Three papers on households and child health in Botswana

The primary objective of this thesis is to examine the change in Botswana households between 1971 and 2011, show the relationship of household composition to child health outcomes and to demonstrate the association of community factors with stunting in Botswana. The thesis focusses on three related papers.Paper one uses a descriptive approach, and it provides an overview of how households have changed over time in Botswana. This paper analyses data from census reports for the years 1971, 1981, 1991, 2001 and 2011, and three nationally representative surveys: the 1988 Botswana Standard Demographic Health Survey, the 2000 Multiple Indicator Survey (MICS) and the 2007 Botswana Family Health Surveys (BFHS). The findings from this paper indicate a change towards smaller households. The results show a decrease in the average household size, growth in the number of households, and a growing importance in the distribution of other household types (e.g., a rise of one person households, living with family and less with extended and not related household members). At the same time there is a decrease in childbearing and fertility.Using data from the 2007 BFHS and multi-level logistic models, paper two examines the relationship between household composition, stunting and diarrhoea prevalence. The findings indicate that stunting varies by whom the child lives with. Stunting is higher among children living with no parents compared to those living with both parents, and among those living with not related household members. Stunting is less prevalent among children living with an aunt and among those living with other relatives. Also, children living in mother only households and with a grandparent present have a higher level of stunting compared to those living with both parents. The findings on diarrhoea prevalence show that children living in mother-only households, and those living with no parents are less likely to have the condition than those living with both parents. Across all households, those who are richer, regardless of who is in the household, have lower rates of stunting and diarrhoea prevalence than those that are poorer.Paper three draws on census and survey data to examine the relationship between community factors and stunting. The results from paper three show a negative association between percentage of professionals in a district and stunting. Children living in the West, East/North East and Central parts of the country had higher levels of stunting compared to those living in the North. In addition, household wealth was found to be a significant moderator between the percentage of professionals in a district and stunting. Finally, the analysis in paper two and three show a clustering effect at household level.The implications of these findings is that policies aimed at reducing stunting and diarrhoea may be best targeted at individual and household factors, and other adults in a household beside biological parents. While further research is needed to understand the mechanisms by which community factors influence stunting