This is a joint post with Heather Lanthorn.
The Global Fund Board’s decision over the Affordable Medicines Facility – malaria (AMFm) rapidly approaches, and tensions within the malaria community are acute. In her global health blog for The Guardian, Sarah Boseley characterizes the rift as one of
“huge arguments and intense passions…[because] it is about the belief on one side that the private sector is the most effective way to get medicines to those who need them – and the certainty on the other side that bolstering the public sector to diagnose and treat people is a fairer and safer way to go. These are not just practical matters, but highly political.”
No doubt, the debate on the AMFm has devolved into bickering and accusations from many sides. But the overstated rhetoric obscures genuine differences of opinion on how best to move forward with an evidence-based decision-making process, and what counts as “evidence” sufficient to approve, modify, or scrap the program. Both the sufficiency and the relevance of available evidence on AMFm have been questioned, particularly regarding its impact on children and the most socio-economically vulnerable. Below, we try to cut through the noise and make sense of the relevant evidence.
The AMFm was designed to expand access to the most effective treatment for malaria – artemisinin-based combination therapies (ACTs). The basic logic of AMFm is that by dramatically reducing the price of quality-assured ACTs and by leveraging all existing distribution systems (public and private), the availability, market share and usage of these drugs would increase. Proper usage of these drugs against malaria, in turn, would decrease morbidity and mortality from malaria. Moreover, the low-cost, quality-assured ACTs could potentially crowd-out less effective drugs on the market and delay the onset of drug resistance to artemisinin.
So what’s the hitch? Much of the debate about AMFm’s effectiveness centers around one point: Who uses these subsidized ACTs, and how?
Here, the word ‘Use’ raises several points for debate. First, has use of ACTs changed with AMFm? Second, has use of ACTs changed specifically for children and vulnerable populations? Third, is use commensurate with (epidemiological) need? Fourth, what is the effect of AMFm on misuse and overuse, both of which have implications for drug resistance, overall morbidity, and efficient use of aid dollars?
Overall use: Critics argue that there is scant evidence regarding AMFm’s impact on ACT use at the household level. The most vocal critic has been Oxfam, whose recent report, entitled “Salt, Sugar, and Malaria pills,” argues for AMFm’s immediate discontinuation.
The lack of evidence on household-level effects was partly self-inflicted; for reasons apparently related to budgetary constraints and concerns that downstream effects would not be apparent in such a short time frame, the Global Fund chose not fund the measurement of ACT use as part of its independent AMFm evaluation. Yet the lack of evidence does not necessarily mean evidence of no effect. Aiming to fill the evidence gap, a newly published brief by the Clinton Health Access Initiative (CHAI) summarizes the results of a systematic review which shows that national ACT subsidies produced huge increases in ACT uptake versus other anti-malarials in a sample of sub-Saharan countries.
Use by population: As reported in the CHAI brief, the proportion of anti-malarials purchased for children under 5 which were ACTs in both the public and private sectors rose substantially in three AMFm countries – Madagascar (37 percentage points), Nigeria (16 percentage points), and Uganda (42 percentage points) – during the implementation period (plus or minus) for the AMFm subsidy (though the underlying data has not yet been made public). CHAI’s survey results also showed increased ACT among children in the lowest wealth quintile (i.e. the poorest children), and those in rural areas. The CHAI brief supplements the formal AMFm evaluation, for which the main outcomes were facility-level availability and price of ACTs, in both the private and public sector (see the full evaluation here and the newly published Lancet paper here). But even with only facility-level information, it seems highly improbable that private facilities and shops were fully stocked, only to sell them to ghost patients or to throw them away.
Indeed, the design of AMFm was predicated, in part, on evidence indicating that 50% or more of fevers in sub-Saharan Africa are treated through the private sector, where anti-malarials have de facto, if not always de jure, over-the-counter status. The CHAI findings are consistent with the logic that poor patients tend to seek care more frequently in the private sector than less poor patients. We are unaware of evidence on the extent of anti-malarial use in the public sector being more pro-rich or pro-poor than the private sector (see some indirect but related evidence here and here). A simple exercise would be to count the number of public and private providers in a country by urban and rural areas; one would likely find that public health-care providers and facilities, more than private providers, are disproportionately located in urban and wealthier areas.
Use as need: Because many fevers are treated presumptively rather than by diagnostic test, the extent of over-treatment of patients presenting with fever is not well known. The Oxfam report makes a broad comparison of some 32 million cases of malaria vis-à-vis some 156 million doses of ACTs, which may suggest over-use, although we are unable to confirm how these numbers were calculated and whether the figure on courses refers to adults, children, or something else.
A brand new Science paper published on Friday by CHAI authors suggests that:
- Appropriate use is heterogeneous geographically; the likelihood of a fever being malaria differs widely across space, time-of-year, and age of patient. This problem will remain so long as there are weak or non-existent incentives to use and follow proper diagnostics.
- Overtreatment is a problem in both the public and private sectors in many countries, and there are always incentives for doctors and pharmacists alike to overprescribe medicines, in hopes that the patient feels better from some combination of an ACT, an antibiotic, a painkiller, a fever-reducer, and a vitamin.
Distinguishing ideology and evidence
The debate between AMFm detractors and supporters has not been fully grounded in operational realities. Instead, it has too often centered on perceptions about rights, norms and ideals, e.g. Oxfam’s position that people should not pay for medicines at all, and a broader discussion about the appropriate roles of the public and private sectors in delivering health commodities – especially those commodities whose misuse can generate a great public bad. These perceptions, in turn, are too often based on stereotypes and idealizations of the equity, fairness, and effectiveness of each sector. AMFm has such been sucked into such debates, even though it has done well (if not far better than others) in conducting impact evaluation on those outcomes feasible to measure given the timeframe.
As a condition of its coming into being, funders have required a higher standard of evidence for AMFm’s private sector delivery strategy than for the traditional delivery approaches used by multilateral or bilateral health organizations that are also aimed at expanding access to ACTs. For example, it is absolutely important to promote appropriate use of ACTs through diagnostics – but inappropriate and over-prescriptions are not exclusive to the private sector (though possibly more common), and private sector delivery mechanisms are not necessarily incompatible with better diagnosis. Presumptive treatment of fever with anti-malarials is a concern that requires further attention and potential modifications; it is not, by itself, grounds for invalidating private-sector delivery strategies.
There is no question that AMFm must be considered within the bigger picture of malaria efforts globally, particularly by other donors. The malaria community, particularly within the United States, has faced tremendous political pressure to ensure “value for money”—arguably with far more intensity than exists for other disease areas. Value for money in a resource-constrained environment often means facing hard trade-offs – trade-offs between target countries, disease areas, and yes, intervention choice – including the means to scale-up appropriate ACT use.
It is good that the malaria community is seriously assessing trade-offs and is vigorously debating the most effective use of scarce resources and there are many things to applaud about AMFm being launched as a pilot with a set review date. Evidence needs to be at the core of these discussions. Ultimately, all malaria advocates share the same goal: to reduce the burden of malaria and the burden it places on human and economic development.
The authors thank Amanda Glassman, April Harding, Rachel Silverman, Jenny Ottenhoff, and several others for helpful comments