Sage Journals: Discover world-class research

Abstract

Introduction

Patient-centricity calls for accurate consideration of patients’ preferences at every step of the care pathway. Decision aids (DAs) are interventions that provide patients with information about treatment options, enabling shared decision-making (SDM). This study investigated the practice of SDM by breast cancer (BC) specialists and the diffusion of DAs in the treatment of BC in Europe. The secondary objective of the study was to explore heterogeneity across respondents’ subgroups.

Methods

A prospective, cross-sectional, cross-country questionnaire targeting breast specialists was distributed online across breast cancer networks at both national and European level. The survey results were analyzed descriptively. Mann-Whitney-U tests, and probit regressions were performed to explore differences between subgroups.

Results

Respondents were mostly medical oncologists and surgeons. Respondents expressed high comfort with SDM approach (mean: 4.29, CI95%: 4.15-4.44). The majority reported DAs were available, and used, in their organization. Statistically significant gender differences were observed, indicating that female physicians might be more prone to adopting SDM approaches with patients.

Conclusions

Patient-centricity calls for consideration of patients’ values. The specialists in this survey showed a promising attitude towards SDM. DAs availability translates into use, although this mainly refers to paper tools, whereas digital tools seem mostly illustrated in the literature.

Keywords

survey shared decision-making patient-centricity decision aids breast cancer

Introduction

Background

Thanks to the latest scientific and therapeutic advancements, breast cancer (BC) can now be considered a chronic disease for many patients, rather than a life-threatening illness.¹ In the past few years, plenty of treatment options became available to patients, ranging from chemo-to radiotherapy, or from endocrine-based to surgical interventions. In this context, accounting for patients’ preferences and values becomes not only ethically appropriate, but also crucial for therapeutic success.^2,3 Patient decision aids (DAs) are interventions that support patients by making their decisions explicit, providing information about treatment options, and helping clarify congruence between decisions and personal values. A recent systematic review revealed that people exposed to DAs feel more knowledgeable and clearer about their values, which leads to a more active role in the decision-making process.⁴

DAs facilitate shared decision-making (SDM), namely a process of collaboration between patients and clinicians to reach a joint decision about care involving multiple medically appropriate options.^5,6 SDM is achieved by using appropriate communication techniques, understanding of the context in which patients live, and embracing their values on what is a good treatment outcome.⁷ SDM is known to increase quality of cancer care, and is associated with less decision regret, better treatment adherence, less treatment failure.^8-10 Recent updates of international BC clinical guidelines emphasized the need to foster the adoption of SDM approaches in clinical practice.^11,12 Relatedly, attention towards the impact that patient-clinician gender-concordance might have on clinical decision-making has recently emerged.¹³ On one hand, gender of treating physicians has been shown to have an impact on patient outcomes^14,15; on the other hand, the physicians most involved in the treatment decision-making of women diagnosed with BC have the potential to significantly contribute to the spread of SDM approaches, and as such the influence of specialists like oncologists and surgeons has been investigated in previous studies.¹⁶

The characteristics of DAs have changed over time, evolving from plain, paper-based interventions to increasingly sophisticated digital solutions. Quality standards, such as the International Patient Decision Aids Standards (IPDAS), have been developed, thus establishing a shared evidence-informed framework to guide developers and researchers in designing, building, and evaluating these tools.¹⁷ However, previous works called for more knowledge on DAs, revealing the need for stronger evidence base on their development, implementation and cost-effectiveness.¹⁸ Despite the growing interest shown both at the policy and professional levels, current use of DAs in clinical settings, as well as factors associated to their adoption, are therefore not clear.

Objectives

Using a cross-sectional survey, this work was primarily aimed at investigating the perceived practice of SDM by breast specialists and the uptake and diffusion of DAs in the treatment of BC care in Europe. The secondary objective of the study was to identify and explore heterogeneity across subgroups of respondents.

Methods

Survey Development

This is a prospective cross-sectional study. The reporting of this study conforms to STROBE guidelines.¹⁹ The process of developing the survey questionnaire was iterative and followed a systematic approach.²⁰

Step 1: Item generation

To explore the individual attitude towards SDM and the use of DAs as tools to support breast specialists in engaging their patients in treatment-related choices, a scoping review of the literature was performed to inform the development of the study questionnaire. The survey items were initially drafted by two researchers based on existing questionnaires or interview guides retrieved in the research streams of SDM and DAs in BC care. The questionnaire was developed in English.

Step 2: Survey draft

The Questionnaire Appraisal System (QAS) framework was used, to assess questionnaire items’ reading, instructions, clarity, assumptions, knowledge/memory, sensitivity/bias, response categories and other.²¹ Then, the questionnaire flow was evaluated jointly by the research team and an oncologist, focusing on the order, logic, and relevance of the questions. As result, the survey structure was updated, grouping questions differently or shifting question blocks. The resulting survey draft that formed the basis of cognitive testing contained up to 17 questions organized across three sections: (i) profiling, (ii) communication style and decision-making process, (iii) patient DAs. As for the profiling section, we used the term “gender” to refer to women and men as social categories. We believe this is consistent with our research focus on the behavioral and social dimensions of healthcare professionals’ experience.²²

Step 3: Cognitive testing

Cognitive interviewing is widely employed for pre-testing survey instruments, to evaluate whether respondents interpret the questions consistently, as intended by the investigator. Both think-aloud and verbal probing techniques have been used.²³ To select participants in the process of cognitive testing, a purposive sampling selection was applied satisfying criteria related to country, age, experience and gender. Overall, nine healthcare professionals were interviewed, representing 7 countries (ie, Albania, Belgium, France, Germany, Italy, Portugal, Spain), and with seniority that ranged from 2 to 40 years. All interviews were recorded. The questionnaire was not translated in national languages as English was deemed appropriate for the target respondents (ie, BC professionals regularly consult documents, guidelines, and various sources written in English).

Step 4. Final questionnaire

The final version of the questionnaire received ethical approval in November 2021 from the Ethical Review Board of Bocconi University (Milan, Italy), with application ID FA000375. The final questionnaire contained 17 questions, with filtering questions distinguishing between users, non-users, and potential users of DAs. Figure 1 displays the structure of the survey, reported in its integral version in Appendix Aand uploaded on Zenodo (10.5281/zenodo.7925762).

Figure 1.

Survey Flows by Users, Non-users and Potential Users. Abbreviations: PtDA = Patient Decision Aid; Q = Question; SDM = Shared Decision-Making.

Recruitment and Data Collection

Target respondents were breast specialists, specifically medical oncologists, surgeons, radiologists, radiation oncologists or any other professionals who may interact with breast patients at some point of the care pathway. The survey was disseminated through BC networks and scientific associations, both at national and international level in Europe. These included European Society of Breast Cancer Specialists (EUSOMA), European Commission Initiatives on Breast and Colorectal Cancer (ECIBS), Breast Surgical Oncology Platform (BRESO), Central-Eastern European Breast Cancer Surgical Consortium (CEEBCSC), Breast Centers Network (BCN), Seno Network Italia, Unicancer France, German Breast Group (GBG), Europa Donna Italia Scientific and Technical Committee. The breast specialists were approached via dedicated mailings or internal newsletters. Informed consent was obtained from all participants prior to survey access. Data was collected from April to August 2022 through the commercial electronic platform for survey services Qualtrics XM©. Due to the open distribution strategy of the survey, the response rate could not be calculated nor estimated. The final sample considered for the analyses included respondents who completed at least 70% of the survey items.

Data Analysis

Descriptive statistics of survey data were computed. We reported absolute and relative frequencies for categorial variables and means with associated standard deviations for metric variables. No missing data imputation was performed. Results were analysed both by gender, and professional roles (ie, medical oncologists, surgeons, and any other medical specializations). The breast units (BUs) could be classified as being certified based on nation-specific and/or European quality standards (ie, EUSOMA Breast Centres Certification), as opposed to not being classified under any certificates. Perceived barriers and facilitators to DA uptake were analyzed with the dual perspective of current users and non- or potential users. Means were compared using non-parametric Mann Whitney U tests. A series of ordered probit regression analyses were carried out for categorical variables that showed significant differences in the means across population subgroups. Each outcome was regressed on five theoretically grounded predictors: seniority, professional role, type of BU, the endorsement of “shared” decision-making archetype, and gender. Seniority was included based on evidence from diffusion-of-innovation research showing that more clinicians with longer tenue tend to adopt SDM practices more slowly and often perceive fewer relative advantages in doing so.²⁴ Professional role was considered because surgeons, oncologists, and radiologists play distinct roles in information exchange and preference elicitation.¹⁶ Identification with the “shared” SDM archetype (Charles et al) was used as a proxy for a clinician’s general propensity to involve patients in treatment decisions. The type of breast unit (certified vs not certified) was included to account for the organizational culture and resource endowment, known to modulate the implementation of SDM interventions in breast-cancer services.²⁵ Finally, gender was retained due to documented differences in communication style and decision-making authority in oncology settings. Statistical significance was set at P < 0.05. 95% confidence intervals (95%CI) were reported. Given the exploratory nature of the analyses, P-values were interpreted descriptively and no power calculation was performed. As a sensitivity check, we conducted additional analyses excluding countries with fewer than 10 respondents. Analyses were performed using the Stata 17.0 statistical package (StataCorp LLC, College Station, TX, USA).

Results

Participants’ Profiling

After excluding 238 invalid responses (response rate below 70%, or specialists not primarily specialized in breast cancer), a total of 198 valid responses were considered. Of these, over 98% (194/198) had a 100% completion rate, whereas 4 had a completion rate between 70% and 100% due to partially incomplete responses in the profiling section. Table 1 summarized the characteristics of the study participants. Respondents were mostly female (117, 62%). Participants came from 26 different countries, although most responses were from Italy (72, 36%), Germany (26, 13%), France (22, 11%), United Kingdom (15, 8%), and Portugal (13, 7%). Seniority was defined as the years of practice after the residency. Average seniority was 19 years (min 1 year; max 40 years; median 20 years), although with some cross-country variability (eg, average of 24 years in Italy, 13 years in Portugal). Most respondents were medical oncologists (70, 35%) and breast surgeons (70, 35%), whilst other professional roles, aggregated in the table, included radiation oncologists, radiologists, and self-reported roles (open field). In the latter, nurses and gynecological oncologists were somewhat recurring. Respondents reported working in certified breast units, either at national (91, 48%) or European (63, 33%) level, and in relatively large centers, with only 4% (7) of respondents treating less than 100 patients on a yearly basis in their center.

Table 1.

Descriptive Information of the Study Participants (Total Sample)

	Total (N = 198)
Gender
Female	117 (62%)
Male	72 (38%)
Prefer not to say	1 (0%)
Total	190 (100%)
Seniority (years)
Average (SD)	19 (10.9)
Minimum	1
Maximum	40
Professional role
Medical oncologist	70 (35%)
Surgeon	70 (35%)
Radiation oncologist	18 (9%)
Radiologist	10 (5%)
Other	30 (15%)
Total	198 (100%)
Type of BU
Certified - national level	91 (48%)
Certified - European level	63 (33%)
Not certified	24 (13%)
I don’t know	13 (7%)
Total	191 (100%)
Patients treated yearly*
≤100	7 (4%)
101-300	79 (40%)
301-500	44 (22%)
≥501	61 (31%)
Total	191 (100%)
Country
Italy	72 (37%)
Germany	26 (13%)
France	22 (11%)
United Kingdom	15 (8%)
Portugal	13 (7%)
Others¹	47 (24%)
Total	195 (100%)

Abbreviations: BU = Breast Unit; SD = Standard Deviation.

*At BU level.

Notes. ¹ Others include: Albania, Austria, Belgium, Bosnia, Bulgaria, Croatia, Cyprus, Denmark, Georgia, Greece, Iceland, Latvia, Lithuania, Netherlands, Norway, Poland, Russia, Serbia, Slovenia, Spain, Turkey.

Decision-Making Archetypes, Process and Influencing Factors

The survey examined patient-clinician interaction in a treatment decision process, with a focus on SDM (Table 2). The former was investigated using Charles et al (2004)²⁶ treatment decision-making framework and Emanuel et al (1992)²⁷ models of patient-physician relationship. The authors derived alternative approaches to engage patients corresponding to different degrees of transparency over available treatments, risks and benefits, consideration of patients’ values and preferences, and patient involvement in the decision itself.^16,26,28 Respondents were asked to quantify their level of comfort with the four archetypes on a 1-5 scale. Interestingly, the “shared” approach scored the highest values when the whole sample is considered (mean: 4.29, CI95%: 4.15-4.44). Virtually overlapping values were observed when responses were stratified by role (results shown in Appendix B), indicating that all specialists embrace the need to reach a shared decision with their patients, at least in theory.

Table 2.

Decision-Making Archetypes, Process and Factors, Total Sample and by Gender: Mean (SD) and Mann–Whitney U Test Results

	Total (N = 198)	Female (N = 117)	Male (N = 72)	Mann-Whitney U test
Decision-making archetypes^a
1 Paternalistic	3.10 (1.34)	2.99 (1.34)	3.31 (1.36)
2 Informed	3.67 (1.08)	3.64 (1.05)	3.74 (1.11)
3 Deliberative	2.61(1.29)	2.51 (1.21)	2.71 (1.39)
4 Shared	4.29 (1.04)	4.32 (0.95)	4.19 (1.21)
Decision-making process^b
1 Clinician giving information to the patient	4.65 (0.72)	4.73 (0.52)	4.56 (0.93)
2 Patient giving information on what is important to her/him	4.52 (0.75)	4.63 (0.61)	4.33 (0.90)	*
3 Clinician giving a treatment recommendation	4.30 (0.80)	4.36 (0.7)	4.19 (0.96)
4 Patient deciding on their treatment without the clinician	2.34 (1.24)	2.5 (1.24)	2.00 (1.13)	**
5 Discussing pros and cons of treatment options with the patient	4.61 (0.66)	4.62 (0.60)	4.58 (0.75)
6 Clinician and patient agreeing on a treatment option	4.58 (0.73)	4.62 (0.63)	4.50 (0.89)
7 Engaging with patients’ family members or caregivers	3.87 (0.99)	4.11 (0.87)	3.43 (1.03)	***
Shared decision-making factors^c
1 Patients can interpret SDM as a sign of incompetence	1.98 (0.83)	2.00 (0.82)	1.92 (0.82)
2 Patients with lower levels of education are more challenging to engage in SDM	2.86 (0.81)	2.87 (0.82)	2.79 (0.82)
3 Patients need DAs (leaflets, videos, apps) to engage in their health decisions	2.98 (0.78)	3.10 (0.71)	2.74 (0.84)	**
4 Organizational support (top management, peers) is necessary to implement SDM	3.21 (0.72)	3.24 (0.73)	3.17 (0.75)
5 SDM requires appropriate training of health professionals	3.53 (0.64)	3.57 (0.59)	3.4 (0.73)
6 Patient associations can aid SDM by disseminating material and giving support to patients	3.24 (0.72)	3.27 (0.71)	3.18 (0.76)
7 Putting SDM into practice will lead to improved patient satisfaction, mental health and quality of life	3.67 (0.55)	3.71 (0.49)	3.60 (0.64)
8 Lack of time and resources (staff, equipment, space) makes it difficult to engage in SDM	3.02 (0.94)	3.34 (0.90)	3.00 (1.02)

Abbreviations: DA = Decision Aid; SD: standard deviation; SDM = Shared Decision-Making.

Notes: Values are means with standard deviations in parentheses. Gender-group differences (Female vs Male) were tested with Mann-Whitney U; corresponding P-values are reported in the right-hand column. Statistical significance is marked with asterisks: * for P-values ≤0.05, ** for P-values ≤0.01, *** for P-values ≤0.001. Non-significant P-values are not reported.

^aOn a 1-5 Likert-scale where 1 = Not comfortable and 5 = Extremely comfortable.

^bOn a 1-5 Likert-scale where 1 = Not at all important and 5 = Extremely important.

^cOn a 1-4 Likert-scale where 1 = Strongly disagree and 4 = Strongly agree.

Then, respondents were asked to state how relevant they found factors commonly involved in a decision-making process on a 1-5 scale. Providing patients with information was considered the most important dimension (mean: 4.65, CI95%: 4.55-4.75), followed by discussing pros and cons of treatment options with the patient (mean: 4.61, CI95%: 4.52-4.70), and agreeing with patients upon a treatment regimen (mean: 4.58, CI95%: 4.48-4.68). Female respondents reported significant higher scores compared to males with respect to item 2 (Patient giving information on what is important to her/him, P = 0.0179), item 4 (Patient deciding on their treatment without the clinician, P = 0. 0031) and item 7 (Engaging with patients’ family members or caregivers, P = 0.000).

Finally, the survey investigated the perceived barriers and facilitators of SDM using a 1-4 scale (a non-neutral scale was used to accrue respondents’ preferences). Females reported higher agreement scores to any SDM-related items than male responders. For some items, the gender difference was significant.

Sensitivity analyses excluding countries with fewer than 10 respondents, reported in Appendix C, revealed consistent patterns of item relevance across gender groups, supporting the validity of the associations identified in the overall sample. To explore whether cultural factors associated with different countries might influence shared decision-making (SDM) practices, additional analyses were conducted by clustering responses according to the type of healthcare system, namely tax-based, insurance-based, or mixed systems; however, no relevant relationship across groups were detected (see Appendix D).

Distinct ordered probit regressions confirmed the role of gender in changing in the predicted probability of the dependent variable being in a higher category for each one-unit increase in the independent variable, holding all other predictor variables constant. Table 3 reports the effect of professional role, gender, seniority, “shared” decision-making archetype, and type of BU on selected items (1-4) related to decision-making process and factors. Supplemental pairwise comparisons for every predictor/outcome combination are reported in Appendix E.

Table 3.

Ordered Probit Regression Results for Four [(1)-(4)] Decision-Making Related Outcomes: Effect of Professional Role, Gender, Seniority, Decision-Making Archetype, and Type of Breast Unit

	Dependent variables
	(1)	(2)	(3)	(4)
Professional role	0.130 (0.119	0.113 (0.100)	0.137 (0.102)	0.061 (0.102)
Gender	0.523*** (0.193)	0.473*** (0.167)	0.717*** (0.169)	0.514*** (0.171)
Seniority	0.019** (0.009)	−0.003 (0.007)	−0.007 (0.007)	0.001 (0.008)
“Shared” decision-making archetype^a	0.436*** (0.085)	0.201** (0.079)	0.238*** (0.076)	0.23*** (0.077)
Type of BU	0.400*** (0.125)	−0.147 (0.099)	0.076 (0.102)	−0.036 (0.103)
Observations	189	189	189	189
Pseudo R²	0.135	0.033	0.068	0.044
Prob > chi2	0.000	0.003	0.000	0.002

Abbreviations: SDM = Shared Decision-Making; BU = Breast Unit.

Notes: Entries are ordered-probit coefficients; robust standard errors are shown in parentheses. Significance levels: *P < 0.10, **P < 0.05, ***P < 0.01.

Dependent variables (column headers).

(1) Patient giving information on what is important to her/him.

(2) Patient deciding on their treatment without the clinician.

(3) Engaging with patients’ family members or caregivers.

(4) Patients need DAs (leaflets, videos, apps) to engage in their health decisions.

^aPredictor “Shared decision-making archetype”: Assessed with the vignette from Charles at al. (2004): “After reviewing the medical records and examining the patient, I present the available treatment options. Information about risks and benefits of each option is discussed. I invite the patient to ask any question. Then I ask her/his preference for a treatment given her/his lifestyle and the issues that are important to her/him. Together we decide on a treatment to implement”.

Patient Decision Aids

Table 4 reports the descriptive statistics of the availability and use of DAs. DAs were said to be available as a support to treatment decisions by more than half of the respondents (N = 109, 55%). In terms of types of DAs, paper-based tools were by far the most common type. These can include any kind of support provided to patients in paper formats, such as booklets, brochures, workbooks, or letters from cancer survivors. Conversation aids, namely simple communication tools intended to support clinicians ensuring that patients’ information needs are appropriately met (eg, question prompt lists), were the second-most used tool across survey respondents (N = 41, 44%). Coaching sessions with patients, consisting in sharing information about possible treatment courses by trained nurses, cancer survivors, patient advocates, or different stakeholders, were also recurrent. Surprisingly, digital, web-based, interactive tools, such as online learning modules, mobile apps, websites, or risk calculators, were reported to be used by less than one-fourth of overall number of respondents. Allowing the patient to familiarize with the surgical implant (touch and see) or making personalized drawings and sketches that patients could bring home, were provided as additional examples of DAs by respondents in the open text field.

Table 4.

Availability and Use of DAs (Distribution of Responses)

N (%)	Total	Females	Males
Availability of DAs
Yes	109 (55%)	65 (56%)	39 (54%)
No/I don’t know	89 (45%)	52 (44%)	33 (46%)
Total	198 (100%)	117 (100%)	72 (100%)
Use of DAs
Yes	94 (86%)	57 (88%)	33 (85%)
No	15 (14%)	8 (12%)	6 (15%)
Total	109 (100%)	65 (100%)	39 (100%)
DAs used
Paper-based tools	72 (77%)	44 (77%)	25 (76%)
Conversation aids	41 (44%)	23 (40%)	14 (42%)
Coaching sessions	38 (40%)	25 (44%)	11 (33%)
Web-based tools	23 (24%)	14 (25%)	7 (21%)
Option grids	16 (17%)	8 (14%)	7 (21%)
Video	8 (9%)	2 (4%)	6 (18%)
Others	5 (5%)	3 (5%)	1 (3%)

Abbreviations: DAs = Patient Decision Aids.

The survey also investigated when DAs were used, as well as frequency of usage (Table 5). DA could be employed before, during, after the consultation with the clinicians, or in a combination of moments. To assess usage frequency, the number of patients provided with the DAs out of a 10-patient hypothetical sample was used as a proxy. The data suggests that it is more common to distribute paper-based tools to a larger number of patients (ie, 6-10 patients; N = 53, 74%), compared to web-based tools, that are typically shared with smaller groups of patients (ie, 1-5 patients; N = 16, 70%).

Table 5.

Moment of DA Use and Usage Frequency (Distribution of Responses, Total Sample)

	Moment of use							Usage frequency
N (%)	Before	During	After	During & after	Before & during	Before & after	Before & during & after	1-5 patients	6-10 patients
Paper-based tools	8 (11%)	27 (38%)	20 (28%)	11 (15%)	3 (4%)	-	3 (4%)	19 (26%)	53 (74%)
Conversation aid	5 (12%)	30 (73%)	4 (10%)	1 (2%)	-	-	1 (2%)	15 (37%)	26 (63%)
Coaching sessions	6 (16%)	5 (13%)	18 (47%)	5 (13%)	-	1 (3%)	3 (8%)	15 (39%)	23 (61%)
Web-based tools	4 (17%)	9 (39%)	6 (26%)	1 (4%)	2 (9%)	-	1 (4%)	16 (70%)	7 (30%)
Option Grid	1 (6%)	13 (81%)	1 (6%)	1 (6%)	-	-	-	10 (63%)	6 (38%)
Video	1 (13%)	5 (63%)	2 (25%)	-	-	-	-	6 (75%)	2 (25%)
Other	-	3 (60%)	-	1 (20%)	-	1 (20%)	-	1 (20%)	4 (80%)

Notes: Percentages are calculated out of the total number of observations in each DA category.

Clinicians’ responses to the question What aspects do you think are relevant for a patient DA? are presented in Table 6. A 1-5 scale was used to provide a two-fold perspective of both users and potential users, stratified by gender.

Table 6.

Perceived Enablers to the Uptake of DAs

Mean (SD)¹		Total	Females	Males	P-value
Time of usage	Users	3.75 (0.92)	3.81 (0.83)	3.63 (1.06)
Time of usage	Potential users	3.90 (0.81)	3.91 (0.70)	3.87 (0.99)
Layout and interactivity	Users	3.63 (0.89)	3.82 (0.73)	3.34 (1.03)	*
Layout and interactivity	Potential users	-	-	-	-
Accessibility in multiple formats and languages	Users	3.60 (1.12)	3.89 (0.86)	3.09 (1.29)	**
Accessibility in multiple formats and languages	Potential users	3.97 (0.83)	4.18 (0.75)	3.68 (0.91)	*
Evidence-based information	Users	4.09 (0.91)	4.23 (0.76)	3.86 (1.12)
Evidence-based information	Potential users	4.29 (0.85)	4.47 (0.76)	4.03 (0.95)	*
Integration in the workflow/eHR	Users	3.41 (1.09)	3.56 (1.00)	3.14 (1.24)
Integration in the workflow/eHR	Potential users	3.94 (0.84)	4.11 (0.71)	3.68 (0.98)	*

Abbreviations: eHR = Electronic Health Records; DA: decision aids; SD: standard deviation.

Notes: Statistical significance is indicated with: * for P-values ≤0.05, ** for P-values ≤0.01, *** for P-values ≤0.001. P-values are calculated using Mann Whitney U test. On a 1-5 Likert-scale where 1 = Not at all relevant and 5 = Extremely relevant.

In terms of DA quality, leveraging on evidence-based information had the highest score both amongst current users of DAs (mean: 4.09, CI95%: 3.91-4.27), and potential users (mean: 4.29, CI95%: 4.10-4.48), whereas the factors that are perceived less important are the integration within the hospital IT system (mean: 3.41, CI95%: 3.19-3.63) and time of usage (mean: 3.90, CI95%: 3.72-4.08) for current users and potential users, respectively. In each item analyzed, self-reported values of current users are lower compared to the corresponding self-reported values of potential users.

Lastly, Figure 2 displays the perceived barriers to the uptake of DAs in the clinical practice from the non-user perspective (ie, specialists who declared not to use DAs even when in principle they were available to them, and specialists employed in centers where DAs were not provided). Interestingly, amongst respondents who could use available tools to facilitate treatment decisions, 9 (60%) nevertheless adopt different possibly unstructured or uncodified strategies, to engage their patients. The percentage of non-users is higher than the corresponding share of users in most of the question items. This gap in self-reported values is particularly relevant with respect to the need of an organized system to distribute the DAs within an organization (N = 6, 46% vs N = 3, 20%), and training for an accurate use of the DAs (N = 4, 31% vs N = 1, 7%).

Figure 2.

Perceived Barriers to the Uptake of DAs

Discussion

In an era of sustained commitment to patient-centricity at every step of the health care pathway, this survey investigated clinicians’ attitudes toward SDM and current use of DA in the area of BC.²⁹ The findings highlight a vigorous interest for participatory communication approaches with patients. Notably, our results suggest that the self-reported inclination to genuinely adopt a participatory and inclusive communication style is not randomly distributed among professionals, but is more often observed among female clinicians and those who, in line with the SDM archetype proposed by Charles et al,²⁶ demonstrate a greater tendency to engage patients as active partners in the decision-making process. This pattern was consistent across all dependent variables analyzed, indicating a broader alignment between individual orientations toward SDM and actual behaviors. While the intensity of DAs usage can vary based on the gender, DAs appear to be usable in different moments, ranging from first visits to follow-ups. On one hand, the availability of DAs seems to translate into actual use; on the other hand, professionals who do not have access to DAs within their organization still exhibit a strong inclination to use these instruments. DAs are accessible in multiple formats, with paper tools emerging as the most common in practice. Overall, the survey illustrates a promising landscape for DAs, yet a few specialists who were surveyed expressed a preference for not using these tools despite their availability. In such cases, long-established, tacit strategies for patient engagement still prevail and represent a latent obstacle to the wider diffusion of more sophisticated, evidence-based instruments. Lastly, female respondents appeared better inclined to use DAs in the clinical encounters, suggesting that gender of treating physician might play a role in the acceptance and use of these tools. Although this evidence is only reflective of the self-reported perceptions of respondents, this survey seems to suggest that there is not a one-size-fits-all approach with respect to how DAs are currently employed with patients.

Through widespread dissemination with BC networks, this study contributed to increasing awareness of what factors might influence SDM among health practitioners and has the potential to stimulate further adoption of SDM practices in the future. Yet, the fact that the literature on SDM and the use of DAs in BC care is more extensively developed in North America, particularly the United States (US), suggests that cultural factors, such as norms, values, and underlying assumptions, may influence how clinicians engage with patients and, ultimately, how health technologies are adopted in practice.²⁸ Indeed, several studies from the US have shown that structured policies, reimbursement mechanisms, and legal mandates have supported the systematic integration of SDM and DAs into clinical practice, especially in oncology. For example, the implementation of the Patient Protection and Affordable Care Act in the US explicitly emphasized patient engagement and SDM (establishing a program “to provide for the phased-in development, implementation, and evaluation of SDM using patient decision aids to meet the objective of improving the understanding of patients of their medical treatment options”),³⁰ leading several health organizations to adopt institutional SDM requirements. In contrast, many European healthcare systems lack unified policies or incentives supporting SDM, with substantial variation across countries in the degree to which SDM is embedded in national cancer strategies or clinical guidelines. Moreover, specific features of European healthcare systems, such as varying degrees of centralization or differences in primary vs specialist care pathways, may contribute to inconsistent implementation of DAs. Efforts to scale up SDM practices across Europe may benefit from coordinated strategies that combine policy incentives, clinician training, and integration of DAs into clinical workflows and national care standards.

In general, applying research findings in real-world settings can be challenging, and there may be difficulties in adopting new clinical innovations into routine practice.³¹ The underlying factors facilitating or impeding the adoption of DAs in BC have been synthetized in a prior review of the literature under the lens of a consolidated implementation science framework,³² concluding that knowledge to support the sustained implementation of SDM interventions in BC care was still limited.²⁵ Building on this, this survey interestingly highlighted that many of the proposed barriers to DAs uptake were viewed as obstacles more by individuals who did not use nor have the tools, possibly suggesting that the absence of access or usage of these instruments may result in an overestimation of the challenges involved in their use.

Based on these findings, we suggest that healthcare administrators prioritize integrating SDM practices and DAs into routine workflows by embedding them in eHRs, supporting staff training on their effective use, and allocating protected time during patients’ visits for SDM conversations. From a policy perspective, national and regional health authorities could develop formal guidelines and incentives that recognize SDM as a quality-of-care indicator, particularly in oncology. Moreover, ensuring equitable access to validated, culturally adapted DAs across institutions, regardless of size or geographic location, could reduce variation in practice and promote patient empowerment. Investments in education and awareness campaigns targeted at both clinicians and patients may also facilitate a shift in attitudes and organizational culture, fostering greater acceptance and normalization of SDM across the care continuum.

Contribution to the Literature

To our knowledge, this is the first survey extensively investigating clinician attitudes towards SDM and the uptake of DAs in the field of BC at the European level. Previous studies had a narrower focus in terms of perspective of analysis and geographical scope. A survey by Maes-Carballo et al (2021) only investigated the knowledge, awareness and attitude towards SDM amongst Spanish-speaking clinicians from Spain and Argentina,³³ not covering the use of DAs. A recently published study using focus group methodology investigated Portuguese general practitioners’ awareness and perception of PDAs.³⁴ Other published surveys investigated the level of SDM from the perspective of BC patients, as is the case of a German cross-sectional study in which BC patients who underwent surgery evaluated SDM using Shared Decision Making Questionnaire (SDM-Q-9).^35,36

By revealing that the use of DAs with female BC patients might be influenced by the gender of treating physicians, this work also contributes to a growing body of literature that examines the impact of patient-clinician gender-concordance on patient outcomes.¹³ A previous study showed that gender affects the relation between specific nonverbal behaviors and satisfaction, concluding that, for female physicians, the latter was related to female gender stereotypical nonverbal behavior (eg, more gazing at patient, less interpersonal distance, softer voice, less looking at medical chart).³⁷ Furthermore, for female physician–female patient consultations, doctor interruptions were positively related to patient satisfaction, whereas in male physician–male patient consultations interruptions worsened satisfaction levels.³⁸ A recent analysis of more than 20 000 US adult respondents to the 2018 and 2020 Medical Expenditure Panel Survey found that women who saw a female usual clinician were 6 – 9 percentage points more likely to be up-to-date with influenza, pneumococcal and shingles vaccination as well as breast, cervical and colorectal cancer screening; the same sex-concordance pattern was not observed among male patients, hinting at gender-specific mechanisms according to which females with female physician have a higher likelihood to adhere to preventive services guidelines.³⁹ However, the underlying factors explaining why patient-provider concordance might impact on patient experience and outcomes have been less extensively explored in the literature. In this area, another recent mixed-methods study investigated how patient-clinician gender concordance relates to trust, concluding that gender concordance showed no consistent effect, with participants often indicating that rapport, communication style and feeling respected mattered more than sharing the clinician’s gender.⁴⁰ Building on these qualitative insights. The statistically significant gender differences observed in our survey study with respect to SDM, namely the use of DA to explain clinical contents or the stronger tendency to involve caregivers, contributed to shed some light in this regard, offering concrete behavioral pathways through which gender concordance may shape patient experience and outcomes.

Limitations

This study comes with limitations. The overall number of responses collected is relatively small. Furthermore, some countries were only sporadically represented in the survey, and the category “Other countries” grouped together a heterogeneous mix of contexts, further limiting the granularity of country-level interpretation. This affects the representativeness of the findings across all European BC centers. However, the results of this survey were treated descriptively and interpreted with an exploratory approach, and should therefore be considered as indicative rather than definitive. Additionally, online surveys are susceptible to selection bias, as certain population subgroups may be difficult to reach (eg, older respondents).⁴¹ Social desirability bias can influence survey scores, as respondents may answer questions based on what is socially expected rather than their true attitudes toward patient-physician relationship style and SDM. To minimize the impact of these biases, the study purpose was disclosed, and anonymity of the responses was ensured.⁴² The survey was administered in English, which could have posed a challenge for respondents from different nationalities and backgrounds in accurately interpreting the survey questions. To mitigate the risk of collecting inconsistent responses, a cross-country cognitive testing exercise was carried out to validate the survey structure and items. Lastly, this survey illustrates self-reported perceptions from the clinicians’ perspective, and it is not informative of the actual diffusion of DAs in Europe. The results do not reflect the specificities of any country nor of the underlying health systems, and as such should be interpreted only as exploratory and with caution. Nonetheless, the study contributes valuable insights into shared decision making in Europe, a topic that remains relatively underexplored.

Conclusions

Patient-centricity calls for accurate consideration and reflection of patients’ values, preferences and choices at every step in the treatment pathway. The progress in BC care has revived interest in SDM, given to a broad spectrum of alternative therapeutic options available to patients, and the potential for better patient outcomes and improved quality of care arising from treatment plans aligned with patient values. Published literature suggests that participatory approaches and the use of DAs to support treatment decisions have been more extensively studied and implemented in North America. Nonetheless, the level of interest in SDM demonstrated by the BC specialists in this study points to a promising outlook for the continued advancement of SDM practices in Europe. Future research should investigate patient-perspective studies, prioritizing the efforts towards implementation trials evaluating the real-world effectiveness of DAs.⁴³ Notably, there are promising ongoing initiatives in this regard, like the EU-funded CINDERELLA project that is assessing the impact of an AI platform to support SDM in locoregional breast cancer treatment.^44,45 Future work should therefore clarify which factors can lead to the sustained adoption of SDM practices and DAs in clinical settings, with particular attention to barriers and facilitators linked individual characteristics – including physician gender-related differences and patient-clinician gender-concordance, which may shape communication styles, decision-making dynamics, and openness to shared decision-making, as observed in this study – so as to move toward increasingly patient-centric healthcare systems.

Supplemental Material

Supplemental material - Evidence on Patient Decision Aids and Shared Decision-Making across Breast Centers in Europe: A Cross-Sectional Survey Study

Supplemental material for Evidence on Patient Decision Aids and Shared Decision-Making across Breast Centers in Europe: A Cross-Sectional Survey Study by Vittoria Ardito, MSc, Natalia Oprea, PhD, Oriana Ciani, PhD in Cancer Control.

Footnotes

Acknowledgements

The author would like to thank the networks of breast cancer professionals involved in disseminating this survey.

ORCID iDs

Vittoria Ardito

Natalia Oprea

Ethical Considerations

The authors confirm that all methods were performed in accordance with the relevant guidelines and regulations. In November 2021, the Ethical Committee of Bocconi University (Milan, Italy) approved the survey protocol (FA000375).

Consent to Participate

All survey participants signed informed consent and agreed to the publication of the study results.

Authors contribution

Vittoria Ardito: Conceptualization, data curation, formal analysis, methodology, writing – original draft, writing – review and editing, validation. Natalia Oprea: Conceptualization, data curation, methodology, writing – review and editing, validation. Oriana Ciani: Conceptualization, data curation, methodology, writing – review and editing, validation, funding acquisition.

Funding

This study was conducted as part of the ShareView project, “Supporting shared decision-making and communication in metastatic breast cancer”, funded by Pfizer Global Medical Grants in partnership with Sharing Progress in Cancer Care (SPCC), and the CINDERELLA project, which was funded by the European Union (EU grant HORIZON-HLTH-2021-DISEASE-04. Project number 101057389). The funder did not have a role in study design, data collection and analysis, decision to publish, or preparation of this manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Survey data will be made available upon request to the corresponding author.*

Supplemental Material

Supplemental material is available online.

Appendix

References

Pizzoli

SFM

Renzi

Arnaboldi

Russell-Edu

Pravettoni

. From life-threatening to chronic disease: is this the case of cancers? A systematic review. In: Monacis L, ed. Cogent Psychology. 2019;6(1):1577593. doi:10.1080/23311908.2019.1577593

Veenstra

Hawley

. Incorporating patient preferences into cancer care decisions: challenges and opportunities. Cancer. 2020;126(15):3393-3396. doi:10.1002/cncr.32959

Seidman

Masi

Gomez-Rexrode

. Personalizing value in cancer care: the case for incorporating patient preferences into routine clinical decision making. J Particip Med. 2019;11(3):e13800. doi:10.2196/13800

Stacey

Lewis

Smith

, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Consumers and Communication Group. Cochrane Database Syst Rev. 2024;2024(1):cd0014331. doi:10.1002/14651858.cd001431.pub6

Bae

. Shared decision making: relevant concepts and facilitating strategies. Epidemiol Health. 2017;39:e2017048. doi:10.4178/epih.e2017048

Elwyn

Frosch

Thomson

, et al. Shared decision making: a model for clinical practice. J Gen Intern Med. 2012;27(10):1361-1367. doi:10.1007/s11606-012-2077-6

European code of cancer practice. https://www.europeancancer.org/2-standard/66-european-code-of-cancer-practice

Chichua

Brivio

Mazzoni

Pravettoni

. Shared decision-making and the lessons learned about decision regret in cancer patients. Support Care Cancer. 2022;30(6):4587-4590. doi:10.1007/s00520-021-06725-5

Han

Jungsuwadee

Abraham

. Shared decision-making and women’s adherence to breast and cervical cancer screenings. IJERPH. 2018;15(7):1509. doi:10.3390/ijerph15071509

10.

Josfeld

Keinki

Pammer

Zomorodbakhsch

Hübner

. Cancer patients’ perspective on shared decision-making and decision aids in oncology. J Cancer Res Clin Oncol. 2021;147(6):1725-1732. doi:10.1007/s00432-021-03579-6

11.

Carmona

Crutwell

Burnham

Polak

Guideline Committee . Shared decision-making: summary of NICE guidance. Br Med J. 2021;17:n1430. doi:10.1136/bmj.n1430. Published online June.

12.

Iwata

Saji

Ikeda

, et al. Correction to: the Japanese Breast Cancer Society Clinical Practice Guidelines, 2018 edition: the tool for shared decision making between doctor and patient. Breast Cancer. 2021;28(4):987. doi:10.1007/s12282-021-01253-w

13.

Otte

. Improved patient experience and outcomes: is patient–provider concordance the key? J Patient Exp. 2022;9:237437352211030. doi:10.1177/23743735221103033

14.

Wallis

CJD

Jerath

Coburn

, et al. Association of surgeon-patient sex concordance with postoperative outcomes. JAMA Surg. 2022;157(2):146-156. doi:10.1001/jamasurg.2021.6339

15.

Wallis

Ravi

Coburn

Nam

Detsky

Satkunasivam

. Comparison of postoperative outcomes among patients treated by male and female surgeons: a population based matched cohort study. Br Med J. 2017;10:j4366. doi:10.1136/bmj.j4366. Published online October.

16.

Charles

Whelan

Gafni

Willan

Farrell

. Shared treatment decision making: what does it mean to physicians? J Clin Orthod. 2003;21(5):932-936. doi:10.1200/JCO.2003.05.057

17.

Stacey

Volk

for the IPDAS Evidence Update Leads Bekker

, et al. The international patient decision aid standards (IPDAS) collaboration: evidence update 2.0. Med Decis Mak. 2021;41(7):729-733. doi:10.1177/0272989X211035681

18.

Ankolekar

Dekker

Fijten

Berlanga

. The benefits and challenges of using patient decision aids to support shared decision making in health care. JCO Clin Cancer Inform. 2018;2:1-10. doi:10.1200/CCI.18.00013

19.

Von

Altman

Egger

Pocock

Gøtzsche

Vandenbroucke

STROBE Initiative . The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344-349. doi:10.1016/j.jclinepi.2007.11.008

20.

Burns

KEA

Duffett

Kho

, et al. A guide for the design and conduct of self-administered surveys of clinicians. Can Med Assoc J. 2008;179(3):245-252. doi:10.1503/cmaj.080372

21.

Willis Gordon

Lessler

Judith T.

(1999). Question Appraisal System (QAS-99). Research Triangle Institute. Rockville, MD: Research Triangle

Institute. August

1999.

22.

APA style guidelines. https://apastyle.apa.org/style-grammar-guidelines/bias-free-language/gender

23.

Willis

. Cognitive Interviewing. Sage Publications, Inc.; 2005. doi:10.4135/9781412983655

24.

Spinnewijn

Aarts

Braat

Scheele

. Unravelling clinicians’ shared decision-making adoption: a qualitative exploration through the lens of diffusion of innovations theory. BMJ Open. 2024;14(6):e080765. doi:10.1136/bmjopen-2023-080765

25.

Oprea

Ardito

Ciani

. Implementing shared decision-making interventions in breast cancer clinical practice: a scoping review. BMC Med Inf Decis Making. 2023;23(1):164. doi:10.1186/s12911-023-02263-8

26.

Charles

Gafni

Whelan

. Self-reported use of shared decision-making among breast cancer specialists and perceived barriers and facilitators to implementing this approach. Health Expect. 2004;7(4):338-348. doi:10.1111/j.1369-7625.2004.00299.x

27.

Emanuel

. Four models of the physician-patient relationship. JAMA. 1992;267(16):2221-2226.

28.

Charles

Gafni

Whelan

. Shared decision-making in the medical encounter: what does it mean? (or it takes at least two to tango). Soc Sci Med. 1997;44(5):681-692. doi:10.1016/S0277-9536(96)00221-3

29.

Ciani

Ardito

Oprea

. Shared Decision-Making (Chapter 24; page: 445-454). Handbook of Generalized Pairwise Comparisons: Methods for Patient-Centric Analysis. Curated by Marc Buyse, Johan Verbeeck, Everardo D. Saad, Mickaël De Backer, Vaiva Deltuvaite-Thomas, Geert Molenberghs. Chapman and Hall/CRC; 2025.

30.

The affordable care Act. https://www.hhs.gov/healthcare/about-the-aca/index.html

31.

Kirchner

Smith

Powell

Waltz

Proctor

. Getting a clinical innovation into practice: an introduction to implementation strategies. Psychiatry Res. 2020;283:112467. doi:10.1016/j.psychres.2019.06.042

32.

Feldstein

Glasgow

. A practical, robust implementation and sustainability model (PRISM) for integrating research findings into practice. Joint Comm J Qual Patient Saf. 2008;34(4):228-243. doi:10.1016/s1553-7250(08)34030-6

33.

Maes-Carballo

Martín-Díaz

Mignini

Khan

Trigueros

Bueno-Cavanillas

. Evaluation of the use of shared decision making in breast cancer: International survey. IJERPH. 2021;18(4):2128. doi:10.3390/ijerph18042128

34.

Proença-Portugal

Heleno

Dias

Gama

Baptista

. General practitioners’ perceptions on decision aids in healthcare: a qualitative study in Portugal. BMC Med Inf Decis Making. 2025;25(1):202. doi:10.1186/s12911-025-03044-1

35.

Kriston

Scholl

Hölzel

Simon

Loh

Härter

. The 9-item Shared Decision Making Questionnaire (SDM-Q-9). Development and psychometric properties in a primary care sample. Patient Educ Counsel. 2010;80(1):94-99. doi:10.1016/j.pec.2009.09.034

36.

Feiten

Scholl

Dünnebacke

, et al. Shared decision making in routine breast cancer care in Germany—a cross sectional study. Psychooncology. 2022;31(7):1120-1126. doi:10.1002/pon.5898

37.

Mast

. On the importance of nonverbal communication in the physician–patient interaction. Patient Educ Counsel. 2007;67(3):315-318. doi:10.1016/j.pec.2007.03.005

38.

Hall

Irish

Roter

Ehrlich

Miller

. Gender in medical encounters: an analysis of physician and patient communication in a primary care setting. Health Psychol. 1994;13(5):384-392. doi:10.1037/0278-6133.13.5.384

39.

Green

Rodriquez

Nápoles

Pérez-Stable

Strassle

. Patient-clinician sex and race and/or ethnicity concordance and adherence to preventive services guidelines: meps 2018–2020. J Gen Intern Med. 2025. doi:10.1007/s11606-025-09631-2. Published online June 3.

40.

Greene

Silver

Verrier

Long

. Is patients’ trust in clinicians related to patient-clinician racial/ethnic or gender concordance? Patient Educ Counsel. 2023;112:107750. doi:10.1016/j.pec.2023.107750

41.

Bethlehem

. Selection bias in web surveys: selection bias in web surveys. Int Stat Rev. 2010;78(2):161-188. doi:10.1111/j.1751-5823.2010.00112.x

42.

Gordon

. Social desirability bias: a demonstration and technique for its reduction. Teach Psychol. 1987;14(1):40-42. doi:10.1207/s15328023top1401_11

43.

Henderson

Soltani

Suckow

, et al. AMPDECIDE amputation level patient decision aids: a feasibility study. BMC Med Inf Decis Making. 2025;25(1):218. doi:10.1186/s12911-025-03084-7

44.

Borsoi

Listorti

Ciani

the CINDERELLA Consortium . Artificial-Intelligence cloud-based platform to support shared decision-making in the locoregional treatment of breast cancer: protocol for a multidimensional evaluation embedded in the CINDERELLA clinical trial. Pharmacoecon Open. 2024;8(6):945-959. doi:10.1007/s41669-024-00519-1

45.

Kaidar-Person

Antunes

Cardoso

, et al. Evaluating the ability of an artificial-intelligence cloud-based platform designed to provide information prior to locoregional therapy for breast cancer in improving patient’s satisfaction with therapy: the CINDERELLA trial. In: Akand M, ed. PLoS One. 2023;18(8):e0289365. doi:10.1371/journal.pone.0289365

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.25 MB

Evidence on Patient Decision Aids and Shared Decision-Making Across Breast Centers in Europe: A Cross-Sectional Survey Study

Abstract

Introduction

Methods

Results

Conclusions

Keywords

Introduction

Background

Objectives

Methods

Survey Development

Step 1: Item generation

Step 2: Survey draft

Step 3: Cognitive testing

Step 4. Final questionnaire

Recruitment and Data Collection

Data Analysis

Results

Participants’ Profiling

Decision-Making Archetypes, Process and Influencing Factors

Patient Decision Aids

Discussion

Contribution to the Literature

Limitations

Conclusions

Supplemental Material

Supplemental material - Evidence on Patient Decision Aids and Shared Decision-Making across Breast Centers in Europe: A Cross-Sectional Survey Study

Footnotes

Acknowledgements

ORCID iDs

Ethical Considerations

Consent to Participate

Authors contribution

Funding

Declaration of Conflicting Interests

Data Availability Statement

Supplemental Material

Appendix

References

Supplementary Material