On Wednesday, President Bush signed the Genetic Information Nondiscrimination Act, affording genetic information special protections. A product of more than a decade of debate, this moment was bittersweet. GINA may harm many of the individuals it is designed to protect.

The act increases protections only for some medical information, privileging those with genetic conditions over those with nongenetic ones. This necessarily places greater pressure on insurers to use nongenetic medical information to segregate risk. Further, it creates unequal disability protections. GINA protects individuals with genetic conditions from health insurance and employment discrimination even if they have no symptoms of the condition. The Americans with Disabilities Act protects only individuals with symptoms in regard to employment, services and public accommodation.

GINA also creates the perception that genetic testing is unique and not basic health care. This may further limit already restricted coverage of such technologies under public and private health insurance. In addition, heightened protections for genetic testing may discourage insurers from covering such services out of fear that they will open themselves to increased risk of litigation for breaching patient privacy. Further, insurers and employers are likely to lose incentive to provide genetic testing when they are not allowed access to the results, as they are for other diagnostic tests.

Well-intentioned members of Congress were wrongly persuaded that genetic information is a unique form of medical information. Insurers must discriminate in differentiating between risks in order to function as insurers. Whether it is unjust to deny anyone entitlement to basic health care based on randomly chosen risk criteria —- genetic or nongenetic —- is a separate question from whether insurers should be able to use only nongenetic information for actuarial purposes.

The fact that genetic information is both shared (other family members might also have the same condition) and predictive of a future illness is not unique. Nongenetic tests may reveal medical information about others as well. Consider sexually transmitted disease testing of one member of a sexual partnership, tuberculosis testing within families and communities, and nongenetic cancer testing of individuals within a group exposed to environmental carcinogens.

Knowledge of exposure in these situations is also predictive.

The view that genetic information is unique —- what legal scholars term "genetic essentialism" —- is outmoded. This is evident when examining the statutory language of some state genetic privacy laws, which invoke strong notions of privacy originally embraced in the realm of reproductive freedom in the classic abortion case Roe v. Wade.

These laws appear to associate genetic testing with its use in the prenatal screening and abortion context and treat genetic information as uniquely and fundamentally private.

The reproductive model of privacy does not reflect the reality of applied genetics today, especially genetic testing. Information derived from genetic testing is no longer used almost exclusively in reproductive decisions. The ability to test for predispositions to polygenic (multigene) and multifactorial (environmentally influenced) genetic diseases brings genetic testing into the realm of routine health care delivery.

For most people, genetic testing —- like cholesterol testing —- is simply a tool to obtain medical information that may be beneficial to current or future health. Only a small percentage of positive genetic tests can determine future health outcomes.

The Health Insurance Portability and Accountability Act is consistent with this view.

When privacy protections are increased, they should be strengthened for all patients. Increasing protections for all medical information need not mean that genetic privacy will be insufficiently protected.

Ani B. Satz is an associate professor of law and public health at Emory University.

 MARK WEBER / NewsArt