Ethics Toolbox: In Praise of Transparency and Opacity in Healthcare
November / December 2006
Ethics Toolbox
In Praise of Transparency and Opacity in Healthcare
On August 22, 2006, President Bush issued an executive order on “transparency of healthcare quality and pricing.” Michael Leavitt, secretary of Health and Human Services, stated the rationale for such an order, saying, “People deserve to know what their healthcare costs, how good it is, and the choices available to them.” (WSJ, 8/23/2006).
At first blush, many of us applaud and embrace this executive order. Implicit in this embrace however, may be a disdain of transparency’s opposite — opacity. Closer inspection reveals that transparency and opacity are two sides of the same coin, and that the interrelationship between the two is substantially more complex than it initially appears. On the one hand, in order to act in a rational way, reliable and credible information is required (transparency), but on the other hand, trade secrets and other forms of intellectual property are important assets of healthcare products and services we enjoy and accordingly must be protected (opacity).
Opacity is something that at times we abhor and at other times we savor. It merits careful attention and is much more complex than a facile value assessment of “good” or “bad.” Designations of this sort make little sense without a consideration of context.
Imagine, for example, playing poker without opacity. What fun is poker if you know the cards your fellow players are holding, and they know the cards in your hand? It makes the game hollow and boring. If a player inadvertently shows you a glimpse of his cards, you would remind him to keep his cards close to ensure a more fair and sporting game.
On the other hand, the rules of the game must be clear and acceptable to both players: how many cards make up the deck, how many cards must be dealt in a certain game and whether three of a kind beats two pair and a straight beats three of a kind and so forth, so we can play the game with continuity and consistency. In essence, while we must be transparent about the rules and generalities of the games, we must keep certain specifics (who holds which cards) opaque.
Indeed, we believe there is a lack of clarity, if not opacity, to the transparency issue! To state it simply, “transparency” is not always good, and “opacity” is not always bad. A preliminary question is whether transparency should mandate unlimited and unrestrained access to all there is to know. Unlimited transparency means no more trade secrets, protected algorithms, or certain forms of intellectual property … and no more poker. Throw in hide-and-seek without “opacity” and we generate cross-generational grieving of the loss of something previously enjoyed.
Thus, we require some transparency with our opacity, and vice versa. The methods employed in deriving a certain outcome or result must be both transparent and easily accessible — for in the absence of transparency we cannot examine or understand how a certain result is achieved or a given claim is justified. Let’s be clear about what should be transparent.
Guiding Ethic
Bottom line: neither total disclosure nor total opacity are acceptable or, for that matter, realistic. Establishing a guiding ethic to offer some logic, process, and justification of how we handle these matters is in order. Such a guiding ethic should not only define how we determine right, wrong, and propriety, but also should serve as a common thread to interweave our actions with the essence of who we are and what is demanded of us.
That being the case, we must also look more carefully at this transparency/opacity interplay and ask, “What does this have to do with healthcare?” A great deal, it turns out. The products and services of healthcare services, not covered by patents (patents are for commercial products, i.e., many prescription drugs, or services that are both legally transparent and proprietary), are often based on other forms of intellectual property and are covered by non-disclosure agreements or licenses. One example is the “predictive modeling” tools that assist with assessing the risk of individuals developing disease-related costs. These are legitimate assets in healthcare companies. To put it another way, the internal operations of these products and services are protected from disclosure by the legitimate principle of opacity, akin to the cards in one’s poker hand. However, the “secret sauce” of these protected products and services must NOT also include the methods and metrics by which these services are evaluated for impact. These rules of impact evaluation, like the rules of poker, must be “patently” transparent. Certainly most would agree that we can enjoy that flavorful chicken with its secret sauce, but the methods by which we assess its value compared with other chicken products with their own secret sauces has to be transparent to the public. This means, for example, that the scores of the risk adjustment models must be transparent, so we can compare one tool against the next.
The transparency of quality metrics in healthcare — per Bush’s executive order — is essential to evaluating the impact of different healthcare products and services and must be followed by a transparent (and replicable) methodology for comparing and evaluating different products and services. The actual value of these quality metrics can be changed by many different causal factors that occur at the same time (e.g., disease management, quality improvements in hospitals, payment for performance programs for doctors and hospitals, the advent of electronic medical records, medical school training, better risk adjustments algorithms, etc.). Moving from method and methodology to fairness, and an ethical assessment of intermediate outcomes or degrees of contributions, is a critical step that is often ignored. Currently, we’ve seen multiple parties each taking 100% credit for the changes that occur, for example, in HEDIS® metrics. That’s not fair! Eyebrows and suspicion would be raised if more than 100% benefit, in toto, were being claimed.
We must develop and agree on standards for fairness as well as for the credit we give to an individual, activity, or intervention for its contribution to a given result. This must serve as a critical component of our guiding ethical framework. The manner in which our personal values harmonize or conflict with such a framework should be subordinated to “reverence” for the way we make and support claims. We need norms to determine “conformity of action.”
A foundation for the ethics of measurement must be developed. It must begin with transparency of quality (and other measures), but is not complete without transparency of methods — even when the business processes by which outcomes are achieved remain opaque. Fairness as an ethical principle must drive metrics and methodology. As the ancient Greek philosophers reminded us, “Knowledge and Ethics are inextricably linked.” While healthcare products and services may be opaque, metrics and methodology that stand behind the relative benefits (including quality) of these products must be transparent.
Dennis Robbins is a healthcare innovator, ethicist, author, and thought leader. He is president of Integrated Decisions, Ethics, Alternatives, and Solutions (IDEAS). He has served as an advisor on ethics and related issues for major national organizations, associations, law firms, hospital systems, private industry, and government. Robbins holds a PhD in philosophy from Boston College and a postdoctoral MPH from Harvard. He has been awarded three fellowships, including a National Fund for Medical Education Fellowship at Harvard in the Dept. of Legal Medicine in the Division of Health Management and Policy. He is a member of the Editorial Advisory Board for Patient Safety and Quality Healthcare. Robbins has written and published nine books on healthcare ethics as well as many chapters, articles, and other contributions to the literature. Robbins may be contacted at dennisrobbins@cox.net or at 602-368-6455.
Thomas Wilson is an epidemiologist specializing in the evaluation of the financial and clinical impact of interventions that are designed to effect defined populations (e.g., disease management, case management, payment-for-performance programs, and electronic medical record use). He received his doctorate in epidemiology from UCLA. Former positions include the corporate epidemiologist at Anthem Blue Cross Blue Shield, faculty research positions at Columbia University and UC Berkeley, a NIH post-doctoral fellowship position in cardiovascular disease, and a Fulbright award lectureship. He is currently the principal of Trajectory Healthcare, LLC, a strategic epidemiologic consultancy firm, and the founder and board chair of the Population Heath Impact Institute (www.PHIinstitute.org), a non-profit organization set up as an educational and research organization advocating for scientific, independent, and transparent evaluations of organized population health management programs. Wilson can be reached at twilson@trajectory-inc.com or 513-289-3743.
References
Executive Order: Promoting Quality and Efficient Health Care in Federal Government Administered or Sponsored Health Care Programs. www.whitehouse.gov/news/releases/2006/08/20060822-2.html