Patients and providers in the courts: fractures in the Americans with Disabilities Act.

• Author: Bobinski, Mary Anne
• Position: Symposium on Health Care Policy: What Lessons Have We Learned from the AIDS Pandemic?

1. INTRODUCTION

   As fragile living beings we walk the world at risk for injury. Human society is organized--in theory and in fact--to reduce the risk of harm. Social safety nets reduce the risk of starvation. Armies and police forces reduce the risk of violent death from war or anarchy. We use law to reduce the risk of harm as well. Criminal and tort rules deter and punish conduct that creates a risk of harm to others. Elaborate regulatory regimes attempt to reduce the risks of human activities ranging from the development of new pharmaceuticals to the safety of the industrial workplace.(1)

   Our systems of risk identification and reduction have become so elaborate that we--those who are the beneficiaries of these protections--sometimes complain. Not all of us want to wear seatbelts or motorcycle helmets.(2) We grumble at a legal system that requires vendors of coffee to provide detailed warnings regarding the obvious risks of consuming hot beverages.(3) We balk at applying coercive social policies against groups, even when those policies might be supported by statistically greater probabilities of harm. Pregnant women present unique risks of harm to their fetuses, for example, yet we hesitate to apply coercive social controls to their conduct.(4)

   Our resistance to some types of legal regulations designed to reduce risk suggests that we value more than safety alone. In some contexts, we are willing to trade a risk of harm for some other valued good. We can explore the tension between our interests in risk reduction and our other values and concerns by examining where and how we make decisions about risk. What risks are significant enough to warrant legal intervention? What values are sufficiently strong to outweigh our interests in reducing risk? How will the risk be reduced? Will creation of the risk be prohibited? Or will we be given information about the risk and permitted to decide whether to accept it?

   This Symposium's analysis of the past and future of our responses to the AIDS epidemic can illuminate our understanding of the legal response to risk. The last fifteen years of the AIDS epidemic have given us ample opportunities to study social and legal responses to health threats in the context of an apparently new and frightening risk of harm. The epidemic has sparked a wide range of social and legal policies designed to minimize the risks of HIV infection.(5) As demonstrated elsewhere in this Symposium issue, courts and legislatures have focused on the use of criminal, tort, and public health law to reduce perceived risks. This Article focuses on the process of risk identification and reduction in the relationship between patients and physicians, nurses, and other health care professionals. It explores whether and how laws prohibiting discrimination against persons with disabilities--which reflect important social values--have constrained efforts to eliminate the risk of HIV transmission between provider and patient. The Article also examines whether the legal response to HIV is consistent with the legal response to other types of risks in the provider-patient relationship.

   In Part II, the Article summarizes current research attempting to assess and to quantify the risk of HIV transmission in the health care treatment relationship.(6) These studies indicate that there is a risk of HIV transmission from patient to provider and from provider to patient, though there is no agreement on the precise probability of harm. In Part III, the Article focuses on the parallel development of two seemingly inconsistent lines of decisions under the Americans with Disabilities Act (ADA).(7) One line of cases, decided under the Act's employment provisions, permits employers to discriminate against HIV-infected health care workers on the theory that they present a "direct threat" to the health and safety of others, especially patients.(8) The second line of cases, decided under the Act's public accommodation provisions, prohibits health care providers from discriminating against HIV-infected patients because those patients do not present a "direct threat" to providers.(9) The apparent inconsistency of these decisions is even more striking when they are compared to the medical evidence about the risks presented by HIV-infected health care workers and patients.(10)

   Part IV analyzes whether the outcomes in these two lines of cases are truly inconsistent.(11) I conclude that the divergence in the treatment relationship decisions cannot be explained by the statutory language (which is identical) or by traditional risk assessment (which would appear to lead to the opposite results in the two lines of cases).(12) Instead, the courts and litigants have reached results which reflect normative assumptions about the health care provider relationship, in which the duty to accept risk is not reciprocal. The decisions reflect the view that health care providers have a duty to confront risks presented by patients and to refrain from presenting additional risks to those patients. This Article will explore how these normative conceptions of risk have surfaced in the courts--a matter of some interest because the relevant statutes do not appear to permit courts to consider the normative justification for differential discrimination.(13)

2. THE RISKS OF HIV TRANSMISSION IN HEALTH CARE SETTINGS

   The precise risk of HIV transmission in health care settings is unknown. The opportunity for transmission is clear. HIV is found in blood and in other body fluids and tissues, although at quite variable concentration levels.(14) HIV can be transmitted from one person to another when infected fluids contact an open sore or mucosal membrane of an uninfected person, although at variable and relatively low levels of efficiency.(15) An infected health care worker who accidentally bleeds into a patient's open wound-perhaps because the health care worker sustains a cut during surgery--can expose the patient to the risk of infection.(16) An uninfected health care worker can acquire HIV after being exposed to the blood or bodily fluids of infected patients, perhaps, by sustaining a cut during surgery.(17) The fact that we can imagine easily how transmission could take place does not mean that we know the probability that transmission will take place in a single exposure incident or over time. The probability of transmission in any incident probably depends on a number of factors, including the concentration of virus in the infected material, the volume of material introduced into the uninfected person, the characteristics of the virus itself, the characteristics of the immune system of the uninfected person, and the use of any post-exposure prophylactic measures by the person exposed.(18)

   There are three major sources of data on the transmission of HIV in health care settings. First, United States Centers for Disease Control and Prevention (CDC) collects data on the "source" for all reported HIV and AIDS diagnoses.(19) The CDC collects information on cases of HIV infection apparently arising from the health care treatment relationship.(20) Second, public health officials have used available data and computer modeling techniques to attempt to quantify the risk of HIV transmission in health care.(21) Finally, researchers have conducted "look back" investigations of the risk of HIV transmission from health care workers to patients.(22)

   1. Transmission from Health Care Provider to Patient

      The risk of HIV transmission from health care provider to patient moved from "theoretical" to "actual" for many researchers when a young woman named Kimberly Bergalis was diagnosed with AIDS.(23) Epidemiologists traced her infection to the practice of an HIV-infected dentist, David Acer, and eventually identified five additional cases of possible transmission from dentist to patient in this one practice.(24)

      The Acer case sparked even more intense efforts to identify other cases of possible transmission from health care worker to patient.(25) It also led to the development of public health guidelines governing the continued practice of HIV-infected health care workers.(26) Congress enacted legislation requiring that states either adopt these CDC guidelines or face the loss of federal funds.(27)

      The CDC guidelines for HIV-infected health care workers are based on the concept of "exposure-prone procedures" (EPPS).(28) Health care workers who perform EPPs have a duty to know their own HIV-status.(29) If they are HIV positive, then they must refrain from performing EPPs, unless they have secured the permission of an expert review panel and have obtained the informed consent of their patients.(30) Unfortunately, the CDC did not define the specific subset of invasive procedures that were exposure-prone.(31) The guidelines referred to particular types of invasive surgical and dental procedures in which transmission of hepatitis had been shown to occur and to other types of invasive procedures in which injuries to surgeons had been demonstrated.(32) Other than these examples, the CDC offered only a broad definition of EPPs as those which involved the use of sharp instruments in circumstances where the health care worker could not easily see his or her work area.(33) The guidelines did suggest that the risk of HIV transmission was sufficiently great to warrant intervention for at least some invasive procedures.

      Attempts to use computer modeling techniques to quantify the risk of transmission have yielded various results. In 1991, the CDC estimated the risk of HIV transmission in surgery to be from .0024% to .00024%.(34) This data then can be manipulated to yield a lifetime exposure risk for transmission from a particular infected health care provider. The resulting cumulative risk will depend on both the type and number of procedures performed over the course of the infected provider's working life. In its 1991 study, the CDC estimated that the lifetime cumulative risk for an...