalence of stunting would need to reduce by another 17 percentage points and the prevalence of underweight an additional 3.5 percentage points.

The data for Lira district indicate that there has been little to no improvement in UNAP targets for children and mothers/caregivers. Child underweight and wasting prevalence did not significantly improve. And while the changes are positive in most cases, for some indicators the change is negative. For instance, there is an increase in the prevalence of stunting, although this was not statistically significant. However, there was a positive and significant (P < 0.05) increase by 4% in food-secure households, based on the Household Dietary Diversity Score, which can reflect the ability of a household to access a variety of foods. The prevalence of exclusive breastfeeding also declined.

Discussion

In Uganda, multi-stakeholder initiatives have proved to be influential, locally owned and more sustainable than single-stakeholder initiatives. However, ascribing a given change to a specific project is difficult. The authors believe that a strong District Nutrition Coordination Committee (DNCC) role in nutrition planning and implementation, particularly in Kisoro, may explain the significant changes in child-related nutrition indicators indicated. The Kisoro DNCC has been very influential in building strong coordination, monitoring and collaboration of a variety of stakeholders in helping district leadership align its nutrition targets with different implementing partners. A strong DNCC can help coordinate the work of multiple actors in the district context and form a bridge to the various district administrative actors that characterise a multi-sector effort. Continually assessing nutrition indicators at the district level can support district decision-makers to achieve UNAP-derived nutrition goals. This can also help districts to align their development partners to district goals by mapping district needs to possible interventions. Using data for monitoring district performance also signals commitment to nutrition goals by district leadership and the DNCC. Monitoring data can be used by a DNCC to persuasively advocate for comprehensive, multi-sector planning at the district level.

National endorsement of nutrition plans does not in itself result in desired outcomes; hence the allocation of scarce resources has to be based on rigorous evidence. Ugandan districts should develop better capacity to acquire and use data relevant to nutrition targets. Empowering DNCCs to use data in planning is critical, since they are the custodians of nutrition planning at district level. This study shows that NIL data can be used to track UNAP indicators until more routine systems are developed. Furthermore, the authors suggest that districts should explore incorporating UNAP target indicators into existing systems, such as District Health Information System (DHIS-2) monitoring data, for district local government planning.