Plasmodium falciparum malaria remains a major public health problem in much of the world, despite decades of interventions [1]. The tragedy remains that many more malaria patients would survive if they had timely access to good quality, affordable and efficacious medicines. With the implementation of pivotal artemisinin-based combination therapy (ACT) throughout malarious Africa and attempts to make it accessible and affordable, hope of controlling malaria has been rekindled [2,3]. Of 42 African countries with P. falciparum malaria, ACT is now national policy in 40 (95%)[4].

A diverse range of important problems reduce ACT effectiveness, including inaccessibility, poor prescribing, poor adherence and poor medicine quality. There are two main categories of poor quality medicine. Counterfeit (or falsified, spurious) medicines (i.e. 'deliberately and fraudulently mislabelled with respect to identity and/or source;[5,6]) and substandard medicines (i.e. 'genuine medicines produced by manufacturers authorized....which do not meet quality specifications set for them by national standards';[6-8]). Substandard medicines frequently, and counterfeits occasionally, contain sub-therapeutic amounts of active pharmaceutical ingredients (API) and/or may show suboptimal release of API (dissolution), exposing parasites to sub-lethal concentrations of API(s) [8-10]. However, the percentage API in genuine medicines may also be reduced after manufacture if they are degraded by extremes of temperature and humidity [11].

Substandard and counterfeit anti-malarials were major problems in pre-ACT Africa. Those medicines with subtherapeutic amounts of API are likely to have contributed to the spread of resistance to previous generations of anti-malarials, such as sulphadoxine-pyrimethamine (SP) and chloroquine [10,12,13](Additional file 1). With evidence that SP and chloroquine-resistant P. falciparum entered Africa from SE Asia [14,15], the recent descriptions of artesunate resistance there [16] suggests that, in addition to major local repercussions, it is very likely to spread to Africa.

There are increasing reports of poor quality artemisinin monotherapies in Africa (Figures 1, 2, Additional file 1). However, monotherapies, even when of good quality, should be replaced by ACTs [1]. Although poor quality ACT has yet to be reported in Asia, there has been an alarming increase in reports of poor quality ACTs in Africa [6,17-23](Additional file 1, Figure 2). Poor ACT quality, along with poor prescribing and poor adherence, would provide a fertile environment for the spread of artemisinin-resistant parasites. This would destroy the renewed hope for malaria control in Africa and killing many patients who would otherwise survive. Therefore, the authors offered to analyse anti-malarial medicines of suspicious quality in sub-Saharan Africa via meetings, INTERPOL and the Counterfeit Drug Forensic Investigation Network (CODFIN [24]).

Analyses of packaging, chemistry and botany were performed blindly. The results have been reported using the MEDQUARG guidelines where possible [25]. Samples were stored at +4°C for between 1 month and 6 years before analysis. Counterfeit, substandard and degraded were defined based on the above definitions, with packaging analysis, in comparison to authentic samples, as the crucial evidence for the distinction between counterfeit and substandard/degraded. This was performed without any intellectual property considerations. When data did not allow such classification but the sample contained API% outside reference ranges, such medicines are referred to as 'poor quality'.

Seven sets of anti-malarials, of ten different types, were collected in eleven African countries 2002-2010 (Table 1, Additional file 2). The results were reported to the appropriate national Medicines Regulatory Agency (MRA) and the company as stated on the packaging.

This wide diversity of different counterfeit and substandard anti-malarials from eight sub-Saharan African countries are cause of great concern. However, this study has important limitations, especially since it involved ad hoc collection and estimates of the frequency of poor quality anti-malarials cannot be deduced from these data. Tablet dissolution was not measured. The sampling method will bias towards finding counterfeit, rather than substandard, anti-malarials. It provides early warning, worrying for public health, as would a case series of new rapidly fatal epidemic influenza strains. However, that counterfeit formulations of ACT have been found at all is extremely alarming and will increase treatment failure, death and morbidity, yield covert unprotected monotherapy, increase the frequency of anti-malarial drug resistance and produce unexpected and clinically confusing adverse events. That 32% of artesunate and all DHA and DHA-piperaquine in the West Africa and DRC collections had one analysed tablet outside %API reference range is difficult to interpret. Analysis with larger numbers of tablets/sample is needed to understand the clinical implications.

There has been a dramatic rise in reports of poor quality non-artemisinin and artemisinin containing anti-malarials in Africa, suggesting an important worsening situation and/or an emerging interest (Figure 2)[34-36]. There are no data available that allow accurate estimation of the prevalence of poor quality anti-malarials in Africa, but enough information is available to know that it is a serious problem. Public health bodies should not wait for large-scale evidence based on random surveys to decide on interventions, as these data will take years to acquire. Unless action is taken quickly, poor ACT quality and profligate use of monotherapy (whether genuine or poor quality) will contribute to the failure of ACT. Although, the correlation between artemisinin resistance and poor drug quality has not been proven, modeling strongly suggests that underdosing is an important factor in the spread of P. falciparum drug resistance [13]. Poor quality anti-malarials, usually substandards, resulting in blood concentrations between those that kill resistant and sensitive parasites and those that kill only sensitive parasites will select for drug resistance. Counterfeits may also aid and abet this process by increasing the risk of hyperparasitaemia and recrudescence and the co-circulation of substandard and counterfeit medicines may be especially prone to engender drug resistance especially where patients 'shop' around when treatments fail.

The initial misclassification of a counterfeit artemether-lumefantrine as genuine emphasizes the importance of analysing both packaging and the pharmaceutical chemical composition and the importance of access to samples of the authenticated packaging, which was found to be difficult with only 32% companies responding. In addition, we may have classified counterfeit samples, containing correct %API, as genuine because we were unable to compare all samples with genuine packaging.

The discovery of unexpected APIs in counterfeit anti-malarials have important public health implications [37]. Unexpected pyrimethamine, especially if taken repeatedly, could give rise to clinically-confusing adverse effects, such as bone marrow suppression, rash and insomnia [38]. Depending on the background level of P. falciparum antifolate resistance [39,40], they may-at least initially-alleviate some malaria symptoms but are extremely unlikely to be curative. Pyrimethamine, in combination with sulphadoxine, is still used in sub-Saharan Africa, especially for Intermittent Preventive Treatment in pregnancy (IPTp) [4]. The presence of pyrimethamine as hidden monotherapy in counterfeits will engender the further spread of P. falciparum dihydrofolate reductase mutations in Africa, increasing therapeutic failure and reducing the useful life of SP for IPTp. Covert consumption of anti-malarials will also confuse our understanding of changes through time of the frequency of clinical failure and molecular markers of chloroquine and SP resistance [39-41]. The median concentration of sildenafil, a wrong API in DHA-piperaquine, was 10.4 mg/tablet, whilst the smallest dose/tablet of Viagra is 25 mg, suggesting that covert administration may cause unexpected penile erection and, as it is not usually taken by the acutely ill, unknown complications. The consequences could be severe especially as this drug is contraindicated in those with hypotension and myocardial ischaemia. In addition, patients may be exposed to dangerous drug interactions between covert consumption and other medicines patients may take, such as sildenafil with anti-HIV medication and chloroquine and pyrimethamine with anti-epileptics [38]. Pollen analyses of the counterfeit anti-malarials were consistent with an origin in eastern Asia, but do not prove this. In 2001 Guangzhou police arrested Nigerian and Chinese men for production of counterfeit halofantrine [42]. No evidence was found, from pollen analysis, of counterfeit pharmaceutical production in Africa. However, production facilities for counterfeit anti-malarial packaging have been seized in Nigeria [43].

What should be done [6,8,10,12,44]? Multiple parallel strategies are urgently needed to improve the quality of medicines people take and to ensure that they are available and taken at the recommended doses. The enormous investment in the development, evaluation and deployment of anti-malarials is wasted if the medicines that patients actually take are, due to criminality or carelessness, sub-therapeutic. Objective data on the epidemiology of poor quality anti-malarials are needed to allow quantification and mapping of the problem [40], the relative public health importance of counterfeit and substandard 'products', determine intervention prioritization and, through following changes through time, evaluate their effectiveness. Second, WHO estimated that 30% of countries have either 'no drug regulation or a capacity that hardly functions' [45,46] and presumably many of these are economically-poor and malarious. MRAs are keystones for crucial interventions to improve medicine quality and without them most interventions are doomed. There are only three countries with WHO pre-qualified Quality Control medicine analysis laboratories in the whole of malarious Africa [47]. Investment in African MRAs and quality-assured medicine quality laboratories would facilitate countries ability to regulate medicines. There is a danger of poor quality medicines use in clinical trials, likely to bias results and therefore mandatory testing of such medicines should be carried out, preferably at one of the WHO-prequalified laboratories. Third, increasing the reach of affordable good quality ACT will reduce mortality [2,3] and undercut the counterfeiters, by reducing their profit margins. Stockouts of ACT may encourage poor quality anti-malarial distribution. Fourth, artemisinin monotherapies are still very widely available in large quantities [37,48], despite appeals to restrict their use, and their removal where patients have access to ACT is a key intervention. Fifth, much more attention needs to be paid to substandard medicines, with inspection and facilitation of good quality production. Sixth, increased cooperation between MRAs, police, customs, malaria control programmes, pharmaceutical companies and international organizations is vital in countering the trade in counterfeit medicines. Seventh, new portable and rapid techniques, based on Raman and Near-Infrared spectroscopy, have been used in the screening of medicine quality and assisted in recent seizure of imported counterfeit ACT in Nigeria [21]. Although which technique is the most accurate and appropriate remains unclear, they could potentially empower drug inspectors in the screening of pharmacy stock for poor quality medicines. Eighth, we are woefully ignorant as to how best to tackle poor medicine quality in different situations and there has been a damaging lack of public health, civil society and political will to tackle the problem, which those combating the fake Chinchona bark and quinine scandals in the 17^th-19^th centuries would have found puzzling [10]. Importantly, African heads of state and President Chirac issued the Cotonou Appeal for more action against counterfeit medicines in Africa [49].

With artemisinin resistance in Asia, authorities there have a duty to contain resistant parasites so that they do not spread to Africa. African countries may wish to lobby for more political will in Asia for containment and for financial and human capacity support of African MRAs. The International Health Regulations (IHR)[50] may be a method of facilitating change. In the IHR "disease" means an "illness or medical condition, irrespective of origin or source, that presents or could present significant harm to humans". Poor quality medicines and inappropriate monotherapies-being man-made public health hazards-fall within this definition and the IHR could be invoked to try to stop the spread of poor quality medicines. In addition, a treaty, drafted under the auspices of the WHO, to bring international agreement on interventions to reduce the frequency of both substandard and counterfeit medicines would allow coordinated action [51]. Action is needed immediately or the hopes of controlling malaria in Africa may, again, be dashed.

The description of a wide diversity of different counterfeit and substandard anti-malarials from eight sub-Saharan African countries are cause of great concern. Criminals are producing diverse harmful anti-malarial counterfeits with important public health consequences. The presence of artesunate monotherapy, substandard and/or degraded and counterfeit medicines containing sub-therapeutic amounts of unexpected anti-malarials will engender drug resistance. With the threatening spread of artemisinin resistance to Africa, much greater investment is required to ensure the quality of ACT and removal of artemisinin monotherapies. Support for MRAs is likely to be a key intervention. The International Health Regulations may need to be invoked to counter these poor quality medicines.

The authors declare that they have no competing interests except CF who owns a consultancy company, East West Pharmaceuticals, which assist Chinese companies in clinical pharmaceutical development.

AP, AET, AAA, SKO, SB, CF, JT, PF, HK, BA collected and documented samples. MDG, DCM, HN, LN, DMH, IS, GAH, KP, CWRS, KF, FMF performed the chemical/botanical analysis. PN performed packaging analysis. PN and KS performed the statistical analysis. PN, MDG, DCM, AET, PJG, FMF wrote the first draft. All authors revised and approved the final manuscript