National Academies Press: OpenBook

HIV Screening of Pregnant Women and Newborns (1991)

Chapter: Appendix A: The Conference: Program and Summary

« Previous: References
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

APPENDIX A
THE CONFERENCE: PROGRAM AND SUMMARY

Appendix A contains, first, the program of the conference, "Prenatal and Newborn Screening for HIV Infection: Opportunities for Prevention and Treatment?" The second part of the appendix is a summary of the conference presentations and discussions. It is not intended to present a consensus of opinion on the range of issues pertinent to the development of perinatal HIV screening policy. Rather, the views expressed here are those of the speakers and do not necessarily correspond to the conclusions and judgments reached by the IOM committee. Because the committee drew on the conference presentations during its deliberations and the preparation of its report, portions of this summary may overlap with material covered in the main body of the report.

The speakers who presented the material on which the summary is based are listed below by individual conference section.

HIV Infection and AIDS in Women and Children: Diagnosis and Treatment —Alfred Saah, Marta Gwinn, James Oleske, and Rhoda Sperling

Principles and Pitfalls of Mass Population Screening—Norman Fost and George Cunningham

Who Should be Screened?—Larry Wissow, Howard Minkoff, and Claire Brindis

Consent and Counseling—Ruth laden, Ann Sunderland, John Arras, and George Annas

HIV Screening Policy Implementation—Hermann Mendez, Christine Grant, and Richard Schwarz

Economic Considerations in Screening for Perinatal HIV Infection—Jesse Green, Peter Arno, and Sara Rosenbaum

Evaluating the Effectiveness of HIV Screening—Milton Weinstein, Renata Kiefer, and Donald Francis

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

The Advocates' Voices: Thoughts on Prenatal and Newborn HIV Screening —Milagros Davila, Dazon Dixon, and Sallie Perryman

HIV Screening: Are Human Rights Threatened?—Larry Gostin, Katherine Franke, and Patricia King

CONFERENCE PROGRAM

Auditorium, National Academy of Sciences

2101 Constitution Avenue, N.W., Washington, D.C.

May 14-15, 1990

Monday, May 14

8:25

Welcome and Opening Remarks

 

- Marie McCormick, Associate Professor of Pediatrics, Joint Program in Neonatology, Harvard Medical School and Chair, IOM Committee on Prenatal and Newborn Screening for HIV Infection

8:30

HIV Infection and AIDS in Women and Children

 

Moderator: Kathleen Nolan, Associate for Medicine, The Hastings Center

 

HIV Infection and AIDS in Women: Magnitude of the Problem

 

- Tedd Ellerbrock, Medical Epidemiologist, Pediatric and Family Studies Section, Epidemiology Branch, Division of HIV/AIDS, Centers for Disease Control

 

HIV Infection in Infants and Children: Magnitude of the Problem

 

- Marta Gwinn, Medical Epidemiologist, Population Studies Section, HIV Seroepidemiology Branch, Division of HIV/AIDS, Centers for Disease Control

 

HIV Diagnostic Technology: How Good Are the Tests?

 

- Alfred Saab, Associate Professor of Epidemiology, Johns Hopkins University School of Hygiene and Public Health

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

 

Medical Management of HIV Infection in Children: How Good Is the Treatment?

 

- James Oleske, Professor of Pediatrics, New Jersey Medical School and Medical Director, Children's Hospital of New Jersey AIDS Program

 

Medical Management of HIV Infection in Pregnancy

 

- Rhoda Sperling, Director of Obstetrics and Gynecology-Infectious Diseases, Mount Sinai Medical Center

9:55

General Discussion

10:45

Principles and Pitfalls of Mass Population Screening

 

- Norman Fost, Professor and Vice Chairman, Department of Pediatrics, University of Wisconsin

11:15

Discussant

 

- George Cunningham, Chief, Genetic Disease Branch, California State Department of Health Services

11:30

General Discussion

11:50

Who Should be Tested? What Could be Achieved?

 

Moderator: James Curran, Director, AIDS Program, Centers for Disease Control

 

Screening Newborns

 

- Larry Wissow, Assistant Professor of Pediatrics, Johns Hopkins University School of Medicine

 

Screening Pregnant Women

 

- Howard Minkoff, Professor and Director, Division of Obstetrics and Maternal-Fetal Medicine, State University of New York Health Science Center at Brooklyn

 

Screening Women of Childbearing Age

 

- Claire Brindis, Co-director, Center for Reproductive Health Policy Research, University of California at San Francisco

12:45

General Discussion

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

2:15

Consent and Counseling

 

Moderator: Peter Selwyn, Assistant Professor of Epidemiology and Social Medicine, Albert Einstein College of Medicine

 

Means and Ends of Informed Consent

 

- Ruth Faden, Director, Program in Law, Ethics, and Health, and Professor of Health Policy and Management, Johns Hopkins University School of Hygiene and Public Health

 

One HIV Counseling Program for Women of Reproductive Age

 

- Ann Sunderland, Social Worker, HIV Perinatal Transmission Study, State University of New York Health Science Center at Brooklyn

 

Directive vs. Nondirective Counseling

 

- John Arras, Philosopher in Residence, Department of Epidemiology and Social Medicine, Albert Einstein College of Medicine

 

- George Annas, Professor of Health Law and Director, Law, Medicine, and Ethics Program, Boston University Schools of Medicine and Public Health

3:25

General Discussion

4:10

HIV Screening Program Implementation: Providing the Services

 

Moderator: Reed Tuckson, Senior Vice President for Programs, March of Dimes

 

Providing Follow-Up Care for Women and Children with HIV Disease

 

- Hermann Mendez, Assistant Professor of Pediatrics, State University of New York Health Science Center at Brooklyn

 

Implementing an HIV Screening Program: New Jersey's Experience

 

- Christine Grant, Deputy Commissioner of Health, New Jersey State Department of Health

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

 

Professional Education and Standards of Care in Obstetrics

 

- Richard Schwarz, Provost and Vice President for Clinical Affairs, State University of New York Health Science Center at Brooklyn

5:05

General Discussion

Tuesday, May 15

8:20

Economic Considerations in Screening for Perinatal HIV Infection

 

Moderator: Molly Coye, Head, Division of Public Health Practice, Department of Health Policy and Management, Johns Hopkins University School of Hygiene and Public Health

 

Current Costs and Services Utilization Associated with Pediatric AIDS

 

- Jesse Green, Director for Health Policy Research, New York University Medical Center

 

Economic Implications of Early Intervention in Women and Children

 

- Peter Arno, Assistant Professor of Epidemiology and Social Medicine, Albert Einstein College of Medicine

 

Financing Strategies for Screening and Follow-Up Care

 

- Sara Rosenbaum, Director of Programs and Policy, Children's Defense Fund

9:15

General Discussion

10:00

Evaluating the Effectiveness of HIV Screening

 

Moderator: Sandy Schwartz, Executive Director, Leonard Davis Institute of Health Economics, University of Pennsylvania

 

Program Evaluation: Process and Outcomes

 

- Donald Francis, Centers for Disease Control Regional AIDS Consultant, California State Department of Health Services

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

 

Cost-Effectiveness Analysis of Prenatal and Newborn HIV

 

Screening: Methodologic Issues and Data Needs

 

- Milton Weinstein, Henry J. Kaiser Professor of Health Policy and Management, Harvard School of Public Health

 

A Cost-Effectiveness Analysis of Prenatal HIV Screening

 

- Renata Kiefer, Fellow, Center for AIDS Prevention Studies, University of California at San Francisco

10:55

General Discussion

11:30

The Advocates' Voices: Thoughts on Prenatal and Newborn HIV Screening

 

Moderator: Janet Mitchell, Chief of Perinatology, Department of Obstetrics and Gynecology, Harlem Hospital Center

 

- Milagros Davila, Coordinator, Prenatal Care Guidance Program, Department of Health Services, San Diego County

 

- Dazon Dixon, Director, Sister Love Women's AIDS Project

 

- Sallie Perryman, Special Assistant to the Director of Policy, New York AIDS Institute

12:30

General Discussion

2:00

HIV Screening: Are Human Rights Threatened?

 

Moderator: Carol Levine, Executive Director, Citizens Commission on AIDS for New York City and Northern New Jersey

 

Confidentiality, Disclosure, and Discrimination

 

- Larry Gostin, Executive Director, American Society of Law and Medicine

 

Discrimination in Access to Reproductive Health Services

 

- Katherine Franke, Supervising Attorney, Fair Housing and Public Accommodations Division, New York City Commission on Human Rights

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

 

Judicial Intervention in Obstetrical Medicine

 

- Patricia King, Professor of Law, Georgetown University Law Center

2:55

General Discussion

3:45

Developing Policy: Moving Toward Consensus

 

Moderator: Neil A. Holtzman, Professor of Pediatrics, Johns Hopkins University School of Medicine

 

- Edward Connor, Associate Director, Division of Allergy, Immunology, and Infectious Diseases, Children's Hospital of New Jersey

 

- Sheldon Landesman, Associate Professor of Medicine and Director, AIDS Study Group, State University of New York Health Science Center at Brooklyn

 

- Kristine Gebbie, Secretary, Department of Health, State of Washington

 

- Ronald Bayer, Associate Professor, Department of Sociomedical Sciences, School of Public Health, Columbia University

5:30

Summary and Closing Remarks

 

- Neil A. Holtzman, Professor of Pediatrics, Johns Hopkins University School of Medicine

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

CONFERENCE SUMMARY PRENATAL AND NEWBORN SCREENING FOR HIV INFECTION: OPPORTUNITIES FOR PREVENTION AND TREATMENT?

HIV INFECTION AND AIDS IN WOMEN AND CHILDREN: DIAGNOSIS AND TREATMENT

Assessment of proposals for HIV screening of pregnant women and newborns requires an examination of thc technology currently available to diagnose HIV infection in women and children. It also requires consideration of the treatment options and possible drug therapy that can be offered to those in whom infection is identified. Part of the challenge of developing rational perinatal HIV screening policy lies in the fact that HIV diagnostic technology and medical therapy are evolving. Therefore, present screening policies need to be flexible to respond to future technological developments.

HIV Diagnostic Technology

The serological tests most commonly used for the diagnosis of HIV infection are the Enzyme-linked Immunosorbent Assay (ELISA) combined with the Western Blot. The ELISA is reliable, relatively inexpensive, easy to perform—and therefore appropriate for screening purposes. The Western Blot is used to confirm the results of the ELISA. The ELISA, which detects antibodies to virus antigens (i.e., HIV viral proteins), is performed first on the patient's serum specimen. If the test is positive, it is repeated on the same blood sample. If positive again, the specimen is referred to as ''repeatedly reactive'' on the ELISA. This test should not be considered (or be reported as) truly positive, however, until the specimen is subjected to a more specific confirmatory test (usually the Western Blot), which is used to validate the ELISA result (i.e., to determine whether the ELISA-reactive specimen is a true or false-positive result). A "repeatedly reactive" ELISA test that is confirmed positive by a Western Blot is generally diagnostic of HIV infection and can be reported as positive.

In some cases, the Western Blot results will be "indeterminate" or considered nondiagnostic of infection. This nonspecificity can be reduced

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

to a minimum by using more robust Western Blot interpretive criteria, such as those currently recommended by the Association of State and Territorial Public Health Lab Directors or the American Red Cross. In many cases, indeterminate Western Blot results can be resolved by additional antibody tests, particularly the new assays that employ recombinant or synthetic peptide HIV antigens. These assays could be inserted in the current ELISA/Western Blot algorithm either before or after the Western Blot.

Although these HIV antibody tests are useful in screening pregnant women for HIV infection, they are not diagnostic of infection in newborns; that is, they are unable to distinguish between infected and uninfected infants. (All babies are born with passively acquired maternal antibodies, which may persist for up to 15 months; consequently, all infants born to HIV-infected mothers will test HIV-antibody positive at birth using these serological tests.) Several tests (e.g., assays for IgM and IgA HIV antibodies, which do not cross the placenta and therefore reflect infant status; in vitro antibody production; polymerase chain reaction) currently under development may prove useful in the diagnosis of infection in young infants. The polymerase chain reaction (PCR) in particular appears promising. To enhance the ability to detect the virus, PCP, amplifies specific viral nucleic acid sequences (e.g., HIV proviral DNA) in an individual's peripheral blood mononuclear cells, a capability of particular interest in neonates, given the difficulty of distinguishing infants who are infected rather than merely seropositive. Because of PCR's high level of technical sensitivity, however, inadvertent laboratory contamination or presence of maternal cells can lead to false-positive results. Moreover, the technique's sensitivity and specificity in diagnosing infection in young infants have yet to be determined. For the moment, it appears that it is not sufficiently sensitive to diagnose infection in infants who do not develop AIDS within the first year of life. PCR requires further evaluation before wide-scale use is considered and should, at this point, be regarded as a research rather than a diagnostic tool.

Medical Management and Treatment of HIV Infection in Children

The manifestations and course of HIV disease in children are distinct from those in adults. For example, the tempo of disease progression from asymptomatic infection to AIDS is generally more rapid: children with perinatally acquired infection most often develop signs and symptoms of disease within the first year of life, although the true median age at diagnosis is probably closer to two years. (Overall, about half of all

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

pediatric AIDS cases reported to date were diagnosed before their first birthday; some children, however, did not develop HIV-related conditions until they were seven, eight, nine, or even ten years of age.) Infants diagnosed within the first year of life tend to have the poorest prognosis: fewer than half survive for 12 months. In contrast, about two-thirds of children diagnosed at one year of age or older were still alive 12 months after diagnosis.

HIV infection in children is a multisystem disease, characterized by progressive immunologic deterioration and thus susceptibility to opportunistic infections and other HIV-related illnesses. The most common opportunistic infection among perinatally infected children is Peneumocystis carinii pneumonia (PCP), which usually appears in the first year of life and is associated with a grim prognosis. Other common HIV-associated conditions or infections include failure to thrive, recurrent bacterial infections, persistent oral thrush, and lymphoid interstitial pneumonia. In addition, the majority of infected children experience some degree of neurodevelopmental delay, and some will ultimately develop progressive encephalopathy.

Because the manifestations of HIV infection in children are many and varied, the medical management and treatment of these children are fairly complicated. Such care can include multiple therapeutic interventions, nutritional and psychosocial support, and even pain management. Early identification of infected children allows anticipatory management and the consequent vigorous treatment of bacterial infections and other opportunistic diseases. In addition, it permits modification of routine pediatric vaccination practices—for example, inactivated rather than oral poliovirus vaccine can be administered. Early identification of children who are at risk for infection but have not yet been diagnosed as infected allows medical caregivers to carefully monitor the child's condition for evidence of immunologic impairment and signs and symptoms of HIV disease.

Despite the apparent rationality of these approaches, there is little evidence to confirm that such aggressive care and management makes a difference over the course of the disease. As a result, some pediatric experts question whether these treatment and care measures actually prolong life. Others speculate that quality of life may be improved, but this, too, is unsubstantiated by anything except anecdotal evidence. Notwithstanding this uncertainty regarding the benefits of intervention, the tendency of most caregivers has been to assume that early recognition of infection in children can lead to more effective medical management and treatment.

As noted earlier, perinatally infected children often present with symptomatic disease within the first year of life, and they frequently experience an accelerated decline. This rapid progression creates a narrow

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

"window" of opportunity for early therapeutic intervention—for example, antiretroviral therapy and prophylaxis against opportunistic infection. Yet the effectiveness of primary PCP prophylaxis in HIV-infected infants is uncertain. Clinical experience with HIV-infected children seems to suggest that it may be beneficial in preventing pneumonia and may even help prevent other bacterial infections. Nevertheless, most pediatric experts agree that PCP prophylactic therapy should not be administered to all seropositive infants. This regimen should still be reserved for children with known HIV infection.

Questions also remain regarding early intervention with antiretroviral therapy such as zidovudine. Evidence suggests that zidovudine therapy in children with symptomatic HIV infection or AIDS leads to improvements in neurological function and weight and growth parameters, and may even prolong life. Some clinicians argue that early intervention with antiretroviral therapy may help reverse or prevent the neurodevelopmental deterioration commonly seen among HIV-infected children. It is as yet unclear, however, whether such therapy will ultimately be of benefit to asymptomatic HIV-infected children.

Medical Management of HIV-Infected Pregnant Women

Appropriate prenatal management of HIV-infected pregnant women includes assessing maternal immunologic function, treating HIV-related complications and infections, and safeguarding and monitoring the intrauterine environment. Some guidance in the management of HIV-infected pregnant women can be gleaned from experience in the management of HIV-infected nonpregnant adults.

For example, among HIV-infected nonpregnant adults, immunologic function is now generally monitored through serial CD4+ cell counts to predict disease progression and indicate when to initiate antiretroviral therapy and prophylaxis against opportunistic infection. There is good evidence of effectiveness of such regimens among nonpregnant infected adults: trials of zidovudine in nonpregnant adults (with CD4+ cell counts of less than 200 and counts between 200 and 500) have demonstrated its effectiveness in delaying the progression of disease, and PCP prophylaxis has been successful in delaying or preventing the onset of Pneumocystis pneumonia (one of the most common HIV-related opportunistic infections) in nonpregnant adults with CD4+ cell counts below 200.

Many obstetrical experts concur that zidovudine therapy should be administered to those pregnant women who are at highest risk of disease progression (i.e., those with CD4+ cell counts of less than 200). Although the safety of zidovudine therapy during pregnancy is still relatively

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

unconfirmed, its use among these women is supported by arguments that the maternal benefits outweigh potential fetal risks and that it is unethical to deny pregnant women a standard of care that would be considered life-prolonging if they were not pregnant. For pregnant women with CD4+ cell counts of between 200 and 500, the benefits associated with antiretroviral therapy are less dear, and many practitioners would defer therapy in this group until the postpartum period. Obstetrical experts believe that, in all cases, the maternal benefits should be weighed against the possible fetal risks, and decisions regarding initiation of therapy should be made in concert with the patient.

Because infected pregnant women with CD4+ cell counts of less than 200 are at greatest risk of developing PCP, many experts believe that administering prophylactic therapy to this group is appropriate. The most common therapies now in use are trimethoprim-sulfamethoxazole (TMP-SMX) and aerosolized pentamidine. TMP-SMX therapy during pregnancy raises some concern because the drug readily crosses the placenta. Sulfonamide therapy has been associated with displacement of bilirubin from its protein-binding site, resulting in increased circulating unconjugated bilirubin, which moves easily across the placental barrier. However, unconjugated bilirubin would be cleared in the maternal circulation, and hence the fetal risks of such therapy appear to be minimal. In neonates, high levels of unconjugated bilirubin have been associated with kernicterus (i.e., the accumulation of bilirubin in the brain, which can lead to severe neurological damage). To date, there have been no reports of increased incidence of newborn kernicterus following maternal sulfonamide therapy; nevertheless, infants born to mothers who have received TMP-SMX therapy during pregnancy should be carefully monitored. There has been little study of the use of aerosolized pentamidine among pregnant HIV-infected women. However, because it has little systemic absorption, one would expect little effect on the fetus or neonate.

Pregnancy also brings special considerations to the treatment of other infectious complications associated with HIV infection, including tuberculosis, fungal disease, herpes simplex, and syphilis. Also germane to the management of pregnant HIV-infected women are the potential effects of pregnancy itself on maternal disease progression. In the case of "normal" pregnancies (i.e., among uninfected individuals), the body experiences immunologic adaptations (reductions in lymphocyte function and in CD4+ cell counts) to prevent maternal rejection of the fetus as a foreign graft. Additionally, during pregnancy women can be prone to increased morbidity from a number of vital illnesses. Specific concern about HIV-infected pregnant women has focused on whether pregnancy accelerates HIV-induced depletion of CD4+ cells and enhances the risk of developing AIDS.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

Unfortunately, few studies have examined the impact Of pregnancy on the progression of maternal HIV disease. One recent study compared changes in T-lymphocytes in 102 pregnant women (39 of whom were HIV positive and 63 of whom were HIV negative) during pregnancy and 12 months postpartum.1 Among the HIV-infected women, the researchers observed a steady decrease in CD4+ cell levels throughout and following pregnancy, although the rate of CD4+ cell decline during pregnancy appeared to be steeper than during the postpartum period. These data have several limitations, however. No statistical approach was used to compare the slope of changes in CD4+ cell levels before and after delivery. Moreover, the clinical significance of the above observation is unclear because control or comparison data from nonpregnant HIV-infected women are lacking. The authors noted that the rate of CD4+ cell decline in the HIV-infected pregnant women was somewhat faster than that in HIV-infected homosexual men or hemophiliacs over the same time period. They concluded that pregnancy may be an important determinant in the further destruction of CD4+ cells. However, further studies are needed to compare the rate of CD4+ cell loss among HIV-infected pregnant women with that of infected nonpregnant women.

Another area of research interest is the potential interruption of vertical transmission through maternal chemoprophylaxis. Studies of maternal therapy to reduce the risk of perinatal transmission are currently under development. One such study protocol, which is being revised to incorporate suggestions from the Food and Drug Administration's Antiviral Drugs Advisory Committee, is a randomized, controlled trial of zidovudine therapy among HIV-infected pregnant women (perhaps beginning as early as the second trimester) followed by newborn treatment during the first weeks of life.

Other strategies to prevent perinatal transmission are also under consideration. These include giving mothers purified human immunoglobulin containing high titers of antibodies to HIV structural proteins (e.g., gp120), which are capable of crossing the placenta beginning in the late second trimester. The rationale for this approach stems from recent reports suggesting that the absence of certain maternal antibodies (i.e., anti-gp120 antibodies) appears to be correlated with vertical transmission of infection.

In sum, many obstetrical experts agree that therapies should not be withheld from pregnant women unless there are known adverse fetal or neonatal effects that outweigh maternal therapeutic benefits. In addition,

1  

 R. J. Biggar, S. Pahwa, H. Minkoff, H. Mendes [sic], A. Willoughby, S. Landesman, and J. J. Goedert, "Immunosuppression in Pregnant Women Infected with Human Immunodeficiency Virus, "American Journal of Obstetrics and Gynecology 161(1989):1239-1244.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

they maintain that guidelines for treatment during pregnancy should be based on a standard of care for HIV-infected pregnant women that has first been proven effective in nonpregnant adults. Any decisions regarding therapy and acceptable maternal and fetal risks should be made in concert with the patient.

PRINCIPLES AND PITFALLS OF MASS POPULATION SCREENING

In the search for an acceptable public policy on prenatal and neonatal HIV screening, it is helpful to look to past experience with mass screening programs. By examining problems in previous screening programs from an historical perspective, and ascertaining the extent to which these problems have relevance to proposals for perinatal HIV screening, it may be possible to avoid repeating mistakes.

Screening programs typically have served three major purposes. The first is to identify presymptomatic individuals so that disease can be treated or prevented. The second is to counsel individuals of reproductive age who are making family planning decisions. (Both of these purposes are partially motivated by public health concerns about horizontal and vertical transmission of an infectious disease.) The third is research, including enumeration, natural history studies, and recruitment for experimental treatment protocols.

Whether screening programs should be implemented at all, and whether they should be mandatory or voluntary, depends in part on the extent to which these goals are achievable. Implementation depends as well on ethical concerns about the relative balance of benefit and harm. Many who either oppose mandatory screening or urge extreme caution before implementing a mass screening program refer to the harm caused by therapeutic misadventures of the past, all of which were premature attempts to screen for "urgent" conditions that could "obviously" benefit from treatment.

The best-documented mass newborn screening for purposes of treatment is screening for phenylketonuria (PKU). When first introduced, this mandatory screening program was based on two ultimately flawed assumptions. First, it was expected that most infants with high blood phenylalanine were destined to be retarded; in fact, many had normal intelligence. The second problem was a failure to appreciate the toxicity of a low phenylalanine diet. The consequences of these mistaken premises were twofold. First, because many children who were destined to be of normal intelligence were recruited into trials investigating the effects of low phenylalanine diets, it appeared that the diet had been effective in

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

preventing retardation. Second, and perhaps more important, the toxicity of the diet itself caused retardation and some deaths, not just among infants with PKU but for some normal children. The problems of specificity in newborn screening for PKU have since been resolved.2 This program is now considered w be successful at identifying those infants who will ultimately benefit from therapy.

The PKU experience highlights several recurring themes in mass screening for the purpose of treatment. First, when the manifestations of a disorder are heterogeneous, there will be uncertainty with respect to benefit and risk. Just as some children with elevated blood phenylalanine are destined to be normal, so, too, most newborns with HIV antibodies are in fact uninfected. Even if truly infected newborns could be identified, they are a heterogeneous population with respect to their prognosis. Some may die within the first year of life, and others may live a decade or longer before symptoms develop. A second aspect of the PKU experience is that it was instituted without adequate pilot studies. Because the benefits and risks of HIV screening are not yet fully understood, some would argue that HIV screening programs should be treated as experiments and subjected to peer and institutional review board evaluation as well as stringent criteria for informed consent. Others maintain that, given a substantial medical benefit that appears to outweigh harm, screening programs may be implemented but should be carefully monitored. A third lesson of the PKU program is to avoid mandating screening programs by law before their adverse consequences are identified and minimized. It is difficult to build into law the flexibility required for midcourse correction.

The most instructive historical example of the potential pitfalls of screening for purposes of reproductive counseling is that of sickle-cell trait (to be distinguished from more recent advances achieved through newborn screening for sickle-cell disense). The goal of sickle-cell trait screening was to identify individuals who were carriers of the trait, thus providing them with additional information on which to base their reproductive decisions. In retrospect, reproductive choices were often made based on inaccurate

2  

 With the use of additional tests, it is now possible to distinguish children with true PKU from those with benign forms of hyperphenylalaninemia, as well as other variations that cause retardation but may not respond to simple phenylalanine restriction. Other problems with the PKU program are more complex and difficult to resolve. Women who were successfully treated for PKU in childhood continue to have high blood phenylalanine levels if they consume a normal diet. Maternal hyperphenylalaninemia is one of the most potent human teratogens. The vast majority of newborns who are exposed in utero have severe malformations and mental retardation. Because a low PKU diet is extremely unpalatable, and prevention of fetal damage may require beginning a low phenylalanine diet before conception, this problem is more difficult to resolve. This situation illustrates the need to monitor treatment programs for long-term effects.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

assumptions (e.g., the belief that sickle-cell trait itself was a disease), resulting in irrational decisions. For example, some couples chose to forego procreation altogether based on a false belief that they were at risk. Also present was the risk of stigmatization, that is, the unwelcome or adverse social consequences of accurate information. In the case of sickle-cell trait, which appears predominantly among Afro-Americans, stigmatization on the basis of trait status resulted in reduced eligibility for marriage and reproduction and led to discrimination in insurance and employment. Like the earlier example of PKU screening for the purpose of offering treatment, screening with the goal of informing reproductive counseling has at times been accompanied by substantial risks (generally psychosocial rather than medical) that were unanticipated at the outset of the program.

There are several lessons from the sickle-cell experience that may be applicable to HIV screening. The first is the danger of discrimination when a screening program or disorder targets racial groups. Because the prevalence of HIV infection is increasing among blacks and Hispanics (groups already disadvantaged with respect to obtaining insurance, employment, and health care), policies formulated by the white majority to identify infected persons may be viewed by these groups with suspicion and concern. A second, related lesson is that medical and political motivations are often difficult to separate. If public monies are devoted to HIV screening for political reasons, then HIV screening will occur regardless of whether medical benefit has been demonstrated. Third, if reproductive choices are to be made individually, based on personal values and beliefs,3 then counseling and education needs to precede and follow screening.

As a result of previous experience with screening programs, the National Research Council,4 the Hastings Center,5 and the President's Commission for the Study of Ethical Problems in Medicine and Biomedical

3  

 In the case of sickle-cell screening, several reproductive options were available to individuals to prevent affected offspring: they could decide to have no children, they could consider adoption of healthy children, or they could pursue impregnation by another partner, either informally or through the use of new reproductive technologies (e.g., artificial insemination by a donor who had been screened and found not to be a carrier). They could also, being aware of the risks, decide to have a child.

4  

 National Research Council Genetic Screening: Programs, Principles, and Research, Report of the Committee for the Study of Inborn Errors of Metabolism, Division of Medical Sciences, Assembly of Life Sciences (Washington, D.C.: National Academy of Sciences, 1975).

5  

 M. Lappé, J. M. Gustafson, and R. Roblin (co-chairmen), Research Group on Ethical, Social and Legal Issues in Genetic Counseling and Genetic Engineering of the Institute of Society, Ethics and the Life Sciences, "Ethical and Social Issues in Screening for Genetic Disease," New England Journal of Medicine 286(1972):1129-1132.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

and Behavioral Research6 independently developed criteria for ethically and socially acceptable mass screening. Several themes seem to have attracted consensus. First, the goals should be carefully defined and shown to be achievable. Second, if reproductive planning is the purpose of the screening, education should be shown to be effective. Third, informed consent should be feasible in terms of the capacity of those being screened to understand the benefits and risks of the proposed tests. Fourth, if the goal of screening is treatment, services should be available. Fifth, quality control needs to be ensured. Finally, the costs should be acceptable.

Several problems are evident when these criteria are applied to newborn HIV screening. First, newborn screening is, de facto, maternal screening. If one of the achievable goals of newborn screening is to identify maternal infection and provide voluntary counseling and treatment to the woman, it would be more justifiable ethically to screen the woman herself. Second, a major goal of neonatal screening is likely to be early recognition of infection and subsequent therapeutic intervention (e.g., antiretroviral therapy or prophylaxis against opportunistic infection). It is currently unclear, however, just how great a benefit may be derived from treatment of newborns. This problem is compounded by the difficulty in distinguishing truly infected infants from those who are seropositive as a result of passive transfer of maternal antibodies. Third, even if these interventions are shown to be beneficial the relative weight of benefits and burdens to be derived from screening for any given infant is a judgment that is traditionally made by the infants parents. Some refusals of standard, established treatment constitute neglect. If a parent were negligent in refusing to allow his or her infant to be tested for a therapeutic intervention with known benefit, established mechanisms for authorizing treatment could be utilized.

A fourth problem is the difficulty in obtaining services, for financial and clinical reasons. There is unpublished evidence that health care professionals may change their treatment decisions (e.g., forego aggressive measures, such as surgery, if life-threatening complications arise) if they believe a baby is HIV-antibody positive. Finally, in addition to the problems encountered in meeting these criteria, there are questions regarding the accuracy of the screening tests now available (especially for newborns), stigmatization of newborns and their mothers, concerns about discrimination, and uncertain cost-effectiveness.

Because the benefits of HIV screening are unproven and the risks of psychosocial injury substantial, most health experts argue that it is

6  

 President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, Screening and Counseling for Genetic Conditions (Washington, D.C.: Government Printing Office, 1983).

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

particularly important to pay strict attention to the ethical and legal duty to secure an informed consent from an individual before proceeding with HIV testing. Indirect maternal screening (i.e., through the newborn) without consent subjects the woman to a range of potentially serious harms, and there is evidence that concern about these harms may deter some women from seeking health care. Consequently, newborn HIV screening does not qualify as a routine procedure for which consent may be presumed or inferred. Yet informed consent may not always be feasible. Parental involvement in these cases is often limited because of the mother's poor health and sometimes poor mental status. If a parent is absent or legally incompetent, the child needs a guardian appointed who can make reasonable judgments about the child's care, including participation in treatment or research protocols.

Although such problems are acknowledged by proponents of mass screening, they argue that, if well organized, screening can be one of the most important, cost-effective contributions to the nation's health that health agencies provide. Under the direction of skilled public health screening specialists, with careful program design, a flexible implementation mechanism, and broad-based input in the form of public scrutiny and debate, many of the pitfalls of screening can be avoided. Well-organized mass screening can also permit large-scale collaborative studies that help identify and resolve program flaws or unanticipated side effects more quickly than small-scale programs. Critics of mass screening emphasize that the first moral responsibility is to do no harm (nonmaleficence); supporters, on the other hand, claim that, if screening is to be performed at all, the ethical mandate is to do good (beneficence) and to do it for the greatest number (utility). They argue that harm is done by omission as well as commission.

When the state mandates screening, it incurs moral obligations. First, the state must provide the means to comply with the screening mandate. Second, it must ensure the protection of privacy and confidentiality. Third, it must provide access to appropriate prevention or treatment services. Some experts hold the controversial view that people should be screened, a need identified, and demand created so that appropriate treatment or prevention services will then be made available.

Some supporters of universal screening point to the program of the California Department of Health Services' Genetic Disease Branch as exemplifying the benefits such screening can offer. Advocates of the program believe that confidentiality of test results and medical records can be effectively provided. Some argue that such universal programs may be better able than selective screening programs to protect the privacy of those screened because screening is offered to everyone regardless of risk history. Furthermore, universal screening is likely to identify cases in so-

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

called low-risk groups, which may dilute some of the stigma attached to positive test results by demonstrating that the problem is not restricted to a small segment of the population.

Before (widespread perinatal or any other type of) HIV screening is undertaken, goals need to be clearly defined, and they must be achievable. In addition, attention must be paid to the risks involved such as the potential for stigmatization, deterrence from seeking health care, and limited access to services. Finally, standards of informed consent should be observed (see the later discussion on consent and counseling).

WHO SHOULD BE SCREENED?

Screening programs typically have three common characteristics. First, they share the goal of identifying persons in the early, presymptomatic stages of an illness so that some action can be taken to favorably alter the course of the illness, an action that would not be as effective if one waited until the disease manifested itself overtly. Second is the existence of some proper authority for conducting the screening, either the consent of the Persons being screened or some public mandate. Third is the requirement that the process be efficient in both human and economic terms (i.e., the benefits of early identification and treatment should exceed the Personal and societal costs associated with screening).

The first and most compelling goal of neonatal HIV screening is to provide potential benefits to the child. These benefits include, most notably, prevention of Pneumocystis pneumonia, as well as early detection and prevention of other HIV-associated illnesses; changes in pediatric management, such as substitution of an inactivated for a live poliovirus vaccine; and early intervention to ameliorate some of the social disruption that occurs in families stricken with HIV disease. A second but less important goal involves case-finding, including the infant's parents or siblings. (The presumption is that early detection and treatment would be beneficial to all of the individuals identified.) A third goal is prevention of secondary transmission. Although available data overwhelmingly suggest that there is little risk of casual transmission of HIV infection without direct mucous membrane or wound contact with contaminated body fluids, in some select cases there may be a reason to let caretakers know that a child is infected.

Whether these goals can be achieved by newborn HIV screening depends on the epidemiology, natural history, and nature of transmission of HIV infection in children, and the effectiveness of HIV screening tools and medical interventions. It is likely that the prevalence of HIV infection among newborns will continue to be highest among disadvantaged

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

populations, but indications are that a growing number of cases may be found outside of the most common risk groups and geographic areas. In addition, there will continue to be a range of expression of perinatally acquired HIV infection from severe manifestations with early mortality to increasing numbers of infected children living to school age and beyond. In all probability, zidovudine will show promise for treating asymptomatic children who are infected; however, given the difficulties of diagnosing infection in infants, uncertainties will remain about when such therapy should begin. Consequently, at least in the first few months of life, there will continue to be a risk of treating infants who are, in fact, not infected.

HIV screening programs for any group must continue to consider the problem of discrimination. Although the stigmatizing effect of HIV infection can be eased through intensive counseling and support services, it may remain a significant deterrent to individuals who might otherwise wish to know their own or their child's HIV-antibody status. By the same token, medical and public health authorities are likely to consider any sort of mandatory HIV screening as a major imposition that could only be justified by much larger and more concrete benefits than have been demonstrated to this point.

With respect to vertical transmission, about 25 to 35 percent of children born to HIV-positive mothers will ultimately be infected themselves. Some experts believe that a medical approach to prevent vertical transmission is unlikely in the near future, although the avoidance of breast feeding may come to be advocated more strongly (particularly in countries where safe alternatives to breast milk exist). Behavioral change will remain the only effective means of preventing viral transmission. Finally, it is unlikely that the discovery of new tests to allow earlier and more precise identification of infection in infants will completely solve the problem of determining, in the first days or weeks of life, which infants are truly infected.

Although conventional HIV-antibody tests are not specific enough to identify truly infected infants and interventions do not yet exist to prevent transmission or cure HIV infection, potential changes in medical management might justify neonatal screening even with the current limitations in the state of knowledge. For example, close medical follow-up of seropositive infants may allow early identification of symptomatic infants who are likely to benefit from intervention.

Of the three major interventions that seem most practical—PCP prophylaxis, the alteration of immunization schedules, and close medical follow-up—the first and third present certain risks for those who prove ultimately not to be infected. PCP prophylaxis for these children exposes them to therapeutic risks without accompanying medical benefits. Close medical follow-up may subject infants to iatrogenic risks either through

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

excessive interventions for benign illness or participation in experimental protocols. The medical risks should be minimal, however, if HIV-specific interventions such as zidovudine are reserved for infants who develop symptoms and if presymptomatic treatment is confined to nutritional support and prompt treatment of intercurrent infections.

Close follow-up may also have the effect of labeling infants as HIV infected, a label that will be difficult to remove even though later definitive testing shows no infection. Avoiding such a label could well spare a child the burdens of discrimination in early child care programs. These risks might be justified if they were balanced by improvements in the quality of medical and social support. The benefits of prophylaxis and early treatment may, indeed, outweigh the social and medical risks inherent in case identification. However, establishing the necessary service network to deliver those benefits remains the most significant barrier to the accomplishment of screening goals.

Once the benefits of screening and the likelihood of their effectiveness are established, a decision must be made whether to target screening or make it universal. There is a compelling rationale for geographically targeted newborn screening so that limited resources are not used inefficiently and medical attention is not diverted from other important concerns. Yet there are epidemiological and sociological arguments against geographic targeting that are at least as compelling. For example, as HIV infection spreads among the heterosexual population, it will also spread in the pediatric population. In fact, almost every state now has at least some pediatric AIDS cases. Another concern about geographic targeting is that, because HIV infection currently is most prevalent among minority populations, targeting might subtly reinforce discrimination.

The next question is whether there can or should be targeting at the individual level. A logical approach would be one that combined several strategies. For example, the prevalence in different geographic regions could determine the aggressiveness with which prenatal HIV screening is carried out and individual risk factors are considered. Prevalence of infection in a locale (determined with the use of methodologically sound population estimates) of less than 1 infected woman per 1,000 women, would be considered ''low.'' To apply such a threshold, it would be necessary to determine how such geographic boundaries were to be defined. Whereas it might be most efficient for public health authorities to implement a program at the state level, clinically it might be more meaningful to develop screening programs based on prevalence in specific communities.

Some experts believe newborn HIV screening should be universally offered but voluntary, because it is still questionable whether its goals can be achieved. (The thoroughness with which parents are approached should

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

be dictated by the extent of prenatal HIV screening, local seroprevalence, and the existence of specific individual risk factors.) Advances in diagnostic technology, presymptomatic treatment, and social safeguards will further strengthen the case for neonatal HIV screening. There is some consensus, likewise, regarding the routine offering of HIV screening to pregnant women in all communities regardless of prevalence. (In higher prevalence communities, such screening might be accompanied by the recommendation that women be specifically encouraged to take the test.) Such a policy of universal voluntary prenatal screening has several justifications. First, the autonomous decisions of competent adults should be respected. Moreover, it is arguably the case that prenatal screening, no matter how forcefully or mandatorily it is implemented, will not substantively change the epidemiology of HIV disease in the next generation. There are several lessons to be learned in this regard from the public health system's experience with syphilis, a sexually transmitted disease that is highly correlated with HIV infection. Despite the implementation of mandatory prenatal screening and treatment for syphilis, as well as the availability of a curative agent, syphilis in the United States is now occurring in epidemic proportions. In the case of HIV, identifying infected women is an important task and potentially beneficial, but screening carried out in health care settings will in fact identify only a small proportion of those infected because the women at greatest risk are the ones who do not receive health care. The use of crack cocaine (which may increase a woman's risk of becoming infected because of the bartering of sexual favors for money to obtain the drug) is a major barrier to identification because of its association with decreased health-seeking behavior. Pregnant women who do not receive care until the time they give birth are far more likely than women who do receive prenatal care to be users of crack cocaine—and to be HIV positive. Thus, testing conducted only in prenatal clinics will never identify the majority of pregnant women who are infected with HIV.

There are other barriers as well to the identification of infected women. Social factors such as a loss of status in the community, stigmatization, and a loss of social supports may deter some women from undergoing testing; inadequate insurance coverage that limits access to treatment can discourage others. A barrier to identification may also result from a reliance on self-reported risk factors to determine who will be offered testing. Some research has shown that a substantial percentage of women who are HIV positive acknowledge no risk factors. In some cases, women may be reluctant to disclose the drug use or sexual behaviors that place them at risk; in other cases, they may simply be unaware of their risk. One cause of this lack of patient awareness may be dishonesty

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

on the pan of male partners (in an attempt to "make a good impression") regarding their sexual history and past or present drug use.

Another barrier to identification may stem from the ignorance of physicians regarding HIV (and risk-associated behaviors) and their fear of finding a case of infection. It is sometimes difficult for clinicians to confront their own inadequacies regarding the detection and treatment of HIV infection; in addition, the referral networks that exist for other diseases are not yet firmly established for HIV. Obstetricians in particular often experience painful conflicts: society appears to be asking them to become champions for the needs of and benefits to the child, when their traditional, and to them, appropriate, focus has been the needs of the mother (although not to the exclusion of considerations for the child).

Even if these barriers to diagnosing HIV infection can be overcome and women can obtain testing, the number of women who are actually tested will still vary among programs. The best predictor of how many women will be tested in a given setting is the type of program there. Some evidence suggests that if the test is selectively offered (on the basis of risk factors), about 10 percent of women will be tested. If the test is universally offered, using a neutral (nondirective) counseling model, about 40 percent of women will take the test. If the test is universally recommended (i.e., a more directive approach is used), 70 to 80 percent of women will be tested. Finally, if a screening program ostensibly gives the woman the right to refuse the test but does not inform her that she has that right, it is indistinguishable from a mandatory program and virtually 100 percent of women will be tested.

Many experts agree that screening all women of reproductive age for the purpose of identifying and treating HIV-infected women does not appear to be an efficient use of limited resources. As indicated earlier, even when testing is widely available, not all women who are at risk will benefit from such testing or even avail themselves of the service. HIV screening instituted for the purpose of changing behavior is also of uncertain merit. Experience has shown that it is very difficult to change risk-taking and sexual behaviors, even among HIV-seropositive individuals. Indeed, large proportions of adolescents and adult women, both infected and uninfected, experience difficulties with contraceptive use, as well as with planning their pregnancies. The results of several studies of pregnant and nonpregnant HIV-infected women's behavioral response to knowledge of their HIV infection indicated that approximately 20 to 25 percent chose abstinence, 30 to 40 percent practiced fairly reliable safer sex techniques, and another third or more continued their sexual practices unchanged. Even within a relatively empowered group of educated individuals, knowledge of the risks of HIV infection in many cases does not produce behavioral change.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

Knowledge of HIV-antibody status does not seem to affect pregnancy termination decisions any more strongly than it affects contraceptive behavior. One study in New York found that HIV-positive women do not choose to terminate their pregnancies more frequently than seronegative women7 Neither the perceived personal risk of perinatal HIV transmission nor the presence of concern about AIDS appeared to be associated with an increased rate of pregnancy termination. Decisions to terminate were best predicted by whether the pregnancy was unplanned and whether the woman reported feeling upset, sad, or frightened when she first learned about it.

Some health officials contend that networks already exist within the United States where women of reproductive age have access to HIV testing services. These experts argue that, rather than funding wide-scale screening, confidential, voluntary testing services should be made available in a number of sites serving women, such as family planning clinics, prenatal care clinics, sexually transmitted disease clinics, substance abuse treatment centers, or alternative testing sites. Neutral counseling and education provided in such settings can facilitate informed reproductive decision making by seropositive women (and, where appropriate, their partners). Counseling regarding the use of condoms for purposes of both contraception and prevention of transmission of infection is also likely to be offered. Even in these already established settings, however, substantial resources are required to provide appropriate, effective counseling and testing and the social support and health care services needed for both infected women and their children.

CONSENT AND COUNSELING

Creating an adequate informed consent process for HIV screening can be guided by knowledge of informed consent generally. Information about the status of informed consent in clinical medicine, how consent requirements shift as one moves from a clinical to a public health perspective, and alternative models for consent in the context of screening programs will contribute to formulating an appropriate informed consent process for prenatal HIV screening.

7  

 P. A. Selwyn, R. J. Carter, E. E. Schoenbaum, V. J. Robertson, R. S. Klein, and M. F. Rogers, "Knowledge of HIV Antibody Status and Decisions to Continue or Terminate Pregnancy Among Intravenous Drug Users," Journal of the American Medical. Association 261(1989):3567-3571.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

Informed consent is not something one gives to a patient, nor is it something one does to a patient. Rather, a consent is something one gets from a patient. Sometimes, in the context of research, consent is merely a permission that satisfies the operative institutional or legal rules, most classically, a signature on a form. Informed consent, however, has been taken to mean much more, both in law and in medical ethics, than a signature on a form.

An informed consent is an autonomous action by a subject or patient that authorizes a professional to initiate a specific research intervention or therapeutic medical plan or to withdraw treatment. Conversely, an informed refusal would be the autonomous act of refusing or withholding such authorization. Informed consents and refusals are two sides of the same coin. Indeed, there is no point in having an informed consent policy if there is no simultaneous commitment to respect refusals. Informed consent need not be fully autonomous as long as the person makes a decision based on a substantially accurate understanding of what is at stake and makes that choice reasonably free of constraint by others. Underlying the concept of informed consent are the principles of respect for autonomy and respect for privacy.

Informed consent is a relatively new tradition, the theory and rhetoric of which has gained rapid acceptance in the medical community. There are grave questions, however, about whether the theory, rhetoric, and formality of informed consent procedures have any meaning in practice or whether they have changed the fundamental character of the physician-patient relationship. The lines of authority and control seem roughly to be what they have always been. Patients routinely acquiesce to medical interventions rather than autonomously authorizing them.

Patient acquiescence is not a problem for many circumstances in clinical medicine (e.g., noninvasive interventions that pose literally no risk) in which proceeding without specific patient authorization can be plausibly defended. Committing the necessary staff resources to obtain informed consent for these interventions would entail shifting resources from other competing needs and interests, such as other educational efforts. The use of such resources to obtain informed consent is not justified, either from an efficiency or a justice standpoint. At the other extreme are experimental interventions, which always require informed consent.

The general problem in the area of consent is what to do with the medical interventions that fall in between these polar extremes. The challenge in clinical medicine is to distinguish the conditions under which it is inappropriate for a clinician to proceed without the express authorization of a patient from those situations in which the clinician may proceed on his or her own initiative. Specifically, it must be determined where HIV-antibody testing fits on this continuum.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

The extent to which resources are devoted to a meaningful informed consent requirement turns, in part, on whether HIV testing during pregnancy is framed as a public health issue or an issue in clinical medicine. Insofar as public good motivates the screening, considerations for informed consent will be somewhat different than if the motivation is clinical—that is, to provide benefits to the individual pregnant woman. The question becomes whether the autonomy interests of pregnant women are, in this instance, validly overridden by the interests of society in controlling the epidemic.

This question cannot be answered without examining the extent to which screening of pregnant women would contribute to furthering society's interests. Moreover, various screening models require examination to determine the extent to which they must be mandatory to serve the public good. Different screening approaches range from voluntary with specific informed consent, to voluntary with right of refusal, to mandatory. In practice, it is difficult to maintain these distinctions in such a way that one type of program does not fold into another. If there is a policy of informed consent but the educational process that must accompany it is nonexistent, there will be no difference between a policy of informed consent and a policy of voluntary screening with the fight to refuse. The latter policy will shade into a mandatory policy to the extent that no attention is paid to notifying or educating women that testing is going to take place. Even some mandatory policies, which are those regulated by law, have allowed women to opt out of the test for reasons of conscientious objection or specific religious beliefs.

Some argue that mandatory screening programs actually hurt rather than promote the public good. A screening policy that requires informed consent can be defended, not only on the grounds of protecting individual women's rights but also on the grounds that mandatory programs have the potential to create social divisiveness and both gender and ethnic discrimination. A final consideration in screening policy is the extent to which the consent that a patient signs limits disclosure to third parties, thereby maintaining confidentiality.

Once a program establishes that informed consent is required for screening, the elements of pretest counseling, during which consent is either given or refused, must be set in place. Pretest counseling should be viewed as an opportunity for education regardless of the patient's decision about testing. In one successful program, pretest counseling consists of two pans: information and assessment. The information (education) component of this program includes discussion of such topics as the spectrum of HIV infection, modes of transmission, and risk reduction practices (e.g., condom use). This information can be imparted using a mix of methods depending on the particular setting, its resources,

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

and the patients being counseled. Another useful feature of the education component is a discussion of the importance of involving the partner in testing and counseling.

Assessment is more difficult and resource intensive, for it involves a determination of the patient's coping mechanisms, suicide risk, psychiatric problems, and support network. One of the major benefits of assessment is that it helps to provide a more meaningful informed consent. Moreover, discussion of support networks offers another opportunity to introduce the notion of including the partner in the process of HIV testing and counseling.

In the program noted above, posttest counseling begins immediately after blood is drawn (because some people will decide that they do not want to learn their results). The actual communication of test results always occurs in person. Negative results are accompanied by reinforcement of education regarding how to stay negative. Positive results are offered by two people, one of whom was responsible for conducting the pretest counseling. Result sessions are scheduled early in the week so there is adequate time for follow-up.

Posttest discussions in this program include the distinction between HIV infection and AIDS, the impact of HIV infection on the pregnancy and the pregnancy's impact on the woman's disease progression, the risk-benefit ratio of various treatments and the impact that treatment could have, and the option of pregnancy termination if the woman is early enough in her pregnancy. If abortion is an option, the counselor helps the woman explore her feelings about the pregnancy, recognizing that many HIV-infected women choose not to abort. Partner notification is discussed, and assistance in telling the woman's partner is offered. Literature is provided so that women who have been unable to absorb what is being said can have something to read when they go home and something to share with their families. Referrals for abortion or prenatal care, medical evaluation of the woman, assessment for other family members, and appropriate pediatric care are arranged, as well as follow-up counseling sessions with the woman. Although these components of counseling have contributed to this particular program's success, clinicians need to determine the model that works best in their own settings with their own resources. Whatever model is chosen, it should be culturally sensitive and mindful of the social realities (including poverty, lack of access to services, and drug addiction) facing HIV-infected women who are pregnant or considering pregnancy.

Once a woman has decided to be tested and is identified as seropositive, posttest counseling must focus not only on anxiety and coping skills but also on reproductive decisions. Often, because treatment is not always offered to pregnant women, a woman must choose between her own health

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

and remaining pregnant. Increasingly in the near future, she may be asked to decide between the risk of perinatal transmission and the possible fetotoxic risk of antiretroviral therapies intended to reduce the risk of transmission. There is considerable controversy over the degree to which counseling regarding these issues is and should be nondirective (as in the value-neutral genetic counseling model) or directive (which, in theory, can be subtle but more often tends to be overtly coercive).8 Several arguments in favor of directive counseling have been put forth. First is the lack of capacity for reasoned decision making in some women by virtue of immaturity, drug use, or other forms of incompetence. The argument is that an objective third party may be better positioned to ascertain what is in the best interests of such a person. Second is the high risk of great harm to the child. If the case could be made that being born with HIV infection would be "a wrongful life," it can be distinguished from many genetic diseases that present children with a burdened life but by no means one that is not worth living. A third justification for directive counseling focuses on social burden. Because many HIV-positive women will not be able to assume responsibility for their choices, in particular, the continuation of their pregnancies, some have questioned whether it is fair for society to pay the already substantial—and escalating—costs of treatment, social services, and foster care. The final argument is that nondirective counseling seems to be failing in altering the behavior of infected women—either not to reproduce in the first place or to abort if pregnant.

Although proponents of directive counseling acknowledge that such counseling for abortion is more difficult to defend than counseling for contraception, they argue that one form of directive counseling, called negotiated directive counseling, can be entirely noncoercive and respectful of reproductive rights. The claim is made that such a counseling model, by permitting the counselor to recommend a particular course of action only with the prior consent of the woman, poses the least threat to her autonomy. Although the vulnerability of the group singled out for such counseling raises questions about whether their choice would, indeed, be autonomous, requiring their consent would respect moral concerns about equal protection. Moreover, such a model would help to redress the power imbalance that characterizes many clinical encounters. Experimenting with different types of negotiated directive counseling, such as peer counseling, would allow assessment of the effectiveness of such a model.

Opponents of directive counseling for HIV-infected women likewise offer several reasons for their position, many of which are grounded in concerns about women's rights. They argue, first, that women are the

8  

 This controversy is relevant to pre- as well as posttest counseling—that is, not only for a woman's reproductive decisions but also for her decision about whether to be tested at all.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

most neglected group of all HIV-infected individuals. Second, pregnant women, unless they abort, are often systematically deprived of state-of-the-art treatment. Third, the women in question are already discriminated against because of their gender, race, poverty, and drug addiction, and opponents of directive counseling consider it politically and morally unacceptable to single out this group for any kind of counseling innovation without helping them with their other problems (e.g., homelessness, drug addiction). Moreover, to apply concerns about wrongful life more equitably, the same type of counseling would have to be implemented with any genetic disease that has an equivalent level of risk. Fourth, it is argued that the entire concept of a maternal-fetal conflict is flawed. The conflict is not between the mother and fetus but between the mother and someone else in a position of authority who thinks they know what is best for her baby. In this view, it is justifiable for a doctor to give advice when it is sought but not to imply that there are no other options.

Those who defend nondirective counseling base their arguments on the notion of reproductive liberty wherein only individual women and not their physicians or the government have the right to determine whether a pregnancy is continued. Caution is urged in regard to the slippery slope between directive counseling for abortion and mandatory sterilization. Proponents also assert that no directive counseling was never intended to alter behavior and therefore cannot be criticized for failing to do so. What is important in a pregnant woman's self-determined decision making is not what decision she makes but whether it is informed, and whether it is the right decision for her.

Opponents of directive counseling offer practical as well as moral arguments. They maintain that such counseling cannot possibly work in the current health care system because most women are not going to be seen in time to counsel them about abortion. Their primary claim is that efforts at prevention have been misdirected. Instead of targeting the population of women who are already HIV infected, efforts should be directed toward drug-addicted women and teenagers. In addition, they argue, work should continue toward developing some kind of nationally financed system with universal access to health care.

In determining the appropriate approach to reproductive counseling, several temptations must be resisted. The first is the tendency to view counseling as taking one form or the other. In reality, counseling combines directive and nondirective approaches because it is a human communication process that involves both speech and subtle nonverbal cues. Second is the tendency to superimpose the goals of public health, an area where recommendations for medical care are made so as to reduce morbidity, onto reproductive goals. It is easy to see why this confusion occurs in the context of HIV infection. Currently, vertical transmission

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

can only be interrupted (a public health goal) if the woman avoids having a child. In this case, the disease comes "attached" to something that the woman might want very much, namely, a child—a crucial difference between perinatal HIV infection and other diseases for which preventive health measures are available. Finally, reproductive decision making is one form of clinical decision making, and it is important to distinguish the general clinical context from the public health context. Goals that may be appropriate in the public health context, for example, discouraging HIV-infected women as a group from reproducing, may not be appropriate in the individual encounter.

HIV SCREENING POLICY IMPLEMENTATION

Perinatal HIV disease occurs predominantly in milieus characterized by a lack of material and personal resources, social disorganization, and drug use. Women and children in these environments often are not in contact with the health care system; when they are, their care tends to be underfinanced, crisis oriented, and fragmented because there is virtually no coordination of health and human services. In addition, care is often available only at sites located some distance away from the families' own communities.

Any policy of prenatal or neonatal HIV screening must consider the extent to which necessary follow-up services are available and accessible. In addition, it must be determined whether such services can be organized and delivered in a family-oriented, community-based fashion, with a regionalized network of referrals and linkages with other maternal and child health programs. Such a system requires that health professionals involved in the care of infected women and children collectively come to an understanding of what it takes to test, care for, and treat families with HIV infection (rather than focusing exclusively on women, children, or male drug users). It is likely that changes in practice standards will be necessary and that they will involve modifications in professional education for all primary care specialties including obstetrics, pediatrics, internal medicine, and family practice.

Optimal care of families with HIV infection is founded on the health care practitioner's knowledge of the family's HIV infection status and on his or her understanding of the course of the disease and the requirements of infected persons. Care of the HIV-positive pregnant woman and her partner requires close medical supervision and a range of psychosocial interventions. A child born to an HIV-seropositive mother, in addition to routine pediatric visits, requires careful medical and immunologic surveillance for signs and symptoms of HIV infection. Early detection of

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

severe immune deficiency will justify the offering, either singly or in combination, of prophylaxis against opportunistic infection, antiretroviral therapy, and participation in clinical trials. Because many of these children are also at risk for developmental disabilities, they may require formal evaluation and referral for early developmental intervention.

These families need, at the very minimum, a general, family-oriented physician and a case manager to coordinate services. In addition to providing medical services and counseling, these providers should be part of a network for consultation and referral. They should have ready access to the full range of human services providers required to meet the family's needs for housing, transportation, income, food, home care, visiting nurses, drugs and medical equipment, respite care, foster care, day care, drug treatment programs, and individual and group counseling. In this ideal multidisciplinary-team approach to care, obstetricians, pediatricians, family planning specialists, internists, psychiatrists, social workers, nurses, and human resource specialists would work together simultaneously at the same site.

Whether in an inner-city or rural area, such an arrangement could be based in a family care facility or in a clinic attached to a tertiary care center close to the patient's home. A program like this could function both in high-and low-prevalence areas and be incorporated into already existing programs for maternal and child health, programs to reduce infant mortality, and drug treatment programs. Ideally, the system of care for the HIV-infected family should be regionalized. Its design, financing, and implementation should be seen as the responsibility of the entire society carried out as a joint effort by representatives of both the public and private sectors.

One state (New Jersey) has approached prenatal and newborn HIV screening by developing a policy of universal HIV education about the virus and its transmission, risk reduction counseling on drug use and sexual behavior, and voluntary testing for all pregnant women and all women contemplating pregnancy. Other relevant New Jersey policies include AIDS prevention and control programs that involve men, and a partner notification program, wherein all persons, male or female, who are found to be HIV positive are counseled concerning partner notification and offered assistance in notifying their sexual or drug use partners of their infected status. Implied but not stated in the prenatal screening policy is also the recommendation that parents of newborns whose mothers have not been screened should be counseled and advised to consent to HIV testing. Specific written, informed consent or refusal is a cornerstone of this testing policy.

The service system in place in New Jersey to implement the screening policy includes a network of 16 free, anonymous counseling and testing

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

sites that regularly accept all people seeking an HIV test, all Health Start (a prenatal program in New Jersey) providers, family planning clinics, and maternal and child health clinics. All are expected to meet the standard of practice requirements for counseling and for offering the test. For women and infants found to be HIV positive, there are five federally funded programs that not only provide pediatric care but case managers as well, who routinely assess the mother's needs and access to care. The special child health services program has a home health Medicaid waiver program, a demonstration perinatal program, and a clinical trial at a major teaching hospital. In addition, a demonstration early intervention program was recently launched in five of the state's largest infectious disease hospitals. There are four pediatric residences in the state, one hospital is planning a home for mothers and children, and a pediatric nursing home now has beds for children with HIV disease.

Implementing the screening policy as a statewide standard of medical practice rather than by statute or regulatory requirement was considered most likely to achieve the goals of screening without the divisiveness—and extended time—of a legislative struggle. Therefore, the policy was intentionally couched as a strongly recommended standard of medical practice and sent to all licensed physicians, certified nurse-midwives, directors of obstetric and pediatric services, and health care facility administrators. Simultaneously, with the health department's support, New Jersey passed legislation regarding confidentiality. If the policy does not achieve its goals as a standard of care, the state may consider implementing the same approach but in more forceful terms (i.e., as a statute or regulation). Before any changes are made, however, evaluation is necessary to determine whether in its current form the policy is having the desired effects. New Jersey will soon begin a process-oriented, three-phase evaluation to make that determination. In the first phase, a formal questionnaire will be administered to directors of publicly funded clinics to ascertain if they are aware of the policy and adhering to it. Phase 2 will investigate basic knowledge and program implementation among a larger sample of providers including private individual and group obstetrics practices. The third phase will survey counselors to obtain information on the qualitative content of counseling and any concerns raised by the public.

New Jerseys policy analysis and choice of options were driven by both epidemiological findings from state seroprevalence data on childbearing women and the state's assessment of the proportion of infected women who were actually receiving care. Blinded seroprevalence data in 1989 showed that there were approximately 600 HIV-positive women in the state (a seroprevalence rate of 0.52 percent), but only about 25 percent of them were, in fact, receiving care. New Jersey health officials made the case that both the unacceptably high rate of infection and the lack of

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

identification and referral of infected patients for care indicated the need to introduce some statewide approach to screening.

The perspectives of several parties were included in the development of the screening policy. There was consensus, among a group composed of pediatricians with expertise in HIV infection, nurse clinicians, ethicists, and a pediatric society representative, that counseling and testing should occur prenatally, not after birth. The state also consulted community groups: the Women in AIDS Network, a New Jersey advocacy group that includes lawyers, the American Civil Liberties Union, a state public advocate, a psychotherapist, and family planning associations, as well as individual women, some of whom have AIDS. The main argument against the policy being formulated as a standard of practice was put forward by groups of obstetricians and regional American College of Obstetrics and Gynecology (ACOG) representatives, who were afraid that office-based physicians would not comply without a mandatory testing law. Indeed, the greatest challenge in implementing this policy is to help the office-based obstetrician and pediatrician counsel HIV-infected persons and secure informed consent. New Jersey has developed a clinical protocol, including a draft informed consent form, for the identification and management of asymptomatic pregnant women and children, which can be incorporated into office-based practice. In addition, a series of roving symposia for office-based practitioners are being planned. State health officials hope that the potential for lack of compliance will be overcome by making physicians feel somewhat at risk of the ''threat'' of liability if they fail to follow the protocol and miss an HIV-positive patient as a result. The Department of Health and the Women in AIDS Network are also developing several brochures to ensure that the information given to women is adequate, readable, and supportive of informed consent.

Although a statewide approach led by the state health department is one way to implement policy, there are other means that deserve consideration. Physician groups, university medical centers, local health departments, clinics, or group practices could also develop and implement policy. Many would argue that professional societies are best suited for this task, and in fact some have begun such efforts, although most have limitations. For example, current practice guidelines of the American College of Obstetrics and Gynecology9 and the American Academy of Pediatrics10 recommend HIV counseling and testing but only for those

9  

 American College of Obstetrics and Gynecology, Human Immune Deficiency Virus Infections, Technical Bulletin, No. 123 (New York: American College of Obstetrics and Gynecology, December 1988).

10  

 American Academy of Pediatrics, Task Force on Pediatric AIDS, "Perinatal Human Immunodeficiency Virus Infection," Pediartics 82(1988):941-944.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

individuals who are identified as being at risk. ACOG's most recent standards suggest history-taking at the initial prenatal visit as a way of identifying risk behaviors but do not specifically mention HIV. ACOG is also preparing a statement on physician responsibility in the area of HIV infection, including the obstetrician's need to be knowledgeable about HIV and to provide appropriate, current information to patients in a way that respects individual reproductive choice. The statement, however, does not indicate who should be counseled or tested. Although the Public Health Service11 expert panel on the content of prenatal care recommended that HIV testing be offered to all patients before conception and advised practitioners to counsel their patients concerning drug use and safer sexual practices, preconception counseling is clearly not yet the accepted standard of practice in this country. Even if it were, no standards for the content of such counseling have been set. Moreover, it is questionable whether the ACOG recommendation for selective counseling and testing is being observed.

If the standard of practice is to be changed from selective to universal counseling and voluntary testing of prenatal or preconception patients, the prevention and treatment services that are now available will have to be expanded. In addition, physicians will need considerable guidance regarding methods for eliciting key information and for securing informed consent. Unless they feel competent to carry them out, physicians are unlikely to comply with practice guidelines. Yet even if compliance with counseling and testing standards could be ensured by legislating the standards and establishing sanctions for noncompliance, the quality of patient counseling, and therefore the rational and appropriate basis for patient choices, could not be ensured without provider education and support, Regulation or legislation could, in fact, result in universal testing with perfunctory or no counseling at all.

Part of the solution to this problem depends on the involvement of professional medical societies. These organizations can play a primary role in educating and training providers so they can inform and advise their patients about HIV infection and its ramifications. Provider education can take a number of forms ranging from written publications to video and audio tapes to seminars and postgraduate courses. Training in counseling skills could also be offered; one effective technique for building confidence in such skills is the range of interactive methods that involve role-playing.

11  

 Public Health Service, Caring for Our Future: The Content of Prenatal Care, A Report of the Public Health Service Expert Panel on the Content of Prenatal Care (Washington, D.C.: Department of Health and Human Services, 1989).

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

These techniques, however, are time-consuming and instructor intensive, and they must be brought to the providers in their practice settings. For a variety of reasons—resource intensity, training needs, and quality of counseling—other health care workers may be better suited than physicians to counsel patients about HIV. Such counselors might include nurses, nurse practitioners, social workers, and retrained genetics counselors. Nevertheless, the primary providers of maternity services still require a thorough working knowledge of HIV infection and its perinatal implications. Under certain circumstances, physicians will have to be prepared to enter the educational and counseling process.

Professional societies will require financial assistance to mount such an educational effort. The grants that are now in place for supporting the education of health professionals about HIV infection will need to be expanded and focused more on the problems of women and children. Professional societies can be powerful advocates for the development and allocation of sufficient resources to train and educate their members in the care of HIV-infected women and children.

ECONOMIC CONSIDERATIONS IN SCREENING FOR PERINATAL HIV INFECTION

There are three economic aspects of perinatal HIV disease whose examination may inform the development of screening policy for pregnant women and newborns. First, to provide an economic context for the design and implementation of screening programs and early intervention schemes, it is necessary to consider the current costs of pediatric AIDS care, the patterns of health services utilization (particularly inpatient care), and the predominant sources of payment for such care. Second, to determine the magnitude of financial investment required to mount an early intervention program for HIV-infected children, estimates (albeit tentative) are needed of the costs of providing early intervention services (e.g., medical surveillance and early therapeutic intervention) for these children. Third, and perhaps most important, there is the challenge of paying for these services (and ensuring access to appropriate care), given currently available health care financing mechanisms and the possible development of innovative payment strategies in the future.

Current Costs of Pediatric AIDS Care

A complete picture of the economic impact of pediatric AIDS is difficult to discern because studies of the costs of care for children with

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

symptomatic HIV infection and AIDS are limited. Most studies to date have been confined to hospitalization costs, which do not fully represent the range of costs associated with comprehensive pediatric AIDS care. Despite this limitation, these studies do offer some insight into hospital utilization rates and associated costs among children with HIV disease. In particular, average length of hospital stay is examined because it contributes substantially to the overall costs of medical care.

One 1987 study of 37 children with AIDS at Harlem Hospital reported a mean length of hospital stay of 164 days, with an average length of stay (ALOS) of 43 days per hospitalization.12 These estimates are derived from a small sample with widely varying lengths of stay (the range was 2 to 659 days). The estimated average lifetime inpatient cost per child was $90,347 (with a range of about $7,000 to $184,000). (This estimate was based on medical expenses incurred by nine children who died and who had received all of their treatment at Harlem Hospital.) Most experts now agree that the above estimated cost is probably too high, particularly given the wide range of lifetime costs presented. The researchers in this study also investigated the factors that might explain the considerable variation observed in length of stay. They found that the primary predictors of length of stay were maternal drug use and boarder baby status. Moreover, a substantial proportion of the hospital days used by these children were for social rather than medical reasons. That is, many of the children were medically eligible for discharge, but because they lacked homes, they could not be released from the hospital Overall, social factors were responsible for one-third of all hospital days used by these children and 20 percent of total costs.

Another 1987 study of 34 HIV-infected children at Yale-New Haven Hospital discovered that about 54 percent of their hospital days were medically unnecessary.13 Again, this unnecessary inpatient care was most often associated with difficulties in finding suitable residential alternatives for these children. The authors noted, however, that the rate of unnecessary days had declined from 64 percent of all hospital days in 1983 and 1984 to 30 percent in 1987, largely because of improved outpatient care and access to foster care placement.

More recently, several studies have examined the average length of stay for pediatric AIDS hospitalizations, as well as the average per diem

12  

 J. D. Hegarty, E. J. Abrams, V. E. Hutchinson, S. W. Nicholas, M. S. Suarez, and M. C. Heagarty, "The Medical Care Costs of Human Immunodeficiency Virus-Infected Children in Harlem," Journal of the American Medical Association 260(1988):1901-1905.

13  

 K. Kemper and B. Forsyth, "Medically Unnecessary Hospital Use in Children Seropositive for Human Immunodeficiency Virus," Journal of the American Medical Association 260(1988):1906-1909.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

costs and payment sources for such care. One study reviewed pediatric HIV-related discharge data from 43 member hospitals of the National Perinatal Information Center, a nonprofit hospital membership organization and health services research center.14 The investigators found that the average length of stay for pediatric AIDS discharges was 12.5 days, with an average per diem cost of $955. They also noted that more than 75 percent of funds to pay for inpatient care of children with AIDS came from public sources, primarily Medicaid.

Another study collected survey data from 45 member hospitals of the National Association of Children's Hospitals and Related Institutions, a national voluntary association of children's hospitals in the United States and Canada.15 Although the average length of stay reported in this study was slightly higher (ranging from 16.3 to 17.6 days) than that in the study noted above, the average cost per day was comparable—about $900, with average charges of about $1,200. This study also found that Medicaid was the primary source of payment for pediatric AIDS inpatient care (i.e., 45 percent of all admissions were covered by Medicaid).

Currently, Jesse Green and his colleagues at New York University are conducting an investigation of hospital utilization patterns among children with AIDS. Using hospital discharge data bases maintained by the states of New York, California, and Florida on all hospitalizations in short-term general hospitals, they extracted all pediatric AIDS cases reported in the period 1983-1986. Surprisingly, they found that, for this four-year period, children with AIDS had an average length of hospital stay in both California and New York that was lower than that for adults. In New York, for example, ALOS for children with AIDS was 12.5 days; it was 21.3 days for adults with AIDS. However, the length-of-stay data for children included some very short (e.g., 1 day) and long (e.g., more than 30 days) hospital stays, which skewed the data. When one-day stays are removed, the average length of stay for children with AIDS rises to 26.6 days in New York. In addition, children under one year of age have longer average hospital stays (i.e., about 40 days) than older children.

The data from New York and California also indicate that the number of hospitalizations for pediatric AIDS have increased substantially from 1983 to 1986. Similar trends have been observed in San Francisco and

14  

 S. Allison-Cooke, "Cost and Utilization of Inpatient Services for Pediatric AIDS," Challenges for Public Health Statistics in the 1990s. Proceedings of the 1989 Public Health Conference on Records and Statistics July 17-19, 1989. (Hyattsville, Md.: National Center for Health Statistics, 1990).

15  

 D. P. Andrulis, V. B. Weslowski, E. Hintz, R. H. Parrot, and M. Brady, Pediatric AIDS and Hospital Care in the U.S.: Report on the 1987 U.S. Hospital Pediatric AIDS Survey (Washington, D.C.: National Public Health and Hospital Institute, 1990).

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

Los Angeles. Most notable, however, is the comparison of hospitalizations in California and New York. In 1986, there were 126 hospitalizations for children with AIDS in California but nearly 10 times that number in New York. Over the four years combined, the number of hospitalizations for children with AIDS in New York were also approximately 10 times higher than in California. Pediatric AIDS cases constituted just slightly more than 1 percent of all AIDS hospitalizations in California, whereas in New York they accounted for more than 8 percent.

Additionally, the hospitalizations of children with AIDS are not evenly distributed throughout the hospital system but are highly concentrated in a few hospitals—quite often, public hospitals. This is due in part to the insurance status of many of the children. In both California and New York, Medicaid pays for a substantial proportion of care. Eighty-one percent of children (as compared with 43 percent of adults) hospitalized with AIDS in New York were covered by Medicaid; an additional 5 percent lacked insurance altogether. In California, 27 percent of children hospitalized with AIDS had Medicaid coverage.

Because children with AIDS rely heavily on the public sector for insurance coverage, their care is generally provided in public hospitals. In New York City, for example, public hospitals provide a disproportionate share of services to children with AIDS: although they account for only 20 percent of all hospital beds in the city, they provide nearly 50 percent of all bed days for children with AIDS.

Developing Early Medical Intervention Services for HIV-Infected Children

The reliance on the public health care delivery and financing system noted above underscores the socioeconomic realities of pediatric HIV infection that make the development of early intervention services for children a demanding task. Other factors that complicate the development of a successful early treatment program for HIV-infected children include difficulties in diagnosing infection in young infants, variability in the natural history of infection, the limited armamentarium of medical therapies, and the almost uniform economic privations that characterize the lives of HIV-infected women and their children. The diagnostic technology and therapeutics required for an effective early intervention program are likely to improve substantially in the future; the social and organizational impediments to such a program may be more difficult to overcome.

Whether it is possible to mount an early medical intervention program for HIV-infected children remains to be determined. The first step toward

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

this goal is an assessment of the magnitude of the costs of providing early intervention services to the pediatric population. To develop such an assessment, one must first determine the size of the treatment population. In one preliminary model that has been offered, it is assumed that 5,000 to 10,000 children are infected with HIV and that 2,500 have been either diagnosed with AIDS or have died (and thus are subtracted from the number of infected children). Using the Centers for Disease Control (CDC) estimate of approximately 1,700 new cases of HIV infection in children per year (which is added to the above range of total infected children) results in a potential treatment population of 4,200 to 9,200 cases of pediatric HIV infection in the United States.

Next, one must estimate the annual early intervention treatment costs per child. To derive this estimate, an assumption has been made that 25 percent of infected children will be treated with zidovudine in their first year of life and 75 percent will be treated in their second year. The estimate also includes costs for other drugs and dosage schedules, medical monitoring, and primary medical care; it does not include hospitalizations. The estimated cost per child per year for pharmaceuticals is $1,583 (55 percent of total costs); ambulatory care costs total $1,319 (45 percent of total costs).16 Total estimated early intervention treatment costs per child per year are $2,902. This estimate should be regarded only as a baseline figure, however, because it does not include hospitalizations. A richer picture of the economic burden of pediatric HIV disease would include not only hospital length-of-stay data but an assessment of the fiscal repercussions for the foster care system and other social service and housing providers.

Combining the estimates of the population eligible for intervention and the annual treatment costs allows construction of a national costing model for early medical intervention among HIV-infected infants and young children. For example, using a midrange estimate of the number of infected children in the United States (i.e., 6,700), the cost of early intervention, if 50 percent of this population were in treatment, is $9.7 million; if 100 percent of children were in treatment, the figure would rise to $19.4 million. Even given a high-range estimate of the number of infected children (i.e., 9,200), the cost of early treatment of 100 percent of the pediatric population is $26.7 million—still less than 1 percent of total national expenditures for AIDS treatment.

16  

 Because of the need to monitor infants closely, their primary care coats are relatively higher than those of adults.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

Paying for Services for HIV-Infected Children

Despite what appears to be a relatively small financial investment for the provision of early medical intervention services for HIV-infected children, delivering and paying for such services in the context of the current health care delivery and financing system remain a challenge. Pediatric HIV disease has underscored the existing limitations and inadequacies of this system, particularly for children with chronic illnesses. Providing and financing comprehensive, long-term care for chronic conditions have always been problematic in this country, a situation certainly not unique to AIDS, although the difficulties involved may be more severe for families affected by HIV than for those affected by other chronic diseases. Chronically ill children have often relied on their families to provide the needed home care, custodial care (as well as room and board), informal nursing care, and other services that are not covered under insurance plans. In the context of HIV infection, family-provided care may not be generally available because many families affected by this disease often live under conditions of social disorganization, poverty, and substance abuse, as well as homelessness in some instances.

Another problem is the economic burden that women and children with HIV infection represent for state Medicaid programs. As noted earlier, Medicaid is likely to be the primary source of health insurance for these individuals. Poor pregnant women in particular are the population most likely to be insured through Medicaid as a result of changes in eligibility criteria. In all states, pregnant women are currently eligible for Medicaid if their family incomes are no higher than 133 percent of the federal poverty level, which is about $13,000 per year for a family of three. (In some states, they remain eligible if their incomes are as high as 185 percent of the poverty level.) Medicaid eligibility for these women concludes at 60 days postpartum, however, leaving many of them without insurance altogether unless they continue to qualify on another basis (e.g., eligibility for Aid to Families with Dependent Children [AFDC] through which a woman automatically qualifies for Medicaid coverage).17

Medicaid eligibility criteria for children also vary widely across states. For infants, eligibility now ranges from 133 percent to 185 percent of the federal poverty level, depending on the state of residence. In all states,

17  

 Some children cannot reside with their mother because of her illness or inability to provide adequate care. This raises the question of whether states should continue to consider these children dependents (despite the fact that they are not living with their mother) and whether the mother should be eligible for AFDC payments.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

Medicaid coverage is extended to children up to age six in families with incomes of up to 133 percent of the poverty level.

One mechanism that may prove useful for providing and paying for the care of HIV-infected children (and young mothers under age 21) is the Early and Periodic Screening, Diagnosis, and Treatment (EPSDT) program, a mandatory service under Medicaid that is best known for its preventive orientation. The EPSDT program provides preventive care and treatment for low-income, Medicaid-eligible children through age 21 and covers periodic evaluation of the child's health, developmental, and nutritional status; it also offers vision, dental, and hearing care as well as treatment for basic acute and chronic problems. There is no requirement in the program, however, that treatment be provided only for basic needs.

Congress recently modified and expanded the EPSDT program through amendments to the Omnibus Budget Reconciliation Act of 1989. These amendments mandate state reimbursement for interperiodic screenings and for vision, hearing, and dental services, whenever they are medically necessary to identify or diagnose a suspected illness or condition. In addition, once a condition or illness is discovered, states must provide any medically necessary follow-up, diagnostic, or treatment services reimbursable under Medicaid, even if the state plan does not otherwise cover such services. These program changes became effective April 1, 1990; it will be some time before their impact can be assessed.

The expansion of EPSDT may offer the most logical source of financing for pediatric HIV-related care; however, several other programs may also provide some assistance. For example, the Title V Maternal and Child Health Block Grant program (which allocates funds directly to states for various services for pregnant women, infants, and children) could be used to finance services for HIV-infected women and children. The Community and Migrant Health Centers program may offer another financing stream. This program funds primary health care clinics in medically underserved urban and rural areas, and many of these clinics, by virtue of their location, can be found in areas of highly concentrated HIV infection. Over the past several years, these clinics have been recognized as logical providers of HIV-related care and have begun to receive supplemental perinatal grants and funds for the development of programs for persons with HIV infection. Both the Title V and the Community Health Centers programs are in urgent need of additional resources, however, because general funding levels for these programs have been tapering off during the past decade.

Despite Medicaid coverage for many HIV-infected women and children, the access of these groups to appropriate care is not ensured. Many pregnant women, as noted earlier, may lose their coverage 60 days after delivery; in some states, coverage for children beyond the age of six is not

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

guaranteed. Moreover, Medicaid reimbursement rates may be inadequate, and as a result health care providers and institutions may be reluctant to accept Medicaid patients.

Medicaid is inherently a disease-neutral program (i.e., specific diseases are not targeted for preferential consideration), and many experts argue that such disease neutrality should be sustained in the development of remedies to the above problems. One approach has been to use home-and community-based waivers, which allow states to cover various outpatient and community-based services under Medicaid for individuals (including persons with AIDS) who would otherwise require institutional or inpatient care. In addition, some experts suggest looking to other programs, such as Supplemental Security Income (SSI) and AFDC, as possible sources of funding for care, particularly custodial care. In short, a single solution to the problem of adequate financing of care for HIV-infected women and children is unlikely. Instead, a consortium of approaches may be necessary to address the inadequacies that exist in the current system of delivering and paying for care and, in particular, chronic and long-term services.

EVALUATING THE EFFECTIVENESS OF HIV SCREENING

In the current environment of limited health care resources and fiscal restraint, policymakers must make difficult decisions regarding the prudent allocation of resources among various health programs. Whether HIV screening of pregnant women and newborns constitutes an efficient, appropriate use of resources must be determined in light of other competing interests and programs, quite often in the absence of complete data. Once a screening program is in place, evaluation is necessary to determine its effectiveness. Thus, before screening is introduced, a well-articulated evaluation component must be incorporated into the program's design to monitor its effects and assess whether program objectives have been achieved.

Methodological Issues in Cost-Effectiveness Analysis of Prenatal and Newborn HIV Screening

An often useful economic tool in making resource allocation decisions is cost-effectiveness analysis, which measures how well a particular program uses health resources to improve health or extend life. The cost-effectiveness ratio is the yardstick for cost-effectiveness analysis. This ratio compares the net burden of the program on health resources (i.e., its cost)

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

with some measure of its health impact—for example, years of life saved or quality-adjusted years of life saved18 as a result of the program. Drummond and colleagues19 have developed a helpful framework for designing a cost-effectiveness evaluation of a health program, which comprises the following steps:

  • Identify the alternative policy strategies.

  • Specify the viewpoint of the evaluation.

  • Review the evidence of program effectiveness.

  • Identify the health consequences and costs of the program.

  • Define the units for counting up these consequences and costs.

  • Specify the valuation scheme (or weighing) of health consequences and costs.

  • Discount future costs and health consequences to present value.

  • Compare incremental costs and health consequences of competing alternatives to derive a cost-effectiveness ratio.

  • Perform sensitivity analyses with respect to uncertain parameters, costs, and valuations.

  • Consider other issues of concern to decision makers that lie outside the scope of the analysis.

The application of this model to HIV screening and early intervention programs for pregnant women and newborns, although complex, offers some direction for the types of data required to conduct a cost-effectiveness analysis.

Three policy alternatives are currently under consideration. The first is to screen pregnant women for HIV. Infected women are then counseled about the implications of this diagnosis for their own health and that of their infants, the possible impact of pregnancy on disease progression, and reproductive options. Next, the infected woman and her caregiver must decide whether to initiate antiretroviral therapy or prophylaxis against opportunistic infection. If the mother does not elect to terminate her pregnancy, a decision must be made regarding treatment of the newborn, that is, whether to begin antiretroviral or prophylactic therapy in the neonatal period or to follow the infant closely with the intent to treat once HIV infection can be confirmed. A second strategy would be to test the newborn and not the mother. If this alternative were selected, the above considerations regarding treatment of the infant (and mother) would

18  

 This measure attempts to reflect the quality of the years of life saved-that is, the health status or quality of life of an individual during the years of extended life.

19  

 M. F. Drummond, G. L. Stoddart, and G. W. Torrance, Methods for the Economic Evaluation of Health Care Programmes (New York: Oxford University Press, 1987).

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

apply. A third strategy, of course, would be to test neither the pregnant woman nor the newborn.

In conducting a cost-effectiveness analysis of prenatal or newborn HIV screening, assessment of the evidence of program effectiveness involves an examination of HIV diagnostic capacity in women and infants and consideration of the interventions that are possible for these populations. The HIV testing algorithm—composed of the ELISA and Western Blot tests—is highly sensitive and specific in adult populations, although the predictive value of a positive test (i.e., the probability that an individual with a positive test result is, indeed, infected) diminishes as the prevalence of infection in the population declines. In newborns, however, conventional HIV antibody tests cannot differentiate those who are truly infected from those who carry passively acquired maternal antibody.

Several aspects of intervention for HIV-infected pregnant women should be considered. Whether a woman's knowledge of her infection will necessarily promote the adoption of less risky behaviors is unclear, although the possibility exists that some women may alter their sexual or drug-using practices in light of their infected status. However, a substantial reduction in horizontal transmission of infection is unlikely to result from prenatal HIV screening at this time. Therapeutic interventions (e.g., zidovudine therapy and PCP prophylaxis) among HIV-infected pregnant women with severe immune deficiency are likely to be effective in delaying progression and minimizing symptoms of disease, given the evidence of effectiveness of these regimens for nonpregnant adults. The potential effectiveness of maternal chemoprophylaxis or immunotherapy to prevent perinatal HIV transmission, however, remains an empirical question. There is also the possibility of teratogenicity of maternal therapy, although the maternal risks from deferring therapeutic intervention would seem to outweigh the potential fetal risks at this time.

The effectiveness of therapeutic intervention among HIV-positive infants is still open to question. Because about two-thirds of these infants are likely to be uninfected, administering therapy to all seropositive infants means that uninfected infants would be exposed to substantial toxicity without deriving any medical benefit. Hence, most pediatric experts generally agree that such therapy should be reserved for children with known HIV infection. Delaying treatment of the infant until infection is confirmed, however, risks the loss of the child to follow-up care. Even when infection has been confirmed, the medical benefit associated with early therapeutic intervention for asymptomatic HIV-infected children is still uncertain.

The costs associated with prenatal HIV screening, subsequent follow-up, and medical intervention are varied. There are the costs of identifying the target population for screening, which may include outreach costs; the

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

costs of HIV-antibody testing itself; and the costs of contacting and notifying the pregnant woman of her test results. Therapeutic abortion costs, should that option be elected by the woman, must also be considered. In addition, there are the costs of medical surveillance and care, as well as antiretroviral therapy and PCP prophylaxis (which should actually be calculated as the difference between the costs of early treatment and the costs of delayed treatment—that is, the net costs or savings of early treatment versus later treatment). Also to be included are the costs associated with the follow-up of infants born to HIV-infected mothers and subsequent therapeutic interventions for these infants, again, measured as the difference between the costs of early treatment versus later treatment.

Consideration of the costs of possible adverse effects, such as drug toxicities in the mother or child, is another aspect of calculating cost-effectiveness. Additionally, one might wish to include the costs of medical care that would be incurred during the years of life added by early intervention for the mother or child. If affected infants were aborted, there might also be savings of the costs associated with treating and caring for these children—both the health care costs and the costs to the parent or government of custodial and long-term care.

In determining the cost-effectiveness of a screening program, these program costs would be measured in current dollars, with future costs corrected for inflation. Nevertheless, some outstanding questions remain. Should health care cost savings as a result of elective abortion be included in the analysis? Should the health care costs arising from life extensions be included and added to the costs attributable to this early intervention program?

The health consequences of the program can be assessed from the viewpoint of the woman, the child, the potential contacts of the woman, and other interested parties, such as insurance companies and society at large. Health outcomes for the woman might include gains in life expectancy as a result of early medical intervention and improved quality of life arising from the avoidance, delay, or amelioration of HIV-related symptoms. There may, however, be losses in quality of life because of the toxicity of treatments. Although difficult to value in such an analysis, there is also the psychological impact of learning of one's HIV infection (and of possible pregnancy termination).

Two aspects of the program's health consequences for the child require examination. First, treatment of the pregnant woman may result in major gains in life expectancy and quality of life for her child, if chemoprophylaxis were, indeed, successful in interrupting perinatal HIV transmission. Yet, as noted earlier, this is still only a theoretical possibility. There may also be possible losses of life expectancy or quality of life for the child as a result of potential teratogenic effects of maternal therapy. Second, early

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

therapeutic intervention for the HIV-infected child may prolong or improve the quality of his or her life, although whether such an improvement is, indeed, the case remains unconfirmed at this point. Alternatively, possible toxicities associated with treatment may negatively affect quality of life.

Other individuals that may be affected by this program include the woman's potential sexual and needle-sharing partner(s). They might gain life expectancy or quality of life as a result of the avoidance of HIV infection or as a result of early medical intervention made possible through partner notification. There may also be psychological effects of the knowledge of one's own or one's partner's HIV infection.

These health consequences (for the woman, child, and others) can be measured in terms of the number of years of life gained, improvements in health status or quality of life (e.g., physical and social functioning, physical, psychological, and emotional symptoms), and the number of abortions that might be performed as a result of the program. Valuation of these health consequences is problematic, however. For example, how does one value an extension of the life of a child, from what would have been the child's life expectancy with untreated HIV infection to the potential life expectancy if the HIV-infected child were treated early? How does one weigh the health consequences for the mother against the possible health consequences for the child and others, particularly when these consequences may be in conflict? What value should be assigned to an aborted HIV-affected fetus? Is this a benefit? Is life with HIV infection worse than no life at all? What value or disvalue does one assign to an aborted fetus that would not have been infected?

Sensitivity analyses provide a means to address the uncertainties inherent in a cost-effectiveness analysis of a prenatal or newborn HIV screening program. Aspects of such sensitivity analyses are listed below:

  • effects of zidovudine on the interruption of perinatal HIV transmission;

  • effects of early treatment on the survival and quality of life of the HIV-infected mother and infant;

  • toxicity of zidovudine in infants;

  • probability that a woman would elect an abortion, given knowledge of her HIV infection;

  • probability that a woman might change her behavior, which might affect subsequent HIV transmission;

  • probability of HIV infection in a woman, given her risk status;

  • possibility of new, effective treatments, after the child has been kept alive;

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
  • probability of the loss of the child to follow-up care, if the mother is not screened prenatally; and

  • future costs of zidovudine and other therapeutic agents.

Finally, there are a number of concerns that emerge when prenatal and newborn HIV screening are considered, but incorporating them directly into the cost-effectiveness analysis may not be possible. Decision makers must nevertheless be mindful of these issues—the psychological effects of knowledge of one's infection status, ethical questions that arise when high-risk pregnant women (which might translate into minority pregnant women) are targeted for screening, concerns about confidentiality, and the ethical status of abortion.

A Cost-Effectiveness Analysis of Prenatal HIV Screening

Renata Kiefer and colleagues at the Center for AIDS Prevention Studies, University of California, San Francisco, have conducted a cost-effectiveness analysis of voluntary (with informed consent), confidential prenatal HIV screening, comparing maternal and infant outcomes and the costs of alternative, prenatal HIV-antibody screening strategies. The analysis examines three strategies: (1) universal screening (HIV testing is offered to all pregnant women regardless of risk status) with posttest counseling for all women screened; (2) universal screening with posttest counseling for HIV-infected women only; and (3) selective screening (HIV testing is offered only to women with self-identified risk factors, based on prior risk assessment) with posttest counseling for all women screened.

The outcomes and costs considered in the analysis are the number of HIV-infected women identified through the screening program, the cost per identified woman, the number of infants detected who are at risk for infection, the number of averted births of infected infants if the women elect to terminate their pregnancies after learning of their infected status, the number of pregnancies with uninfected fetuses that would be terminated, and the cost per averted birth of an HIV-infected infant. The probabilities used in the decision analysis include estimates of several possible prevalence levels, ranging from high to low. For the HIV testing sequence (i.e., the ELISA and Western Blot tests), a sensitivity of 0.99 and specificity of 0.99999 were chosen.20 A perinatal HIV transmission rate

20  

 These figures are consistent with false-positive and false-negative rates reported in CDC publications and those documented by reference laboratories and the military screening program.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

of 35 percent was selected, although current estimates range from 25 to 35 percent. Other variables are as follows:

  • The prevalence of risk factors in the pregnant population—that is, an estimate of the distribution of (rather than actually acknowledged) risk factors.

  • The probability that HIV-infected women will be tested, which depends on the overall acceptance of testing. If HIV testing is offered to all women (universal screening), an estimated 95 percent of infected women might accept testing. If, however, testing is offered only to women with acknowledged risk factors, an estimated 50 percent of infected women might accept testing. (This latter probability is the product of two other probabilities—that of detecting risk factors through pretest risk assessment and of acceptance of testing, given the identification of risk.)

  • The probability of pregnancy termination in light of HIV infection. This probability, estimated at 30 percent, is also the product of two probabilities—the probability that a woman will seek prenatal care early enough for pregnancy termination to be an option and the probability that she will elect termination on learning of her infection.

The cost assumptions used for this cost-effectiveness analysis are detailed in Appendix B.

The results of the analysis indicate that universal screening identifies almost twice as many HIV-infected women as selective screening (Table A-1). This is an important consideration if the goal of the screening program is ultimately to identify all infected pregnant women for the purpose of early intervention. As expected, the table shows that the cost per HIV-infected woman identified declines with increasing prevalence of infection. Additionally, the costs of universal screening with posttest counseling for only HIV-positive women compare favorably with the costs of selective screening with posttest counseling for all women. Universal screening with posttest counseling for all women is considerably more expensive.

The infant outcomes for each screening strategy (assuming a pregnancy termination rate of 30 percent and a perinatal transmission rate of 35 percent) are presented in Table A-2. In short, two uninfected infants would be lost for each HIV-infected infant. With universal screening, twice as many births of infected infants would be averted as with selective screening.

The cost per averted birth of an HIV-infected infant is shown in Table A-3. The cost-effectiveness of all screening strategies improves as HIV prevalence levels increase. Universal screening with posttest counseling for HIV-infected women only offers the lowest cost per averted birth of an

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

TABLE A-1 Number of HIV-Infected Pregnant Women Identified (per 100,000 pregnant women) and Cost per HIV-Infected Woman Identified Using Selective or Universal Screening at Various Prevalence Levels and with Two Counseling Scenarios

 

 

Selective Screening

Universal Screening

Estimated Prevalence

Total No. of Infected Pregnant Women

No. Detected

Cost for Counseling All

No. Detected

Cost for Counseling All

Cost for Counseling HIV-Infected Women Only

0.02

2,000

1,000

$ 300

1,900

$ 700

$ 300

0.01

1,000

500

600

950

1,400

600

0.002

200

100

2,700

190

6,600

2,600

0.001

100

50

5,400

95

13,100

5,100

0.0002

20

10

26,600

196

5,100

25,100

0.0001

10

5

53,100

9.5

130,100

50,100

Note: Calculations prepared by R. Kiefer, E. Washington, N. Hearst, and S. Hulley, adapted from their study, ''Costs and Benefits of Prenatal HIV Screening,'' presented at the Sixth International Conference on AIDS, San Francisco, California, June 20-24, 1990. The authors are affiliated with the Center for AIDS Prevention Studies, University of California, San Francisco.

TABLE A-2 Infant Outcomes (per 100,000 pregnant women) for Universal, Selective, and No Screening Strategies at Various Prevalence Levels

Prevalence and Strategy

Number of Infants at Risk

Number of Infants at Risk Identified

Number of Averted Births of Infected Infants

Number of Pregnancies with Noninfected Infant Terminated

0.02

Universal

2,000

1,900

200

371

Selective

2,000

1,000

105

195

No Screening

2,000

0

0

0

0.002

Universal

200

190

20

37

Selective

200

100

10

20

No Screening

200

0

0

0

0.0002

Universal

20

19

2

4

Selective

20

10

1

2

No Screening

20

0

0

0

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

TABLE A-3 Cost per Averted Birth of an HIV-Infected Infant Using Selective or Universal Screening at Various Prevalence Levels and with Two Counseling Scenarios

 

 

Universal Screening

Prevalence

Selective Screening with Counseling for All

Counseling for All

Counseling for HIV-Infected Women Only

0.02

$ 3,300

$ 7,000

$ 3,300

0.002

26,000

62,700

24,700

0.0002

253,200

619,800

239,000

infected infant. If the lifetime cost of HIV-related care for an infected infant were $50,000, then both selective screening and universal screening with counseling for infected women only would result in a net gain at an overall HIV prevalence of 2 infected women per 1,000 women. If counseling costs were to increase beyond the conservative estimates used in this analysis, however, then all screening strategies would be more costly than under the current assumptions.

Other outcomes of prenatal HIV screening that should be considered when selecting a particular screening strategy are the possibility of reduced maternal and infant morbidity as a result of early diagnosis and treatment, facilitation of health resources planning, potential reduction of vertical HIV transmission, and possible reduction of horizontal HIV transmission to partners (which will depend to some extent on the capacity of the program to identify and counsel infected women). Additional benefits can be pursued by counseling seronegative women who continue to be at risk for infection, particularly women in selective screening programs. Policymakers must decide the most efficient use of resources. In areas of high prevalence, universal screening with posttest counseling for all women screened may be the most appropriate option.

Program Design and Evaluation

The design of a screening program and its evaluation require that the program's overall goals be clearly defined at the outset. (For example, the objectives of a prenatal HIV screening program might be to prolong, through early medical intervention, the productive lives of infected individuals identified by screening, to prevent further (vertical or horizontal)

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

transmission of HIV, and to minimize the social costs of HIV disease.) Decisions must also be made regarding the target population to be screened and the site(s) at which such screening is to be conducted. If a majority of infected pregnant women are to be identified and offered treatment, it may be most appropriate to offer voluntary HIV testing to all pregnant women prenatally or at the time of delivery, in a given geographic area. Reaching women at increased risk for infection means that HIV testing services need to be available in settings where these women frequently receive care—prenatal, family planning, and sexually transmitted disease clinics, as well as substance abuse treatment centers. As mentioned earlier, HIV screening confined to the prenatal care setting may miss a substantial proportion of at-risk pregnant women because those women who receive little or no prenatal care are often at greatest risk of infection.

Consideration must also be given to the development of a follow-up system of care for infected individuals who are identified through screening. Indeed, some health officials have questioned the wisdom of instituting screening in the absence of an adequate system of follow-up services for infected women and their children. Others have suggested that only when infected individuals are identified through screening will the necessary services be developed.

Once an individual has been identified as infected, there should be a two-stage system of early intervention. The first stage would be oriented toward prevention and would include behavioral intervention, adequate psychosocial support services to minimize the social disorder that accompanies the lives of many HIV-infected women, and couple (or partner) counseling. Such intervention may facilitate behavioral change to limit further HIV transmission. (Prevention efforts can also be targeted at the community level in areas where HIV infection is highly concentrated.) The second stage would focus on early medical intervention—for example, continuing medical surveillance, drug therapies, acute hospitalization (if necessary), community-based care, and substance abuse treatment (if needed), as well as clinical, psychosocial, and practical support services.

Any HIV screening effort relies heavily on laboratory services, and as a result laboratory quality assurance and performance evaluation are critical components of a screening program. Considerable variation across laboratories in the costs of test performance is an important factor in selecting laboratories to process specimens collected through the screening program. Other crucial elements of screening programs are a mechanism for contacting and notifying individuals of their test results and posttest counseling services (particularly for HIV-infected individuals).

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

In developing and performing an evaluation of a prenatal HIV screening program, several aspects of the program may be assessed. For example, one can evaluate whether the program has been implemented as intended and whether program components are in place. To assess the acceptance and use of screening services among pregnant women, one might compare the estimated number of infected women expected to be reached through screening with the number of seropositive women actually identified. Examination of the proportion of these women who are receiving appropriate medical monitoring, treatment, and other support services can also provide some sense of the adequacy and accessibility of the system of follow-up care. Evaluation of prenatal HIV screening will undoubtedly present a challenge, in part because the HIV epidemic continues to evolve and new information continues to accumulate regarding the effectiveness of therapy. Ongoing evaluation will therefore be necessary to assess program effectiveness and to adjust program design in response to evaluation findings and technological developments.

THE ADVOCATES' VOICES: THOUGHTS ON PRENATAL AND NEWBORN HIV SCREENING

The epidemic of HIV infection and AIDS among women (and children) has disproportionately affected minority populations, particularly those in large urban areas of the United States. In developing prenatal HIV screening policy with the goal of early identification and treatment of infected women, this fact—and the implications it brings—must be taken into consideration. Screening policy cannot be developed in a vacuum, that is, without reference to the populations it is meant to serve. In the case of HIV, the social circumstances, culture, and character of these populations will shape their members' acceptance of and degree of participation in the screening process.

Screening policy development must consider variations among ethnic communities as well as among individuals with respect to educational level, prior experience with the health care system, attitudes, values, and fears, and the extent to which there are social and familial support systems. For example, women in different areas of the country may have received varying levels of information about HIV infection and testing, depending on whether the area has a high or low prevalence of infection. A woman's level of awareness about HIV may affect what she understands during the informed consent procedure and how much information must be provided by the counselor to secure an informed consent. Eliciting informed consent may be particularly problematic with women who have limited education or who are functionally illiterate and to whom the meaning and

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

intent of the forms and, indeed, the whole process may be unclear and confusing.

Diversity with respect to previous experience with the health care system may affect the way women respond to offers of HIV testing. Pregnant women who have had experience with preventive or prenatal care are familiar with medical testing. They may not necessarily understand the specific details of such testing, but they believe the tests are being performed for the benefit of their unborn child. Many minority women, however, have had little contact with the health care system. For example, about 20 percent of Latinos nationally lack a regular source of health care. In 1988, at least 12 percent of all Latino mothers received late or no prenatal care, compared with 4 percent of white mothers.21 At the same time, the birth rate was 24.1 births per 1,000 persons for Latinos and 15.7 per 1,000 for non-Latinos.22

Latinos as a population appear to be relatively unfamiliar with the health care system, for reasons that go beyond the lack of available services or accessibility of care. Some individuals, for example, recent immigrants, are concerned about their legal status and the effect that use of the system may have on them and their families (e.g., the potential for deportation). Others may be unaware of existing resources. Even when it is known that services are available, people may be reluctant to enter a system that is unfamiliar, intimidating, and characterized by long waits for care and overcrowded, understaffed facilities. This relative lack of use of health care resources is an important factor that must be considered when formulating policy that will ultimately be implemented in physicians' offices and prenatal care facilities.

Policymakers also need to recognize the diversity of values, cultures, and social circumstances among individuals within a minority group as well as among minority groups. For example, there is a wide range of opinion among minority women about whether screening should be mandatory. There are contextual differences among women as well. Some pregnant women face drug abuse problems, some are coping with the effects of poverty, and some have empowerment concerns. Some women, in fact, confront all of these issues. The implications of a positive HIV test result may differ depending on the particular individual involved. A woman's reality is not limited to the medical facts of her situation, which may or may not be fully understood. It also includes psychological and social components often associated with unemployment, undermined self-esteem, perceptions of locus of control, poverty, and drug addiction. HIV

21  

 In 1988, 11 percent of black mothers received late or no prenatal care.

22  

 National Center for Health Statistics, "Advance Report of Final Natality Statistics, 1988," Monthly Vital Statistics Report 39(1990):1-48.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

infection often compounds these existing problems. Moreover, adequate support mechanisms may simply not be available for many women. Will the institution of a screening policy lead to immediate and long-term emotional and practical support? Will it assist in the implementation of support systems that will facilitate behavioral change? (These range from additional treatment slots for intravenous drug users to child care.)

Additionally, the counseling that accompanies HIV testing should be sensitive to differences among the women participating in the screening program and the varying cultural contexts of their lives, especially the way that context may shape their perceptions of risk and their ultimate decisions or behavior. For example, women who live in a high-risk environment may not view a roughly 30 percent perinatal HIV transmission rate as high. Those who begin life with very little must take risks to survive; to them, a 70 percent chance of having a healthy baby may seem to be a reasonable risk.

In many ethnic communities, particularly the Latino community, the family is extremely important, and decisions are often discussed with family members. For example, before signing a consent form, a woman might want an opportunity to confer with her family and other members of her support system about HIV testing and its ramifications. Any attempt to deal with the management of HIV infection and its prevention should integrate the family in that process and use it as an emotional resource. Male partners of infected women should be involved in any behavioral change efforts. Indeed, because economic and social dependence on men is largely responsible for the lack of empowerment many minority women experience, the successful implementation of behavioral changes depends on the involvement of men.

Reaching minority pregnant women to encourage them to enter care (providing services are actually available) is another consideration in the development of screening policy. Prenatal HIV screening programs must find a way to identify women early in their pregnancy, which is not an easy task. In addition to the difficulties presented by early identification, some of these women will not enroll in prenatal care, some will discontinue care, and others, such as the substance abuser who is afraid that her child will be taken away from her, will only receive care in the emergency or delivery room.

An HIV screening program must also facilitate and support the reproductive choices a woman makes after being informed she is HIV-antibody positive. Contraception, sterilization, and abortion may have different meanings to different women—and different meanings to health care providers. Many women are afraid that screening will be used to subtly convince or pressure them to be sterilized after delivery. Mandatory

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

or recommended HIV testing may, in fact, deter women from much needed prenatal services.

Other factors affecting a woman's reproductive decisions are the relevance that this pregnancy or this child may have for that woman, her religions and spiritual beliefs, and her sense of empowerment. Issues of empowerment and self-esteem have a substantial impact on why some women choose to participate in risky behavior and on their inability to extricate themselves from their predicaments. Some have questioned whether positive attitudes about wellness and disease prevention can exist among individuals (e.g., poor women of color) whose energies must be devoted to confronting the exigencies of a marginalized existence.

An important concern in prenatal HIV screening is whether health practitioners are aware of their own feelings and attitudes about minority patients. Some practitioners may have formed judgments and developed biases about pregnancy and HIV infection in the ethnic patient. Prejudice among practitioners, however, is only part of a much broader problem of racism. Unlike prejudice, racism is systematic, and to the extent that it is part of U.S. society, it will also be part of the nation's medical centers.

Finally, although funding for AIDS is a politically sensitive issue, for the black and Hispanic community to embrace HIV disease as a political issue is problematic for two reasons. One is the potential for discrimination and stigmatization of the communities involved. The other is that when funding requests are issue specific, a certain amount of money is allocated for that particular issue but other, interconnected problems are ignored. Advocates for women of color fully support HIV screening but not at the expense of other programs and only if it is part of a long-term process to provide pregnant women with services and support that will help improve their lives and the lives of their children.

HIV SCREENING: ARE HUMAN RIGHTS THREATENED?

Concerns about discrimination in the context of the AIDS epidemic have centered around employment, insurance, and housing, as well as social ostracism by friends and family. The potential for discrimination within the health care system itself is another area of particular concern, especially for pregnant HIV-infected women.23 Such discrimination can

23  

 L. O. Gostin, "The AIDS Litigation Project: A National Review of Court and Human Rights Commission Decisions, Part 1, The Social Impact of AIDS," Journal of the American Medical Association 263(1990):1961-1970, and L. O. Gostin, "The AIDS Litigation Project: A National Review of Court and Human Rights Commission Decisions, Part 2, Discrimination," Journal of the American Medical Association 263(1990):2086-2093.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

be relatively direct—a physician's reluctance to provide treatment—or subtle—a decision made by a health care provider to refer all cases of HIV infection to other practitioners because of insufficient expertise or resources. Infected persons may also experience problems gaining access to nursing care, foster care, and adoption services.

To avoid such discrimination, some pregnant women refuse to be tested, a decision that raises the questions of a woman's rights versus her obligations toward others who have an "interest" in her pregnancy, that is, her sexual partner, her health care practitioner, and her baby. In terms of her own interests, the most compelling rationale for testing is that, if a woman is found to be infected, she may have access to medical care, such as monitoring of CD4+ cell counts, prophylaxis for PCP, and antiretroviral medication. Neither Medicaid coverage nor any other kind of support, however, can guarantee that she will actually obtain such care because services may simply not be available or they may be inaccessible w pregnant women. Moreover, for many women, the negative effects of being identified as HIV infected may appear to outweigh any benefits that might accrue to the early identification of future illness. (Such effects may include the immediate costs of violence, which can ensue when a woman's sexual partner discovers her infection or when she tries to make changes in their sexual practices; the loss of a husband or boyfriend; and ostracism by family and friends.)

Once a woman is known to be infected, the question of disclosing that knowledge to various parties, in particular, her sexual partner or partners, is raised. Some women may prefer not to disclose such information (for the reasons noted above, among others), but health care professionals have certain legal and ethical responsibilities to warn third parties at risk. Doing so without the woman's permission, however, may in fact generate other public health risks. For example, a major public health goal is to ensure that pregnant women seek prenatal care. Actual or perceived breaches of confidentiality by a woman's obstetrician may cause her to lose trust in that caregiver (and sometimes in all such caregivers), with the result that she will not continue to seek care.

The case has been made that health care providers have a right to know the infection status of their patients to protect themselves from transmission of the virus. Yet the bulk of this type of infection risk can be virtually eliminated by strict adherence to universal infection control procedures. Because the risk of transmission in the health care setting is relatively low, many believe that there is little justification for compelling a woman to disclose her infection status to the health care practitioner treating her.

With regard to the interests of the baby, the lack of clear medical benefit to be gained from neonatal testing—and the difficulty in distinguish-

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

ing those babies who are actually infected—argue against coercion of either testing (of mother or infant) or a woman's reproductive choices. Many pregnant women believe that their baby's best interests are served by being born and nurtured, despite their own potential disability. In such cases, fundamental constitutional rights to privacy and reproductive choice make it difficult, both ethically and legally, to authorize the state to unduly influence a pregnant woman's decision either to continue or terminate the pregnancy. Moreover, the U.S. tradition of parental authority gives mothers the right to make decisions about their babies' health care—unless such a decision would harm the child. Thus, if a reliable HIV test that could detect infection in neonates were developed, and if treatment were shown to be effective, available, and of benefit to the child, the state would have the right to mandate HIV testing of children, regardless of the mother's choice. Absent these conditions, the mother's decisions regarding her baby's welfare must be respected.

A related issue involves the discrimination some HIV-infected women experience in attempting to gain access to reproductive health care. One attempt to document the degree to which HIV-infected women were discriminated against in gaining access to abortions revealed that two-thirds of the abortion facilities contacted in the study canceled appointments made by allegedly HIV-infected women once their infection status was disclosed. Some of the facilities attempted a plausible response, such as their inability to handle that type of procedure. Others changed the vacation schedules of the physicians or quoted inflated prices for abortion services to discourage those seeking care. Still others openly reported that their staff refused to care for HIV-infected patients. These types of action by caregiving facilities may be actionable under law as discriminatory, and civil sanctions could be applied in most states under handicap legislation.

There are problems, however, in invoking the legal system and filing lawsuits against such offenders. A victory in a discrimination case of this kind may force certain medical professionals to treat people they do not want to treat, but it does not ensure good care. Only a minority of the problems faced by HIV-infected women seeking reproductive health care can be addressed by antidiscrimination laws because such laws cannot check the social forces of prejudice that leave infected persons vulnerable to discrimination. These laws are based on neutral principles of fairness that presuppose an homogenization of experience and culture rather than acknowledge the diversity that actually exists. Perhaps a more productive approach to antidiscrimination would be to work with city and state health departments to inform reproductive health care providers that these practices run contrary to current medical knowledge and violate federal and local civil rights laws. Peer education, not litigation, is most likely to

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

be effective in improving the quality of reproductive health care provided to HIV-infected women.

Identification of a woman as HIV positive puts her at risk not only of discrimination but of coercion by the judicial system. The proclivity to view women as either "vectors or vessels" is reflected in the increasing tendency to arrest women who use drugs during pregnancy and to seek so-called judicial interventions into obstetrical medicine (e.g., forced cesarean section, sterilization). An analysis of precedents in this area sheds some light on the degree to which these practices would be constitutionally permissible and judicially sanctioned for HIV-infected women.

In Buck v. Bell (1927), the Supreme Court upheld the right of the state to require that some citizens be forcibly sterilized. One justification for requiring this "sacrifice" was the state's concern about the costs of caring for institutionalized children. Beginning in the 1950s, attention was directed toward women on welfare. Legislation was introduced to require sterilization in situations where the state did not want to pay for the rearing of children born to parents who could not take care of them. Although this legislation was not passed, many women on welfare were in fact sterilized without their consent. Therefore, at several points in U.S. history, public opinion has coalesced around solutions to prevent society from caring for children born to parents who are deemed unworthy. These parents were all too often the economically disadvantaged and, in many instances, racial and ethnic minority women.

Counteracting such discriminatory laws are laws that protect the right to privacy. Griswold v. Connecticut, although currently being challenged, suggests a legal basis for a fundamental right to make individual reproductive decisions.24 These cases, however, deal with the right not to have a child, that is, to use contraception or abortion. Whether the Court would deem sterilization unconstitutional because of a right to procreate is more controversial.25 Moreover, the "image" of sterilization is changing, at least among middle-class couples, for whom it is becoming a preferred method of birth control, and this change may soften the abhorrence of coerced sterilization that has developed over the past few decades. If coerced sterilization remains constitutional, it will be very difficult to mount a constitutional argument against mandating long-term, possibly injectable contraceptives that will have the same result.

24  

 Griswold v. Connecticut held that a couple has a constitutional right to access to contraceptives, which stems from their right to privacy in making reproductive decisions.

25  

 Many believe that there is stronger protection of the right to have children than there is of the right not to have children. Thus, forced sterilization would be unconstitutional even if abortion were no longer considered to be constitutionally protected.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×

The recent precedent of coerced cesarean sections raises the question of whether maternal treatment (e.g., therapy to prevent perinatal transmission) could be forced on an infected woman to benefit her infant. Recent legal trends lean toward ensuring benefit to the fetus and may require some women to act in a manner that they perceive as disadvantageous to them. Ways must be found to limit coercion and ensure protection of the rights of HIV-infected women as well as those of their infants.

Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
This page in the original is blank.
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 69
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 70
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 71
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 72
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 73
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 74
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 75
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 76
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 77
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 78
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 79
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 80
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 81
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 82
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 83
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 84
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 85
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 86
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 87
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 88
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 89
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 90
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 91
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 92
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 93
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 94
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 95
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 96
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 97
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 98
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 99
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 100
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 101
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 102
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 103
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 104
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 105
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 106
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 107
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 108
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 109
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 110
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 111
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 112
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 113
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 114
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 115
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 116
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 117
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 118
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 119
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 120
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 121
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 122
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 123
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 124
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 125
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 126
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 127
Suggested Citation:"Appendix A: The Conference: Program and Summary." Institute of Medicine. 1991. HIV Screening of Pregnant Women and Newborns. Washington, DC: The National Academies Press. doi: 10.17226/1746.
×
Page 128
Next: Appndix B: Cost and Probability Assumptions Used in Cost-Effectiveness Analysis of Prenatal HIV Screening »
HIV Screening of Pregnant Women and Newborns Get This Book
×
Buy Paperback | $50.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Proposals for screening pregnant women and newborns for HIV infection have provoked much controversy. This volume analyzes the possible goals of such screening programs and assesses whether these goals can currently be achieved. It also provides guidance to policymakers in developing and implementing sound screening policy.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!