Is genetic counseling for cancer predisposition always associated with distress? A pre-post intervention study to assess probands’ pre-and post-counseling level of anxiety and satisfaction

Genetic counseling for cancer predisposition is associated with a potentially underestimated emotional impact. The Genetic Counseling Service of the Institute of Oncology of Southern Switzerland evaluated the degree of anxiety before and after counseling and its correlation with the level of satisfaction of counselees. The STAI (State-Trait Anxiety Inventory) questionnaire was submitted to 80 counselees to measure trait (constitutive) and state (contingent) anxiety. It was submitted before and at the end of the fi rst interview. A specifi c questionnaire named genetic counseling satisfaction (GCS), was developed to evaluate the quality of information, submitted to counselees at the end of the fi rst meeting. The mean state and trait anxiety levels before the interview were respectively 41.4 and 40.7. The mean decrease after the interview was 4.3 for state anxiety (p<0.0001) and 1.2 (p=0.0054) for trait anxiety. The GCS showed that most counselees appreciated the clarity of information (85%) and the simple and understandable terminology used during counseling (88%). Higher levels of satisfaction were associated with higher level of state anxiety reduction (ρ=-0.23, p=0.03). Genetic counseling slightly impacts trait anxiety but results in an important reduction in state anxiety. The satisfaction from the interview infl uences the reduction of state anxiety associated with genetic counseling for cancer predisposition Research Article Is genetic counseling for cancer predisposition always associated with distress? A pre-post intervention study to assess probands’ pre-and post-counseling level of anxiety and satisfaction Simona Di Lascio1*, Elena Scaffi di1, Vincenzo Bagnardi2,3, Monica Taborelli1, Gabriella Bianchi Micheli1,5, Piercarlo Saletti1, Cinzia Cafaro-Greco1, Davide Disalvatore4 and Olivia Pagani1,5 1Genetic Counseling Service (CCGO), Institute of Oncology of Southern Switzerland (IOSI) 2Department of Statistics and Quantitative Methods, University of Milan-Bicocca, Milan, Italy, 3Clinical Trial Unit of the Ente Ospedaliero Cantonale (EOC) 4European Institute of Oncology, Milan, Italy (IEO) 5Breast Unit of Southern Switzerland (CSSI) Dates: Received: 13 July, 2017; Accepted: 24 August, 2017; Published: 28 August, 2017 *Corresponding author: Simona Di Lascio, MD, Oncology Institute of Southern Switzerland, Genetic Counseling Service (CCGO), Switzerland, Tel: +41918119039; E-mail:


Introduction
Approximately 5-10% of all breast cancers (BC) and about 3% of all colorectal cancers (CRC) are related to inherited genetic defaults [1]. Genetic testing became available for BC patients and their families after the identifi cation of two BC susceptibility genes, BRCA1 and BRCA2. Inherited BRCA1/2 mutations are associated with an increased risk of both breast and ovarian cancer [2]. Patients harboring inherited predisposition to CRC (Lynch Syndrome) can be identifi ed by both a microsatellite instability (MSI) test and an immunohistochemistry analysis for mismatch repair (MMR) genes, performed on CRC tumor DNA. In Lynch syndrome, almost all CRCs show high (positive) MSI [3][4][5][6]. Psychological distress after BC diagnosis and treatment has been well recognized and investigated.
Psychological distress was reported in 20-30% of BC patients within the fi rst year after diagnosis [7]. Burgess identifi ed anxiety, depression or both in nearly 50% of BC patients in the year following the diagnosis [8]. In addition, high frequency of intrusive thoughts and negation were noted in 18% and 14 % of BC patients six weeks after surgery, respectively [9].
Moderate distress has been reported in patients with stage I/ II CRC [10]. Cancer genetic counseling (GC) allows to identify individuals at increased risk for hereditary breast/ovarian cancer and Lynch Syndrome and to plan surveillance programs and cancer reduction strategies [11,1]. The psychological impact of GC and testing on unaffected women with a family history of BC has been widely studied. Non-carriers derive psychological benefi t from genetic testing, while no adverse effects are observed among gene carriers [12]. In a multicenter study on distress? A pre-post intervention study to assess probands' pre-and post-counseling level of anxiety and satisfaction. Arch Depress Anxiety 3(2): 030-037. DOI: http://doi.org/10.17352/2455-5460.000020 the psychological impact in patients recently diagnosed with CRC, disclosure of the MSI test result was not followed by high levels of distress in the majority of patients [13].
A growing number of women undergoes GC to assess genetic predisposition to breast/ovarian cancer. Several studies show women at high BC risk due to hereditary predisposition better adhere to surveillance programs and preventive strategies recommended after GC and positive test results [14][15][16][17][18][19]. The patient-doctor relationship also proved to be important to support high-risk patients and increase their compliance to risk managing recommendations [19,20]. On the contrary, women who tested negative experienced a signifi cant reduction in the perceived BC risk [21] which was associated with decreased adherence to the suggested check-up program [22,23]. This attitude in mutation negative subjects was confi rmed in the 3 years following genetic testing in ≥50 year's old women [24]. In other series, GC and testing, regardless of their results, induce women to undergo mammography controls as compared to clinical or self-breast examination [20,14,15]. While substantial evidence is available on the short term (≤1 year) effect of genetic testing on adherence to surveillance and prevention strategies, few studies addressed the long-term impact [15,27,24]. The impact of genetic testing on prophylactic mastectomy and the related reduction on BC incidence is less clear [24,17,25,26].
From a psychological perspective, few studies explored the infl uence of genetic testing on acute distress in mutation negative/positive women. Some studies report a steady preand post-testing distress both in mutated and non-mutated subjects either in the year immediately following testing or in subsequent years [23,24,15]. Van Oostrom [27], on the contrary, suggests increased anxiety and depression several years after testing both in mutation positive and negative women. Butow [28], reports mutation carriers do not experience a signifi cant increase in depression and anxiety, while women who tested negative feel released. In addition, while some studies found that anxiety and risk perception are associated with increased frequency of check-ups [29], others did not demonstrate this correlation [21].
Little is known about the acute psychological impact of genetic testing on BC patients who undergo DNA testing. The evidence available suggests that BC patients diagnosed <1 year before testing experienced high anxiety and BC-specifi c distress prior to GC and more depression after testing than patients assessed long term after diagnosis [30]. Patients diagnosed <1 year before testing seemed as interested as patients diagnosed >1 year before testing and showed more interest when advised by a physician [31]. So far, psychological distress during GC after a recent BC diagnosis and treatment has not been fully assessed.
The central role of communication in the relationship between patients and caregivers is well recognized. In some fi elds of healthcare, such as Phase I studies and GC, the increased fragility and vulnerability of the specifi c context mandates a special emphasis on communication aspects.
The informative and relational dimensions of the complex approach to cancer predisposition can potentially impact the quality of life of patients and healthy subjects attending a cancer genetic service and possibly contribute to their level of distress and psychic suffering. Assessment of both cancer predisposition understanding and its psychological impact is therefore needed to allow more tailored and effective patientdoctor communication in this sensitive fi eld.
In previous research projects in Phase I studies, we assessed the quality of the information given [32,33] in patients and families, concluding that it's possible to provide clear and correct information even in diffi cult situations. These studies have also shown that the way information is provided infl uences the comprehension of the risk, the related fears and concerns and the level of anxiety. These aspects have been rarely investigated in the GC area: in particular, no prospective data is available on both the informative and anxiety domains and their possible correlations.
We investigated the informative and relational aspects that potentially infl uence the quality of life of individuals undergoing GC at the Genetic Counseling Service (CCGO) of the Institute of Oncology of Southern Switzerland (IOSI). Our aim was fi rst to check the quality of the information process and then create a model for managing the counseling process while limiting the level of stress and psychic suffering.

Material and Methods
People involved in this study were patients or unaffected with a strong familiar cancer history or clinical elements suspicious for genetic predisposition to cancer. They received a fi rst interview in person and individual with an oncologist and a psychologist or a geneticist and after that, based on the results of multidisciplinary discussion between oncologist and geneticist, eventually performed genetic test. The interview with patients were taken after acute treatment phase and in condition of clinical wellness, if possible.
At the time of this study, the GC process included a fi rst interview in hospital, of about one hour, between the proband (patient or unaffected) and, depending on the predisposition syndrome, the oncologist and the psychologist (mainly breast/ ovarian families) or the geneticist (mainly colorectal cancer). This difference was originally decided based on the typology of counselees, as subjects with a predisposition to breast/ovarian cancer were mainly unaffected and those with a predisposition to CRC were mainly patients. After collecting data on personal and family history and explaining the meaning of frequency, absolute and relative risk and the management of the information for an adequate prevention, the family tree of the proband is designed and the mutation risk calculated using mathematical models (BRCAPRO) [34]. All the cases are then discussed multidisciplinary with the geneticist. If the overall assessment clearly demonstrates the absence of the minimum criteria to propose genetic testing, the GC process is concluded with a second explanatory interview. If a signifi cant mutation probability is detected, the second meeting is a critical step, as the possibility of genetic testing is discussed with the counselee. In the present study, all individuals were given two different self-compilation instruments: The level of global satisfaction has been calculated by summing the individual total questions' scores, to a three escalating scale of global satisfaction (low <18, medium 18-20, high >20 points). In addition, we calculated the relationship between the answer to each question and the changes in the level of anxiety before and after GC.

Counseling sessions and surveys have been conducted in
Italian language, native tongue for all subjects involved.

Pre-testing of the provisional questionnaire
Based on the development model of the EORTC qualityof-life questionnaires, the initial version was reviewed by the medical and coordinating staff taking part, in different ways, at the GC process (2 oncologists, 2 psychologists, 1 geneticist, 1 data manager/coordinator). The questionnaire was also tested in 20 subjects who undertook GC, who were asked to make comments and suggest changes. After discussion within the team, some questions were deleted and/or reformulated, leading to the defi nitive version of 25 items (Appendix 1).

Field testing
The STAI and the fi nal version of the GCS was administered to 100 consecutive individuals referring to the CCGO.

Timing of administration
Step 1: The counselee received at home a presentation letter from the CCGO and the appointment details together with the fi rst STAI questionnaire. The purpose of this initial administration was to assess the counselee's state and trait anxiety in a condition of relative neutrality (baseline). The CCGO coordinator checked that the counselee returned the completed questionnaire in a sealed envelope before the fi rst interview. The counselee was not informed on purpose about the STAI questionnaire before receiving it at home to avoid any possible infl uence on her/his answers.
Step 2: At the end of the fi rst interview, the counselee received the GCS questionnaire together with a 2 nd STAI: in this way, any difference in the degree of the state anxiety compared to the baseline could be related to the GC interview. The counselee was not informed in advance he/she will be asked to complete both questionnaires at the end of the interview not to affect in any way the answers.
After the collection of the questionnaires was completed, the CCGO approached the Institute of Communication and Health (ICH) of the Faculty of Communication Science at the University of Southern Switzerland (USI) to analyze the results. As no previous research in the fi eld included all the different areas investigated in the current research project, ICH fi rst conducted a qualitative analysis of the instrument in a selected number of counselees (data not shown) that allowed to identify its weaknesses and propose possible adjustments.
The study was approved by the Institutional Ethical Committee. Participants signed an informed consent.

Patients' selection
The eligibility criteria were: age ≥18 years, patients with BC or CRC or healthy relatives.

Statistical analysis
Differences in the distribution of subject characteristics between affected and unaffected probands were evaluated by the Chi-square test. Changes in subject-specifi c anxiety levels before and after GC were compared using the paired T-test. Baseline anxiety levels and changes before and after GC were compared among groups by the analysis of variance. The relationships between changes in individual state-and trait-anxiety levels and between the perceived quality of the information received during the GC and anxiety level changes were evaluated by the Pearson correlation coeffi cient ().
When this project was planned, no formal sample size calculation and power analysis was performed. However, a post hoc power calculation showed that this study (sample size=80) had adequate statistical power (>80%) to detect an overall mean change from baseline of anxiety levels distress? A pre-post intervention study to assess probands' pre-and post-counseling level of anxiety and satisfaction. Arch Depress Anxiety 3(2): 030-037. DOI: http://doi.org/10.17352/2455-5460.000020 greater than 2 points, assuming a standard deviation of the change equal to 6 and a two-sided 5% type I error rate.
Regarding the difference between two subgroups, the minimum detectable difference at 80% power was 3.4 points (assuming SD=6, two sided 5% type I error rate and balanced subgroups).

Evaluation of anxiety (STAI)
The mean baseline values of state and trait anxiety were evaluated according to age, sex, level of education, presence or absence of the psychologist during the interviews, the specialty of the referring physician, the type of predisposition syndrome and the belonging to the patient or unaffected group.
The baseline values found in our sample are within the average levels of the European population, indicating uniformity of our population to the average variability of the general population. (Table 2).
Overall, there is also no statistically signifi cant difference in the distribution of baseline levels of state and trait anxiety in all the subgroups examined: only the counselees with primary education show higher trait anxiety (score 46) compared to those with middle school (score 43), high school (score 37) or university (score 42) degrees.
The state anxiety is not distributed differently than trait anxiety according to the typology of counselees (affect and unaffected), sex, and age. Regarding the level of education, subjects with primary education reached a higher score (50.7), compared to those with secondary school (42.7), high school (37.4) and university (42.5) degrees.
Interestingly, counselees who appreciated the presence of the psychologist during the interview, reported a slightly higher level of state anxiety than those who did not.
Overall, the mean decrease for state anxiety was 4.3 (p<0.0001) and 1.2 for trait anxiety (p=0.01). The 2 nd STAI  The reduction of both types of anxiety was not signifi cantly different among different subgroups (Table 3); however, there is a trend towards a more consistent reduction in state anxiety in patients sent by the family doctor than by any of the specialists. be related not to anxiety but instead to the increased awareness, as outlined by the high proportion of respondents (72.5%) who declared the specifi c information on genetic testing did not increase their concern (question 16) and confi rmed by the results of the STAI questionnaire. Nevertheless, 90.0% of the counselees wanted to know if an hereditary cancer risk was present in their family (question 26), despite a comparable or even increased fear measured after the consultation in 75% of the respondents (question 24).

Evaluation of the genetic counseling consultation
Higher levels of satisfaction were associated with higher levels of state anxiety reduction (=-0.23, p=0.03) but not with higher levels of trait anxiety reduction (=-0.08, p=0.42) ( Figure 2).

Discussion
Overall, the implemented instrument to assess the quality of the information provided during the GC process has been very well quoted by responders, both affected and not affected, who found it quick and easy to be fi lled and with no unclear, intrusive, or repetitive questions. A high proportion of probands (66.2%) seemed to fully understand the implications of GC and subsequent genetic testing. Eighty-seventy-fi ve percent of the participants said they could ask any questions during the counseling and 81.2% reported complete satisfaction with the consultation process. Data also show a high degree of satisfaction in terms of well-being during the consultation, clarity and understanding of the information given.
Our working hypothesis was based on the idea that a substantial degree of anxiety is associated with the cancer  Trait-anxiety before the genetic counseling, as assessed by the STAI, did not show a substantial difference between affected and non-affected counselees and according to gender and age, with just a slight difference based on the education level, primary school respondents resulting more anxious. In addition, state-anxiety was also not differently distributed at baseline with regard to the type of counselee (affected/ non affected), gender and age, but primary school responders showed a higher level before consultation. We think these result are possibly infl uenced by the idea to be exposed to an unknown and complex situation. It is also interesting to notice that those preferring the presence of the psychologist in the GC team report a level of state-anxiety slightly higher than those who did not appreciate it. This increased state anxiety could be related to the counselee perception that the psychologist was present in view of the emotionally challenging content of the interview, requiring intellectual and decisional skills superior to his/her capabilities. As a consequence, the psychologist was favorably considered an additional support provided by the CCGO.
We also believed that once the diffi cult concepts pertaining to GC have been understood and the interviews properly conducted, respondents' anxiety should decrease. Results showed a slight decrease of trait-anxiety after the genetic counseling interviews, because this kind of anxiety is less dependent on the contingent situation and mainly determined  6) The analysis allowed to draw the profi le of the standard counselee (mostly unprepared by the referral physician, confi dent, calm) and to make some practical changes to the GC process. This qualitative analysis brought to reconsider how to prepare and organize the interview. Counselees are now routinely verbally informed by the CCGO coordinator on the way the consultation will be structured, receive at home a family tree to fi ll in advance and a summary leafl et to take home after the 1 st interview.

Future plans
The next planned step includes the integration of the information gained from both the quantitative and qualitative analysis of the data and the emerged weaknesses of the questionnaire (type and formulation of some questions) to develop a new tool: a new version of the questionnaire has been created and will be submitted for future analysis in cooperation with other genetic counseling services of Northern Italy.