What do we know about the economics of emergency care systems?

Essential emergency health services are not readily available in much of the world.2 6 Without a well-organised approach to emergency care service delivery and financing, the costs to individuals can lead to significant financial hardship and impoverishment. Methods and indicators to measure this are now well established. Catastrophic health spending is typically defined as out of pocket payments (OOPs) that exceed a certain percentage of household annual income minus necessities, often set between 10% and 25%.7 Impoverishing health spending occurs when OOPs push a household below the poverty line. Catastrophic health spending is now part of regional WHO monitoring frameworks and has been adopted as an indicator for assessing SDG 3.8 related to achieving universal health coverage.8 Recent research in Ecuador, for example, found that 24.4% of households had experienced catastrophic health spending for emergency care in the last year.9

Individuals needing acute care services must often borrow funds or sell assets in order to cover the necessary OOPs.10 Patients face both direct costs (user fees, medication costs, diagnostics) and indirect costs (lost wages, cost of travel). Take the example of traumatic injuries: direct medical costs per person from injuries have been estimated to average US$291, or 15% of gross domestic product per capita across a range of LMICs, and when indirect costs are added this rises to US$4085, or 97% of gross domestic product per capita.11

When people do not have access to quality emergency care services due to lack of availability (service coverage) or high cost, they delay or forgo care leading to further deterioration in their health, resulting in higher morbidity and mortality.5 12 13 This ultimately means higher costs to both individuals and society as a whole. Higher mortality and morbidity lead to lost wages, lower productivity, and higher ongoing costs related to long-term care. Together, these factors diminish human resources, economic potential and reduce the gross domestic product. These losses can be mitigated with the introduction of health coverage and cost-sharing mechanisms, as even a small reduction in the OOP financial burden increases care seeking and improves overall solvency.14

Emergencies occur regardless of the presence of well-organised ECSs; without effective design, these unavoidable emergencies yield poor outcomes, wasted resources, and high OOPs. Thoughtful reorganisation of health system processes towards the purpose of delivering quality emergency care has the potential to be highly cost-effective, as many of these changes can be accomplished with minimal new funding or resource requirements. An ECS offers the opportunity to sort and triage at a population level, allocating existing resources based on clinical acuity rather than social determinants of access to care and personally perceived need. A comprehensive review of published evidence, in PubMed, SCOPUS, EMBASE and Cochrane databases, has yielded a few key examples of cost-effectiveness evaluations on emergency care interventions in LMICs (table 1). The review was conducted using search terms for economic evaluations and emergency care, with Cochrane LMIC filters and no date of publication limitations. All of the economic evaluations identified after screening are presented here, and most would be considered highly cost-effective in all settings by WHO Choosing Interventions that are Cost-Effective (CHOICE) standards.15

Even in emergencies, OOPs have an adverse effect on care-seeking. While some argue that health facility user fees are necessary to mobilise resources and to decrease inappropriate use of the ECS for non-emergent conditions,16 such fees create a barrier to access care and falsely assume that lay people can accurately differentiate between emergent and non-emergent symptoms. Furthermore, requirements of payment prior to care may effectively prevent or delay care provision long enough to increase mortality or morbidity. Cases of imprisonment in hospitals for inability to pay have been well documented.17 Health financing focused on sustainable and equitable prepayment and pooling for emergency care, reducing OOPs, will support the goals of ECS along with promoting universal health coverage.18 Some countries have been successful in generating new streams of funding that can be earmarked for emergency care, such as gasoline taxes or the introduction of vehicle registration fees in Botswana.19 Though substantial knowledge gaps remain, our assessment of the available evidence suggests that investment in emergency care can produce economic gains and at times, cost savings.

Research challenges and future priorities in health economics and emergency care

There are several unique challenges to conducting research in LMICs that lead to gaps in understanding of the economics of emergency care delivery. There is a paucity of published economic data from settings in LMICs that provide emergency care and the rigour of many economic evaluations is difficult to assess given poor adherence to reporting guidelines. Research has demonstrated wide variance in methodology and poor ability to directly compare the results of cost-effectiveness evaluations of healthcare service delivery interventions in LMICs.20 International consensus guidelines, such as the 2013 Consolidated Health Economics Evaluation Reporting Standards,21 should be followed to allow for transparency and to facilitate well-informed resource allocation by health systems planners and policy makers. The WHO has also compiled extensive resources for conducting economic evaluations and conceptualising the role of financing in health systems.15 22

Accurate estimations of cost are particularly difficult to produce in the setting of emergency care, where the complexity of services, medications, equipment, human resources and locations leads to multiple costing perspectives. Both direct and indirect costs should be considered. Broader costs to the individual and their social support network, such as lost time and wages cannot be ignored. Given the range of interlinked and cross-cutting interventions used to provide emergency care, researchers should measure the broadest cost and effect categories possible without sacrificing methodological rigour (known in economics as the ‘societal perspective’). Improvements in costing methodology are still needed, even when restricting costing analysis to a single intervention within the emergency unit itself.23

For reasons of feasibility, the majority of cost-effectiveness studies are limited to single interventions (eg, oxygen for pneumonia) or implementation of a single function within the ECS (eg, triage protocol implementation). It can be difficult to accurately assess the incremental costs of intervention packages added to an existing platform of care delivery, so methods must be transparent to avoid misleading or misinterpreted conclusions. For example, an analysis of a single medication added to the delivery scope of an existing paramedic service should address whether all service costs or just the costs of the additional medication are considered. Attention should be given to fixed and variable costs, as well as clear definition of the time window during which costs are assessed and the time horizon over which effects are measured. Protocols that implement a package of ECSs interventions (eg, prehospital lay provider training, triage and the WHO Trauma Care Checklist) will better assess the cost-benefit implications of ECSs while accounting for economies of scale.

The economic value of emergency care cannot be determined without accurate assessment of the impact of emergency care interventions and systems. Yet, measuring patient outcomes in LMICs is severely restricted when data gathering infrastructure is inadequate. Implementation of standardised charts and centralised registries will allow for preintervention and postintervention assessment of emergency care outcomes. With better data available, more accurate estimates using standard metrics of disease burden and utility, such as the Disability-Adjusted Life Year or the Quality-Adjusted Life Year, can be produced. Expressing outcomes using these metrics is critical to allow for comparison across interventions and appropriate resource allocation. However, methodologies must continue to account for the confounding influence of factors outside of the ECS when attempting to attribute a change in health outcomes directly to an emergency care intervention.

Even with improved data collection systems, there remain conceptual and methodological challenges to assessing efficacy across such a broad range of care interventions. Emergency presentations are largely characterised by chief complaint (eg, chest pain), yet many interventions target a specific emergency diagnosis (eg, acute myocardial infarction). Furthermore, the range of settings, from urban to rural and prehospital to tertiary facility, creates challenges in identifying and capturing the correct population for outcomes (effectiveness) assessment. Measurement of time and its effects on outcome need to be carefully considered, as outcomes can differ radically as time to initial stabilisation and definitive care increase. At the health systems level, reliably identifying emergency presentations within a spectrum of chronic disease can be difficult.24 It can be challenging to isolate the outcomes of interventions that are often bundled or whose scope extend beyond the traditionally defined ECS. Additionally, emergency care interventions that target a particular chief complaint or diagnosis may have an unintended benefit (eg, due to improved training or resources) on another patient population that goes unnoticed and therefore unmeasured. These challenges make it difficult to estimate the economic value of ECS components and interventions.

The economics of large-scale health emergencies and disasters are an additional area where research challenges can, at times, seem insurmountable. The influx of outside funding and rapid local reallocation of existing resources creates numerous methodological challenges in accurate costing. The cost-effectiveness of interventions to enhance preparedness can be difficult to accurately assess and is similar conceptually to the added cost of 24 hours availability of emergency unit resources and capacity, regardless of daily and seasonal waxing and waning demand. Furthermore, research focus on outcomes linked directly to the disaster tends to lose sight of avertable mortality and morbidity from conditions that are not directly attributable to the disaster but are exacerbated by the reallocation of resources (eg, vaccine-preventable illnesses, obstetric emergencies). Impact measurement is also fraught with difficulty, as data collection mechanisms are often absent or lack sufficient quality. The long-term effects of large-scale health emergencies can be profound, and accurate cost-effectiveness assessment must consider an extended time-horizon that captures this.

Finally, more research is needed to understand the implication and effects of various methods for financing emergency care. Given the profound economic vulnerability of low-income households to catastrophic health expenditures, the impact of various financing mechanisms on social inequality should be further evaluated. Fragmentation of health coverage for emergencies may exacerbate these inequalities, particularly in places that utilise voluntary health insurance markets that primarily protect wealthier households or employed individuals able to access to insurance. A research agenda that furthers our understanding of financing as a means to reduce inequities aligns well with the ongoing attention towards universal health coverage.

Agenda for understanding the economic value of emergency care

Acknowledging that the academic literature regarding the economic value of emergency care in LMICs is often fragmented and focused on single interventions within the larger ECS, and that our conception of ECSs financing is largely theoretical at this time, we advocate for a research agenda that will deliberately advance our understanding of: (1) the advantages and disadvantages of various models for financing ECSs in a variety of resource contexts, (2) the best practices for accurately costing comprehensive packages of emergency care in LMICs, (3) the attributable morbidity and mortality benefits of ECSs interventions, and (4) the intangible value of emergency care toward fulfilling human rights obligations and the realisation of Universal Health Coverage. Table 2 provides examples of key questions that can be addressed by a thoughtful economic research agenda for emergency care.