Household survival and changes in characteristics of households in rural South-western Uganda through the period of 1989 to 2008

Background: A better understanding of household dynamics has become increasingly important as researchers consider the broader impact of HIV on communities. This has led to the need for longitudinal investigation of household response to HIV infection. Objectives: Using the General Population Cohort (GPC) household data collected between 1989 and 2008 in rural South-west Uganda, households are traced over time. Structurally classified, overall household behaviour (dissolution, migration and structural change) is observed. Whether and how adult HIV infection or mortality alters the overall household patterns is investigated. Methods: The GPC households were retrospectively traced between 1989 and 2008. For each year, households are classified to identify children (<15 years), “middle-aged” (15-59 years) and “older” (≥60 years) adults, or according to family relationships of residents to the household head. Cross-sectional and longitudinal analyses of household survival and structural change patterns are undertaken. Results: Least household dissolution or migration (‘stable’) is observed in extended, households with a couple with children; or those with a child, middle-aged man, middle-aged woman living with or without an older adult. The overall patterns show households transitioning to these stable structures over time through reproduction, marriage, individual growth, and individual in-migration. However, household head or spouse HIV infection hinders this transition particularly promoting couple separation; being a barrier to reproduction; or increasing the out-migration of non-productive residents (dependants) eventually increasing cases of household non-survival (dissolution or migration). Household head or spouse death results to widowed heads which corresponds to an increase in migration of initially stable households. Alternatively other adult resident HIV infection or mortality promote household growth and increased household survival. Conclusion With the GPC households traced, longitudinal household investigations are possible. This enables a better understanding of how HIV infection and adult mortality influence household dynamics.