Emerging infectious diseases: the Fifth stage of the epidemiologic transition?

Emerging infectious diseases: the Fifth stage of the epidemiologic transition? S. Jay 0/shansky', Bruce A. Carnesb, Richard G. Rogersc, & Len Smifhd Epidemlologlc transition theory The demographic record of anatomically modern human beings suggests that, for the vast majority of our 130 OOO-year existence, birth rates and death rates have remained at very high levels - somewhere between 30 to 50/1 OOO (Omran, 1971). Although birth rates have probably been quite stable for most of human history, death rates have been volatile - a volatility caused by the ever-present deadly force of locally-concentrated epidemic infectious and para- sitic diseases (IPDs) in our early ancestors, but punc- tuated by the episodic influence of pandemic plagues and infections that wiped out relatively large seg- ments of the population (McNeil, 1977). The histori- cal record also suggests that persistently high levels of "extrinsic" mortality (dominated by IPD deaths, but also including accidents, homicide, suicide, and pre- dation) are the primary reason that survival to sexual maturity has been denied to the majority of human beings ever born. It has only been within the last few hundred years that social and economic changes have permitted people in the developed world to avoid the causes of death that have killed the vast majority of our ancestors. Sadly, the same cannot be said for developing nations. In these countries, where the majority of the human population resides, IPDs are still a major influence on the day-to-day lives of most people. The transition from high to low vital rates has been referred to as the demographic transition - a phenomenon that has been documented and dis- cussed extensively in the scientific literature (Davis, 1963; McNamara, 1982; Teitelbaum, 1975). The mortality component of this transition has been further elaborated within what is called epidemio- logic transition theory (Omran, 1971). According to Omran, the transition from high to low death rates has been characterized by a secular transfor- mation of vital rates along a three-stage process - population changes that have led to rapid increases in life expectancy at birth, a shift in the age distribu- tion of death from the young to the old, and an accompanying shift in the underlying causes of death. The First stage of the epidemiologic transition - Omran's Age of Pestilmce and Famine - refers to a a The University of Chicago b National Opinion Research Center < University of Colorado at Boulder d Australian National University Wld hlth statist. quatt., 51 (1998) mortality state that prevailed for most of human history. During this stage, extremely high death rates were vacillating between peaks and troughs in re- sponse to Malthusian positive checks such as epi- demics, famines, and war. In the middle of the 19th century, in many now developed nations, the Sec- ond stage - Omran's Age of &ceding Pandemics - began. Epidemic peaks in mortality became less frequent, and in some cases disappeared as endemic infections were reduced. As a consequence, life ex- pectancy at birth climbed rapidly from about 35 to 50 years. The Second stage also produced two of the most significant demographic changes in human history; explosive population growth and popula- tion ageing. During the Third stage of the epide- miologic transition, Omran's Age of Degenerative and Man-made Diseases, death rates stabilized at relatively low levels (about 10/1 OOO). In addition, the major causes of death shifted from infectious and parasitic diseases to chronic disorders associated with ageing, such as cardiovascular diseases, and cancers. This stage was thought to occur at varying rates across nations because the mortality transition was being driven by socially determined factors in developed nations, and by medical technology throughout the developing world. It was also suggested that the early health and longevity benefits of this transition would be first experienced primarily by infants, chil- dren, and women of childbearing ages, subgroups of a population most susceptible to infectious and parasitic diseases. Early in the epidemiologic transition, life ex- pectancy at birth rose rapidly because declines in death rates were concentrated among the young. Later, in the 20th century, the rise in life expectancy at birth was attributed mostly to declines in death rates that occurred in those surviving to middle and older ages. A shift in the underlying causes of death, and in the ages at which these deaths occurred explains why the rapid rise in life expectancy at birth observed during this century has decelerated (Olshansky, Carnes & Cassel, 1990). Saving the lives of children adds many years to life, whereas the survivors at older ages are confronted by high over- all mortality risks even though death rates for major causes of death (such as cardiovascular diseases) are declining. In his original formulation, Omran ( 1971) never argued explicitly that infectious and parasitic dis- eases would be eliminated, nor did he discuss future trends in age-related diseases. The key concept of 207 epidemiologic transition theory was the suggestion that the causes of death that had dominated human history would be replaced by the chronic degenera- tive diseases that are associated with survival extended into older ages. In other words, age at death shifted primarily from the young to the old, as the principal causes of death shifted from infectious and parasitic diseases to chronic degenerative diseases. Extending epidemlologlc transition theory In the latter half of the 20th century, a totally unex- pected phenomenon occurred. Death rates from some chronic degenerative diseases began to decline rapidly. For example, between 1968 and 1995 (par- ticularly between 1968 and 1982), death rates from heart disease in the United States (responsible for 1 of every 4 deaths in 1968) declined by over 25 percent- a dramatic change over a remarkably brief period of time. Furthermore, these declining death rates were concentrated in people at middle and older ages, a phenomenon neither anticipated, nor discussed in epidemiologic transition theory. This new trend in mortality occurred not just in the United States, but throughout the nations of the developed world, and in more advantaged subgroups within the developed nations. In fact, this new pat- tern of mortality was so dramatically different from the Third stage that some researchers proposed that Omran's epidemiologic transition theory should be extended to account for this phenomenon ( Olshansky & Ault, 1986; Rogers & Hackenberg, 1989). One extension, proposed by Olshansky and Ault (1986), called this "new" pattern of mortality "the fourth stage of the epidemiologic transition," labeling it The Age of Delayed Degenerative Diseases. This Fourth stage was still characterized by the lin- gering presence of the major degenerative causes of death described in Omran's Third stage, but with an average age at death that was significantly de- layed. The delays in age at death from degenerative causes were attributed to the influence of medical technology, a force used by Omran to explain the mortality transition observed in developing nations. The term now used to describe these improvements in survival is "manufactured time", which is a phe- nomenon that can be quantified by examining sur- vival trends for people diagnosed with chronic fatal diseases (Olshansky, Carnes & Grahn, 1998). The Fourth stage was also characterized by gains in life expectancy at birth that had become more influ- enced by declining death rates at older than at younger ages. The influence of infectious and para- sitic diseases was inappropriately dismissed in this proposed Fourth stage of the epidemiologic transi- tion, because it was believed that these causes of death had come almost entirely under the control of modern medicine. Another Fourth stage extension, The Hylmstic Stage, was developed by Rogers and Hackenberg (1989) to address factors of potential demographic 208 importance not considered in the original three- stage model. These factors included 1) an interplay between infectious and chronic diseases within indi- viduals that could give rise to an increase in infec- tious and parasitic diseases associated with the rising prevalence of chronic conditions, a phenomenon influenced by rapid population ageing, 2) the im- portant influence of extrinsic social pathologies such as accidents, homicide, and suicide (particularly at younger ages), 3) the critical role of individual be- haviours in the timing of expression of ageing-re- lated mortality, which are factors that can have both positive (e.g., improved lifestyles) and negative (e.g., cigarette smoking) effects on longevity, and 4) the negative effects on longevity associated with a com- bination of social and behavioural factors, includ- ing economics, public policy, and human behaviour (e.g., the relatively recent emergence and spread of HIV I AIDS, and the inability to eliminate some in- fectious and parasitic diseases in spite of the avail- ability of immunizations). The Age of Delayed Degenerative Diseases model did not account for the four mortality modifiers identi- fied in the Hylmstic Stage model. The emergence of causes of death judged to be strongly influenced by individual behaviour and lifestyle including such major causes as cardiovascular disease and cancer was predicted by the Hybristic model as the so- called "hybristic diseases." The Hylmstic Stage model did not consider, however, the potential influence of genetic factors on the age of expression of chronic degenerative diseases. Neither of the proposed ex- tensions of Omran 's model addressed the possible re-emergence of infectious and parasitic diseases. Although Omran recognized the interaction between environment and resistance to infectious disease, epidemiologic transition theory focused primarily on how the manipulation of environmental factors could facilitate reductions in death rates from IPDs (Omran, 1971:520). The dominance of the Omran paradigm made it difficult for scientists to foresee the important influence that the re-emergence of these ancient killers could have on human mortal- ity (Olshansky, et al., 1997; Wilson, Levins, & Spielman, 1994). Global re-emergence of infectious and parasitic diseases Any discussion about the re-emergence of infectious and parasitic diseases should begin with a recogni- tion that nearly all of the communicable diseases causing morbidity and mortality in the world today could be avoided with technologies that already exist - clean water, sewage disposal and treatment, more sanitary living conditions, antibiotics, immunizations, and changes in behaviour. In other words, controlling the health and mortality consequences associated with most known IPDs is more a financial, or political challenge than a technological obstacle. It should be noted, however, that just a few decades ago this Rapp. trimest. statist. sanit. mond., 51 (1998) attitude led the medical community to declare that humanity was on the verge of one of the most impor- tant social revolutions in history - the complete elimination of IPDs (Burnet & White, 1962). Ironi- cally, this optimistic announcement coincided with the emergence of serious obstacles that blocked the pathway to the elimination of these ancient killers. As alluded to earlier, the persistence of health threats posed by IPDs is due, in part, to an inability to use existing technology to control or eliminate these diseases. Of more serious concern, however, are the unanticipated threats to public health that have been caused by "new" diseases that have sur- faced during the last quarter of the 20th century. The new diseases recognized since 1973 are listed in Table I. Several diseases listed in Table 1 are notable: the emergence of Ebola and Hanta virus in 1977, the E-coli bacterium 0157:H7 in 1982, the HIV identi- fied in 1983, the Hepatitis C virus in 1989 and the Sin nombre virus in 1993. These diseases have already Table 1 had a profound effect on human health in every part of the globe, as evidenced by their global map- ping (see Figure 1). In a literal sense, it is probably incorrect to refer to these diseases as new. For example, it is likely that the HIV has existed for hundreds, or even thousands of years, crossing over from animal populations occa- sionally to kill small isolated subgroups of human beings. Encroachments into yet-to-be, unidentified HIV environments, caused by expanding human populations, and by a population density that is large enough to sustain the virus have given the HIV the opportunity to use human beings as a permanent host. It is estimated that the HIV has infected more than 47 OOO OOO people throughout the world. HIV I AIDS has quickly become the fourth leading cause of death globally (WHO, 1996), and its relative impor- tance to total mortality is still growing. To illustrate the magnitude of the problem, more than 2 200 OOO people died from HIV I AIDS-related illnesses in 1998, a number equivalent to the total number of deaths Disease-Causing Microbes and Infectious Diseases Recognized Since 1973 Year Mlcrebe/dlsease 1973 Rotavirus 1975 Parvovirus 819 1976 Cryptosporidium parvum 1977 Ebola 1977 Legione/la pneumophila 1977 Hantavirus 1977 Campy/obacter jejuni 1980 Human Hymphotropic virus 1 (HTLV-1) 1981 Toxic strains of Staphylococcus Aureus 1982 Escherichia coli0157:H7 1982 HTLV·II 1982 Borrelia burgdorferi 1983 Human immunodeficiency virus (HIV) 1983 Helicobacter pylori 1985 Enterocytozoon bieneusi 1986 Cyclospora cayetanensis 1988 Human herpesvirus-6 (HHV-6) 1988 Hepatitis E 1989 Ehrlichia chaffeensis 1989 Hepatitis C 1991 Guanarito virus 1991 Encephalltozoon hellem 1991 New species of Babesia 1992 Bartone/la henselae 1993 Sin nombre virus 1993 Encephalitozoon cuniculi 1994 Sabia virus 1995 Human herpesvirus 8 HHV-8 Source: WHO, The World Health Report 1996: 112. Wld hlth statist. quart., 51 (1998) Type virus virus parasite virus bacterium virus bacterium virus bacterium bacterium virus bacterium virus bacterium parasite parasite virus virus bacterium virus virus parasite parasite bacterium virus parasite virus virus Health preblem Major cause of infantile diarrhoea worldwide Severe anemia Acute and chronic diarrhoea Ebola hemorrhagic fever/uncontrolled bleeding and kidney failure Legionnaires' disease Hemorrhagic fever Short-term diarrhoea T-cell lymphoma-leukemial cancer of the blood Toxic shock syndrome Hemorrhagic colitis; hemolytic uremic syndrome Hairy cell leukemia Lyme disease Acquired immune deficiency syndrome (AIDS) Peptic ulcer disease Persistent diarrhoea Persistent diarrhoea Roseola subitum I skin rash liver infection, epidemic hepatitis Human ehrlichiosis I influenza-like infection Chronic liver infection Venezuelan hemorrhagic fever Conjunctivitis Atypical babesiosis I infection with fever, chills and fatigue Cat-scratch disease I bacillary angiomatosis Adult respiratory distress syndrome Infection with fever, chills and fatigue Brazilian hemorrhagic fever Associated with Kaposi's sarcoma in HIV/AIDS patients 209 Fig. 1 Selected outbreaks of infectious and parasitic diseases (IPDs) reported between July 1996 and December 1999 .. Crlmean·Congo hemorrll1glc ftver Cllolera Typhoid E. coll Ence,11111111 Dlplllherla Cllolera Mytcanlltls Source: WHO Outbreaks report (online). available at http://Www.who.inVemc/outbr,ak flfWS/index.html (Oecember 1999) from all causes combined in the United States in the same year (WHO, 1999 fact sheet). The Ebola story is similar. Outbreaks of the Ebola virus have been restricted to small geographically isolated subpopulations in Africa. As in the case of the HIV, the Ebola virus has probably existed within animal populations for a long time. Unlike the HIV, however, the rapid progression of debilitating symp- toms and lethality in human beings continues to keep the Ebola virus contained within a small geo- graphic area. The consequences could be devastat- ing if this deadly virus were somehow transported to a densely populated urban area. The Helicobacter frilori bacterium has also been labeled as new when it was discovered in 1983 that this bacterium was the primary cause of peptic ulcer disease. Previously, the general consensus was that this disease was caused by stress. As with the HIV and Ebola virus, the only thing new about H. frYlori was the discovery of a causal relationship between this bacterium and a specific disease. The Hepatitis C Virus (HCV) is an example of a genuinely new disease that spread rapidly across the globe since it was identified in 1989 (Lavanchy, 1999). Approximately 3% of the world's population (about 170 OOO OOO persons) are now infected with HCV. It is estimated that 20% of the carriers of HCV develop cirrhosis of the liver, and that between 1 % to 5% of those with cirrhosis will develop liver cancer within lOyears (WHO, 1999factsheet). Ifthevirusindeed plays a causal role in this disease etiology, a wave of liver cancer could sweep across the globe in the coming decades, comprising a global threat to health that is linked to a virus that first appeared a mere two decades ago. 210 Another genuinely new disease is a genetic vari- ant of the bacterium that causes cholera - Vibrio cholerae0139. This virus first appeared in the Bay of Bengal in 1992, and then spread quickly to 10 other South Asian countries (WHO, 1998). Although the spread of this genetic variant has waned in recent years, the prevalence of V. choleraeOl biotype El Tor, which first appeared in Indonesia in 1961, more than doubled in a single year between 1997 and 1998 (WHO, 1999a). This dramatic rise in cholera to almost 300 OOO cases, and over 10 OOO deaths in 1998 alone occurred mostly in West Africa, and was believed to result from climate changes related to El Niiio (WHO, 1999a). In recent years, dramatic increases in incidence have been reported for a number of infectious dis- eases that have probably been around for thousands of years. For example, dengue is a mosquito-borne disease that is common in tropical and sul>tropical regions of the world. The relatively mild dengue fever was first identified in the 18th century and its deadly complication, dengue hemorrhagic fever (DHF), was first described at the end of the 19th century (Halstead, 1992). Although the prevalence of both diseases rose rapidly after the 1954 epi· demic in the Philippines spread to other Asian na· tions, the most dramatic increases followed the 1981 epidemic in Cuba (Pinheiro & Corber, 1997). To- day, dengue is endemic in virtually every continent except Europe. The deadly DHF form of dengue had been reported in only 9 countries prior to 1970, but has now reached epidemic levels throughout Asia, and most of the Americas. The rapid spread of the mosquito vector across the globe has put an estimated 2 500 OOO people at risk for dengue. Be. Rapp. trimest. statist. sanit. mond., 51 (1998) tween 1981 and 1995, the number of dengue/DHF cases have more than tripled in many Asian coun- tries, and the number of cases more than doubled in the Americas during the 3-year period between 1995 and 1998 (Pinheiro & Corber, 1997). In the first 10 months of 1998, the total number of den- gue/DHF cases in Brazil ( 475 000) was greater than the total number of cases reported just a few years earlier for all of South America. Over the last two decades of the 20th century, the number of yellow fever epidemics has risen, and more countries than ever are reporting cases of the disease. Deforestation, urbanization and global cli- mate changes have expanded the habitat for the mosquito that carries the virus responsible for yel- low fever (Aedes aegypti). Just as in the case of den- gue, the mosquitos have responded by increasing in number, and becoming more widely distributed geo- graphically. The proliferation and ease of interna- tional travel are also thought to contribute to the ongoing spread of this disease (WHO, 1999, fact sheet). There are an estimated 200 OOO cases of yellow fever every year, and this number is rising. Influenza is a disease that has been known for thousands of years, and still contributes to global trends in mortality. Probably the first influenza pan- demic recorded was in 1580. There have been 31 subsequent pandemics described in the literature that could be attributed to influenza. The ones most familiar to people today include the Spanish Flu pandemic (designatedA(HlNl)) ofl918-1920that killed more than 20 OOO OOO people, and the 1957 Asian Flu and 1968 Hong Kong Flu pandemics (both designated A(H2N2)) that together killed more than 1 500 OOO people. Given the regularity with which pandemics punctuate the flu epidemics that occur across the globe every year, scientists thought that another pandemic had begun when it was discov- ered in 1997 that a child in Hong Kong was infected with a virus ( designated A(H5Nl)) previously known to infect only birds. Although there were only 18 confirmed cases and 6 deaths associated with what is now known as the Avian Flu virus, public health officials were worried because of the apparent direct transmission of the disease from bird to human host (Centers for Disease Control, 1998). After the virus had jumped to human beings, another pandemic was possible. The most troublesome aspect of the annual flu epidemics, and of the less common, but more dangerous flu pandemics is that the most vul- nerable hosts are children and the elderly. Global population ageing has dramatically increased the number of elderly residing in nursing homes and extended care facilities, resulting in high concentra- tions of people who are particularly vulnerable to the flu viruses that will appear in the coming years. Prior to the 20th century, the bacterial disease Diphtheria was one of the leading causes of child- hood mortality. Improved living standards, and the development of immunizations caused the preva- Wld hlth statlst. quart., 51 (1998) lence and death rate from this disease to decline. Periodic outbreaks still occurred among unvacci- nated people living in developing and developed countries, especially during World War II, and among people of lower socioeconomic status (Vitek & Wharton, 1998). In the late 1970s, a new strain of the disease-causing bacterium made it possible for diphtheria to make a comeback in the former Soviet Union. When the Russian Federation was formed, the frequency and content of the childhood and adult vaccinations for diphtheria were altered. As a consequence, a diphtheria epidemic began in 1990, and by 1995 there were more than 50 OOO cases of the disease reported across the states making up the Federation (Vitek & Wharton, 1998). Although a diphtheria vaccine was quickly made more widely available, the incident raised serious concerns about the efficacy of this, and of other infectious disease programmes. As the majority of cases were adults, officials were concerned that adults who were immunized as children had not acquired immunity to the disease, and that without an adult revaccination programme, the older population would be vulnerable to the disease (Galazka & Robertson, 1995). Of all the infectious diseases that have plagued humanity throughout history, malaria would be one of the most devastating of them in terms of either morbidity or mortality. The global malaria eradica- tion programme begun in the middle of the 20th century was successful at first, when dramatic reduc- tions occurred as treatments for the disease were developed, and insecticides were used to attack the mosquito vector. Since the early 1970s, however, the prevalence of malaria has risen at an alarming rate due to a reduction in vector control programmes and the lack of an effective replacement for DDT. There are now an estimated 300 to 500 OOO OOO new cases of malaria each year that cause between 1 500 OOO and 2 700 OOO deaths (World Health Organization, 1996). Now that malaria is endemic in 91 countries, and almost 50% of the world's population is at risk (90% of the population of Africa), this disease is a serious and growing threat, especially for children and pregnant women (Gubler, 1998; Nchinda, 1998). The re-emergence and spread of malaria have become commonplace. A typical course of events occurred in the western highlands of Kenya. In the early 20th century, an expanding network of roads and railways led to the inadvertent transport of the malaria-carrying mosquitos from the low-lying area of Lake Victoria to the previously uninfected high- land areas (Malakooti, Biomndo & Shanks, 1998). The development of tea estates and agriculture in the highlands also contributed to the rise in malaria. Deforestation increased the breeding grounds for mosquitos, and laborers already infected with mala- ria migrated to work on the estates. Yet mosquito reduction programmes and disease treatments (pyri- methamine and chloroquine) were so successful 211 that this part of Kenya had been declared free from malaria throughout the 1960s (Roberts, 1964). The resurgence of malaria in Africa has been attributed to many factors: a protozoan parasite that has become resistant to treatment, mosquito vectors that have grown resistant to insecticides, the aban- donment of DDT for environmental reasons, cli- mate and land use changes that have been favour- able to the mosquitos, inadequate financial resources to combat malaria and other diseases (such as HIV I AIDS), conflicts forcing migration into malarial ar- eas, civil unrest and for economic factors (Malakooti, Biomndo & Shanks, 1998; Nchinda, 1998). The ma- laria problem is not restricted to Africa. During the 1990s, there has been a dramatic increase in the prevalence of malaria in South America (Roberts et aL, 1997), and epidemics have arisen in parts of Asia and India (Gubler, 1998). Recurring themes in many of the recent IPD histories have been the growing problem of resist- ance to antibiotics by the organisms that cause dis- ease, and the resistance to insecticides by the vec- tors that spread disease (Levy, 1998). Short genera- tion times, and rapid population growth permit mi- crobes to adapt rapidly to antimicrobial agents (Murray, 1991). For example, when first reported in Europe in 1988, the vancomycin-resistant enterococ- cus (VRE) bacterium appeared to be restricted to hospital settings (Utley et aL, 1988). Recent evidence suggests that VRE has escaped hospitals, and has been introduced into other health-care settings (McDonald et aL, 1998). In the United States, it is estimated that 15% of nosocomial (hospital-acquired) enterococcal infections are resistant to treatment by vancomycin (Centers for Disease Control, 1993). Alarmingly, recent data suggest that the antimicro- bial agents used to enhance the growth rate of feed animals may have given rise to a VRE that can be transmitted from animals to human beings (Wegener et aL, 1999). Evidence for a rising tide of multi-drug resistant bacteria is growing. The list includes, but is not restricted to, E. coli, Staphywcoccus aureus, Shigella flexneri, Shigella dysenteriae, Vibrio chokrae, Streptococ- cus pneumoniae, Streptococcus pyogenes, Haemophilus injluenz.ae, Neisseria meningitidis, Helicobacter pylmi, Campywbacter jejuni, Salmonella Typhimurium, Salm~ nella Hadar, Bacteroides .fragilis, Neisseria gonorrhoeae, Mycobacterium tuberculosis (Aubry-Damon & Courvalin, 1999; Okeke, Lamikanra & Edelman, 1999). Some of these bacteria are resistant to all presently known antibiotics. The importance of the emergence of resistant disease vectors should not be underestimated. In recent decades, the emergence, or re-emergence of many diseases can be attributed to the growing re- sistance of arthropod vectors to insecticides. As ex- amples, consider the mosquitos that transmit ma- laria, dengue/DHF, yellow fever, encephalitis and filariasis, the ticks that carry Lyme disease and 212 ehrlochiosis, the fleas and lice that carry Bartonella and rickettsiosis, and the sand flies that transmit leishmaniasis (Brogdon & McAllister, 1998). The combination of resistant bacteria and resistant dis- ease vectors represents an important and alarming development in the constantly evolving relationship that exists between pathogens and human health. Factors other than the biology of IPDs and their vectors have also contributed to the re-emergence and spread of IPDs throughout the world. Some of these factors are new phenomena in the history of human-microbial interactions (Olshansky et aL, 1997; WHO, 1999b). Population ageing, growth and move- ment are aspects of human demography that can have a profound impact on the maintenance and spread of IPDs. Tremendous advances in transpor- tation technology not only move goods and people around the globe in a matter of hours, but they also transport microbes and their vectors. Social, politi- cal and economic factors that cause the movement of people can increase the contact between microbes and people - whether by voluntary movements of people from rural to urban areas, or by the involun- tary movement of political refugees into IPD-rich environments. The disruption of air, land, and wa- ter by earthquakes, droughts, floods, hurricanes, El Nino, volcanoes, and other acts of nature can have a significant impact on the movement and geographic dispersion of microbes, their vectors and their sub- sequent interactions with human beings. Environ- mental changes caused by human beings (e.g., dam and road building, deforestation, flood control projects, irrigation) also contribute to the spread of IPDs. The overuse of antibiotics and insecticides is a major concern - a nearly unregulated activity that could have disastrous consequences for human be- ings in the ongoing arms race with microbes. Fi- nally, an inadequate or deteriorating infrastructure of public health can cause delays or failures in the response to health threats posed by communicable diseases, a phenomenon that has already contrib- uted to the re-emergence ofIPDs in both developed and developing countries. There have also been notable changes in the transmission routes of infectious diseases. Histori- cally, IPDs were transmitted through contaminated food and water, through the air, or by disease vectors such as mosquitos and ticks. Now, many of the most deadly infectious diseases are transmitted through drug abuse, blood transfusions, and through sexual contact. Such diseases now include HIV and hepatitis B and C. Thus compared to the pre-existing infectious diseases, some new diseases possess fundamentally different characteristics - diseases with different modes of transmission that require the development of new preventive efforts. For example, while it is relatively easy to control the spread of cholera by purifying water, it is difficult to control salmonella infection by completely supervis- ing food handling, and it is perhaps even more Rapp. trimest. statist. sanit. mond., 51 (1998) difficult to control HIV I AIDS through encouraging safe sexual practices. Many infectious diseases throughout the world, but particularly in less developed countries (LDCs), arise from the direct or indirect effects of poverty (Gwatkin & Heuveline, 1997). Gwatkin et aL (1999) examined the world's poorest 20% of the popula- tion based on per capita incomes adjusted for pur- chasing power. Overall, most of the world's 20% came from states in India, su~aharan Africa, Bang- ladesh, and Southeast Asia, with relatively fewer com- ing from provinces in China, the Middle East, Latin America and the Caribbean. Furthermore, 59% of the communicable diseases occurred among the poorest 20% of the world's population. Even within developing and developed countries, there are dra- matic interclass variations in the risk of infectious diseases - with the poor much more likely to die from IPDs (Gwatkin & Hueveline, 1997). Ironically, most of the deaths now occurring from IPDs through- out the world, regardless of social class, can be pre- vented with existing medical technology and inter- ventions (Murray & Lopez, 1996). The last two years of the 20th century have been particularly notable for the re-emergence of IPDs. The global epidemics associated with HIV and hepa- titis C continue to flourish. In addition, there have been major outbreaks of a number of other debili- tating and deadly diseases. In Africa, there have been outbreaks of cholera, yellow fever, meningitis, Ebola, typhus, and Crimean-Congo hemorrhagic fever. In Asia, there have been outbreaks of cholera, diphtheria, dengue, typhoid, and resistant forms of meningitis. In North and South America, there have been notable examples of dengue, diphtheria, chol- era, and yellow fever. In 1999, a strain of West Nile-like encephalitis appeared in the state of New York, a disease that had never before been seen on the North American continent. Does the re-emergence of infectious and para- sitic diseases represent a new Fifth stage in the epi- demiologic transition? Or, alternatively, does the decline in death rates observed for IPDs over the past 200 years represent only a temporary lull in the force of extrinsic mortality. If so, does the re-emer- gence of IPDs in recent decades signal a return to the First stage of the epidemiologic transition, the stage that encompassed most of human history? In the remainder of this paper, a case will be made for both arguments. The Flffh stage of the epldem/ologlc transition? The re-emergence of IPDs during the last quarter of the 20th century is associated with demographic features that distinguish it from the Fourth or Hybristic stage of the epidemiologic transition. A unique set of demographic and health circumstances in low mor- tality populations has contributed to a significant rise in IPDs, and to a pronounced shift toward older ages in the age groups of a population that are most Wld hlth statist. quart., 51 (1998) affected by their re-emergence. The uniqueness of the demographic features has led some researchers to hypothesize that some population subgroups have entered a new Fifth stage of the epidemiologic tran- sition - The Re-emergence of Infectious and Parasitic Diseases stage. As population ageing sweeps across the globe, the absolute number and proportion of the total population of elderly is rising rapidly- approaching 20-25% in some developed nations. This represents a new and dramatic shift in the age structure of the human population. For most of human history, less than 1 % of the population was aged 65 and over. In addition to ageing-related diseases, the compromised immune system of the elderly makes them particu- larly vulnerable to infectious diseases, especially pneumonia and influenza. The demographic mo- mentum of population ageing ensures that the number of people who are immunologically com- promised will grow - a trend that will accelerate rapidly over the coming decades as the baby boom cohorts of the mid-20th century reach older ages. Population ageing has a societal consequence that also contributes to the rise and spread ofIPDs. As the size of the older population increases, so does the number of people living within health care facilities for the elderly. Although serving a critical need, these facilities promote the rapid spread of infectious disease because they usually contain dense populations of people whose immune systems are either fragile, or compromised - compromised by age, as well as by medical treatments for degenera- tive diseases. Acquired or nosocomial infections are often quite virulent under these conditions, but until recently they have been contained within the buildings in which they arose. Evidence is accumu- lating, however, that virulent strains of the nosoco- mial infections are escaping to the general popula- tion due to rapid increases in the number of nurs- ing homes and assisted living facilities (Centers for Diseases Control, 1999). Although not linked to population ageing, a comparable situation exists in prisons where HIV I AIDS and tuberculosis have spread rapidly. In both cases, infectious diseases have spread rapidly through crowded populations and, by escaping geographical containment, have augmented the infectious disease burden of the general population. The human immunodeficiency virus (HIV) and the disease it causes (AIDS) have created a dramatic and unprecedented impact on the rise and spread of IPDs. Two decades ago, HIV was an obscure virus of little concern to public health officials. Today, it is globally distributed, and has infected an estimated 47 OOO OOO people. HIV is particularly insidious because it operates, in part, by attacking the im- mune system. As a consequence, HIV not only con- tributes directly to the IPD burden, but by compro- mising immune systems it also contributes to the burden indirectly by increasing the chances of op- 213 portunistic infections by other IPDs. For example, the recent increase and spread of antibiotic-resist- ant forms of tuberculosis has been directly attrib- uted to rising rates of infection from HIV (WHO, 1999b). HIV and survival to old age are not the only pathways to a weakened immune system. One of the defining attributes of the Fourth stage of the epide- miologic transition is the postponement of deaths from chronic degenerative diseases. Medical inter- ventions are an important contributor to the post- ponement of these diseases. For example, chemo- therapy and radiation therapy are responsible for reductions in case fatality rates, and the extension of survival for many patients with cancer. An una- voidable side effect of these therapeutic interven- tions that delay death and extend survival is that the immune system of the patient is weakened. As a consequence, these interventions also expand the pool of individuals at risk for IPDs, and this sub- group of the population will continue to expand as population ageing continues. Ironically, the conse- quence of medical interventions that are successful can be assimilated to the reaction of a water balloon to a finger prod; poke the balloon in one spot (successful intervention), and a bulge (unintended consequence) will appear elsewhere. Another factor used in the argument for a Fifth stage of the epidemiologic transition has been the "new" diseases that have emerged as a result of human action. Technically, these diseases are not new. What is new is the recent appearance of drug- resistant strains (genetic variants) of the bacteria and other organisms that cause familiar diseases - diseases such as malaria, meningitis, pneumonia, and tuberculosis among others. The demographic and longevity effects of these re-emerging IPDs also differ from those in any previous stage of the epide- miologic transition. Sexually transmitted diseases such as chlamydia, gonnorhea and syphilis are on the rise, and are known to reduce fecundity and fertility rates in those infected. Some genuinely new infectious diseases, such as hepatitis C, can actually raise the risk of what are otherwise known (perhaps inappro- priately) as chronic degenerative diseases, such as cancer and cardiovascular diseases. Furthermore, if some degenerative diseases of late life are caused by IPD exposure early in life, then a rise in IPDs could either push back, or reverse the delay of degenera- tive diseases that would characterize a population in the Fourth stage of the epidemiologic transition. Large increases in the absolute size of the immunocompromised population resulting from global population ageing, a rise in the number of people living in nursing homes, and in prisons, medical treatments for cancer, and a growing number of people infected with HIV have together created an entirely new set of conditions that con- tribute to the rise in IPDs, and a shift in their age distribution to older ages. Other forces influencing 214 today's unique characteristics ofIPDs are permanent changes in the human age structure, forces that will have lasting and profound effects on future trends in IPDs. All of these forces combined represent a strong argument according to which the recent emer- gence ofIPDs is a distinctively novel development in the history of human mortality. There are enough new attributes of this period of rising IPDs for it to be considered a Fifth stage in the epidemiologic transition. As is the case with Omran 's original three- stage model, it is anticipated that the Fifth stage of the epidemiologic transition will occur at varying speeds, and for somewhat different reasons depend- ing on the country. The First stage of the epldem/ologlc transition revisited? Although the arguments for a Fifth stage are compel- ling, an equally plausible argument for a totally different interpretation about mortality transitions can also be made. The re-emergence of infectious and parasitic diseases can be viewed as neither new nor novel, but rather a continuation of a mortality pattern that was initially used to define the First stage of the epidemiologic transition. Several arguments support this view. During the last 200 years, literally billions of people have benefitted from the progress that has been made in gaining some control over IPDs - including antibiotics, immunizations, medical diag- nosis and treatment, sanitation, and environmental manipulations. Despite this impressive progress, IPDs are still the single greatest source of human mortal- ity today. Proponents of epidemiologic transition theory distinguish modern patterns of mortality from those of the past. They provide descriptive labels for their classification of demographic attributes, and the effects that these attributes have on the health and life expectancy of a population. These classifi- cations can be useful when making mortality com- parisons between population subgroups, or across short periods of time within a population. These transition models can also be misleading, however, because their classification of short-term trends in mortality can lose, or obscure the significance of the larger time frame within which these trends are occurring. The vast majority of human history falls within what is termed the First stage of epidemiologic tran- sition theory, a volatile stage of mortality character- ized by peaks and troughs. To those living within a trough that persists for several generations, or a century, it would appear as though a "new" mortal- ity pattern had emerged, one characterized by death rates that are lower than those experienced by pre- ceding generations. As death rates have fluctuated throughout human history, relatively long periods of declining or increasing mortality are not unusual in either the short, or the long term. If history repeats itself, as it has countless times, the last 200 Rapp. trimest. statist. sanit. mond., 51 (1998) years of declining mortality should not be viewed as a "turning point" in human mortality that is worthy of its own special designation. This trend may be nothing more than a temporary period of favorable mortality that will be followed by an equally tempo- rary period of unfavorable mortality. From this per- spective, the re-emergence of IPDs is simply a pre- dictable and typical phase of the First stage, rather than a "new" stage in the epidemiologic transition of human health. A biological perspective provides another rea- son that the current re-emergence of IPDs should be considered nothing more than an ongoing inter- action between human beings and microbes. It is natural, but inaccurate, for human beings to think of this interaction in only one direction. How do microbes cause sickness, and what can be done to lessen their impact on human health? Microbes ex- isted for billions of years before anatomically mod- ern human beings appeared some 130 OOO years ago. The biological reality is that life on earth could thrive without human beings, but could not survive without microbes. In terms of their numbers, biomass, diversity, and geographic distribution, microbes are the most successful organisms on earth. They have survived and thrived because their short generation times and high reproductive rates enable them to adapt quickly to environmental challenges. To these sim- ple but elegant life forms, human beings are just another component of their biotic environment. Biologists and the medical community were aware of the potential consequences for public health when antibiotics were developed and introduced, but the benefits of the present outweighed the theoretical costs in the future. Similarly, scientists were aware of the potential dangers when they developed pesti- cides to protect agricultural plants, and to kill the insect vectors that carry disease. The rapid adaptive responses made by microbes and their vectors are recurrent confirmation of evolution by natural se- lection 's being the organizing principle upon which life is based. Antibiotic-resistant bacteria and viruses that resist treatment (including HIV) are predict- able by-products of the severe selection pressures that have been placed upon these organisms by human beings. They have adapted successfully to every biological weapon thrown at them in the past, and their prodigious capacity for reproduction will defeat any new weapons used against them in the future. Although microbes may lose eventually nearly every battle, the fear in the biomedical community is that human beings will lose the war, if we continue using the current rules of engagement. In the 20th century, human beings have achieved notable (sometimes short-term) victories over such diseases as smallpox, poliomyelitis, tuberculosis, measles, mumps, and diphtheria. In terms of how human beings perceive elapsed time, 50 years is long. As such, it would be easy to think that the Wld hlth statist. quart., 51 (1998) microbial killers are no longer a serious threat, and that the milieu of interactions between human be- ings and disease-causing microbes has been perma- nently and favourably altered. Neither history, nor biology supports this position. Elapsed time of 50 years or even 500 years is minuscule when consid- ered on an evolutionary time scale. The importance of infectious diseases as a source of mortality has fluctuated widely throughout human history. Or- ganisms flourish or become extinct with, or without the influence of human beings. The human response to microbial threats depends on science, finance, and the organization and implementation of public health programmes, which are actions requiring uninterrupted long-term reflection. Microbes do not think, and more importantly, they do not need to depend on thought for their survival. They sim- ply respond to the environmental challenges that confront them. Academic scholars can identify stages of epidemiologic transition, but for infectious micro- organisms, the ebb and flow of their ecological and evolutionary success simply unfold. Trends toward emergence, re-emergence, or disappearance ofIPDs are real short-term trends, but have little signifi- cance for the ongoing interactions between human beings and the microbes that cause human disease. Conclusions McKeown raised the interesting question whether medicine was a dream come true, saving individuals from otherwise fatal diseases; a mirage, substituting one disease for another; or a nemesis, actually doing harm (McKeown, 1979). Investigating trends in causes of death reveals some of the elusiveness of the answer. Certainly on an international scale there have been some great achievements in combating infectious diseases, most notably the eradication of smallpox, and the near elimination of polio. Examining trends in causes of death from the perspective of epidemio- logic transition theory is like following a mirage, however, due to the growing amorphous distinction between chronic and infectious diseases. Despite these concerns, epidemiologic transi- tion theory (Omran, 1971) has proved to be a use- ful tool for categorizing levels of mortality, and for comparing the relative risks of death across popula- tion subgroups over time. Extensions of this model to a Fourth (Olshansky & Ault, 1986), or Hybristic (Rogers & Hackenberg, 1989) stage may have been conceptually justified, but neither of these exten- sions, nor the original model itself captures the unique set of mortality conditions that are associ- ated with the rising threats from infectious diseases that have surfaced during the last quarter of the 20th century. A significant number of "new" IPDs have emerged in recent years. These IPDs have already had a pro- found impact on human mortality, and there is scien- tific justification to suspect that infectious diseases will continue to have a significant impact on human 215 health in the future. In fact, an argument can be made that an unique set of demographic and health conditions has now made the human population more vulnerable to IPDs than at any previous time in history. These conditions include an age structure that will lead to a growing population of elderly who are vulnerable to IPDs, the global spread of a "new" virus (HIV) that weakens the immune system needed to ward off IPD attacks, an international network of transportation that can deliver IPDs to just about any place in the world within a day, dramatic en- croachments into new habitats that expose human populations to "new" zoonotic diseases, and an alarm- ing rise in the number of antibiotic-resistant strains of bacteria, and of insect vectors that are resistant to pesticides and insecticides, all near-perfect condi- tions for a global epidemic, or pandemic ofIPDs. Although human beings are now facing mortal- ity risks that are uniquely different from any earlier time period of the modern era, it is unlikely that this time frame warrants being labelled a new stage of the epidemiologic transition. The typical lifespan of human beings defines the time scale that is most easily perceived by human beings, a time frame that rarely exceeds 100 years. As such, a human perspec- tive can easily overlook the biological time frame that is required to fully understand and interpret the never-ending interactions that occur between human beings and the organisms that cause human disease, because it is an evolutionary time scale that encompasses the origin of anatomically modern human beings. Should the "unique" mortality patterns and risks that are observed today be labelled a new Fifth stage of the epidemiologic transition, another expansion of the Fourth stage, or the re-emergence of the First stage? This question is probably only of concern to academic scholars. Labels are far less important than the elevated risks from IPDs that human popula- tions are now experiencing, and the reasons why these risks are climbing. Warning signals have sur- faced repeatedly in recent years, reminders that the ongoing battle with IPDs is far from over. From the first sign of Staphywcoccus aureus resistance to vanco- mycin in Japan in 1977 (Hiramatsu et aL, 1997), and person-to-person transmission ofhantavirus pulmo- nary syndrome in Argentina (Enria et aL, 1996), to the Avian flu virus (H5Nl) transmitted directly from bird to human being for the first time in Hong Kong in 1997, and the presence of West Nile-like encephalitis in New York State in 1999, there have been plenty of warning signs. To respond, the con- tinued global monitoring of these alarming events should be intensified, and the continued use and misuse of antibiotics, and of other pharmaceuticals in the escalating battle with microbes should be re- evaluated within the context of recent trends in morbidity and mortality for the diseases that they are intended to influence. 216 At:know/1dgm1nts Funding for this work has been provided by the National Insitute on Aging to Dr Olshanksy and Dr Carnes (AG13698-0l, AG00894-01), the National Aeronautics and Space Administration to Dr Carnes (98-HEDS-02-257), and the National Science Foun- dation to Dr Rogers (SES 9906145). References Aubry-Damon, H., Courvalin, P. Bacterial Resistance to Antimicrobial Agents: Selected Problems in France, 1996 to 1998. Emerging Infectious Diseases 5(3): 315-320 (1999). Brogdon, W.G., McAllister,J.C. Insecticide Resistance and Vector Control." Emerging lnfecti(JUSDiseases4(4): 605-613 (1998). Burnet, M., White, D.O. Natural Hist""Y of Infectious Disease. London: Cambridge University Press, 1962. Centers for Diaeue ConcroL Nosocomial enterococci resistant to vancomycin-United States l 9~1993. Morbidity and Morlality Wtdcly Report 42: 597-600 (1993). Centers for Diaeue Concrol. Tuberculosis outbreaks in prison housing units for HIV-infected inmates- c.alifornia 1995-1996. MorbidityandMmality WtdclyReport48 (4): 79-82 (1999). Centers for Disease Concrol. Update Isolation of Avian Influenza A(H5NI) Viruses From Humans- Hong Kong, 1997-1998. Morbidity and Morlality Wtdcly Report46: 1245-1247 (1998). Davis, Kinpley. The Theory of Change and Response in Modern Demographic History. Popul.ation Index. 29(0ctober): 345-66 (1963). F.nria, D., Padula, P., Segura, E.L, Pina, N., Edelstein, A., POS1e, C., et al. Hantavirus pulmonary syndrome in Argentina: possibility of person to person transmission. Medicina 56: 709-711 (1996) in Schmaljohn, C., Hjelle, B .. Hantaviruses: A Global Disease Problem. Emerging lnfecti(JUS Diseases 5 (2): 95-104 (1997). Galma, A.M., Robenson, S.E. Diphtheria: Changing Patterns in the Developing World and the Industrialized World. European Journal of Epidemiology 11: 107-117 (1995 ). Gubler, D. Resurgent Vector-Borne Diseases as a Global Health Problem. EmerginglnfectiowDiseases4 (3): 442-450 (1998). Gwaddn, D.R., Guillot, M., Heuveline, P. The Burden of Disease among the Global Poor. Lanat554 (9178): 58~589 (1999). Gwaddn, D.R., Heuveline, P. Improving the Health of the World's Poor. BritisJ&MedkalJourna/515 (7107): 497-498 (1997). llalltead, S.B. The XXth Century Dengue Pandemics: Need for Surveillance and Research. World Health Statistics Quarterly 45: 292-298 (1992). Hiramaau, K. Hanaki. H., Ino, T., Yabuta. K., Oguri, T., Tenover, F.C. Methicillin-resistant Staphylococcus aumu clinical strain with reduced vancomycin susceptibility. Journal of Antimicrobial Chemotherapy 40: 135-136 (1997). I...immcby, D. Global Surveillance and Control of Hepatitis C. Journal a/Viral Hepatitis 6: 35-47 (1999). Levy, S.B. The Challenge of Antibiotic Resistance. Scientific American218 (3): 4~53 (1998). Malaltooti, M.A., Biomnclo, K., Shanks, G.D. Reemergence of Epidemic Malaria in the Highlands of Western Kenya. Emerging lnfectj(JUSDiseases4 (4): 671-676 (1998). McDonald. LC., Kuehnert, MJ., Tenover, F.C., Janis, W.R. Vancomycin-resistant Enterococci outside the health-care setting: prevalence, sources, and public health implications. Emerginglnfectj(JUSDiseasesJ (3): 311-317 (1999). McKeown, T. The Role of Mtdicine: Dream, Mirage, "" Nemesis 7 Princeton, NJ.: Princeton University Press, 1979. MeNamara, R. Demographic Transition Theory. International Encycloptdia of Popul.ation. Volume I. Edited by John Ross. The Free Press, 1982, pp. 14~7. McNeil, W. PI.agues and Peoples. New York: Doubleday, 1977. Rapp. trimest. statist. sanit. mond., 51 (1998) Murray, B.E. New Aspects of Antimicrobial Resistance and the Resulting Therapeutic Dilemmas. Journal of Infectious Diseases 163: 1185-1194 (1991). Murray, C.J.L, Lopez, A.D. (eds). The Global Burden of Disease: A Cmnprehensive Assessment of M<rrtality and Disability from Diseases, Injuries, and Risk Fact<rrs in 1990 and Projected to 2020. Cambridge, MA: Published by the Harvard School of Public Health on behalf of the World Health Organization and the World Bank; Distributed by Harvard University Press, 1996. Ncbinda, T.C. Malaria: A Reemerging Disease in Africa. Emerging InfecliousDiseases4 (3): 398-403 (1998). Okeke, I.N., Iaroikann, A., Edelman, R. Socioeconomic and Behavioral Factors Leading to Acquired Bacterial Resistance to Antibiotics in Developing Countries. Emerging Infectious Diseases5 (2): 18-27 (1999). Olsbaosky, SJ., Carnes, B.A., Cassel, C. In Search of Methuselah: Estimating the Upper Limits to Human Longevity. Sciena 250: 634-640 (1990). Olsbaosky, SJ., Carnes, B.A., Rogen, R.A., Smith, L Infectious Diseases: New and Ancient Threat to World Health. Population Bulletin52 (2): 1-58 (1997). Olsbmsky, SJ., Carnes, B.A., & D. Grabo. Confronting the Boundaries of Human Longevity. American Scientist 86 (I): 52-61 (1998). Olsbmsky, S.J., Ault, B. The Fourth Stage of the Epidemiologic Transition: The Age of Delayed Degenerative Diseases. The Mil.bank Quarterly64 (3): 355-391 (1986). Omrao, A. The Epidemiologic Transition: A Theory of the Epidemiology of Population Change. Mil.bank Memorial Fund Quarterly 49:509-538 (1971). Omran's epidemiologic transition theory was also published in 1977 by the Population Reference Bureau. Pinheiro, F.P., Comer, SJ. Global Situation of Dengue and Dengue Haemorrhagic Fever, and its Emergence in the Americas. World Health Statistics Quarterly 50: 161-169 (1997). Wld hlth statist. quart., 51 (1998) Robens,J.M.D. The Control of Epidemic Malaria in the Highlands of Western Kenya, Part III. After the Campaign. Journal of Tropical Medicine and Hygiene 59 230-7 (1964). In Malakooti, Biomndo, and Shanks. 1998 (op cit.). Roberts, D.R., Laughlin, LL, Hsbeih, P., Legten, LJ. DDT, Global Strategies, and a Malaria Control Crisis in South America. Emerging Infectious Diseases 3 ( 3): 295-302 (1997). Rogers, R., Hackenberg, R. Extending Epidemiologic Transition Theory:ANewStage. SocialBiology34 (3-4): 234--243 (1989). Teitelbaum, Michael S. Relevance of Demographic Transition Theory for Developing Countries. Sciena. 188 (May): 420-25 (1975). Uttley, A.H.C., Collios, C.H., Naidoo,J., George, R.C. Vancomycin· resistant Enterococci. Lanai 1: 57-58 (1988). Vitek, C.R., Wharton, M. Diphtheria in the former Soviet Union: Reemergence of a pandemic disease. Emerging Infectious Diseases4 (4): 539-550 (1998). Wegener, H. C., Aarestrup, F.M.,Jeasen, LB., Hammenun,A.M., Bager, F. Use of Antimicrobial Growth Promoters in Food Animals and Enterococcus faecium Resistance to Therapeutic Antimicrobial Drugs in Europe. EmerginglnfeaiuusDisease.s5 (3): 329-335 (1999). Walson, M.L, Levins, R., and A. Spielman, (eds.) Disease io Evolution: Global Changes and Emergence of Infectious Diseases. Volume 740. New York: The New York Academy of Sciences, 1994. World Health Organization. The W<rrld Health Report 1996: Fighting Disease, Fostering Development. Geneva: World Health Organization, 1996. World Health Organization. Weekly Epidemiological Rec<rrd 73: 201- 208 (1998). World Health Organization. Weekly Epidemiological Record 74:257- 264 (1999a). World Health Organization. Global Microbial Threats in the 1990s. Geneva: World Health Organization, l 999b. World Health Organization. 1999 fact sheet. hup://www.who.int/ inffs/en/factl 00.html. 217