VI. Riding the Currents

As stated earlier, a central purpose of this Article is to consider whether patients who have been the victims of biased medical decisions have an avenue for obtaining legal recourse. The analyses in the preceding two Parts suggest that those patients' prospects of obtaining a legal remedy through either a civil rights action or an action alleging breach of some professional duty are fairly bleak. Only those patients who fall within a class protected by a civil rights statute that arguably applies to medical decisions can even state a claim, and then they have to prove intentional discrimination on the part of the defendant physician. While victims of overtly biased decisions may thus be able to recover, most patients will have trouble proving intentional discrimination unless courts are willing to entertain the premise that discrimination fueled by subconscious bias can be deemed intentional. By contrast, an action based on the defendant physician's violation of some professional duty is theoretically open to all patients. As a practical matter, while medical malpractice and informed consent actions may allow patients to recover when bias leads to substandard treatment or disclosure, the thrust of those causes of action would be the physician's substandard care; physician bias is essentially irrelevant to liability. Although a claim alleging that biased medical decision making violated a doctor's fiduciary duty to her patients better tracks the wrong claimed, the law regarding physician liability for breach of fiduciary duty is fairly embryonic, as is the law regarding the availability of dignitary damages against physicians in the absence of physical injury. Thus, recovery on these grounds in any but the most blatant and egregious cases would require significant development in these areas.

The conclusion that existing law does not offer a clear avenue of redress for all patients who are the subjects of biased medical decisions, however, does not negate the importance of the potential for liability in these areas. Even if the practical challenges of proving discrimination and recovering dignitary damage prevent this Article's outlines for imposing liability for biased medical decisions from translating into actual recoveries for patients in a large number of cases, recognition of the mere potential for recovery has value. This Article's message is that whatever the number of biased medical decisions, the operation of bias--whether conscious or unconscious--is wrong and may be grounds for imposing liability. The near invisibility of the practice of biased medical decisions does not justify a tacit legal acceptance of those decisions. Whether biased medical decisions are common or rare, they present a problem that the law may play *297 some role in addressing--even if only by calling medicine's attention to the immediacy of the problem for individual doctors.

In the short run, the most effective role of law in addressing bias in clinical decision making may lie precisely in flagging doctors' attention to the need to scrutinize their own decision-making processes more carefully to screen out any biases. Many of the editorial comments in the medical literature on treatment disparities call for self-reflection by physicians as they make medical decisions. In 1990, the AMA's Council on Ethical and Judicial Affairs called on physicians to “examine their own practices to ensure that inappropriate considerations do not affect their clinical judgment.” Perhaps a few legal judgments against doctors for biased decisions, even if in the more egregious cases, might add significant force to these aspirational statements. Although some research raises questions about the ability of individuals to identify and counteract the subconscious biases that infect their decisions, other research indicates that persons who seek to act fairly may be able to counteract the influence of unconscious bias on their decisions. Thus, physicians' awareness of the mere potential for legal liability might serve to ignite their willingness to spend the time and emotional energy required to examine their own decision-making processes with an eye to detecting possible biases. In the long run, if this (admittedly optimistic) forecast comes to pass, then the development of a more thorough-going and inclusive system of liability for biased medical decisions may seem less necessary.

This Article has focused on surveying the evidence of biased medical decisions and identifying possible legal responses to those decisions, as well as the shortcomings of those responses under current law. Notwithstanding the conclusion that existing law will only rarely offer any effective remedy to a patient who is the victim of a biased medical judgment, some signs of hope-- both for improving patients' remedial prospects and for reducing the role that physician bias plays in medical decisions--can be found in currents already flowing in the areas of civil rights enforcement and medical quality improvement. A brief consideration of how these currents might feed into efforts both to hold doctors legally accountable for their biased decisions and to decrease the incidence of biased medical decisions follows.

First, policy makers and commentators have already issued calls for more aggressive enforcement of civil rights laws in response to evidence of *298 racial and ethnic disparities in medical care. One action sought is for the federal government to articulate standards for what constitutes discrimination in the context of health care treatment. In addition, a common refrain that is gathering volume is the need for federal agencies to begin collecting data regarding the provision of medical services to members of racial and ethnic minorities in order to permit assessment of compliance with those standards. Numerous commentators have emphasized the crucial role of collecting and reporting data that track patient race in enhancing Title VI enforcement efforts. Likewise, a recently issued report commissioned by the Commonwealth Fund concludes that DHHS must assume a leadership role in mandating the collection and reporting of racial and ethnic data by all programs supported by departmental funds. This sound and fury over the importance of data collection might well signify nothing, however, were it not to capture the attention of policymakers who have authority to compel the data collection. It did so in 2000, when Congress enacted the Minority Health and Health Disparities Research and Education Act. Among this Act's provisions is a charge to the National Academy of Sciences to study DHHS's existing data collection and reporting systems and practices and to report to Congress on what data is needed to support an effective response to racial and ethnic disparities. Although this report was not submitted by the Act's November 2001 deadline, it is expected to recommend a data collection system for DHHS.

A commonly articulated purpose for proposed data collection efforts is to enable DHHS to carry out its administrative responsibility of enforcing Title VI. Certainly, data regarding racial and ethnic disparities in the provision of services by particular providers would enable the agency's Office of Civil Rights to target those providers for enforcement action. Public *299 availability of this data, however, would also enable both quasi-governmental entities such as the Joint Commission on Accreditation of Healthcare Organizations (JCAHO), and private bodies to use the information in their certification and purchasing decisions. Within the past year, groups both of health plans and of major self-insuring employers have announced plans to establish incentive systems that will financially reward providers who meet certain quality standards. If provider-specific data regarding health services provided to racial and ethnic minorities becomes available, groups of private insurers or health care purchasers will similarly be able to establish financial incentives for providers to reduce disp

Moreover, the public availability of such data could ease significantly the burden for some plaintiffs attempting to prove racially biased medical decisions in civil rights actions. Although data showing race-based disparities in a physician's prescription of a specific treatment does not directly prove that the physician's treatment decision was biased by a particular patient's race, it can provide circumstantial evidence of bias from which discrimination against the plaintiff may be inferred. The difficulty and expense in collecting such data currently erects a major obstacle for plaintiffs considering a private Title VI action against a health care provider; that obstacle could be eliminated by the governmental collection and reporting of data.

Finally, the reporting of provider-specific data regarding racial and ethnic disparities could prove a potent tool in drawing physicians' attention to hitherto unrecognized patterns in their own clinical decisions. If most physician bias that operates today is indeed subconscious, then most physicians who make biased decisions do not perceive their own biases. Presenting a cardiologist with data showing, for example, that she has ordered bypass surgery significantly more often for her white patients than for her black patients may motivate her to probe more consciously how she makes decisions between surgical and non-surgical interventions. And if a good faith desire to promote patient welfare doesn't sufficiently motivate physician self-reflection, concern regarding the potential for adverse action by a federal agency, private health plans and individual patients *300 may add the necessary encouragement. Thus, increased federal data collection and reporting may serve not only to enable more vigorous administrative enforcement of civil rights laws, but may also support both patients' private remedial actions and physicians' efforts at self-policing.

A second current that is gaining momentum within the medical community may also feed into efforts to reduce the incidence of biased medical decisions. The past several years have witnessed skyrocketing public attention to issues of health care quality, and a variety of proposed approaches to improving the quality of patient care have issued forth. In addition to increased use of “report cards” that contain provider-specific quality information that can be used by large purchasers of health care services, another approach to improving quality is the push for evidence-based medicine. This movement seeks “to evaluate the safety, effectiveness, and cost of medical practices using tools from science and social science and to base clinical practice on such knowledge.” In short, evidence-based medicine emphasizes medicine's nature as a science, rather than an art.

The movement towards evidence-based medicine figures prominently in a recent report issued by a committee of the Institute of Medicine, Crossing the Quality Chasm: A New Health System for the 21st Century. The authors of the report undertake no less daunting a task than proposing an “agenda for redesigning the 21st-century health care system.” Notably, two of the six aims that the authors articulate for 21st-century health care are that it should be both effective (“providing services based on scientific knowledge to all who could benefit and refraining from providing services to those not likely to benefit”) and equitable (“providing care that does not vary in quality because of personal characteristics such as gender, ethnicity, geographic location, and socioeconomic status”).

The Institute of Medicine report highlights a decrease in practice variations as one desired outcome of a move towards scientifically grounded clinical practice. For if patients receive care based on the best existing medical science, then “[c]are should not vary illogically from clinician to clinician or from place to place.” Rather than leaving treatment *301 decisions to the influence of local practice styles or a physician's personal experience or training, proponents of evidence-based medicine assert that incorporating scientific evidence into widely disseminated clinical practice guidelines will give individual practitioners the ability to practice evidence-based medicine. The central purpose of clinical practice guidelines, which may be issued by professional medical societies, government bodies or health care payers, is to inform physicians about what treatment is appropriate for a specific medical condition or constellation of symptoms.

The connection between using evidence-based medicine to improve the quality of patient care and using evidence-based medicine to decrease the incidence of biased medical decisions is straightforward. As discussed in Part II, disparities in the medical services rendered to different patient groups are particularly likely when the treatment for a particular condition is discretionary, i.e., when it is not clear what treatment is appropriate for the condition. Through the development of guidelines incorporating the best available scientific evidence, however, the number of conditions for which the choice of treatment is discretionary is reduced, and thus the opportunities for bias to influence the choice of treatment are similarly decreased. Concededly, because it is impossible for clinical guidelines to account for and address all the possible variations in the clinical details of specific patients, clinical guidelines will never entirely eliminate physician discretion. Commentators have also raised concerns about the quality of some guidelines and the ability of developers to keep practice guidelines current with advances in scientific knowledge. Nonetheless, a systematic effort to replace physician discretion with science-based guidelines seems likely to shrink many existing disparities.

Moreover, in contrast to efforts to increase data collection regarding racial disparities in order to invigorate civil rights enforcement activities, efforts to promote evidence-based medicine --if they deliver on their supporters' claims--stand to decrease the operation of all types of bias, as well as to improve the quality of patient care generally. Thus, its proponents might characterize evidence-based medicine as a rising tide that raises all (patients') ships. While riding the civil rights current promoting increased *302 data collection may allow targeted attacks on racially and ethnically biased medical decisions, riding the evidence-based medicine current may enable a broader assault on suboptimal care generally.

Fortunately, these two contrasting approaches do not appear to be incompatible. Although it has been argued, in the context of the development of informed consent doctrine, that increasing physicians' potential legal exposure to patients ultimately works to the detriment of patients by diminishing the level of trust in a patient-physician encounter, a recognition of a physician's legal accountability for biased decisions seems unlikely to produce such results. Rather than motivating physicians to replace real conversation with a legal form entitled “informed consent,” the existence of a “discriminatory medical decision” cause of action under civil rights statutes should motivate doctors to identify relevant clinical information about their individual patients and to elicit those patients' actual preferences, rather than employing group-based stereotypes. Of course, as Sidney Watson astutely points out, a civil rights enforcement approach to health care disparities focuses--perhaps counterproductively--on backwards-looking “blaming and sanctioning” rather than on moving forward to design systems that maximize the likelihood that patients will receive unbiased advice.

And perhaps, as the project of considering how best to decrease the incidence of biased decisions continues beyond this Article, we may conclude that holding physicians legally liable for the operation of bias that occurs on a subconscious level is not productive. On the other hand, we may conclude that holding physicians to a high standard of screening their own decisional processes to prevent any “negligent discrimination” is appropriate. What is clear is that current discussions of race-based data collection and the role of evidence-based medicine provide fertile ground for further exploring potential solutions to the problem of biased medical decision making.