Fever is amongst the most common reason that people seek healthcare in the tropics (Snow et al., 2003; Crump and Kirk, 2015). Whilst the vast majority of febrile illness is self-limiting, a substantial minority of patients develop severe illness. Sepsis, defined as acute life-threatening organ dysfunction caused by a dysregulated host response to infection (Singer et al., 2016), is a severe consequence of infection which carries significant morbidity. Over 30 million cases of sepsis are estimated to occur each year with one in five of these patients dying as a direct consequence of their sepsis (Fleischmann et al., 2016). Children and neonates are disproportionately affected (Turner et al., 2013; Fleischmann-Struzek et al., 2018). The true incidence is likely to be far greater: the highest disease burden occurs in low-income countries and no population-level estimates are available for these settings.
Even in well-resourced health systems, differentiating patients with a mild illness who can be treated in the community from those that will develop severe disease and will benefit from admission is challenging. In remote settings across LMICs this challenge is exponentially greater: health workers in peripheral health facilities have limited training and few diagnostic tests to inform their management of patients with febrile illness. Strikingly little is known about fever aetiology outside urban centres in the region (Acestor et al., 2012). Consequently, patient outcomes are compromised by erratic referral practices and inappropriate antimicrobial prescribing (Phuong et al., 2006; Lubell et al., 2016).
For patients suspected to have severe illnesses, the challenge is not whether to administer antibiotics – they would universally be recommended in these circumstances – but to identify patients whom, in addition to treatment at the peripheral health facility, require referral to higher-level care. One proposed approach is to train health workers to perform systematic clinical assessments, which focus on eliciting symptoms and signs that predict poor outcomes. Whilst tools such as the World Health Organization’s Integrated Management of Childhood Illness (IMCI) aim to achieve this, adherence is poor (Lange et al, 2014): integration of several syndrome-specific algorithms requires diagnostic calculus that is impractical for limited-skill health workers (Keitel and D’Acremont, 2018).
Numerous clinical severity scoring systems are already in use to predict deterioration in patients in hospital settings (Bone et al., 2003; Singer et al., 2016). Although many require physiological variables that are not routinely measured outside the intensive care unit, simplified versions remain strongly predictive of severe outcomes (Seymour et al., 2016) including in LMIC settings (Rudd et al., 2018). Used in combination with clinical assessment, there is evidence that biomarker tests could improve this process both in high-income (Verbakel et al., 2016) and LMIC settings (Kain et el., under review).
However, importantly, these studies focused on predicting whether patients presenting to hospital were likely to develop severe outcomes and thus even patients with low severity scores and low biomarker levels benefited from, often inpatient, hospital-level care. Whether these tools can be adapted or new tools developed to identify patients in need of referral who present to remote healthcare facilities (i.e. at a more proximal, less-resourced point in the health system), is the primary research question of this study.
Identify simple clinical features and point-of-care biomarkers that are predictive of mortality in patients with febrile illnesses presenting to remote healthcare settings in LMICs.
Create a decision-support tool incorporating clinical features and point-of-care tests to assist limited-skill health workers assess and treat patients with febrile illnesses in remote healthcare settings in LMICs.
Data from previous fever studies carried out in our centre and findings in the literature will be used to shortlist clinical parameters and biomarkers that could be measured in the context of remote settings by relatively unskilled healthcare workers. An observational study will then be carried out to assess which of these parameters and their combinations are most predictive of severe outcomes. These findings will then inform an intervention study to evaluate the impact of a proposed decision-support tool for the treatment of patients with febrile illness in remote settings in Southeast Asia.
The project will be of a multi-disciplinary nature, but most suitable for candidates with previous experience in clinical trials and global health, and an interest in applying these skills in resource-limited tropical settings. The student will benefit from training in biostatistics, clinical epidemiology and health economic modelling, as well as from opportunities to engage with and evaluate microbiological laboratory and point-of care diagnostics. The project will also require involvement in multi-centre observational and clinical trials, exposing the student to a variety of research environments.
Project reference number: 1022
|Professor Nicholas PJ Day FMedSci FRCP||Tropical Medicine||Oxford University, Bangkok||THAemail@example.com|
|Professor Yoel Lubell||Tropical Medicine||Oxford University, Bangkok||THAfirstname.lastname@example.org|
There are no publications listed for this DPhil project.