A conversation with Amanda Glassman on November 13, 2013


Note: These notes give an overview of the major points made by Ms. Glassman in the conversation.


Amanda Glassman is the Director of Global Health Policy and a senior fellow at the Center for Global Development.

Good Ventures and GiveWell spoke to Ms. Glassman about promising opportunities for philanthropy in global health, including mental health interventions in the developing world, behavioral-economics-inspired interventions in health and nutrition, and improving monitoring and evaluation (M&E) and prioritization among major global health funders.

Mental health

According to The Lancet's 2010 Global Burden of Disease Study, depression is one of the biggest contributors to the burden of disease in developing countries. Mental health services in low- and middle-income countries tend to use antiquated techniques and to make little use of therapy and cost-effective interventions.

There are studies that demonstrate the cost-effectiveness of certain mental health interventions in low- and middle-income countries. Trials are currently ongoing for community-based mental health care models, which are fairly affordable (for example, the MANAS trial in India http://www.ncbi.nlm.nih.gov/pubmed/21159375?dopt=Abstract). These promising mental health interventions are likely more cost-effective than many interventions subsidized by aid today in terms of health impact.

Most mainstream funders do not fund mental health interventions. This may be because of stigma around mental health issues or a perception that it is difficult to improve mental health, and –as a non-communicable disease- there are no cross-border externalities that characterize infectious diseases and the rationale for their control. Exceptions are the Wellcome Trust, which funds experiments, some at a large scale, in cost-effective mental health care, and Grand Challenges Canada, led by Dr. Peter Singer. It would be beneficial to encourage funders to put more resources into this area; however, there are currently few potential grantees that could scale up quickly.

Early childhood nutrition affects cognitive skills later in life, impacting learning, wages, and mental health. Christina Paxson and Norbert Schady found that among children in Nicaragua who had been poorly nourished, over one half qualified as functionally impaired based on their performance on cognitive tasks. Interventions to improve childhood nutrition are thus one way to improve mental health.

Behavioral economics and interventions in health and nutrition

There has been a great deal of research on using financial incentives to encourage better health and nutrition behaviors in the developing world. That research continues to produce exciting results.

Non-financial incentives have been less well-studied. Paternalistic policies that prevent people from making decisions that have high negative externalities should be studied further. For example, to encourage saving, financial transfers could be deposited into bank accounts rather than given in cash. This might help people accumulate enough money to invest in productive assets.

One-time interventions can lock people in to healthy habits. For example, a program in South Africa subsidized vegetables. People continued to buy vegetables at an elevated rate six months after the end of the program, showing that the program caused lasting change in their habits. In the US, Dan Buettner's Blue Zones project encouraged mayors to compete to adopt a package of interventions designed to promote health in their cities. The interventions included walking groups, bike paths, and reduced portion sizes in restaurants. These packages of interventions have the potential to lead to large improvements in health outcomes.

Improving practices of global health funders

Within global health, one of the most cost-effective activities for a philanthropist may be to work to improve the practices of major global health funders. About $28 billion per year is spent on global health aid. This accounts for about a third of total aid. Health is the biggest category for US aid. It is important to make sure that funds are spent as efficiently as possible.

Monitoring and evaluation

More funders should publicly advocate for transparency and evidence of results in global health aid. The Gates Foundation attempts to tacitly influence decision makers to improve M&E, and the efficiency and effectiveness of global health funders. However, in the public domain, it is much more vocal about fundraising and advocacy for global health.

The Gates Foundation has recently provided a grant to PATH to collect more accurate data on immunization coverage, and to Johns Hopkins to support more frequent and accurate measurement of contraceptive prevalence. The Foundation and its partners have yet to explicitly connect those monitoring results to GAVI's funding of countries, but it likely will in the future. To Ms. Glassman’s knowledge, the Foundation has not undertaken similar M&E for AIDS, tuberculosis (TB) or malaria. The Global Policy and Advocacy Group at the Foundation funded CGD's research and working group on improving the Global Fund to Fight AIDS, Tuberculosis and Malaria's (GFATM) cost-effectiveness. CGD worked closely with the AIDS team at the Foundation, which emphasizes efficiency (procuring health supplies and services at the lowest possible cost). Ms. Glassman also sees performance management (ensuring that health results are measured accurately and achieved) as the first priority.

The Office of the Inspector General (OIG) of GFATM monitors GFATM's work. The OIG focuses on ensuring that funds were spent for the precise purpose that they were budgeted for rather than on determining whether programs had the desired results. The Secretariat itself has very limited staff and budget to monitor the results of GFATM programs (2 full-time it seems), yet the OIG –dedicated mainly to financial audit- employs about 36 people. It is an expression of the relatively low priority of performance measurement and evaluation, versus financial and fiduciary audit. The majority of GFATM's employees work on grant management rather than M&E. GFATM should make improving performance management its top priority.

GFATM and its partners should document the interventions that are funded by its monies within each country. It should also rigorously measure a few, important indicators related to the diseases it focuses on for each of the 20 or so countries receiving the most funds to address each disease. Collecting good data is difficult, but this level of monitoring should be feasible. The monitoring should be observational rather than experimental. Experiments can be expensive, and GFATM is one of many actors working on each disease, making attribution of impact difficult.

For example, for malaria, GFATM should track case rates (if possible), bed net use by households, and appropriate diagnosis and treatment of malaria cases. In AIDS, retention rates are critical, as is incidence and prevalence among key populations. These indicators could be tracked via a random sample of healthcare facilities and households and program enrollees in target areas. GFATM could pay for performance on these indicators. Paying against objective indicators would likely improve the quality of data collection.

The World Bank's Health Results Innovation Trust Fund undertakes this kind of verification of health coverage indicators. It primarily uses data from healthcare facilities. It employs data from other sources, such as household surveys, as checks on the healthcare facility data. It is experimenting to determine how many different checks on the primary data should be used to strike the right balance between expense and data quality.

GFATM may hesitate to more closely monitor performance because it was heavily criticized when its audits found evidence of fraud and corruption. However, if GFATM could document impressive results from its programs, it could argue that it is an effective organization even though some small-scale corruption occurs. Another reason that GFATM may be reluctant to improve performance management is that recipients have a seat on its board. Cost is unlikely to be the main restriction on GFATM's M&E.

Mark Dybul, the executive director of GFATM, could choose to start a major M&E initiative, ideally in cooperation with PEPFAR and a consortium of the most AIDS aid-dependent countries. Donors seem to lack the coordination and interest in M&E to push the GFATM secretariat to undertake more M&E.

The President's Emergency Plan for AIDS Relief (PEPFAR) has expressed interest in paying for more rigorous M&E in countries that receive the most AIDS funding from GFATM. This M&E would track how money is spent and offer suggestions for improvement. GFATM would likely welcome such an initiative.

More information on these ideas is included in CGD’s recent report More Health for the Money.

Priority setting

By governments

In most countries, health spending by governments vastly outpaces international health aid, so governments set most health priorities. There are a few highly aid-dependent countries where donors have huge influence on health priorities. However, there are only about four countries in the world where aid accounts for more than 40% of health spending. If current growth trends continue, low- and middle-income countries will become less dependent on aid over time.

CGD ran a working group to assess whether developing countries have explicit mechanisms to allocate new funds and reallocate existing funds to cost-effective interventions and products, and how well they work to inform resource allocation. CGD found that there are successful efforts to explicitly prioritize in many countries using essential medicines lists, health benefits plans and health technology assessment. However, these efforts are nascent and require greater support.

The UK's National Institute for Health and Clinical Excellence (NICE) advises that country's National Health Service on the most cost-effective uses of funds within each disease area. This is part of the reason why the UK spends less on health than the US yet has better health outcomes. NICE International is setting up a support hub to facilitate health priority-setting pilots with interested countries.

By global funders

Large funders such as GAVI and GFATM require co-financing from recipient countries for certain projects. It is unclear whether co-financing makes recipient countries allocate funds more or less efficiently. New allocation rules have recently been established to determine the split of resources between countries and between diseases, based on country GDP and disease burden, among other factor. Under discussion but not yet in place is the use of cost-effectiveness criteria to determine which technologies (medicines, diagnostics, devices) should be eligible for funding by both organizations. CGD will continue to study the implications and where possible the effects of the new allocation and eligibility policies on efficiency and effectiveness.

Donors have not yet been able to reach consensus on additional diseases to add to the mandate of GFATM beyond AIDS, TB, and malaria. However, GFATM might accept funds to start working on a new disease if such consensus exists.

The Institute for Health Metrics and Evaluation (IHME) does high-quality research on disease prioritization based on disease burden, but this research could receive greater attention from the global health community. This limited attention may be because many in the global health community are primarily concerned with raising money for the diseases that they work on rather than prioritizing between diseases.

More information on CGD work on priority-setting is available here: http://www.cgdev.org/publication/priority-setting-health-building-institutions-smarter-public-spending