Pneumonia is the leading cause of death in children worldwide. Every year it kills an estimated 1.2 million young children, accounting for approximately 18 percent of all deaths in this age group – more than AIDS, malaria and tuberculosis combined. This means that nearly one in five children under the age of five is killed by pneumonia. Children in Uganda are particularly susceptible to the disease, largely due to the fact that 33 percent of Ugandan children under five are stunted, and 14 percent are moderately to severely underweight - leading to immune systems too weak to fight infection.
Due to the overall complexity of diagnosing pneumonia symptoms, combined with high mortality rates and growing evidence around resistance to antibiotics, there is an urgent need for improved tools used by community health workers (CHWs) and first level health facility workers to diagnose at the early stages of infection in resource poor settings. When diagnosing pneumonia CHWs are trained to count the respiratory rate of children with cough or breathing difficulties, and to then determine whether they have fast breathing relating to age-specific respiratory cut-off points. However, counting respiratory rate is difficult even for trained health workers and for CHWs with low numeracy rates the task can lead to incorrect classification of breathing rate in the sick child.
In Uganda and Zambia, where pneumonia kills an estimated 24,000 and 8,400 children under five every year respectively, Malaria Consortium has been conducting research to evaluate the use of different devices to facilitate the counting of respiratory rate. Children from two to 11 months with a respiratory rate of 50 or more per minute, and children from 12 to 59 months with a rate of 40 or more, are classified as having pneumonia and require immediate treatment with antibiotics. This means that an accurate assessment is crucial if children are to receive the appropriate treatment quickly.
According to the World Health Organization, only around 30 percent of children with pneumonia currently receive the antibiotics they need. However, one of our research studies, conducted through the COMDIS-HSD research programme consortium, found that 73 percent of children with fast breathing assessed by a CHW were correctly prescribed antibiotics, as verified by an expert.
Studies have shown that almost half of CHWs with low numeracy in Uganda were not able to accurately apply the age-specific cut-off points to the respiratory rates whilst using a stop watch and counting breaths taken per minute. However, when moving a bead for every breath – with colour-coded sections representing respiratory rate cut-off points – the same groups demonstrated vastly improved accuracy when classifying children with symptoms of pneumonia.
The use of beads also makes communicating results much easier, as the colours flagged are a visual indicator of cut-off points and therefore whether or not the child has pneumonia. CHWs have expressed support for alternative methods to counting respiratory rates with a timer. “What I have liked about this method is that I don’t have to count the beads as I move them,” explained one CHW. “Counting usually comes last and this gives me more concentration on observing the child and moving the beads.”
In spite of some success however, not all CHWs found the counting beads helpful. Many literate health workers found combining the stop watch timer with counting beads to be much more tasking. They often exhibited slow bead movements and therefore reduced accuracy in counting. Overall, the ability of literate CHWs in recording respiratory rates accurately was reduced when using colour-coded counting beads. It is unclear whether this decline in accuracy was due to an unfamiliarity of using beads to classify respiratory rates, or whether it was due to the more complicated task of including beads in the diagnosis process.
Malaria Consortium’s research into improved tools for diagnosing symptoms of pneumonia in Uganda will positively impact on illiterate CHWs by supporting the scale up of the use of counting beads by CHWs who need them, which have been found to increase correct classification of pneumonia by up to 36 percent. However, our findings suggest that more research is necessary to discover the efficacy of counting beads for the accurate diagnosis of pneumonia symptoms when performed by literate CHWs.
To learn more about Malaria Consortium’s work on pneumonia please click here.
Read the journal article on the Health Policy and Planning journal here.