Results

A total of 105 eligible women from 70 different families were mailed a questionnaire, including 71 women from the University of Toronto and 34 from McGill University. The women were white and of French-Canadian, Ashkenazi Jewish, European, or Hispanic ethnicity. Sixty four questionnaires were returned by mail and 15 were completed by telephone. Analysis was performed on a total of 79 completed questionnaires (17 from McGill and 62 from the University of Toronto).

There were 26 non-respondents (24.8% of questionnaires mailed). Seven women declined to participate (via returned consent form indicating refusal to participate) and 19 women did not respond to the invitation to participate in this study. The mean age of the non-respondents was 52.6 years (range 39 to 65 years), compared to 50.4 years in respondents (range 24 to 81 years) (p=0.41). Seventy seven percent of the non-respondents had a previous diagnosis of cancer, whereas 58% of the respondents had a personal history of cancer. The average level of education for the respondents was some college or university, 11% had less than high school education, and 14% had a postgraduate degree. The education level for non-respondents was not known.

Table 2 summarises the sources of referral. Forty four percent of respondents were referred to a genetics clinic by a family member. Fifteen percent were self-referred and 11% were referred by an oncologist. Forty three percent of respondents belonged to a family with a knownBRCA1/2 mutation. Of these, 62% (21/34) were referred by family members.

Fifty eight percent of respondents had a previous diagnosis of cancer. Of these, thirty seven had unilateral breast cancer, three had bilateral breast cancer, and six had ovarian cancer. Other reported cancers were colon, uterine, melanoma, and cervical cancers. The median year of cancer diagnosis was 1990 (range 1961 to 1998). The mean number of breast cancer cases in the families in first or second degree relatives was 3.0 (range 0-7) and the mean number of deaths from breast cancer was 1.1 (range 0-5). The mean number of ovarian cancer cases in the families was 1.2 (range 0-3) and the mean number of deaths from ovarian cancer was 1.0 (range 0-3).

The mean time between respondents receiving mutation results and completing the questionnaire was 17.3 months.

The top three reasons for seeking genetic services were “to learn about personal risk of cancer” (34.2%), “to learn about children's risk of cancer” (16.5%), and “to learn about family's risk of cancer” (11.4%) (table 3). Among women with children, personal risk was still cited as the principal reason for undergoing genetic testing.

Open ended questions were used to assess the patients' emotional responses to their positive mutation test result. When asked to describe initial feelings upon receivingBRCA1/2 results, responses included: “not surprised”, “relieved”, “angry”, “sad”, “devastated”, “shocked”, “concerned about children”, and “guilty for passing gene on to my daughters”. These responses were all cited by multiple respondents. Subjects were also asked if feelings and emotions had changed since first learning about being a carrier ofBRCA1/2. Responses given by more than one woman included: “I now realise what it really means to have the gene”, “I am more realistic about the future”, and “I am more concerned about how my children will be affected”.

INFORMATION NEEDS

Seventy seven percent of women (59/77) were satisfied with the information they received during genetic counselling. Twenty one percent (16/77) said there was additional information that they would have liked, but did not receive, including information on cancer treatment. This included information on preventive surgery, chemoprevention, and new screening modalities (for example, MRI). In the study sample there were no significant differences in information needs based on age or education (table 4), but there was a significant difference by cancer status. Women affected with cancer were more likely to state that there was information they needed but did not receive at the result disclosure session than were unaffected women (p=0.05). There were no differences in information needs based on centre (Toronto v Montreal) (p=0.40). There were also no differences in information needs based on the time since the subject had received a positive mutation result (p=0.95).

Twenty eight percent of women felt they had not received enough information about prophylactic surgery during their genetic counselling sessions. Eighteen percent of the respondents felt they did not receive enough information about cancer screening during their genetic counselling sessions. There were no significant differences in need for information about prophylactic surgery or screening based on education, age, or cancer status (table 4).

The great majority of women (95%) shared the information they received, including their test result, with other family members. One third of the women indicated that the experience of genetic testing had affected either family or personal relationships. There was a significant difference in this response based on whether the woman was the proband or has another family member who received predictive testing (p=0.02). Probands were more likely to state that their personal or family relationships had been altered as a result of the testing process. Of the 26 women who had indicated that their family relationships had changed as a consequence of the genetic testing, two did not indicate how these had changed. Twelve stated that their family relationships were strengthened by this information: “we value each day more and each other”; “(genetic testing) has brought the issue to the surface to be discussed openly and honestly”; “(I am) not waiting to do things any more with my kids”. Twelve women stated that genetic testing had a negative impact on their family or personal relationships: “my husband is even more distant”; “my daughter is terrified”; “my preoccupation with the test results and its implications has affected my availability to my family (spouse and children)”.

SUPPORT RESOURCES

The genetic counsellor was cited as the most frequent source of psychosocial support, followed by a family member and spouse (table 5). The genetic counsellor is the primary contact for each woman in the genetics clinics. This may explain the importance of the genetic counsellor in providing psychosocial support. The majority of women (84%) did not want help or support in sharing the genetic information (their test result) with their family. Nearly one third of women (23/76) would have liked their families to have been more involved in the genetic testing process.

Nineteen percent of women felt they needed more support than was received. There were no significant differences in whether women required more support according to cancer status, age, or education. Half (39/77) of the women said they were interested in being referred to another health care professional after they had received their test result. The most common request was for a gynaecologist (22/39), followed by a breast specialist (7/39). When asked if they were successful in obtaining the referral, 74.4% said yes.

Participants were asked about the need for a support group forBRCA1/2 mutation carriers and about their desire to participate in one. Over two thirds (68.4%) felt a support group was necessary forBRCA mutation carriers, but only 34.2% said they wished to participate. Women with an educational level beyond high school without a previous diagnosis of cancer were more likely to be interested in taking part in a support group than those with less education (p=0.009) or without cancer (p=0.034).

HEALTH CARE PRACTICES

Screening

More than half of women (45/78) reported their cancer screening practices had changed since they learned of theirBRCA1/2 carrier status. Women younger than 50 years old (p=0.08) and women with no previous diagnosis of cancer (p=0.0004) were most likely to have increased their cancer screening practices. The majority of women (63/75) did not have any difficulties obtaining regular screening. Table 6 summarises the self-reported screening practices of respondents for breast cancer and ovarian cancer.

Prophylactic surgery

Nearly two thirds of the respondents (47/72) said they had considered prophylactic surgery either of the breasts or ovaries. Women below the age of 50 were more likely to have considered prophylactic surgery (80.5%) than women aged 50 years and above (45.2%) (p=0.002). Among those with no previous diagnosis of cancer, women aged less than 50 years were more likely to have considered prophylactic surgery (86.4%) than older women (50%) (p=0.054). Similarly, among women with a previous diagnosis of cancer, younger women (<50 years) were more likely to have considered prophylactic surgery (73.7%) than older women (44%) (p=0.049). Women with an educational level above high school were more likely to have considered prophylactic surgery (79.6%) than those with an educational level of high school or below (34.8%) (p=0.0002).

Twenty two of the 76 participants (29%) had either contralateral or bilateral prophylactic mastectomy. Sixteen of these women (73%) had prophylactic mastectomy following the genetic test result disclosure, whereas six women had prophylactic mastectomy before receiving results. Sixty four percent (14/22) of women who had prophylactic mastectomy completed the questionnaire less than 18 months after receiving their genetic test results. Of the 37 women who had unilateral breast cancer, 40.5% had contralateral prophylactic mastectomy following the initial diagnosis. Of women with no previous breast cancer, 15.4% (6/39) had prophylactic bilateral mastectomy.

The number of breast cancer cases and breast cancer deaths in the family were not found to be related to the decision to have prophylactic mastectomy. Women who had a prophylactic mastectomy had a mean of 2.4 breast cancers in the family and 1.1 deaths from breast cancer, whereas carriers who did not have prophylactic mastectomy had a mean of 3.2 breast cancers and 1.3 deaths from breast cancer in the family (p=0.211 and p=0.712, respectively). Similarly, the number of ovarian cancer cases and ovarian cancer deaths in the family were not found to be related to the decision to have prophylactic oophorectomy. Women who had prophylactic oophorectomy had a mean of 1.2 ovarian cancers in the family and 1.1 deaths from ovarian cancer, whereas carriers who did not have prophylactic oophorectomy had a mean of 1.0 ovarian cancers and 0.8 deaths from ovarian cancer in the family (p=0.353 and p=0.229, respectively).

The rates of oophorectomy were higher. Among women who had no previous diagnosis of ovarian cancer, 53.6% (30/56) had prophylactic bilateral salpingo-oophorectomy. Fifty percent of women who had prophylactic oophorectomy completed the questionnaire less than 18 months since receiving their genetic test result. There was a statistically significant difference based on educational level; those with an educational level of high school or less (92.9%) were more likely to have a prophylactic oophorectomy than those with an educational level above high school (40.5%) (p=0.0007). When educational level was adjusted for age, there was still a significant difference (p=0.01). These results included all women who had undergone a prophylactic oophorectomy before and after genetic testing. When eight women who had undergone a prophylactic oophorectomy before genetic testing were excluded from the analysis, education level was still significantly associated with prophylactic oophorectomy (p=0.026).

SATISFACTION

Overall, 81.3% (61/75) of women believed that they were receiving adequate medical care in the prevention of breast and ovarian cancer. Over one third (33.8%) felt the need to visit the genetic counsellor again after receiving their test result. An overwhelming majority (92.3%) said they would recommend testing forBRCA1/2 to other women in their situation. The respondents were asked to rate their overall satisfaction with their experience of undergoing genetic testing on a scale of one to five, one indicating extremely dissatisfied and five indicating extremely satisfied. Nearly 65% (51/79) said they were very or extremely satisfied. Only three patients were very or extremely dissatisfied, including one respondent who believed she was denied disability insurance based on her genetic test result.

Patient satisfaction was also analysed according to the length of time since the mutation result was received. Thirty eight women had received mutation results greater than or equal to 18 months, and 41 women had received results less than 18 months ago. There was no difference in patient satisfaction based on time since receiving mutation result (p=0.95).

Discussion

In this study, we surveyed the needs of women undergoing genetic counselling and predictive testing in two familial breast-ovarian cancer clinics in Canada. The reasons cited by the participants for seeking genetic counselling were similar to those of previous reports on motivations for havingBRCA1/2 genetic testing.16 22 23 The significant proportion of participants (34/79) who underwent predictive testing for a known family mutation may explain why the most common reason for having genetic testing was “to learn about personal risk of cancer”, while in the other studies concern about risk of cancer to children was the primary motivation.16 23

One of the goals of cancer genetic counselling is to promote surveillance for early detection of cancer amongBRCA mutation carriers.24 The majority of women did not have any difficulties obtaining screening. This may be because of the universal access to health care in Canada. There may be financial concerns with health care for US citizens. The potential difficulties to be encountered in the Canadian health care system may relate to waiting time and access for rural communities. We found that younger women (below the age of 50) and women with no previous diagnosis of cancer were most likely to change their screening practices following genetic counselling and testing. It is possible that older women are more likely to be involved in cancer screening protocols based on recommendations currently in place for the general population. It has been shown that women do not necessarily adhere to recommended screening or preventive options discussed when they are seen for results or follow up.6 Adherence to surveillance measures, such as breast self-examination and mammography, appears to be influenced by levels of anxiety, intrusive thoughts about cancer, psychological distress, education, employment, and age.15 25 A high degree of adherence to recommended mammographic screening guidelines has been reported among women at high risk for breast cancer.17 In contrast, in a group ofBRCA carriers previously studied, only 39% reported receiving recommended mammographic screening.6Table 6 describes the self-reported screening practices of our group of patients. The majority of women in all categories reported receiving regular screening for early detection of breast and ovarian cancer. The high adherence observed here may represent errors in recollection and, in future studies, it would be preferable to request the date of last screening or compare attendance for scheduled appointments at a clinic.

Nearly two thirds of the respondents (65.3%) said they had considered prophylactic surgery (either mastectomy or oophorectomy). The option of prophylactic surgery is routinely presented to women as a method of risk reduction during the pre-test genetic counselling sessions at the University of Toronto and McGill University familial cancer clinics. The content of this discussion has not changed considerably for the women who considered this option; in fact, McGill University researchers were counselling women on prophylactic surgery using linkage analysis before the cloning of BRCA1in 1994. Women below the age of 50 and those with an educational level above high school were more likely to have considered this option. Wagner et al 29 recently described how 50% of female carriers in Austria would undergo prophylactic oophorectomy (53% affected and 46% non-affected carriers). Previously, Lerman et al 10 reported that 31% of femaleBRCA mutation carriers considered prophylactic mastectomy and 48% considered prophylactic oophorectomy. However, at six months post-result disclosure, Lermanet al 10 determined that only 1% had undergone prophylactic mastectomy and 2% had undergone prophylactic oophorectomy. In contrast, among our study respondents, 28% had prophylactic mastectomy and 54% had prophylactic oophorectomy on follow up interview. This increased frequency of prophylactic surgery in our study may be explained by a longer follow up period among our study subjects. The difference may also reflect temporal changes in attitudes to prophylactic surgery. Our physicians and patients may be more accepting of the effectiveness of prophylactic surgery in the prevention of breast and ovarian cancers. Recent findings by Hartmann et al 28have suggested that prophylactic mastectomy offers at least an 80% reduction in risk of breast cancer in high risk women. This knowledge may influence the recommendations about prophylactic surgery that are given to BRCA1/2mutation carriers. The differences between our findings and those of Lerman et al 10 may also reflect a difference between the health insurance systems in Canada and the United States. In Canada, prophylactic surgery (including breast reconstruction) is covered by the universal provincial health insurance plans. At present, concerns about health care coverage are minor in Canada, but possible discrimination in life insurance, disability insurance, and long term care insurance are areas of concern in both countries.26 One respondent was dissatisfied with the genetic testing experience and would not recommend it to others because she believed that she was denied disability insurance on the basis of her mutation carrier status. Further research is needed in the area of genetic information and insurance discrimination in Canada.

Women with a previous diagnosis of breast cancer were more likely to opt for prophylactic surgery (of the contralateral breast) than were healthy women (table 4). Carriers with an educational level above high school were more likely to have had a prophylactic mastectomy compared to less educated women, although the association was not statistically significant (p=0.07). In contrast, women with an educational level above high school were less likely to have undergone prophylactic bilateral oophorectomy than less educated women (p=0.0007); this effect remained after adjusting for age (p=0.01). We believe that surgical removal of the breasts has a greater impact on a woman's self-image than oophorectomy. Level of education may be associated with the relative importance that a woman places on self-image and on the consequence of surgical early menopause. This may explain the effect observed for educational levels. For the carriers in this study, the number of breast or ovarian cancer diagnoses or deaths in the family were not found to be related to the woman's decision to proceed with surgery.

Lerman et al 10 concluded that those who opt for genetic testing might represent a psychologically vulnerable subgroup of high risk women. In our study, probands were more likely than their relatives to report that their personal or family relationships had been affected by the genetic counselling process (p=0.02). This probably represents the added anxiety of being the initiator of the genetic testing for the family or “gate keeper” role in disclosing this information to other relatives at risk. One third of women stated that their family or personal relationships had changed since they received their mutation result. There was an equal distribution of positive and adverse effects on family relationships. Although 12 of the women indicated that genetic testing had a negative impact, the majority of the women would recommend genetic testing to other women in their situation. This suggests the value of genetic test results to the subject at risk for cancer, as was described by the women in this study. As previously stated, the primary motivation for undergoing genetic testing in this population was to learn about personal risk for cancer (table 3). The information that a person receives from genetic testing has potentially life altering consequences, including both positive and negative effects on relationships. Health care professionals may wish to make clients and their families aware of these effects and offer support when necessary.

Lynch et al 18 reported that those who are told they have a germline mutation express a variety of reactions including acceptance because the results are not a surprise to them, relief from anxiety with the removal of uncertainty about their genetic risk status, a positive attitude with regard to prevention, and feelings of sadness or even anger. Similarly, our respondents reported a wide range of emotional reactions. We agree with Lynch et al 18 that genetic counsellors should be responsive to all of these emotions.

Overall, 81% of respondents indicated they did not need additional support. Younger women (below the age of 50) were more likely to report a need for greater support, although this was not statistically significant (p=0.11). Audrain et al 20 assessed the preferences for the content and process of genetic counselling and testing for breast-ovarian cancer susceptibility among women at high risk. Younger women (below the age of 50) were more likely to request a discussion of emotional reactions to testing and emotional support in post-test counselling. Women with higher levels of education and no previous diagnosis of cancer expressed a greater interest in taking part in a support group (table4).

Bleiker et al 22 found high levels of satisfaction with clinical genetic services received at family cancer clinics in The Netherlands. Communication of information regarding the possible impact of genetic testing on daily life and psychosocial support during and after the process of genetic counselling were identified as areas that needed attention. Client satisfaction is often cited as a measure of success in patient-physician communication, but this item is inherently flawed. The fact that clients take the time to complete a questionnaire is itself an indication of satisfaction, in that dissatisfied clients would not commit their time and effort to answer a physician survey. In our study, a relatively high response rate of 82.3% (79/96) was achieved and nearly 65% (51/79) of respondents said they were very or extremely satisfied. An overwhelming majority (92.3%) of the women said that they would recommend genetic testing to other women in their situation. This, combined with high satisfaction and response rates, suggests that counselling is effective in helping a woman throughout the genetic testing process. Based on our respondent who believed she was denied disability insurance based on her genetic test result, it is important for health care professionals working in the genetics field to communicate clearly the current knowledge of employment and insurance implications and potential discrimination based on a positive BRCA mutation result.

Genetic counsellors and other health care professionals must also be sensitive to the emotional sequelae of disclosing positive mutation results to women, particularly in the period following disclosure, and must attempt to meet the needs of female carriers as shown in this study. Equally important is the effect of the woman's genetic test result on her family relationships as recorded in one third of our respondents. These findings have a direct impact on the counselling and risk management of BRCA carriers. Further work is under way to assess the effectiveness of a group counselling intervention for BRCA mutation carriers.