Injuries are one of the leading causes of death and disability in low- and middle-income countries (LMICs), especially in sub-Saharan Africa. High-quality, effective and accessible healthcare systems are needed to reduce the burden of death and disability after injury. But, many health systems operate within budget constraints, so where should they invest scarce resources improve health systems and ensure that quality care is delivered efficiently and effectively? This represents a complex or wicked problem.
And it’s a challenge that a group of fiery female Fellows are tackling. The group comprises Kathryn Chu of the Centre for Global Surgery, Department of Surgical Sciences, Stellenbosch University; Justine Davies of the Department of Applied Health Sciences, University of Birmingham; Laura Bojke of the Centre for Health Economics at the University of York; Antouela Takou of the Nottingham Business School, Nottingham Trent University and Lucia D’Ambruoso of the Centre for Health Data Science at the University of Aberdeen.
“We are trying to understand where to invest to achieve the most equitable solution in resource-limited settings from a health-systems perspective,” said Chu. “Solutions that reduce the barriers and enhance the facilitators.”
Chu explained that injury (which includes motor-vehicle accidents, interpersonal violence and trauma caused by environmental disasters) causes four million global deaths annually.
“Sub-Saharan Africa experiences a disproportionate amount of injury,” she continued. “South Africa is the most dangerous country to drive in for the second year running and Cape Town is notorious for gang violence and has one of the highest homicide rate in the world.”
Countries also experience unique challenges in dealing with injury. “In South Africa, for example, there are parallel ambulance systems depending on whether or not you have medical aid,” said Chu. “Also ambulances going into red zones in townships may need police escorts to protect the healthcare workers which may mean they miss the golden hour (the optimal time between the injury happening to receiving healthcare) to save a life.”
Interesting but wicked
“Injury care is a microcosm of the health system,” said Justine Davies. “It’s perhaps not as complex as the whole system but it follows similar pathways to many complex diseases.”
She described health systems as a wicked problem because they are intractable, difficult to solve, hard to predict and comprise multiple interacting factors that must be understood in multidisciplinary ways.
“If you add human biases like culture, language, and even religion, these add further to the complications,” continued Davies. “There are no perfect solutions, it’s about finding the least-worst solution.”
Healthcare overall is complex with 14 000 different diseases according to the International Classification of Diseases (ICD 10), and influenced by a countries’ geography, demography, epidemiology, resources, culture, religion, and education; then add to this mix patients, families and communities, as well as the challenge of people having comorbidities.
Health services themselves are also complex comprising the physical space, the services offered, the staff, the stuff (or equipment) and the support (including things like electricity, water and IT).
“In health services we know we don’t know a lot and there may be many things we haven’t identified,” said Davies. “The Holy Grail is to improve access to quality health services and maximise outcomes with minimal resource investment.”
This is even more challenging currently when US and UK development aid cuts have created a rapidly shifting landscape for many LMICs.
For the purposes of developing a model, the group is looking at the journey the patient takes through healthcare and has divided the injury-care pathway into four steps – seeking, reaching, receiving and remaining in care – all influenced by multiple interactions leading to outcomes which may include death, full recovery or disability.
For seeking care, people need to know what care is needed; reaching care needs to be quick or survival can be diminished; in receiving there is a need to know what is needed in terms of treatments, staff, space and resources; and, remaining in care comprises referral to ongoing services.
Models start the journey
The group is using systems dynamic modelling to better understand these complex dynamics of injury care. By producing a conceptual model of the health system, they will show the patient journey from being injured to full rehabilitation, and identify factors experienced by patients moving through the system which limit optimal outcomes.
With funding from the UK National Institute for Healthcare Research (NIHR) they have collected data from 10 000 patients in four countries − Pakistan, Ghana, Rwanda and South Africa − involving 20 hospitals and eight districts in each country. These data will eventually be used to populate the conceptual model and give it its moving parts.
Chu explained that they have mapped the factors at the micro, macro and meso levels and the many interlinks.
“The micro includes patients, family, community reflecting wider social norms and cultural beliefs – for example 80% of South Africans seek care from traditional healers first. The macro includes society, culture, the economy and environmental disruptions, while the meso are the health-service derived factors. We want to see how all these factors influence each other.”
She also emphasised that many of the factors are hard to see. “For example, willingness to seek care is influenced by trust in the health system, perceptions and experiences of the health system and the ability to pay. Trust influences the odds that the patient takes the transport, reaches the health facility and remains in care. If they receive effective care they may tell the community, influencing other people’s choices. With more trust there are more possible benefits but it’s also vice versa.”
To complement the health system dynamic model approach; Laura Bojke is also developing an additional data-informed risk-based approach using econometric and extrapolation methods. This will identify the high-level interactions between factors and quantify the value of investments to reduce barriers and maximise facilitators such that patient outcomes are optimised, adverse consequences are minimised, and resources well spent.
“We are trying to identify which injury-care barriers contribute most to lifetime costs and outcomes, looking for the best pathway. We are also exploring data heterogeneity, and prioritising data and information gaps, and where to invest in the evidence,” said Davies.
“We will be able to measure the net health effect,” explained Laura Bojke. “We will generate measures of health using Quality-Adjusted Life Years (QALYs) and Disability-Adjusted Life Years (DALYs) that allow us to convert the health effect into money terms – the opportunity cost of health. This allows us to highlight where investment increases health effects.”
“But it is an area not without controversy. Costs vary and it’s hard to capture the wider social costs,” she added.
They hope that in addition to academic outputs they can influence policy with this work. To that end they held a workshop at STIAS including representatives from the Provincial Department of Health. “We hope work with the National Department as well,” said Davies, “and we have had interest from the World Health Organization which wants to develop informed guidance on injury care.”
“The fellowship has been about understanding the key factors and their relationships to influence care delivery,” said Antouela Takou. “The next step will be to turn this into quantitative models – that predict scenarios with different outcomes. We can explore with models instead of real-life which interventions have the biggest impact on outcomes. Hopefully this empowers policy makers.”
In discussion they addressed the inclusion of care outside of formal health-system structures including community care and traditional healers.
“Community and non-formal healthcare provision has been overlooked historically but there are moves to incorporate it more. Such informal care provides extra health knowledge and is often the eyes and ears of the community. Collective community responses are particularly important in disasters,” said Chu. “The model can be used for all care-seeking options. We could also map the non-formal. If you don’t record the informal you don’t see the whole picture. There is much more awareness of this.”
“Equity questions are so important,” added Lucia D’Ambruoso. “The things we can’t see well with statistical data. In fact, data can perpetuate inequalities – obscuring or distorting issues such as who gets care, who bears the costs of injury, and who is seen in data. There are many different ways of knowing, being, living and dying. It requires interpretive systems thinking.”
They also agreed that no one solution can be applied across all 54 African countries. “But you can develop global health-system models into which countries input their own data,” explained Davies. “You can test for similar or different results between countries and for many health conditions. The differences between countries may not always be as big as you may expect”.
Michelle Galloway: Part-time media officer at STIAS
Photograph: SCPS Photography