Developing regional weight-for-age growth references to optimise age-based dosing of anti-malarials

Location: Global

What we know: Age is often used as a proxy for weight in anti-malarial drug dosing. Accurate knowledge of the weight-for-age distribution of a given at risk population could help optimise treatment regimens but country/regional specific reference data are lacking; the WHO Child Growth Standards describe optimal growth, rather than actual growth in a given place/time.

What this article adds: A recent study derived regional weight-for-age growth references to help optimise age-based dosing of anti-malarials in Africa, the Americas, South-East Asia and the Western Pacific. A model was used to combine data with varying distributions from a range of sources (largely DHS). Growth curves were modelled for adolescents and adult males for whom data were most lacking. Countries were weighted by their population at risk of malaria. The authors recommend that age-based anti-malarial dosing should shift from global to region-based regimens.    

Manufacturers of anti-malarials recommend that dosing should be based on body weight. However, in many low and middle income countries, the dose is frequently based on age, which is used as a proxy for body weight because these drugs are often sold over the counter or prescribed in settings without weighing facilities. The use of inadequate age-based dosing regimens is responsible for a considerable proportion of treatment failures. Substantial improvements could be made by optimising age-based dosing regimens but this idea has received little attention from manufacturers or policy-makers. Accurate knowledge of the weight-for-age distribution of the population at risk of malaria is vital for establishing the optimal dosing regimen. Previously, the optimal age-based dosing regimen and drug ratio of a fixed-dose combination of artesunate and amodiaquine anti-malarials, for use in sub-Saharan Africa, were predicted using weight-for-age data from 88,054 individuals in several African countries. A recent analysis of the efficacy of artesunate and amodiaquine using pooled data on 5,410 patients from 24 studies showed that administered doses of the fixed combination were significantly better with both weight-and age-based regimens than when non-fixed combinations were used.

Currently weight-for–age reference data are lacking for most middle and low income countries. Existing global growth standards are the WHO Child Growth Standards for children aged 0-59 months – based on the Multi Centre Growth Reference Study and the 2007 WHO growth reference for school-aged children and adolescents aged 5-19 years. Optimal growth can be assessed against these standards but they are inadequate for establishing optimal dosing of anti-malarials as they do not describe how children actually grow at a particular time and place. Since growth varies between regions, regional or country-specific reference data would enable dosing regimens to be tailored to the population affected.

Over the past two decades, a wealth of population-representative anthropometric data has become publicly available from low and middle income countries. A recent study has set out to derive regional weight-for-age growth references to help optimise age-based dosing of anti-malarials in Africa, the Americas, South-East Asia and the Western Pacific. This was done by compiling individual level weight-for-age population representative data sources in countries where malaria is endemic, principally in three areas: the WHO African Region; the WHO Region of the Americas; and the WHO South-East Asia and Western Pacific regions combined. Malaria endemic areas in WHO Eastern Mediterranean Region were initially considered but insufficient data were available.

The method involved constructing a weight-for-age database from pre-existing population-based anthropometric data obtained from household surveys and research groups. It contained data collected between 1995 and 2012 on 1,263,119 individuals older than 14 days and younger than 50 years in 64 malaria-endemic countries. Regional growth references were generated using a generalised additive model for location, scale and shape by combining data with varying distributions from a range of sources. Countries were weighted by their population at risk of malaria to enable references to be used in optimising the dosing of anti-malarials.

The authors of this work found that large differences in weight-for-age distributions existed between the three regions and between these regions and optimal growth curves developed by WHO, which indicates that age-based dosing should shift from global to region-based regimens. In all three regions, the models achieved good fits with the original data, which confirmed that the generalised additive model for location, scale and shape extension method is a robust way of establishing growth references using mixed-source data in situations where multicentre growth reference studies are not feasible. The relative homogeneity of countries within the regions justifies modelling pooled country data sets in each region.

The large majority of the data came from Demographic Health Surveys (DHS), which are representative of national or subnational populations and have well-established designs and quality assurance methods. Extensive data were available for most of the age range of interest which helped in modelling the age range for which fewer data were available (i.e. 5-14 years). Data for adolescents and adult males were missing for many countries, however, they were available across the full age spectrum for at least one country in each region. Since the two step modelling approach was designed to use data from adjacent countries and age groups, where these showed similar growth distributions, the researchers were able to model the growth curves for adolescents and adult males for all three regions. In order to increase the representativeness of the curves, data sets from before 1995 were not used. Even so, growth curves should be updated periodically (e.g. every 5-10 years) using the latest data from representative surveys.

The authors conclude that the growth charts generated for individual countries as part of the modelling process could serve as powerful public health tools to support decision-making at national level. The logical next step would be to further validate these charts for countries where data were limited. Country-specific growth references could be improved further by extrapolating female data to derive male growth curves.

The method provides a way of deriving regional growth references by collating weight-for-age data available for populations. Furthermore, the method facilitates the transition from generic, universal, age-based dosing practices to more data-driven optimised, regional regimens for anti-malarials. The method could also help monitor nutrition and optimise age-based dosing of other drugs. 

Footnotes