Pallas Sarah Wood, Ruger Jennifer Prah
Current address: Perelman School of Medicine and Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia, PA, USA.
Health Policy Plan. 2017 May 1;32(4):493-503. doi: 10.1093/heapol/czw164.
Previous literature suggests that increasing numbers of development aid donors can reduce aid effectiveness but this has not been tested in the health sector, which has experienced substantial recent growth in aid volume and number of donors.
Based on annual data for 1995-2010 on 139 low- and middle-income countries that received health sector aid from donors reporting to the OECD's Creditor Reporting System, the study used two-step system generalized method of moments regression models to test whether the number of health aid donors and an index of health aid donor fragmentation affect health services (measured by DTP3 immunization rate) or health outcomes (measured by infant mortality rate) for three subsectors of health aid.
For total health aid and for the general and basic health aid subsector, controlling for economic and political conditions, increases in the number of donors were associated with increases in DTP3 immunization rate and reductions in infant mortality while increases in the donor fragmentation index were associated with decreases in DTP3 immunization rate and increases in infant mortality, though none of these relationships were statistically significant. For the population and reproductive health aid subsector, a one percent increase in the number of donors was associated with a 0.23 percent decrease in DTP3 immunization ( P < 0.01) while a one percent increase in donor fragmentation was associated with a 0.54 percent increase in DTP3 immunization rate ( P < 0.01); associations with infant mortality rates for this subsector were similar to those for total health aid.
The results do not provide clear evidence in support of the hypothesis that donor proliferation negatively impacts development results in the health sector. Aid effectiveness policy prescriptions should distinguish responses to donor proliferation versus donor fragmentation and be adapted to specific subsectors of health aid.
以往文献表明，发展援助捐助方数量增加可能会降低援助效果，但这一点尚未在卫生部门得到验证，而卫生部门近年来援助规模和捐助方数量都大幅增长。
该研究基于1995年至2010年向经合组织债权国报告系统汇报的向139个低收入和中等收入国家提供卫生部门援助的捐助方的年度数据，使用两步系统广义矩回归模型，来检验卫生援助捐助方数量以及卫生援助捐助方碎片化指数是否会影响卫生援助三个子部门的卫生服务（以三联疫苗第三针免疫接种率衡量）或卫生成果（以婴儿死亡率衡量）。
对于总体卫生援助以及一般和基本卫生援助子部门，在控制经济和政治条件后，捐助方数量增加与三联疫苗第三针免疫接种率上升以及婴儿死亡率下降相关，而捐助方碎片化指数上升与三联疫苗第三针免疫接种率下降以及婴儿死亡率上升相关，不过这些关系均无统计学意义。对于人口与生殖健康援助子部门，捐助方数量增加1%与三联疫苗第三针免疫接种率下降0.23%相关（P < 0.01），而捐助方碎片化增加1%与三联疫苗第三针免疫接种率上升0.54%相关（P < 0.01）；该子部门与婴儿死亡率的关联与总体卫生援助的情况类似。
研究结果并未提供明确证据支持捐助方增多会对卫生部门发展成果产生负面影响这一假设。援助效果政策规定应区分针对捐助方增多与捐助方碎片化的应对措施，并应适用于卫生援助的特定子部门。
