There is an urgent need for innovative methods to generate information to evaluate post-conflict reconstruction, especially in the health sector. In order to translate the immediate response into a systematic medium and long term health strategy, baseline estimates of health service performance are required. In war torn Afghanistan, faced with a monumental task of laying the foundations for an equitable and quality oriented health system, the Ministry of Public Health used the outdated information from the 1979 census to conduct the first population based health assessment in 2003. Results from a comparison of these estimates with those generated using the pre-census conducted in 2004 indicate that the originally reported estimates provided information that was adequate for cross-sectional assessment but of limited use for assessing trends over time.

Logistical concerns of restricted access in post-conflict countries like Afghanistan require a measure of living standards to be based on information that is easy to collect, observe and verify. Comparison of principal components analysis based asset index with expenditure estimates based on out of sample prediction indicated that predicted expenditure was a more reliable measure to track poverty over time. Regular collection of data on asset variables to predict expenditure can improve the accessibility of this information to policy makers. In addition, the predicted expenditure can form the basis for poverty mapping and targeting through the social protection programs.

Equitable health systems require a sustained increase in utilization of health services by groups that suffer from the greatest burden of disease--females and the poor. In Afghanistan, a trend towards equitable utilization in the last four years has occurred along with simultaneous improvement in quality of health services. The association between different characteristics of health care delivery system and utilization rate differed across the three outcome groups in the facility catchment area--total population, the poor and females. Improvement in staffing and service capacity was associated with a significantly greater increase in utilization in a later year as compared to a similar improvement in an earlier year. User fee collection was associated with a decrease in utilization rate, especially by the poor.