INTRODUCTION

About 20 years ago, the international community embarked on the project entitled “Health for All by 2000” (4). This goal was considered attainable because, among many other things, it was argued that the technologies needed to carry out the major desired interventions (oral rehydration solutions, food supplements, antibiotics, vector control agents, water pumps, latrines) were known, effective, and inexpensive. Despite some early success, the “Health for All by 2000” campaign was largely a failure (5) (see Malaria and Failed Technology sidebar). For example, infant mortality rates have climbed in some of the target countries since the project's inception.

If the technologies were known, effective, and inexpensive, how could they have failed? There are many reasons: Vaccines could not be stored effectively because the cold chain technology required fuel and maintenance, oral rehydration therapy did not cure diarrhea and was then rejected by the patients, vector control agents created resistant strains and were perceived to damage the environment, water pumps broke and could not be repaired in remote areas, and latrines became disease-concentration points when they were not properly maintained (5), to name just a few.

Although all of the proposed technologies were correctly assessed to be known, effective, and inexpensive, they failed. One of the major health care technology lessons of the “Health for All by 2000” campaign is that being known, effective, and inexpensive is not a sufficient set of criteria to guarantee successful implementation of a health care technology in the developing world. The focus of this article is to review what additional obstacles exist to the successful design of health care technologies and suggest some paths for the future of biomedical engineering design for the developing world.

THE DEVELOPING WORLD

The United Nations (UN) sees the developing world through the lens of its human development indicators. Human development is “about creating an environment in which people can develop their full potential and lead productive, creative lives in accord with their needs and interests” (6). Of the five billion people living in the developing world, one billion are illiterate, one billion lack access to safe drinking water, and 2.5 billion lack access to basic sanitation (7). The life expectancy in the poorest country in the world is 38 years, as compared with 77 in the United States and 78 in much of Europe (6). In the United States, $4000 is spent per capita per year on health care and approximately $2000 per capita per year in Europe, but less than $100 per capita per year in the 35 nations with very low human development—less than $50 in most (7).

Both technical and nontechnical factors contribute to a country's development. The principal nontechnical conditions that have led to a low human development index vary. Military or one-party (person) rule can lead to low human development. Lacking a commitment to international standards in human rights leads to torture and other treatment that can lower human development. Civil war has torn apart many of the nations at the bottom of the UN human development index (7).

It is sometimes stated that there are inherent, nontechnical factors, such as cultural biases, climate, or geography, in developing world regions that contribute to their condition. These factors may play a role in some cases. However, in many developing nations, there is a thriving private sector for health care delivery. Despite the climate and other inherent conditions, the private system typically delivers excellent health care, often rivaling that available in the United States.

TECHNOLOGY IN THE DEVELOPING WORLD

Technology Is a Distraction

Before discussing technology and health care in the developing world, it is important to realize that not all observers agree that technology plays a positive role in health care in the developing world. Some argue that technology plays little role in the improvement of health care (8). For example, safe drinking water and basic sanitation may contribute more to life expectancy than health care technology (9). Those who take this position argue that where money is to be spent, investment in social services is a better investment (8, 10, 11).

There are examples where the introduction of seemingly appropriate and functioning health care technology is harmful to the developing world (12) and, for that matter, to the developed world (13). Even in cases where the technology is appropriate in the United States and Europe, that technology can be inappropriate in the developing world; for example, a new clinical laboratory instrument that helps diagnose a disease that is rarely contracted in the target country (14). The introduction of CT scanners was found to rarely change the diagnoses, and did not change the length of stay in the hospital or the outcome for stroke victims in the developing world (15). Such donations are often for the benefit of the bureaucracy (14) or the physician (16) rather than the patient.

While these are instructive examples of how health care technology may not help the developing world—and may even do harm in some cases—they do not refute the position of most biomedical engineers: Technology need not be universally helpful to be valuable. It is enough that technology can help sometimes to justify our continued efforts.

Biomedical engineers see things from a different perspective. For example, the argument for investing in social services instead of health care technology is based on the assumption that there is a limited pool of resources. However, engineering is based on the assumption that infrastructure improvements create new resources: treatments extend life (creating a larger pool of trained workers), bridges create avenues of commerce (creating new business opportunities), etc.

The inappropriate introduction of technology does occur in the developing world. However, engineers approach this problem with the assumption that the customers’ needs are primary and superior aid results from careful listening to individual doctors (14).

The Technology Landscape in the Developing World

Technology in developing countries is concentrated in the cities and in the private hospitals. The relationship between the private sector and public sector is somewhat different in the developing world. In the developing world, the private sector has many excellent hospitals and clinics serving a very small percentage of the population (perhaps 3%–4%). Users in the private sector must pay the full fee for their services. There are only a few private sector, mostly religious, hospitals serving the public. These few private hospitals in combination with the public hospitals and clinics treat the majority of the population. Users of the public system in the developing world still must, typically, pay a small co-payment, while the hospitals receive government support (17) and foreign support. In many cases, foreign support represents the majority of the health care expenditures (18) [often badly targeted (19)].

More than 95% of medical equipment in public hospitals is imported (Figure 1). There is essentially no local production of medical equipment (20), and when there is, it is controlled by multinational corporations (17). Most of the imported equipment does not meet the needs of the health care facility according to the World Health Organization. Citing an orally delivered study of donations to Columbia from 1974–1979, Pena-Mohr states that 96% of foreign-donated equipment was not working just 5 years after donation (17). In that study, of the 1289 pieces tracked, 39% never worked owing to lack of training, manuals, or accessories.

BARRIERS TO HEALTH CARE TECHNOLOGY

Every technology introduction encounters barriers. Therefore, there are some barriers that are common between the developed and the developing world. However, there are also many unique barriers to technology design and introduction for the developing world. There are also misconceptions: barriers that are assumed but do not exist.

Barriers Common to All Introductions

The training of staff to use a new piece of equipment is challenging. A reluctance to change is always present. Also, as Parsloe, an anesthesiologist with extensive developing world experience, points out, “few, if any, anesthesiologists unfamiliar with new equipment will admit their ignorance” (21). The additional barrier of having to frame and pose a question in a foreign language (as the trainer is often not from the physician's home country or region) makes it even more difficult to delicately broach the barrier raised by ignorance or a reluctance to change (Figure 2).

Nevertheless, introduction of a new health care technology must cause a change in behavior within the hospital. Therefore, it is reasonable to discuss the introduction and design of a piece of health care technology within the context of organizational behavior (1). Models have been presented analyzing the introduction of technology from the individual-level, the technology point of view and the systems level (22). Although the systems level is probably more applicable to biomedical engineering, as more sophisticated medical equipment is concentrated at the hospitals, each level is important.

A major problem with most of the research on technology diffusion is the fact that the developing world is fragmented with a lack of clear power and organization (1). This is particularly apparent when a technology is imposed from a central authority (such as The World Health Organization or the Ministry of Health). There is some recognition in the developing world that the individual, local health care provider can be the agent of change (1), a system where doctors select the technology they will use to treat patients, but this is not typical.

Barriers Unique to the Developing World

There are very little firsthand data on the unique barriers to health care technology development and introduction for the developing world. To fill this gap, Engineering World Health (EWH) has undertaken a four-year study of medical equipment in developing world hospitals. Engineering World Health is a nonprofit corporation focused on developing and delivering appropriate health care technology to the world's poorest hospitals. Largely staffed by student volunteers, EWH visits dozens of hospitals from Africa to Honduras each year. The primary question in the EWH study was, “Why does equipment fail?” These failures give us insights into how to design equipment that will not fail. Only a small fraction of these data has been previously published (23).

METHODS

Engineering students in the Duke-EWH Summer Institute are given one month of training in the identification of medical equipment failures. Basic language training for the target hospital is also given. After training, the students spend one month working in a developing world hospital. The first, approximately, two weeks are spent interviewing staff and taking inventory. The second two weeks are typically spent analyzing broken equipment. Some data are from alumni of the Institute and Institute staff traveling to hospitals outside of the usual summer program. Alumni and staff trips are typically two weeks long and focused on interviews or equipment, but not both. Students, alumni, and staff are called “participants” for the purposes of this review.

The list of hospitals visited by participants where interviews were conducted and equipment was analyzed is shown in Table 1a. Hospitals where only interviews were conducted are listed in Table 1b.

Table 1a

Hospitals visited for equipment surveys and interviews

Table 1b

Hospitals visited for interviews only

For the hospitals listed in Table 1a, broken equipment was identified by (a) the participants during their inventory process, (b) the staff during the in-depth interviews, and (c) the staff of the hospital during the final two weeks. A piece of equipment was tentatively defined as broken if it was not being used by the staff. If the cause of the disuse was uniquely due to the lack of consumables, the piece was not recorded and was returned to the staff (although note was made as to the prevalence of this occurrence). If the piece appeared to have no possibility of repair (parts were missing, excessively complicated problems arose without service manual, etc.), then the piece was abandoned and returned to where it was found. Abandoned pieces are not reported in this study. If the problem was not the lack of consumables and the piece was not abandoned, then it was considered broken and its disposition is reported here.

For every broken piece of equipment, a repair was attempted. The first step in the repair was to determine the cause of the problem. The categorization of cause was based on what action would be (or was) required to return that piece to service.

If the piece required repair relating to any aspect of the power supply, for example, batteries, fuses, power cords, surge protection, or frequency incompatibility (50 Hz versus 60 Hz), then that piece was labeled as being broken owing to “power supply” failure. If the only repair required was training of the user, for example, the manual is in a language they do not speak, the manual is missing, or the manual is unclear, then that piece was labeled as being broken owing to “user error.” All other sources of error were categorized as “other.”

Every piece of equipment was labeled as either repaired or not repaired. A piece was labeled as repaired only if that piece was returned to the patient bedside or clinical laboratory use. Testing at the bench was not considered sufficient evidence of working equipment. With this definition of repaired, we can assume that the cause of the failure was correctly identified.

Interviews were conducted with the technical staff, doctors, nurses, and administrative staff in every hospital listed in Table 1a,b. Basic information was gathered once for every hospital using a standard form. Some of the questions used in the basic interviews are shown in Table 2a. In-depth interviews were conducted with several staff in most hospitals. There was some variation in the questions for the in-depth interviews. However, the basic outline of the in-depth interview is shown in Table 2b.

Table 2a

Basic questions asked of all hospitals

Table 2b

Questions asked at in-depth interviews

All questions were posed in terms of health care technology. If the answer did not deal with healthcare technology, the participant asked the question again or left the answer blank. In many cases, the participant would ask and record answers to additional questions not listed in Table 2, but discussed below.

Basic information (Table 2a) was gained during approximately one-hour discussions and tours of the hospital. In-depth interviews lasted 30–60 min and were conducted at the interviewee's work station or work area. Most interviews were conducted in a mixture of the native language and English (the rest were conducted only in English).

RESULTS

Data on equipment were collected from 33 hospitals in 10 countries. Interviews were conducted in an additional 21 hospitals in an additional 6 countries. A total of 97 engineering students, alumni, and staff participated in this study. Several participants made multiple trips to multiple hospitals and countries.

Basic interviews were conducted in every hospital. Every hospital had at least one operating room. A few of the hospitals conducted only out-patient surgeries (where no patients stayed overnight), but most of the hospitals had between 50 and 200 beds, with a few large referral centers having 500 or more beds. Very few hospitals had engineering staff. Where staff was present, approximately one in ten was trained on medical equipment.

A total of 975 pieces of equipment were labeled as broken during the study (see Table 3). Of the broken pieces, 644 were repaired. It is reasonable to assume that every piece that was repaired was correctly categorized as to the cause of the failure. Therefore, at least 66% of the reported failures were correctly identified as to cause.

Table 3

Disposition of equipment reported as broken

The most common single cause of failure was the power supply (29.9% of repaired equipment), followed by user error (23.3% of repaired equipment). Of the remaining equipment (broken and repaired with a cause labeled as “other”), the most common problems related to a failure to complete preventative maintenance (cleaning of filters or recalibration required).

There were 331 pieces of equipment that were not repaired. For 180 of the pieces that were not repaired, the participant could not identify the cause. For 120 pieces, the participant identified the cause, but could not obtain the required spare part.

One-hundred and thirty three (133) in-depth interviews were conducted. Approximately 50% of the interviews were conducted with doctors. The remaining in-depth interviews were conducted with nurses and other staff. Very few of the staff interviews were with technical staff because very few of the listed hospitals had technical staff to interview.

DISCUSSION OF THE THREE PRINCIPAL BARRIERS

From the data gathered by EWH and the few other studies that do exist, three principal, design-related barriers to health care technology emerge: cost, spare parts, and consumables. However, as discussed below, none of them are universal.

Certainly, the cost of the technology is a major barrier. For example, a single MRI machine can cost US$10,000,000. This is approximately 10% of the entire health care budget for Sierra Leone (6). For at least this reason, no public hospital in Sierra Leone, a country of 5.6 million people, has an MRI machine (whereas many medium-sized cities in the United States boast dozens).

The second most serious problem reported in the interviews, particularly the interviews with technical staff, was the lack of spare parts. Equipment inevitably breaks and fixing it requires spare parts. Spare parts may not be available in the developing world because the parts may not be made anymore or the part may require a credit card to purchase (few people in the developing world can use a credit card). Even when the parts are made and are available, the cost may be prohibitive or the hospital may lack the expertise or tools required to execute the repair.

However, the data from the EWH study suggests that the lack of spare parts may be a perceived problem more than a true dilemma. When examining equipment, participants in the EWH study only identified 12.3% (120 out of 975) of the broken pieces of equipment as requiring a spare part that could not be found or manufactured in the developing world. Therefore, lack of spare parts may be a relatively less significant problem than often claimed.

On the other hand, staff in the hospitals may not have access to spare parts that participants obtained because the hospital perceives the expenditure to be a poor use of resources. One interview revealed that a hospital found it easier to request a new oxygen concentrator from their European sponsor then to spend the $5 required to repair the concentrator they owned.

However, expense was not universally the cause of interviewees reporting a lack of spare parts. Some equipment was repaired by the participants, even though the hospital had trained staff and owned the spare part. Many participants reported that staff frustration led to their inaction. The lack of tools and manuals and corruption in the government, perhaps extending to the public hospitals or centralized biomedical engineering facilities, can create frustration. Long-standing frustration can decay to the point where no work is attempted, even when the job could be accomplished. When interviewed it may have been easier to blame the lack of spare parts as the problem rather than implicate the true cause.

The third most commonly reported barrier was the need for consumables. Consumables are liquids or supplies required for the use of the equipment, but allowing only limited, or no, reuse. Common examples are clinical laboratory test strips, ECG electrodes, blood pressure transducers, and electrosurgery tips. The lack of consumables was a frequently cited cause of a piece of equipment being out of service and delivered to a participant in the EWH study (these pieces were not labeled as broken).

In some cases, consumables cannot be avoided. However, in the United States consumables are sometimes simply preferred over an available nonconsumable item. This preference can be the cause of ineffective donations. For example, when a developing world hospital receives a donation of an electrosurgery unit and disposable electrosurgery tips, the piece of equipment soon sits idle owing to the lack of additional tips once the original donation has been consumed. Similarly the lack of ECG pads frequently idled bedside cardiac monitors in the EWH study.

Discussion of Other Barriers

Besides cost, spare parts, and consumables, there were other barriers identified in the EWH study. One clear barrier was the lack of technical staff. Most administrators expressed a willingness to hire staff and pay for training if eligible workers could be found. In countries where the literacy rate can be 50%, eligible workers can be difficult to find. When they are found, they are often lost to “brain drain.” Brain drain refers to educated workers emigrating from their developing world nations (24). A “brain leak” also occurs, where workers are educated to a specific task in the hospital, making them eligible to be “drained.” The brain leak makes some hospitals reluctant to invest in employees’ education. Unfortunately, modern medical equipment often requires highly skilled technicians to operate and maintain the technology.

The lack of reliable power and water (25) was frequently cited as a barrier to health care technology. Much of today's equipment assumes an existing infrastructure of at least water and electricity (26). Sometimes distilled or de-ionized water must be available. In small quantities, specialized water is not an obstacle. However, continuous sources are not typically available. Electric power is rarely available on a continuous, reliable basis in developing world hospitals as well.

The lack of water and electricity is often bundled with a general lack of public infrastructure such as good roads. Oxygen is an example of how dependent the hospital is on roads. To deliver oxygen to a patient from an oxygen cylinder, the hospital must have a vehicle to get the cylinder, fuel for the vehicle, roads that can support the vehicle, a factory (reasonably close by) that produces the oxygen to fill the cylinder, and, of course, the money (27). In fact, vaccines and most technologies that are introduced depend on such an abundant public infrastructure (28). Largely because of a lack of public infrastructure, as much as 75% of developing world has no oxygen supply for their patients (27). What is available is typically from oxygen concentrators.

The use of embedded service contracts for sophisticated equipment, although not specifically a biomedical engineering phenomenon, has become a barrier incorporated into some equipment by biomedical engineers. Embedded service contracts exist where equipment is expressly designed to be serviced at regular intervals, whether it is broken or not. In some cases, the service is arbitrary: The technician must simply type in a secret code (available only to graduates of the companies’ training course) to restart the machine. These embedded service contracts are sometimes justified or bundled with required preventative maintenance. However, when donated to the developing world, these pieces are soon idle.

Kochar points out that “every technology has a cultural load” (29). The most striking hospital example is the mismatch in the economical model. Few hospitals in the developing world are strongly driven to reduce hospital stays or procedure costs the way that American and European hospitals are. Technology to reduce the amount of labor required to complete a task is a low priority. Several interviewers were surprised to learn that members of the hospital staff were given a twelve-inch knife to trim the hospital's two-acre lawn! In agricultural efforts (30), and some medical efforts, this mismatch between economic conditions has led to a altered design processes (31), where the technology's design is not considered complete until it is adopted (as opposed to being considered complete when it meets standards). For example, WHO developed a standard for oxygen concentrators that many manufacturers have met (32), but this has not solved the developing world's problems with oxygen supply (33–36).

Very few data are available on the role of medical device standards in the developing world. Equipment standards are sometimes claimed to aid the developing world (37). This view was captured by the U.S.-European Mutual Recognition Agreement of 1997 and the 1994 World Trade Organization (WTO) Agreement on Technical Barriers to Trade, which obligates all WTO member states to use international standards as the basis of their product regulations (38). The typical argument is that indigenous companies can sell their product internationally only if it meets the international standard.

Technology standards were not directly identified as a barrier in the EWH study, but the data suggest that standards probably play only a negative role. First, of 975 pieces of equipment examined in the EWH study, none of them were manufactured in a country identified as having low human development by the UN. From this we can conclude that international standards are not creating an indigenous medical equipment manufacturing capability. In fact, imposing standards probably amounts to “kicking out the ladder” (39)—preventing the emergence of local companies that might have produced low-cost devices, devices that work in the developing world, but do not meet the international standard.

For example, participants in the EWH study frequently reported defibrillators failed because of the batteries. Original replacement batteries for a defibrillator can cost US$200–$300. It might be possible for an indigenous manufacturer to produce a lower cost alternative, but it would not meet the international standard (for example, perhaps it would not operate at zero degrees Celsius, as the standard requires). However, they would be prevented from selling such an alternative by the WTO Agreement on Technical Barriers to Trade. In other words, standards may have the effect of removing the lower rungs of the health care ladder (39), preventing development, not promoting it.

Nonbarriers to Health Care Technology

The interviews also revealed two commonly held misconceptions concerning health care technology in the developing world. One misconception is that instruments must be simple. Of the 331 pieces of equipment that could not be returned to the patient bedside or clinical laboratory, none of these failed repairs were due to a failure to be able to train the users. That is, all 150 attempts to train a user on a piece of equipment were successful. In fact, the interviews often revealed knowledge about medical instruments that can be difficult to find outside of the developing world (how to use animal skin to seal a manometer or how to manufacture metal ECG electrodes from roofing materials).

In fact, the interviews suggested that simplicity can cause problems in the developing world. For example, American and European labs are becoming increasingly dependent on laboratory test kits largely because they are simple to use (40). However, these kits depend on reagents from single vendors (40). Point-of-care laboratory tests, for example, i-STAT, require manufacturer-specific strips or cartridges. In many cases, the vendors provide the equipment at low cost and profit from the sale of the reagents. None of the hospitals visited in the EWH study had a functioning (or broken) i-STAT machine, largely owing to the cartridge costs. Unable to afford these kits, the developing world has devised replacement reagents (41). In some cases, alternative, more complicated systems are more successful in the developing world than the original, e.g., PCR (42) and photospectometry (43).

A second misconception is that the cost of the medical equipment is always a barrier. In the case of the simple laboratory kits, the cost of the equipment is negligible compared to the reagents. Even when a piece of equipment is very expensive, hospitals can pool resources to purchase it. Many of the 975 pieces of broken equipment, and many pieces that were not recorded as broken, were in fact expensive X-ray and ultrasound machines. When these items fail, it is because the hospital cannot maintain the equipment, not because it cannot afford it.

POSSIBLE BLUEPRINTS FOR SUCCESSFUL DESIGN

Whereas doctors have, in many cases, learned to adapt their practice to developing world conditions (44), biomedical engineers have not developed medical equipment design practices specifically for the same conditions.

One solution to filling the health care technology gap is to donate equipment to the developing world. Besides all the barriers already mentioned, donated equipment can also be ineffective because a cable or accessory is forgotten, the equipment breaks in shipping, the hospital technical staff cannot install the equipment, or the staff is not trained to use the equipment.

An alternative to sending first world equipment into the developing world is what the Duke University-Engineering World Health Competition for Underserved and Resource Poor Economies (CUREs) is attempting (45). CUREs is a not-for-profit business plan competition that works with student teams and nonprofit corporations around the world to develop new medical devices that specifically target the unique needs of people in developing countries. CUREs is conducted like any other a business plan competition, where student teams conduct (a) need finding through on-the-ground market research in developing world hospitals, (b) nonprofit business development with a national panel of experts, and (c) prototype development through a formal design class at Duke University. If the participants complete these steps successfully, they have a defined customer, a working prototype, and a viable business plan. The winner of CUREs receives $100,000 (in cash and resources) and incubation in the Pratt School of Engineering. CUREs is currently the largest not-for-profit business plan competition in the country and the only one focused exclusively on health care technology.

The CUREs model can be considered a mirror of the competitive, entrepreneurial model that creates small-business, health care technologies in America and Europe. Although not directly involved in biomedical engineering, Benetech is similarly using traditional for-profit start-up models and incubator models and applying them to nonprofit technologies (46).

The Program for Appropriate Technology in Health (PATH) has taken a different approach based on large-scale collaboration. The PATH approach is to select problems where the public and private sectors can work in harmony. Such harmony requires that (a) the need be clearly defined, (b) there be a consensus among the public health community, and (c) there is a public-private collaboration to fund, design, field test, and promote the product. Despite these formidable obstacles, PATH has proven that its model can be effective (47) and can be a powerful tool to attack health care technology problems (5).

Project Impact may represent yet another approach to implementing health care technology in the developing world (48). Project Impact is a nonprofit in the earned-income model. Earned-income nonprofits do not use grants or donations as the principal source of revenue. Rather, they sell products instead (as nonprofit corporations, they are still prevented from distributing any net profits to individuals) (49).

Project Impact places manufacturing facilities, such as those for intraocular lenses, in the developing world and sells the product from the manufacturing line. However, as an earned-income nonprofit, Project Impact can focus on maximizing service to the developing world, instead of maximizing profits. So, for example, they can sell some lenses below their cost of manufacturing to the poorest in the target regions while selling them above their cost to the more wealthy in the same region.

THE FUTURE

A bright future for the developing world would include biomedical engineers being trained in the developing world, designing medical equipment for the developing world, and manufacturing the equipment in the developing world for the developing world. Unfortunately, the first step, training biomedical engineers in the developing world, is at a stagnation point. For example, IEEE-EMBS lists no chapters of biomedical engineers on the continent of Africa and no chapters in any nation with low human development. Indeed, there is no consistent policy toward education in the developing world by biomedical engineering societies. Even engineers trained in the United States and Europe receive very little, if any, training on design for resource-poor environments. Only a very small number of universities are offering biomedical engineering classes focused on designing for the developing world. Engineering design problems for the developing world can be envisioned for X-ray, ultrasound, electrosurgery, clinical laboratory equipment, etc. Alternative designs could avoid disposables; consider the low cost of labor; and require little power, little service (or easily delivered service), and little specialized training for servicing.

It was with an eye to the future that Drs. Mohammad Kiani, William Novick, and I met in 1999 to discuss what biomedical engineers could do. At that time there was no organization focused on harnessing the unique capabilities of biomedical engineers for the developing world. From those initial discussions was born the EWH corporation. Starting with a single visit to Nicaragua in 2001, EWH has grown to provide more than 10,000 man hours per year to developing world hospitals, most of which have no other source of engineering support. More than US$2,000,000 worth of medical equipment has been repaired by EWH volunteers. Many hundreds of biomedical engineers volunteer each year through EWH chapters in the United States, Canada, and Europe. Almost all EWH volunteers are students or recent engineering graduates. Inasmuch as these volunteers represent the future, the future is bright (Figure 3).

CONCLUSIONS

The group that could have the most impact on the state of health care in the developing world is biomedical engineers. There are billions of people who have a strong need for health care technology solutions. Engineers can design and develop tools and techniques to meet these needs. Although the financial gains may be slight, the gains in human development promise to be enormous. Engineering health care for the developing world could be one of the next great challenges for biomedical engineering as a profession.

SUMMARY POINTS

RELATED RESOURCES

Supplemental Appendices containing more information on the topics discussed in this review can be found online.