In recent years, the links between HIV/AIDS, food security and livelihoods have been the subject of much research and many direct interventions. The impacts of HIV/AIDS on livelihoods are multiple and diverse: it can reduce the ability of sick household members to work; increase the demands on remaining household members’ time to care for the ill; increase the burden of healthcare costs; and lead to problems with the inheritance of land and other assets by the bereaved. The ways in which different aspects of the household economy can be affected by HIV/AIDS is illustrated here.

The HEA framework can be used to examine the situation of HIV/AIDS-affected households and to illustrate the effects of HIV/AIDS. In this case, we would take our baseline as being the period before the effects of HIV/AIDS were felt, and look at assets, food, income and expenditure as usual. HIV/AIDS would then be treated as a shock, or as a collection of shocks, with those affected providing information both on how they have been affected and how they have responded. For example, the problem specification may show that the loss of labour as a result of illness resulted in a 100% loss of casual labour income if the ill household member was the only one working, or in a 300% increase in the cost of healthcare, etc.  The response may be, for example, an increase in the number of livestock sold as a coping mechanism, or in an increase in work by other members of the household to compensate for the loss of labour of the ill member.

The added value of HEA in this case is that it gives a holistic view of the impacts of HIV/AIDS rather than focusing, for example, on the impacts on agricultural production alone. It enables us to see how the household adapts to the illness, recognising that while the overall impacts will almost invariably be negative, households will try to re-allocate their labour and other assets to minimise those negative impacts.

In terms of the methods needed for collecting this information, some adaptation of the purposive sampling usually used for HEA assessments is required in that it is necessary to define quite clearly the type of people who will be interviewed in order to get the clearest results. For example, the broad category of ‘HIV/AIDS-affected’ can include families with someone who is chronically ill, families who have recently lost an income-earner, or families who have taken in an orphaned child. The nature of the ‘shock’ resulting from HIV/AIDS may differ significantly in each case, and so care should be taken not to over-aggregate information by placing all HIV/AIDS-affected families in a single category. Similarly, disaggregating by wealth remains necessary, as the impact of HIV/AIDS on a family that was initially better off could vary substantially from the impact on a poor family, especially in the short term. Because of the sensitivity of the issue of HIV/AIDS in some cases, it may be more effective to interview people on an individual basis rather than in focus groups.

Once information on HIV/AIDS-affected families has been collected, then their status can be compared with those unaffected. But to do so effectively, it is again vital to ensure that you are comparing the affected and unaffected within the same wealth group. Comparing, for example an affected family from a better off background with an unaffected family from a poor background would produce confusing results. This type of comparison is also important as a means of trying to separate out the effects of other shocks from the shock of HIV/AIDS. For example, if there has also been a drought between the baseline period and the current period, then it is useful to compare the change over time between affected and unaffected households to try to determine how much of the change for the affected households is caused by HIV/AIDS and how much by drought.