Injectable Artesunate Assessment Report

Executive summary  
Malaria is one of the leading causes of morbidity and mortality in children under five in Africa. With timely and effective treatment, the risk of death can be significantly reduced. Following the results of clinical trials showing the high efficacy of injectable artesunate over quinine, the World Health Organization (WHO) recommended injectable artesunate as the preferred first line treatment for severe malaria in adults and children in 2011. However, uptake of the product remained slow. To help catalyze injectable artesunate procurement and uptake, Unitaid funded the “Improving Severe Malaria Outcomes” (ISMO) project from 2013-2016. Implemented in Cameroon, Ethiopia (two regions), Kenya, Malawi, Nigeria (13 states), and Uganda, the project supported each of these countries with the  quantification, procurement, distribution, case management trainings, supervision, and M&E systems to introduce and drive the uptake of injectable artesunate.

Since the ISMO project, few efforts have been undertaken to understand to what extent injectable artesunate has become (or remained) the preferred drug to treat severe malaria in the ISMO countries or in other malaria-endemic countries globally and, particularly, in sub-Saharan Africa, where most deaths from malaria occur. This report describes the findings of an assessment intended to answer this question.  The assessment consisted of the following key activities:

1) a global desktop review of published and grey literature on injectable artesunate introduction, availability, use, and prescription;

2) interviews with relevant stakeholders at the global level (i.e., donors, NGOs);

3) an analysis of available procurement data and 4) an in country ‘deep dive’ in three high endemic countries (DRC, Nigeria, Uganda) that included interviews with stakeholders at the central and subnational level, a supply chain analysis, and the extraction and analysis of health management information system (HMIS) and logistics management information system (LMIS) data where available. 

Key findings from the assessment presented in this report are: 

Guidelines and policy:
The majority of malaria high-endemic countries (80%) have adapted their malaria policies and guidelines to reflect the WHO recommendation for injectable artesunate as the preferred severe malaria treatment. However, 74% of countries do not make an explicit recommendation to use artemether over quinine when injectable artesunate is not available, and only 25% specify the dosing recommendations for children under 20kg, as per the latest WHO recommendations.

Procurement:
Injectable artesunate procurement has steadily increased since its adoption in the WHO guidelines in 2011 with an year-over-year growth rate of 73% over period 2011-2018. There are currently two WHO pre-qualified (PQ) suppliers of injectable artesunate: Guilin and IPCA. Guilin received WHO pre-qualification in 2011 and produces injectable artesunate in 30mg, 60mg, and 120mg formulations under the brand name Artesun. IPCA received WHO pre-qualification in December 2018 and only produces a 60mg vial product, under the brand name Larinate. Key findings from the procurement analysis include:

• Almost all (98%) injectable artesunate procurement is the 60mg vial; small quantities of 30mg and 120mg vials were procured in 2017-2018 through the Global Fund.
• Majority of procurement went to just five countries- DRC, Uganda, Kenya, Tanzania, and Nigeria have received almost 55% of the 123 million 60mg Artesun vials procured in the
• 2011-2018 period. 
• 87% of countries have received multiple donor shipments of injectable artesunate.
• Actual injectable artesunate procurement has been lower than the forecasted procurement - Forecasted procurement has been relatively consistent in the 2017-2020 
• Unitaid global quantification report but actual procurement has been under 85% of the forecasted need for 2017-2018.
• Prices have remained relatively stable (average of $1.53 per 60 mg vial) throughout the 2011-2018 period even though there was only one prequalified supplier for the drug.

 
Early adoption challenges:
We found four key challenges among countries introducing injectable artesunate. First, the required volumes of injectable artesunate were poorly quantified. In Nigeria, for
example, the first Global Fund procurement in 2014/2015 was based on a quantification that used a flat 5% severe malaria rate (i.e., 5% of uncomplicated cases) that proved to be too high, causing overstock in the Global Fund supported states, which had to be redistributed to non-Global Fund supported states.

Secondly, procurement delays during the ISMO project plagued the initial injectable artesunate orders due to extended price negotiations with Guilin and either delayed introduction in project countries or had them rely on other donors for procurements.

Third, there was (and in some countries still is) a preference for quinine over injectable artesunate due to the relatively inexpensive cost of quinine compared to injectable artesunate, ample supply, and the preference of some health providers to continue to use quinine. Fourth, there was, and in some countries continues to be, competition from non-WHO prequalified injectable artesunate products and alternative treatments that discouraged local procurement and uptake of PQ injectable artesunate. In countries such as DRC and Nigeria there are strong local markets for quinine or other artemisinin derived products that were established prior to injectable artesunate introduction. 

Poor adherence to treatment guidelines:
Following its introduction, misuse of injectable artesunate has been common and pervasive. In some countries, such as Uganda and DRC, injectable artesunate is used to treat cases of uncomplicated malaria, contrary to national guidelines. This excessive use of injectable artesunate for non-severe cases drives up consumption leading to stock outs, complicates patient treatment, and increases the cost of treatment for simple cases.

We found five drivers of this misuse. First and foremost, health workers receive insufficient training and a lack of regular supervision. A 2017 survey in DRC found only 42% of providers had been trained on PNLP guidelines.

Similar challenges  have been found in Uganda: a 2018 survey reported only 45% of health providers trained in severe malaria case management.

Second, insufficient health system financing encourages providers to prioritize services and treatments for which they can charge:

In some countries, consultations and hospitalization for severe malaria provide opportunities for providers to charge for services which results in patients opting for alternative and/or less expensive treatments (i.e., quinine). Third, injectable artesunate is used at lower levels of the health system, where its use is unauthorized, and where lower level providers and facilities often do not have the skill level, training, or resources to properly care for severe malaria and related complications. Fourth, the relatively high cost to patients of injectable artesunate and severe malaria treatment, in certain countries, can be a barrier to accessing treatment and forces patients to rely on less effective treatments or on drugs of unknown quality. Fifth, use of quinine is encouraged by its persistent availability and accessibility, its lack of donor funding which enables it to be charged back to patients, and its perception as efficacious.

Finally, ACTs may not be provided after the patients has received injectable artesunate, contrary to national guidelines. Reasons for this are not certain but likely due to a lack of knowledge and negligence on the part of the provider or stock outs of ACTs. 

Based on the findings in the assessment and identified best practices, six key recommendations are suggested to improve injectable artesunate availability and use:

1. 1) Country malaria programs should improve quantification and forecasting of injectable artesunate by strengthening severe malaria data availability and quality through interventions such as regular HMIS data quality audits and support to facilities on reporting and data use. Improving facility reporting and data use can help improve the quality of consumption data used in quantification, supply and distribution planning, and stock monitoring.
2. 2) Country malaria programs should strengthen the capacity of supply chain staff to minimize stock outs and improve quantification and forecasting capacity through facility and warehouse logistics training, mentoring, and supervision. More accurate quantification, reporting, and ordering can reduce product rationing.   
3. 3) Country malaria programs, in collaboration with partners, should work to encourage injectable artesunate rational use. Examples could include: 

• a. Revise training curricula and supervision through approaches such as on-the-job training, mentoring, and clinical audits to ensure proper use of the drug and patient care. Targeted clinical audits, for example, may improve reporting, severe malaria case management, and overuse of injectable artesunate. Training and supervision particularly need to highlight the prescription of ACTs following injectable artesunate treatment as well as the superiority and higher efficacy of injectable artesunate compared to quinine. 
• b. Work with WHO to ensure alignment between severe disease classifications and management practices at different levels of the health system such that severe malaria patients receive injectable artesunate while those with uncomplicated malaria or other illnesses do not. Countries that have not yet made a clear recommendation of injectable artesunate as a first-line treatment, do not having dosing specifications for children under 20kg, or do not recommend artemether as the preferred alternate treatment, should also work with WHO to align its guidance to global WHO recommendations. Donors could play an advocacy role or provide technical support to make any revisions.  
• c. Develop and implement patient registration/tracking systems for those patients with severe malaria who were prescribed injectable artesunate. For example, in Uganda, a hospital rolled out an injectable artesunate-patient registration/tracking system (for every person who receives the product, the pharmacist records the name and the number of vials received), which significantly reduced overconsumption.  Improved rational product use could lead to lower and more targeted injectable artesunate use and reduce stock outs, rationing, and reliance on the open market. 

1. 4) Donors should work closely together with country programs and other in-country stakeholders to ensure free severe malaria treatment by leveraging comprehensive primary health care and maternal and child health projects and initiatives (e.g., free treatment schemes for children under five, flat service fees, insurance schemes). Country malaria programs should ensure that malaria drugs and treatment are included in these non-malaria-specific interventions.
2. 5) National regulatory authorities such as the MoH Department of Pharmacy and Medicine and National Drug Authorities should work to strengthen their pharmacoviligence systems in countries to reduce the availability of sub-standard or fake artemisinin derived injectable drugs, particularly those not registered for use with the national drug authorities.
3. 6) Donors and implementing partners should support severe malaria case management initiatives that provide comprehensive care (rather than funding procurement of injectable artesunate alone) to ensure that the necessary drugs and services to treat severe malaria complications,such as blood transfusions are consistently available and provided to patients.  

Overall, it seems that the ISMO project has had a lasting effect on procurement and use of injectable artesunate for the treatment of severe malaria in the two ISMO project countries examined here, particularly in Uganda. The procurement analysis found that procurement has steadily increased over the years since 2012, through and then following the ISMO project. The ISMO project found that training and supervision can reduce misuse, particularly for uncomplicated malaria cases, although findings suggest that discontinuation of such activities may result in these effects waning. Since the ISMO project, use of injectable artesunate use has increased, albeit with challenges around misuse. Finally, analysis of the (very limited) data available (from DRC and Uganda) was unable to identify a discernable difference in the case fatality rate among facilities with and without injectable artesunate. However,
poor data quality and confounding factors, such as the possibility that facilities with injectable artesunate might have started off with higher case fatality rates than those without, limit the conclusions that should be drawn from this analysis.