We implemented a semi-quantitative investigative framework using the semi-quantitative evaluation of access and coverage (SQUEAC) method to assess and evaluate access to and coverage (uptake) of initiatives established by MoHS and other partners to improve nutrition of women and children (e.g. the Mother Support Groups). Our key investigation question was:
What are the blockages to preventing chronic malnutrition at the district and community levels in Kambia district, Sierra Leone?
Specifically, the investigation’s objectives were:
Assess barriers and boosters to behaviour change in relation to maternal, infant and young child feeding practices in Kambia district.
Conduct a causal analysis of stunting in Kambia district.
SQUEAC combines quantitative and qualitative data to provide in-depth analysis of causes of undernutrition, coverage of specific behaviours and/or interventions and of barriers and boosters to service uptake. The SQUEAC approach is well suited for application to address the core research question stated above.
This report details the research design that was developed for this research project.
SQUEAC investigation flow
SQUEAC will identify the barriers and boosters that act to determine whether these initiatives ultimately impact the outcomes expected i.e. maternal, infant and young child feeding practices. Interviews and focus group discussions with a range of key stakeholders/institutions at community level will collect qualitative data the validity of which is ensured by triangulation by source and method. This data will be used to develop hypotheses on level of various service uptake and on barriers and boosters to adoption of critical maternal, infant and young child feeding practices. To test these hypotheses small studies will be performed at community level that either randomly sample households/carers for interview or that purposively select cases and controls for interview.
The collected data and information from the semi-quantitative investigation will be used to generate hypotheses relating to chronic malnutrition causality to be tested using a matched case-control study. A matched case-control design will be adopted as this requires a smaller sample size than an unmatched design for the same statistical power. Matching will be done on location, sex and age. The causal analysis will identify risk factors and risk markers (e.g. poor maternal diet diversity, infection, early introduction of fluids other than breastmilk – a marker for poor IYCF practices) that are significantly associated with stunting.
Figure 1 below presents the proposed SQUEAC investigation flow for this research.
Research respondents and participants
We will be engaging primarily with pregnant and lactating women (PLW) and their children age 6 to 24 months old.
Research indicators and instruments
Maternal and child nutrition status
Maternal nutrition status will be assessed using the mid-upper arm circumference (MUAC). A cut-off of less than 230 mm will be used to indicate that a mother is undernourised.
Child nutrition status will be assessed using the mid-upper arm circumference (MUAC) and the height-for-age z-score (HAZ). A cut-off of MUAC less than 125 mm will be used to assess wasting (with MUAC less than 115 mm used to indicate severe wasting). A cut-off of less than -2SD z-score will be used to assess stunting (with a -3SD z-score used to indicate severe stunting).
Infant and young child nutrition (IYCN) practices
We propose to use the Infant and Child Feeding Index (ICFI) developed by Arimond and Ruel1 as a framework to assess IYCN practices and behaviours. There is existing data / information from an ongoing IYCN monitoring and surveillance system initiated by UNICEF and implemented by MoH and local partners. Figure 2 describes the hierarchical nature of the ICFI which facilitates easy and straightforward diagnosis of current IYCN practices and helps the identification of specific practices that need improvement.
Coverage of IYCN services / interventions
We propose to use the Tanahashi model2 of health services coverage assessment. Five hierarchical coverage indicators will be used: 1) availability coverage; 2) accessibility coverage; 3) acceptability coverage; 4) contact coverage; and, 5) effectiveness coverage. Figure 3 illustrates these different coverage levels and describes what each one measures.
Women’s dietary diversity score
Diets of pregnant and lactating women will be assessed using the Women’s Dietary Diversity Score (WDDS) which is a proxy for minimum nutritional adequacy of a woman’s diet developed by Arimond and colleagues3.
Selection of research sites
Kambia was selected as the focus district for this research.
For SQUEAC, the selection of villages or communities to visit to conduct investigation will be based on the topic-at-hand and the current investigation lead that is being followed by the investigators.
For the matched case-control, a centric-systematic area sampling (CSAS) approach4 will be used to select the villages to be included in the study. Figure 4 shows a collection of maps illustrating the selection conducted using CSAS. This is a tentative list of 10 sampling sites proposed and may change during conduct of the research.
We will use standard questionnaires and survey instruments whenever possible.
The research has found that the issues causing malnutrition are both multifactorial and complex, suggesting a range of initiatives are needed in order to prevent it. Many of these initiatives already exist at community level, but they require strengthening and expansion. Our evidence also suggests that, in order to achieve sustained behaviour change at the local level, there is a need to target key decision-makers within the household, such as fathers and elders, and to engage seriously with traditional practices.
This output forms part of a wider package of research conducted by the SLRC in Sierra Leone. Together with a forthcoming qualitative report, it seeks to answer the country programme’s second overarching research question: ‘What are the blockages to preventing malnutrition at the local level?’.
Arimond, M. & Ruel, M.T., 2004. Summary Indicators for Infant and Child Feeding Practices: An Example from the Ethiopia Demographic and Health Survey 2000 (pub. 2002). pp.1–78. ↩
Tanahashi, T., 1978. Health service coverage and its evaluation. Bulletin of the World Health Organization, 56(2), pp.295–303. ↩
Arimond, M. et al., 2009. Dietary Diversity as a Measure of the Micronutrient Adequacy of Women’s Diets in Resource-Poor Areas: Summary of Results from Five Sites, Washington, DC: FANTA at AED. ↩
Milne, A., 1959. The centric systematic area-sample treated as a random sample. Biometrics, 15(2), pp.270–297. ↩