RCTs IN CHILD HEALTH IN THE DEVELOPING WORLD

http://www.ichrc.org/sites/www.ichrc.org/files/RCTs20112012.pdf

Four trials reported significant reductions in mortality (marked with *** in the booklet), among these:

§ In India the introduction of a program: Integrated Management of Maternal, Neonatal and Child Health reduced neonatal and infant mortality. In this program community health workers were trained to conduct postnatal home visits and women's group meetings, where physicians, nurses, and community health workers were trained to treat or refer sick newborns and children. Supply of drugs and supervision were strengthened.

§ In rural Pakistan application of 4% chlorhexidine to the umbilical cord reduced neonatal mortality and omphalitis

§ In Uganda a trial of zinc in the treatment of severe pneumonia showed a significant reduction in deaths in the zinc treated group. This is the first trial of zinc treatment in pneumonia with the power to show a mortality difference. The effect was especially strong in children with HIV. Two other trials this year – from India and Nepal - did not show a significant beneficial effect of zinc on resolution of pneumonia signs.

§ In Bangladesh, antenatal treatment of pregnant women from poor communities with multiple micronutrients, including iron and folic acid combined with early food supplementation decreased the risk of mortality in their children.

Other important results in 2011-12:

In South Africa, extended nevirapine during breast-feeding significantly reduced the risk of HIV infection: 1.1% (95% CI 0·3-1·8) of infants who received extended nevirapine developed HIV-1 between 6 weeks and 6 months compared with 2·4% (1·3-3·6) of infants who only received nevirapine for the first 6 weeks of life. However in a trial in Ethiopia, children who received nevirapine for 6 weeks and had prophylaxis failure - i.e. they developed HIV - had a higher risk of resistant strains of HIV.

In the Americas, post-natal treatment with zidovudine for 6 weeks plus three doses of nevirapine during the first 8 days of life, or zidovudine for 6 weeks plus nelfinavir and lamivudine for 2 weeks was more effective than zidovudine for 6 weeks at reducing parent-to-child transmission of HIV in mothers who did not receive ART during pregnancy.

In 6 African countries initiation of HIV treatment in children who had no prior exposure to nevirapine, ART with zidovudine, lamivudine, and ‘ritonavir-boosted lopinavir’ resulted in lower virological failure than zidovudine, lamivudine and nevirapine. Nevirapine resistance was a common feature of treatment failure.

In 7 African countries in a phase III trial the RTS,S/AS01 malaria vaccine provided protection against both clinical and severe malaria in African children, with vaccine efficacies of 50% for first episode of malaria, and 35% against severe malaria. Another study from 3 African countries in a phase II trial showed similar efficacy (53% and 59%) against the first episode of malaria and all malaria episodes, respectively, when children were followed up at 19 months. A third study of seroresponse in children in Mozambique showed protective anti-circumsporozoite antibodies at 42 months. The RTS,S/AS02 vaccine also induced high levels of anti-hepatitis B surface antigen antibodies.

In a meta-analysis of 7 trials in malaria endemic countries in West Africa involving 12,000 children, intermittent preventative therapy of malaria (IPTc) during the malaria season prevented approximately three quarters of all clinical malaria episodes and a similar proportion of severe malaria episodes. These effects remain present even where insecticide treated net (ITN) usage is high.

In Mali, a program for intermittent preventative treatment of malaria along with routine vaccines increased vaccine coverage. In Ghana health care delivery costs were less and coverage was the slightly higher when IPTi was delivered by village health workers, compared with when IPTi was delivered by clinic or outreach EPI nurses.

In a large study in Uganda involving over 100,000 children with suspected malaria, use of rapid diagnostic tests (RDT), compared with presumptive diagnosis, significantly reduced the prescribing of artemether-lumefantrine. However 23% of children with negative RDT were still prescribed antimalarials. Compared with microscopy, RDTs reduced waiting time and were considered more convenient for patients and health workers. In Tanzania community health workers could use RDT: no fatal or severe malaria occurred among 682 RDT negative children who were not treated with antimalarials by CHWs. This suggests that it is safe to withhold malaria treatment to RDT negative patients and that lower level health workers can make decisions based on RDT.

As has been found in studies in previous years, in a multi-country study in Africa, dihydroartemisinin-piperaquine was as effective as artemisinin-based therapy for uncomplicated P. falciparum, and resulted in a lower malaria recurrence risk.

In Lao, China, and Uganda trials of albendazole and mebendazole for the treatment of worm infestation showed that albendazole is more efficacious than mebendazole for hookworm. However single-dose albendazole had low efficacy against hookworm, and treatment daily for 3 days (in Lo and China), or 2 doses 8 hours apart (in Uganda) was better. Albendazole had lower efficacy than mebedazole against Trichuris trichiura, where 3 days of treatment (or 2 doses in the one day) was optimal for cure.

In Kenya, the combination of albendazole and di-ethyl carbamazine (DEC) was more effective than either drug alone for filariasis. This is important for mass administration programs aiming to interrupt transmission of W. bancrofti in endemic areas.

In Columbia, oral Meltifesone given for 28 days by directly-observed treatment was shown to be as effective as antimonial drugs given by intramuscular injection daily for 20 days in the treatment of cutaneous Leishmaniasis. Meltifesone is the first oral drug to be effective against visceral or cutaneous leishmaniasis, and is good news for efforts to eradicate the disease.

In a trial involving over 66,000 people in Kolkata, India, the 2-dose killed whole-cell oral cholera vaccine provided 65% protection for at least 3 years. One case of cholera was averted for every 404 people vaccinated.

In the Gambia, the 7-valent pneumococcal conjugate vaccine showed a marked herd immunity among children in neighbouring non-vaccinated villages, with no significant serotype replacement.

In Malawi, South Africa, and Kenya, rotavirus vaccine given in the first 3 months of life remained effective against severe rotavirus diarrhoea in the second year of life. Three doses of RV vaccine in the first 3 months of life provided greater second year protection than two doses.

In Papua New Guinea a single dose of oral azithromycin was as effective as a single injection of benzathine penicillin for the treatment of yaws. This may overcome the operational difficulties associated with administering an injection, raising the prospect of tackling yaws through the mass treatment of populations at risk.

For Indian children with type I diabetes, drinking 500ml of camel milk daily improved glucose tolerance and reduced insulin requirements.

In Angola, 12-hour infusions of cefotaxime resulted in a lower rate of the combined outcome of mortality and severe neurological sequelae in children with pneumococcal meningitis, than boluses of cefotaxime every 6 hours.

In Bangladesh simple guidelines and training on child TB case detection together with basic logistics support were integrated into the existing National TB Control Programme and markedly improved case funding for children with TB.

There were some important negative trials:

§ Despite strong evidence that children with vitamin D deficiency are at increased risk of pneumonia and bronchiolitis in some populations, two trials showed there was no beneficial effect of vitamin D as adjuvant therapy for severe pneumonia.

§ Despite previous positive trials, a large trial in South Africa showed no evidence that isoniazid preventative therapy improved tuberculosis-disease-free survival among HIV-

infected children or tuberculosis-infection-free survival among HIV-uninfected children who had received BCG vaccine.

It is important to understand the context in which benefit (or harm) occurs in a trial. This context may include: individual or population characteristics, comorbidities; the health care environment and health care providers; geographical factors; other interventions; the delivery mechanism for the drug, vaccine or other intervention; the disease stage and specific aetiology; economic, social and cultural characteristics of the population and individuals within it…and other unknown factors. This can be even more complex in understanding systematic reviews of randomised trials (where heterogeneity is often reported incompletely), and is one reason why there is a need for more large-scale implementation trials – not necessarily randomised - that provide insight into local context.

In the last 10 years there have been 1342 trials summarised in the various editions of this booklet. The public health benefits that have come from the huge number of trials on malaria (about 22% of all RCTs in the last decade) can be seen in the uptake of new interventions and reductions in malaria in each affected country in the world. The funding of comprehensive programs of research to “roll-back” malaria and implement the results of trials is a good example of the optimum benefit of research. While malaria rates are falling, the same reductions are not being seen in pneumonia, malnutrition or neonatal illness – and taking similar comprehensive approaches to the research agenda and to research-driven public health interventions are needed. It is striking that despite over 60 randomised trials of zinc sulphate over the last decade, most children with diarrhoea or malnutrition in developing countries still do not have access to zinc, and many not even access to oral rehydration solution – proven by many decades of RCTs.

In 2011-12 the impact of economic transition, Western morbidities and high-technology research was more evident, with clinical trials this year from India and China on issues related to non-communicable diseases, including obesity, diabetes, congenital heart disease, allergy, and modifying risk factors in childhood for adult cardiovascular disease.

More support is needed for developing public health research capacity in developing countries. This would improve the quality, scale and relevance of future trials, and improve the process of local analysis and implementation. High quality local trials need to be valued higher. At present, mechanisms of research funding and publication have a bias towards international agency supported and organised trials. Flourishing local research efforts are essential for development.

Trevor Duke

July 2012