Article Text

Download PDFPDF
Health economic evaluation: cost-effective strategies in humanitarian and disaster relief medicine
  1. Thomas Stewart1 and
  2. P Bird2
  1. 1Institute of Naval Medicine, Royal Navy, Gosport, UK
  2. 2Insititute of Aviation Medicine, Royal Air Force, Henlow, UK
  1. Correspondence to Surg Lt Thomas Stewart, Royal Navy, London PO12 2DL, UK; Thomas.stewart116{at}mod.gov.uk

Abstract

The health economic evaluation is a tool used in disaster relief medicine to generate a cost–benefit analysis. Like all areas of healthcare, disaster relief operations must use finite financial resources, much of which comes from charitable donations and foreign aid. Interventions can be assessed using cost-effectiveness tools and equity assessments. Through these tools, interventions that maximise benefit for a given cost are highlighted in the immediate rapid response phase where food, clean water and shelter are prioritised, often with military support. Beyond this, applications of technology and pre-response training are discussed as cost-effective investments made in anticipation of a disaster. In particular, novel technology-based approaches are being explored to deliver medical advice remotely through telemedicine and remote consulting. This strategy allows medical specialists to operate remotely without the logistical and financial challenges of forward basing at the disaster site. Interventions in disaster relief medicine are often expensive. A specific and regularly reviewed health economic assessment ensures that healthcare interventions yield a maximal impact while limiting waste and working within the budgetary constraints of a disaster medicine response. This is a paper commissioned as part of the humanitarian and disaster relief operations special issue of BMJ Military Health.

  • health economics
  • telemedicine
  • international health services
  • organisation of health services
  • public health

Data availability statement

Data sharing not applicable as no datasets generated and/or analysed for this study. No datasets were generated or analysed for this review.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key messages

  • Health economic evaluations allow the most cost-effective strategies to be identified.

  • Intervening early delivers compound savings throughout a disaster relief operation.

  • Computer modelling and artificial intelligence are cost-effective areas for investment.

  • Prepositioning of aid can be optimised using data modelling and simulation.

  • Training in internationally recognised frameworks ensures standardisation and improves responses.

  • The military is commonly involved in the initial phases, but must operate within international guidelines.

Introduction

Disaster relief medicine often requires deployment of extensive healthcare resources into an area where the health infrastructure has been overwhelmed. Rapidly reinstating a functioning medical capability is expensive, and is made more difficult by the additional costs posed by damage to logistics chains and the challenges inherent to natural disasters. Economic costs in disaster relief medicine often reach into millions of dollars. The estimated health sector damage and losses in Indonesia alone following the 2004 tsunami was $91.9 million.1 It is vital that limited financial resources are appropriately channelled to prevent waste and to maximise benefit.

A health economic assessment (HEA) is a crucial tool to assess need and direct limited resources to maximise their benefit. An HEA is used to apply economic parameters to the healthcare needs, enabling comparison of interventions. Through an HEA, areas where limited resources can best be deployed are identified. When undertaken in conjunction with a health needs assessment, a disaster relief response can be made specific and cost-effective, maximising benefits offered by the military and external agencies.

The core component of an HEA is the economic evaluation.2 This is a comparison of two or more courses of action, assessing both the cost and the extent of benefit to be gained.3 There are a number of tools available to guide this assessment, including guidance set out by the WHO CHOICE (CHOosing Interventions that are Cost-Effective) tool.4 Important to this, and more familiar in its application to non-disaster medicine, is the cost–utility assessment. This takes an intervention and equates its cost to a benefit to generate a standardised value, which can be used to compare with other interventions. An example of the standardised value is quality-adjusted life years, commonly used to compare cost-effectiveness of drug treatments.4 Another important measure in the disaster relief context is ‘equity’, a measure of fairness, usually of access to resources.3 Equity can be vertical or horizontal (Table 1). Horizontal equity and vertical equity are both crucial in a balanced and effective HEA.3

Table 1

Important definitions in health economics

Applying health economic evaluation in disaster relief medicine demonstrates that there are many interventions that offer cost-effectiveness. Many of these interventions are grouped in the early response phase of disaster relief operations (Figure 1).5 Preparedness, through sufficient training and planning, is crucial in maximising cost-effectiveness. As time passes and the simple interventions are made, costs rise significantly and benefits tend to be less tangible. Novel approaches including use of technology and multiyear funding programmes to prepare responses may offer strategies to maximise benefit from the finite resources available in disaster relief operations.6

Figure 1

Disaster management cycle showing the four phases of the management in disaster medicine with accompanying summary of cost-effective interventions for each stage. Health economic evaluations must be conducted at all stages in this cycle.

Maximising cost–benefit during initial disaster responses

The early phases of natural disasters are instantly recognisable from images of displaced people, destruction of infrastructure, trauma and loss of life. Typically, local medical care facilities, if they survive the initial incident, are rapidly overwhelmed. Availability of essential resources including clean water, shelter and basic medications may be scarce. In the response phase of the disaster management cycle (Figure 1), assistance from military and aid agencies is crucial in enabling medical responses. The initial priority is a rapid damage assessment.7 Included in this is an initial health needs assessment. Applying the principles of health economic evaluation to the findings of the initial health needs assessment allows first responder agencies to identify specific interventions as early as possible.8 Most commonly the needs in the first hours and days are shelter, fresh water and food. The rapid needs assessment conducted 15 days after Hurricane Katrina in 2005 identified that up to one-third of homes in Hancock country, Mississippi had been destroyed with many of the remainder lacking basic services including water and electricity.7 Large-scale shelters set up in stadiums and conference venues offered shelter and acted as a hub for relief efforts in these early stages. The provision of hygienic emergency living conditions and basic lifesaving supplies reduces disease spread. Only 22 cases of cholera were attributed to Hurricane Katrina compared with over 600 000 cases in Haiti, with cholera being declared endemic in the country following the 2010 earthquake.9 10 While this initially looks like a positive ratio of benefit to cost, there are issues with this approach. Widespread reports of squalid conditions and crime in these sites emerged and of those interviewed in shelters in Houston, Texas, over 50% reported going without adequate food or water, and over one-third had been threatened with violence.11 The overall cost of delivery, stock renewal and storage of prepositioned aid must also be considered. When conducting an HEA, understanding of the opportunity costs, the potential negative ramifications, of initial measures must be incorporated into the economic evaluation.

The delivery of aid remains difficult in natural disasters. It is for this reason that the military is often looked to for support in these early stages. In addition to reconnaissance as part of the initial damage and needs assessments, military assets can be used to reinstate communications and transport routes, reach cut-off communities and deliver aid (Figure 2). The military offers a highly trained and rapidly deployable asset, well equipped for humanitarian relief operations. The importance of military assets in disaster relief is demonstrated by the forward-based presence maintained by the UK government in the Caribbean which is regularly called on to support British Overseas Territories during the core hurricane season from 1 June to 30 November. The cost of maintaining ships, helicopters and other vehicles, as well as trained staff for this initial response, is prohibitively expensive for aid agencies. The military capability uses assets deployable in other roles throughout its sphere of operation to help spread this cost. Though this forward-based, equipment-heavy approach is unavoidably expensive, in the context of other military operations, HEAs demonstrate an acceptable cost for delivery of a significant early benefit when deployed in the initial days of a disaster. Without the military, aid becomes more difficult to deliver in the crucial first response phase after a disaster, potentially leading to increased hardship for affected people.

Figure 2

Summary of phases of a disaster response with the agencies best placed to offer assistance. An initial prepositioned military response lasting days allows a medium-term response from international groups including non-governmental organisations. Finally, the international community and private investors reinstate infrastructure and assist in the enduring economic recovery.

There are drawbacks to this strategy, particularly regarding concerns over political influence enacted through military aid. The Oslo guidelines were created to prevent this, setting out rules of collaboration between military and charitable aid to prevent political influence driving interventions, rather than the health economic and needs assessment.12 They state: Foreign military and civil defence assets should be requested only where there is no comparable civilian alternative and only the use of military or civil defence assets can meet a critical humanitarian need.12 In all disaster relief operations, it is important that central coordination is achieved where possible by the home nation. Military aid facilitates, rather than runs, the response. In most cases, it forms part of the HEA providing a service. Reducing political tensions or motivations in disaster relief operations is a core component of the Oslo guidelines. This ensures vertical equity is assessed, and unequal needs can be addressed with aid delivered based on need rather than the loudest voice of greatest political benefit.

Challenges after the initial response

Moving beyond the initial days of a disaster relief operation, the HEA must be reviewed and adjusted to reflect the changing needs assessment which, at this point, is often more specific to the disaster and community. In this phase, the basic amenities of shelter, food, and water become more difficult to deliver as aid resupply is required with limited resources and rising healthcare costs. Non-governmental organisations are often best placed to deliver this medium-term response. They act in support of the local governments to avoid worsening of the crisis by providing basic amenities and experienced decision-making (Figure 2). As pressure on resources increases, intelligent use of resources is important in cost-effectiveness. Analysis of flood reactions in Kenya demonstrates that proactive liberation of funds is essential. Sale of livestock before they lose weight and their value falls provides a greater fund from which to buy food, preventing reliance on food aid.13 Similar reactions can generate health benefits if employed in a timely manner. The cholera vaccination programme following Hurricane Matthew in 2016 reached 90% of people in three communities studied, preventing further outbreaks which could have devastated an already affected nation.14 Identifying similar areas in which benefit can be gained in a cost-effective way is often opportunistic and commonly relies on a community-based approach by experienced professionals. Interviews with disaster relief experts have identified 11 core themes, these can guide the HEA in this phase.15 Included in these are technology and training. These areas in particular lend themselves to achieving a positive cost–utility ratio.

The advantage of a focus on technology and training is that an initial expenditure delivers benefits across multiple separate disasters. Trained staff will often attend multiple disasters in their career and technologies effectively used in one disaster may provide a similar benefit again.

Telemedicine

Telemedicine is a relatively new but rapidly evolving method of responding to disasters, with increasing potential for delivering high benefit with minimal financial outlay. The WHO defines this as [using] information and communication technologies to exchange information for diagnosis, treatment and prevention of disease and injuries.16 In a disaster medicine context, telemedicine can provide clinical support and improve health outcomes, crossing geographical barriers to connect clinicians in different physical locations. Telemedicine can be broadly classified into two different types. ‘Asynchronous’ telemedicine involves exchanging prerecorded data at different times, such as email communications. In contrast, ‘synchronous’ telemedicine requires simultaneous information exchange, such as videoconferencing.17 In both subtypes, a variety of media may be transmitted, such as images, text, audio and video.

Many developed countries have prioritised incorporating telemedicine into their healthcare system, with significant expansion over the past year driven by COVID-19 pandemic.18 19 However, developing countries can benefit most from telemedicine due to obstacles such as lack of specialists and medical infrastructure, especially in rural areas.20 The applications in disaster medicine are also significant. After a disaster event, there is often a lack of forward-based specialists, especially in the early phase. Telemedicine provides solutions to involve disaster medicine subject matter experts early and provide affected individuals with healthcare access. This has the benefit of coordinating the medical response, potentially increasing the number of patients attended to, while limiting overhead costs. A smaller team of generalist medical professionals can be forward deployed to the area to provide triage and facilitate treatment based on assessment and advice given remotely by specialists. Access to specialists is increased without the expense and logistical challenges of deployment to the disaster epicentre.

Despite this potential, telemedicine had been underused in disaster responses.21 There have only been a handful of reports of its use. In 2014, a telemedicine hub was successfully established to coordinate triage and treatment from over 700 miles away in response to a nightclub fire in Brazil.22 Portable video-capable devices including mobile phones are almost ubiquitous among medical professionals or can be purchased at a relatively low initial cost. Establishing telemedicine networks in both developing countries and high-risk areas is therefore likely to be highly cost-effective. If these networks are developed in anticipation of disasters in high-risk areas, the infrastructure will be in place to employ these existing networks as opposed to attempting to set up networks in the post-incident chaos.

The British military in conjunction with the NHS has begun use of an app-based technology to achieve this.23 Photos and messages can be sent from remote locations to UK-based specialists for advice. Technology is secure and confidentiality is assured, a previous barrier for using other messaging applications for medical advice and guidance. This application is already being used to provide secure clinical communication to facilitate teleophthalmology during the COVID-19.23 Using this versatile technology could prove vital in the wake of a natural disaster. The obvious limitation of this technology-driven capability is the requirement for internet connection. Establishing connection via satellite technology and rapid repair of local infrastructure is often a priority in disaster relief, therefore enabling these technologies to be used.

Pre-incident training

First responders to disasters often feel unprepared in the acute phase due to limited experience, inadequate training and lack of effective prior medical planning.24 Simple training is a cost-effective way to ensure accurate health needs assessments are carried out and vital resources are not wasted. Whether for delivery of emergency medical care or through delivery of essential infrastructure or aid, training allows immediate response to be delivered efficiently and cost-effectively. Future costs can also be reduced through the reduction of undue morbidity and mortality. First responders, or those working within a high level of threat, should be in a constant state of readiness.

This training can take various forms. Disaster medicine simulations, mass-casualty scenarios, and ‘table top’ planning exercises provide methods of role-playing and planning for circumstances and responses to conventional healthcare being overwhelmed by a disaster. High-fidelity simulation, including features such as varied injuries, surging patient numbers and time constraints as well as conflicting information, can further improve the realistic nature of simulation training.25 This preparation, when put into practice, proves to be cost-effective with these refined skills providing the best likelihood of a slick and efficient early disaster response, one that does not waste valuable time or resources.

Incorporating the use of needs assessment tools and standardised frameworks for rapid cost–utility analysis ensures familiarity and experience, making these assessments more effective in real-world events. This planning forms the basis of an incident command system, an internationally recognised system providing the staff structure and guidance on organising assets required to respond to all types of major incident, including those requiring multiple agencies to liaise with each other (Figure 3).26 Potential conflicts and shortcomings can be addressed and contingency plans implemented. Early planning using common systems is extremely beneficial to create a smooth and effective response that will not compromise time and money in the initial phases of disaster. For example, data show secondary care facilities are on average at 90% capacity, so strategies to manage a sudden patient surge with finite financial and material resources need to be in place long before disaster strikes.27 Using a single common system familiar to multiple agencies allows coordinated responses across multiple nations delivered by whichever agency is best placed to lead. This drives a less wasteful response, able to maximise healthcare outputs for the given expenditure available. Use of a simple tool in this way represents cost-effectiveness in disaster medicine.

Figure 3

Summary of the incident command system with a staff structure of an incident commander with the command staff supported by the general staff. This standardised response ensures multiple agencies can operate under a common framework.

Future directions

The use of artificial intelligence may assist the response to disaster relief operations and provide cost-effectiveness in this difficult environment. Computer modelling is relatively low cost. It uses existing tools and equipment ahead of time to analyse previous data of disaster healthcare needs, enabling prediction of future needs and therefore identification of essential resources.28 This information can be used to determine how and where vital medical resources are prepackaged and prepositioned for use quickly in high-risk areas, saving vital time and money. Computer-based modelling has been used in the southern states of the USA to identify the optimum cost–utility ratio achieved when selecting how much aid to preposition ahead of hurricanes.29 Too much, and there is a risk of ‘maximum regret’ where damage from the storm risks wiping out the aid. Too little, and the cost and time to transport limits the effectiveness of the response. This algorithm and computer modelling-based approach has the potential to save years of trial-and-error learning and the inherent human cost of this.

The applications of artificial intelligence and computer learning are diverse. From designing disaster-resilient buildings to predicting flood dynamics, computer learning can refine the preparation for disaster relief. The National Hurricane Centre in the USA devised a HUrricane RISK (HURISK) programme to map the landfall locations of hurricanes using historical data (Figure 4).30 This can be used to determine where to preposition aid. This return period map for the USA, generated using HURISK, shows that southern Florida is particularly vulnerable, however northern Florida has historically been less affected. When combined with impact data and other metrics such as insurance pay-outs and damage assessment, this can be used to refine the prepositioning of aid and plan disaster relief responses to maximise speed and logistical efficiency. Ultimately this saves both lives and money.

Figure 4

Return period for major hurricanes passing within 50 nautical miles of various locations on the US coast—generated using the HUrricane RISK programme.30

Conclusion

Medical responses to disasters require vast resources in order to meet the needs of affected people. The initial phase of a disaster relief operation should happen before the disaster. Planning is a cheap and effective tool to ensure rapid responses, specific to the expected needs, can be mounted. The systems-based learning driven by computer modelling and artificial intelligence is being used to improve how and where aid is prepositioned, and training of individuals using standard methods is enabling multiple agencies from different spheres of a disaster response to understand each other and work together. This streamlines responses and reduces waste.

When disasters occur, the military has an important role to play. Often the first to respond, and well positioned to deliver a trained workforce, aid and support in restoring vital infrastructure, the military is able to deliver cost-effectiveness in the early phase. In lieu of civilian organisations for whom this early phase is often logistically and financially more challenging, the military can prevent escalation of a crisis and allow time for other aid agencies to mount a response. Within this ongoing response, telemedicine is well positioned to offer benefit. Though the potential is yet to be fully realised, the COVID-19 pandemic has accelerated medical communication technologies and many more regions globally are now well placed to use telemedicine to deliver specialist medical care in a cost-effective way in response to disasters.

Data availability statement

Data sharing not applicable as no datasets generated and/or analysed for this study. No datasets were generated or analysed for this review.

Ethics statements

Patient consent for publication

References

Footnotes

  • Twitter @drpetebird

  • Contributors TAS coauthored this submission contributing to the researching and writing of this article as well as submitting the article. PB coauthored this submission contributing to the research and writing of this article.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Map disclaimer The depiction of boundaries on the map(s) in this article does not imply the expression of any opinion whatsoever on the part of BMJ (or any member of its group) concerning the legal status of any country, territory, jurisdiction or area or of its authorities. The map(s) are provided without any warranty of any kind, either express or implied.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.