Epidemics, Impacts, And Responses

The impact of AIDS has many dimensions, only a few of which are captured in official statistics or analysis by the research community. The numbers of AIDS cases and HIV infection count as an impact: cumulatively, they state the effect on the population of the United States and on particular subpopulations. Each case has many dimensions—personal, professional, and institutional—through the many social organizations that touch the life of each infected person. Each set of interactions creates an impact, and the diverse impacts have generated equally diverse responses by individuals, groups, and communities.

The panel set out to study these impacts, and it immediately confronted the problem of defining the terms of reference. "Impact" is an overused word that in common parlance has become a synonym for "effect." In this sense, it indicates that one action or state of affairs is caused or influenced by some other action or state of affairs and is used to describe both major and minor effects. Reaching deeper into the language, however, impact has a more powerful meaning—collision. In this use of the word, an impact is an effect that radically changes the previous state of affairs or even destroys it.

After much discussion, the panel adopted a definition of impact that fits somewhere between these two meanings. "Impact" as used by the panel describes a concentrated force producing change, a compelling effect. We adopted this hybrid meaning not only because it more accurately describes the impact of AIDS on contemporary America—social institutions have not been destroyed—but because we quickly realized that social impact does not merely destroy; it evokes a reaction or a response. It is more organic than physical. Persons and societies do not merely feel the impact of an event; they remake their lives and institutions to accommodate, negate, or preserve its effects. In this report, we attempt to capture and describe the process of impact and response of selected social institutions to the HIV/AIDS epidemic.

The task of this panel was to go beyond, to the extent possible to limited human vision, the impression of the extraordinary impacts of AIDS on individual lives and on social institutions. We have tried to sort out those that will endure in such a way as to force, or to invite, Americans to take them into account in the next decade. This epidemic is not ordinary in one quite specific way: it can be determined many years in advance of the onset of actual illness in a patient that the illness will come. This epidemic is not, like many historical epidemics, an invasion of morbidity and mortality that rapidly sweeps through a population. It comes and will stay for years, not only in the population, but in the individual people infected, and its presence will often be known to them and to others long before they suffer the disabling, lethal effects. Similarly, rough estimates can be made of the numbers of people who will begin to experience those disabling, lethal effects years from now. Thus, Americans must think about this epidemic for many years into the future.

The institutions we studied appear to have absorbed the impact of AIDS and accommodated to it in a very limited way. However, even a response that is partial and apparently transitory may mark the beginning of more fundamental change. Several of the institutions we studied may follow this trajectory of limited initial response, followed some years later by very significant changes. These longer term responses would be interesting to follow, and we hope that researchers will attempt to do so. However, the panel did not attempt to suggest a methodology for longer term monitoring: the data needs and methods of observation would be very different for the individual institutions studied.

After extensive deliberation, the panel determined that it had sufficient information and understanding to describe social impact and response for six institutions (broadly defined):

• the public health system
• health care finance and delivery
• clinical research and drug regulation
• religion
• voluntary organizations
• the correctional system

These institutions were selected for several reasons: the panel members had the competence to study and evaluate them, we judged that sufficient empirical data and informed opinion existed to formulate our own assessment of impact and response, and they had not been treated in previous reports of the parent committee. The six institutions chosen are very different in structure, degree of centralization, and other dimensions. Such differences affect the level of generalization appropriate to each area.

In the course of our work we also began to see another kind of impact and response—on public policies not necessarily connected to institutions. The HIV/AIDS epidemic has clearly had an impact on policies related to families, and we thus decided to add to our study an examination of two policy areas: issues related to newborns and children and issues related to intimate nonmarital relationships.

Finally, in addition to examining institutional systems as a whole and selected family policies, the panel wanted to look at the impact of HIV/AIDS on communities, where several institutions converge and where the synergy resulting from that convergence is most clearly seen. Originally, three case studies were envisioned: New York, Miami, and Sacramento. We were able to complete only New York—a city that could never be described as typical, but one that does vividly illustrate the impact and response to AIDS among major social institutions.

It is common to find references to the impact of AIDS and HIV. It is also rather common to find such references expressed in quite strong terms. For example, Milbank Quarterly's two-volume study, A Disease of society: Cultural Responses to AIDS, opens with these words (Nelkin et al., 1991:1):

AIDS is no "ordinary" epidemic. More than a devastating disease, it is freighted with profound social and cultural meaning. More than a passing tragedy, it will have long-term, broad-ranging effects on personal relationships, social institutions, and cultural configurations. AIDS is clearly affecting mortality—though in some communities more than others. It is also costly in terms of the resources—both people and money—required for research and medical care. But the effects of the epidemic extend far beyond their medical and economic costs to shape the very ways we organize our individual and collective lives.

It is not clear what an "ordinary" epidemic would be. No epidemic seems ordinary to those who experience it. The AIDS epidemic has invoked comparison with many epidemics of the past. Most commonly, the bubonic plague (the Black Death) that devastated Europe in the fourteenth century is recalled: between 1348 and 1350, some 20 million people, one-third of the population of Europe, died. (Additional tends of millions had died in Asia during the preceding decade [McNeil, 1976].) This epidemic had unquestionable impacts. Historians attribute to it, at least in part, the emergence of nation states, the rise of mercantile economies, and the religious movements that led to the Reformation (Campbell, 1931; McNeil 1976; Tuchman, 1978). As Anna Campbell (1931) noted, the Black Death "changed the minds of men" bringing new ways of understanding God, the meaning of death, the place of tradition, and the role of authority in religious and social life. Changes in the collective mind of a society might be the most profound of all impacts, for the new ideas generated by a major social tragedy can propel institutional change and outlast immediate changes to affect lives far in the future. Difficult though it might be to predict the future import of the present impact of the HIV/AIDS epidemic, one should not shrink from the task, especially when one must plan for that future.

AIDS has been compared with other epidemics, too: the resurgence of bubonic plague in England in the mid-seventeenth century, the cholera epidemics of the nineteenth century, the venereal disease epidemics of the sixteenth century and the early twentieth centuries, and the polio epidemics of the twentieth century (Brandt, 1988; Risse, 1988; Slack, 1988). AIDS has its analogies to each of these epidemics—number of deaths, methods of prevention, stigmatization of sufferers and presumed carriers, and responses of authorities—all can be compared in general or in detail. The comparisons are often illuminating, but sometimes misleading (Fee and Fox, 1988). It can be said with some assurance, however, that none of the historical epidemics was "ordinary." Each had impacts that struck its sufferers and subsequent commentators as "extraordinary."

The comparison with epidemics of the past invokes the features that are remembered about those plagues. They have, in this respect, had an impact on history or, as Campbell wrote (1931), "on the minds of men." They also left social institutions that sometimes affect present-day thinking about the AIDS epidemic: cholera, for example, left a public health approach to epidemic disease that stressed quarantine; venereal diseases gave rise to the public health approach of contact tracing. These established public health practices have had to be reconsidered in the current epidemic. Many of the prominent, even dramatic impacts of past epidemics, however, have so melded into the social fabric that people are often astonished to hear of them today, and some, interesting though they be, seem of little relevance to the current problem. For example, to attribute the existence of Protestant Christianity to the effects of the Black Death on religious ideas and sentiments has little influence on the ways in which people today think about religion or about epidemics. Similarly, to attribute the existence of Canada as an independent nation to the fact that British troops had been vaccinated against smallpox before the Battle of Quebec, but American troops were decimated by the disease, is certainly to point to an effect of epidemic and, indeed, an impact. Yet that impact has been of little relevance to subsequent citizens and governments, except that "some Canadians to this day worship smallpox as the deliverer from United States citizenship" (Foege, 1988:332).

Many features of epidemics are no longer remembered and have left little imprint on the societies that they ravished for a time. Indeed, one of the greatest of epidemics, the influenza of 1918-1920, has been called by its historian "the forgotten epidemic" (Crosby, 1989). Worldwide, perhaps 30 million people died; in the United States, 675,000 people died, most of whom were not the usual victims of influenza (the very old, infants, and children), but men and women in their 20s and 30s. This terrible scourge might have had a great impact, but it passed and left almost no mark on the social institutions and practices of the time. Many people were mourned, but life quickly returned to normal. Even the absence of impact has a lesson for this study: it is possible that many of the effects currently taken as important and lasting will pass or be absorbed into the course of American life and culture. It is not entirely clear how confidently one should accept the words of Milbank editors Nelkin, Willis, and Parris (1991:1, 2):

More than a passing tragedy, it [AIDS] will have long-term, broad-ranging effects on personal relationships, social institutions, and cultural configurations … AIDS will reshape many aspects of society, its norms and values, its interpersonal relationships and its cultural representations … the future will be different from both the past and the present.

Our report suggests that, in some respects, the AIDS epidemic may be more like the influenza of 1918 than the bubonic plague of 1348: many of its most striking features will be absorbed in the flow of American life, but, hidden beneath the surface, its worst effects will continue to devastate the lives and cultures of certain communities.

General Findings And Conclusions

Historically, certain epidemics have done great damage to social institutions: the Black Death in a 3-year sweep through Europe wiped out enough laborers to cause a major restructuring of the economy of the continent. The HIV/AIDS epidemic, although often compared to the Black Death, has not affected U.S. social institutions to any such extent. Although it had by the end of 1991 infected perhaps 1 million people, brought devastating sickness to 206,392, and death to 133,233, it had not significantly altered the structures or directions of the social institutions that we studied. Many of the responses have been ad hoc and may be reversed when pressures subside. Others may be more lasting, but only because they reinforced or accelerated changes already latent or budding within the institutions.

It is the panel's opinion that the limited responsiveness of institutions can in part be explained because the absolute numbers of the epidemic, relative to the U.S. population, are not overwhelming, and because U.S. social institutions are strong, complex, and resilient. However, we believe that another major reason for this limited response is the concentration of the epidemic in socially marginalized groups. The convergence of evidence shows that the HIV/AIDS epidemic is settling into spatially and socially isolated groups and possibly becoming endemic within them. Many observers have recently commented that, instead of spreading out to the broad American population, as was once feared, HIV is concentrating in pools of persons who are also caught in the "synergism of plagues" (see Wallace, 1988): poverty, poor health and lack of health care, inadequate education, joblessness, hopelessness, and social disintegration converge to ravage personal and social life. These coexisting conditions foster and aggravate HIV infection and AIDS. Our study of New York City (see Chapter 9) illustrates this dramatically for one epicenter of the epidemic. We believe that the patterns shown there are repeated throughout the country: many geographical areas and strata of the population are virtually untouched by the epidemic and probably never will be; certain confined areas and populations have been devastated and are likely to continue to be.

This epidemiological direction reveals the disconcerting implications of our major conclusion. The institutions that we studied are particularly weak at those points at which the epidemic is likely to be most destructive. For example, the health care system, which responded to the appearance of a new disease with some alacrity, is weakest organizationally and economically in those places where the affected populations are concentrated. The problems of caring for those who are infected are magnified by the particular configuration of the U.S. health care system, which emphasizes to a greater extent than other developed countries private insurance and ability-to-pay criteria. Providers, hospitals, and public health mechanisms can and have responded to a flood of patients with AIDS, but those responses were most successful where health care was better organized and financed and where the populations to be served had sufficient knowledge to understand the disease and its modes of transmission and were capable of organizing themselves in ways that supported and supplemented the health care system.

Thus, our most general conclusion about the epidemic is that its impact has hit institutions hardest where they are weakest: serving the most disadvantaged people in U.S. society. Predictions of the imminent collapse of the health care system due to the epidemic, for example, now look shrill, but, conversely, hopes that the epidemic would force the country toward more rational and equitable reform of the system now also seem unrealistic. In the panel's judgment, the HIV/AIDS epidemic has effected many transient changes in the institutions that we studied and relatively few changes that we expect to be permanent. Among the more permanent, however, two are particularly noteworthy.

First, the institutions of public health, of health care delivery, and of scientific research have become more responsive to cooperation and collaboration with "outsiders." Policies and practices have been modified in these three institutions under pressure from and in collaboration with those who are affected by the epidemic and their advocates. Many of these changes are positive and will contribute to the efficiency and efficacy of the institutions. Similarly, volunteer organizations stimulated by the challenge of the epidemic have discovered ways not only of supplying help where extant institutions were lacking, they have also influenced the policies and practices of those institutions.

Second, even in institutions with very defined purposes and strong constraints—institutions as different as religious groups and correctional agencies—the response to the epidemic has reflected awareness of the scientific realities, as well as the social implications, of HIV/AIDS. Traditionally based doctrinal constraints in the case of religious groups and the stringent requirements of civil punishment in the case of correctional agencies are powerful forces that could and did dictate rigid and narrow response. Yet, powerful as those forces were, they did not negate more reflective responses that contributed to containment of the epidemic and respected the rights of individuals. We are concerned, however, that as the epidemic strikes with greater force in socially and economically deprived communities, the directions toward more communal involvement and respect for civil and personal liberties might be constricted and diverted.

The panel believes that a failure by scientists and policy makers to appreciate the interaction between social, economic, and cultural conditions and the propagation of HIV/AIDS disease has often led to public misunderstanding and policy mistakes about the epidemic. Although in the beginning of the epidemic, gay life and behavior were certainly at the center of attention, even then they were noted primarily as "modes of transmission" and not as social contexts in which the disease had particular meanings around which strong forces for care, prevention, and political action could rally. Similarly, intravenous drug use was understood as a social behavior that could transmit infection, but its place in a matrix of social, cultural, and economic conditions was ignored.

A constant theme of this report and of the AIDS literature is the stigma, discrimination, and inequalities of the AIDS epidemic. At its outset, HIV disease settled among socially disvalued groups, and as the epidemic has progressed, AIDS has increasingly been an affliction of people who have little economic, political, and social power. In this sense, AIDS is an undemocratic affliction. In "democratic epidemics" (Arras, 1988), communicable illnesses cut across class, racial, and ethnic lines and threatens the community at large. In traditional societies with limited medical knowledge and technology, epidemics fall on most, if not all, of the people. In the modern world, particularly in industrial societies, inequalities in morbidity and mortality are often more social than biological phenomena. With HIV/AIDS, the concentration of the epidemic from its beginnings was among those who were, for a variety of reasons, members of marginalized social groups. In this case, the biology of viral transmission matched existing social inequalities and resulted in an unequal concentration of HIV/AIDS in certain regions and among certain populations (see Grmek, 1990). This pattern has created tension between the social and geographical localization of the epidemic and the need to mobilize resources to deal with the epidemic from among individuals, groups, and institutions that are removed from the social groups that are at the epicenter of the epidemic. As the epidemic becomes endemic in already deprived and segregated populations, this tension will be intensified.

If the current pattern of the epidemic holds, U.S. society at large will have been able to wait out the primary impact of the epidemic even though the crisis period will have stretched out over 15 years. HIV/AIDS will "disappear," not because, like smallpox, it has been eliminated, but because those who continue to be affected by it are socially invisible, beyond the sight and attention of the majority population.

Specific Findings And Conclusions

Technical Note

It is instructive to review briefly the procedures for estimating HIV incidence and prevalence to gain an understanding of the basis for the current uncertainty regarding the size of the epidemic and its prevalence among specific population groups. Given that there are no nationally representative seroprevalence surveys, the incidence and prevalence of HIV infection must be inferred from the reported cases of AIDS. New AIDS cases at time t result from infections acquired over a considerable period before that time. Suppose that the incubation distribution of duration from infection to AIDS is denoted by I(d) and that the number of new HIV infections at time t is denoted by H(t). Then the number of new AIDS cases at time t, A(t), is given by the equation

where ω is the maximum incubation period. Although this equation may appear complicated to those without training in mathematics, the concept underlying it is quite simple: the number of new AIDS cases in the current year, for example, is the sum of the number of new infections last year times the fraction of infections that are manifested in AIDS 1 year later, the number of new infections 2 years ago times the fraction of infections that are manifested in AIDS 2 years later, and so forth.

What one observes, of course, is only the time series A(t). One needs to invert the equation to estimate H(t), the time series of new infections. It is well known that without knowledge of I(d)—the fraction of infections that progress to AIDS at each duration d since infection—there is no unique solution for H(t); in fact, there are infinitely many combinations of H(t) and I(t) that satisfy equation (1). Therefore, to make progress, one must obtain external knowledge of I(d). Estimates of the incubation distribution can be derived only from individuals whose date of infection is known. The most reliable data come from cohort studies of hemophiliacs and homosexual men (Brookmeyer, 1991). Note that these data provide information only on the waiting time from seroconversion to AIDS. The time from infection to seroconversion is assumed to be from 3 to 6 months, although seroconversion in some individuals is apparently much longer. The result is that if I(d) is taken from these studies, then the resulting estimates of H(t) pertain to seroconversions, not infections. In any event, one clear source of uncertainty in the estimates of H(t) is the degree to which the I(d) distribution obtained from either of these sources is applicable to the larger population of infected persons.

Assume that I(d) is known. A very simple model will illustrate both the principles and the problems involved in recovering H(t) through a procedure known as back-projection. Suppose it is known that of all HIV infections acquired in year t, 20 percent will progress to AIDS in year t + 2, 30 percent in year t + 3, and 50 percent in year t + 4. Suppose that 1980 is the first year in which AIDS cases appear. It follows that these can have resulted only from infections in 1978, since if there had been infections in 1977, there would have been AIDS cases in 1979. Suppose that one observes 100 AIDS cases in 1980, 400 in 1981, and 1,000 in 1982. One can then solve the following three equations for the three unknowns,

to obtain H₇₈ = 500, H₇₉ = 1,250, and H₈₀ = 1,875. Thus, the prevalence of HIV infection in 1978 is 500, in 1979 is 1,750 (500 + 1,250) (minus any who have died of AIDS), and in 1980 is 3,625 (500 + 1,250 + 1,875) (minus any who have died of AIDS). Note that even at the end of 1982, one cannot obtain any estimate of the number of new infections in 1981 or 1982 since there is at least a 2-year delay between infection and AIDS. In order to get an estimate of H₈₁ and H₈₂ before the end of 1984, one must forecast A₈₃ and A₈₄ by extrapolating the prior time series of new AIDS cases or directly forecast H₈₁ and H₈₂ by extrapolating the estimated prior time series of new HIV infections.

This simple model can be used to identify three sources of uncertainty in the estimates of HIV incidence and prevalence (Centers for Disease Control, 1990):

(1)

Estimates of new HIV infections for very recent periods are less likely to be accurate than estimates for the more distant past.

(2)

It is likely that the incubation distribution I(d) has changed over time as new therapies have been introduced for people who are HIV positive but are asymptomatic. In addition, changes over time in the stage of infection at which infected people are diagnosed would include changes in I(d). Furthermore, there is uncertainty about whether the I(d) distributions obtained from cohort studies of homosexual men or hemophiliacs accurately reflect the I(d) distribution for all infected people.

(3)

The time series of CDC counts of AIDS cases is not the time series of new AIDS cases, for two reasons. First, the case definition of AIDS has been expanded twice (in 1985 and 1987); consequently, the growth in AIDS cases is exaggerated. Second, there are delays from the time of diagnosis to the time a case is reported to the CDC. These delays would result in counts of diagnosed cases that are increasingly less complete as one moves from the past to the present. Some cases are never reported; these can be regarded as cases with infinite delays in reporting. The CDC estimates that its surveillance system identifies 70-90 percent of HIV-infection-related deaths and that it therefore provides a minimum estimate of HIV-infection-related mortality (Centers for Disease Control, 1991). Consequently, CDC counts must first be corrected for the changes in the case definition and for reporting delays before they can be used. Reporting delays may have changed over time, imparting further uncertainty into the corrected series, A_t (Harris, 1990).

References

• Arras, J.D. (1988) The fragile web of responsibility: AIDS and the duty to treat. Hastings Center Report 8(Suppl.):10-20. [PubMed: 3391759]
• Brandt, A.M. (1988) AIDS and metaphor: toward the social meaning of epidemic disease. Social Research 55:413-432. [PubMed: 11650269]
• Brookmeyer, R. (1991) Reconstruction and future trends of the AIDS epidemic in the United States. Science 253:37-42. [PubMed: 2063206]
• Campbell, A.M. (1931) The Black Death and Men of Learning. New York: Columbia University Press.
• Centers for Disease Control (CDC) (1990) HIV prevalence and AIDS case projections for the United States: report based on a workshop. Morbidity and Mortality Weekly Report 39:(RR-16):1-31. [PubMed: 1979423]
• Centers for Disease Control (CDC) (1991) Mortality attributable to HIV infection/AIDS—United States, 1981-1990. Morbidity and Mortality Weekly Report 40:41-46. [PubMed: 1898757]
• Centers for Disease Control (CDC) (1992) HIV/AIDS Surveillance Report. Atlanta, Ga.: Centers for Disease Control.
• Crosby, A.W. (1989) Epidemic and Peace, 1918: America's Forgotten Pandemic. New York: Cambridge University Press.
• Fee, E., and D. Fox (1988) AIDS: The Burdens of History. Berkeley, Calif.: University of California Press.
• Foege, W.H. (1988) Plagues: perceptions of risk and social responses. Social Research 55:331-342.
• Grmek, M.D. (1990) The History of AIDS: Emergence and Origin of a Modern Pandemic. Princeton, N.J.: Princeton University Press.
• Harris, J. (1990) Reporting delays and the incidence of AIDS. Journal of the American Statistical Association 85:915-924.
• McNeil, W.H. (1976) Plagues and Peoples. New York: Doubleday.
• Miller, H.G., editor; , C.F. Turner, editor; , and L.E. Moses, editor. , eds. (1990) AIDS: The Second Decade. Committee on AIDS Research and the Behavioral, Social, and Statistical Sciences, Commission on Behavior and Social Sciences and Education, National Research Council. Washington, D.C.: National Academy Press. [PubMed: 25144007]
• Nelkin, D., D.P. Willis, and S.V. Parris (1991) Introduction. In D. Nelkin, editor; , D.P. Willis, editor; , and S.V. Parris, editor. , eds., A Disease of Society: Cultural Responses to AIDS. New York: Cambridge University Press.
• Risse, G.B. (1988) Epidemics and history: ecological perspectives and social responses. In E. Fee, editor; and D.M. Fox, editor. , eds., AIDS: The Burdens of History. Berkeley, Calif.: University of California Press.
• Slack, P. (1988) Responses to plague in early modern Europe: the implications of public health. Social Research 55:433-453. [PubMed: 11650270]
• Tuchman, B. (1978) A Distant Mirror. The Calamitous 14th Century. New York: Knopf.
• Turner, C.F., editor; , H.G. Miller, editor; , and L.E. Moses, editor. , eds. (1989) AIDS: Sexual Behavior and Intravenous Drug Use. Committee on AIDS Research and the Behavioral, Social, and Statistical Sciences, Commission on Behavioral and Social Sciences and Education, National Research Council. Washington, D.C.: National Academy Press. [PubMed: 25032322]
• Wallace, R. (1988) A synergism of plagues: ''planned shrinkage," contagious housing destruction and AIDS in the Bronx. Environmental Research 47:1-33. [PubMed: 3168963]