On not wanting to know and not wanting to inform others: choices regarding predictive genetic testing

Abstract

Recent advancement in genetics testing for late-onset diseases raises fundamental decision dilemmas. The first study surveyed people's willingness to undergo predictive testing to find out about their own predisposition for certain incurable, late-onset diseases. The second study investigated the respondents' willingness to be tested as a function of the base rate of the disease, test diagnosticity, and the availability of treatment for the disease. In addition we surveyed (in the first study) people's willingness to disclose to others personal information about their genetic predisposition. The findings show that people often prefer not to know, as if they are choosing “protective ignorance.” Respondents' verbal justifications of their choices were also analyzed. Respondents offered emotional, cognitive-instrumental, and strategic reasons for their preferences. The findings are compared with other issues in behavioral decision theory, including attitudes toward uncertainty and desire for control. The implications of the findings for policies and legislation on genetic tests are also considered.

Acknowledgement

This research was supported by grant no. 822/00 to Ilan Yaniv from the Israel Science Foundation. Deborah Benador submitted to the department of psychology portions of this research for her masters thesis. The work benefited from the help of Liat Pollak-Basis, discussions with Gideon Keren, and comments from the participants at the London JDM seminar are also acknowledged.

Notes

¹Legislation considered in the United Kingdom (“Insurers to DNA test for genetic illnesses,” Daily Telegraph, March 20, 2000) and the United States (“Clinton proposes legislation to bolster privacy protections,” Wall Street Journal Europe, May 2, 2000).

²The gene for Huntington's disease is, in technical language, autosomal dominant, namely, it is not sex-linked and is certain to produce the disease if it is inherited from just one parent.

³The common technical distinction between risk and uncertainty is not strictly followed here. This is because for some types of disease the probabilities are precisely known, whereas for others they are not.

⁴The decision tree is presented only for illustrative purposes. Standard decision theory applies the expected value calculation to this decision. The application of such calculations is beyond the scope of the present research.

Entries in the table show the percentages of respondents who mentioned each reason category. Since respondents could mention one or more reasons (or none) the numbers do not add to 100%. Reason categories that occurred only infrequently are not presented in this table.

Entries in the table show the percentages of respondents who mentioned each reason category. Since respondents could mention one or more reasons (or else none) the numbers do not add to 100%. Reason categories that occurred only infrequently are not presented in this table.

⁵Effect size is measured by the difference between the marginal means divided by the pooled standard deviation. We follow Cohen's (1962) advice and treat effect sizes larger than 0.8 as high and those smaller than 0.2 as low.

⁶ One might argue, contrary to our position, that choosing the status quo (in Figure 1) indicates a preference for certainty (rather than uncertainty) since the outcome, ignorance, is guaranteed with certainty; thus, it is guaranteed with certainty is that one will remain uncertain about one's genetic status. The paradox arises since the outcomes are knowledge states, rather than being goods, wealth states, or health states. Consider however an individual who chooses to undergo diagnosis of an easy-to-cure illness. Such a choice is likely to be viewed as a preference for resolution (certainty), rather than for uncertainty. Therefore we argue that the knowledge state denoted as status quo would be aptly labeled “preference for uncertainty.”

⁷See Knesset site (in Hebrew) http://www.knesset.gov.il/asp/hebframe.asp?content=/laws/heb/laws.asp?type=2