It has now been fifteen years since the first description of AIDS was published in the medical literature. Those fifteen years have taught us many things, but above all else they have taught us that HIV disease is here to stay. Even if one or more of the HIV vaccines now in development pass the rigors of clinical trials and become available for general use, they are unlikely to be 100% effective, to provide sterilizing immunity, or to reach 100% of the population -- especially in the parts of the world that need them most. Thus far, the search for an effective vaccine has produced no truly promising candidate, and there is little expectation, at this point, that a universally effective preventive vaccine will be developed in the next decade.
This disheartening news means that, for the foreseeable future anyway, the primary tools of HIV prevention will be 1) educational efforts intended to promote social and behavioral changes, and 2) biomedical interventions such as virucides and improved barrier devices (see "New female condom provides barrier against HIV transmission"). Until such time as partially or wholly protective vaccines are widely available, healthcare professionals will have to depend upon a combination of these educational efforts and biomedical interventions to mount -- and maintain -- effective HIV prevention programs in both developed and developing countries (2, 3).
The rate of new HIV infections is increasing rapidly in this country among adolescents and young adults, ethnic minorities, and the urban poor. While male-to-male sex and injection drug use remain the most prominent modes of transmission, male-to-female and, less frequently, female-to-male heterosexual transmission -- always the major routes of transmission in the developing world -- are on the rise in the United States (4). In 1996, five chilling facts about the epidemiology of HIV transmission drive our prevention priorities:
AIDS is the leading cause of death among all Americans aged 25 to 44, irrespective of race or gender.
Half of all new HIV infections occur in persons younger than 25, and one fourth occur in persons younger than 22.
Heterosexual transmission of HIV infection, especially to women, is rising exponentially.
HIV disease disproportionately affects the urban poor, especially those who belong to ethnic minorities.
HIV remains present -- and potent -- among the first groups affected by the disease: gay and bisexual men and injection drug users (IDUs) (5).
Every healthcare professional who sees sexually active patients can be an effective agent in the effort to reduce transmission of HIV. To have a positive impact, clinicians should incorporate seven essential prevention practices into their medical practice (Table 1). First, clinicians should screen all patients for HIV risk, just as they screen patients for other behavioral risks such as smoking and alcohol consumption.
Half of the deaths that occur in this country each year are the result of risky behavior: drinking and smoking, automobile and firearm accidents, unwise dietary choices and unprotected sex. Recognizing that certain behaviors place patients at higher risk of premature death, clinicians have incorporated many important aspects of disease prevention and health promotion into their daily medical practice. A recent survey found that an overwhelming majority of office-based practitioners -- 94% -- "usually" or "always" ask their adult patients about cigarette smoking, and 84% ask about the use of alcohol. However, practitioners are more reticent when it comes to asking about HIV risk behaviors. Only 49% ask about the use of illicit drugs, 31% ask about condom use, 27% ask about sexual orientation, and only 22% ask how many sexual partners a patient has (6).
Screening for HIV risk need not be laborious or intensive. One-on-one AIDS prevention education is not required for each and every patient, because HIV risk is not spread evenly across the population. The office-based practitioner should focus on learning which patients are at high risk of infection, and prevention efforts should be concentrated on these patients. A simple set of screening questions (Table 2) will help clinicians identify most of the individuals who need intensive HIV education.
Any patient who is identified as being at risk of infection should be encouraged to agree to be tested for the presence of antibodies to HIV -- a safe, simple, sure, and relatively inexpensive means of establishing whether an individual has been exposed to the virus. Women who are pregnant or are contemplating pregnancy should receive particular encouragement to undergo HIV testing, because the administration of antiretroviral agents during gestation and parturition can reduce by at least 67% the likelihood that the child will be born infected (see "AZT Diminishes Transmission of HIV-1 from Mothers to Their Infants," Vol. 1, No. 2).
It is now possible, in every state in the union, for individuals who want to know their HIV status to receive that information anonymously and with no breach of confidentiality. Professional counseling is also available nationwide, and it is offered both before and after testing occurs. The F.D.A. recently approved a testing device that insures complete anonymity. Individuals are now able to purchase the testing kit at a pharmacy, take a blood sample, send it to a designated laboratory and, after an interval of one week, call to receive the results. Specially trained personnel will offer an unlimited number of telephone counseling sessions to all those who test HIV-positive (see "New HIV testing program guarantees anonymity, offers telephone counseling").
Effective prevention of HIV transmission is possible only when physicians are willing to talk openly, frankly, and non-judgmentally with patients about sexuality and illicit drug use (7). Nothing in the training most physicians receive in medical school prepares them adequately for this aspect of their professional duties, and it is little wonder that many practitioners feel uncomfortable discussing sexual practices and recreational drug use with their patients.
Clinicians who are familiar with the current statistics on HIV transmission rates and the current guidelines on safer sex practices will find themselves better equipped, and more relaxed, in discussions with at-risk patients. "What to Tell Your Patients About Safer Sex and Condom Use," in this issue, provides readers with that information.
All sexually active individuals run some risk, however small, of acquiring HIV, and the more partners they have, and the less they know about those partners, the higher the risk. Therefore, all patients who report having multiple sexual contacts should be urged to reduce the number of those contacts. They should also be encouraged to use condoms, male or female, with all of their sexual partners. The pull out and save feature in this issue explains, in straightforward lay language, the correct use of a condom. Practitioners may want to photocopy these instructions and hand them out to at-risk patients.
While most clinicians -- and most of their patients -- recognize that unprotected vaginal and anal intercourse are high-risk sexual activities, on a par with sharing contaminated syringes, there is no such consensus regarding the riskiness of oral sex. What we do know is that the risk of HIV transmission through oral sex is very, very low. But it is not zero, as my colleague Dr. Joshua Schechtel explains in "What to Tell Your Patients About Safer Sex and Condom Use."
Because the risk of transmitting HIV through oral-genital contact is so minimal, the San Francisco-based Gay and Lesbian Medical Association recently declared that this form of sexual activity should be regarded as low-risk, at least among men who have sex with men (see "National association of gay physicians classifies oral sex as 'low-risk' but not 'no-risk'"). The organization's rationale, in taking this controversial position, is that sexual expression is a fundamental aspect of human nature, and when sexual activity is restricted to zero-risk activities, frustration builds to the point of sexual rebelliousness, which can lead to distinctly unsafe sexual behaviors. It is the group's hope that, by offering gay men some sexual leeway, a trade-off will be effected: more very-low-risk activity will occur, but as a result fewer breakouts of very-high-risk activity will take place.
Because all responsible educators, the Gay and Lesbian Medical Association included, accept that oral-genital contact -- especially when vaginal secretions or semen are taken into the mouth -- involves some degree of risk, patients should be told that this activity is not without risk. And all patients who engage in oral sex with partners whose HIV status is unknown should be advised to use condoms.
Knowledge alone is rarely sufficient to motivate patients to change risky behaviors, and individuals who persistently engage in high-risk behavior do so for a variety of reasons. In order to offer such patients constructive advice, the clinician must make an effort to understand why patients are taking these risks. Effective counsel is possible only when those reasons are appreciated.
The easiest way for a clinician to discover why a particular patient is engaging in unsafe activities is to ask that individual about the circumstances surrounding the last risky act. Inebriation is a common explanation; so is the unavailability of condoms, clean needles, or both. When asked why they did not insist that their sexual partner use a condom, women often cite concerns about their physical safety, and both women and gay men say that they feel they lack the power to demand that condoms be used. Referring such individuals to agencies that specialize in AIDS education and risk reduction can go a long way toward helping these patients reduce the risks that they will acquire, or transmit, HIV infection (see the pull out and save feature, "A State-by-State Guide to AIDS Education Services" in Vol. 1, No. 4).
It is the rare individual who deliberately infects another with HIV, but every new infection is the result of the unwitting or unintentional transmission of the virus from an infected to an uninfected individual. Encouraging patients whose risk profile suggests that they may be HIV-positive to undergo antibody testing is a crucial first step in reducing transmission rates, and in this regard the healthcare system in the United States has been remiss. In Australia, 90% of gay men know their antibody status, versus 65% of gay men, 50% of IDUs, and 30% of heterosexuals with multiple sexual partners in this country (9). Physicians should recommend antibody testing for all patients who may have been exposed to HIV, no matter how remote the risk, since asymptomatic infection in low-risk populations is an important reservoir of virus.
To date, few HIV prevention programs have specifically appealed to the HIV-positive individual's sense of moral obligation and his personal responsibility to avoid infecting others, but this may be a fruitful approach to prevention (9). The reluctance to focus on people who know they are seropositive, and to invoke morality, has stemmed from legitimate concerns about discrimination and commendable efforts to avoid stigmatization, but it has often diffused prevention efforts and confused target audiences.
There is nothing which says that individuals who know they are HIV-positive are any less responsive to appeals to personal responsibility than, say, individuals who know they are carriers of hepatitis B or tuberculosis, and these models may be usefully invoked by clinicians when they counsel seropositive patients. Indeed, the evidence suggests that such appeals can be effective in reducing transmission rates, even among the ill-educated, indigent, and homeless (see "Effective Intervention Is Possible in the Inner City").
We now know that during acute HIV infection -- before antibodies to HIV develop -- infectivity is 50- to 2,000-fold greater than it is during the ensuing period of symptom-free disease. As a result, the period of primary HIV infection is one of particular danger: standard tests will not detect the presence of antibody, and individuals often feel no ill effects of acute infection.
Acute retroviral syndrome occurs in approximately 50% of infected individuals (range: 30% to 80%), and those flu-like symptoms generally resolve within a matter of days. Few infected individuals recognize these symptoms for what they are, and so they rarely seek medical attention. Moreover, when they do it takes an alert and informed healthcare professional to diagnose acute retroviral syndrome. The likelihood of transmission is very high during this period, and identifying recently-infected individuals may therefore be an especially effective means of reducing new infections.
Aggressive treatment of HIV infection itself is another effective way of reducing transmission rates. Therapies that combine a protease inhibitor with one or more nucleoside analogs have been shown to reduce viral burden to undetectable levels in a significant majority of treated patients (see "Update: The Protease Inhibitors," Vol. 2, No. 3). Drug regimens that are effective in reducing the amount of virus circulating in the blood will also reduce the amount of virus in semen and vaginal secretions -- and these reductions will result in concomitant reductions in the likelihood that HIV will be transmitted to a sexual partner or during the sharing of syringes and needles.
HIV prevention is one of the medical success stories of the past decade. Although we still record roughly 40,000 new cases of HIV infection every year -- many of them long-standing infections that have only recently been diagnosed -- successful intervention strategies have been developed and implemented. These programs include a number that were specifically devised to reduce transmission rates among men who have sex with men and among IDUs (Table 3).
A common feature of these successful interventions is that they are based on the principle of harm reduction. The goal of these programs is not to achieve perfection, or reduce transmission rates to zero; it is to move individuals in the direction of safer, if not completely safe, practices.
The indisputable effectiveness of these programs dispels the three major myths of HIV prevention: 1) that condoms don't work; 2) that early sex education increases promiscuity; and 3) that providing clean syringes and needles to drug users increases drug use. All three of these propositions are absolutely false (3, 10).
We could halve the number of new HIV infections in this country if these programs were implemented completely and without reservation. The proof of this assertion comes from many sources but particularly from cities where the prevalence of HIV infection among drug users has remained very low (< 5%) (11). Those cities include: Glasgow, Scotland; Lund, Sweden; Sydney, Australia; Liverpool, England; Toronto, Canada; and, in the United States, Tacoma, Washington. All of these cities began HIV prevention programs early, initiated programs that made clean syringes widely available, and developed outreach programs to bring IDUs into the system. Cities that did not implement such strategies early enough have seen prevalence jump to 20% or more in a single year.
Australia and New Zealand both undertook sound HIV prevention practices during the early stages of the epidemic, and as a result the spread of new infections has been largely contained in those countries. Through aggressive public prevention programs Australia has retarded the spread of HIV among IDUs and heterosexuals as well as gay males. As a direct result of these efforts, the number of newly diagnosed infections in men who have sex with men dropped from a peak of 2,284 in 1987 to 772 in 1994. A marked decline in unsafe sexual practices among homosexual men has also been documented (12). In both Australia and New Zealand the epidemic has been halted in IDUs; there has been no increase in HIV infection rates or AIDS among members of this risk group in the last five years.
We could achieve the same results in this country. Doing so would require adequate funding, aggressive public health measures, and cooperation among governmental and non-governmental agencies, advocacy groups, healthcare professionals, and concerned individuals. Absent such funding, such programs, and such cooperation, physicians should work by themselves -- and through medical societies and other professional organizations -- to do what is right, and necessary, to prevent the spread of this terrible disease.
Gerbert B, Maguire BT, Coates TJ. Are patients talking to physicians about AIDS? Am J Pub Health 1990; 80: 467-8.
Stryker J, Coates TJ, Haynes-Sanstad K, et al. HIV prevention: Looking back, looking ahead. JAMA 1995; 273: 1143-8.
Coates TJ, Aggleton P, Gutzwiller F, et al. HIV prevention in developed countries. Lancet (in press).
Holmberg SD. The estimated prevalence and incidence of HIV in 96 large U.S. metropolitan areas. Am J Pub Health 1996; 86: 642-54.
First 500,000 cases -- United States, 1995. MMWR 1995; 44: 849-53.
Centers for Disease Control and Prevention. HIV prevention practices of primary care physicians -- United States, 1992. MMWR 1994; 42: 988-92.
Hearst N. AIDS risk assessment in primary care. J Am Board Fam Prac 1994; 7: 44-8.
Berrios DC, Hearts N, Coates TJ, et al. HIV antibody testing among those at risk for infection. The National AIDS Behavior Surveys. JAMA 1993; 270: 1576-80.
Bayer R. AIDS prevention -- sexual ethics and responsibility. N Engl J Med 1996; 334: 1540-2.
Francis D, Chin J. The prevention of acquired immunodeficiency syndrome in the United States. JAMA 1987; 257: 1357-66.
Des Jarlais DC, Hagan H, Friedman SR, et al. Maintaining low HIV seroprevalence in populations of injecting drug users. JAMA 1995; 274: 1226-31.
National Centre in HIV Epidemiology and Clinical Research. Australian HIV Surveillance Report 1995; 11 (No. 3).
Thomas J. Coates, Ph.D., is Professor of Medicine at UCSF and Director of the Center for AIDS Prevention Studies, San Francisco, CA.