Using evidence, expert opinion and epidemiological model to understand pathways to survival and mortality: The Pathways to Survival (PATHS) Tool

BACKGROUND: The reasons why episodes of illness can lead to fatal outcomes in affected persons in low resource settings are numerous and complex. A tool that allows policy makers to better understand those complexities could be useful to improve success of programmes that are implemented globally to reduce mortality. METHODS: We developed a “Pathways to Survival” (PATHS) tool: an epidemiological model using decision trees, available evidence and expert opinion. PATHS visualises the “architecture” of mortality in the population by following the entire population cohort over a certain period of time. It explains how initially healthy persons progress through health systems to lethal outcomes at the end of the specified time period. We developed an illustrative example based on the 136 million newborns and an estimated 907 000 deaths from newborn sepsis in the year 2008. This allowed us to develop an epidemiological model that described pathways to deaths from neonatal sepsis globally in 2010. RESULTS: The model described the “status quo’ situation in 2010 with 907 000 deaths to allow an assessment of the potential impact and feasibility of different interventions and programmes at various level of health systems in reducing this cause of mortality. A useful model should incorporate both a ‘horizontal’ and a ‘vertical’ component. The ‘horizontal’ would track the progress of all neonates globally through time, ie, their first 28 days of life, and separate them into different ‘pathways’ every time a change in their risk of dying from neonatal infection occurs because of their specific contextual circumstances. The ‘vertical’ would track their position within the health systems of their countries and separate them into different categories based on the ability of health system to intervene and reduce their risk of dying. Based on those requirements, PATHS tool was developed which is based on decision trees where different “branches” of the trees are associated with varying case-fatality rates. CONCLUSIONS: The application of the PATHS tool on the example of newborn sepsis revealed that novel diagnostic tests could save many lives, so we should continue to invest in them to improve their validity, deliverability and affordability. However, PATHS showed that investments in better diagnostics have limited impact unless they are coupled with improvements of the context. Programs for parental education improve compliance and care seeking. Promoting legislation change to empower community health workers (CHWs) to actively engage in prevention, diagnosis and care also makes a difference, as well as programs for training CHWs to use diagnostic tests and administer treatments correctly. Care-seeking behaviour can also be improved through programs of conditional cash transfers. Finally, PATHS demonstrated that improving access to primary and secondary health care for everyone is the most powerful contextual change.