- Email: [email protected]
2-mutation carriers
Patient Education and Counseling 102 (2019) 1925–1931
Contents lists available at ScienceDirect
Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou
Accuracy in risk understanding among BRCA1/2-mutation carriers Dorothee Speisera,1,* , Felix G. Rebitschekb,1, Markus A. Feufelb,c, Hannah Brandd , Laura Beschd , Friederike Kendeld a
Department of Gynecology with Breast Center, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany Harding Center for Risk Literacy, Max Planck Institute for Human Development, Berlin, Lentzeallee 94, 14195 Berlin, Germany Division of Ergonomics, Department of Psychology and Ergonomics, Technische Universität Berlin, Marchstr. 23, 10587 Berlin, Germany d Institute of Medical Psychology, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 18 October 2018 Received in revised form 7 March 2019 Accepted 4 May 2019
Objective: BRCA1/2-mutation carriers are at high risk of developing cancer. Since they must weigh clinical recommendations and decide on risk-reducing measures, the correct understanding of their 10-year cancer risks is essential. This study focused on the accuracy of women’s subjective estimates of developing breast and ovarian cancer within ten years as prerequisite to reduce unnecessary prevention. Methods: 59 and 52 BRCA1/2-mutation carriers provided their individual risks of developing breast or ovarian cancer in the next 10 years, along with self-reported sociodemographic and psychosocial variables. Women’s risk estimates were compared with their objective cancer risks that had been communicated before. Results: 22.6% of counselees under- and 53.2% of the counselees overestimated their 10-year risk of developing breast cancer. As for ovarian cancer, 5.6% under- whereas 51.9% overestimated their risk. Neither demographic factors such as education, parenthood and age, nor a prior diagnosis of breast cancer or prophylactic surgery accounted for these variations in risk accuracy. Conclusion: Currently, risk communication during genetic counseling does not guarantee accurate risk estimation in BRCA-mutation carriers. Practice Implications: Counselors must be prepared to prevent overestimation. Counselees’ risk estimates need to be assessed and corrected to enable informed decision-making and reduce risks of unnecessary preventive efforts. © 2019 Elsevier B.V. All rights reserved.
Keywords: BRCA1/2 gene mutation Genetic counseling Subjective risk estimation Risk perception Risk accuracy
1. Introduction Women carrying a deleterious mutation in BRCA1 or BRCA2 have a high lifetime risk of developing breast or ovarian cancer. The lifetime risks for breast cancer in BRCA1-mutation carriers range from 65 to 79%, and from 61 to 77% for BRCA2-mutation carriers, respectively. Ovarian cancer risks for BRCA1-mutation carriers range from 36 to 53%, for BRCA2-mutation carriers from 11 to 25% as lifetime risks [1]. For mutation carriers, the question of risk management, that is, early detection and/or preventive measures, arises as soon as the mutation status is revealed. A precondition for making informed decisions regarding risk management strategies
* Corresponding author at: Charité – Universitätsmedizin Berlin, Department of Gynecology with Breast Center, Campus Charité Mitte, Charitéplatz 1, 10117 Berlin. E-mail addresses: [email protected] (D. Speiser), [email protected] (F.G. Rebitschek), [email protected] (M.A. Feufel), [email protected] (F. Kendel). 1 DS and FGR are shared first authors. https://doi.org/10.1016/j.pec.2019.05.007 0738-3991/© 2019 Elsevier B.V. All rights reserved.
is that affected women understand their individual disease risks [2]. During counseling on risk management, women’s objective disease risks are presented as lifetime risks together with several age-specific cumulative risks. As the objective risk depends on clinical parameters such as age, affected genes, pre-existing cancer, and preventive measures already taken, the complex interplay of these parameters is likely to influence also the individual’s risk perception and understanding. Furthermore, risk perception is associated with psychosocial variables such as personal experience [3,4], education, and cultural differences [5]. Women with a higher risk have higher levels of cancer worry [6]. Also, anxiety seems to increase the perception of vulnerability and is thus likely to inflate risk perceptions [7]. Last but not least, the counseling process itself, in which large amounts of information need to be communicated, could lead to inaccurate assessments of risk [8]. Over the last years, the number of prophylactic surgeries, such as bilateral risk reducing mastectomies (BRRM), has increased substantially. In part, this increase can be traced back to the enormous public attention the actress Angelina Jolie attracted in
1926
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931
2013 when she made public her BRCA1/2-gene mutation and her decision for a prophylactic mastectomy [9]. Since then, BRRMs have also increased in women with a negative BRCA1/2-test result, although prophylactic surgery in this population has not been shown to improve overall survival [10–12]. One reason for the increased BRRM-rate among BRCA-mutation carriers and noncarriers could be a lack of risk understanding in counselees at tumor risk clinics [13]. Many women may opt for prophylactic mastectomy despite an unclear survival benefit because they overestimate their disease risk [14]. In a study with women undergoing prophylactic surgery, risk overestimation and confusion about risks prevailed even after surgery and were accompanied by increased anxiety [15]. These findings underline the clinical relevance of accurate risk perception for informed risk management decisions. Studies on risk estimation thus far predominantly focused on perceptions of lifetime risks [16–23] or on how women perceived their own risk in relation to the risk in the general female population (comparative risk estimates; [16,24,25]). However, it is becoming increasingly clear that, clinically, the most important time parameters on which relevant screening and prophylactic recommendations should be based are 10-year risks [26]. Only two studies [13,27] assessed the accuracy of subjective risk estimates in comparison with objective 10-year-risks in women with a family history of breast cancer. Abbott et al. [28] compared different clinical subgroups with respect to their subjective 10-year-risk estimates but did not compare these risk estimates to their objective risk. Both studies explicitly excluded women with known BRCA1/2- mutations and focused on women with a known family history of breast or ovarian cancer instead. Although risk perception was measured differently in the above-mentioned studies (either with verbal labels or numerically), it can be concluded that counselees overestimated rather than underestimated their breast cancer risks. Assessing the individual’s subjective estimation of risk is challenging. Generally, the constructs “risk understanding” and “risk perception” need to be distinguished. Whereas risk understanding depends on an individual’s knowledge and is assessed with numerical estimates (e.g. 3 out of 100), risk perception also includes subjective vulnerability to health threats and is usually assessed with verbal labels (description of a risk as “low” or “high”) [7]. Most studies, so far, have focused on risk perception. Guideline-based counseling requires that women are informed about their individual cancer risk which is composed of various factors, including mutation type, familial and personal cancer history, and age [29]. From a clinical perspective, the 10-year risk of developing breast or ovarian cancer usually guides the counseling and recommendations. In this sense, women's understanding of the 10-year risk is the basis of informed decision-making in this context [28]. Therefore, the main goal of our study was to compare women’s subjective 10-year risk estimates of developing breast or ovarian cancer to the objective risks after guideline-based counseling. Depending on whether counselees have already been diagnosed with cancer or undergone prophylactic surgery, the preconditions for genetic counseling vary. The fact that the conditions for genetic counseling vary depending on whether counselees have already had a diagnosis of cancer or prophylactic surgery was taken into account. In addition, we sought to identify clinical and psychosocial predictors of risk accuracy. 2. Methods The cross-sectional, observational, mono-center study was conducted at the Center of Hereditary Breast and Ovarian Cancer at Charité – Universitätsmedizin Berlin, which belongs to the German Consortium of Hereditary Breast and Ovarian Cancer (GC-HBOC). The
GC-HBOC comprises 18 university centers from all over Germany, which offer counseling, pedigree analysis, risk calculation, genetic testing, surveillance programs and prophylactic surgery in accordance with GC-HBOC guidelines. Ethical approval was obtained from Charité – Universitätsmedizin Berlin (EA1/222/15). 2.1. Participants and instruments Initial inclusion criteria for study participants were women between the age of 18 and 70 years and either a statistical high risk for breast and ovarian cancer or detection of a pathogenic mutation in a gene that predisposes for breast and ovarian cancer, not longer than two years ago. Initial exclusion criteria were insufficient German language skills. In order to keep the heterogeneity at an acceptable level, we only included women with a pathogenic BRCA1- or BRCA2- mutation for data analysis in this study. Women were excluded from data analysis when their objective risk could not be calculated using the Breast and Ovarian Analysis of Disease Incidence and Carrier Estimation Algorithm (BOADICEA2 ) program (e.g. women with bilateral breast cancer or ovarian cancer) and when questionnaires missed variables. Women with a diagnosis of breast cancer were only asked to estimate their contralateral breast cancer risk to avoid misunderstandings concerning unilateral relapse risks (see Fig. 1). After initial counseling, the genetic testing process at our center consists of three steps as soon as counselees opt for genetic analysis: (a) genetic testing; (b) communication of test results; (c) optional genetic counseling concerning age-specific risks for breast and/or ovarian cancer. This additional counseling session focuses on 10-year-risks combined with personalized clinical risk management options following the guidelines for counseling (GC-HBOC and the Association of the Scientific Medical Societies in Germany (AWMF)). Risk perception is not explicitly addressed in this setting. Between August 2015 and April 2017, 300 women with a positive mutation test result or who had a statistical high breast cancer risk and decided to get additional optional counseling were screened for our initial inclusion criteria (Fig. 1). 250 women, who met the initial inclusion criteria, were invited to participate in the study. Of these, 207 women provided written consent and were mailed questionnaires subsequently. Of 127 women who returned the questionnaire, 39 were excluded from data analysis (see Fig. 1 and exclusion criteria above for more information). In total, 88 counselees who had obtained a positive BRCA1/2mutation test result on average 14.9 12.6 months ago were included in the analysis. From this group those women were excluded who already had a (secondary) prophylactic mastectomy (26) and those who had not reported subjective risks (3). This allowed us to present objective and subjective breast cancer risk assessments for 59 women. To illustrate the objective and subjective ovarian cancer risks, we had to exclude 34 women who had already undergone prophylactic ovarian removal and 2 women who had not reported any subjective risks. Thus, the group for which we could report objective and subjective ovarian cancer risks was 52 women. Therefore, the difference between objective and subjective risk estimates could be calculated for 59 counselees in terms of breast
2 The risk calculation tool BOADICEA is a web-based risk prediction model for hereditary breast and ovarian cancer used to compute lifetime and age-specific breast and ovarian cancer risks for BRCA1/2-mutation carriers [30][31],]. It is based on complex segregation analyses of families with multiple affected individuals carrying BRCA1/2-mutations [[32]]. The usage of this tool has been incorporated in the guidelines of several countries for the management of hereditary breast and ovarian cancer [[33][34],]. All participants were informed that their risks as calculated from BOADICEA were used for research purposes only.
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931
1927
Fig. 1. Flowchart of study participants. *reasons for exclusion: BOADICEA could not be calculated (bilateral breast cancer (4), history of ovarian cancer (7), missing variables (6), low-risk-mutation (9) or no mutation (13).
cancer risk and for 52 counselees for ovarian cancer risk (Fig. 1). Counselees were assigned to several subgroups according to clinical parameters: whether or not they had already undergone prophylactic surgery, whether or not they had a diagnosis of cancer, and the type of the diagnosed cancer (ovarian or breast cancer, respectively). This resulted in eight subgroups, for which we depict the 10-year-risks and subjective risk estimates separately (see Supplement). 3. Procedures Participants’ objective risks of developing breast and ovarian cancer were calculated using the risk calculation tool BOADICEA described above [30,31], which is available online. Clinical data that determine disease risks were taken from case report forms and entered in BOADICEA. Sociodemographic variables (age, level of education, parenthood) and cancer worry were assessed with self-report questionnaires. Four items of the cancer worry scale were used to assess cancer worry. The sum score range was 4 to 16, with higher values indicating higher levels of cancer worry [35]. Women were asked to estimate their subjective numeric risks of developing breast or ovarian cancer within the next 10 years: “Imagine 100 women in exactly your situation. How many of these women do you think will (again) develop breast cancer within the next 10 years?”, and, accordingly, “How many of these women do you think will develop ovarian cancer within the next 10 years?”. Women with breast cancer were only asked to indicate their risk estimation for the contralateral breast to avoid misunderstandings concerning unilateral relapse risks. In addition, women were asked how well informed they felt about the gene test result after the consultation. The subjective risk was assessed with simple frequencies. Compared with percentage formats, simple frequencies can be more easily understood and interpreted [36], allowing for more accurate lifetime risk estimates [37]. 3.1. Data analysis We report means and standard deviations for metrically and ordinally scaled variables for the entire sample and for women with and without breast cancer separately. For categorical variables, we display frequencies and percentages. Group
comparisons for continuous variables were carried out using Student’s t-test to determine if the two patient groups differed significantly. To indicate the accuracy of risk estimation, we calculated a difference score between the objective disease risks calculated with BOADICEA and women’s estimated probabilities of developing breast and ovarian cancer. Participants estimates were scored as correct with an error tolerance of 5 (5% of scale). Accordingly, three groups were considered: underestimation, correct estimation, and overestimation. We tested for the differences between mean objective and subjective risk estimates with the t-test for dependent samples. Additionally, the Bartlett’s test for homogeneity of variances was used to test if the variances of objective and subjective risk estimates differed. Associations between clinical and psychosocial variables on the one hand, and the difference score on the other hand were determined with Pearson’s r. An alpha level of p < .05 was considered statistically significant. 4. Results Of all women who returned the questionnaires (N = 127), 89.8% felt well informed about the genetic test result. Risk calculations for either the risk of breast cancer or ovarian cancer, or both, were possible for 88 women (Fig. 1). Women were on average 42.1 10.6 years old (range 19–67 years). Of these, 41 study participants had a diagnosis of breast cancer. These women reported higher levels of cancer worry and were on average 4.5 years older than the 47 women without breast cancer. No woman had developed metastases at the time of the study. Women without a former diagnosis of breast cancer had more often undergone prophylactic surgery (e.g. risk-reducing salpingo-oophorectomy or secondary prophylactic mastectomy). The sample characteristics are shown in Table 1. The objective 10-year risk for breast cancer could be calculated for 62 women (not possible for n = 22 women with mastectomy or bilateral breast cancer). Of these, 59 provided subjective risk estimates for breast cancer. The mean objective risks differed significantly from the subjective estimates (Mobjective risk = 20.44 9.05 vs. Msubjective risk = 33.88 25.52; t(58) = 4.12, p < .001) with an effect of Cohen’s d = 0.70. Bartlett’s test for homogeneity of
1928
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931
Table 1 Sample characteristics differentiated by breast cancer status. No breast cancer (n = 47)
Diagnosis of breast p cancer (n = 41)
Characteristic
Total sample (N = 88)
Age, yrs, mean (SD) Higher education, n (%) Living with partner Parenthood, n (%) Prophylactic surgery, n (%) No prophylactic surgery Prophylactic mastectomy Prophylactic salpingoophorectomy Mastectomy and salpingoophorectomy Time since test result in months (SD) Cancer worry, mean (SD)
42.10 (10.6) 44.68 (9.23) 57 (64.8) 22 (57.9)
40.14 (11.30) 35 (70.0)
.047 .170
71 (80.7) 61 (69.3)
28 (73.7) 28 (73.7)
43 (86.0) 33 (66.0)
.120 .296 .001
39 (44.3)
9 (23.7)
30 (60.0)
14 (15.9)
9 (23.7)
5 (10.0)
23 (26.1)
10 (26.3)
13 (26.3)
12 (13.6)
10 (26.3)
2 (4.0)
14.86 16.12 (11.88) (12.63) 10.47 (3.29) 11.35 (2.90)
risk estimation for developing breast or ovarian cancer did not associate with age, education, parenthood, prophylactic surgery, or cancer worry. Only the desire to undergo prophylactic mastectomy was related to the overestimation of breast cancer risk (r = .34, p = .010) (Table 2). We performed additional analyses with subgroups that were built according to the presence of a cancer diagnosis and/or prophylactic surgery. These analyses indicated a tendency towards risk overestimation across the subgroups (Figure S1). 5. Discussion and conclusion 5.1. Discussion
13.95 (13.16)
.425
9.80 (3.43)
.029
variances indicated a significant difference in the variances of the objective risks and the subjective risk estimation (X = 14.89; p < .001). The respective distributions are depicted in Fig. 3a. Permitting a tolerance of +/-5% for an accurate risk estimation, 14 (23.7%) of the women underestimated their breast cancer risk, 12 (20.3%) provided correct numbers, and 33 (55.9%) overestimated their risk (Fig. 2). The objective ten-year risk for ovarian cancer was calculated for 54 women (not possible for n = 34 women with ovarian cancer or prophylactic salpingo-oophorectomy). Of these, 52 provided subjective risk estimates for ovarian cancer. Again, the mean objective risk differed from the subjective risk estimation (Mobjective risk = 3.25 3.12 vs. Msubjective risk = 18.45 18.62; t (51) = 5.80, p < .001) with an effect of d = 1.14, as did the variances of the objective risks and the subjective risk estimates (X = 36.09; p < .001) (Fig. 3b). Only 3 (5.6%) of the women underestimated their ovarian cancer risk, whereas 21 (38.9%) provided correct numbers, and 28 (51.9%) overestimated their risk. The accuracy of
Fig. 2. Proportion of women underestimating and overestimation their risk of breast and ovarian cancer (%). A tolerance of (+/5%) was permitted for the correct estimation.
The objective of this study was to evaluate the accuracy of risk estimation in BRCA1/2-mutation carriers after genetic counseling. To our knowledge, it is the first study comparing women’s subjective risk estimates with the objective 10-year-risk. The key findings of our study are that, even though women generally felt well informed, they markedly overestimated their breast and ovarian cancer risks after guideline-based genetic counseling. Women with past cancer diseases or prophylactic surgeries did not provide more accurate risk estimates. In case of ovarian cancer risk, the magnitude of overestimation is particularly striking. Several reasons can be assumed for the stronger overestimation of the ovarian cancer risk as compared to the breast cancer risk: (1) During counseling, women must be informed about the lack of screening options and still unsatisfying therapy options for ovarian cancer, which could intensify risk perception and influence risk understanding; (2) women may confuse numbers for ovarian cancer and breast cancer risks, which could lead to higher estimates for ovarian cancer risk; (3) ovarian cancer is a rare disease, and events with a lower probability are more likely to be overestimated [7]; (4) finally, there may be people who just guess, which is likely to result in a number somewhere in the middle of the scale (“can be/ cannot be” or “50:5000 ). As a consequence, the discrepancy between the subjective and the objective risk due to pure guessing could be larger in cases with a low objective risk. Our findings further indicate that the women’s subjective 10-year risk estimates are close to their actual objective lifetimerisks. The differentiation of age-specific and lifetime risks is a challenge. Possibly, these two risks are confused, which is not surprising as predominantly lifetime risks are communicated in the media. Because 2-, 5- and 10-year risks are substantially more important for risk-management [26], special attention should be paid to explaining the distinction between lifetime risk and agespecific risks in the consultation. Surprisingly, in our study risk overestimation was independent from mutation type, level of cancer worry, education, age and parenthood. The fact that no clear correlations between psychosocial or clinical factors and overestimation could be identified in this population may imply that the consultant cannot rely on, for instance, less anxious or more educated women to accurately assess their risks. Thus, a similar effort is required to make the risks transparent and easy to understand for all women. It should be examined which graphics, tabular formats [38] and verbal explanations are particularly suitable for supplementing the numerical information to support counseling [39]. In addition, 23.7% and 5.6% of the responders underestimated their objective breast and ovarian cancer risks. Again, we could not find any correlation as to prophylactic surgeries or age. Future analyses should address this phenomenon in detail to make sure that this underestimation will not impair compliance to intensified surveillance programs. This group of women should also be kept in mind in the design of counseling-supporting tools.
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931
1929
Fig. 3. Objective risks and estimates for the 10-year risk of breast (3a) and ovarian (3b) cancer.
Table 2 Correlations between sociodemographic, clinical, and psychosocial variables and difference scores between objective risks and subjective estimates. Characteristic
Correlation with the breast cancer risk difference score (n = 59)
p
Correlation with the ovarian cancer risk difference score (n = 52)
p
Age, yrs, mean (SD) Higher education, n (%) Living with partner Parenthood, n (%) Prophylactic surgery, n (%) Time since test result in months (SD) Cancer worry, mean (SD)
.03
.810
.01
.967
.17
.194
.23
.101
.13
.327 .06
.659
.17
.192
.05
.751
.25
.055 .03
.816
.19
.167
.19
.182
.10
.444 .13
.353
5.2. Practice implications Numbers are often presented in a way that is difficult to understand and patients are challenged by extracting the relevant numbers and interpreting them adequately [40,41]. Physicians should make a great effort to present the relevant numbers in the most transparent way possible in order to enable informed decisions when managing cancer risks. The complexity and amount of information requires new ways in risk consulting. 5.3. Study strengths and limitations One strength of our study is the inclusion and differentiated evaluation regarding clinical characteristics. We postulate that extensive analysis and survey of the actual individual situation of each counselee is a prerequisite of successful counseling. Limitations include that the cross-sectional design does not allow analysis of changes in risk assessment over time. This might be interesting because risk assessment may vary due to changes in family situation or through risk management measures. Second,
we did not measure health literacy and therefore cannot say what proportion of overestimation is due to limited abilities to understand health information. However, education, which was associated with health literacy in other studies [13], was not correlated with inaccurate estimates in our study. Third, although this is one of the largest samples on this topic, the power was not sufficient to confirm variations in risk estimation accuracy across the different subgroups with past diseases states and prophylactic surgeries. In addition, due to the ‘right-not-to-know’, which is part of the German Genetic Diagnostics Act (“Gendiagnostikgesetz”), there is a clear bias in our study population towards more interested counselees. Due to the current limitations of the calculation tool, 10-year risks for ovarian cancer could not be calculated for patients who had bilateral breast cancer before. Neither, 10-year risks for breast cancer could be calculated for patients with ovarian cancer. Therefore, even if we emphasized the differentiated depiction of subgroups these two important group of counselees could not be considered. Finally, we did not focus on counselees’ specific personal family history with breast and ovarian cancer with regard to their overestimation. However, this individual familial background seems to be important in the complex process of risk assessment and should be investigated in further studies. 5.4. Conclusion and research recommendations Women with a BRCA1/2-gene mutation strongly overestimate their 10-year risk of breast cancer. Overestimation was even more pronounced in the case of ovarian cancer. This effect was not restricted to women with a particular psychosocial background, such as lower education, a prior diagnosis of cancer, or children. Thus, counselors must be prepared to prevent overestimation in all of their counselees. On the other hand, given large individual differences in objective risks based on clinical parameters, counselors must also adapt their advice to meet the individual requirements and situation of each counselee. In other words, counseling must be standardized and personalized at the same time. Future studies should investigate which counseling-supporting tools are particularly suitable to achieve the right balance between standardization and personalization in counseling to improve risk understanding and as a prerequisite for preventing over-treatment.
1930
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931
Competing interests The authors declare that they have no conflict of interest.
[7] [8]
Authors’ contributions DS, FR, MF and FK designed the study, carried out the research, analyzed the data and wrote the manuscript. LB and HB substantially contributed to the design of the study and participated in the acquisition and interpretation of the data. All authors read and approved the final manuscript.1 Human studies and informed consent All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000 (5). Informed consent was obtained from all patients for being included in the study. Funding The study was supported by an unrestricted grant from the Berliner Krebsgesellschaft e.V. [grant number KEFF201607]. The funding source had no role in design, conduct of the study, interpretation of the data, or publication. We confirm all patient/personal identifiers have been removed or disguised so the patient/person(s) described are not identifiable and cannot be identified through the details of the story. Acknowledgements We wish to thank the study participants, who provided their time and showed great interest in this study. We would also like to thank Nanette Kalmbach and Marina Kniehase for their support in data acquisition.
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.pec.2019.05.007. References [1] K.B. Kuchenbaecker, J.L. Hopper, D.R. Barnes, K.-A. Phillips, T.M. Mooij, M.-J. Roos-Blom, S. Jervis, F.E. van Leeuwen, R.L. Milne, N. Andrieu, D.E. Goldgar, M.B. Terry, M.A. Rookus, D.F. Easton, A.C. Antoniou, BRCA1 and BRCA2 Cohort Consortium, L. McGuffog, D.G. Evans, D. Barrowdale, D. Frost, J. Adlard, K.-R. Ong, L. Izatt, M. Tischkowitz, R. Eeles, R. Davidson, S. Hodgson, S. Ellis, C. Nogues, C. Lasset, D. Stoppa-Lyonnet, J.-P. Fricker, L. Faivre, P. Berthet, M.J. Hooning, L.E. van der Kolk, C.M. Kets, M.A. Adank, E.M. John, W.K. Chung, I.L. Andrulis, M. Southey, M.B. Daly, S.S. Buys, A. Osorio, C. Engel, K. Kast, R.K. Schmutzler, T. Caldes, A. Jakubowska, J. Simard, M.L. Friedlander, S.-A. McLachlan, E. Machackova, L. Foretova, Y.Y. Tan, C.F. Singer, E. Olah, A.-M. Gerdes, B. Arver, H. Olsson, Risks of breast, ovarian, and contralateral breast cancer for BRCA1 and BRCA2 mutation carriers, JAMA 317 (2017) 2402–2416, doi:http://dx.doi.org/10.1001/jama.2017.7112. [2] D. Stacey, R. Samant, C. Bennett, Decision making in oncology: a review of patient decision aids to support patient participation, CA. Cancer J. Clin. 58 (2008) 293–304, doi:http://dx.doi.org/10.3322/CA.2008.0006. [3] F.M. Walter, J. Emery, Perceptions of family history across common diseases: a qualitative study in primary care, Fam. Pract. 23 (2006) 472–480, doi:http://dx. doi.org/10.1093/fampra/cml006. [4] A. Werner-Lin, Building the cancer family: family planning in the context of inherited breast and ovarian cancer risk, J. Soc. Soc. Work Res. 1 (2010) 14–27, doi:http://dx.doi.org/10.5243/jsswr.2010.3. [5] D. Leblond, A. Brédart, S. Dolbeault, A. De Pauw, D. Stoppa Lyonnet, C. Flahault, S. Sultan, [Cognitive, emotional and behavioral impact of an uncertain outcome after study of BRCA1/2: review of the literature], Bull. Cancer (Paris). 98 (2011) 184–198, doi:http://dx.doi.org/10.1684/bdc.2011.1309. [6] D.P. French, J. Southworth, A. Howell, M. Harvie, P. Stavrinos, D. Watterson, S. Sampson, D.G. Evans, L.S. Donnelly, Psychological impact of providing women
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
with personalised 10-year breast cancer risk estimates, Br. J. Cancer. 118 (2018) 1648–1657, doi:http://dx.doi.org/10.1038/s41416-018-0069-y. W.M.P. Klein, M.E. Stefanek, Cancer risk elicitation and communication: lessons from the psychology of risk perception, CA. Cancer J. Clin. 57 (2007) 147–167. C.J. Frost, V. Venne, D. Cunningham, R. Gerritsen-McKane, Decision making with uncertain information: learning from women in a high risk breast cancer clinic, J. Genet. Couns. 13 (2004) 221–236, doi:http://dx.doi.org/10.1023/B: JOGC.0000027958.02383.a9. D.G. Evans, J. Wisely, T. Clancy, F. Lalloo, M. Wilson, R. Johnson, J. Duncan, L. Barr, A. Gandhi, A. Howell, Longer term effects of the angelina jolie effect: increased risk-reducing mastectomy rates in BRCA carriers and other high-risk women, Breast Cancer Res. BCR. 17 (2015) 143, doi:http://dx.doi.org/10.1186/ s13058-015-0650-8. N.E. Carbine, L. Lostumbo, J. Wallace, H. Ko, Risk-reducing mastectomy for the prevention of primary breast cancer, Cochrane Libr., John Wiley & Sons, Ltd, 2018, doi:http://dx.doi.org/10.1002/14651858.CD002748.pub4. A.W. Kurian, D.Y. Lichtensztajn, T.H.M. Keegan, D.O. Nelson, C.A. Clarke, S.L. Gomez, Use of and mortality after bilateral mastectomy compared with other surgical treatments for breast cancer in California, 1998-2011, JAMA 312 (2014) 902–914, doi:http://dx.doi.org/10.1001/jama.2014.10707. K. Rhiem, K. Pfeifer, R.K. Schmutzler, M. Kiechle, risk-reducing surgery in women at risk for familial breast or ovarian cancer, Geburtshilfe Frauenheilkd. 72 (2012) 833–839, doi:http://dx.doi.org/10.1055/s-0032-1315362. E.J. Rutherford, J. Kelly, E.A. Lehane, V. Livingstone, B. Cotter, A. Butt, M.J. O’Sullivan, F.O. Connell, H.P. Redmond, M.A. Corrigan, Health literacy and the perception of risk in a breast cancer family history clinic, Surg. J. R. Coll. Surg. Edinb. Irel. 16 (2018) 82–88, doi:http://dx.doi.org/10.1016/j.surge.2016.06.003. S. Schott, L. Vetter, M. Keller, T. Bruckner, M. Golatta, S. Eismann, N. Dikow, C. Evers, C. Sohn, J. Heil, Women at familial risk of breast cancer electing for prophylactic mastectomy: frequencies, procedures, and decision-making characteristics, Arch. Gynecol. Obstet. 295 (2017) 1451–1458, doi:http://dx. doi.org/10.1007/s00404-017-4376-y. R. Glassey, M. O’Connor, A. Ives, C. Saunders, k Con Fab Investigators, S. O’Sullivan, S.J. Hardcastle, Heightened perception of breast cancer risk in young women at risk of familial breast cancer, Fam. Cancer 17 (2018) 15–22, doi:http://dx.doi.org/10.1007/s10689-017-0001-2. S. van Dijk, W. Otten, M.W. Zoeteweij, D.R.M. Timmermans, C.J. van Asperen, M.H. Breuning, Ra.E.M. Tollenaar, J. Kievit, Genetic counselling and the intention to undergo prophylactic mastectomy: effects of a breast cancer risk assessment, Br. J. Cancer. 88 (2003) 1675–1681, doi:http://dx.doi.org/10.1038/sj.bjc.6600988. D.J. Bowen, C.M. Alfano, B.A. McGregor, M.R. Andersen, The relationship between perceived risk, affect, and health behaviors, Cancer Detect. Prev. 28 (2004) 409–417, doi:http://dx.doi.org/10.1016/j.cdp.2004.08.008. M. Broadstock, S. Michie, J. Gray, J. Mackay, T.M. Marteau, The psychological consequences of offering mutation searching in the family for those at risk of hereditary breast and ovarian cancer–a pilot study, Psychooncology. 9 (2000) 537–548. G. Cicero, R. De Luca, P. Dorangricchia, G. Lo Coco, C. Guarnaccia, D. Fanale, V. Calò, A. Russo, Risk perception and psychological distress in genetic counselling for hereditary breast and/or ovarian cancer, J. Genet. Couns. 26 (2017) 999–1007, doi:http://dx.doi.org/10.1007/s10897-017-0072-0. D.G.R. Evans, V. Blair, R. Greenhalgh, P. Hopwood, A. Howell, The impact of genetic counselling on risk perception in women with a family history of breast cancer, Br. J. Cancer. 70 (1994) 934, doi:http://dx.doi.org/10.1038/ bjc.1994.423. P. Hopwood, A. Shenton, F. Lalloo, D.G.R. Evans, A. Howell, Risk perception and cancer worry: an exploratory study of the impact of genetic risk counselling in women with a family history of breast cancer, J. Med. Genet. 38 (2001), doi: http://dx.doi.org/10.1136/jmg.38.2.139 139–139. E.A. Lobb, P.N. Butow, B. Meiser, A. Barratt, C. Gaff, M.A. Young, J. Kirk, M. Gattas, M. Gleeson, K. Tucker, Women’s preferences and consultants’ communication of risk in consultations about familial breast cancer: impact on patient outcomes, J. Med. Genet. 40 (2003), doi:http://dx.doi.org/10.1136/jmg.40.5. e56 e56–e56. Y. Rothemund, S. Paepke, H. Flor, Perception of risk, anxiety, and health behaviors in women at high risk for breast cancer, Int. J. Behav. Med. 8 (2001) 230–239, doi:http://dx.doi.org/10.1207/S15327558IJBM0803_5. A. Bish, S. Sutton, C. Jacobs, S. Levene, A. Ramirez, S. Hodgson, Changes in psychological distress after cancer genetic counselling: a comparison of affected and unaffected women, Br. J. Cancer. 86 (2002) 43–50, doi:http://dx. doi.org/10.1038/sj.bjc.6600030. K. Kelly, H. Leventhal, M. Andrykowski, D. Toppmeyer, J. Much, J. Dermody, M. Marvin, J. Baran, M. Schwalb, Using the common sense model to understand perceived cancer risk in individuals testing for BRCA1/2 mutations, Psychooncology. 14 (2005) 34–48, doi:http://dx.doi.org/10.1002/pon.805. R. Schmutzler, K. Rhiem, Beratung junger frauen mit hereditärer belastung für brust- und eierstockkrebs, Forum (Genova). 32 (2017) 37–41, doi:http://dx.doi. org/10.1007/s12312-017-0212-3. D. Keohane, E. Lehane, E. Rutherford, V. Livingstone, L. Kelly, S. Kaimkhani, F. O’Connell, H.P. Redmond, M.A. Corrigan, Can an educational application increase risk perception accuracy amongst patients attending a high-risk breast cancer clinic? Breast edinb, Scotl. 32 (2017) 192–198, doi:http://dx.doi. org/10.1016/j.breast.2017.02.009. A. Abbott, N. Rueth, S. Pappas-Varco, K. Kuntz, E. Kerr, T. Tuttle, Perceptions of contralateral breast cancer: an overestimation of risk, Ann. Surg. Oncol. 18 (2011) 3129–3136, doi:http://dx.doi.org/10.1245/s10434-011-1914-x.
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931 [29] A. Albada, M.G.E.M. Ausems, J.M. Bensing, S. van Dulmen, Tailored information about cancer risk and screening: a systematic review, Patient Educ. Couns. 77 (2009) 155–171, doi:http://dx.doi.org/10.1016/j.pec.2009.03.005. [30] A.C. Antoniou, A.P. Cunningham, J. Peto, D.G. Evans, F. Lalloo, S.A. Narod, H.A. Risch, J.E. Eyfjord, J.L. Hopper, M.C. Southey, H. Olsson, O. Johannsson, A. Borg, B. Pasini, B. Passini, P. Radice, S. Manoukian, D.M. Eccles, N. Tang, E. Olah, H. Anton-Culver, E. Warner, J. Lubinski, J. Gronwald, B. Gorski, L. Tryggvadottir, K. Syrjakoski, O.-P. Kallioniemi, H. Eerola, H. Nevanlinna, P.D.P. Pharoah, D.F. Easton, The BOADICEA model of genetic susceptibility to breast and ovarian cancers: updates and extensions, Br. J. Cancer. 98 (2008) 1457–1466, doi: http://dx.doi.org/10.1038/sj.bjc.6604305. [31] A.C. Antoniou, P.P.D. Pharoah, P. Smith, D.F. Easton, The BOADICEA model of genetic susceptibility to breast and ovarian cancer, Br. J. Cancer. 91 (2004) 1580–1590, doi:http://dx.doi.org/10.1038/sj.bjc.6602175. [32] A.J. Lee, A.P. Cunningham, K.B. Kuchenbaecker, N. Mavaddat, D.F. Easton, A.C. Antoniou, D. of P.H. and P.C. The consortium of investigators of modifiers of BRCA1/2Centre for cancer genetic epidemiology university of Cambridge, strangeways research laboratory, Cambridge CB1 8RN, UK, D. P.H. and P.C. The breast cancer association ConsortiumCentre for cancer genetic epidemiology university of Cambridge, strangeways research laboratory, Cambridge CB1 8RN, UK, BOADICEA breast cancer risk prediction model: updates to cancer incidences, tumour pathology and web interface, Br. J. Cancer 110 (2014) 535– 545, doi:http://dx.doi.org/10.1038/bjc.2013.730. [33] B.D. Riley, J.O. Culver, C. Skrzynia, L.A. Senter, J.A. Peters, J.W. Costalas, F. CallifDaley, S.C. Grumet, K.S. Hunt, R.S. Nagy, W.C. McKinnon, N.M. Petrucelli, R.L. Bennett, A.M. Trepanier, Essential elements of genetic cancer risk assessment, counseling, and testing: updated recommendations of the national society of genetic counselors, J. Genet. Couns. 21 (2012) 151–161, doi:http://dx.doi.org/ 10.1007/s10897-011-9462-x.
1931
[34] R.A. Smith, V. Cokkinides, O.W. Brawley, cancer screening in the United States, 2012: a review of current American cancer society guidelines and current issues in cancer screening, CA. Cancer J. Clin. 62 (2012) 129–142, doi:http://dx. doi.org/10.3322/caac.20143. [35] C. Lerman, B. Trock, B.K. Rimer, C. Jepson, D. Brody, A. Boyce, psychological side effects of breast cancer screening, health psychol, Off. J. Div. Health Psychol. Am. Psychol. Assoc. 10 (1991) 259–267. [36] G. Gigerenzer, U. Hoffrage, How to improve bayesian reasoning without instruction: frequency formats, Psychol. Rev. 102 (1995) 684–704, doi:http:// dx.doi.org/10.1037/0033-295X.102.4.684. [37] M.M. Schapira, S.L. Davids, T.L. McAuliffe, A.B. Nattinger, Agreement between scales in the measurement of breast cancer risk perceptions, Risk Anal. Off. Publ. Soc. Risk Anal. 24 (2004) 665–673, doi:http://dx.doi.org/10.1111/j.0272-4332.2004.00466.x. [38] M. McDowell, F.G. Rebitschek, G. Gigerenzer, O. Wegwarth, A simple tool for communicating the benefits and harms of health interventions: a guide for creating a fact box, MDM Policy Pract. 1 (2016)2381468316665365, doi:http:// dx.doi.org/10.1177/2381468316665365. [39] N.T. Brewer, J.P. Tzeng, S.E. Lillie, A.S. Edwards, J.M. Peppercorn, B.K. Rimer, Health literacy and cancer risk perception: implications for genomic risk communication, Med. Decis. Mak. Int. J. Soc. Med. Decis. Mak. 29 (2009) 157– 166, doi:http://dx.doi.org/10.1177/0272989X08327111. [40] O. Wegwarth, W. Gaissmaier, G. Gigerenzer, Deceiving numbers: survival rates and their impact on doctors’ risk communication, Med. Decis. Mak. Int. J. Soc. Med. Decis. Mak. 31 (2011) 386–394, doi:http://dx.doi.org/10.1177/ 0272989X10391469. [41] J.C. Tilburt, K.M. James, P.S. Sinicrope, D.T. Eton, B.A. Costello, J. Carey, M.A. Lane, S.L. Ehlers, P.J. Erwin, K.E. Nowakowski, M.H. Murad, Factors influencing cancer risk perception in high risk populations: a systematic review, Hered. Cancer Clin. Pract. 9 (2011) 2, doi:http://dx.doi.org/10.1186/1897-4287-9-2.
Contents lists available at ScienceDirect
Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou
Accuracy in risk understanding among BRCA1/2-mutation carriers Dorothee Speisera,1,* , Felix G. Rebitschekb,1, Markus A. Feufelb,c, Hannah Brandd , Laura Beschd , Friederike Kendeld a
Department of Gynecology with Breast Center, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany Harding Center for Risk Literacy, Max Planck Institute for Human Development, Berlin, Lentzeallee 94, 14195 Berlin, Germany Division of Ergonomics, Department of Psychology and Ergonomics, Technische Universität Berlin, Marchstr. 23, 10587 Berlin, Germany d Institute of Medical Psychology, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 18 October 2018 Received in revised form 7 March 2019 Accepted 4 May 2019
Objective: BRCA1/2-mutation carriers are at high risk of developing cancer. Since they must weigh clinical recommendations and decide on risk-reducing measures, the correct understanding of their 10-year cancer risks is essential. This study focused on the accuracy of women’s subjective estimates of developing breast and ovarian cancer within ten years as prerequisite to reduce unnecessary prevention. Methods: 59 and 52 BRCA1/2-mutation carriers provided their individual risks of developing breast or ovarian cancer in the next 10 years, along with self-reported sociodemographic and psychosocial variables. Women’s risk estimates were compared with their objective cancer risks that had been communicated before. Results: 22.6% of counselees under- and 53.2% of the counselees overestimated their 10-year risk of developing breast cancer. As for ovarian cancer, 5.6% under- whereas 51.9% overestimated their risk. Neither demographic factors such as education, parenthood and age, nor a prior diagnosis of breast cancer or prophylactic surgery accounted for these variations in risk accuracy. Conclusion: Currently, risk communication during genetic counseling does not guarantee accurate risk estimation in BRCA-mutation carriers. Practice Implications: Counselors must be prepared to prevent overestimation. Counselees’ risk estimates need to be assessed and corrected to enable informed decision-making and reduce risks of unnecessary preventive efforts. © 2019 Elsevier B.V. All rights reserved.
Keywords: BRCA1/2 gene mutation Genetic counseling Subjective risk estimation Risk perception Risk accuracy
1. Introduction Women carrying a deleterious mutation in BRCA1 or BRCA2 have a high lifetime risk of developing breast or ovarian cancer. The lifetime risks for breast cancer in BRCA1-mutation carriers range from 65 to 79%, and from 61 to 77% for BRCA2-mutation carriers, respectively. Ovarian cancer risks for BRCA1-mutation carriers range from 36 to 53%, for BRCA2-mutation carriers from 11 to 25% as lifetime risks [1]. For mutation carriers, the question of risk management, that is, early detection and/or preventive measures, arises as soon as the mutation status is revealed. A precondition for making informed decisions regarding risk management strategies
* Corresponding author at: Charité – Universitätsmedizin Berlin, Department of Gynecology with Breast Center, Campus Charité Mitte, Charitéplatz 1, 10117 Berlin. E-mail addresses: [email protected] (D. Speiser), [email protected] (F.G. Rebitschek), [email protected] (M.A. Feufel), [email protected] (F. Kendel). 1 DS and FGR are shared first authors. https://doi.org/10.1016/j.pec.2019.05.007 0738-3991/© 2019 Elsevier B.V. All rights reserved.
is that affected women understand their individual disease risks [2]. During counseling on risk management, women’s objective disease risks are presented as lifetime risks together with several age-specific cumulative risks. As the objective risk depends on clinical parameters such as age, affected genes, pre-existing cancer, and preventive measures already taken, the complex interplay of these parameters is likely to influence also the individual’s risk perception and understanding. Furthermore, risk perception is associated with psychosocial variables such as personal experience [3,4], education, and cultural differences [5]. Women with a higher risk have higher levels of cancer worry [6]. Also, anxiety seems to increase the perception of vulnerability and is thus likely to inflate risk perceptions [7]. Last but not least, the counseling process itself, in which large amounts of information need to be communicated, could lead to inaccurate assessments of risk [8]. Over the last years, the number of prophylactic surgeries, such as bilateral risk reducing mastectomies (BRRM), has increased substantially. In part, this increase can be traced back to the enormous public attention the actress Angelina Jolie attracted in
1926
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931
2013 when she made public her BRCA1/2-gene mutation and her decision for a prophylactic mastectomy [9]. Since then, BRRMs have also increased in women with a negative BRCA1/2-test result, although prophylactic surgery in this population has not been shown to improve overall survival [10–12]. One reason for the increased BRRM-rate among BRCA-mutation carriers and noncarriers could be a lack of risk understanding in counselees at tumor risk clinics [13]. Many women may opt for prophylactic mastectomy despite an unclear survival benefit because they overestimate their disease risk [14]. In a study with women undergoing prophylactic surgery, risk overestimation and confusion about risks prevailed even after surgery and were accompanied by increased anxiety [15]. These findings underline the clinical relevance of accurate risk perception for informed risk management decisions. Studies on risk estimation thus far predominantly focused on perceptions of lifetime risks [16–23] or on how women perceived their own risk in relation to the risk in the general female population (comparative risk estimates; [16,24,25]). However, it is becoming increasingly clear that, clinically, the most important time parameters on which relevant screening and prophylactic recommendations should be based are 10-year risks [26]. Only two studies [13,27] assessed the accuracy of subjective risk estimates in comparison with objective 10-year-risks in women with a family history of breast cancer. Abbott et al. [28] compared different clinical subgroups with respect to their subjective 10-year-risk estimates but did not compare these risk estimates to their objective risk. Both studies explicitly excluded women with known BRCA1/2- mutations and focused on women with a known family history of breast or ovarian cancer instead. Although risk perception was measured differently in the above-mentioned studies (either with verbal labels or numerically), it can be concluded that counselees overestimated rather than underestimated their breast cancer risks. Assessing the individual’s subjective estimation of risk is challenging. Generally, the constructs “risk understanding” and “risk perception” need to be distinguished. Whereas risk understanding depends on an individual’s knowledge and is assessed with numerical estimates (e.g. 3 out of 100), risk perception also includes subjective vulnerability to health threats and is usually assessed with verbal labels (description of a risk as “low” or “high”) [7]. Most studies, so far, have focused on risk perception. Guideline-based counseling requires that women are informed about their individual cancer risk which is composed of various factors, including mutation type, familial and personal cancer history, and age [29]. From a clinical perspective, the 10-year risk of developing breast or ovarian cancer usually guides the counseling and recommendations. In this sense, women's understanding of the 10-year risk is the basis of informed decision-making in this context [28]. Therefore, the main goal of our study was to compare women’s subjective 10-year risk estimates of developing breast or ovarian cancer to the objective risks after guideline-based counseling. Depending on whether counselees have already been diagnosed with cancer or undergone prophylactic surgery, the preconditions for genetic counseling vary. The fact that the conditions for genetic counseling vary depending on whether counselees have already had a diagnosis of cancer or prophylactic surgery was taken into account. In addition, we sought to identify clinical and psychosocial predictors of risk accuracy. 2. Methods The cross-sectional, observational, mono-center study was conducted at the Center of Hereditary Breast and Ovarian Cancer at Charité – Universitätsmedizin Berlin, which belongs to the German Consortium of Hereditary Breast and Ovarian Cancer (GC-HBOC). The
GC-HBOC comprises 18 university centers from all over Germany, which offer counseling, pedigree analysis, risk calculation, genetic testing, surveillance programs and prophylactic surgery in accordance with GC-HBOC guidelines. Ethical approval was obtained from Charité – Universitätsmedizin Berlin (EA1/222/15). 2.1. Participants and instruments Initial inclusion criteria for study participants were women between the age of 18 and 70 years and either a statistical high risk for breast and ovarian cancer or detection of a pathogenic mutation in a gene that predisposes for breast and ovarian cancer, not longer than two years ago. Initial exclusion criteria were insufficient German language skills. In order to keep the heterogeneity at an acceptable level, we only included women with a pathogenic BRCA1- or BRCA2- mutation for data analysis in this study. Women were excluded from data analysis when their objective risk could not be calculated using the Breast and Ovarian Analysis of Disease Incidence and Carrier Estimation Algorithm (BOADICEA2 ) program (e.g. women with bilateral breast cancer or ovarian cancer) and when questionnaires missed variables. Women with a diagnosis of breast cancer were only asked to estimate their contralateral breast cancer risk to avoid misunderstandings concerning unilateral relapse risks (see Fig. 1). After initial counseling, the genetic testing process at our center consists of three steps as soon as counselees opt for genetic analysis: (a) genetic testing; (b) communication of test results; (c) optional genetic counseling concerning age-specific risks for breast and/or ovarian cancer. This additional counseling session focuses on 10-year-risks combined with personalized clinical risk management options following the guidelines for counseling (GC-HBOC and the Association of the Scientific Medical Societies in Germany (AWMF)). Risk perception is not explicitly addressed in this setting. Between August 2015 and April 2017, 300 women with a positive mutation test result or who had a statistical high breast cancer risk and decided to get additional optional counseling were screened for our initial inclusion criteria (Fig. 1). 250 women, who met the initial inclusion criteria, were invited to participate in the study. Of these, 207 women provided written consent and were mailed questionnaires subsequently. Of 127 women who returned the questionnaire, 39 were excluded from data analysis (see Fig. 1 and exclusion criteria above for more information). In total, 88 counselees who had obtained a positive BRCA1/2mutation test result on average 14.9 12.6 months ago were included in the analysis. From this group those women were excluded who already had a (secondary) prophylactic mastectomy (26) and those who had not reported subjective risks (3). This allowed us to present objective and subjective breast cancer risk assessments for 59 women. To illustrate the objective and subjective ovarian cancer risks, we had to exclude 34 women who had already undergone prophylactic ovarian removal and 2 women who had not reported any subjective risks. Thus, the group for which we could report objective and subjective ovarian cancer risks was 52 women. Therefore, the difference between objective and subjective risk estimates could be calculated for 59 counselees in terms of breast
2 The risk calculation tool BOADICEA is a web-based risk prediction model for hereditary breast and ovarian cancer used to compute lifetime and age-specific breast and ovarian cancer risks for BRCA1/2-mutation carriers [30][31],]. It is based on complex segregation analyses of families with multiple affected individuals carrying BRCA1/2-mutations [[32]]. The usage of this tool has been incorporated in the guidelines of several countries for the management of hereditary breast and ovarian cancer [[33][34],]. All participants were informed that their risks as calculated from BOADICEA were used for research purposes only.
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931
1927
Fig. 1. Flowchart of study participants. *reasons for exclusion: BOADICEA could not be calculated (bilateral breast cancer (4), history of ovarian cancer (7), missing variables (6), low-risk-mutation (9) or no mutation (13).
cancer risk and for 52 counselees for ovarian cancer risk (Fig. 1). Counselees were assigned to several subgroups according to clinical parameters: whether or not they had already undergone prophylactic surgery, whether or not they had a diagnosis of cancer, and the type of the diagnosed cancer (ovarian or breast cancer, respectively). This resulted in eight subgroups, for which we depict the 10-year-risks and subjective risk estimates separately (see Supplement). 3. Procedures Participants’ objective risks of developing breast and ovarian cancer were calculated using the risk calculation tool BOADICEA described above [30,31], which is available online. Clinical data that determine disease risks were taken from case report forms and entered in BOADICEA. Sociodemographic variables (age, level of education, parenthood) and cancer worry were assessed with self-report questionnaires. Four items of the cancer worry scale were used to assess cancer worry. The sum score range was 4 to 16, with higher values indicating higher levels of cancer worry [35]. Women were asked to estimate their subjective numeric risks of developing breast or ovarian cancer within the next 10 years: “Imagine 100 women in exactly your situation. How many of these women do you think will (again) develop breast cancer within the next 10 years?”, and, accordingly, “How many of these women do you think will develop ovarian cancer within the next 10 years?”. Women with breast cancer were only asked to indicate their risk estimation for the contralateral breast to avoid misunderstandings concerning unilateral relapse risks. In addition, women were asked how well informed they felt about the gene test result after the consultation. The subjective risk was assessed with simple frequencies. Compared with percentage formats, simple frequencies can be more easily understood and interpreted [36], allowing for more accurate lifetime risk estimates [37]. 3.1. Data analysis We report means and standard deviations for metrically and ordinally scaled variables for the entire sample and for women with and without breast cancer separately. For categorical variables, we display frequencies and percentages. Group
comparisons for continuous variables were carried out using Student’s t-test to determine if the two patient groups differed significantly. To indicate the accuracy of risk estimation, we calculated a difference score between the objective disease risks calculated with BOADICEA and women’s estimated probabilities of developing breast and ovarian cancer. Participants estimates were scored as correct with an error tolerance of 5 (5% of scale). Accordingly, three groups were considered: underestimation, correct estimation, and overestimation. We tested for the differences between mean objective and subjective risk estimates with the t-test for dependent samples. Additionally, the Bartlett’s test for homogeneity of variances was used to test if the variances of objective and subjective risk estimates differed. Associations between clinical and psychosocial variables on the one hand, and the difference score on the other hand were determined with Pearson’s r. An alpha level of p < .05 was considered statistically significant. 4. Results Of all women who returned the questionnaires (N = 127), 89.8% felt well informed about the genetic test result. Risk calculations for either the risk of breast cancer or ovarian cancer, or both, were possible for 88 women (Fig. 1). Women were on average 42.1 10.6 years old (range 19–67 years). Of these, 41 study participants had a diagnosis of breast cancer. These women reported higher levels of cancer worry and were on average 4.5 years older than the 47 women without breast cancer. No woman had developed metastases at the time of the study. Women without a former diagnosis of breast cancer had more often undergone prophylactic surgery (e.g. risk-reducing salpingo-oophorectomy or secondary prophylactic mastectomy). The sample characteristics are shown in Table 1. The objective 10-year risk for breast cancer could be calculated for 62 women (not possible for n = 22 women with mastectomy or bilateral breast cancer). Of these, 59 provided subjective risk estimates for breast cancer. The mean objective risks differed significantly from the subjective estimates (Mobjective risk = 20.44 9.05 vs. Msubjective risk = 33.88 25.52; t(58) = 4.12, p < .001) with an effect of Cohen’s d = 0.70. Bartlett’s test for homogeneity of
1928
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931
Table 1 Sample characteristics differentiated by breast cancer status. No breast cancer (n = 47)
Diagnosis of breast p cancer (n = 41)
Characteristic
Total sample (N = 88)
Age, yrs, mean (SD) Higher education, n (%) Living with partner Parenthood, n (%) Prophylactic surgery, n (%) No prophylactic surgery Prophylactic mastectomy Prophylactic salpingoophorectomy Mastectomy and salpingoophorectomy Time since test result in months (SD) Cancer worry, mean (SD)
42.10 (10.6) 44.68 (9.23) 57 (64.8) 22 (57.9)
40.14 (11.30) 35 (70.0)
.047 .170
71 (80.7) 61 (69.3)
28 (73.7) 28 (73.7)
43 (86.0) 33 (66.0)
.120 .296 .001
39 (44.3)
9 (23.7)
30 (60.0)
14 (15.9)
9 (23.7)
5 (10.0)
23 (26.1)
10 (26.3)
13 (26.3)
12 (13.6)
10 (26.3)
2 (4.0)
14.86 16.12 (11.88) (12.63) 10.47 (3.29) 11.35 (2.90)
risk estimation for developing breast or ovarian cancer did not associate with age, education, parenthood, prophylactic surgery, or cancer worry. Only the desire to undergo prophylactic mastectomy was related to the overestimation of breast cancer risk (r = .34, p = .010) (Table 2). We performed additional analyses with subgroups that were built according to the presence of a cancer diagnosis and/or prophylactic surgery. These analyses indicated a tendency towards risk overestimation across the subgroups (Figure S1). 5. Discussion and conclusion 5.1. Discussion
13.95 (13.16)
.425
9.80 (3.43)
.029
variances indicated a significant difference in the variances of the objective risks and the subjective risk estimation (X = 14.89; p < .001). The respective distributions are depicted in Fig. 3a. Permitting a tolerance of +/-5% for an accurate risk estimation, 14 (23.7%) of the women underestimated their breast cancer risk, 12 (20.3%) provided correct numbers, and 33 (55.9%) overestimated their risk (Fig. 2). The objective ten-year risk for ovarian cancer was calculated for 54 women (not possible for n = 34 women with ovarian cancer or prophylactic salpingo-oophorectomy). Of these, 52 provided subjective risk estimates for ovarian cancer. Again, the mean objective risk differed from the subjective risk estimation (Mobjective risk = 3.25 3.12 vs. Msubjective risk = 18.45 18.62; t (51) = 5.80, p < .001) with an effect of d = 1.14, as did the variances of the objective risks and the subjective risk estimates (X = 36.09; p < .001) (Fig. 3b). Only 3 (5.6%) of the women underestimated their ovarian cancer risk, whereas 21 (38.9%) provided correct numbers, and 28 (51.9%) overestimated their risk. The accuracy of
Fig. 2. Proportion of women underestimating and overestimation their risk of breast and ovarian cancer (%). A tolerance of (+/5%) was permitted for the correct estimation.
The objective of this study was to evaluate the accuracy of risk estimation in BRCA1/2-mutation carriers after genetic counseling. To our knowledge, it is the first study comparing women’s subjective risk estimates with the objective 10-year-risk. The key findings of our study are that, even though women generally felt well informed, they markedly overestimated their breast and ovarian cancer risks after guideline-based genetic counseling. Women with past cancer diseases or prophylactic surgeries did not provide more accurate risk estimates. In case of ovarian cancer risk, the magnitude of overestimation is particularly striking. Several reasons can be assumed for the stronger overestimation of the ovarian cancer risk as compared to the breast cancer risk: (1) During counseling, women must be informed about the lack of screening options and still unsatisfying therapy options for ovarian cancer, which could intensify risk perception and influence risk understanding; (2) women may confuse numbers for ovarian cancer and breast cancer risks, which could lead to higher estimates for ovarian cancer risk; (3) ovarian cancer is a rare disease, and events with a lower probability are more likely to be overestimated [7]; (4) finally, there may be people who just guess, which is likely to result in a number somewhere in the middle of the scale (“can be/ cannot be” or “50:5000 ). As a consequence, the discrepancy between the subjective and the objective risk due to pure guessing could be larger in cases with a low objective risk. Our findings further indicate that the women’s subjective 10-year risk estimates are close to their actual objective lifetimerisks. The differentiation of age-specific and lifetime risks is a challenge. Possibly, these two risks are confused, which is not surprising as predominantly lifetime risks are communicated in the media. Because 2-, 5- and 10-year risks are substantially more important for risk-management [26], special attention should be paid to explaining the distinction between lifetime risk and agespecific risks in the consultation. Surprisingly, in our study risk overestimation was independent from mutation type, level of cancer worry, education, age and parenthood. The fact that no clear correlations between psychosocial or clinical factors and overestimation could be identified in this population may imply that the consultant cannot rely on, for instance, less anxious or more educated women to accurately assess their risks. Thus, a similar effort is required to make the risks transparent and easy to understand for all women. It should be examined which graphics, tabular formats [38] and verbal explanations are particularly suitable for supplementing the numerical information to support counseling [39]. In addition, 23.7% and 5.6% of the responders underestimated their objective breast and ovarian cancer risks. Again, we could not find any correlation as to prophylactic surgeries or age. Future analyses should address this phenomenon in detail to make sure that this underestimation will not impair compliance to intensified surveillance programs. This group of women should also be kept in mind in the design of counseling-supporting tools.
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931
1929
Fig. 3. Objective risks and estimates for the 10-year risk of breast (3a) and ovarian (3b) cancer.
Table 2 Correlations between sociodemographic, clinical, and psychosocial variables and difference scores between objective risks and subjective estimates. Characteristic
Correlation with the breast cancer risk difference score (n = 59)
p
Correlation with the ovarian cancer risk difference score (n = 52)
p
Age, yrs, mean (SD) Higher education, n (%) Living with partner Parenthood, n (%) Prophylactic surgery, n (%) Time since test result in months (SD) Cancer worry, mean (SD)
.03
.810
.01
.967
.17
.194
.23
.101
.13
.327 .06
.659
.17
.192
.05
.751
.25
.055 .03
.816
.19
.167
.19
.182
.10
.444 .13
.353
5.2. Practice implications Numbers are often presented in a way that is difficult to understand and patients are challenged by extracting the relevant numbers and interpreting them adequately [40,41]. Physicians should make a great effort to present the relevant numbers in the most transparent way possible in order to enable informed decisions when managing cancer risks. The complexity and amount of information requires new ways in risk consulting. 5.3. Study strengths and limitations One strength of our study is the inclusion and differentiated evaluation regarding clinical characteristics. We postulate that extensive analysis and survey of the actual individual situation of each counselee is a prerequisite of successful counseling. Limitations include that the cross-sectional design does not allow analysis of changes in risk assessment over time. This might be interesting because risk assessment may vary due to changes in family situation or through risk management measures. Second,
we did not measure health literacy and therefore cannot say what proportion of overestimation is due to limited abilities to understand health information. However, education, which was associated with health literacy in other studies [13], was not correlated with inaccurate estimates in our study. Third, although this is one of the largest samples on this topic, the power was not sufficient to confirm variations in risk estimation accuracy across the different subgroups with past diseases states and prophylactic surgeries. In addition, due to the ‘right-not-to-know’, which is part of the German Genetic Diagnostics Act (“Gendiagnostikgesetz”), there is a clear bias in our study population towards more interested counselees. Due to the current limitations of the calculation tool, 10-year risks for ovarian cancer could not be calculated for patients who had bilateral breast cancer before. Neither, 10-year risks for breast cancer could be calculated for patients with ovarian cancer. Therefore, even if we emphasized the differentiated depiction of subgroups these two important group of counselees could not be considered. Finally, we did not focus on counselees’ specific personal family history with breast and ovarian cancer with regard to their overestimation. However, this individual familial background seems to be important in the complex process of risk assessment and should be investigated in further studies. 5.4. Conclusion and research recommendations Women with a BRCA1/2-gene mutation strongly overestimate their 10-year risk of breast cancer. Overestimation was even more pronounced in the case of ovarian cancer. This effect was not restricted to women with a particular psychosocial background, such as lower education, a prior diagnosis of cancer, or children. Thus, counselors must be prepared to prevent overestimation in all of their counselees. On the other hand, given large individual differences in objective risks based on clinical parameters, counselors must also adapt their advice to meet the individual requirements and situation of each counselee. In other words, counseling must be standardized and personalized at the same time. Future studies should investigate which counseling-supporting tools are particularly suitable to achieve the right balance between standardization and personalization in counseling to improve risk understanding and as a prerequisite for preventing over-treatment.
1930
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931
Competing interests The authors declare that they have no conflict of interest.
[7] [8]
Authors’ contributions DS, FR, MF and FK designed the study, carried out the research, analyzed the data and wrote the manuscript. LB and HB substantially contributed to the design of the study and participated in the acquisition and interpretation of the data. All authors read and approved the final manuscript.1 Human studies and informed consent All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000 (5). Informed consent was obtained from all patients for being included in the study. Funding The study was supported by an unrestricted grant from the Berliner Krebsgesellschaft e.V. [grant number KEFF201607]. The funding source had no role in design, conduct of the study, interpretation of the data, or publication. We confirm all patient/personal identifiers have been removed or disguised so the patient/person(s) described are not identifiable and cannot be identified through the details of the story. Acknowledgements We wish to thank the study participants, who provided their time and showed great interest in this study. We would also like to thank Nanette Kalmbach and Marina Kniehase for their support in data acquisition.
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.pec.2019.05.007. References [1] K.B. Kuchenbaecker, J.L. Hopper, D.R. Barnes, K.-A. Phillips, T.M. Mooij, M.-J. Roos-Blom, S. Jervis, F.E. van Leeuwen, R.L. Milne, N. Andrieu, D.E. Goldgar, M.B. Terry, M.A. Rookus, D.F. Easton, A.C. Antoniou, BRCA1 and BRCA2 Cohort Consortium, L. McGuffog, D.G. Evans, D. Barrowdale, D. Frost, J. Adlard, K.-R. Ong, L. Izatt, M. Tischkowitz, R. Eeles, R. Davidson, S. Hodgson, S. Ellis, C. Nogues, C. Lasset, D. Stoppa-Lyonnet, J.-P. Fricker, L. Faivre, P. Berthet, M.J. Hooning, L.E. van der Kolk, C.M. Kets, M.A. Adank, E.M. John, W.K. Chung, I.L. Andrulis, M. Southey, M.B. Daly, S.S. Buys, A. Osorio, C. Engel, K. Kast, R.K. Schmutzler, T. Caldes, A. Jakubowska, J. Simard, M.L. Friedlander, S.-A. McLachlan, E. Machackova, L. Foretova, Y.Y. Tan, C.F. Singer, E. Olah, A.-M. Gerdes, B. Arver, H. Olsson, Risks of breast, ovarian, and contralateral breast cancer for BRCA1 and BRCA2 mutation carriers, JAMA 317 (2017) 2402–2416, doi:http://dx.doi.org/10.1001/jama.2017.7112. [2] D. Stacey, R. Samant, C. Bennett, Decision making in oncology: a review of patient decision aids to support patient participation, CA. Cancer J. Clin. 58 (2008) 293–304, doi:http://dx.doi.org/10.3322/CA.2008.0006. [3] F.M. Walter, J. Emery, Perceptions of family history across common diseases: a qualitative study in primary care, Fam. Pract. 23 (2006) 472–480, doi:http://dx. doi.org/10.1093/fampra/cml006. [4] A. Werner-Lin, Building the cancer family: family planning in the context of inherited breast and ovarian cancer risk, J. Soc. Soc. Work Res. 1 (2010) 14–27, doi:http://dx.doi.org/10.5243/jsswr.2010.3. [5] D. Leblond, A. Brédart, S. Dolbeault, A. De Pauw, D. Stoppa Lyonnet, C. Flahault, S. Sultan, [Cognitive, emotional and behavioral impact of an uncertain outcome after study of BRCA1/2: review of the literature], Bull. Cancer (Paris). 98 (2011) 184–198, doi:http://dx.doi.org/10.1684/bdc.2011.1309. [6] D.P. French, J. Southworth, A. Howell, M. Harvie, P. Stavrinos, D. Watterson, S. Sampson, D.G. Evans, L.S. Donnelly, Psychological impact of providing women
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
with personalised 10-year breast cancer risk estimates, Br. J. Cancer. 118 (2018) 1648–1657, doi:http://dx.doi.org/10.1038/s41416-018-0069-y. W.M.P. Klein, M.E. Stefanek, Cancer risk elicitation and communication: lessons from the psychology of risk perception, CA. Cancer J. Clin. 57 (2007) 147–167. C.J. Frost, V. Venne, D. Cunningham, R. Gerritsen-McKane, Decision making with uncertain information: learning from women in a high risk breast cancer clinic, J. Genet. Couns. 13 (2004) 221–236, doi:http://dx.doi.org/10.1023/B: JOGC.0000027958.02383.a9. D.G. Evans, J. Wisely, T. Clancy, F. Lalloo, M. Wilson, R. Johnson, J. Duncan, L. Barr, A. Gandhi, A. Howell, Longer term effects of the angelina jolie effect: increased risk-reducing mastectomy rates in BRCA carriers and other high-risk women, Breast Cancer Res. BCR. 17 (2015) 143, doi:http://dx.doi.org/10.1186/ s13058-015-0650-8. N.E. Carbine, L. Lostumbo, J. Wallace, H. Ko, Risk-reducing mastectomy for the prevention of primary breast cancer, Cochrane Libr., John Wiley & Sons, Ltd, 2018, doi:http://dx.doi.org/10.1002/14651858.CD002748.pub4. A.W. Kurian, D.Y. Lichtensztajn, T.H.M. Keegan, D.O. Nelson, C.A. Clarke, S.L. Gomez, Use of and mortality after bilateral mastectomy compared with other surgical treatments for breast cancer in California, 1998-2011, JAMA 312 (2014) 902–914, doi:http://dx.doi.org/10.1001/jama.2014.10707. K. Rhiem, K. Pfeifer, R.K. Schmutzler, M. Kiechle, risk-reducing surgery in women at risk for familial breast or ovarian cancer, Geburtshilfe Frauenheilkd. 72 (2012) 833–839, doi:http://dx.doi.org/10.1055/s-0032-1315362. E.J. Rutherford, J. Kelly, E.A. Lehane, V. Livingstone, B. Cotter, A. Butt, M.J. O’Sullivan, F.O. Connell, H.P. Redmond, M.A. Corrigan, Health literacy and the perception of risk in a breast cancer family history clinic, Surg. J. R. Coll. Surg. Edinb. Irel. 16 (2018) 82–88, doi:http://dx.doi.org/10.1016/j.surge.2016.06.003. S. Schott, L. Vetter, M. Keller, T. Bruckner, M. Golatta, S. Eismann, N. Dikow, C. Evers, C. Sohn, J. Heil, Women at familial risk of breast cancer electing for prophylactic mastectomy: frequencies, procedures, and decision-making characteristics, Arch. Gynecol. Obstet. 295 (2017) 1451–1458, doi:http://dx. doi.org/10.1007/s00404-017-4376-y. R. Glassey, M. O’Connor, A. Ives, C. Saunders, k Con Fab Investigators, S. O’Sullivan, S.J. Hardcastle, Heightened perception of breast cancer risk in young women at risk of familial breast cancer, Fam. Cancer 17 (2018) 15–22, doi:http://dx.doi.org/10.1007/s10689-017-0001-2. S. van Dijk, W. Otten, M.W. Zoeteweij, D.R.M. Timmermans, C.J. van Asperen, M.H. Breuning, Ra.E.M. Tollenaar, J. Kievit, Genetic counselling and the intention to undergo prophylactic mastectomy: effects of a breast cancer risk assessment, Br. J. Cancer. 88 (2003) 1675–1681, doi:http://dx.doi.org/10.1038/sj.bjc.6600988. D.J. Bowen, C.M. Alfano, B.A. McGregor, M.R. Andersen, The relationship between perceived risk, affect, and health behaviors, Cancer Detect. Prev. 28 (2004) 409–417, doi:http://dx.doi.org/10.1016/j.cdp.2004.08.008. M. Broadstock, S. Michie, J. Gray, J. Mackay, T.M. Marteau, The psychological consequences of offering mutation searching in the family for those at risk of hereditary breast and ovarian cancer–a pilot study, Psychooncology. 9 (2000) 537–548. G. Cicero, R. De Luca, P. Dorangricchia, G. Lo Coco, C. Guarnaccia, D. Fanale, V. Calò, A. Russo, Risk perception and psychological distress in genetic counselling for hereditary breast and/or ovarian cancer, J. Genet. Couns. 26 (2017) 999–1007, doi:http://dx.doi.org/10.1007/s10897-017-0072-0. D.G.R. Evans, V. Blair, R. Greenhalgh, P. Hopwood, A. Howell, The impact of genetic counselling on risk perception in women with a family history of breast cancer, Br. J. Cancer. 70 (1994) 934, doi:http://dx.doi.org/10.1038/ bjc.1994.423. P. Hopwood, A. Shenton, F. Lalloo, D.G.R. Evans, A. Howell, Risk perception and cancer worry: an exploratory study of the impact of genetic risk counselling in women with a family history of breast cancer, J. Med. Genet. 38 (2001), doi: http://dx.doi.org/10.1136/jmg.38.2.139 139–139. E.A. Lobb, P.N. Butow, B. Meiser, A. Barratt, C. Gaff, M.A. Young, J. Kirk, M. Gattas, M. Gleeson, K. Tucker, Women’s preferences and consultants’ communication of risk in consultations about familial breast cancer: impact on patient outcomes, J. Med. Genet. 40 (2003), doi:http://dx.doi.org/10.1136/jmg.40.5. e56 e56–e56. Y. Rothemund, S. Paepke, H. Flor, Perception of risk, anxiety, and health behaviors in women at high risk for breast cancer, Int. J. Behav. Med. 8 (2001) 230–239, doi:http://dx.doi.org/10.1207/S15327558IJBM0803_5. A. Bish, S. Sutton, C. Jacobs, S. Levene, A. Ramirez, S. Hodgson, Changes in psychological distress after cancer genetic counselling: a comparison of affected and unaffected women, Br. J. Cancer. 86 (2002) 43–50, doi:http://dx. doi.org/10.1038/sj.bjc.6600030. K. Kelly, H. Leventhal, M. Andrykowski, D. Toppmeyer, J. Much, J. Dermody, M. Marvin, J. Baran, M. Schwalb, Using the common sense model to understand perceived cancer risk in individuals testing for BRCA1/2 mutations, Psychooncology. 14 (2005) 34–48, doi:http://dx.doi.org/10.1002/pon.805. R. Schmutzler, K. Rhiem, Beratung junger frauen mit hereditärer belastung für brust- und eierstockkrebs, Forum (Genova). 32 (2017) 37–41, doi:http://dx.doi. org/10.1007/s12312-017-0212-3. D. Keohane, E. Lehane, E. Rutherford, V. Livingstone, L. Kelly, S. Kaimkhani, F. O’Connell, H.P. Redmond, M.A. Corrigan, Can an educational application increase risk perception accuracy amongst patients attending a high-risk breast cancer clinic? Breast edinb, Scotl. 32 (2017) 192–198, doi:http://dx.doi. org/10.1016/j.breast.2017.02.009. A. Abbott, N. Rueth, S. Pappas-Varco, K. Kuntz, E. Kerr, T. Tuttle, Perceptions of contralateral breast cancer: an overestimation of risk, Ann. Surg. Oncol. 18 (2011) 3129–3136, doi:http://dx.doi.org/10.1245/s10434-011-1914-x.
D. Speiser et al. / Patient Education and Counseling 102 (2019) 1925–1931 [29] A. Albada, M.G.E.M. Ausems, J.M. Bensing, S. van Dulmen, Tailored information about cancer risk and screening: a systematic review, Patient Educ. Couns. 77 (2009) 155–171, doi:http://dx.doi.org/10.1016/j.pec.2009.03.005. [30] A.C. Antoniou, A.P. Cunningham, J. Peto, D.G. Evans, F. Lalloo, S.A. Narod, H.A. Risch, J.E. Eyfjord, J.L. Hopper, M.C. Southey, H. Olsson, O. Johannsson, A. Borg, B. Pasini, B. Passini, P. Radice, S. Manoukian, D.M. Eccles, N. Tang, E. Olah, H. Anton-Culver, E. Warner, J. Lubinski, J. Gronwald, B. Gorski, L. Tryggvadottir, K. Syrjakoski, O.-P. Kallioniemi, H. Eerola, H. Nevanlinna, P.D.P. Pharoah, D.F. Easton, The BOADICEA model of genetic susceptibility to breast and ovarian cancers: updates and extensions, Br. J. Cancer. 98 (2008) 1457–1466, doi: http://dx.doi.org/10.1038/sj.bjc.6604305. [31] A.C. Antoniou, P.P.D. Pharoah, P. Smith, D.F. Easton, The BOADICEA model of genetic susceptibility to breast and ovarian cancer, Br. J. Cancer. 91 (2004) 1580–1590, doi:http://dx.doi.org/10.1038/sj.bjc.6602175. [32] A.J. Lee, A.P. Cunningham, K.B. Kuchenbaecker, N. Mavaddat, D.F. Easton, A.C. Antoniou, D. of P.H. and P.C. The consortium of investigators of modifiers of BRCA1/2Centre for cancer genetic epidemiology university of Cambridge, strangeways research laboratory, Cambridge CB1 8RN, UK, D. P.H. and P.C. The breast cancer association ConsortiumCentre for cancer genetic epidemiology university of Cambridge, strangeways research laboratory, Cambridge CB1 8RN, UK, BOADICEA breast cancer risk prediction model: updates to cancer incidences, tumour pathology and web interface, Br. J. Cancer 110 (2014) 535– 545, doi:http://dx.doi.org/10.1038/bjc.2013.730. [33] B.D. Riley, J.O. Culver, C. Skrzynia, L.A. Senter, J.A. Peters, J.W. Costalas, F. CallifDaley, S.C. Grumet, K.S. Hunt, R.S. Nagy, W.C. McKinnon, N.M. Petrucelli, R.L. Bennett, A.M. Trepanier, Essential elements of genetic cancer risk assessment, counseling, and testing: updated recommendations of the national society of genetic counselors, J. Genet. Couns. 21 (2012) 151–161, doi:http://dx.doi.org/ 10.1007/s10897-011-9462-x.
1931
[34] R.A. Smith, V. Cokkinides, O.W. Brawley, cancer screening in the United States, 2012: a review of current American cancer society guidelines and current issues in cancer screening, CA. Cancer J. Clin. 62 (2012) 129–142, doi:http://dx. doi.org/10.3322/caac.20143. [35] C. Lerman, B. Trock, B.K. Rimer, C. Jepson, D. Brody, A. Boyce, psychological side effects of breast cancer screening, health psychol, Off. J. Div. Health Psychol. Am. Psychol. Assoc. 10 (1991) 259–267. [36] G. Gigerenzer, U. Hoffrage, How to improve bayesian reasoning without instruction: frequency formats, Psychol. Rev. 102 (1995) 684–704, doi:http:// dx.doi.org/10.1037/0033-295X.102.4.684. [37] M.M. Schapira, S.L. Davids, T.L. McAuliffe, A.B. Nattinger, Agreement between scales in the measurement of breast cancer risk perceptions, Risk Anal. Off. Publ. Soc. Risk Anal. 24 (2004) 665–673, doi:http://dx.doi.org/10.1111/j.0272-4332.2004.00466.x. [38] M. McDowell, F.G. Rebitschek, G. Gigerenzer, O. Wegwarth, A simple tool for communicating the benefits and harms of health interventions: a guide for creating a fact box, MDM Policy Pract. 1 (2016)2381468316665365, doi:http:// dx.doi.org/10.1177/2381468316665365. [39] N.T. Brewer, J.P. Tzeng, S.E. Lillie, A.S. Edwards, J.M. Peppercorn, B.K. Rimer, Health literacy and cancer risk perception: implications for genomic risk communication, Med. Decis. Mak. Int. J. Soc. Med. Decis. Mak. 29 (2009) 157– 166, doi:http://dx.doi.org/10.1177/0272989X08327111. [40] O. Wegwarth, W. Gaissmaier, G. Gigerenzer, Deceiving numbers: survival rates and their impact on doctors’ risk communication, Med. Decis. Mak. Int. J. Soc. Med. Decis. Mak. 31 (2011) 386–394, doi:http://dx.doi.org/10.1177/ 0272989X10391469. [41] J.C. Tilburt, K.M. James, P.S. Sinicrope, D.T. Eton, B.A. Costello, J. Carey, M.A. Lane, S.L. Ehlers, P.J. Erwin, K.E. Nowakowski, M.H. Murad, Factors influencing cancer risk perception in high risk populations: a systematic review, Hered. Cancer Clin. Pract. 9 (2011) 2, doi:http://dx.doi.org/10.1186/1897-4287-9-2.