Antimalarial drug resistance is forcing newly developed pharmaceuticals into widespread use at an accelerating pace. To have the greatest public health impact, new pharmaceuticals will need to be deployed effectively in sub-Saharan Africa. Achieving effective antimalarial drug deployment over the short- to medium-term will require an appreciation of how drugs are currently used in Africa and the development of innovative approaches to optimize that use. Over the long-term, fundamental changes in the way that drugs are deployed will probably be required. There are many new strategies and initiatives that, to a greater or lesser degree, will influence how drugs are used. These influences may have a positive or negative effect on reducing malaria morbidity and mortality. The concept of analyzing and monitoring programmatic effectiveness allows for a more holistic understanding of these influences and allows for more unbiased, evidence-based decision making related to drug policy and deployment.

Antimalarial drug resistance is forcing newly developed pharmaceuticals into widespread use at an accelerating pace. While other regions of the world may have greater problems with multi-drug resistant malaria, the public health threat of drug resistance is greatest in Africa, which bears the bulk of the burden of malaria. While newly developed antimalarial drugs would probably be useful in any of the malarious regions of the world, to have any substantial global public health impact, new pharmaceuticals will need to be deployed effectively in sub-Saharan Africa. In addition, the particular intensity of malaria transmission in Africa, coupled with severely constrained health delivery systems that prevail in the continent, present unique challenges to effective deployment of these new treatments.

Achieving effective antimalarial drug deployment over the short- to medium-term will require an appreciation of how drugs are currently used in the real world and development of innovative approaches to optimize that use. Over the long term, however, effectively responding to antimalarial drug resistance will come to a choice between maintaining business as usual (which would include an on-going need for rapidly developing and deploying new malaria treatments in order to keep up with, if not stay ahead of, developing drug resistance) or making fundamental changes in how drugs against malaria are used in practice, in order to sustain their useful life spans as long as possible.

Despite malaria being an illness that claims as many as a million lives each year in Africa, antimalarial drugs are used within an environment that is characterized by a great degree of nonchalance. This nonchalance is largely the result of many years of surprisingly successful use of the most common antimalarial drugs, particularly chloroquine (CQ). In many situations,practices that could best be described as misuse of drugs have become routine,and in some cases, institutionalized and promoted.

Over years of deployment, there was little incentive to improve the way that antimalarial drugs are used. Both CQ and sulfadoxine/pyrimethamine (SP)were inexpensive, making it more cost-effective to treat presumptively rather than to attempt to get microscopic confirmation (especially when this is unavailable, a common situation in Africa). As both drugs were relatively safe, treating uninfected people, including infants and pregnant women, for malaria carried minimal health risk. More importantly, because of the risk of severe or fatal malaria in these groups, the potential for placing the individuals with malaria infection at even greater risk due to delays in obtaining a definitive diagnosis was felt to far outweigh that of providing malaria treatment to the many women and children who were not infected.

As is true with many different drugs in practically all cultures, it is recognized that antimalarial drugs are often taken in incorrect or incomplete doses. For years, CQ was highly efficacious, so incomplete dosing was still likely to reduce the parasite load, if not eliminate it. CQ has an antipyretic effect and can provide some relief of symptoms even if parasites persist. Among patients with sufficiently developed acquired immunity, underdosing of CQ may still be sufficient to prevent progression to severe disease or death in many, if not most, circumstances. In fact, lack of parasite clearance post-treatment was deemed by some experts to be a desirable outcome, based on an unproven assumption that this would be necessary so as not to interfere with the acquisition and maintenance of partial immunity.

For decades, malaria therapy was also relatively easy to give. In Africa,CQ was initially given as single dose treatment (10 mg kg-1) (WHO, 1973, 1986). As resistance developed, the recommended dose increased to 25 mg kg-1 given over 3 days (although the proportion of patients actually taking all three doses was low) (WHO, 1986). As CQ resistance further intensified and spread, some countries switched to SP,going back to an easy-to-give, single-dose regimen(Bloland and Ettling,1999).

Until recently, therefore, first-line drugs for malaria were inexpensive,easy to give, safe and, at least initially, highly efficacious. While there was some concern over adherence, as demonstrated by a few studies looking at this phenomenon, there was little effort or incentive to devise strategies to improve the way that CQ or SP were deployed(McCombie, 1996; Deming et al., 1989; Slutsker et al., 1994; Ruebush et al., 1995). This indifference has changed only recently.

While most of the newer malaria treatments currently recommended do offer much-improved parasitologic efficacy over failing treatments like CQ and SP,that increased efficacy typically comes at a cost, both an increased economic cost and a cost of increasing complexity. Newer treatments tend to be much more expensive and are more difficult to administer. Furthermore, their safety is relatively unproven, especially among the highest risk groups for malaria in sub-Saharan Africa, young children and pregnant women. In some countries,these newer pharmaceuticals are about to be deployed on a relatively large scale long before the country's medical community has gained any practical experience with them. For the most part, few countries have considered how the introduction of newer treatment regimens might affect the delivery system. While the drug names and dosing schedules may change, the environment in which they are deployed is expected to remain unchanged.

This situation (comparatively high drug cost, complex regimens, uncertain safety, poor diagnosis, heightened concern over resistance) argues strongly in favor of a fundamental change in the way that antimalarial drugs are deployed and used. Such fundamental changes would require a new vision of malaria treatment policy and practice and substantial investments into health infrastructure within both the public and private sectors.

Two basic and, at times, conflicting paradigms of antimalarial drug deployment are evident in discussions of antimalarial drug policies. While actual practice falls somewhere in between the two extremes, it is enlightening to understand how the two differ and the implications of these differences.

The first paradigm opts for sensitivity in case finding over specificity and maintains that the best approach to reduce malaria morbidity and mortality is to make effective treatment widely and freely available down to the most peripheral level, the household. Essentially, anyone with even a small chance of being infected receives treatment (even in situations where as few as 5% of febrile patients are actually infected)(WHO, 1997). This approach is a fundamental part of the Global Strategy for Malaria Control and Roll Back Malaria, and is a major component of related strategies, such as the Abuja Declaration goals, and the Integrated Management of Childhood Illnesses (IMCI)program. Such an emphasis on deploying efficacious malaria treatment as widely as possible is supported by observations that, under experimental conditions,it can reduce severe malaria-related morbidity and overall mortality among young children (Pagnoni et al.,1997; Kidane and Morrow,2000).

The alternative paradigm, which favors specificity over sensitivity,maintains that a primary objective of malaria therapy should be to limit the advent and spread of drug resistance. Because drug pressure is a leading contributor to intensification of resistance, this paradigm stresses that access to treatment should be controlled sufficiently to ensure that only those with confirmed diagnosis receive treatment. This approach recognizes,even emphasizes, the existence of a very limited antimalarial armamentarium,and a slow and costly process involved in developing new antimalarial drugs.

The two paradigms are clearly at odds on a number of issues. To date, no data have been systematically collected to examine the impact that widespread,easy access to antimalarial drugs (such as described above) would have on drug resistance or other important health outcomes. Similarly, the desire to limit provision of antimalarial drugs to those who need them fails to account for the systematic realities that limit the availability of accurate diagnosis to only a small fraction of the African population.

Some of the programmatic implications of these approaches on how drugs should be deployed are outlined in Table 1. The way forward for Africa will most likely depend on a rational compromise between these two polar-opposite paradigms. Currently,compromise between these paradigms is occurring, but in an ad hoc, or even accidental, manner rather than through coordinated effort. Achieving durable and programmatically effective implementation of new treatment strategies will require a more purposeful and informed merging of these approaches.

Table 1.

Comparisons of paradigms for antimalarial drug deployment

FactorComponentImproved accessLimitation of drug resistance
Drug Dosage Single dose, therefore long half-life Short half-life, therefore multiple doses likely
Regimen Simple - i.e. single dose treatment ideal, but the fewer doses the better Combinations of drugs typically increase complexity
Availability Widely available at most peripheral level Available at the most peripheral level offering definitive diagnosis and directly observed therapy (DOT)
Administration By patient or patient's guardian (in home); directly observed therapy possible if single dose given at clinic/source DOT is the ideal administration method to ensure adherence to full dosing
Diagnosis  Anyone with reasonable risk of needing malaria therapy, therefore clinical is acceptable Only of confirmed cases, therefore based on lab-based diagnosis
Sources of drugs  As many as possible - private sector (including public sector;informal/itinerate drug sellers); traditional healers Only from licensed and trained personnel/pharmacies and with prescription by appropriately trained and supervised HCW
Regulatory requirements  Minimal - primarily assurance of quality of drugs being distributed within community Maximal: Assurance of quality, licensing and monitoring of drug outlets,formal training and supervision of HCW
Goals of therapy Clinical relief Yes Yes
Stop progression to severe disease Proven Assumed
Prevent death Proven Assumed
Parasitologic cure Ideal, but not necessary Necessary
Interruption of transmission Ideal, but not necessary Highly desirable
Cost per case treated  Relatively low Likely to be very high
Follow-up of cases to identify failures  Desirable, but not necessary Necessary
Use of 2nd line treatment  Ideally on identification of clinical failure - in practice, multiple attempts with 1st line Immediately on identification of parasitologic failure
Need to coordinate with vector control Reduction in transmission pressure Not an essential component of drug policy, per se Very important if not essential
Information, Education and Communication (IEC) Needs  Complex: improved recognition of symptoms, understanding drug choices, dosages and adherence Simple: go to health sector for diagnosis and treatment
Health Worker Training and Supervision needs  Simpler: no specific interventions aimed at reducing overtreatment (no change from status quoComplex: maintenance of diagnostic competency, recognition of treatment failures, many drug regimens possible
Evaluation Process/Implementation Indicators reflecting availability of drugs in communities; penetration of IEC materials/messages into community; others Indicators reflecting availability of treatment algorithms, availability of drugs and diagnostic resources at health centers
Outcome/Effectiveness Difficult: requires understanding of complex community beliefs, behaviors, and practices Simpler: indicators reflecting understanding and correct use of diagnostic and treatment algorithms by health workers
Impact Difficult: HIS systems; demographic surveillance systems; special studies; in vivo evaluations as currently used Difficult: long term tracking of changes in molecular markers, in vitro and in vivo resistance; demographic surveillance systems,special studies
FactorComponentImproved accessLimitation of drug resistance
Drug Dosage Single dose, therefore long half-life Short half-life, therefore multiple doses likely
Regimen Simple - i.e. single dose treatment ideal, but the fewer doses the better Combinations of drugs typically increase complexity
Availability Widely available at most peripheral level Available at the most peripheral level offering definitive diagnosis and directly observed therapy (DOT)
Administration By patient or patient's guardian (in home); directly observed therapy possible if single dose given at clinic/source DOT is the ideal administration method to ensure adherence to full dosing
Diagnosis  Anyone with reasonable risk of needing malaria therapy, therefore clinical is acceptable Only of confirmed cases, therefore based on lab-based diagnosis
Sources of drugs  As many as possible - private sector (including public sector;informal/itinerate drug sellers); traditional healers Only from licensed and trained personnel/pharmacies and with prescription by appropriately trained and supervised HCW
Regulatory requirements  Minimal - primarily assurance of quality of drugs being distributed within community Maximal: Assurance of quality, licensing and monitoring of drug outlets,formal training and supervision of HCW
Goals of therapy Clinical relief Yes Yes
Stop progression to severe disease Proven Assumed
Prevent death Proven Assumed
Parasitologic cure Ideal, but not necessary Necessary
Interruption of transmission Ideal, but not necessary Highly desirable
Cost per case treated  Relatively low Likely to be very high
Follow-up of cases to identify failures  Desirable, but not necessary Necessary
Use of 2nd line treatment  Ideally on identification of clinical failure - in practice, multiple attempts with 1st line Immediately on identification of parasitologic failure
Need to coordinate with vector control Reduction in transmission pressure Not an essential component of drug policy, per se Very important if not essential
Information, Education and Communication (IEC) Needs  Complex: improved recognition of symptoms, understanding drug choices, dosages and adherence Simple: go to health sector for diagnosis and treatment
Health Worker Training and Supervision needs  Simpler: no specific interventions aimed at reducing overtreatment (no change from status quoComplex: maintenance of diagnostic competency, recognition of treatment failures, many drug regimens possible
Evaluation Process/Implementation Indicators reflecting availability of drugs in communities; penetration of IEC materials/messages into community; others Indicators reflecting availability of treatment algorithms, availability of drugs and diagnostic resources at health centers
Outcome/Effectiveness Difficult: requires understanding of complex community beliefs, behaviors, and practices Simpler: indicators reflecting understanding and correct use of diagnostic and treatment algorithms by health workers
Impact Difficult: HIS systems; demographic surveillance systems; special studies; in vivo evaluations as currently used Difficult: long term tracking of changes in molecular markers, in vitro and in vivo resistance; demographic surveillance systems,special studies

Table adapted from P. B. Bloland, CDC (unpublished) and Robb et al.(2003).

HCW, health care workers.

Malaria treatment policy has been driven by a very narrow concept of what it means for a malaria drug to work'. This narrow concept appears to be a reflection of the influence of the type of training that malariologists tended to have: biology or, more specifically, parasitology. As a result, working'was traditionally equated with drug efficacy (i.e. under ideal conditions of correct dose of quality drug given under direct observation). The methods used to measure drug efficacy have evolved over the years and this evolution has changed the actual definition of drug efficacy from parasitologic clearance to clinical response to some degree of both(WHO, 1996). Contemporaneous development of laboratory-based methods, such as in vitro cultivation and sensitivity testing and identification of molecular markers linked to drug resistance, have both complemented and confused this evolution, due to the uncertain and variable correlations between these laboratory-based indicators and actual treatment outcomes.

More recently, broader concepts of what it means for a malaria drug towork' have been introduced, due in large part to the development of epidemiology and public health practice as disciplines in their own right, as well as an increased interest in malaria on the part of anthropologists,economists, and other social scientists. For example, for a drug to work', it must not only clear parasites, but must also be perceived to have worked on both an individual and cultural level. In some situations, the perception of treatment is more important than its biological effect.

With the input of these additional disciplines, not only have concepts of what it means for a drug to work changed, but also greater emphasis has been placed on understanding why drugs might not work. Greater emphasis on the influence of cultural beliefs, treatment-seeking behavior, household economics, actual behaviors and practices, among others, have all contributed to a more comprehensive concept of antimalarial treatment.

This more inclusive concept of malaria treatment, programmatic effectiveness, is the first that truly addresses the issues that are central to actually achieving reductions in malaria morbidity and mortality. Programmatic effectiveness attempts to identify and maximize the favorable outcomes at the critical junctures that occur between infection of an individual with malaria parasites and ultimate clearance of those parasites and survival of the infected individual. It is through this process of maximization that the most reasonable compromise between the two conflicting paradigms of treatment will occur and, in turn, malaria control programs can achieve the greatest impact that treatment can have on malaria and morbidity.

Two examples of the concept of programmatic effectiveness related to malaria case management have been published. In the most recent, a Piot model,previously used successfully to evaluate TB and sexually transmitted infection control programs, was developed for malaria case management(Mumba et al., 2003). According to this model there are six steps that a patient would have to achieve in order to be successfully cured of malaria: awareness of a need for treatment, motivation to seek treatment, diagnosis, initiating correct treatment, completing that treatment, and the treatment being efficacious. The second example takes a similar approach and applies it to a specific situation(Krause and Sauerborn, 2000). In this model, the authors have identified what they felt were the critical steps in correct management of malaria: (i) patient seeks care at a health facility, (ii) a complete history is taken, (iii) a complete physical examination is performed, (iv) the correct drug at the correct dose is recommended, (v) the correct drug at the correct dose is bought by patient,(vi) the complete dose is taken and (vii) the drug is efficacious. Using data collected in Burkina Faso, the authors suggest that only 3% of patients are actually managed correctly. Furthermore, both examples illustrate that optimizing a single factor while leaving the other factors alone can have minimal impact in terms of improved case management: in the Burkina Faso example, increasing the efficacy of the antimalarial drug used from 85% to 100% increased overall programmatic effectiveness by <1%. The authors correctly point out that patients could still achieve a favorable outcome even if not correctly managed (for example, receiving and taking the correct drug at the correct dose could occur even though the health care worker did not take a complete history or perform a proper physical examination), and clearly the selection of which steps are truly critical for optimal case management is open to debate. Nonetheless, it is a logical conclusion that morbidity and mortality will decline as more patients receive correct management, and that optimizing correct management will require a holistic approach rather than single-minded focus on isolated factors taken one at a time.

Programmatic effectiveness has implications for more than just patient outcomes. Many of the components of effectiveness that contribute to improved patient health will also contribute to selection for and/or intensification of drug resistance: critical components such as the quality of the drug being used and the ability and willingness of the patient to correctly take a complete treatment will also either facilitate or inhibit drug resistance.

As mentioned previously, over the short- to medium-term, newly developed pharmaceuticals will need to be deployed within the epidemiological and health care environment that exists today in Africa. This environment is changing. However, a number of recent or developing trends and initiatives will have an impact on drug deployment, as well as on the programmatic effectiveness of that deployment, both in terms of patient outcome and drug resistance. This impact may be positive, negative or mixed, depending on the activity and how well it is actually implemented.

### Lack of recognition of populations at risk in drug development and deployment

In tropical sub-Saharan Africa, the two groups at greatest risk of severe malaria morbidity and mortality are very young children and pregnant women(WHO/UNICEF, 2003). Antimalarial drugs must be both safe and effective for the treatment of malaria in these high risk groups to be of greatest use. Unfortunately, drug companies often do not conduct the appropriate research to establish safety in these groups before marketing their products. For example, one potentially important addition to the antimalarial pharmacopoeia, lumefantrine-artemether(Co-artem), is not approved for use in the very groups that are likely to need it the most, children <10 kg in weight and pregnant women, making the drug either of limited value or forcing health care providers (and patients) to bear the risk of off-label use.

In addition to use for treatment of acute illness, preventive use of antimalarial drugs is being promoted for both pregnant women and infants (see below). The safety profile of drugs needs to be even more favorable to justify their use among otherwise healthy individuals, as the risk:benefit ratio is very different from that among acutely ill patients.

### Combination therapy

Much has been written about combination therapy(White et al., 1999; White, 1999; Bloland et al., 2000; Nosten and Brasseur, 2002). There is also a review on combination therapy in this issue(Olliaro and Taylor, 2003). Combining antimalarial drugs, especially when one of the components is an artemisinin compound, offers increased efficacy and the potential for inhibition of development of resistance and reductions in overall transmission, at least in some environments. Use of drugs in combination(especially artemisinin-containing combinations or ACTs) is currently the World Health Organization's recommended strategy for coping with drug resistance globally (WHO,2001).

Unfortunately, as is true for any drug (whether in combination or not,whether coformulated or coadministered), the mere existence of a new treatment does not guarantee that the treatment will have any public health impact(Bloland et al., 2000). These therapies must not only exist, but must be affordable, accessible and acceptable to the end user. Additionally, the end-user must be able to take them in correct quantities and for correct amounts of time. They must be sufficiently safe, especially among users in the highest risk groups. Finally,they must be robust enough in terms of their ability to withstand the misuse that is likely to occur and the selective pressure that this misuse will place on the parasite.

### The decline of the public health system

In much of the region, public health infrastructure is inadequate to fully meet existing needs (Kager,2002; Moerman et al.,2003). This manifests in many ways that affect the provision and use of antimalarial agents. Relatively few health facilities have or use laboratory-based diagnostic tests for identifying patients with malaria infections, therefore the majority of febrile patients receive malaria treatment, regardless of whether or not they are actually infected (see section on diagnosis below). Staff training and motivation can be poor and many facilities operate without reliable access to medicines, electricity or clean water (Gilson et al.,1994; Isra et al.,2000).

Nonetheless, the formal public health sector provides an important service to communities, especially when home-based treatments for fever fail. Strategies to address improvements in public health sector should address the specific reasons why people might choose to avoid those facilities, such as poor quality service, leakage of drugs for private resale, informal patient charges, mismanagement of patient user fees, distance to facilities, lack of drugs and other desired services, cost and lengthy waiting times(Agyepong, 1995; Baume et al., 2000; Lindblade et al., 2000; McPake et al., 1999; Mumba et al., 2003; Ndyomugyenyi et al., 1998; Nyamongo, 2002; Tarimo et al., 2000; Williams et al., 1999).

### The growing importance of the private sector

In Africa, private sector sources of antimalarial drugs often refer not only to officially recognized businesses, such as private pharmacies or general merchandisers, but also to informal sources, such as small kiosks or even itinerant drug sellers. Regardless of the permanence or formality of the shop, the importance of private sector sources of antimalarial drugs is substantial in many African communities(Adome et al., 1998; Baume et al., 2000; Gilson et al., 1994; Ndyomugyenyi et al., 1998; Thera et al., 2000). In some studies, as much as 60% of patients (or more) seeking help for febrile illness receive medicines from the private sector(McCombie, 1996; WHO/UNICEF, 2003; S. P. Kachur, CDC, 2003, unpublished data).

The popularity of the private sector is, in part, a result of the poor state of the public sector, but also because private sector outlets tend to be more numerous, closer to home, offer rotating credit schemes, have drugs in stock, and involve less time to obtain the desired treatments(Armstrong-Schellenberg et al.,2001; Marsh and Mutemi,1997; Molyneux,2002; Ongore and Nyabola, 1996; Reynolds-Whyte and Birungi,2000). These outlets also tend to interact with the customer in a friendlier fashion than health care workers interacting with patients. Unfortunately, it has been well established that the private sector is poor at providing appropriate advice, complete doses, or even the correct drug for the problem (Djimde et al., 1998; Mwenesi, 1994; Massele et al., 1998; Oketch-Rabah et al.,1998).

Due to the importance of private sector sources of medicines, many groups are investigating ways to improve the process by which these drug sellers provide drugs (Marsh et al.,1999; Reynolds-Whyte and Birungi, 2000). In Kenya, providing shop keepers with specific training on how best to provide malaria medicines and what advice to give patients resulted in an increase in the appropriate use of over-the-counter chloroquine by at least 62% (Marsh et al.,1999). Other efforts are aimed at social marketing of licensed or franchised drug shops where provision of quality treatment advice is available and a higher degree of quality assurance can be achieved (for example, see http://www.msh.org/features/gates/kenya-release.html). It is now being recognized that the educational system could be used to target messages about antimalarials and the need for prompt and effective treatment,supplementing efforts at improving health care through the public system(Bundy et al., 2000; Geissler et al., 2001). Some novel programs are now being conducted that focus on training school children in the appropriate use of drugs (van der Geest and Geissler, 2003). Many countries are now facing the choice between continuing to support malaria treatment primarily through a failing public health system or divert those needed funds to improve access and use of malaria medicines through a largely uncontrollable private sector,or to try to do both and run the risk of doing neither well.

### Pre-packaging medicines

To improve both the private sector's ability to provide malaria drugs correctly as well as the likelihood that patients will take the correct dose and complete the full regimen, schemes have been developed to pre-package medicines for specific age ranges. In most circumstances, such pre-packaging has not only been well received by the end-users but well-designed drug packages have increased patient compliance by, on average, 20%(Yeboah-Antwi et al., 2001; Pagnoni et al., 1997). Initial findings are encouraging for the use of pre-packaged antimalarial drugs. In a recent study in Burkino Faso, mothers could recognize and treat malaria in a prompt and correct manner, given appropriate training and adequately packaged drugs (Sirima et al.,2003).

Pre-packaging has its own limitations. For logistical reasons related to drug management and supply, a small number of packages are created, covering broad age ranges (such as for patients under 1 year, 1 to 6 years, 7 to 11 years, and 12 years and above). This situation raises the possibility of over-or under-dosing at the extremes of the age range, especially among populations where malnutrition is prevalent.

### Labeling of medicines

Packaging of pharmaceuticals, malaria drugs included, is woefully inadequate given the settings in which they are used (C. Goodman, London School of Hygiene and Tropical Medicine, personal communication). When drugs are given with printed information, that information is typically written using complicated or technical language, or written in a language foreign to the end user, or of little use to a predominantly illiterate population. More often, especially for drugs provided through the private sector, no information is given at all: pills and syrups are provided to the patient wrapped in slips of paper or unmarked bottles. Due to their lower costs,medicines are often bought in bulk jars that are labeled with nothing more than Use as directed by physician' (if labeled at all), leaving both vendor and patient clueless as to how to properly dose the medicine.

Increasingly, these shortcomings are being recognized and addressed. Efforts to produce pre-packaged medicines are often accompanied by efforts to improve patient information, including graphic representations of proper use for illiterate patients and written information in a locally appropriate language. These innovations, however, add to the cost of the drug and,particularly within the private sector, may not be sufficient to ensure adequate patient education.

### Integrated management of childhood illness (IMCI)

IMCI is an algorithm that, when used properly, assists health care workers to identify and treat the most common childhood illnesses on the basis of clinical signs and symptoms. The algorithm is intended to rationalize the diagnosis and treatment of these illnesses in settings where health care workers are minimally trained and have poor or non-existent access to laboratory or radiographic diagnosis(Perkins et al., 1997). This approach is intended to be an improvement over an unstructured clinical diagnostic approach in that it encourages health care workers to spend more time and effort conducting a more thorough assessment of children.

This approach does not change malaria diagnosis much, as any child with fever or history of fever is assumed to have malaria, even in areas where as little as 5% of febrile children actually have parasites on blood smear. The result is considerable misdiagnosis and unnecessary treatment. In a health facility survey recently conducted in Tanzania, for example, a greater proportion of children with IMCI-diagnosed malaria' was found to be aparasitemic than among children with malaria' diagnosed using traditional,non-structured clinical diagnosis (62% vs. 40%, respectively) (L. Causer, CDC, 2003, unpublished data).

### Improved access and use of laboratory-based diagnostics

As mentioned previously, the prevailing method for diagnosing malaria in sub-Saharan Africa is by clinical impression, which, in turn, typically amounts to treating all fevers as malaria. The inaccuracies of this method are well known. In various studies of clinical diagnosis of malaria using IMCI,the sensitivity of clinical diagnosis (i.e. the proportion of clinically diagnosed malaria patients that actually have malaria infection) can range from 87% to 100%. The specificity of clinical diagnosis (the proportion of all patients that do not get a diagnosis of malaria who actually do not have malaria infection) can range from 0% to 8%(Perkins et al., 1997; Weber et al., 1996).

Laboratory-based diagnostic tests can improve this situation dramatically;however, access to laboratory diagnosis is rare. Even in facilities with laboratories capable of performing diagnostic tests, the results of these tests are often ignored (Barat et al.,1999).

The advent of simple-to-use, rapid-diagnostic tests (RDTs) for malaria holds the promise of more definitive diagnosis occurring even in settings lacking laboratory capacity (Moody,2002). Under controlled conditions, RDTs reportedly have very high sensitivity and specificity (ranging between 81% and 100% for both), although their sensitivity drops with very low parasite densities(Moody, 2002). Although South Africa uses RDTs for primary malaria diagnosis (and requires laboratory-based confirmation before treatment), and other countries have used RDTs during emergency situations, these tests are currently far too expensive for the majority of endemic countries in Africa to sustain (US$0.80-2.50 per test)(National Research Council,2003). Nonetheless, in some settings, microscopic diagnosis of malaria has been shown to be cost-effective, even cost-saving(Jonkman et al., 1995). ### Home-based management of malaria/'community IMCI' There is increasing interest in pushing malaria treatment closer to home,or even to within the home (Winch et al.,2002). The argument in favor of this strategy is based on the assertions that (i) malaria infections can progress to severe or fatal illness very rapidly, (ii) many children die within the home without visiting a health facility and (iii) a majority of treatment already occurs in the home with medicines either bought from shops or left over from previous clinic trips. Although there is no single accepted definition of what exactly home-based malaria therapy actually is, the general desire is to vastly improve access to efficacious medicines at the most peripheral level (in terms of availability and price) and to increase community members' knowledge about how to properly use them. A number of studies have shown that such approaches can be highly effective. In Ethiopia, for example, local mothers were trained to provide their neighbors within their village with information on how to identify likely malarial illness and to supply chloroquine packaged with pictorial instructions for appropriate use. This strategy led to a reduction of 41% in the under-five mortality rate over 2 years, compared to villages with the more typical community health workers who were not specifically supplied with malaria treatment (Kidane and Morrow,2000). ### Intermittent Preventive Treatment for pregnant women (IPT) and infants (IPTi) A relatively new approach to preventing the effects of malaria in pregnancy has been developed and is being actively promoted. In this strategy, pregnant women are given full treatment doses of an antimalarial drug regularly during pregnancy regardless of malaria infection or illness status (Intermittent Preventive Treatment or IPT). Such treatment should occur, at a minimum, once in the second and once in the third trimester of pregnancy, although in most settings, more frequent treatment is advisable(WHO, 2002). This strategy has been associated with substantial reductions in maternal anemia, fetal loss and low birth weight and, because low birth weight is a primary risk factor for infant mortality, IPT may improve child survival(WHO/UNICEF, 2003). A similar approach, Intermittent Protective Treatment for Infants (IPTi), has been proposed and is currently being evaluated extensively. In this strategy,children are given periodic treatment doses of antimalarial drugs with the aim of reducing morbidity and mortality during infancy. The only published study of this strategy suggests that, when linked to a child's routine immunization schedule, IPTi resulted in a decrease in clinical malaria and severe anemia of about 50-60% (Schellenberg et al.,2001). ### Global Fund for AIDS, Tuberculosis and Malaria (GFATM) and other financing schemes The newer malaria medicines, including artemisinin-containing combination therapies (ACTs), are substantially more expensive than the single-drug treatments in common current usage (e.g. US$0.15-0.20 for an adult dose of chloroquine or SP compared with US\$1.20-2.40 for ACTs). The ability of most African economies to sustain these increased malaria treatment costs is limited. The GFATM, which delivered its first grants in 2001, has provided substantial funds to a number of countries in sub-Saharan Africa to support malaria control activities (see www.globalfundatm.org). Some countries have applied these funds to changing their antimalarial drug policies, including the purchase of ACTs. The World Bank has indicated that its loans could be used to buy malaria medicines. In other situations,non-governmental organizations (NGOs) have offered to supply governments with ACTs.

While these mechanisms offer some hope for governments not otherwise able to afford the expensive, new medicines, there are lingering concerns over financial sustainability. Concerns over the longevity of the Global Fund have already been raised (Kapp,2002). Countries are reluctant to take on additional national debt for malaria drugs. While donations via NGOs can be helpful in the short-term for specific situations, the long-term sustainability of those donations is doubtful.

Finally, all of these mechanisms are best suited to providing drugs via the public health sector. None of these funding initiatives have identified a way to provide free or heavily subsidized malaria treatment via the private sector, which is relied upon so extensively in sub-Saharan Africa.

As the above initiatives and trends illustrate, there remains a distinct division in paradigms for malaria case management. On the one side,technologic advances offer improved application of antimalarial drugs at the cost of diminished access, either intentionally (moving towards specific diagnosis-based treatment or restricting distribution of drugs to controlled and licensed outlets) or unintentionally (high-cost treatment regimens). On the other side is a willingness to accept and work within the realities of malaria treatment in sub-Saharan Africa in order to minimize obstacles to obtaining efficacious treatments (i.e. better ensure that people who might need malaria treatment, get treated with the recommended drug at an effective dose), but at the cost of control (home-based treatment; IMCI; increased utilization of private sector outlets for treatment).

While a few attempts at reconciling these divergent paradigms have occurred(pre-packaging; improved labeling and instructions; improved education of private sector drug vendors), far more needs to be done in order to ensure that new, highly efficacious treatments are in fact, programmatically effective and, therefore, offer the best chance of contributing positively towards reducing the burden of malaria in sub-Saharan Africa. This will be likely to require investment in large-scale studies to ascertain the likely impact of these strategies, both to illustrate improvements in health outcomes associated with wider and easier access to antimalarials as well as to better understand the implications of this wider and easier access to antimalarials might have on development of drug resistance.

It is clear, however, that Africa cannot and will not wait until everything is known before moving ahead. The risk is real that the promise of these new therapies might be squandered by deploying them in a business as usual'fashion. But by being cognizant of the implications of the realities of the environment in which antimalarial drugs are used in sub-Saharan Africa and deploying these various new treatments in a manner that maximizes programmatic effectiveness, the positive impact of these promising new therapies can be realized.

Adome, O. R., Whyte, S. R., Ortenblad, L., Ezati, E.,Nsabagasani, X., Owor, J. and Turinde, A. K. (
1998
). The community epidemiology of drug use: A case of three districts in Uganda. In
Proceedings of Workshop on People and Medicines in East Africa,16-20 November 1998
(ed. W. Geissler and L. Meinert). Copenhagen:Danish Bilharziasis Laboratory.
Agyepong, I. (
1995
).
Improving malaria control in the context of health sector reform. Technical Report prepared for the WHO/TDR
. Geneva: World Health Organisation.
Armstrong-Schellenberg, J., Abdulla, S., Nathan, R., Mukasa, O.,Marchant, T., Kikumbih, N., Mushi, A., Mponda, H., Minja, H., Mshinda, H.,Tanner, M. and Lengeler, C. (
2001
). Effect of large-scale social marketing of insecticide-treated nets on child survival in rural Tanzania.
Lancet
357
,
1241
-1247.
Barat, L., Chipipa, J., Kolczak, M. and Sukwa, T.(
1999
). Does the availability of blood slide microscopy for malaria at health centers improve the management of persons with fever in Zambia?
Am. J. Trop. Med. Hyg.
60
,
1024
-1230.
Baume, C., Helitzer-Allen, D. and Kachur, P.(
2000
). Patterns of care for childhood malaria in Zambia.
Soc. Sci. Med.
51
,
1491
-1503.
Bloland, P. B. and Ettling, M. (
1999
). Making malaria-treatment policy in the face of drug resistance.
Annu. Trop. Med. Parasitol.
93
,
5
-23.
Bloland, P. B., Ettling, M. and Meek, S.(
2000
). Combination therapy for malaria in Africa: Hype or Hope?
Bull. WHO
78
,
1378
-1388.
Bundy, D. A. P., Lwin, S., Osika, J. S., McLaughlin, J. and Pannenborg, C. O. (
2000
). What should schools do about malaria?
Parasitol. Today
16
,
181
-182.
Deming, M. S., Gayibor, A., Murphy, K., Jones, T. S. and Karsa,T. (
1989
). Home treatment of febrile children with antimalarial drugs in Togo.
Bull. WHO
67
,
695
-700.
Djimde, A., Plowe, C., Diop, S., Dikco, A., Wellems, T. and Doumbo, O. (
1998
). Use of antimalarial drugs in Mali: Policy versus reality.
Am. J. Trop. Med. Hyg.
59
,
376
-379.
Geissler, P. W., Meinert, L., Prince, R., Nokes, C.,Aagaard-Hansen, J., Jitta, J. and Ouma, J. H. (
2001
). Self-treatment by Kenyan and Uganda schoolchildren and the need for school-based education.
Health Pol. Plan.
16
,
362
-371.
Gilson, L., Alilio, M. and Heggenhougen, K.(
1994
). Community satisfaction with primary health care services:An evaluation undertaken in the Morogoro Region of Tanzania.
Soc. Sci. Med.
39
,
767
-780.
Isra, S. M., Razum, O., Ndiforchu, V. and Martiny, P.(
2000
). Coping strategies of health personnel during economic crisis: A case study from Cameroon.
Trop. Med. Int. Health
5
,
288
-292.
Jonkman, A., Chibwe, R. A., Khoromana, C. O., Liabunya, U. L.,Chaponda, M. E., Kandiero, G. E., Molyneux, M. E. and Taylor, T. E.(
1995
). Cost-saving through microscopy-based versus presumptive diagnosis of malaria in adult outpatients in Malawi.
Bull. WHO
73
,
223
-227.
Kager, P. A. (
2002
). Malaria control:constraints and opportunities.
Trop. Med. Int. Health
7
,
1042
-1046.
Kapp, C. (
2002
). Global Fund faces uncertain future as cash runs low.
Lancet
360
,
1225
.
Kidane, G. and Morrow, R. H. (
2000
). Teaching mothers to provide home treatment of malaria in Tigray, Ethiopia: a randomised trial.
Lancet
356
,
550
-555.
Krause, G. and Sauerborn, R. (
2000
). Comprehensive community effectiveness of health care. A study of malaria treatment in children and adults in rural Burkina Faso.
Ann. Trop. Paediatr.
20
,
273
-282.
Lindblade, K., O'Neill, D., Mathanga, D., Katungu, J. and Wilson, M. (
2000
). Treatment for clinical malaria is sought promptly during an epidemic in a highland region of Uganda.
Trop. Med. Int. Health
5
,
865
-875.
Marsh, V. and Mutemi, W. (
1997
).
A community educational intervention to optimize the home use of shop-bought antimalarial drugs in the management of uncomplicated childhood fevers. Technical Report prepared for the KEMRI/CRC/DVBD
. Nairobi: Kenya Medical Research Institute.
Marsh, V., Mutemi, W., Muturi, J., Haaland, A., Watkins, W.,Otieno, G. and Marsh, K. (
1999
). Changing home treatment of childhood fevers by training shop keepers in rural Kenya.
Trop. Med. Int. Health
4
,
383
-389.
Massele, A., Nsimba, S. E. D., Warsame, M. and Tomson, G.(
1998
). A survey of sources, availability and use of antimalarial drugs in households and drug stores in Kibaha, Tanzania. In
Proceedings of Workshop on People and Medicines in East Africa,16-20 November 1998
(ed. W. Geissler and L. Meinert). Copenhagen:Danish Bilharziasis Laboratory.
McCombie, S. C. (
1996
). Treatment seeking for malaria: a review of recent research.
Soc. Sci. Med.
43
,
933
-945.
McPake, B., Asiimwe, D., Mwesigye, F., Ofumbi, M., Ortenblad,L., Streefland, P. and Turinde, A. (
1999
). Informal economic activities of public health workers in Uganda: implications for quality and accessibility of care.
Soc. Sci. Med.
49
,
849
-865.
Moerman, F., Lengeler, C., Chimumbwa, J., Talisuna, A., Erhart,A., Coosemans, M. and D'Alessandro, U. (
2003
). The contribution of health-care services to a sound and sustainable malaria-control policy.
Lancet Inf. Dis.
3
,
99
-102.
Molyneux, C. S., Murira, G., Masha, J. and Snow, R. W.(
2002
). Intra-household relations and treatment decision-making for childhood illness: A Kenyan case study.
J. Biosoc. Sci.
34
,
109
-131.
Moody, A. (
2002
). Rapid diagnostic tests for malaria parasites.
Clin. Microbiol. Rev.
15
,
66
-78.
Mumba, M., Visschedijk, J., van Cleeff, M. and Hausman, B.(
2003
). A Piot model to analyse case management in malaria control programmes.
Trop. Med. Int. Health
8
,
544
-551.
Mwenesi, H. (
1994
). The role of drug delivery systems in health care: the case of self-medication.
African J. Health Sci.
1
,
43
-48.
National Research Council (
2003
).
Malaria Control During Mass Population Movements and Natural Disasters
(ed. P. B. Bloland and H. A. Williams), pp.
60
-62. Washington, DC: National Academies Press.
Ndyomugyenyi, R., Neema, S. and Magnussen, P.(
1998
). The use of formal and informal services for antenatal care and malaria treatment in rural Uganda.
Health Pol. Plan.
13
,
94
-102.
Nosten, F. and Brasseur, P. (
2002
). Combination therapy for malaria: the way forward?
Drugs
62
,
1315
-1329.
Nyamongo, I. (
2002
). Health care switching behavior of malaria patients in a Kenyan rural community.
Soc. Sci. Med.
54
,
377
-386.
Oketch-Rabah, H. A., Oduol, E., Oluka, M. A. and Nyamwaya,D. (
1998
). Use of traditional and pharmaceutical medicines in Kenya: The case of Kisumu and Rachuonyo Districts in Luo Nyanza. In
Proceedings of Workshop on People and Medicines in East Africa,16-20 November 1998
(ed. W. Geissler and L. Meinert). Copenhagen:Danish Bilharziasis Laboratory.
Olliaro, P. L. and Taylor, W. R. J. (
2003
). Antimalarial compounds: from bench to bedside.
J. Exp. Biol.
3753
-3759.
Ongore, D. and Nyabola, L. (
1998
). The role of shops and shopkeepers in malaria control. In
Proceedings of Workshop on People and Medicines in East Africa, 16-20 November 1998
(ed. W. Geissler and L. Meinert). Copenhagen: Danish Bilharziasis Laboratory.
Pagnoni, F., Convelbo, N., Tiendrebeogo, J., Cousens, S. and Esposito, F. (
1997
). A community-based programme to provide prompt and adequate treatment of presumptive malaria in children.
Trans. R. Soc. Trop. Med. Hyg.
91
,
512
-517.
Perkins, B. A., Zucker, J. R., Otieno, J., Jafari, H. S.,Paxton, L., Redd, S. C., Nahlen, B. L., Schwartz, B., Oloo, A. J., Olango, C.,Gove, S. and Campbell, C. C. (
1997
). Evaluation of an alogrithm for integrated management of childhood illness in an area of Kenya with high malaria transmission.
Bull. WHO
75
,
33
-42.
Robb, A., Sukwa, T. and Walker, O. (
2003
).
Framework for developing, implementing and updating national antimalaria treatment policy: a guide for country malaria control programmes. AFR/MAL/03.02.
Brazzaville, Republic of Congo: World Health Organization, Regional Office for Africa.
Ruebush, T., Kern, M., Campbell, C. and Oloo, A.(
1995
). Self-treatment of malaria in a rural area of western Kenya.
Bull. WHO
73
,
229
-236.
Reynolds-Whyte, S. and Birungi, H. (
2000
). The business of medicines and the politics of knowledge. In
Global Health Policy, Local Realities
(ed. L. M. Whiteford and L. Manderson), pp.
127
-148. Boulder: Lynne Rienner.
Schellenberg, D., Menendez, C., Kahigwa, E., Aponte, J., Vidal,J., Tanner, M., Mshinda, H. and Alonso, P. (
2001
). Intermittent treatment for malaria and anemia control at time of routine vaccinations in Tanzanian infants: a randomized, placebo-controlled trial.
Lancet
357
,
1471
-1477.
Sirima, B. S., Konate, A., Tiono, A. B., Convelbo, N., Cousens,S. and Pagnoni, F. (
2003
). Early treatment of childhood fevers with pre-packaged antimalarial drugs in the home reduces severe malaria morbidity in Burkino Faso.
Trop. Med. Int. Health
8
,
133
-139.
Slutsker, L., Chitsulo, L., Macheso, A. and Steketee, R. W.(
1994
). Treatment of malaria fever episodes among children in Malawi: results of a KAP survey.
Trop. Med. Parasitol.
45
,
61
-64.
Tarimo, D. S., Lwihula, G. K., Minjas, J. N. and Bygbjerg, I. C. (
2000
). Mothers' perceptions and knowledge on childhood malaria in the holoendemic Kibaha district, Tanzania: implications for malaria control and the IMCI strategy.
Trop. Med. Int. Health
5
,
179
-184.
Thera, M. A., D'Alessandro, U., Thiero, M., Ouedraogo, A.,Packou, J., Souleymane, O. A. D., Fane, M., Ade, G., Alvez, F. and Doumbo,O. (
2000
). Child malaria treatment practices among mothers in the district of Yanfolila, Sikasso region, Mali.
Trop. Med. Int. Health
5
,
876
-881.
van der Geest, S. and Geissler, P. W. (
2003
). Editorial: Should medicines be kept away from children?
Trop. Med. Int. Health
8
,
97
-99.
Weber, M. W., Mulholland, E. K., Jaffar, S., Troedsson, H.,Gove, S. and Greenwood, B. M. (
1997
). Evaluation of an algorithm for the integrated management of childhood illness in an area with seasonal malaria in the Gambia.
Bull. WHO
75
,
25
-32.
White, N. J. (
1999
). Delaying antimalarial drug resistance with combination chemotherapy.
Parassitologia
41
,
301
-308.
White, N. J., Nosten, F., Looareesuwan, S., Watkins, W. M.,Marsh, K., Snow, R. W., Kokwaro, G., Ouma, J., Hien, T. T., Molyneux, M. E. et al. (
1999
). Averting a malaria disaster.
Lancet
353
,
1965
-1967.
Williams, H., Kachur, P., Nalwamba, C., Hightower, A., Simoonga,C. and Mphande, P. (
1999
). A community perspective on the efficacy of malaria treatment options for children in Lundazi District,Zambia.
Trop. Med. Int. Health
4
,
641
-652.
Winch, P. J., Leban, K., Casazza, L., Walker, L. and Pearcy,K. (
2002
). An implementation framework for household and community integrated management of childhood illness.
Health Pol. Plan.
17
,
345
-353.
WHO (
1973
).
Chemotherapy of malaria and resistance to antimalarials. Report of a WHO scientific group. Technical Report Series #529
. Geneva: World Health Organisation.
WHO (
1986
).
WHO Expert Committee on Malaria, Eighteenth Report. Technical Report Series #735
. Geneva:World Health Organisation.
WHO (
1996
).
Assessment of therapeutic efficacy of antimalarial drugs for uncomplicated falciparum malaria in areas with intense transmission. WHO/MAL/96.1077
. Geneva and Brazzaville:World Health Organisation.
WHO (
1997
). Integrated Management of Childhood Illnesses Adaptation Guide. Part 2. C. Technical basis for adapting clinical guidelines, feeding recommendations, and local terms.
Working Draft Version 3. Division of Child Health and Development, World Health Organisation
, pp.
49
-51. Geneva: World Health Organisation.
WHO (
2001
).
Antimalarial Drug Combination Therapy. Report of a WHO Technical Consultation, 4-5 April,2001. WHO/CDS/RBM/2001.35.
Geneva: World Health Organisation.
WHO (
2002
).
Strategic framework for malaria control during pregnancy in the WHO Africa Region
. Final Draft document, Nov 1, 2002. Geneva: World Health Organisation.
WHO/UNICEF (
2003
).
The African Malaria Report, 2003
. (WHO/CDS/MAL/2003.1093) Geneva: World Health Organization.
Yeboah-Antwi, K., Gyapong, J. O., Asare, I. K., Barnish, G.,Evans, D. B. and Adjei, S. (
2001
). Impact of prepackaging antimalarial drugs on cost to patients and compliance with treatment.
Bull. WHO
79
,
394
-399.