Printer Friendly

Hedging their bets.

Predictive medicine--monitoring and even treating potential illness in people who are asymptomatic--holds the promise of transforming health care. Armed with foreknowledge of individuals' particular susceptibilities, physicians will be able not only to intervene before disease takes its toll but also to tailor the advice they give patients about diet and life-style, focusing attention on the salient measures that now get lost in the deluge of generalized guidance and admonitions.

Yet this new field holds peril as well as promise since findings about disease susceptibility may interest people other than the ones being tested. Indeed, the possible interest of insurance companies in such data has been a point of concern for those who see a double-edged sword in the techniques being developed by the genome mappers and others on the frontiers of molecular medicine.[1]

Recently, several states have responded by enacting legislation to restrict the uses that insurers can make of "genetic information." In opposing such laws, the insurance industry has claimed that they are unnecessary because insurers have no plans to make use of genetic tests now. The concern that genetic results will render some people unable to obtain health insurance reflects underlying problems in health care coverage and would best be addressed as part of current efforts to reform access to health care generally.

Nonetheless, one major life insurance company's wholesale use of an experimental screening test not only suggests that at least some insurers are interested in predictive medicine but also raises questions about how reliability data for such screening devices will be gathered. On 14 January 1993 the California Department of Insurance opened an investigation of Transamerica Occidental Life Insurance Company's use of a new blood test for cancer antibodies to screen over 50,000 applicants over the previous two years.[2] The department is looking into allegations, denied by the company, that the test was used to rate or deny applicants for insurance policies.

The willingness of Transamerica to use a test characterized by a scientist at the National Cancer Institute as unreliable and inappropriate for screening throws doubt onto the insurance industry's insistence that genetic tests are not being used since they are obviously still just experimental. The company--like any thinking of adopting a new screening tool--understandably wanted to be able to track whether the test provided reliable information for underwriting purposes. But the situation is complicated for Transamerica because it owns 30 percent of the laboratory that developed the test and hopes, if the test pans out, to market it to other insurers. Thus it appears to have used its applicants as experimental subjects.

Furthermore, the company admits that applicants were not told about the test but merely signed forms consenting to blood tests for "tumors." Again, this is problematic for Transamerica because California is one of the few states with a statute that establishes standards for human experimentation.[3] That statute requires explicit, written informed consent after subjects have been given a copy of the "experimental subject's bill of rights."

It would be surprising were insurers totally uninterested in the torrent of predictive tests that the life sciences are starting to unleash. Some tests rely on older methods, such as antibody detection, but the most eagerly awaited are those that look for genetic variation at a molecular level rather than searching for cellular products. While many genetic conditions are so rare that population screening may never be practical (unless technical advances reduce the cost to pennies per test), some diseases are more common, and even for rare ones a genetic test may be justified economically when an applicant wants a very large policy.

Recognizing that premature death or severe illness could have a catastrophic effect on their families, people are willing to pay a small amount to cushion the financial consequences of misfortune. So long as an insurance pool is large enough and participants have not joined because of atypical peril, their collective risk should correspond to the actuarial data on which their premiums were established, and the pool should have enough resources to fund the losses of its members.

Accordingly, insurers try not only to enroll large numbers but also to avoid situations in which those purchasing insurance have greater knowledge about their probability of experiencing a loss than the insurer has. Since predictive medicine involves just such a potential differential in knowledge, insurers have an obvious interest in learning any information of this type that has been provided to policy applicants.

If insurers are unable to take this information into account to set a person's premiums, or perhaps even to decline to write a policy, they fear issuing policies disproportionately to people who know they are more likely than average to experience an insured event--what is known as "adverse selection." In competitive terms, companies particularly dread being the only one that does not screen out such adverse selectors.

The most dramatic recent example of the industry's fears in this regard was provoked by the AIDS epidemic. While people in their twenties and thirties are usually eagerly sought by insurers and are charged low premiums, companies found themselves potentially insuring people who believed themselves to be at risk of being infected by HIV (or who had even already been infected) and whose insurable losses would occur at much higher rates than anticipated for the general population. As one would expect, insurers wanted to screen applicants for HIV. Their desire provoked attempts (successful in several jurisdictions) to bar AIDS testing or at least to insist that tests be performed only with the explicit, informed consent of the persons tested. It also generated a lively debate over the purposes of various types of insurance and the appropriate role of the means used by insurers to differentiate among applicants based on various perceived risks.[4]

The industry claimed that state laws mandating fairness obliged them to treat people with HIV infection differently from other people because failure to do so would mean that healthy people would be charged higher premiums than could be justified based upon their expected losses. Critics replied that the industry already allowed certain factors to override actuarial considerations, and further that society sometimes insists that other considerations, such as gender neutrality, take precedence.

No one disputed, however, that apart from social insurance programs (those that reach an entire population and that typically are funded through taxes or comparable mandatory contributions), insurance plans must be able to take certain predictors of loss into account if they are to be actuarially sound. This need is especially strong when a new illness arises that significantly changes expectations--in terms, for example, of medical costs and death rates.

Such has been the case with AIDS over the past decade; such is not the case, however, with most diseases that will be susceptible to the capabilities of predictive medicine. Rather, these conditions (such as genetic illnesses and cancer) are already included in insurers' actuarial tables. The new molecular tests may simply permit the probable victims of such conditions to be identified much earlier and more precisely.

The dilemma is thus clearly posed: if people are going to undergo predictive screening, both to advance knowledge in the field and to find out information of potential value to their own health and even their own life, they will need to be protected against unfair discrimination by insurance companies, while at the same time these companies will need to be protected against adverse selection.

Over the past two years, several states have enacted laws that explicitly limit the use of genetic tests by insurers. But the actual effect of such laws on insurance may be much more modest than the statutes suggest on their face. Since the 1970s, laws against discrimination based on sickle-cell anemia and several other genetic diseases or carrier states have been on the statute books. In 1989 Arizona became the first state to enact protection based on "genetic conditions" generally, but it merely brought such conditions within the prohibition of unfair discrimination, while permitting insurers to consider genetic risks that substantially affect actuarial predictions.[5]

Last April Florida took the stronger step of requiring informed consent for DNA analyses, except in criminal investigations. Results may not be disclosed without the consent of the person tested, but insurers are apparently free to use them to determine eligibility and rates, provided that the tester reveals this to the person tested. Should insurance be denied, the test "must be repeated to verify the accuracy."[6]

Probably the strongest statute is the one that became effective 1 July 1992 in Wisconsin. It provides a preview of what are certain to be an increasing number of efforts to limit "genetic testing," defined there as a test using deoxyribonucleic acid extracted from an individual's cells in order to determine the presence of a genetic disease or disorder or the individual's predisposition for a particular genetic disease or disorder.[7]

Under the statute, insurers (as well as employers that self-insure) are not only prohibited from requiring genetic tests but also from requesting information about previous tests. Furthermore, coverage may not be conditioned upon having a genetic test, nor may rates be determined upon test results. None of these restrictions apply, however, to life insurers, which are mandated only to behave reasonably when setting rates based on genetic data.

This willingness to allow life insurers to rely on predictive information while excluding it for health coverage is not surprising. Not only is health insurance regarded as more essential but because most of it is written in group policies, the whole notion of underwriting is under heavy attack in many quarters.

For the moment, the wisest course would be a self-imposed but strictly observed moratorium by insurers on any genetic screening of applicants, coupled with equal restraint on the part of state legislatures in enacting formal prohibitions regarding insurers' use of the new means of predictive medicine. Furthermore, research on the usefulness of the new tests ought to be carried out by scientists operating independently of the insurance industry, openly and in accord with ethical guidelines for research, including the subjects' informed, voluntary consent.


[1.] See, among others, Nancy E. Kass, "Insurance for the Insurers: The Use of Genetic Tests," Hastings Center Report 22, no. 6 (1992): 6-11; Thomas H. Murray, "Genetics and the Moral Mission of Health Insurance," Hastings Center Report 22, no. 6 (1992): 12-17.

[2.] Denise Gellene, "State Probes Transamerica's Test for Cancer," Los Angeles Times, 15 January 1993.

[3.] Calif. Health and Safety Code Subsection 24170-78 (1992).

[4.] Norman Daniels, "Insurability and the HIV Epidemic: Ethical Issues in Underwriting," Milbank Quarterly 68, no. 4 (1990): 497-525; K. A. Clifford and R. P. Iuculano, "AIDS and Insurance: The Rationale for AIDS-Related Testing," Harvard Law Review 100 (1987): 1806-24.

[5.] Arizona, HB 2517 (1989).

[6.] Florida Stat. Section 760.40 (1992).

[7.] Wisc. Stat. Ann. Section 631.89 (1992 Supp.).
COPYRIGHT 1993 Hastings Center
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:genetic testing and life insurance
Author:Capron, Alexander Morgan
Publication:The Hastings Center Report
Date:May 1, 1993
Previous Article:Commentary.
Next Article:Are laws against assisted suicide unconstitutional?

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters