Hrev_master Healthcare in Low-resource Settings 2025; volume 13:12874 The short-run effects of health aid in low-income countries: evidence from panel data analysis Keneni Gutema Negeri Health Systems Management and Policy Unit, School of Public Health, College of Medicine and Health Sciences, Hawassa University, Hawassa, Ethiopia Abstract The effectiveness of health-targeted aid in improving health outcomes in developing countries remains a subject of debate. This paper investigates the short-run impact of health aid on health status in low-income countries globally. A panel dataset was constructed from 34 low-income countries spanning 2000 to 2017, with Infant Mortality Rate (IMR) serving as the primary proxy for health status. To estimate the short-run effect, First Difference GMM and System GMM estimators were employed, with a preference for System GMM due to its robustness against weak instrument problems in dynamic panel data models. The model incorporated log transformations for Health Development Aid (HDA), GDP per capita, and cereal yield, while exponential transformations were applied to human capital and governance indices, alongside adolescent fertility rate and elderly dependency rate. The System GMM estimation revealed a statistically signifi- cant and beneficial short-run effect of health aid on health status. Specifically, a doubling of health aid is associated with a reduction of 2 infant deaths per 1,000 live births. Other significant findings include the positive impact of GDP per capita, human capital, and governance, and the negative impact of adolescent fertility rate and elderly dependency rate on infant mortality. The Sargan test confirmed the validity of the over-identifying restrictions (p=0.2279), and the Arellano-Bond test for AR(2) indicated no serial correlation in the idiosyncratic errors (p=0.158). The find- ings strongly suggest that health aid serves as a potent instrument for narrowing the health status gap between high and low-income countries, thereby contributing to the achievement of Universal Health Coverage. However, recipient countries should also prior- itize fostering domestic factors that positively influence the health sector to reduce persistent reliance on external resources. Introduction Low-income countries grapple with a dual burden of disease: the persistent challenge of communicable diseases and a rapidly escalating prevalence of Non-Communicable Diseases (NCDs), such as common diseases found in all income groups, including heart disease, stroke, cancer, diabetes, and chronic lung diseases.1 This epidemiological complexity is exacerbated by healthcare financing mechanisms heavily reliant on Out-Of-Pocket (OOP) payments.2 Such payments impose significant financial hardship, pushing approximately 100 million individuals into poverty and subjecting 150 million to catastrophic health expenditures annual- ly, perpetuating a vicious cycle of poverty and ill health.3 In this context, a critical question emerges: can avoidable infant mortality be averted in the foreseeable future, and what policy instruments can most efficiently achieve this? Globally, Universal Health Coverage (UHC) has been recog- nized as a cornerstone of sustainable development. The United Nations General Assembly enshrined UHC within the Sustainable Development Goals (SDGs) as Goal 3.8, aiming to “Achieve uni- versal health coverage, including financial risk protection, access to quality essential health care services and access to safe, effec- tive, quality, and affordable essential medicines and vaccines for all” by 2030. A key pathway to achieving this goal involves increased health aid, particularly that channeled through public spending on health, which is expected to enhance financial protec- tion. In low-income countries, health aid currently accounts for an average of 30% of health expenditure and has been increasing in absolute terms over time.2 While this raises numerous questions, this paper specifically investigates the effect of health aid in low- income countries on the achievement of the aforementioned UHC Correspondence: Keneni Gutema Negeri, Health Systems Management and Policy Unit, School of Public Health, College of Medicine and Health Sciences, Hawassa University, Hawassa, Ethiopia. E-mail: kenenigut2000@yahoo.com Key words: health function, health aid, infant mortality, low-income countries, panel data. Conflict of interest: the author declares that there is no conflict of interest. Ethics approval and informed consent: not applicable. Availability of data and material: available from the corresponding author on request. Funding: The researcher used fund from his own source. Acknowledgement: I extend my deepest gratitude to my family, whose unwavering encouragement and insightful discussions have been invalu- able in the preparation of this manuscript. Their support, in ways beyond financial contributions, has shaped my thinking and strengthened my commitment to this research. Received: 29 July 2024. Accepted: 28 August 2025. Early access: 18 September 2025. This work is licensed under a Creative Commons Attribution 4.0 License (by-nc 4.0). ©Copyright: the Author(s), 2025 Licensee PAGEPress, Italy Healthcare in Low-resource Settings 2025; 13:12874 doi:10.4081/hls.2025.12874 Publisher's note: all claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organi- zations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher. [Healthcare in Low-resource Settings 2025;13:12874] [page 247] goal, particularly regarding infant mortality. The existing literature on the impact of health-specific aid on recipient countries’ health status presents a notable divergence, contrasting with the more extensively studied effects of aggregate aid on national growth. This section synthesizes the arguments for and against the effectiveness of health aid, highlighting key find- ings and the methodological nuances that contribute to the ongoing debate. A significant body of research suggests that health-targeted aid demonstrably improves health outcomes in low-income countries by augmenting resource availability for health service delivery. Proponents argue that health is a sector where aid’s impact is par- ticularly discernible due to the direct link between health programs and desired outcomes. For instance, Levine and Kinder3 posits that health programs like communicable disease prevention and control (through safe water, sanitation, immunizations, and improved nutrition) are directly linked to positive health outcomes. Easterly,4 similarly, contends that with appropriate accountability, external aid can significantly decrease infant mortality. Empirical support for this perspective is robust. Mishra and Newhouse5 provide strong evidence using donor commitment data from 118 countries (1973-2004), finding that a 1% increase in per capita health aid is associated with a 2% improvement in infant mortality rate. Likewise, Chauvet et al.6 utilizing a panel data of 109 developing countries (1987-2004), reported significant effects of health aid on health improvement. Similar findings are reported by Ebeke and Drabo,7 Mishra and Newhouse,8 and Chauvet and Guillaumont,9 all suggesting a substantial positive impact of health aid on health outcomes in the developing world. These studies often emphasize the heightened effectiveness of health-targeted aid in low-income settings. Furthermore, some scholars argue that the impact of aggregate aid on health status is also evident in low-income countries. Gormanee et al.10 assert that aggregate aid bridges resource gaps, leading to remarkable changes through direct public health proj- ects such as communicable disease prevention and control, improved water supplies and sanitation, malaria control, and immunization programs. This positive effect of aggregate aid on health status has been corroborated by other studies.11,12 Conversely, another group of scholars argues that there is insufficient evidence to conclusively claim that health aid consis- tently improves health outcomes in recipient countries. This skep- ticism often stems from concerns about aid fungibility and poten- tial negative externalities. For example, Williamson,13 examining the impact of foreign aid commitments on the health sector across 208 developed and developing countries (1973-2004), found an insignificant effect. Similarly, Wilson’s14 empirical analysis of a panel data from 96 high-mortality countries (1975-2005) indicated no significant effect of health aid on recipient countries’ infant mortality rate. The primary argument for the ineffectiveness of health-target- ed aid centers on the concept of “fungibility,” where recipient countries may divert aid resources to non-targeted expenditures instead of channeling them into the health sector as intended. Pettersson15 highlights that such non-targeted expenditures, sourced from all development assistance, can be as high as 70%. Beyond fungibility, critics also argue that aid can negatively impact the competitiveness of aid-receiving countries, foster dependency, disincentivize the adoption of sound domestic poli- cies, and exacerbate corruption.16,17 The contrasting findings presented above create a significant dilemma for policymakers: should health aid be considered a com- plementary tool for achieving UHC, or should the focus be exclu- sively shifted to domestic factors? One root of this persistent con- troversy lies in methodological deficiencies within existing empir- ical studies. Specifically, issues such as misspecification problems, encompassing both weak functional forms and omitted variables, are prevalent in health estimating equations. The literature also rarely emphasizes separate short-run health aid effects, despite the fact that the size and significance of estimated marginal effects are strongly dependent on such time spans. Therefore, this research endeavors to address these critical gaps by employing a better-specified estimation equation that is consistent with a sound theoretical framework grounded in utility- maximizing human behavior.18-21 This will involve a rigorous examination of functional forms and the inclusion of relevant con- trol variables to mitigate omitted variable bias. Investigating the short-run effects of health aid on health out- comes in low-income countries, a dimension often overlooked in existing analyses, which predominantly focus on long-term impacts. This will provide more nuanced insights into the immedi- ate responsiveness of health indicators to aid inflows. By address- ing these methodological shortcomings, this paper aims to provide a more robust and reliable assessment of the effect of health aid on achieving UHC goals in low-income countries, offering clearer guidance for policy formulation. Materials and Methods Framework of the study Grossman’s health production model specifies a vector of inputs, where the variables of the vector include: nutrition, educa- tion, consumption of public goods, income, initial individual endowments like genetic makeup, time devoted to health-related procedures, and community endowments such as the environment.20 Following this approach, let the implicit function that relates these factors to health out- come H(t) as [1] where Wj(t) the input j>t variables are unobserved or not meas- ured, H(t) represents the health outcome at time t. In constructing health capital model, Grossman suggested the application of utility maximization constrained with resources which may require appli- cation of optimal control analysis.20 Building on these views it is assumed here that households derive satisfaction from their health status and they strive to maximize their utility constrained by socioeconomic and demographic factors. The common and very important solution from such utility maximization problem is the constancy of marginal effects of the input variable. That is after taking total derivative of equation [1] [2] The marginal effects fj are constants. Based on the constancy of marginal effects, one can integrate equation [2] to get [3] Article [page 248] [Healthcare in Low-resource Settings 2025;13:12874] Where A is some constant. In fact, in empirical analysis, to main- tain the result of optimal control analysis, i.e. constancy of the marginal effects, the input variables have to undergo some mathe- matical transformations, like log transformation, exponential trans- formation, depending on the measure of the input variable, other- wise the estimation equation will face a misspecification problem arising from the wrong functional form. In the specification of the health estimating equation, besides the wrong functional form, one may face the omitted variable problem, the case where a part of the input variables is unobservable, or the data may not be available. From introductory econometrics, we understand that ignoring these variables will make the coefficient estimates of the known variables unbiased. To deal with this issue, here it is assumed that the omitted vari- ables follow autoregressive of order two, which can be expressed as a Second Order Difference Equation whose particular solution and complementary function together form a function of time, i.e., for omitted variable The estimated Auto regressive of order two for Wj(t) can be written as whose complementary function and particular solution will be where complementary function Taking the total derivative of the variable and divide through by Wj(t) to get [4] is the elasticity of Wj(t) with respect to time. Assuming this elastic- ity to be constant (this assumption is derived from the belief that the growth of the input variables declines over time so that in the long run the variables exhibit stability. This stability, together with the constancy of marginal effects, would imply stability of the portion of health status generated by these variables) and integrat- ing both sides of equation [4], one gets [5] Substituting equation [5] in equation [3] one gets the long run health function as [6] Essentially, equation [6] is a long-run health equation since it is grounded on the constancy of marginal effects of the input vari- ables, which holds true in the long run. To drive the short-run health function from equation [6], the Partial Adjustment Model (PAM) is adopted. Intuitively, it is clear that the possibility that the coefficients in the health status estimat- ing equation [6] could be related to the level of change in health status before the input variables change. That is, keeping all other things equal, a one percent change in an explanatory variable in a population with a lower level of health status could have a higher effect than when a similar change takes place in another population with a higher level of health status. When the interest is to know the short-run effect, this phenomenon demands us to control for the previous level of health status. The PAM specifies the observed level of a given dependent variable as a weighted average of its level that existed in the previous time period and its equilibrium level at the present time, as [7] where l such that 00 or HDA>1.0 USD, and down- ward sloping to the right. The linearity of the curve in turn implies that the sought constancy of the marginal effect of the lnHDA is confirmed, i.e., the log transformation of the HDA is appropriate in estimating the health function. This figure also suggests that during the covered period of study for the sample countries, there was a maximum IMR, which was 83.5 infants per 1000 live births. Moreover, Table 1 informs that the average human capital index was 1.60. The least index was observed in Burkina Faso (1.14), whereas the highest index was observed in Tajikistan (3.14). The table also reports that for the income group, the aver- age adolescent fertility was 107.30 births per 1,000 women aged 15-19. The indicator was the highest in Niger (208) and the lowest in the Democratic People’s Republic of Korea (0.61). In fact, the data indicates that this indicator is falling over time at an average decline of -1.9 births per year. Moreover, the table informs that the elderly dependency rate was 5.88 per 100 working-age population. The indicator was the lowest in Sierra Leone (4.53%) and the high- Article Figure 1. Plot of IMR vs HDA per capita. Fractional Polynomial Fit [2000-2017]. Figure 2. Plot of IMR vs lnHDApc. Fractional Polynimial Fit [2000-2017]. [Healthcare in Low-resource Settings 2025;13:12874] [page 251] est in the Democratic People’s Republic of Korea (11.96%). Furthermore, Table 1 shows that during the covered years of study, in the considered income group, the mean composite index of governance was below zero (-0.96). It was below-2.0 in Somalia (-2.17) and South Sudan (-1.75), whereas it was above -0.3 in Senegal (-0.19) and Benin (-0.304). A look at the overall trend of the index reflects that it was declining, at an annual average of - 0.0088 with [95% Conf. Interval] of (-.0163, -.0013) i.e. institu- tional qualities are worsening substantially rather than improving during the covered period of study. Finally, the table reports that in the indicated time period the average cereal yield was 13.59 quintal per hectare of harvested land. It was below 5qt /hr in Eritrea (4.70qt/hr) and Niger (4.34qt/hr) and above 25qt/hr in the Democratic People’s Republic of Korea (35.05qt/hr) and Madagascar (28.76qt/hr). Considering an estimate of the effect of health-targeted aid on health status measure (IMR), whilst first difference GMM estima- tor result is shown for comparison purposes only as indicated on Table 2, the system GMM estimator was considered for a detail description of the results for the reason argued earlier.5,8,26,27 Consequently, like all GMM estimators, system GMM can produce consistent estimates only if the moment conditions used are valid. To test the validity of the over-identified restriction, the Sargan test is employed for it, unlike the Hansen test, which can be weakened by many instruments, but is not weakened by many instruments. In fact, Arellano and Bond show that the one-step Sargan test over-rejects in the presence of heteroskedasticity.23 In the case of the current study, the null hypothesis that the overiden- tifying restrictions are valid is not rejected. In its second half, Table 2 reports that the Sargan test of over-identifying restrictions accepts the null hypothesis that states the over-identifying restric- tions are valid, χ2 (102)=112.388, P=0.2279. Accepting this null hypothesis implies that the current study model or instruments need not be reconsidered. Hence, the test confirms the hypothesis that the instrumental variables should not be correlated with the residuals, and hence they are acceptable. Moreover, the table informs that for these countries the Wald test rejects the null hypothesis that states all the coefficients except the constant term are zero in both estimators. The table also reports that the coefficient of the lagged IMR is 0.5458 and is statistically significant, z=11.27, P=0.0000, confirm- ing the need for controlling for past effects of the independent vari- ables when the interest is in getting their short-term effects. In its robust version, the Arellano-Bond test for AR(2) in the first difference accepts the null hypothesis of no serial correlation in the idiosyncratic errors, which implies the instrumental variables are acceptable, z =1.411, P=0.1580. Besides, the table also informs that for the measured variables, the Wald test rejects the null hypothesis that states all the coefficients except the constant term are zero, Wald χ2 (9)=2890.76, P=0.0000 (Table 2). Moreover, as shown in Table 2, the coefficient estimate of log-HDA was - 1.9818, and this was statistically significant (P=0.0000). Similarly, a statistically significant estimate was observed for the log-GDPP coefficient, -9.6007 (P=0.0000). In the same way, the estimator gives -as a coefficient of expINST, 7.8092 (P=0.0160). The estima- tor also gives -0.8945 as a coefficient estimate of expHC, which is statistically significant at 10% level of significance (P=0.051). Sometimes the short run relative importance of the selected input variables together with their flexibility in policy decisions may be point of interest. Table 3 reports the shares of effects of the chosen variables’ effect in declining IMR from the annual average. In calculating the shares of the effects of the input variables, the previous level of IMR is unchanging for it has already been real- ized. Hence what determine the change in IMR from previous time up to the present time are changes in the input variables. Accordingly, during the covered period of study, in the sample countries, while the annual average change in IMR from the data was -0.9845 infants per 1000 live births the predicted change from the input variables using the chosen estimator was -1.1133 infants per 1000 live births, indicating the estimator predicted very close to what was observed in the short run (Table 3). The table informs that a decline in adolescent fertility, an increase in health aid, and an increase in per capita income play a major role in reducing infant mortality. In explicit terms, in the observed average annual IMR decline, 13.29 percent (- 1.9818x0.0747)/-0.9845) is due to an increase in health aid, 12.54 percent (-9.600x0.0145)/-0.9845) is due to an increase in per capita income and 15.68 percent (0.0921x1.8968)/-0.9845) is due to a decline in adolescent fertility. If left unchecked governance quality and yield were found to play an adverse role in the efforts made to reduce IMR (Table 3). Moreover, from Table 3, it can be under- stood that 46% of the decline in IMR was due to the selected input variables. Discussion In this study, the short-run health function analysis provides important insights for policymakers working to enhance health outcomes in low-income countries, particularly regarding health Article Table 1. Health-related indicators across low-income countries (2000-2017). Variable Obs Mean Std. Dev. Min Max IMR 612 65.25 24.16 13.80 142.00 HDA 440 8.38 7.68 0.00 48.38 GDPP 531 574.00 223.95 193.87 1309.23 HC 375 1.60 0.44 1.07 3.17 AFERT 578 107.30 48.08 0.29 217.16 EDEP 606 5.88 1.37 4.33 14.03 INST 568 -0.96 0.50 -2.43 0.06 YIELD 566 1359.09 715.33 158.20 4439.90 HDA, Health Development Aid, GDPP, Gross Domestic Product per capita, HC, Human Capital, AFERT, Adolescent Fertility Rate, EDEP, Elderly Dependency Rate, INST, Worldwide Governance Indicator, Yield, Cereal Yield. [page 252] [Healthcare in Low-resource Settings 2025;13:12874] aid allocation and governance reforms. The significant negative coefficient of log-Health Development Assistance (HDA) (- 1.9818, p=0.000) underscores the strong positive impact of health aid on reducing Infant Mortality Rate (IMR). Specifically, dou- bling per capita health aid is associated with saving approximately two infant lives per 1,000 live births. This finding aligns with pre- vious research by Mishra and Newhouse5 and Negeri and Haile Mariam,22 reinforcing the argument that properly allocated health aid significantly improves population health in low-income set- tings. The compelling relationship between health aid and IMR highlights the need for optimizing aid allocation to maximize its effectiveness. Policymakers should prioritize investments in inter- ventions proven to reduce infant mortality, such as maternal and child health programs, immunization campaigns, and improved access to essential healthcare services.3,2 However, direct aid is most impactful when it strengthens broader health systems. Investments in healthcare infrastructure, professional training, and efficient medical supply chains are necessary to bolster long-term health outcomes.13 Additionally, aligning health aid with domestic financing mechanisms, such as taxes or health insurance, can mit- igate out-of-pocket expenditures, a key factor contributing to financial hardship among vulnerable households.1 Thoughtfully integrating aid into sustainable financing structures ensures that it complements national efforts rather than replaces them. Beyond the role of health aid, our findings highlight the signif- icant influence of broader economic and social determinants. The strong negative coefficient of log-GDP per capita (-9.6007, p=0.0000) suggests that economic growth plays a pivotal role in improving population health. This aligns with Pritchett and Summers,34 who argue that higher income levels facilitate public health improvements through infrastructure expansion—such as improved access to safe water and sanitation—and enhanced Article Table 3. Estimates of the relative importance of the input variables. Variable Coef. Std. Err. z P>|z| Share in% _b[lnHDA]*0.0746858 -0.1480 0.0367 -4.0300 0.0000 13.29 _b[lnGDPP]*0.0145404 -0.1396 0.0331 -4.2200 0.0000 12.54 _b[lnYIELD]*0.0146193 0.0075 0.0170 0.4400 0.6590 -0.67 _b[expHC]*0.0784127 -0.0701 0.0360 -1.9500 0.0510 6.30 _b[expINST]*-0.0027128 0.0212 0.0088 2.4100 0.0160 -1.90 _b[AFERT]*-1.8968 -0.1746 0.0707 -2.4700 0.0130 15.68 _b[EDEP]*-0.0036 -0.0069 0.0033 -2.0800 0.0380 0.62 _b[lnTIME]*0.1700219 -0.6026 0.1804 -3.3400 0.0010 54.13 Sum -1.1133 0.1654 -6.7300 0.0000 100.00 Mean[d.imr] -0.9845 0.0404 -24.3600 0.0000 [Healthcare in Low-resource Settings 2025;13:12874] [page 253] Table 2. Estimate of IMR estimating equation, 2000-2017, one-step GMM results. Variable First difference GMM System GMM IMR Coef. Std. Err. z P value CI Coef. Std. Err. z P value CI L.IMR 0.6030 0.0541 11.1400 0.0000 (0.50, 0.71) 0.5372 0.0476 11.2700 0.0000 (0.44, 0.63) LnHDA -1.6653 0.6891 -2.4200 0.0160 (-3.02, -0.31) -1.9818 0.4919 -4.0300 0.0000 (-2.95, -1.02) lnGDPP -8.8893 3.1900 -2.7900 0.0050 (-15.14, -2.64) -9.6007 2.2734 -4.2200 0.0000 (-14.06, -5.14) LnYIELD 0.3882 1.2592 0.3100 0.7580 (-2.08, 2.86) 0.5115 1.1601 0.4400 0.6590 (-1.76, 2.79) ExpHC -0.8519 0.7793 -1.0900 0.2740 (-2.38, 0.68) -0.8945 0.4589 -1.9500 0.0510 (-1.79, 0.00) ExpINST -4.7728 4.1800 -1.1400 0.2540 (-12.97, 3.42) -7.8092 3.2433 -2.4100 0.0160 (-14.17, -1.45) AFERT 0.0485 0.0617 0.7900 0.4320 (-0.07, 0.17) 0.0921 0.0373 2.4700 0.0130 (0.02, 0.17) EDEP 1.5025 1.1151 1.3500 0.1780 (-0.68, 3.69) 1.9293 0.9289 2.0800 0.0380 (0.11, 3.75) LnTIME -3.5430 1.2644 -2.8000 0.0050 (-6.02, -1.06) -3.5444 1.0611 -3.3400 0.0010 (-5.62, -1.46) _cons 82.6748 26.8142 3.0800 0.0020 (30.12,135.23) 86.4150 19.7937 4.3700 0.0000 (47.51, 125.32) Sargan test of overidentifying restrictions Sargan test of over identifying restrictions H0: overidentifying restrictions are valid H0: over identifying restrictions are valid χ2 (89)=86.9263 P=0.5424 χ2 (102)=112.388 P=0.2279 Arellano-Bond test for AR(1) in first differences: z=-1.2039 P=0.2286 Arellano-Bond test for AR(1) in first differences: z=-1.247 P=0.2125 Arellano-Bond test for AR(2) in first differences: z =1.2937 P=0.1958 Arellano-Bond test for AR(2) in first differences: z =1.411 P=0.158 Wald χ2 (9)=5121.08 P =0.0000 Wald χ2 (9)=2890.76 P=0.0000 Number of Instruments 99, Number of Countries 24, Number of Instruments 112, Number of Countries 24, Number of Observations 265 Number of Observations 289 HDA, Health Development Aid, GDPP, Gross Domestic Product per capita, HC, Human Capital, AFERT, Adolescent Fertility Rate, EDEP, Elderly Dependency Rate, INST, Worldwide Governance Indicator, Yield, Cereal Yield. healthcare accessibility. Similarly, the statistically significant neg- ative coefficient of human capital (-0.8945, p=0.051) reinforces the view that human capital accumulation serves as a vital policy instrument.19,20 Individuals with higher levels of education and skills are better equipped to adopt health-improving technologies and behaviors, while families with stronger human capital can pro- vide more effective primary healthcare. Although cereal yield was not statistically significant in the short run, extensive literature emphasizes its long-term role in reducing mortality through improved nutrition. Scholars such as Cutler et al.11 and Fogel37 underscore the essential link between sustained agricultural productivity and health outcomes, support- ing the idea that food security remains a key long-term policy con- sideration in reducing mortality. One of the most striking findings of our analysis is the pro- nounced role of governance quality in shaping health aid effective- ness. Institutional quality (expINST) exhibits a significant nega- tive coefficient (-7.8092, p=0.0160), demonstrating that improve- ments in governance can yield health outcomes comparable to doubling health aid. Raising the governance index to 0.5816 from its current negative and declining average could save two addition- al infants per 1,000 live births—underscoring the critical impor- tance of robust governance in maximizing the effectiveness of health aid.31 To enhance aid effectiveness, policymakers must implement targeted governance reforms. The declining governance index sig- nals an urgent need to combat corruption and improve transparen- cy through rigorous anti-corruption measures, enhanced financial accountability, and robust oversight mechanisms.4,17 Strengthening rule of law and institutional frameworks is equally crucial, ensur- ing that health aid is disbursed efficiently and reaches its intended beneficiaries. Public financial management systems must be rein- forced, healthcare regulatory structures improved, and judicial independence safeguarded.9 Participatory governance also plays a crucial role in optimizing aid impact. By engaging local communi- ties and civil society in health policy planning and implementation, governments can foster ownership, ensure resource allocation aligns with local needs, and strengthen accountability in aid deliv- ery.14,15 Additionally, capacity building within government agencies is essential to effectively absorb, manage, and monitor health aid. Policymakers should invest in training programs focused on proj- ect management, financial oversight, and data-driven decision- making to enhance institutional efficiency and responsiveness.24,26 Without these governance improvements, aid effectiveness will remain constrained, reducing its long-term impact on population health. While our selected input variables explain 46% of the decline in IMR—leaving room for further policy exploration—the robust effects of health aid, income, human capital, and institution- al quality provide clear policy pathways. Prioritizing efficient health aid allocation alongside meaningful governance reforms is not merely a best practice; it is fundamental to maximizing aid effectiveness and ultimately saving lives in low-income countries. Recommendations Based on the findings this study, the following recommenda- tions were given for optimizing health financing in low-income countries: i) aligning with WHO guidance, donor funds should increasingly focus on reducing out-of-pocket health expenditures to prevent poverty; ii) low-income countries health policymakers, decision-makers, and all healthcare financing stakeholders should consider private foreign direct investment and workers’ remit- tances as essential complementary sources of health funding; iii) to enhance the effectiveness of aid in health sector, low-income coun- tries should implement strong governance frameworks that ensure transparency, accountability, and efficient resource allocation; iv) health aid recipient low income countries and their policymakers must prioritize the development of alternative domestic mecha- nisms to pool resources, safeguarding populations from cata- strophic health costs and reducing unsustainable dependence on external aid. References 1. Goryakin Y, Suhrcke M. The prevalence and determinants of catastrophic health expenditures attributable to non-communi- cable diseases in low- and middle-income countries: a method- ological commentary. Int J Equity Health 2014;13:107. 2. WHO. New perspectives on global health spending for univer- sal health coverage. Geneva: World Health Organization; 2018. 3. Levine R, Kinder M. Millions saved: proven successes in glob- al health. Washington: Center for Global Development; 2004. 4. Easterly W. The white man's burden. London: Penguin; 2006. 5. Mishra P, Newhouse D. Health aid and infant mortality. IMF Working Paper WP/07/100. Washington: International Monetary Fund; 2007. 6. Chauvet L, Gubert F, Mesple S. Aid, remittances, medical brain drain and child mortality: evidence using inter and intra- country data. J Dev Stud 2013;49:801-8. 7. Ebeke C, Drabo A. Remittances, public health spending and foreign aid in the access to health care services in developing countries. Etudes Doc CERDI 2011. 8. Mishra P, Newhouse D. Does health aid matter? J Health Econ 2009;28:855-72. 9. Chauvet L, Guillaumont P. Aid, volatility, and growth again: when aid volatility matters and when it does not. Rev Dev Econ 2009;13:452-63. 10. Gormanee K, Girma S, Morrissey O. Aid, public spending and human welfare: evidence from quantile regressions. J Int Dev 2005;17:299-309. 11. Cutler D, Deaton A, Lleras-Muney A. The determinants of health disparities. J Econ Perspect 2006;20:97-120. 12. Leunig I, Dijkstra G, Tuytens P. Health aid and health out- comes in developing countries. Forum Dev Stud 2024;51:45– 68. 13. Williamson CR. Foreign aid and human development: the impact of foreign aid to the health sector. South Econ J 2008;75:188-207. 14. Wilson S. Chasing success: health sector aid and mortality. World Dev 2011;39:2032-43. 15. Pettersson J. Child mortality: is aid fungibility in pro-poor expenditure decisive? Rev World Econ 2007;143:673-93. 16. Friedman M. Foreign economic aid. Yale Rev 1958;47:500-16. 17. Rajan R, Subramanian A. What undermines aid’s impact on growth? IMF Working Paper 05/126. Washington: International Monetary Fund; 2005. 18. Mushkin SJ. Health as an investment. J Polit Econ 1962;70:129-57. 19. Becker GS. Human capital. New York: Columbia Univ Press; 1964. 20. Grossman M. On the concept of health capital and the demand for health. J Polit Econ 1972;80:223-55. 21. Berman P, Kendall C, Bhattacharyya K. The household pro- Article [page 254] [Healthcare in Low-resource Settings 2025;13:12874] duction of health: integrating social science perspectives on micro-level health determinants. Soc Sci Med 1994;38:205-15. 22. Negeri G, Haile Mariam D. Effect of health development assis- tance on health status in sub-Saharan Africa. Risk Manag Healthc Policy 2016;9:33. 23. Arellano M, Bond S. Some tests for panel data: Monte Carlo evidence and an application to an employment equation. Rev Econ Stud 1991;58:277-97. 24. Blundell R, Bond S. GMM estimation with persistent panel data: an application to production functions. Econ Rev 2000;19:321-40. 25. Alonso-Borrego C, Arellano M. Symmetrically normalized instrumental variable estimation using panel data. CEMFI Working Paper No. 9612, September; 1996. 26. Blundell R, Bond S. Initial conditions and moment restrictions in dynamic panel data models. J Econom 1998;87:115-43. 27. Yousuf AS. Impact of health aid on infant mortality rate. MPRA Paper No. 4294; 2012. 28. Boone P. Politics and the effectiveness of foreign aid. Eur Econ Rev 1996;40:289-329. 29. World Bank. World development indicators. [Accessed 5 Mar 2018]; 2018a. 30. Feenstra RC, Inklaar R, Timmer MP. The next generation of the Penn World Table. Am Econ Rev 2015;105:3150-82. 31. World Bank. Worldwide governance indicators. [Accessed 5 Mar 2018]. Available from: http://data.worldbank.org/data-cat- alog/worldwide-governance-indicators; 2018b. 32. WHO. Global health expenditure atlas. Geneva: World Health Organization; 2014. 33. Negeri G. The long-run effects of health aid in low-income countries. J Public Health Afr 2023;14:2219. 34. Pritchett L, Lawrence HS. Wealthier is healthier. J Hum Resour 1996;31:841-68. 35. Preston SH. Causes and consequences of mortality declines in less developed countries during the 20th century. In: Easterlin RA, ed. Population and economic change in developing coun- tries. Chicago: Univ Chicago Press; 1980. 36. Wang L. Determinants of child mortality in LDCs: empirical findings from demographic and health surveys. Health Policy 2002;65:277-99. 37. Fogel RW. Economic growth, population theory and physiolo- gy: the bearing of long-term processes on the making of eco- nomic policy. Am Econ Rev 1994;84:369-95. 38. Link BG, Phelan J. Social conditions as the fundamental caus- es of disease. J Health Soc Behav 1995;35:80-94. Article [Healthcare in Low-resource Settings 2025;13:12874] [page 255]