Pilot Study on Patients' and Spouses' Attitudes Toward Potential Genetic Testing for Bipolar Disorder

A 46-item questionnaire was distributed to 90 patients with bipolar disorder and to their spouses (a total of 180 questionnaires) between June and September 1993. Subjects with bipolar disorder were drawn from participants in the Johns Hopkins University bipolar disorder genetic linkage study, members of the Depression and Related Affective Disorders Association, and inpatients and outpatients of the affective disorders division of the Johns Hopkins University Hospital. An introductory letter describing the purpose and nature of the study was sent to 84 linkage study participants; this letter was followed, in turn, by a phone call from a researcher asking permission to send a questionnaire. Questionnaires were sent to 32 of the 84 linkage study participants and their spouses. Support group leaders of the Depression and Related Affective Disorders Association in Maryland, Pennsylvania, and Washington, D.C., were contacted to request permission to distribute questionnaires to their group members. Questionnaires were given to 52 of the members and their spouses. Some individuals completed the questionnaires at their group meeting, and others completed them at home and mailed them back. Questionnaires were also given to three outpatients and their spouses and to three inpatients and their spouses. To be eligible for the study, only one member of a married couple could have a diagnosis of bipolar disorder. All of the outpatient couples mailed the questionnaires back to us, while all of the inpatient couples completed them in the hospital.

After complete description of the study to all subjects, written informed consent was obtained. Patients and spouses were asked to complete the questionnaires independently. In most instances, a researcher was present to answer any questions that the subjects had. In instances where no researcher was present, the phone number of a researcher was given to the subjects so that they could call if they had any questions.

The questionnaire included 40 multiple-choice questions (20 of these in a Likert scale format and seven involving percentages), four questions requiring numerical answers, and two questions requiring an open, written response.

The subjects were asked to give social and demographic information and to designate the clinical diagnosis (bipolar type I or bipolar type II) of the affected member of the couple. Additional questions addressed the following issues: 1) the degree of knowledge that the subjects had about the genetics, severity, and rates of response to treatment of bipolar disorder; 2) the likelihood that these same individuals would take advantage of presymptomatic and prenatal genetic tests for bipolar disorder if they were to become available; 3) perceptions about the potential benefits and risks of knowing whether one carries a gene for bipolar disorder, and whether the benefits of knowing outweigh the risks; 4) whether knowledge that one was carrying a gene for bipolar disorder would affect decisions to get married and to have children; 5) preferences regarding the disclosure of bipolar disorder genetic information; 6) attitudes toward aborting a fetus with a gene for bipolar disorder; and 7) a comparison of subjects' attitudes in a wider range of situations that might justify abortion. We modeled several of the questions about abortion after those designed by Wertz et al. (16).

Paired t tests were done on questions involving percentages, comparing a patient's and spouse's paired responses. McNemar chi-square statistics were used for paired categorical responses. We found no statistically significant differences in the responses to most questions. We have detailed the statistical analysis in this report only in instances where a statistically significant difference was found.

Of the 180 questionnaires distributed, 90 were returned to us. Thirty-four of the questionnaires received were completed by bipolar type I patients and their spouses, 42 were completed by bipolar type II patients and their spouses, and 14 were completed by couples for whom we were unable to ascertain the specific type of bipolar disorder in the affected individual. Of the 45 couples who responded, the patients in 16 were participants in the genetic linkage study, the patients in 23 were members of the Depression and Related Affective Disorders Association, and there were three outpatients and three inpatients of the affective disorders unit of the Johns Hopkins University Hospital. The majority of the questionnaires were fully completed; however, some questions were left blank by certain subjects.

The mean age of the respondents with bipolar disorder was 48 years (SD=10.7), and the mean age of the spouses was also 48 years (SD=12.6). Of the respondents with bipolar disorder, 31.1% (N=14) reported a diagnosis of type I disorder, 51.1% (N=23) reported a diagnosis of type II disorder, and 17.7% (N=8) were not sure which subtype of bipolar disorder they had. Most patients (88.9%, N=40) and spouses (82.2%, N=37) were in their first marriage. The number of biological children of patients and spouses ranged from one to six (for both, mean=1.9, SD=1.5). More than one-half of the patients (53.3%, N=24) and spouses (60.0%, N=27) were college graduates.

Our data indicate that attitudes toward presymptomatic and prenatal testing for bipolar disorder were quite positive. The majority of the bipolar patients and their spouses stated that they would want to take advantage of presymptomatic and prenatal genetic tests for bipolar disorder if they became available. Furthermore, the majority of both patients and spouses were in favor of testing minors for the gene. The greatest degree of uncertainty for both patients and spouses involved the testing of a fetus.

Importantly, the vast majority of patients and spouses believed that the benefits of knowing whether they or their spouse carries a gene for bipolar disorder outweigh the risks. The benefit ranked most important by both patients and spouses was to obtain treatment to prevent episodes of illness. Only a small group of patients (10.0%) and spouses (15.0%) felt that knowing that one carried a gene for bipolar disorder would make it harder for that person to reach his or her goals in life.

The overwhelming majority of the bipolar patients and spouses indicated that they would not choose to abort a fetus with a gene for bipolar disorder. Overall, respondents were less likely to justify abortion for carrying a gene for bipolar disorder than they were for carrying a gene for any of the other hypothetical conditions presented. This is in agreement with a study by Smith et al. (15) which showed that members of bipolar disorder support groups, psychiatric residents, and medical students would all be more inclined to abort a fetus for another life-threatening or severely debilitating disorder than they would for bipolar disorder. However, compared to our patient and spouse groups, a greater percentage of their respondents indicated that they would end a pregnancy if the fetus were to definitely develop an unspecified course of bipolar disorder. This may be because our question left open the possibility that an individual may in fact have a gene for bipolar disorder but not develop any clinical symptoms. (When asked to estimate the likelihood that one could have a gene for bipolar disorder but not show any signs of the disorder at all, our patients gave a mean estimate of 31.7%, and the spouses gave a mean estimate of 30.5%.) Smith et al. also found that the intention to abort a fetus would be influenced by the projected severity of the illness and the likelihood of developing bipolar disorder. However, given the multifactorial nature of bipolar disorder, it is possible that a positive result of a genetic test for the disorder would still leave uncertainty about whether one will in fact develop the disorder. In addition, a positive genetic test result might not provide information about the severity of the illness.

It is interesting to note that there were no significant differences between the attitudes of patients and those of their spouses toward most of the issues addressed.

The relatively high rate of uncertain responses regarding childbearing, prenatal testing, and abortion indicate areas in which patients and spouses perceive a need for more information.

This was a pilot study, and several limitations may have influenced our results. First, we had a 50% response rate to our questionnaire, which is comparable to the general response rate for medical questionnaires (17). Another limitation of the study was the small size of the study group. Trends that did not reach statistical significance in our study might well have reached significance with a larger group. In addition, it is apparent that our subject group was in some respects quite skewed; ascertainment was not systematic, and most respondents were well-educated and in their first marriage. The patients in their first marriage may represent those whose illness has caused less marital disruption. Most of the patient respondents were members of support groups or participants in a genetic study and would be expected to be better-educated and more willing to disclose personal information about the genetics of their disease.

Our study did not explore the length of time that affected individuals had been treated for bipolar disorder or how efficacious their treatment had been. These factors would be expected to affect attitudes toward marriage, childbearing, and disclosure of genetic information. In addition, we had no method of ensuring that patients were euthymic at the time that they completed the questionnaire. Finally, we had no normal (nonclinical) control group.

Despite the limitations of the study, the results are of considerable interest and value in terms of providing a basis for further research. The trends we observed suggest a strong likelihood that individuals will make use of genetic tests for bipolar disorder once they become available. In addition, they convey attitudes that may shed light on some of the concerns that have been raised regarding the possible misuse of genetic information. However, we must be careful about extrapolating from hypothetical intentions to actual requests for genetic testing. In the case of Huntington's disease—an incurable, genetic (autosomal dominant) neuropsychiatric degenerative disease for which genetic testing is now available—the current demands for testing are substantially below the rate predicted by early surveys (18, 19). A study by Quaid and Morris (20) surveyed 66 people at risk for Huntington's disease who chose not to undergo testing for the disorder. The five most important reasons that they cited for declining testing were lack of a treatment, increased risk to their children, the prospect of losing health insurance, financial costs of testing, and “the inability to undo the knowledge.”

The discovery of a genetic test for bipolar disorder will inevitably raise some similar concerns, but bipolar disorder differs in many respects from Huntington's disease. The availability of treatment, high rates of response to treatment, and aspects of the illness that might confer advantages to the individual, to interpersonal relationships, and to society may outweigh the potential disadvantages at the time of actual testing (1, 9–12).

Additional studies with larger study groups and broader sampling of the population would be useful. It would be informative to address the issue of how the degree of accuracy of a genetic test for bipolar disorder would influence people's attitudes toward testing. It is not clear how much genetic information will be required to justify making a genetic test for bipolar disorder available. Careful consideration must be given to the issue of testing minors. The question of whether parents' desire to know that their child has a gene for bipolar disorder outweighs the child's right to privacy—particularly if such knowledge could lead to earlier treatment and avoiding potential harm to the child—remains to be answered. Related issues, such as circumstances under which it might be permissible to deny testing to individuals and whether there might ever be circumstances that warrant mandatory testing, have already become real issues in the genetic testing for Huntington's disease (21) and will undoubtedly have to be addressed in future testing for bipolar disorder. Data from continuing research on the issues identified here will be essential in facilitating the development of appropriate guidelines for the eventual genetic testing for bipolar disorder.

Received April 28, 1997; revision received Jan. 6, 1998; accepted Feb. 9, 1998. From the Department of Psychiatry, Johns Hopkins University School of Medicine; the Department of Psychiatry, Tufts University School of Medicine, Boston; and the Department of Psychiatry, University of Maryland School of Medicine, Baltimore. Address reprint requests to Dr. Trippitelli, Johns Hopkins University Hospital, Meyer 3-181, 600 North Wolfe St., Baltimore, MD 21205-2182. Supported by the Drs. Kienle Center for Humanistic Medicine at the Pennsylvania State University College of Medicine and the Dana Consortium for the Genetic Basis of Manic-Depressive Illness. The authors thank Dr. Peter Houts for assistance with the design of the questionnaire, Dr. Kimberly Quaid and Dr. Neil Holtzman for providing copies of questionnaires they have used for studies on disclosure of genetic information, Dr. Jason Brandt for sharing his work on the ethics of genetic testing for Huntington's disease, Barbara Schweizer and Sally Mink for assistance with referring individuals for the study, and the patients and spouses who participated in the study.