PediSEP1

The challenge

Infections are a very common cause of death in young children. The majority of infections are self-limiting, but in the early stages, it can be difficult to distinguish infections likely to progress to severe disease from those at low risk. Symptoms may provide limited clues, and diagnostic or prognostic tests may not be available. As a result, clinicians tend to take a ‘safety-first’ approach, providing high levels of care to a large proportion of sick children.

This has several consequences. For example, it leads to overuse of antibiotics, driving the development of antimicrobial resistance. It also puts children at risk of developing a hospital-acquired infection, which can then spread in communities once children are discharged. It leads to overcrowding in hospitals and to inefficient use of scarce healthcare resources.

The project

The PediSEP1 project is developing a simple prognostic tool that would enable clinicians to judge the likely severity of infection soon after hospital admission. Low-risk children could then be rapidly discharged, and clinical attention could be focused on the highest-risk cases.

Although some efforts are currently made to stream cases by severity, this is rarely done systematically. Low-risk children are often looked after by clinicians with limited paediatric training, and there are no formal guidelines on the de-escalation of treatment and discharge.

Focusing on first-level referral hospitals in Kenya, Malawi, and Uganda, the PediSEP1 project is analysing existing data from multiple studies to identify key early factors associated with severe disease. These will be integrated into a clinical algorithm that can be used in the early stages of hospitalisation to assess the risk of deterioration, enabling cases to be classified as low-, medium-, or high-risk and managed accordingly. 

The emphasis is on clinical and other measures that can be easily assessed in low-resource settings. The algorithm will be co-designed with health policymakers and clinicians, and consultations will be held with caregivers to check its acceptability.

Once developed, the tool will be evaluated in a clinical trial using a stepped-wedge design, whereby all sites gain access at different time points, with later sites serving as controls for the early adopters. This design will also enable the study of factors associated with successful implementation. The main outcome measure will be safety, to ensure that early discharge does not increase the risk of death. The project will also explore other key factors, including antibiotic use and hospital-acquired infections. Potential impacts on the health system will then be modelled to guide policymaking.

As part of this work, the project will also investigate a new approach for collecting data on hospital-acquired infections, based on systematic faecal sampling at specified times, to augment existing surveys of antibiotic use.

Impact

The PediSEP1 project could have a major impact on how children with infections are cared for in lower-level hospitals in sub-Saharan Africa. It will:

• Create a simple-to-use tool for assessing the risk of deterioration soon after admission.
• Enable the lowest-risk children to be rapidly and safely discharged, ensuring they are less likely to develop a hospital-acquired infection.
• Allow clinicians to focus on the most severely affected children.
• Ensure that antibiotics are used more appropriately. 

Most immediately, the PediSEP1 project will lead to more efficient use of scarce healthcare resources, with care focused on the sickest children. In the long term, it will reduce antibiotic use and the incidence of hospital-acquired infections, thereby helping to prevent the rise of antimicrobial resistance.