Food insecurity among volunteer AIDS caregivers in Addis Ababa, Ethiopia was highly prevalent but buffered from the 2008 food crisis.

Our objective in this study was to assess the validity and dependability of the Household Food Insecurity Access Scale (HFIAS), which was developed for international use, among community health volunteers in Addis Ababa, Ethiopia. The HFIAS was translated into Amharic and subsequently tested for content and face validity. This was followed by a quantitative validation study based on a representative sample (n = 99) of female community volunteers (HIV/AIDS home-based caregivers), with whom the HFIAS was administered at 3 time points over the course of 2008, in the context of the local and global "food crisis." By pooling observations across data collection rounds and accounting for intra-individual correlation in repeated measures, we found that the HFIAS performed well according to standards in the field. We also observed slight amelioration in reported food insecurity (FI) status over time, which seems paradoxical given the increasing inaccessibility of food over the same time period due to inflating prices and disappearing food aid. We attempted to resolve this paradox by appealing to self-report-related phenomena that arise in the context of longitudinal study designs: 1) observation bias, in which respondents change their reports according to changing expectations of the observer-respondent relationship or change their behavior in ways that ameliorate FI after baseline self-reports; and 2) "response shift," in which respondents change their reports according to reassessment of internal standards of FI. Our results are important for the validation of FI tools and for the sustainability of community health programs reliant on volunteerism in sub-Saharan Africa.