Primary Care Physicians and HIV Prevention

Persons continue to become infected with HIV at an alarming rate across the United States, where it is estimated that 40 000 to 80 000 new infections are diagnosed each year [2]. A review of the literature [3] suggests that the increasing rate is not caused by a lack of effective ideas about HIV and AIDS prevention, but rather by a failure of political will and a lack of material resources. Prohibitions on programs that "promote homosexuality," make clean needles available to injection-drug users, or make condoms available to sexually active teenagers, combined with budgets that fail to keep pace with infection rates, have caused a lethal national paralysis in efforts to prevent HIV.

Expert testimony not only documents the need for increased prevention efforts but describes specific programs that work and successful efforts to change community norms. It also suggests that a comprehensive national agenda is required to stem the epidemic [4]. Surprisingly, experts agree unanimously about the key components of such an agenda. To avoid becoming infected with HIV, persons must permanently change some of their behaviors; unfortunately, the margin for error or lapse is small. Programs must provide meaningful information, build new social skills, and connect questions of individual behavior to larger concerns for community survival [5]. A national agenda must incorporate many strategies that target the different vulnerable groups—adolescents, homosexual and bisexual men and women, injection-drug users, members of racial and ethnic minorities, and incarcerated persons—who respond best to uniquely crafted health messages delivered in distinctive styles and contexts [6].

With respect to clinically based prevention efforts, official bodies from the U.S. Department of Health and Human Services, the U.S. Preventive Services Task Force of the American Medical Association, and the American College of Physicians have proclaimed a central role for primary care physicians in preventing HIV infection [1, 7-9]. Research supports this call. Physicians are cited by the public as the most trusted source of health information [10, 11], and patients heed their health-related messages [12, 13]. Most patients are willing to discuss HIV and AIDS with their physicians [14], and teenagers express a clear preference to learn about the disease from their physicians [15]. Moreover, the physician has many opportunities to deliver HIV and AIDS prevention messages. The National Health Interview Survey [16] documents the frequency with which a broad spectrum of persons in the United States annually visit physicians' offices: White adults average 3.66 visits, and African-American adults average 2.80 visits. More than one half of all HIV antibody testing occurs in a physician's office or in a hospital [17]. Physicians can play a vital role in counseling patients. Recent research, such as the finding that zidovudine can dramatically decrease the risk for HIV transmission from mother to fetus [18], should be explained to women during office visits. In addition, preventive measures must be addressed specifically with persons who are already infected [19, 20].

Despite these opportunities, the best intentions of physicians, and the willingness of patients to learn from physicians, actual practice incorporates few clinical prevention strategies. One national survey [21] showed that counseling and advice about HIV transmission were given in fewer than 1% of patient visits to primary care physicians. When another survey queried physicians about the kinds of questions that they would "usually" or "always" ask new adult patients, it showed that 49% of primary care physicians ask about sexually transmitted diseases, 31% ask about condom use, 27% ask about sexual orientation, and 22% ask about numbers of sex partners. In contrast, 94% indicated that they "usually" or "always" ask new adult patients about cigarette smoking. Studies conducted in New York City underscore the strong reluctance of many physicians to undertake HIV and AIDS prevention activities, even in areas with high rates of infection [22-24].

Physicians express greater apprehension than patients do about discussing HIV. One quarter of physicians surveyed by the Centers for Disease Control and Prevention [25] believed that their patients would be offended by questions about their sexual behaviors, whereas a corresponding survey of patients indicated that nearly all would not object to discussing AIDS. Confirming the extent of physician discomfort, this study showed that only 15% of patients surveyed had discussed AIDS with their physicians during the previous 5 years and that 72% of these conversations were initiated by the patients.

Barriers to Clinical Prevention

Physicians' lack of involvement in clinical HIV prevention reflects their general lack of involvement in any prevention activities. Yet, the great disparity between other prevention efforts and those specifically related to HIV indicates that additional factors keep physicians from directly addressing the issue with their patients. A review of the literature suggests that there are four major barriers to providing clinical prevention information: 1) narrow conceptions of medicine and the role of the physician; 2) physicians' discomfort with issues raised by HIV and AIDS and their attitudes toward persons with the virus or the disease; 3) practical constraints of time and resources; and 4) ambiguities inherent in HIV prevention messages. As we outline these barriers, we offer specific recommendations.

Narrow Conceptions of Health Care and the Physician's Role

Our health care system has traditionally given higher priority to the immediate, acute needs of patients than to long-term behaviors that can lead to disease. Physicians have been rewarded for responding to patient-identified complaints rather than for offering services for problems that have not yet developed or have not yet been revealed by the patient [26]. However, this highly focused, technologic perspective on medicine in the United States fails to address the complexities of clinical prevention. Some physicians believe that preventive activities are less valuable than other clinical interventions and that counseling patients will not change their risky behaviors [27-29]. Others may be discouraged by the lack of tangible evidence of the value of prevention efforts. Rather than an immediate improvement in a patient's health or quality of life, the satisfactions of prevention are in better communication, understanding, and the potential for preservation of life.

Many institutions are responding to the challenges of contemporary health care by making radical changes in general medical education [30]. Reform of medical education, like reform of the health care system, could focus attention on the value of clinical prevention [31]. Medical school curricula and health care institutions must support the changing role of the physician and account for the powerful social influences on physician-patient communication. Without such support, physicians may return to focusing only on immediate, quickly identifiable problems. These same approaches should be incorporated into continuing medical education programs offered by universities and professional organizations.

Changing the expectations for clinical prevention efforts may also make the task less discouraging and more accessible for practicing physicians. Emphasizing harm reduction rather than absolute risk elimination will help physicians establish realistic goals [6]. The assumption that only 100% adherence to clinical advice constitutes success or that a single intervention will change life-long patterns of behavior is difficult for physicians and leaves no room for the patient's right to choose, for individual differences, or for development. A risk-reduction model is consistent with approaches to client education developed by genetics counselors that respect patient autonomy and privacy [32]. It is also consistent with the ethics of informed consent, which remind us that prevention services are best offered to the patient rather than mandated, because this acknowledges the patient's right to reject the service or the recommendations of the health care provider [33].

Physicians' Attitudes toward Persons with HIV Infection or AIDS

Although health promotion and disease prevention historically have not been emphasized in medical education in the United States, the task is even more complex when issues such as human sexuality and drug use must be discussed. As in the rest of society, researchers have found unreasonable stigma and prejudicial attitudes associated with AIDS among practicing physicians and medical students [34-36]. Prejudice compromises the care available to persons who have or who are at risk for HIV infection or AIDS. If the virus and disease are characterized as affecting only "other people," then to address prevention is to acknowledge that we are more vulnerable to the disease than we like to admit. Homophobia, along with physicians' lack of knowledge about HIV and AIDS, often compounds a general discomfort in counseling patients about sexual risk factors, and it is associated with the failure to offer preventive education about HIV. Importantly, studies indicate that misdiagnoses occur when cultural and social differences between physicians and patients lead to miscommunication [37, 38]. For example, homosexual persons report that they are more satisfied with their health care and, if they are male, that they are more likely to have been checked for sexually transmitted disease if they are comfortable sharing information about their sexual orientation [39].

For some physicians, the commitment to nonjudgmental clinical prevention may require a considerable shift in practices and values [22, 40, 41]. Educational interventions can improve physician behavior with respect to prevention [42, 43]. Especially relevant for HIV and AIDS prevention are interventions that enhance physicians' skills in taking sexual histories [44, 45]. Attitudes toward treating persons with HIV infection or AIDS have also been improved through education [46]. Most importantly, 87% of physicians believe that professional training could help "increase their comfort in caring for AIDS patients" [17].

How can we improve education in this area? Large-scale studies that document the presence or absence of specific behaviors or attitudes provide little insight into physician decision making. More in-depth studies of physicians' perspectives could show, given the constraints of time and resources, how physicians decide to address certain issues and how they prioritize their efforts. A recent ethnographic study of physicians' practices with respect to domestic violence, which used open-ended, semistructured interviews, is a model of effective qualitative research [47]. It shows how primary care physicians understand domestic violence and clarifies ways in which to address the barriers to effective interventions in the clinical setting. In fields such as elementary and secondary education, qualitative research has enabled educators and policymakers to understand the practitioner's perspective and to design better preparation programs [48, 49]. We also need more research on how patients comprehend prevention information and the messages they take away from clinical encounters.

New problem-based methods can also model the contradictions and conflicts experienced by practicing physicians in ways that are not possible with traditional didactic instruction. Driven by research that documents physicians' concerns and perspectives, these methods can increase a learner's involvement and improve critical-thinking skills. Evaluation of these approaches with respect to their lasting effect on physicians and patients will be important. A recent study using a simulated-patient instructor for role playing and feedback showed that this approach was far superior to the traditional use of mailed educational materials in improving physicians' office-based prevention practices for sexually transmitted diseases [50]. Such efforts must be further studied and expanded.

Practical Constraints of Time and Resources

In many settings, reimbursement for counseling and testing is limited or nonexistent, a factor that decreases the economic efficiency of involving physicians in prevention activities [51]. Many argue that public health strategies, community-based efforts, and the media reach more persons at lower cost. Clinical prevention efforts do not reach those without access to the health care system, but for patients with access, it may be cost-effective to devote a small portion of already-existing patient–physician contact time to this task. The cost of physician silence is far greater than any immediate spending needed to enhance clinical prevention. Although it is unrealistic to expect physicians to change patients' long-term behaviors in these brief encounters, physicians' lack of active involvement sends the wrong message to persons who look to their physicians for guidance on public health matters.

Prevention would be more manageable for the physician and more effective for the patient if it were seen as a matter of health education rather than as a psychological intervention requiring extensive assessment and counseling skills. In this light, physicians' goals could be simplified and limited to 1) helping patients assess their own risks [52]; 2) referring patients for counseling and testing; 3) reinforcing good prevention practices when they exist; 4) determining patients' interest in changing behavior [53]; and 5) offering referrals to patients who want to learn more about HIV and AIDS. Successful clinical prevention would then involve the physician as part of a systems approach that may include other health care professionals in the immediate setting, community-based prevention services, counselors in private practice, and targeted written or computer-based materials and media presentations [54, 55].

Prevention cannot be successfully regulated or mandated without creating systems that predispose, enable, and reinforce physicians' commitment to this kind of work. Studies [26, 29, 56] indicate that practical support and the encouragement of prevention efforts in clinical practice settings can supersede the effects of physicians' attitudes or education. Kottke and associates [57] report that establishing such support systems in a health maintenance organization substantially increased physicians' participation in a smoking cessation campaign. These structures will differ according to setting—private office, public health clinic, municipal hospital, or health maintenance organization—but their efficacy is not in doubt.

Ambiguities in the HIV and AIDS Prevention Message

The final barrier to increasing efforts to prevent HIV infection reflects the ambiguous and sometimes contradictory messages physicians are asked to convey to their patients. Physicians have diverse opinions about the risks of multiple sex partners, oral sex, anal intercourse with condoms, and HIV transmission between women. For example, some data document HIV transmission during oral sex between men, but the interpretation of these data varies greatly [58]. Some providers ignore these data, believing that if they suggest that oral sex is a risk factor for infection, patients will become easily frustrated with safer sex guidelines and disregard them completely. Other providers rate all activities with any risk for HIV transmission equally and thereby confuse patients who rightfully understand that oral sex presents less risk than activities such as unprotected receptive anal intercourse. As a result, patients could become cynical about all physician advice [59].

Similarly, safer sex guidelines have emphasized the importance of limiting the number of sex partners [60]. Yet, this emphasis may detract from a more important need for persons to limit themselves to low-risk activities, despite their number of partners, and to openly discuss their serostatus before engaging in any high-risk activity. For example, a recent report of women in urban health clinics documented that although they may be in single sexual relationships with men they know well, they are still at high risk for contracting HIV infection if their partners continue to have sex outside of the relationship or have a history of drug use [61]. Similarly, new evidence suggests that unprotected anal intercourse between homosexual men is more likely to occur in ongoing relationships than in anonymous sex encounters with multiple partners [62]. Therefore, prevention messages for patients at increased risk must be designed with realistic knowledge of the lives of the persons at whom they are targeted.

Some of the more ambiguous areas may be clarified by developing practice guidelines based on available data that could be used to standardize clinical prevention messages in both clinical and community settings. Clinical guidelines are important tools for improving patient care when development, dissemination, and implementation are appropriate [63, 64]. Information on prevention must be shared with patients in ways that are easily understood, and guidelines will help physicians to be more consistent in the messages they deliver.

Medical educators at all levels should use the complexity of HIV prevention to their pedagogic advantage. Through focus groups, we found that discussing the ambiguity of prevention messages provoked lively debate among physicians and led to new understanding of the need for prevention education. Similarly, physicians and other health care providers can use the complexity of prevention messages to educate their patients. Especially with respect to safer sex, patients must choose those practices with which they are most comfortable. In large measure, our primary goal will be met when patients begin to see the need to make those choices.

Improving Clinical Prevention Efforts

Physicians can play an integral role in a national HIV prevention agenda. Guinan and colleagues [65] estimate that the medical cost savings from each prevented case of HIV is $56 000 to $80 000, excluding indirect and direct costs other than medical care. The importance of recognizing this role and of beginning to remove the barriers to these efforts cannot be understated. In addition to these barriers, HIV prevention raises a fundamental set of issues for physicians about professional goals, personal identity, and physician-patient relationships. The model of physician-patient interaction in which the former is characterized as an expert and the latter as a passive recipient of care and in which a simple exchange of information is understood as meaningful communication, is not conducive to discussing behaviors that place a patient at risk for HIV infection. A more appropriate model emphasizes a reciprocal and dynamic search for mutual understanding, in which culture and values are considered. Such an approach is consistent with the aims of many physicians who are looking for better ways to motivate their patients to live healthier lives. Similarly, a clearer articulation of the difficulties and ethical dilemmas involved in prevention would support more responsive and realistic practical preparation for physicians and long-range commitments to prevention. Clinical prevention could then emerge from the shadows of "high-tech" medicine and illness management as one of the most intellectually challenging and socially relevant components of primary care.

Dr. Silin: Bank Street College of Education, 610 West 112th Street, New York, NY 10025.