Tropical Health Matters

Malaria Journal published a few days ago an article comparing the costs of treating children and adults for malaria at a Nigerian hospital based on clinical diagnosis versus treating only when microscopy was positive for parasites. Normally we would pass abstracts from such articles on to members of our listserve (see link at right), but comments from a colleague in Nigeria gave pause.

He rightly pointed out that normally any research that helps us consider the factors involved in proper malaria diagnosis and treatment is welcome as we move toward universal coverage and elimination. His concern was that the researchers, who conducted their study in 2009, had not followed national malaria treatment policy and guidelines, which had been promulgated in 2005 based on the alarming growth of resistance of malaria parasites to the common, though cheap, antimalarials such as chloroquine and sulphadoxine-pyrimethamine (SP).

First the cost findings from the team at Bowen University Teaching Hospital (aka Baptist Medical Center, Ogbomosho) –

• For children, testing all but treating only Giemsa positives was $6.04/child
• Empiric treatment of all children clinically diagnosed was $4.49/child
• For adults, treating only Giemsa positives was $4.84/adult for treatment option one and $4.97/adult for option two
• Empiric treatment for adults was $4.14/adult for option one and $4.63/adult for option 2

In spite of the cost findings, the researchers did point out the drawbacks of empirical or clinical diagnosis in terms of accuracy and potentials for promoting drug resistance and called for scale up of rapid diagnostic tests (RDTs) to address these concerns.

The treatment regimens in this study included …

• Pediatric patients: artesunate (6-9 tablets of 3 mg/kg on day one and 1.5 mg/kg for the next four days) plus amodiaquine (10 mg/kg on days one to two and 5 mg/kg on day three in suspension)
• Adult option one: four and one-half 50 mg artesunate tablets on day one and nine additional artesunate tablets over the next four days, plus three 500 mg sulphadoxine/25 mg pyrimethamine tablets
• Adult option two: four and one-half 50 mg artesunate tablets on day one and nine additional artesunate tablets for the next four days plus nine 200 mg tablets of amodiaquine at a dose of 10 mg/kg on day one to two and 5 mg/kg on day three

National treatment guidelines specifically stress use of artemisinin-based combination therapy (ACT) for basic, uncomplicated malaria treatment.These guidelines are undergoing further revision with a stronger emphasis on ACT use based on RDTs and microscopy where available and recognition of the dangers of monotherapy drugs like chloroquine, SP and even artesunate itself.

The researchers from Ogbomosho are rightly concerned about cost issues, and being a private/NGO university and hospital, they do witness the direct effects of medication costs on patients that health staff in the public sector may not see.

This is still no excuse for not following national treatment guidelines when these drugs were available for their procurement in 2009. Now with the advent of the Affordable Medicine Facility malaria (AMFm) in Nigeria all health facilities, especially NGO hospitals like Ogbomosho, have no reason not to buy and dispense the correct medicines.

To re-emphasize this point, a press release from November 2010 clearly states –

“The Federal Government has directed all medical doctors and other health officials in the country to henceforth start using Artemisinin-based Combined Therapy (ACT) for the treatment of malaria disease in the country. Minister of Health, Prof. Onyebuchi Chukwu, gave the directive yesterday in Abuja during the ministerial press briefing on Affordable Medicines Facility (AMF) for malaria programme. According to the minister, the spread of malaria had become so critical that everyone in the country was now involved.”

We hope health workers in all sectors get the word! Hopefully national authorities will step up their efforts to disseminate guidelines to all front line health workers whether in public, private or NGO sectors.

During a recent malaria diagnostics assessment to Burkina Faso, we found health workers who were concerned that malaria rapid diagnostic tests showed fewer positive results in the dry season even though they often suspected that their patients had malaria. The big challenge for health workers is accepting the fact that even when they use clinical algorithms, not all febrile illnesses that they suspect to be malaria are actually malaria.

The dry season in the African Sahel is the period for epidemics of meningitis. Seasonal epidemics of meningitis kill thousands in Africa every year.

According to CDC “Meningitis infection is characterized by a sudden onset of fever, headache, and stiff neck.” These are similar to early malaria symptoms, but in addition CDC says that people with meningitis may experience nausea, vomiting, photophobia (sensitivity to light), altered mental status.

BBC reports that, “For the people in Niger, Mali and Burkina Faso, a new meningitis vaccine offers hope of an escape from one of the world’s deadliest, most disabling and infectious diseases. So there is little wonder that the queues were enormous when a pilot project for the MenAfriVac vaccine got underway in the three West African countries in recent weeks.”

This vaccine was developed specially for Africa and costs around 50 US cents per dose and should be effective for 10-15 years. Meningitis A is “caused by the bacterium Neisseria meningitidis group A, which mostly attacks infants, children, and young adults. It accounts for ninety per cent of all meningitis epidemics in Africa. The outbreaks strike during the dry season. In 1997, in the worst epidemic on record, 25,000 people died,” as reported by the BBC.

As with other public health interventions, scaling up of this new vaccine will be on overcoming dependent on logistical challenges, in this case the need for sustained funding. As noted, the effort will being in only 3 countries, but “450 million people … are at risk of this disease … in the very well known African meningitis belt.”

Preventing meningitis in Africa will not only save lives directly, but should reduce the chances that a febrile child is misdiagnosed as having malaria and allowed to die from another disease.

The need for rapid diagnostic tests (RDTs) for malaria case management has never been more clear since of the publication by Gething et al. that only 43% of febrile episodes in malaria endemic countries are actually malaria. They did find that fever is a better indicator of malaria in areas of higher transmission, but in no situation can a clinician be confident that fever equates automatically with malaria.

Unfortunately, that confidence is more the rule than the exception.  In Burkina Faso, for example, health workers are taught to use the algorithm to the right in diagnosing malaria in patients greater that 5 years of age (at present there are not enough RDTs to use on children below 5 years).

In reality, health workers still provide artemisinin-based combination therapy (ACT) antimalarial drugs to most people who present with the classical symptoms of fever and headache.  Even when RDTs are used, negative tests are frequently treated with ACTs.

Health workers explain that during their training they were told that a negative test does not mean the person does not have malaria. While this is true to a very small extent, RDTs in current use are more that 90% accurate if stored and used properly. Gething’s results should cause health workers to think harder.

Earlier this month Bisoffi and colleagues reported that …

According to microscopy, in the rainy season versus the dry season, the prevalence of malaria infection in patients presenting at primary health care centres was three times higher in febrile patients and twice in non-febrile patients. In the dry season, only a small proportion of fevers were attributable to malaria.

The Burkina Faso research found seasonal variations in the sensitivity and specificity of the RDTs, but concluded for adult patients that, “RDTs appear to be most useful during the low transmission season: a negative test safely excludes malaria and would avoid most unnecessary treatments, if prescribers are convinced to rely on the negative result.” Health workers need to hear this information.

Other concerns about positive predictive value of RDT use in Burkina Faso, and people should read the article by Bisoffi, but the take home lesson is that a negative result can be trusted and is an important guide to using expensive stocks of ACTs more judiciously.

Not all fevers are malaria. This should not be an earth shaking statement, but national treatment guidelines in malaria endemic countries often stress presumptive treatment for malaria, especially when children present with fever. Irin explains that even the World Health Organization has been hard pressed to recommend otherwise when accurate parasitological diagnostic resources are unavailable.

The concern about over-diagnosis of malaria is hitting home though because the current first line treatment, artemisinin-based combination therapy (ACT) is quite expensive, and additionally, health experts are concerned that overuse or misuse of these drugs may foster parasite resistance. To make this point even stronger Peter Gething and colleagues found that, “Of the 183 million children with malaria symptoms treated by public health clinics in 2007, only 43 percent were diagnosed with malaria, but many more most likely received anti-malarial medication (IRIN News).”

A variety of febrile illnesses, especially from mosquito-borne diseases, occur in the same community. A news report from Vapi, India states that, “During the (previous week), 13 cases of chikungunya, six of dengue and 25 of malaria have been reported from in and around Vapi with Nehru Street being the most affected. The outbreak of mosquito transmitted diseases has made the health officials in the district rush to the city to initiate measures for vector control.”

A serological-epidemiological household survey in Sudan after a yellow fever outbreak found, “serologic evidence of recent or prior chikungunya virus, dengue virus, West Nile virus, and Sindbis virus infections.”

Källander and colleagues stressed the challenges of clinical diagnosis to differentiate malaria and pneumonia and reported that, “Of 3671 Ugandan under-fives at 14 health centres, 30% had symptoms compatible both with malaria and pneumonia, necessitating dual treatment. Of 2944 ‘malaria’ cases, 37% also had ‘pneumonia’.”

Gething’s group did find that 72% of those febrile cases that actually were malaria were found in locations that had higher parasite prevalence. Possibly clinicians in more highly endemic areas can presume correctly more often that a fever is malaria, but this still does not stop the wastage of ACTs, which will continue until the parasitological testing gap is closed with adequate supplies of rapid diagnostic tests and microscopes (and the skills to use these).

Gething and colleagues stress the need for countries to develop appropriate strategies by adapting the statistical models they developed with more country based data. They sadly conclude that, “Unfortunately, inadequacies in national health management information systems across Africa are in part a cause of the present imperfections in essential  commodity demand and burden estimation.”

It would be even sadder if much of the treatment commodity supplies distributed in 2010 to achieve universal coverage of malaria interventions were wasted on non-malaria fevers.

The Malaria Society of Nigeria is planning a seminar to update members and those concerned about controlling malaria in the country on management of malaria. The event will take place at the Nigerian Institute for Medical Research in Yaba, Lagos, on 12 May 2010 at 10 a.m.

There are many aspects to managing malaria, but to take only one – case management – is a challenge in itself.  The National Malaria Control Program‘s 2010 annual workplan outlines five key activities that need to be accomplished in order to properly treat a person who has suspected malaria:

• Parasitological confirmation of malaria cases by rapid diagnostic tests (RDT) and scaling up of diagnosis by microscopy
• Treatment of uncomplicated malaria with an ACT within 24 hours of fever onset through all health care providers (public and private)
• Expansion of access to free ACTs to community level through local human resources
• Early recognition and improved management of severe malaria cases
• Drug efficacy and quality monitoring

To this we should add ‘counseling’ of those receiving ACTs to ensure adherence to the full course of treatment.  As a recent Malaria No More posting noted, “The only pill that works is the pill that’s swallowed.”

The current national malaria treatment policy, guidelines and training materials were adopted in 2005. While these are technically correct in terms of stressing ACTs, but there is still a reliance on clinical or symptomatic diagnosis. While parasitological diagnosis is addressed in the current workplan, it also needs to be disseminated in easy to read guidelines and training materials.

Much has changed in the five years since the last malaria treatment policy and guidelines were adopted including the pressing need to use rapid diagnostic tests in primary health care facilities, the huge multiplication of brands of ACTs on the market, the impending large scale roll out of home management of malaria through community volunteers and patent medicine vendors, and related to the latter, the award of a pilot Affordable Medicines Facility (malaria) grant.

These changes are built on an unsteady foundation as documented in the 2008 Demographic and Health Survey. Three years after the national treatment policy had been updated, ACTs were very rarely reported in malaria treatment, as seen in the chart.

An ACT Watch survey in December 2008 of 468 medicine outlets (public and private) found that only 16.7% had the national firstline ACT – artemether-lumefantrin.  In all cases, the most common antimalarials in stock were non-artemisinin drugs.

Increased malaria funding for Nigeria from the Global Fund, DfID, USAID and the World Bank Booster Program should make ACTs and RDTs more readily available if supply and distribution systems are strengthened. This will only be effective if health professionals understand the national malaria treatment policy and the case management implications of proper parasitological diagnosis. We hope that the Malaria Society’s upcoming seminar can contribute toward this goal.

TropIKA.net is one of the latest to comment on the World Malaria Report and data from specific countries that show a drop in malaria cases that are ‘believed’ to be a results of stepped up intervention. Zambia, which is reporting a 50% drop in cases, is contrasted with Sierra Leone where there is a reported increase coupled with malaria control program implementation challenges.

In another part of the world, often known as a seed bed for malaria drug resistance, an increase is also reported. In Cambodia …

Figures for malaria cases in 2009 are still being tallied, said Ministry of Health and World Health Organization officials, but are already higher than in 2008, when there were 58,887 cases and 209 deaths. In 2009, 60,157 recorded malaria cases led to 213 deaths from January through September.

In either situation – more malaria or less – the real question is how do we know?  Diagnostics, monitoring, documentation and evaluation systems are not strong yet in most countries.  Until these are improved we may not recognize malaria elimination when we achieve it – or worse, falsely claim victory.

The potential demise of presumptive treatment for malaria was the topic of a ‘Controversies’ session at the 5th MIM Pan-African Malaria Conference on Monday.  One view was expressed by Ambrose Talisuna from Uganda Ministry of Health that as we move into the phases of sustained control toward elimination, there will be a greater need for parasitological diagnosis of malaria and more rational provision of ACTs.  Since the process of policy formulation to full implementation may take 2-4 years, Ambrose Talisuna thought it would not hurt to get started on efforts to update malaria treatment guidelines to emphasize a parasitological diagnostic component as a requirement for prescribing ACTs

Another perspective expressed by Mike English from Kenya Medical Research Institute was that it may be very difficult to change case management norms away from presumptive treatment until we can increase the confidence of clinicians in parasitological diagnostic methods by guaranteeing quality. Also there is concern that at least half of children in endemic areas live in high burden countries where presumptive treatment is still be a rational choice.

An interesting viewpoint came from Franco Pagnoni from TDR who said all treatment is presumptive. Even with parasitological diagnosis there are presumptions based on perceived quality of the diagnostic procedures and their interpretation.

There was a general sense that some Rapid Diagnostic Tests are clearly effective under research conditions, but have not been thoroughly tested in real life clinical conditions.  Another RDT challenge includes the general procurement and supply management difficulties facing all malaria commodities. There are cost issues too – will AMFm or a similar effort guarantee affordable malaria tests?  Another challenge of malaria treatment is the private sector, especially the informal component, and the community/home – how far will RDTs be distributed, and how can quality be maintained under such conditions?

As with all our efforts to move toward elimination we must recognize that different countries and different regions within countries are at different epidemiological stages.  We need development of flexible and appropriate case management and diagnostic guidelines. These must be disseminated in a way that builds diagnostic capacity at all levels – from the research lab to the community – with back-up to ensure 1) RDT and microscopy quality and 2) training that builds clinicians’ and treatment providers’ confidence in the tests and their own ability to use the tests correctly.

USAID’s Essential Health Services Program in Angola – known as Serviços Essenciais de Saúde (SES) – has been developing sentinel surveillance sites at four locations so far.  These have able to document current practices in malaria diagnosis and treatment.

During a regular seminar among representatives of Municipality Health Department staff from Luanda Province in Angola SES raised the issue of Rapid Diagnostic Tests and their value in preventing over-use of expensive artemisinin-based combination therapy (ACT) medicines.  We also acknowledged that there were people who trusted their clinical judgement more than RDTs. Similar skeptics were also at our meeting.

We presented data from other countries that showed how clinical diagnosis of malaria resulted in an over-estimation of malaria cases by anywhere from 25% to 75% of the time.  This may depend on the age of the patient (more accurate in children below five years of age) and in areas of higher endemicity.  But regardless of the context, there were excess cases and wasted treatments when clinical diagnosis was used compared to laboratory diagnosis.

The data so far indicate that a similar problem exists in Angola. Unofficial results show that out of all clinically suspected cases of malaria in the four sentinel clinics only 34% of children below five years of age and 23% of older people had laboratory confirmed malaria parasites.  In contrast half of the children below 5 years and two-thirds of the others who were clinically suspected of having malaria were given ACTs, i.e. much more than the numbers with laboratory confirmed malaria.

If people question laboratory tests, what also of RDTs? One of the skeptical people in the room shared that all the RDT tests in their clinic had turned out negative, and based on his clinical experience, he was sure that many of the patients really had malaria. We called attention to a recent study that found very low prevalence in urban Luanda, making it challenging to identify parasites. He was still not convinced.

Finally participants from two other municipalities came to our rescue.  They agreed that initially all their results had been negative. Then they undertook training and follow-up supervision of all staff who would use RTDs. This resulted in identification of positive cases.  Their stories also confirmed experiences in Tanzania where such training was crucial.

As more countries move closer to elimination and malaria cases become rarer, the importance of correct diagnosis and surveillance will increase.  We must ensure not only that our RDTs are of high quality (see FIND Report), but also that those who combat malaria at the front lines have trust in our diagnostic tools.

Many communities lack access to health facilities due to distance or seasonal rains.  Strategies that ensure residents of these communities get appropriate malaria treatment promptly should be a central part of any country’s national malaria plan. According to WHO the HMM strategic components include –

1. Availability of and access to effective, high-quality, prepacked antimalarial medicines at the community level.
2. Training of community-based service providers to ensure they have the necessary skills and knowledge to manage febrile illness or malaria.
3. An effective communication strategy to ensure correct early care seeking behaviour, and appropriate and effective home care of a febrile illness or malaria.
4. A good mechanism for supervision and monitoring of the community activities.

Elmardi and colleagues describe their efforts to provide home management of malaria in less accessible areas of Sudan, and not only include provision of artemisinin-based combination therapy (ACTs) at the village level but also the training of community volunteers to use rapid diagnostic tests (RDTs).

Research sponsored by WHO/TDR has shown that community volunteers have had an important impact on coverage of appropriate ACT treatment of malaria:

• 77% where there were village volunteers
• 33% through health facilities alone

The Sudan experiment in 20 villages provides some important management lessons. All but one volunteer followed treatment guidelines.  On the other hand only 14 relied on the RDT results when treating, and thus provided ACTs for other febrile conditions.

The importance of supervision to reinforce training was underscored here. Supervision is important even for regular health workers in clinics, let along volunteers in villages, but we know that many health systems do not have or utilize the necessary logistics to carry out supervision on a regular basis. The same rains that make it difficult for villagers to reach clinics may make it difficult for health workers to make supervisory visits.

The community volunteers in Sudan were not much different from health workers in clinics in believing that their judgment is better than RDTs.  This is unfortunate.  The community volunteers were also exposed to pressure from clients who were reluctant to accept that they did not have malaria when they made their complaints.

As mentioned by WHO, home management of malaria needs an effective communication strategy.  Community members have their own perceptions of malaria. Communication must be grounded in an understanding of what the community believes and expects.  Only then will local volunteers be able to convince people on the accuracy of RDTs.

Of course it would help greatly if village volunteers had medicines to treat other common ailments so clients with these complaints will not have to go away empty handed.

A discrepancy “between the perceived and actual level of transmission intensity” has been observed in the ‘Mosquito River’ area of Tanzania near Arusha by Mwanziva and colleagues. Specifically, they found …

Malaria transmission intensity by serological assessment was equivalent to < 1 infectious bites per person per year. Despite low transmission intensity, >40% of outpatients attending the clinics in 2006-2007 were diagnosed with malaria. Prospective data demonstrated a very high overdiagnosis of malaria. Microscopy was unreliable with <1% of slides regarded as malaria parasite-positive by clinic microscopists being confirmed by trained research microscopists. In addition, many ‘slide negatives’ received anti-malarial treatment. As a result, 99.6% (248/249) of the individuals who were treated with ACT were in fact free of malaria parasites.

A similar experience was found in urban Lagos, Nigeria ten years ago*:

• Blood film investigation of 916 children between the ages of 6 months and 5 years yielded a parasite prevalence rate of 0.9%.
• Night knockdown collections of mosquitoes in rooms yielded only C. quinquefasciatus and A. aegypti
• Very low densities of A. gambiae larvae were found in breeding sites (between 0.3 and 0.7)
• Community members, during focus group discussion identified malaria, in it various culturally defined forms, as a major health problem.
• Among the children examined clinically, 186 (20.3%) reported an illness, which they called “malaria” in the previous two weeks, and 180 had sought treatment for this illness.
• Data obtained from 303 shops in the area documented that a minimum of US $4,000 was spent on purchases of antimalarial drugs in the previous week.

This contrasts with a report from Médecins Sans Frontières (MSF) that “ weak distribution and health systems and a lack of qualified staff” are reasons why poor people in many malaria endemic areas do not receive appropriate treatment.

During a recent visit to Mozambique I observed that antenatal clinic staff had Rapid Diagnostic tests. Some explained that if the test was negative, they would send the client to the lab. If the lab results were negative, they would still treat to be on the sfae side.  This reinforces the conclusion by Mwanziva that “rational drug-prescribing behaviour” must be reinforced. Of course as seen in Lagos, this concern goes well past the behavior of orthodox prescribers.

This malaria treatment gap poses serious threats to both lives and resources.  The shame is that health workers and program planners bear as much of the responsibility as the patients themselves, if not more. This is a challenge may be met through development and enforcement of better treatment performance standards.

———–

*Brieger WR, Sesay HR, Adesina H, Mosanya ME, Ogunlade PB, Ayodele JO, Orisasona SA. Urban malaria treatment behaviour in the context of low levels of malaria transmission in Lagos, Nigeria. African Journal of Medicine and Medical Sciences 2002; 30(suppl): 7-15.