Sudan Household Health Survey Cluster Analysis

by

Acknowledgement

I would take this opportunity to thank my research supervisor, family and friends for their support and guidance without which this research would not have been possible.

DECLARATION

I, [type your full first names and surname here], declare that the contents of this dissertation/thesis represent my own unaided work, and that the dissertation/thesis has not previously been submitted for academic examination towards any qualification. Furthermore, it represents my own opinions and not necessarily those of the University.

Signed __________________ Date _________________

CHAPTER 4: RESULTS & DISCUSSION

4.1 Data sources

Data collected by the Sudan Household Health Survey which was carried out were used in this study. This survey collected information on children and women, measuring key indicators that allow the country to monitor progress towards MDGs and other internationally agreed upon commitments. Although SHHS covered both Northern and Southern Sudan, only the data covering the 15 Northern States was used in this study (Figure 1). This was due to the Southern States' data being unavailable. In each state, 25 households from each of 40 clusters (villages) were sampled. A two-stage cluster sampling design was employed to choose the samples in each state. Questionnaires were based on the Standard Multiple Indicator Cluster Survey which were adapted and amended to suit the country.

Figure 1. Conceptual framework for factors influencing neonatal mortality.

For this study, data were obtained from the Women's Questionnaire (for women aged 15-49 years), which included information on demographic characteristics, and then details of births within the 2 years preceding the survey - child mortality, history of antenatal care, delivery assistance, and mode and place of delivery for births.

4.2 Conceptual framework

A previously designed conceptual framework for the study of child survival in developing countries was adopted and slightly modified based on the available data in the SHHS, for identifying independent variables (Figure 1). Some factors were not available from the SHHS, so they were excluded from the analysis (Figure 2) (e.g. community-level factors and post-delivery factors).

Figure 2. Map of Sudan

4.3 Statistical Analyses

First, descriptive analysis using frequency tabulation was conducted. Crude odds ratios and 95% confidence intervals were calculated to assess the rough effect of each explanatory variable on the neonatal mortality rate without adjusting for other variables. Variables found to be significantly associated with neonatal mortality on at least one category (defined as a p value < 0.05) were entered into a multivariate logistic regression model to evaluate neonatal mortality as a function of demographic, socioeconomic and maternal health indicators. Pregnancy complications were removed from the final model because it was highly correlated to delivery complications. Similarly, parity was removed from the final model because it was highly correlated to maternal age. The final model included maternal age, household wealth index, sex of child, mode of delivery and delivery complications.

4.4 Results

In total, 21,942 women were interviewed, and of the 6,198 live-born children who were included in the analysis, 189 (3.0%) had died during the neonatal period. Characteristics of the study population are presented in Tables  2 and 3. Approximately 51% of the children were male, 52% of the children were born to ...