|
|
| |||||||||||||||||||||||||||||||||||||||||
In this issue of the Journal, Green et al.1 describe the results of a randomized, controlled trial of genotype disclosure on the level of psychological distress of adult children of persons with Alzheimer's disease; positive genotyping for the apolipoprotein E (APOE) 
4 allele is associated with an increased risk of Alzheimer's disease. The purpose of the study was to determine whether substantial harm was caused by this testing of offspring and disclosure of the results. As part of the study, genetic counseling was provided to all subjects before phlebotomy, which was followed by random assignment to receive the results of APOE genotyping or not to receive the results. The counselors also interpreted the results of testing. Subjects in the disclosure group did not have significantly more depression or anxiety than those in the nondisclosure group either immediately after receiving the genotyping results or 1 year later, regardless of whether they were in a subgroup of subjects carrying the high-risk APOE 4 allele.
The study by Green et al. is a rare and welcome trial of a process that might inform ethics guidelines. But how reassured should we be that testing and disclosure would not be harmful? None of the subjects had high anxiety or depression scores at baseline, which would have excluded them from the study. Presumably, subjects who agreed to participate were sufficiently indifferent to the potential test results to accept randomization and thus are not representative of those who have a strong perceived need to know or need not to know. It seems possible that some subjects were unable to interpret the data. Despite the rhetoric of a new patient-centered, evidence-based society, we remain largely innumerate2 and poorly equipped to comprehend statistical probabilities. Consumer-based information on the quality of health care has had little effect on subsequent consumer decisions about care.3,4,5,6 Perhaps many subjects understood that a negative result conferred no immunity from Alzheimer's disease and that a positive result introduced too many layers of uncertainty to justify personal planning or psychological distress.
There are two levels of uncertainty with respect to Alzheimer's disease: whether and when the disease might develop and, if so, the likely experience with the disease. Alzheimer's disease can have a trajectory of more than a decade and is famously variable in its course. As Green et al. acknowledge, without such genetic counseling, subjects might not have understood the uncertainties of risk they might have and therefore responded differently. A similar caveat concerns the fact that subjects had at least one parent with Alzheimer's disease. The investigators do not present data on the course of the parents' disease or on the nature and extent of the relationship between the subjects and their affected parent, factors that may influence how a person reacts to learning the results of genotyping. Approximately three quarters of the subjects reported being a caregiver for a relative with Alzheimer's disease, but the term "caregiver" was not defined and could connote various types of involvement with varying intensity and duration.
Fryback and Thornbury7 proposed a six-part framework for analyzing the value of diagnostic tests, including technical accuracy, diagnostic accuracy, effect on diagnosis, effect on treatment, patient outcomes, and effect on society. Knowledge of the results of APOE genotyping has no effect on treatment, although in this area, the science is rapidly evolving, as discussed by Caselli et al.8 in their report in this issue of the Journal on the temporal effect of APOE 4 on cognition. However, the study by Green et al. prompts thought about patient outcomes and the effect on society.
On an individual level, the value of diagnostic testing lies in its ability to produce benefits that outweigh the risks. In this regard, how much solace does the study by Green et al. provide? The investigators measured negative psychological outcomes, but not negative and positive social and financial outcomes. Some APOE 4 carriers might have responded to the disclosure of their increased risk by acting precipitously and unnecessarily on the information to get their affairs in order, fashion advance directives, avoid major residential or occupational changes, and save money or, conversely, spend it rapidly. They might have become ineligible for certain types of insurance; the Genetic Information Nondiscrimination Act of 2008 specifically prohibits the consideration of genetic-testing results for employment or health insurance but does not prohibit companies from considering such results for long-term care, disability, or life insurance. Moreover, APOE 4 carriers might have risked losing various social opportunities and relationships if their heightened risk of Alzheimer's disease had become known.
Green et al. followed the subjects for a year, but persons who tested positive for the risk allele (or who simply received disclosure) might react years later, perhaps when they first forget a name or observe distress in a relative with dementia. Green et al. point out that larger studies with longer follow-up are needed to detect effects such as "delayed emotional repercussions and injudicious life decisions." Also, the subjects were a well-educated group with a mean age in the early 50s; the results of the trial may have limited relevance to younger or less educated subjects.
Could any societal good or harm result from widespread, on-demand genetic testing for APOE 4? Societal benefit seems unlikely, given that it is not possible to prevent Alzheimer's disease, and available treatments have at best a modest effect on disease progression. Societal harm from testing remains possible, given that Alzheimer's disease is widely regarded as being singularly horrific,9,10 although persons with Alzheimer's disease can retain their sense of humor, experience existential joy, and overcome the sum total of so-called negative behaviors that are typically measured in care settings. One hopes that the subjects in this study had experiences with their affected family members that were consistent with this observation. With an increasing public familiarity with Alzheimer's disease and improving choices in community care for those with the disease, perhaps the social effects of genetic testing will be less worrisome by the time a clinical rationale for the test becomes apparent.
Dr. Robert Kane reports receiving consulting fees from SCAN Health Plan, United Health Group, and Medtronic. No other potential conflict of interest relevant to this article was reported.
Source Information
From the School of Public Health and Center on Aging (R.A.K., R.L.K.) and the Center for Biomedical Ethics (R.A.K.) — all at the University of Minnesota, Minneapolis.
References
4 effect. N Engl J Med 2009;361:255-263.
| |||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2010 Massachusetts Medical Society. All rights reserved. |
Previous
