- Shopping Bag ( 0 items )
But the European Union hasn't gotten over it. The E.U. has threatened to impose trade sanctions on the U.S. for failing to protect personal information. That's part of the reason President Clinton appointed a new "privacy czar" earlier this month.
Amid the increasing hubbub over privacy comes this reasonable book from communitarian thinker Amitai Etzioni, a professor at George Washington University. Etzioni says folks like McNealy are scarier than the state. "The main danger to privacy for people who live in free democratic societies in the cyberage comes from within the private sector, not the government," he writes.
Examining a range of issues from encryption to the use of biometric identifiers, Etzioni seeks to demonstrate the shortsightedness of those who make privacy the highest goal. While Etzioni may gloss over the practical problems posed by encryption (for example, he's unconvincing when he argues that the government's public-key system would work in the face of unbreakable overseas codes), his critique of knee-jerk privacy advocates provides some needed balance.
HIV Testing of
Should Public Health
The citizens of forty-nine states and the District of Columbia must come to terms with the question of whether it is acceptable to violate the privacy of mothers in order to save the lives of some of their children. Can mothers, in effect, be tested for HIV and informed of their HIV status, without their consent, so that they can properly treat their infants? So far only New York State has acted on this vital matter. Congress walked away from it by asking the Institute of Medicine (IOM) to study the matter. IOM released its conclusions in October 1998. Congress has yet to take notice.
This specific context for HIV testing should be seen within the larger context of the treatment of AIDS in general. Few public health issues have raised more emotional confrontations and policy disputes than the way AIDS has not been treated. The stigma that surrounded AIDS until very recently led gay activists and others to oppose some very basic public health efforts. These activists feared that those who contracted HIV might face loss of jobs, housing, and insurance. At the same time, various public health advocates denounced the "exceptionalism" with which AIDS was treated; the prevention measures that have worked well to contain other dreaded diseases, such as tuberculosis and syphilis, have rarely been deployed to slow the spread of AIDS. In this confrontation between the individual rights of people at risk to contract HIV and the needs of public health, one issue has garnered much less attention but deserves much more: The fate of infants born to mothers who have HIV.
Few matters of public concern offer a better opportunity to explore the thesis that there is a pro-privacy bias in contemporary American society than that of the opposition to mandatory testing of newborns for HIV antibodies and disclosure of the results to their mothers. The main arguments for testing infants are that if mothers of infants who are found to have been exposed to HIV are properly informed and counseled, hundreds of infants could be spared severe suffering and major illness, and often death. Also, mothers would be spared much grief, and significant public costs could be avoided.
The main objection to such disclosure is that it violates the privacy of the mother by revealing her HIV status without her consent. As a result of such privacy violation, proponents of privacy argue, many mothers may be stigmatized or lose their health insurance, livelihood, and in some cases even their home. Opponents of disclosure also argue that alternative ways of serving the same public health goal — preventing the spread of HIV from pregnant women to their infants — are more effective and judicious, such as testing pregnant women only after gaining their consent (voluntary testing).
As with many of the issues discussed in this book, the debate about HIV testing of infants is often highly charged; each side demonizes the other by accusing it of seeking to subvert fundamental rights, discriminating against the poor or against gay people, or causing numerous, preventable infant deaths. My attempt here is to advance a civil dialogue. I presume that the motives of all parties to the debate are beyond reproach; disagreements reflect differences of values and goals and conflicting interpretations of the evidence.
A word about terminology. "Proponents" is used in the following examination to refer to those who support unblinded testing, and "opponents" refers to those who oppose such testing. "Unblinded testing" refers to disclosing the results of HIV tests of infants to their mothers and the health care personnel who treat them. In "blind" testing the same tests are conducted but for aggregated, statistical purposes only; information about individual infants is not disclosed to any party, parent or otherwise.
Basic Facts About
HIV Testing Of Infants
Since 1987 the Centers for Disease Control and Prevention (CDC), in conjunction with local health authorities in forty-four states and the District of Columbia, has arranged and paid for blind testing of all newborn infants for the presence of HIV antibodies. The CDC has used the resulting data to assess the level of HIV in the populations studied, both infants and mothers. The information was initially used only for epidemiological purposes.
In 1993 New York State Assemblywoman Nettie Mayersohn introduced a bill that mandated HIV testing for all infants born in the state and required that mothers be informed of the results. The bill, which would be called the Baby AIDS Law, was drafted in response to reports estimating that as many as 60 percent of the infants who tested positive for HIV were leaving hospitals unidentified and untreated.
It should be noted that, technically speaking, standard HIV antibody tests, the ELISA and the Western Blot, do not determine a newborn's HIV status. Rather, these tests indicate the presence of the mother's HIV antibodies in the newborn's blood. The test of infants thus reveals whether the mother is infected with HIV, a discovery that in turn raises ethical, legal, and policy issues. It is also firmly established that a significant number of the infants whose mothers have the disease will develop HIV and ultimately AIDS.
Although the number of infants infected with HIV by their mothers has been declining, the number of new pediatric AIDS cases reported in the United States in just one year alone, from July 1996 to June 1997, was still a troublesome 609. By December 1997 the total number of cases due to perinatal transmission totaled 8,086.
It is crucial for the analysis that follows to realize that the number of children born to mothers who have HIV is much larger than the number who will actually develop HIV, roughly four times so. It is estimated that, without proper treatment, approximately 25 percent of infants born to HIV-positive mothers will contract HIV from their mothers during pregnancy or birth (perinatally), while the other 75 percent will eventually clear their systems of their mothers' HIV antibodies and remain HIV-negative, under special conditions to be discussed shortly.
The fate of the infants is deeply affected by the treatments administered — or not — by their mothers and by health care personnel. Those newborns who have not contracted HIV before or during birth (75 percent) are put at risk if they are breast-fed; a given percentage of them will acquire HIV from their mother after birth in this manner. The CDC estimates that "breastfeeding may increase the [overall] rate of [HIV] transmission [from mother to infant] by 10%-20%," and the World Health Organization (WHO) suggests that "up to one-third of HIV-infected infants are infected through breastfeeding." In a meta-analysis of the findings of several studies that examined some infants who were breast-fed and others who were not, the risk of transmission of HIV through breast-feeding was found to be 14 percent for infants who otherwise would not have developed the illness. For mothers who developed HIV infection during the postpartum period, the risk of transmission through breast-feeding was 29 percent. These infants will develop full-blown AIDS, suffer the full course of the disease, and die. Their illness and death could have been prevented if the mothers had merely been warned not to breast-feed their infants and had heeded the warning. This fact alone is an argument for unblinded testing.
Moreover, there is significant evidence that the lives of the 25 percent born with HIV could be greatly improved and prolonged if mothers and health care personnel were informed of the condition of the infants. Caregivers would be able to provide AZT (and possibly newer drugs) to these infants to reduce their viral loads and provide many corollary benefits. Citing various antiretroviral studies, Leonardo Renna notes: "Studies indicate that ZDV [AZT] treatment improves children's appetites, weight gain and CD4 cell counts. Most importantly, ZDV significantly improves the mental and cognitive development of children."
Health care personnel who know that the infants they are treating have HIV can also provide prophylaxis against the opportunistic infections these infants are particularly prone to getting. These illnesses include, among others, tuberculosis, varicella, meningitis, mycobacterium avium complex, thrush, and particularly pneumocystis carinii pneumonia (PCP). Indeed, according to the CDC, PCP occurs in 37 percent of all HIV-infected infants; of these, 53 percent acquire PCP between three and six months of age. The CDC also reports that the median survival time after infection with PCP is nineteen months.
It should be noted that, at the time of this writing, the benefits of postnatal treatment of infants with AZT, in terms of preventing the development of HIV, are not fully documented. Most trials that have been publicly reported have focused on AZT provided to women during pregnancy, during delivery, and for six weeks after birth. Hence AZT's effectiveness when given only after birth is not clearly established. However, the U.S. Public Health Service recommends that if AZT treatment is not begun before or during delivery, such therapy should be "initiated as soon as possible after birth, preferably within 12-24 hours," and should continue for six weeks.
Furthermore, a 1997 study conducted by the AIDS Institute of the New York Department of Health indicates that "initiation of ARV [antiretroviral] prophylaxis in the newborn period (within 48 hours of birth) may be associated with an intermediate decline in transmission." Among the initial group studied, postpartum administration of ARV prophylaxis resulted in a transmission rate of 8.9 percent, a significant improvement over the 26.5 percent transmission rate from mothers to their infants when no prophylaxis was provided. It should be noted that this study examined only seventy-nine infants who were treated in this fashion. Additional studies are necessary before the benefits of postpartum prophylaxis can be fully assessed. However, if these findings hold up, AZT given within forty-eight hours of birth — which can be done only if it is known which infants are at risk — saves lives by preventing transmission.
Early identification of possible HIV infection also allows physicians to tailor immunization schedules, an important consideration for children with HIV infection. For example, HIV-infected infants should receive flu and measles vaccines more frequently than uninfected children, and some infected babies may need one of the different types of vaccines available rather than another. Finally, test disclosure makes possible "close nutritional monitoring, an essential component of care" for HIV-positive infants.
As these facts about the beneficial effects of postnatal treatment became widely known, public support for unblinding tests increased to the point that, in June 1996, Nettie Mayersohn's legislation was passed in New York State, despite strong and prolonged objections by the American Civil Liberties Union (ACLU), the Gay Men's Health Crisis (GMHC), and the New York chapter of the National Organization for Women (NOW), among others.
As is often the case, opponents have raised both principled and practical objections to this legislation. Principled objections focus on the violation of the rights of the mothers caused by involuntary disclosure of the test results. The New York ACLU argues that, "in effect, mandatory legislation is mandating every new mother in New York State to be tested for HIV. The testing of newborns is an underhanded way of testing mothers and circumventing their rights to consent to the test." Jeffrey Reynolds, deputy director for the Long Island Association for AIDS Care, states that newborn testing "amounts to mandatory testing by proxy, and without informed consent or counseling requirements, for all pregnant women in New York a sharp departure for a state that mandates coercive testing for no other population." He calls Mayersohn's bill "misguided," adding that "above all, HIV testing after birth is simply too late to reduce the chances of perinatal transmission." The Gay Men's Health Crisis adds:
Concern over the health of the city's infants does not require that the state ignore their mothers' right to privacy and impose the mandatory newborn HIV testing program you [Newsday editorial] advocate. ... Testing newborns for HIV antibodies ... is tantamount to testing their mothers.... New York State law appropriately prohibits testing any individual for HIV antibodies against his or her will; the fact that you ignore pregnant women's absolute fight to the same privacy as all other adults is disturbingly reminiscent of the current trend among conservative policymakers to consider women and their health only in relation to their role — or potential role — in childbearing.
Turning to more emotive and practical objections, when asked, "Is freedom that important that you might allow 15,000 babies' lives to be poured down the drain?" Ruth Macklin, a bioethicist at the Albert Einstein College of Medicine in New York, replies, "At a certain point, one balances freedom against lives, indeed. We fight wars to preserve our freedom, knowing that a certain number of people are going to die."
Terry McGovern, director of the HIV Law Project, has argued that "information is often not kept confidential.... I can't tell you how many heartbreaking cases we see where women are thrown out by their families, where children are thrown out of schools." The executive director of the National Women's Health Network claims that "mandatory testing of newborns and their mothers will not save lives and will do nothing to stem the spread of HIV." In the face of these strong objections, other states, as of 1998, have declined to follow New York's lead.
On the federal level, Representatives Tom Coburn (R-Okla.) and Gary Ackerman (D-N.Y.) introduced draft legislation in 1995 that would have required all states to follow the New York State example or lose the sizable federal funds they receive under the Ryan White Care Act of 1990. However, the bill raised strong objections from many of the same groups that had objected to it in New York. In response to this initial opposition, the law that was eventually enacted in 1996 reflected a compromise: All states will be required to meet certain goals for pediatric AIDS prevention within five years. By the year 2000 states must reduce their number of pediatric AIDS cases by 50 percent or, through counseling, test 95 percent of pregnant women for HIV. States that do not meet these goals will have to implement unblinded mandatory newborn testing for all mothers who do not receive a prenatal HIV test or lose their Ryan White Care Act funds. The Institute of Medicine was charged with the task of conducting a study of the states' testing procedures and the results they obtained. Thus, through this bill, Congress delayed any final vote on the unblinding of infant tests for a period of close to five years.
Mayersohn's bill, as well as Coburn and Ackerman's congressional bill, also faced opposition from the CDC, which canceled its newborn testing efforts in 1995, shortly after the New York law was enacted and just as the congressional bill was being drafted.
Applying the four evaluative criteria laid out in the introduction, we have seen so far that there is a serious problem: The lives and well-being of a considerable number of infants are at stake. Would second-criterion treatments (which are voluntary and do not involve changes in the law) suffice, or do we need to initiate third-criterion interventions (which entail mandatory acts)?
The Public Policy Debate
Opponents of unblinded testing argue that from a medical viewpoint there is a more effective alternative that does not raise the legal and ethical issues, especially the violation of privacy, invoked by unblinded testing. This alternative policy, voluntary testing of pregnant women as part of prenatal care or a special program, is said to be capable of achieving the same public health goals as unblinded testing. The ACLU, the Gay Men's Health Crisis, and the CDC, which have all adopted this position, feel that it is supported by the results of a 1994 study: If AZT treatment is administered to HIV-positive pregnant women during pregnancy and delivery, and to infants for six weeks after birth, the risk of infection for the infants is reduced by two-thirds, from 25 percent to 8 percent. In the light of these findings, the CDC issued guidelines in 1995 calling for the voluntary testing of all pregnant women for HIV. Thereafter, there has been a significant reduction in the number of sick infants.
It is correctly argued that voluntary testing of mothers during prenatal care would also dispose of the ethical and legal issues raised by unblinding test results in which mothers' privacy is violated. If mothers were tested instead of infants, their consent would first have to be obtained. And if the proper funds were allocated, health care staff could provide mothers with counseling before the test and help them cope with a positive result if the test's findings were in fact positive. Furthermore, many health care professionals argue for prenatal as opposed to postnatal testing because prenatal treatment is believed to be more effective.
As I see it, the issue is not properly framed when put in terms of choosing between voluntary testing of pregnant women and mandatory infant testing followed by disclosure of the results; both alternatives are needed, for reasons spelled out shortly.
Long experience has taught me that whenever one party argues for policy A and against policy B, and the other party for both policy A and policy B, the debate soon makes it sound as if the second party favors B but not A. Hence I repeat: I fully grant that testing pregnant women, counseling them, and encouraging them to take proper medications and undergo appropriate care is a highly desirable public health policy. (Note that this policy requires that considerable resources be allocated and that physicians, other health care personnel, hospitals, and clinics include HIV testing and counseling in their prenatal care programs and make efforts to find the mothers who do not present themselves for prenatal care.) However, successful implementation of this policy does not obviate the need to learn from tests of infants' blood — tests that are already conducted to check for other illnesses, especially phenylketonuria (PKU) — whether they have been exposed to HIV, most notably because many pregnant women do not present themselves for prenatal care despite concerted efforts to encourage them to do so. Many of them can be found and persuaded to provide treatments to their infants only if their infants are tested. One does not choose between seat belts and airbags, both of which are mandated, to protect oneself from auto accidents; similarly, the effectiveness of one policy in reducing HIV transmission to infants does not invalidate another policy.
True, some believe that unblinded infant testing would undermine programs aimed at counseling and testing pregnant women. Alan Fleischman, a physician with the New York Academy of Medicine who strongly opposes unblinded newborn testing, has suggested that if newborn testing is universally conducted, doctors who provide prenatal care may stop advising pregnant women to be tested for HIV under the mistaken assumption that newborn testing eliminates the need for such counseling. This possibility is highly speculative and quite unlikely, especially given current knowledge regarding the benefits of prenatal counseling and testing. A physician who so conducts herself would not discharge her elementary duty to her patient, the woman in her care. Although it is possible that some physicians would conduct themselves in this manner, there seems to be no evidence that more than a very few would do so.
There are several other reasons why infant testing is needed, beyond the very important fact that many pregnant women, including those most likely to develop HIV, do not present themselves for prenatal care. Some pregnant women who do seek prenatal care do not consent to be tested for HIV. And pregnant women who have tested negative may develop HIV after they have been tested but before delivery. It should be noted here that research suggests that the greatest risk of infection occurs during birth. To put it bluntly: Voluntary testing of pregnant women is highly commendable, but numerous infants will still fall through the cracks, so to speak, and need to be "caught" so that they can be helped.
Finally, given that symptoms may take many months or even years to develop, it is quite conceivable that an HIV-positive mother, unaware of her infection, could give birth to yet another child without the benefits of appropriate ARV prophylaxis. This child would also run a significant chance of contracting HIV.
These concerns are supported by the fact that voluntary counseling and testing programs have failed to identify numerous HIV carriers. In New York a study that gathered data from July 1993 through September 1993 determined that only 53 percent of infected infants were identified through prenatal testing and counseling. Although subsequent figures have been more favorable, champions of unblinded testing point out that in New York the improved results were achieved largely after Mayersohn's bill unblinding test results was enacted. (That is, after pregnant women learned that, regardless of whether they decided to undergo prenatal testing, their infant would be tested and the results provided to them.)
Those who favor testing only mothers and not infants also point to the success of voluntary prenatal testing and counseling at one hospital, the Harlem Hospital. This program convinced 90 percent of pregnant women to be tested for HIV. However, as Elaine Abrams, director of pediatric AIDS care at Harlem Hospital notes, "Isn't it troublesome that only this one program has been successful? Counseling has worked in one place and failed everywhere else, and they want to duplicate it?" And Mark Rapoport, a former Westchester County public health commissioner, notes that although voluntary testing works in "the best circumstances, as at Harlem Hospital ... overall, less than half of women know their infant's status and, therefore, their own status when the baby's born."
It is theoretically possible that if more and more resources are invested in locating, counseling, and testing pregnant women, a point may be reached where only a very few would transmit the disease to their children. The best way to establish that this day has arrived is to continue to collect data from tests conducted on the infants. Efforts to reach such a goal are not advanced by canceling infant tests.
Far from being wasteful, the initial results of the New York Baby AIDS Law have been encouraging. Data collected by the New York Department of Health's AIDS Institute and Wadsworth Laboratories indicate that nearly 100 percent of HIV-exposed infants are being identified, and that follow-up efforts have linked 98.8 percent of these infants with care. In the first twelve months of the program, 102 exposed infants were identified whose mother's HIV status had not been known at the time of delivery.
Finally, opponents argue that by the time the results of infant tests are available, mother and child have typically long since left the hospital and some are difficult to locate. However, since 1992 a test has been available that can provide results within ten minutes. This test is rarely used owing to concerns over interpretation of the results, and previous CDC recommendations have suggested that confirmatory tests be performed for all HIV tests before patients are informed of the results. However, the CDC recently changed this policy, urging doctors to make greater use of the rapid test and to inform patients immediately of the results while confirmatory tests are performed.
Relative Costs of Newborn Testing
Some objections have been raised to infant testing and disclosure on the grounds that such measures would be too costly. At the Institute of Medicine's Second Perinatal Transmission of HIV Committee Meeting in 1998, Ellen Mangione of the Colorado Department of Public Health and Environment argued that her state's total HIV prevention budget was limited, and that dedicating dollars to unblinded infant tests was thus unacceptable. Another opponent asserted at the meeting that in the first nine months after enactment of the New York Baby AIDS Law, 185,000 infants were tested and "only" 60 infants who tested positive were born to mothers who had not themselves been tested during pregnancy. (The actual figure is 69.)
Actually, the costs of the test are very small because the blood of babies delivered in the health care system in most states is already collected and tested to determine whether the infant has PKU, hypothyroidism, or galactosemia. Testing for HIV is done on "residual" blood specimens, that is, "specimens collected and tested for other reasons that were temporarily stored prior to being discarded."
Dr. Guthrie Birkhead of New York's AIDS Institute has reported that blood-testing is highly automated and that adding an initial HIV test to the existing battery of tests raises the cost only minimally. Birkhead estimates that the change would require two additional technicians, along with a few other nominal expenses. (It should be noted that confirmatory tests and counseling occur only if the first test is positive, and so these costs are incurred by a much smaller number of people.) In short, the costs of testing infants in addition to pregnant women are surprisingly modest.
Opponents who base their objections on cost should also be reminded that the HIV budget is not a zero-sum game; if additional measures were proposed that would directly save or improve lives, added funds might well be made available. Moreover, the costs could be reduced if testing were concentrated in those areas where more than a given number of HIV infections per capita prevails, a proposal the CDC itself considered at one point for another HIV testing measure. This approach, however, could lead to complaints that the testing is focused on certain minorities or socioeconomic groups. One can look at universal testing as a means to avoid even the appearance of discrimination, surely a goal worthy of some added expenditure.
One last point: Because counseling and testing pregnant women is more costly than testing infants, someone guided by cost alone would be compelled to eliminate or restrict these efforts first — certainly an unacceptable solution.
I am not suggesting that costs are irrelevant to this issue. However, given how nominal any added costs would be, such considerations do not seem to justify opposing unblinded testing and forgoing its very considerable benefits.
Other Arguments Against Unblinded Testing
Opponents of unblinded, mandatory testing of infants argue further that such a program would scare women away from the health care system, forcing them to avoid prenatal care and hospital deliveries. For instance, the ACLU argues that
mandatory testing of pregnant women and newborns would have detrimental public health consequences, most significantly by deterring women, especially low income women, from seeking prenatal care. ... Without trust there is rarely compliance, especially when a woman is confronting not only the possibility that her child has an incurable disease, but the certainty that she does as well.
The Gay Men's Health Crisis concurs: "Mandatory or coercive proposals, which offer no guarantee of services and which ignore the need to involve the mother in the care of her child, neither reduce the possibility of HIV transmission nor increase the likelihood that infants or their mothers will receive the care they need."
These claims have been countered by the Association to Benefit Children (ABC), a foster care provider in New York City. After years of work in this field, this agency observes that "an HIV-positive result typically meant that a mother took strengthened interest in managing her health and that of her child." Observations by both sides on this point are anecdotal, however; until systematic data are generated, this argument cannot be employed to oppose unblinded testing any more than it can be used to support it.
Some opponents claim that there is little or no support in the medical community for unblinded newborn testing. Elizabeth Cooper, an associate professor at the Fordham School of Law, writes that "it is alarming when any arm of the government ... chooses to pursue a path of policy development that not only has not been endorsed by, but in fact has been rejected outright by virtually every well-respected medical, scientific, and public health organization." Although it is true that several health groups, already cited, have opposed unblinding the newborn test, support has arisen from several others, including the New York State Association of County Health Officials, the Council of Family and Child Caring Agencies, the New York chapter of the American College of Emergency Physicians, and the AIDS Institute Committee for the Care of Children and Adolescents with HIV Infection. Additionally, former Surgeon General C. Everett Koop has supported the measure. Although the American Academy of Pediatrics opposes unblinding, the American Medical Association has endorsed it. Clearly unblinding is not a measure without significant professional support.
The CDC's Concerns
The CDC suspended its testing of infants in 1995 as unblinded testing began to gain support. The CDC states that it suspended the testing to "re-evaluate how best to combine prevention and surveillance strategies" in light of new findings on the effectiveness of prenatal intervention. However, the CDC has made no secret of its opposition to unblinded tests. A CDC spokeswoman, Terry Hammond, has commented that "testing newborns is too little, too late. The CDC feels very strongly that voluntary testing is the most effective way to reach women and children who need care." Dr. Helene Gayle of the CDC testified before a congressional hearing that the CDC suspended the program because it intended to call for voluntary testing of all pregnant women. Dr. James Curran, head of the CDC's Division of HIV/AIDS Prevention Surveillance and Epidemiology, commented that through voluntary testing, "every pediatrician will know every baby who's exposed [to HIV]," obviating the need for newborn testing.
Others contend that the CDC feared that unblinding the results could jeopardize the quality of its demographic and statistical work. This point was stressed repeatedly in a long letter to the author from Dr. Robert Janssen, deputy director of the CDC's AIDS division. He states that the CDC is mandated "to examine, evaluate and monitor the extent of the HIV epidemic in this country." The agency's mission is not to find cures or prevent the spread of the dreaded disease. Indeed, the role of the CDC is often to collect information that will assist policymakers and researchers, an effort that does not necessarily entail actual medical research directly seeking to find a cure, let alone administer it to the ill.
Discussing the ethical concerns surrounding the surveillance studies, Dr. Janssen writes:
In response to questions regarding the ethics of these studies, we suspect that this question arises now because of the extensive publicity of the importance of newly available therapies for HIV. Important therapy for HIV ... ha[s] been available for a decade or more.... The recent identification (since 1995) of the class of protease inhibitor drugs has been important for certain individuals with HIV infection; however, these drugs do not cure the infection and its ultimate disease course. Problems with the existing limits on the access to, availability of, and adherence to these drugs also limit their ultimate effectiveness in the population of HIV-infected persons. These comments are not to belittle the advances in therapy, but to simply balance these advances with the ongoing needs to monitor the epidemic, particularly with tools that can evaluate populations where new HIV infections are emerging.
Dr. Janssen crowns his observations with the following comment:
In fact the question is really about whether public health officials need to know the extent and the movement of the HIV epidemic in the United States. Based on the assumption that public health programs and public health leaders need accurate and representative information on important diseases, the survey methodology (unlinked anonymous) is critically necessary.
In effect Dr. Janssen is saying that keeping the data straight is more important than ameliorating the effects of the disease, as far as the work of his agency is concerned. (Note that the data do not evaluate various treatments but merely provide a picture of the level of the epidemic). Sadly he does not consider that, without continued and unblinded tests, infants whose lives could be saved if they were given AZT and/or if they were not breast-fed will die.
Even from a purely social science perspective it is difficult to find fully compelling the CDC's arguments against unblinding. To begin with, testing an entire population rather than smaller samples is usually not a favored procedure. Most important, it does not automatically follow that such a loss to the cause of epidemiology would outweigh either the medical benefits to newborns or the additional normative concerns I explore next.
It should also be noted that in many medical research studies the test results have often been unblinded once it became clear that doing so would enable health authorities to provide better care to those involved — and certainly when lives could be saved. As I noted earlier, infants are already tested for a variety of illnesses, and the results of all but the HIV test are readily disclosed to the mother. These include routine tests for hepatitis and syphilis, the results of which are released to the mother owing to the overriding health benefits of disclosure.
Once one recognizes that such testing has become routine, the course of action seems self-evident. In fact, when health officials implemented mandatory testing of infants for sickle-cell disease, many of the same objections facing unblinded HIV testing were raised. However, it became clear that the medical benefits of the tests outweighed other concerns.
The epidemiological studies in question are not tests of medication that would allow HIV to be cured, in which case one could argue that it might be proper to endanger some lives now to save many more later. At issue here are studies that have much more indirect and limited health benefits than those associated with a vaccine or cure. And, to reiterate, these purposes can be served even if the test is unblinded.
Consent to Disclosure?
The public policy debate has focused on whether test results should be unblinded and mothers informed, but another option comes to mind, one that has in effect already been tried: seeking the mother's consent to be informed about the test's results (or, if the infant's blood is not tested for HIV, to allow it to be tested and the results disclosed). At first blush, this approach would seem to combine the best of both worlds: Practically all the mothers could be reached because even though only some pregnant women present themselves for prenatal care, almost all deliveries take place in hospitals or other health care facilities; in addition, consensual disclosure is assumed not to violate privacy.
One disadvantage of this approach is that the costs would be substantially higher than those entailed in unblinding the test, because all mothers would have to be counseled (before they could be asked to consent), not just the much smaller number whose infants' test results are positive (if no a priori consent is required).
Even more important, the consent given under these conditions is far from compelling and would likely be considered unsatisfactory by a typical bioethical committee. First of all, it has been observed that many of the treating professionals are of a much higher socioeconomic status than the mothers and thus are perceived as strong authority figures. Marcia Angell, the executive editor of the New England Journal of Medicine, notes that "many people can be coerced into submitring to harmful experiments, especially if they are poor and uneducated." Although this statement was made in reference to overseas AZT experiments involving pregnant women, the same principle applies in this country. Even many white middle-class mothers find it difficult to object to suggestions by their physicians.
Most important, the consent must generally be sought as the pregnant woman is in labor, a point at which it would be unethical to ask her to consider agreeing to a test that might establish that she has a horrible, fatal illness, one she might already have transmitted to her about-to-be-born child. Counseling under these conditions would, at the very least, be ethically suspect and difficult to carry out.
Although requests for consent to disclose might be delayed until after delivery, they must still be made very shortly thereafter to prevent breast-feeding and to start whatever medications are necessary; moreover, mothers usually remain in the hospital for only a short time. Hospitals would also have to ask all mothers for consent, because a practice of asking only those whose infants test positive would soon become known and would be tantamount to disclosure without consent.
Last but not least, some mothers will refuse to consent. The fact that their numbers might be small (which is far from well established) does not answer the ethical questions involved. Should a mother who does not consent to be tested be informed that she will gravely endanger the life and health of her child if she breast-feeds and does not provide that child with AZT?
When all is said and done, consensual disclosure (if issues of costs are not allowed to prevail) has some advantages over nonconsensual disclosure, but these are offset by the difficulty of obtaining consent under the typical conditions. In either case, the ethical issues that arise when mothers prefer not to be informed cannot be avoided.
Many oppose testing infants for HIV and unblinding the results on the grounds that doing so violates the mother's right to privacy. The right to privacy includes "the right to bodily integrity, including the right to make personal decisions regarding medical tests and treatments," and the right to make medical decisions on behalf of one's children. These rights are protected by civil tort law, by state statutes, and by the Constitution. In addition, there are specific laws requiring that individuals be given the opportunity to grant written, informed consent before submitting to an HIV test and providing for the strict confidentiality of HIV test results. A law that requires mandatory, unblinded testing of newborns would override a woman's right to informed consent and confidentiality.
Opponents argue that violations of rights besides privacy, as well as other injuries, are likely to follow. The harms they claim include discrimination, stigma, loss of jobs, health and life insurance, and housing, and even domestic abuse or loss of family, all supposedly stemming from the violation of privacy. The ACLU argues that "these woman are susceptible to the same kinds of discrimination faced by others if it becomes known that they are infected with HIV." AIDS Project Los Angeles and the San Francisco AIDS Foundation point out that "there have been numerous court cases involving HIV-positive individuals who have lost their health insurance, their job, or both. The reality is that we cannot legislate away the stigmatization that people with HIV experience." According to data provided by the Equal Employment Opportunity Commission (EEOC), 2,327 HIV-related cases have been filed with the EEOC since 1992. Of these, about 60 percent have centered on wrongful firings.
Karen Rothenberg and Stephen Paskey, writing for the American Journal of Public Health, have opined, "We believe that HIV-infected women are particularly vulnerable to the risk of domestic violence." Although no studies have documented this observation on a large scale, one study conducted in Baltimore found that the fear of violence or abandonment is prevalent among HIV-positive women: "Forty-five percent of all providers surveyed had at least one female patient who expressed fear of physical violence resulting from disclosure of her diagnosis to a partner, while 56% of providers had patients who expressed fear of emotional abuse and 66% had patients who expressed fear of abandonment." If infant HIV testing remains blinded, the reasoning goes, such violent outcomes are far less likely because nobody can identify the subjects who have tested positive. They are merely nameless numbers in statistical tables.
Opponents also argue that unblinded testing raises questions as to whether such a procedure might violate a mother's right to freedom from unreasonable search and seizure. Taking a blood sample has been shown many times in court to be a search within the context of the Fourth Amendment. However, the Fourth Amendment permits searches under certain circumstances, such as when the search is intended to protect the health of an individual. Because the search is intended for precisely such a purpose in this case, some legal scholars argue that a mother's civil rights would not be violated.
As in many ethical deliberations, the conflict here is not between a wrong and a right but between two or more rights. Several questions are at issue: Which rights should take precedence? Should the intervention be modified to minimize the damage to the right, or do the rights need to yield? What are the implications of the new balance struck among the different rights for the common good (a particular concern for communitarians)?
Loss of Life
In ranking the rights and values involved in the question of whether HIV testing of newborns should be mandatory, one can rely on abstract ethical considerations, such as weighing autonomy against beneficence. I draw mainly, however, on the core values of the democratic society in which these issues must be worked out. As I see it, these values provide clear guidance in the case at hand. Our core values and the legal code that expresses them generally rank the loss of life over loss of limb, and both higher than the loss of property. Other concerns are less clearly prioritized but usually do not take precedence over threatened loss of life, including the knowing infliction of a major illness or the failure to treat it when it can be cured.
The Tuskegee Experiment is a case in point: Not informing the syphilis patients who were participating about their condition, and not providing them with the available treatment, is considered one of the great ethical failures of American public health policy. The parallelisms between HIV testing and withholding information and the Tuskegee Experiment have been very well drawn by Marcia Angell.
Laws reflect this type of value "ranking" in court cases where the well-being of children is involved. Although parents have the right to give informed consent for treatment on behalf of their children, the "parents' right to determine the course of treatment for children is limited." Renna explains that "a parent may not deprive a child of lifesaving treatment," and adds that "courts have consistently allowed the state to intervene when a child's health is in danger."
From this standpoint, one set of facts stands out: Given that keeping infant tests blind, or not conducting them at all, will directly contribute to the deaths of a significant proportion of infants born to mothers who have HIV, this is in line with our values and legal tradition.
A Devastating, Prolonged Illness
Untreated infants are condemned not only to early death but also to severe illness. AIDS, of course, differs from most other illnesses in that it entails a very debilitating and extended period of suffering for both infants and their families. As noted earlier, a newborn with AIDS is subject to many devastating, opportunistic infections that will kill him or her painfully over the course of many months or years. Infants with HIV also suffer from "recurrent severe bacterial infections, cancers, specific encephalopathy, and wasting syndrome." Even if keeping the information from the mothers did not cause the deaths of a significant number of infants, the severity of the illness to which they would be exposed is itself a major concern.
Opponents are correct in pointing out the psychological and sociological effects of having one's privacy violated. We must also consider, however, the suffering inflicted on the infants' mothers if they are not informed about positive test results. Here is one far from atypical account.
A baby girl was born to a mother at Mount Sinai Hospital in New York City on the morning of 31 January 1991. The doctor assured the mother that her baby was normal and healthy. Nine months later the mother rushed her child to the same hospital. Her child was having seizures and had stopped breathing. To the mother's astonishment, she learned that her child was HIV-positive and suffering from meningitis, a pneumococcal infection brought on by an HIV-weakened immune system; the infection eventually left her child blind, deaf, brain-damaged, and paralyzed. The mother was incredulous when she discovered two agonizing facts: first, that this debilitating infection could have been prevented if her child had been diagnosed and treated soon after birth; and second, that her child had actually been tested for HIV as a newborn in the hospital but the results were withheld from her "to protect her own privacy."
Similar scenarios have occurred many times across the country for more than ten years. Mothers who take their infants home without being informed of the latter's HIV status may have children who will gradually develop a variety of undiagnosed illnesses that may seem like a severe flu, pneumonia, diarrhea, and so on. After continued treatment for these various illnesses, however, the mothers soon discover the true nature of their child's problem, and its prospects. The guilt these mothers may feel about having inadvertently contributed to the illness and death of their child is a very serious concern, as is their anger at the medical community for denying them information that could have prevented their child's death and suffering. Indeed, one may ask whether these mothers have a legal, or at least an ethical, claim against those who failed to inform them of their child's severe but treatable condition.
Other benefits for mothers may result from unblinded tests, aside from being spared considerable agony. As explained earlier, a positive newborn test result indicates the mother's HIV status. If the tests are unblinded, the mother may obtain medical treatment for herself. Given the favorable results obtained through "drug cocktails" that include protease inhibitors, the medical benefit of knowing one's HIV status is undisputed, provided that care is available.
Not only does the mother's treatment indirectly benefit the infant, who depends on her for care, but "identification of a mother's seropositive status helps her make informed decisions about her future family plans." She can make informed choices regarding future pregnancies and appropriate care if she does plan to have additional children. She can also make early arrangements for the care of her children who may survive her. Finally, the knowledge of her HIV status gives the mother an opportunity to take precautions against further transmission of HIV to sexual partners, certainly no small consideration.
Violation of Privacy
There is no denying that unblinded tests entail a diminution of the mother's privacy. It should be noted, though, that privacy is not an absolute value and does not trump all other rights or concerns for the common good. As Alan Westin has observed, "An individual's desire for privacy can never be absolute, `since participation in society is an equally powerful desire.'" Even civil libertarians recognize that the police have a right to search a person's home if there is specific evidence that, say, a criminal or a murder weapon is hidden therein. It is therefore insufficient simply to argue that unblinding violates privacy and then rest one's case. One must weigh the harm inflicted on the mother, infant, and community when there is no disclosure compared to the harm inflicted by violating privacy.
Moreover, even if newborn test results remain blind, privacy will not be preserved for long. As Renna observes:
The reality is that she [an HIV-positive mother] is likely to learn of her newborn's HIV status in short order.... Without treatment, the probability that an HIV-infected infant will develop an opportunistic infection within the first year is great.... She not only will be forced to deal with the sudden revelation of her and her infant's status, but will be forced to deal with her infant's serious HIV related illness (and possibly her own), as well as the realization that this illness might have been preventable.
In short, the HIV status of both the baby and the mother will eventually be revealed whether the newborn is tested or not; the issue of discrimination and other ill effects of disclosure will have to be faced when either the infant or the mother develops more visible symptoms over time.
All this means not only that the claim of privacy is trumped by concern for saving lives, avoiding severe illness and suffering, and directly improving the well-being of others, but also that the privacy claim is itself rather limited.
Violation of Autonomy
The argument has been made that not only the privacy but also the autonomy of mothers is violated when they are informed about the results of a test that they did not seek and provided with information they do not wish to have. This argument calls attention to the complex and often confused relationship between privacy and autonomy.
Violations of autonomy may occur even when mothers are merely asked to consider whether they wish to know, because the mothers may be perturbed by having to face such considerations.
One could enter here into an extensive discussion of what autonomy encompasses, the difference between a right to know and a right not to know, and related issues. The main question that must be faced at this point, however, is not whether and to what extent autonomy is infringed by unblinded HIV testing in newborns, but whether there is another consideration that justifies a diminution of autonomy. The moral equation is similar to the one invoked by privacy considerations: Given that fully honoring autonomy significantly increases the chance of loss of life, and that a serious illness will fester untreated, and given that the intrusion is limited to sharing information, it seems that the other considerations should take precedence. Indeed, a case can be made that health care personnel have a moral obligation to share such information with mothers and those who treat their children.
An analogy might illuminate the point. Assume that a patient asks her physician for a battery of tests but informs the physician that she would like to know all the results but one: If she is found to have cancer, she would rather not be told. Assume also that a cancer is indeed caught early by these tests, and it is of the highly treatable kind. The physician in this case is able directly to save a life. (Very often the connection between the physician's action and the saving of a life is indirect and comparatively weak — for example, when a physician counsels the patient to exercise and maintain a proper diet.) Assume also that the patient has a very high probability of dying relatively soon if not treated. As I have suggested already, to not offer a warning in such a situation is like not pulling someone away on spotting the approach of a speeding truck. Under these conditions, the physician would seem obligated to do all she can to gain the patient's consent to be informed. (This consent must, of course, be obtained before the test is undertaken because seeking it after the fact might be seen as a communication that the test results were positive.) If the patient still refuses to be informed, the physician seems obligated to at least instruct the patient in the care she needs.
The same argument is even more valid when the patient is a mother who, even if one could completely respect her autonomy, is likely to cause irreparable harm to a defenseless infant if she is not informed. Even strong libertarians concede that our various rights do not include a right to cause serious damage to others, let alone endanger their lives. Indeed, our laws and ethics fully support taking the much more drastic step of removing children from abusive or neglectful parents. Informing mothers about actions that must be undertaken — such as not breast-feeding their infants and giving them antiviral medications — without spelling out the reasons may be a poor but justified compromise, given that lives might well be saved. Such limited disclosure would allow mothers to stay in denial, if they cannot face the truth, and still do right by their children.
It is certainly of grave concern that HIV-positive mothers may be subject to discrimination. Many of these mothers are members of social groups that are already subject to discrimination even when not infected with HIV. Of all children reported with AIDS between 1982 and 1996, 23 percent were Hispanic and 58 percent were black. An estimated two-thirds of mothers whose newborns tested positive were poor and members of minorities. Further, the greatest percentage of the children's mothers contracted HIV through intravenous drug use (41 percent). By the same token, it should be noted that laws against discrimination based on a person's HIV status have already been strengthened in many states. Additional penalties for unauthorized disclosure of HIV status might well be called for and are currently being contemplated for personal medical information in general.
One must not forget, however, that unblinding the HIV tests entails informing only the mother and possibly the health care personnel who directly treat her and her child. It does not require that the information be reported to public health authorities, included in public records (as when a crime is alleged to have taken place, long before conviction), or revealed to employers and neighbors (as in Megan's Laws). If the mother and the health care professionals involved keep the information confidential, as they are required to do by law, professional code, and elementary ethics, there are no obvious ways in which she or the child will suffer discrimination.
As a further objection to newborn testing, Elizabeth Cooper suggests that mothers with HIV fear that their children will be taken from them. To the extent that this fear is warranted, it should be alleviated as much as possible. It does not, however, seem to constitute a compelling reason not to treat the mothers or their children. Children are indeed sometimes removed from their homes, a fact that raises a whole host of new questions that run in two directions. First, are children removed solely because their mothers are HIV-infected or also because their mothers have been found to neglect their children (as a result of severe drug addiction, for instance)? Second, what conditions justify removing children from their mothers (or fathers)? These questions cannot be answered within the confines of this study but must be addressed if evidence arises that infants are being removed primarily because of their mother's HIV status.
Above all, one must avoid the simplistic argument that unblinded testing will lead automatically to additional discrimination. Fourth-criterion considerations are relevant here: When testing is unblinded, additional safeguards can always be introduced so that the information will not travel beyond the mother and those directly involved in treating her and her child. Such protective measures might include unique identifier numbers and audit trails (see Chapter 5).
A Communitarian Note
The communitarian perspective is critical to this discussion in several ways. First, it underscores the important reality that we are not merely rights-bearing individuals but also community members who are responsible for each other. None of our communal bonds and responsibilities is more important than the mother-child bond and the mother's responsibility to attend to her children's well-being. One may argue about what a child's rights encompass and whether these take precedence over the mother's rights; I have already suggested that the child's right to life and health takes precedence over the mother's right to privacy. But even if one were to reject this conclusion, it would still hold that being a parent entails a series of responsibilities, even if these exact various "costs" from the parents.
The communitarian perspective also enters this analysis by reminding us that a given individual right cannot be used to trump all other considerations, including the common good. Obviously, limiting privacy in the case at hand benefits not merely the child but also the community. Allowing children to develop severe illness or to die, when these outcomes can be avoided, is demeaning to the community. That we allow this to happen on other fronts does not mitigate the moral defect here. Unblinding also benefits the community by curbing public costs: Testing and counseling are much less costly than the treatment of infants infected with HIV.
Finally, the communitarian perspective points to the need to respect shared formulations of the good, those societal values that tend to give precedence to life over privacy, especially when the correlation between the privacy-violating act and the saving of a life is direct and strong.
|1||HIV Testing of Infants: Should Public Health Override Privacy?||17|
|2||Sex Offenders' Privacy Versus Children's Safety: Meagan's Laws and the Alternatives||43|
|3||Deciphering Encrypted Messages: A Prolonged Deadlock and an Unholy War||75|
|4||Big Brother or Big Benefits? ID Cards and Biometric Identifiers||103|
|5||Medical Records: Big Brother Versus Big Bucks||139|
|6||A Contemporary Conception of Privacy||183|