The Future of Epidemiology:
Looking Ahead at the Twenty-First Century
David F. Duncan, Dr.P.H., F.A.A.H.B.
Duncan & Associates
Brown University School of Medicine
Lecture to students in the M.P.H. Program,
Fort Valley State University
Fort Valley, Georgia
October 29, 2001
In 1988 a committee of the national Institute of Medicine released a report entitled The Future of Public Health (Committee for the Study of the Future of Public Health, 1988). The Future of Public Health sounded an alarm, describing the state of public health in the United States as being characterized by insufficient funding, inadequate capacity, organizational fragmentation, and disjointed decision-making. The Committee offered a vision of how America could recover from this state of disarray.
In the years since 1988, some of the deficiencies identified by the Committee were corrected but many were not. Public health continues to be underfunded and many public health concerns continue to be addressed in a fragmented fashion -- if at all. Environmental concerns in particular have largely been divorced from health concerns. Even concerns specific to environmental health have too often been isolated in agencies separate from general public health.
The September eleventh assault on America has focused renewed concern on long neglected elements of our public health system. The speed with which injured victims of the World Trade Center disaster were triaged into medical care and with which emergency medical facilities were established and staffed in the wake of the attack showed that our emergency response systems were in good shape. The response to anthrax distributed through the mail has shown that our traditional public health activities in epidemic control are in better shape than many critics had suggested in their warnings about the dangers of bioterrorism. The anthrax crisis has also accentuated the fact that the critical level of response to bioterrorism is in the local public health department. If we hope to achieve any safety from bioterrorist attacks the key element in our response must be the maintenance of effective systems of disease surveillance and investigation of outbreaks in our local public health departments.
New and Reemerging Diseases
Bioterrorism is not the only, or most likely, source of new infectious threats to the public. New infections will emerge from time to time in the future as they have in the past. Old infections that once seemed largely under control, such as tuberculosis and cholera, will remerge as serious public health problems. We are past due for another pandemic of influenza that could be as serious as the 1918 pandemic. Among the influences leading to the emergence of new infections and the reemergence of old ones will be:
When new infectious organisms reach a population they are likely to be characterized by high levels of infectivity, pathogenicity and virulence – that is of ability to spread through the population, to produce disease in those who are infected, and to cause severe illness and even death in those with the disease. Over time, changes in host resistance or in the infectious organism itself tend to reduce all three of these characteristics. Given enough time equilibrium develops between the infectious organism and its host, such that the infection can feed on the host without killing or debilitating the host.
It is even possible that an infection may in time become completely harmless or even beneficial to the host. The bacteria that thrive in the human bowel – the so-called "intestinal flora" -- are an example of a beneficial infection. A newborn infant cannot digest food properly until it is infected with these bacteria and their unintended elimination during antibiotic therapy causes diarrhea in adults.
Viral and host populations can exist in such an equilibrium for centuries, perhaps even millions of years, until changes in environmental conditions shift the equilibrium and favor rapid evolution in the virus. Ecological disruptions have often caused new and dangerous infections to emerge. The introduction of bubonic plague into Europe resulted from the introduction of a different species of rats, which in turn was probably the result of environmental disruption of the rat habitat in South Asia. The emergence of Lyme disease has resulted from changing land use patterns in the Northeastern United States which have resulted in more people living in close proximity to deer, field mice and their ticks. Legionnaire’s disease was a rare and unidentified infection until the development of air conditioning systems provided it with an ideal reservoir and mechanical vector of distribution.
HIV is thought to have originated as a mutation of Simian Immunodeficiency Virus (SIV), a disease of old world monkeys. Crossover of the virus from chimpanzees to humans is thought to have occurred more than three hundred years ago and some researchers believe that it may have taken place as long as eight-million years ago (Doolittle, Feng, Johnson, and McClure. 1989). Disruption of society and family life resulting from civil wars produced increased promiscuity and prostitution and created the ideal social environment for development of a sexually transmissible disease such as HIV, which has continued to mutate at a rapid rate after its many generations of stability.
Hazards of Social Change and Stress
In many parts of the developing world there is increasing social and demographic turbulence -- massive migration from rural regions into urban slums, usually lacking safe water supplies, adequate sanitary facilities and public health infrastructure, often aggravated by conflict and breakdown of law and order. This situation favors epidemic disease and increasing social unrest.
As I have already noted the social disruption resulting from civil wars in Africa probably played an important role in the emergence of HIV. The same is probably true of ebola virus and a number of other emerging and reemerging infectious diseases. Any breakdown of the social structure also has effects on human behavior that may result in increased violence, automobile accidents, and abuse of alcohol and other drugs.
The experience of the former Soviet Union after the fall of communism demonstrates the impact that social change can have on the health of the population. Even before the collapse of communism, mortality in the Soviet Union and its vassal states was excessive – due in large part to high levels of air pollution, smoking and alcoholism. In 1991, life expectancy for men was 63.4 years and 74.3 years for women. By 1995, four years after the dissolution of the Soviet Union, Russian life expectancy levels had declined to 58.2 years for men and 71.1 for women.
In part the disastrous decline in health following the collapse of the Soviet Union may have been due to worsening of the previous causes of excess mortality. But the major reasons appear to be rooted in the loss of organizational infrastructure. The newly independent states created from the former Soviet Union lacked administrative structures necessary to continue the public health activities that had previously been centrally administered from Moscow. Childhood immunization activities, in particular, suffered due to this lack of organization. Diphtheria, for instance, which had previously produced less than 1,000 cases per year in the USSR, struck 47,802 persons in the newly independent states in 1994, resulting in 1,746 deaths. Outbreaks in neighboring nations soon occurred due to the availability of rapid transportation and migration motivated by mass unemployment following failure of the communist system.
World population has undergone unprecedented growth during the past Century. Continued population growth is the engine that drives many of the factors contributing to new public health challenges. While growth has slowed in the developed nations it continues unabated in the developing nations. These developing nations commonly face the double burden of new risks arising from population growth and social change while continuing to be plagued by traditional risks.
The most direct impact of population growth with public health implications is congestion. Overcrowding of housing and other settings increases the likelihood of contagion while the resultant stress may suppress the immune systems of people exposed to congested living conditions. Increased traffic congestion results in more accidents and can contribute to "road rage" induced violence.
Population growth places a strain on necessary infrastructure. Water supply, sanitation, wastewater treatment, and solid waste disposal are hard pressed to keep up with rapid population increase. Housing and medical care facilities can also be overwhelmed by population growth. All of this threatens the health of the population.
Population growth also contributes to increased pollution. Air pollution increases as a result of more cars, increased domestic fuel use, and the expansion of industry. Water pollution also increases as a result of overwhelmed infrastructure for wastewater treatment, more industrial effluent, and both deforestation and more land covered with pavement -- resulting in more run-off.
Global Climate Change
There is no longer any serious question about the fact that global warming is occurring due to the build-up of greenhouse gases. Continued global warming will result in depletion of essential resources, especially fresh water and food; species extinction, reduced biodiversity, and massive changes in the balance among all the living creatures on earth with which humans are interdependent; desertification; and underlying all these, sociodemographic, economic, cultural and political changes among the nations of the earth.
Alpine glaciers and ice caps in Africa, the Americas, Europe, even much of the Himalayas, are receding, some could disappear altogether, with devastating consequences for river flow and downstream irrigation. Wetlands are drying out, rainfall patterns are changing, and weather extremes (floods, hurricanes, droughts, etc) are all becoming more frequent and more severe at a greater rate than originally estimated. The health impacts of all these processes could be devastating, and will affect huge numbers of people. For example, an increase in the average ambient temperature of half to one degree Celsius would expand the range of malaria-carrying mosquitoes into temperate zones and higher altitudes and would put almost a billion additional people at risk. Melting of polar and alpine icecaps and thermal expansion of the sea water mass is expected to raise sea levels by about 50 cm in the next 50 years. Half a billion people live very near sea level; their habitat, especially in small island states will disappear or be threatened with inundation. Climate change also threatens food security, by leading to less predictable weather patterns and growing seasons, and more frequent extreme weather (floods, droughts). Climate change forces people to migrate as environmental refugees, and thereby imposes severe strains on public health services, many of which are already stretched to the limit because tax revolts and budget cuts have led to deterioration of public health infrastructure.
The importance of genetic epidemiology to public health is certain to increase steadily in the post-genome mapping era. We will see this discipline come into its own in the realm of gene-environment interactions. Understanding the interactions between risk factors and a person's genetic make up will allow us to identify those people who are genetically susceptible to an environmental risk factor.
It may well be the case in the future that genotype assessment will become a part of all epidemiological studies. The failure to take into account the genetic make up of study participants may make it more difficult to find out who is at risk, since the risk factor under study may affect only a portion of the population.
At first glance it may seem that genetic epidemiology can offer nothing to public health in the way of practical preventive interventions other than a return to the old ideas of eugenics in a modern, more precise form. It may not seem that genetic epidemiology can serve to identify the sort of modifiable factors that public health can seek to change in order to actually change disease risks in the population.
The truth, however, is that in at least some instances we can influence gene-environment interactions. For instance, some infants, as a result of a genetic fault, are born deficient in the enzyme that metabolizes phenylalanine. The resultant accumulation of unmetabolized phenylalinine causes neurological damage that causes learning disabilities. Since we know this and can test for this defect, babies are routinely screened for the enzyme deficiency and those who test positive are placed on a diet free of phenylalanine, thus preventing them from developing learning disabilities.
It is easy to imagine a similar scenario for cancer. If as we suspect there is a strong genetic predisposition toward certain risk factor-environment interactions in cancer causation, then we will in the future be able to advise people on which risk factors it is most important for them individually It may even be possible in the future, combining genetic epidemiology with molecular biology, to induce changes in the individual’s genotype by introducing a designer virus that will modify the host’s DNA eliminating susceptibility to a selected risk factor.
One way in which molecular biology is having a growing effect on public health is in the identification of hidden viruses. One accomplishment of these new techniques has been the identification of the human papilloma virus (HPV) as the probable viral agent involved in causing cervical cancer. Further identification of viruses associated with major cancers provides one possible route to the elimination of cancer as a major threat to public health.
Molecular biology is also revolutionizing the development and manufacture of vaccines. It is no longer necessary to use the entire viral agent in a vaccine, with the attendant risks that it may not have been fully inactivated or that it may provoke an allergic reaction. Instead specific antigens of the virus – parts of the viral coat that are recognized by antibodies – can be isolated and used to prepare vaccines which are effective and also safe. The hepatitis B (HVB) vaccine is an example of this new technology in practice. Given the fact that chronic infection with hepatitis B is now known to be the major cause of liver cancer in developed nations, this vaccine may be regarded as the first anti-cancer vaccine.
The greatest contribution of molecular biology to public health has been in allowing us to improve the precision of measurement in the epidemiologic studies that provide the basis for public health practice. All measurements contain some element of error. Both the self-report data and physician reports on which epidemiology is so reliant are highly subject to errors, both deliberate and unintentional, which may lead to inaccurate results. Imagine that in an epidemiologic study 10% of all smokers said that they did not smoke. As a result the estimates of relative and attributable risk of disease between smokers and non-smokers arrived at in that study would be diluted because part of the group that we thought were non-smokers were actually smokers, with a higher risk of disease.
Molecular biology is increasingly revealing to us that there are a host of biomarkers of exposure -- biological factors that reflect the individual's exposure levels. Most often these are measured in samples of blood or urine. For example, measuring the amount of cotinine, the major metabolite of nicotine, can tell you whether someone has recently smoked a cigarette.
There are limitations, however, to the use of biomarkers of exposure. First, they tend to be expensive, which limits their use in epidemiological studies and even more so in public health practice. Second, biomarkers often reflect only recent exposures, whereas disease processes are more often influenced by cumulative exposure over a long period. Third, biomarkers do not always translate readily into modifiable behaviors that can be altered to decrease risk of disease. Knowing the relationship between cotinine and disease is useful only because we know that reducing smoking will directly reduce cotinine levels to a predictable degree.
Molecular biology has also allowed us to identify biomarkers of disease. This can contribute both to more effective population screening and more accurate individual diagnosis.
Biomarkers can also permit us to identify disease in its preclinical stage. If we know before clinical signs develop that the disease process has begun in individuals we can channel them into an early intervention thus practicing good secondary prevention. Preclinical biomarkers will also be valuable in cohort studies where we are often faced with a long wait to see whether individuals will develop a disease. Early determination of outcomes using biomarkers will save much research effort and expense.
Newer statistical models
The analytic methods used in epidemiological have changed radically since the introduction of computer statistical packages, such as SPSS, SAS and EpiStat. The continuing expansion of the range of statistical procedures available in these packages and the growth in available computing power has had a major impact on the analysis of public health data of all types. Analysis of the cross-classified data that characterizes most public health research no longer relies on traditional methods of contingency tables and Chi-squared tests. Today such data is more likely to be analyzed using log-linear and logistic models and there is increasing use of the alternate weighted least squares model. Nonlinear models, such as nonlinear canonical correlation, have seen limited use but are likely to play an important role in an increasingly ecological or holistic approach to the understanding of disease causation.
Changes which have taken place in the analytical approaches to public health data have not all been related to advances in computer software and hardware. The introduction and refinement of the randomized controlled trial (RCT) constituted an important advance in obtaining solid evidence of disease causation or preventive efficacy. The realization that the results of many RCTs, which address the same question, could themselves be treated as data in a model known as meta-analysis. Both RCTs and meta-analysis are likely to see expanded use in the future of public health. We can expect further development of statistical methods which allow the analysis of increasingly complex sorts of data resulting from more sophisticated RCTs and meta-analyses.
Information Technology and Medical Informatics
Modern information technology will help us to collect and manage the ever-larger volumes of data necessary to monitor the impact of the new challenges facing public health in this Century. An ecological perspective on health requires simultaneous monitoring of numerous factors and their integration in multidimensional models for analysis. Timely identification of newly emerging and reemerging threats to public health will require faster means of data collection and handling than is possible using traditional paper-based systems, reliant on mailed reports. We will doubtless see increased reliance on email and automated systems for data collection and networked computers for analysis at agency, local, regional, and national levels.
With the computer revolution the always-important function of medical records has been transformed into a new field known as medical informatics. This new discipline has become an increasingly vital element of our healthcare system. In the future medical informatics will play a growing role in preventive medicine.
The central elements of medical informatics are data acquisition, information distribution/storage, access, decision support, and distance education. Once biosignals or images are measured, the information system will acquire this information, preferably through a direct connection with the medical instrument. While this data could be acquired and stored in a home personal computer, local home server, or commercial secure website, a more promising possibility is storage on a "chip card." Such a credit card sized device would contain a tiny microprocessor chip and non-volatile memory storage. Use of a chip card would give individuals direct physical control over their own data and the ability to link it to a personal computer or computer terminal via a special chip card reader. Such readers are likely to become as common as a CD drive on computers in the future..
Powerful decision support systems can then analyze the data, comparing it to the individual’s past information to determine any significant changes. Only those changes considered substantially outside of norms for that individual will be used in an alert, first to the individual, and then to a healthcare provider. The systems that analyze the information will likely be distributed throughout the internet, since it is unlikely all the decision support systems could or would be contained on a single website.
Every bit as important as access to individual information, will be access to appropriate medical educational and reference material. Intelligent agents will be needed to sift through the mass of information to selectively identify information relative to the person’s query. It will be essential to provide information that is accurate and up to date. This information must be provided in a manner that is usable by the individual patient --with a human/user interface that stimulates and encourage healthful behavior rather than confuses or frustrates the person seeking advice.
When necessary, individuals will be able to consult a physician or other health professional using systems that are reminiscent of today’s email but which will greatly facilitate the exchange of medical information and enhance the doctor-patient relationship. It is likely that such tele-consultations will be fully interactive, 3-D immersive experiences which will give participants the sense of the other’s presence. Such virtual reality over the internet has been called tele-immersion and is a cutting edge area of computer science.
Bloom, B. R. (1999). The future of public health. Nature, 402, C63-C64.
Centers for Disease Control (1998). Preventing Emerging Infectious Diseases: A Strategy for the 21st Century. Atlanta, GA: CDC.
Cockerham, W. C. (1999). Health and Social Change in Russia and Eastern Europe. New York: Routledge.
Committee for the Study of the Future of Public Health, Institute of Medicine (1988). The Future of Public Health. Washington, DC: National Academy Press.
Doolittle R.F., Feng D.F., Johnson M.S. and M.A. McClure. 1989. Origins and evolutionary relationships of retroviruses. Quarterly Review of Biology, 64, 1-30
Drexler, M. (2001). The germ front. The American Prospect, 12(19), 26-29.
Duncan, D. F. (1988). Epidemiology: Basis for Disease Prevention and Health Promotion. New York: Macmillan.
Institute of Medicine (1985). Vaccine Supply and Innovation. Washington, DC: National Academy Press.
Lederberg, J., Shope, R. E., and Oaks, S. C., Jr. (1992). Emerging Infections: Microbial Threats to Health in the United States. Washington, DC: National Academy Press.
Miller, J., Engelberg, S., and Broad,W. (2001). Germs: Biological Weapons and America’s Secret War. New York: Simon & Schuster.
Morse S.S. (ed.). (1993). Emerging Viruses. Oxford University Press.
Morse S.S. and A. Schluederberg. 1990. Emerging Viruses: The Evolution of Viruses and Viral Diseases. Journal of Infectious Disease, 162, 1-7
Rudolph, F. B., and McIntyre, L. V. (Eds.) (1996). Biotechnology: Science, Engineering, and Ethical Challenges for the 21st Century. Washington, DC: Joseph Henry Press.
Steinhauer D., and Holland, J. J. (1987). Rapid evoltuion of RNA viruses. Annual Review of Microbiology, 41, 409-433