This paper presents findings on conditions of healthcare delivery in Afghanistan. There is an ongoing debate about barriers to healthcare in low-income as well as fragile states. In 2002, the Government of Afghanistan established a Basic Package of Health Services (BPHS), contracting primary healthcare delivery to non-state providers. The priority was to give access to the most vulnerable groups: women, children, disabled persons, and the poorest households. In 2005, we conducted a nationwide survey, and using a logistic regression model, investigated provider choice. We also measured associations between perceived availability and usefulness of healthcare providers. Our results indicate that the implementation of the package has partially reached its goal: to target the most vulnerable. The pattern of use of healthcare provider suggests that disabled people, female-headed households, and poorest households visited health centres more often (during the year preceding the survey interview). But these vulnerable groups faced more difficulties while using health centres, hospitals as well as private providers and their out-of-pocket expenditure was higher than other groups. In the model of provider choice, time to travel reduces the likelihood for all Afghans of choosing health centres and hospitals. We situate these findings in the larger context of current debates regarding healthcare delivery for vulnerable populations in fragile state environments. The 'scaling-up process' is faced with several issues that jeopardize the objective of equitable access: cost of care, coverage of remote areas, and competition from profit-orientated providers. To overcome these structural barriers, we suggest reinforcing processes of transparency, accountability and participation.
Copyright 2010 Elsevier Ltd. All rights reserved.