Physician‐related facilitators and barriers to patient involvement in treatment decision making in early stage breast cancer: perspectives of physicians and patients

Abstract

Objective  To identify patients’ and physicians’ perceptions of physician‐related verbal and nonverbal facilitators and barriers to patient involvement in treatment decision making (TDM) occurring during clinical encounters for women with early stage breast cancer (ESBC).

Methods  Eligible women were offered treatment options including surgery and adjuvant therapy. Eligible physicians provided care for women with ESBC in either a teaching hospital or an academic cancer centre. In Phase 1, women were interviewed 1–2 weeks after their initial consultation. In Phase 2, women and their physicians were interviewed separately while watching their own consultation on a digital video disk. All interviews were audiotaped, transcribed and analysed.

Results  Forty women with ESBC and six physicians participated. Patients and physicians identified thirteen categories of physician facilitators of women’s involvement. Of these, seven categories were frequently identified by women: conveyed a rationale for patient involvement in TDM; explained the risk of cancer recurrence; explained treatment options; enhanced patient understanding of information; gave time for TDM; offered a treatment recommendation; and made women feel comfortable. Physicians described similar information‐giving facilitators but less often mentioned other facilitators. Few physician barriers to women’s involvement in TDM were identified.

Conclusions  Women with ESBC and cancer physicians shared some views of how physicians involve patients in TDM, although there were important differences. Physicians may underestimate the importance that women’s place on understanding the rationale for their involvement in TDM and on feeling comfortable during the consultation.

Introduction

Decision making in the oncology context is often complex. Many treatments are accompanied by modest potential benefits and significant expected side effects. Patients may live with serious physical and psychosocial side effects that impact their well‐being. ¹ , ² Adding to the complexity, patients may be offered more than one treatment option at each treatment phase (surgery, adjuvant systemic or radiation therapy).

The majority of patients with cancer wish to obtain information about their treatment options and to participate in making treatment decisions to some extent. ³ , ⁴ , ⁵ , ⁶ Charles et al. ⁷ suggest that there are different stages of treatment decision making (TDM), and that involvement of the patient and their physician may vary depending on the decision‐making model (paternalistic, shared and informed) adopted in any given medical encounter. They also note that these models are ‘ideal’ types, and that there may be ‘in‐between’ models that can be adopted. ⁷

Evidence from the decision aid (DA) literature suggests that involvement in TDM is beneficial to patients. ⁸ , ⁹ , ¹⁰ Benefits include increased knowledge of treatment options and increased satisfaction with TDM. ⁸ , ⁹ , ¹⁰ However, research findings also indicate that some patients with cancer fail to achieve their desired level of participation in TDM. ⁶ , ¹¹ , ¹²

The challenges of physician–patient communication in a cancer setting have been well documented. ¹³ , ¹⁴ Physicians can play an important role in helping patients to participate in TDM. Recent systematic reviews by Gravel et al. ¹⁵ and Légaréet al. ¹⁶ reported key barriers and facilitators to shared decision making as perceived by health professionals. Shared decision making was defined broadly to include ‘the active participation of patients in decision making’. ¹⁶ Lack of clinician time was cited as a barrier across all studies. One of the most common facilitators identified was health professional motivation to engage in a shared process. ¹⁶ Légaré used a previously published framework ¹⁷ to classify barriers and facilitators in terms of health professional knowledge, attitudes and behaviours. However, there may be other types of facilitators and barriers such as those pertaining to the cultural background of both the patient and the physician. ¹⁸

The primary objective of this study was to identify both patients’ and cancer physicians’ perceptions of ‘physician‐related’ verbal and nonverbal facilitators and barriers to the involvement in TDM by women with early stage breast cancer (ESBC). This information is important for two reasons. First, there may be a mismatch between patients’ and physicians’ perceptions of what constitutes important facilitators and barriers. Second, such information could inform the development of training programmes or processes of care to promote the involvement of patients with cancer in TDM.

We chose to study women with breast cancer because previous studies suggest that these women often wish some involvement in making decisions about their care. ⁶ , ¹¹ Moreover, within both the surgical and adjuvant setting, women may be offered more than one treatment option. ¹⁹ , ²⁰ , ²¹ , ²²

This study is part of a larger research initiative that examined the perspectives of women with ESBC regarding their involvement in TDM. Our study results pertaining to women’s perceptions of their actual involvement in stages of TDM have been previously published. ²³

Methods

The design was a qualitative descriptive study. ²⁴ This design is appropriate when the goal is to explore and describe experiences and meanings attributed by participants to a particular process they are engaging in (in this case, TDM) in a naturalistic setting (the patient–physician consultation). ²⁴ Secondly, we wished to present participants’ views of physician facilitators and barriers, while staying close to the language that participants used in their descriptions. ²⁴

Study setting and participants

Patients were recruited from two teaching hospitals and a regional academic cancer centre in Hamilton, Canada. Women with ESBC were eligible for inclusion if they were referred for a consultation regarding surgical or adjuvant therapy options, had sufficient English skills to participate in a 1–2‐h interview and gave informed consent. The treating physician affirmed that there was more than one active treatment option. Consecutive women meeting the inclusion criteria were approached. To ensure that study women were similar to those usually seen in the study settings, we sampled across the following categories: age (under 40, 40–49, 50–64, 65 years and older), stage (Stages 1, 2, 3A ²⁵ ), risk of recurrence (moderate and high) and type of treatment (surgical and adjuvant).

Physicians were eligible to participate if they practiced at the study sites and provided treatment for women with ESBC. As part of their routine practice, most medical oncologists used an online calculator to determine the risk of cancer recurrence called Adjuvant online http://www.adjuvantonline.com. A computerized printout of the estimated risk of recurrence with and without adjuvant therapy tailored to the patient’s age and tumour characteristics was shown to the patient. Two physicians routinely used a decision board, a type of DA during the consultation (six patients). ²⁶ , ²⁷ In addition, a randomized trial comparing different versions of a DA ²⁸ was conducted at the same setting. Six patients in this study participated in the DA trial, either in a previous consultation (three patients) or concurrent with this study (three patients).

Recruitment

Patient recruitment and interviews occurred between March 2005 and July 2006. Patients were approached by a nurse or physician, and details were explained by a research assistant. Physicians were asked to participate prior to patients being contacted. All women and physicians provided written informed consent. The study was approved by the local research ethics boards.

Data collection

Data were collected in two study phases. In Phase 1, semi‐structured interviews were conducted with 19 women approximately 1–2 weeks after their initial surgical, medical oncology (MO) or radiation oncology (RO) consultation. These women were interviewed to determine whether they could describe any physician facilitators or barriers to their involvement in TDM. Data from this phase informed the data collection in Phase 2. The ability and willingness of Phase 1 women to identify physician‐related facilitators or barriers occurring during their consultation provided justification for proceeding to the more intensive Phase 2 data collection. In the original design, we intended to analyse these data separately (see analysis). In Phase 2, another group of 21 women and six physicians (surgeons or medical oncologists) were interviewed (separately) regarding their perceptions approximately 1–3 weeks post‐initial consultation, while each participant watched a digital video disk (DVD) recording of his or her own consultation. ²³ The interview guide for both study phases was similar. Patient interviews were 1.5–2.5 h, while those of physicians were 0.5–1.5 h. Most women were interviewed in their home, and most physicians were interviewed in their office. All interviews were audiotaped and transcribed.

Coding and analysis

A coding guide was developed by the research team and refined during the analysis of transcripts. Physician verbal and nonverbal facilitators and barriers of women’s involvement in TDM were inductively derived from the data using the constant comparative method. ²⁹ , ³⁰ , ³¹ Two team members independently coded eight transcripts and compared their results. Discrepancies were resolved through discussion, and an audit trail ³² was kept of how to apply the codes in future transcripts. One analyst coded the remaining transcripts.

Each instance of a reported physician verbal or nonverbal facilitator or barrier to women’s involvement in TDM was coded. Facilitator and barrier codes were organized by study component (Phase 1 or 2), by the type of encounter (surgery or MO) and by type of participant (patient, surgeon or medical oncologist) using MS Excel. These codes were then grouped into categories. For example, the facilitator code ‘the medical oncologist said the patient had time to decide treatment’ from one transcript was listed under the category ‘gave time for decision making’. All codes in each category were examined for similarities and differences by two team members. A new category was assigned when codes did not fit into an existing category. All study transcripts were re‐read, and the categories were checked with the coded data to ensure a good fit between the two. The stages of decision making (information exchange, deliberating about their options and choosing the treatment to be implemented) described by Charles et al. ⁷ were used to further guide the interpretation of the results. After the facilitator and barrier categories had been identified inductively, they were examined for potential fit within one of the TDM stages in the Charles et al. framework. ⁷

In the original design, we intended to analyse patient data from Phases 1 and 2 separately. While there were several differences between the two data set (see results section), we judged that there were sufficient similarities in the categories to justify combining data from each phase. We did not conduct a paired analysis of Phase 2 patient and physician data because we could not predict how many physicians would provide consent, and we did not wish to exclude a patient because her physician did not participate.

Results

Forty women attending a surgical (12), adjuvant systemic therapy (25) or radiation oncology (3) consultation participated in interviews in one of two study phases. Four women refused participation after providing informed consent. One woman was withdrawn by study investigators because of equipment malfunction. Data pertaining to age, type of consultation and relevant clinical findings for women in both study phases are presented in Table 1. Women attending the surgical consultation had received their diagnosis approximately 2 weeks earlier, while women attending the MO consultation had received their diagnosis approximately 12 weeks earlier.

Table 1.

Characteristics of patient participants

	Surgical decision		Adjuvant therapy decision
	Phase 1	Phase 2	Phase 1	Phase 2
Number of women	6	61	13	151
Median age in years (range)	62 (40–70)	61 (48–79)	59 (40–68)	50 (34–69)
			72 (47–74) RT
Treatment choice (number of women)	BCT (5)	BCT (4)	Chemotherapy (10)	Chemotherapy (12)
	MT (1)	MT (2)	RT (3)	AI (3)
Risk of recurrence (number of women)
High‐risk node positive	Unknown2	Unknown2	4	8
High‐risk node negative	Unknown2	Unknown2	4	2
Moderate or low risk	Unknown2	Unknown2	5	5
Completed college/university (number)	0	3	9	7

RT, Radiation Therapy; BCT, Breast conserving therapy; MT, Mastectomy; AI, Aromatase inhibitor.

¹Patients also recalled their surgical treatment decision‐making experiences.

²Nodal status was unknown at the time of the preoperative interview.

Six physicians (two surgeons and four medical oncologists) also participated. There were three female and three male physicians between the ages of 34 and 44 years. Physicians had practiced in the setting between 1 and 11 years.

Thirteen categories of physician facilitators occurring during the consultation were reported (Table 2). In most categories, both patients and physicians identified verbal and nonverbal facilitators. In two categories, ‘the physician conveyed a rationale for decision making by the patient’, and ‘the physician made women feel comfortable’, facilitators were identified more often by patients than by physicians. In contrast, in two other categories (Table 2), facilitators were identified primarily by physicians. There were several differences between the Phase 1 and Phase 2 data. Women in Phase 1 did not identify physician facilitators in four of thirteen categories: assessing patient preference for the type/amount of information, tailoring information to the individual, facilitating question asking and giving take‐home information (Table 2). Two women in Phase 2 did not wish to participate in choosing a chemotherapy treatment, although they wanted information about their options and the physician’s recommendation. These women described feeling impatient, perplexed or distressed when the physician attempted to facilitate their involvement in choosing among treatment options.

Table 2.

Categories of physician facilitating behaviours perceived by women and physicians

Categories1 of facilitator	Identified by patients		Identified by physicians	Potential fit with stage of TDM [7]
The physician	Phase 1 (n = 19)	Phase 2 (n = 21)	Phase 2 (n = 17 interviews)
Assessed patient’s understanding of information acquired prior to the consultation	N	N	Y	Information Gathering
Assessed patient’s preference for type or amount of information	N	Y	Y	Information Gathering
Conveyed rationale for patient involvement in TDM	Y	Y	N	Context for TDM
Explained risk of recurrence/recovery	Y	Y	Y	Information Gathering
Explained treatment options including benefits and risks	Y	Y	Y	Information Gathering/Deliberation
Enhanced the patient’s understanding of information about her tumour or treatment (including use of visual aids)	Y	Y	Y	Information Gathering/Deliberation
Tailored information/treatment to the individual patient	N	Y	Y	Information Gathering/Deliberation
Facilitated patient question asking	N	Y	Y	Information Gathering/Deliberation
Gave take‐home information	N	Y	N	Deliberation
Gave time for decision making	Y	Y	Y	Deliberation
Assessed patient’s preference for treatment	N	N	Y	Deliberation
Gave a treatment recommendation or guidance about treatment options	Y	Y	Y	Deliberation
Made women feel comfortable through interpersonal skills	Y	Y	N	Context for TDM

Y, Yes; N, No; TDM, treatment decision making.

¹The behaviour may have occurred but was not noted unless identified by the patient or physician as a facilitator of patient involvement in TDM.

Women’s perceptions of physician facilitators

Seven of thirteen categories of physician facilitators were frequently described by women in both study phases and are presented in the following section along with quotes from the study participants. Text in square brackets was added by the investigators for clarity or to provide anonymity.

Conveyed the rationale for patient involvement in TDM

Several women thought it was important that the physician conveyed the rationale for their involvement in TDM. They perceived that their physicians did so by explicitly addressing the patient and physician roles in decision making or by using the word choice when describing treatment options. Other women thought that physicians conveyed the reason for patient involvement in TDM implicitly by explaining that the two treatment options were equally effective but differed with respect to side effects. For example, one woman attending a MO consultation said,

Dr. [MO] came in and [s/he] did a little bit of education. General stuff about breast cancer. Very, very sequential, the stages of treatment. Always very, very clear to me that it was my choice. That they don’t make the decision, they present the information. They’re there to ask lots of questions but, the ultimate decision was up to me and my husband.

Explained risk of cancer recurrence

Several women in Phase 1 and most women in Phase 2 commented on the importance of receiving information about the risk of cancer recurrence with and without adjuvant systemic or radiation treatment. Women also talked about the importance of receiving tumour‐related information from the pathology report, which helped them to understand their risk of recurrence. One woman attending a MO consultation said,

[S/he] showed me the stats. I can look at it myself and say… this is the highest possibility of recovery.

Explained treatment options

Closely related to the physician’s explanation of risk of cancer recurrence was the physician’s explanation of treatment options. For women attending the surgical consultation, the options of lumpectomy plus radiation or mastectomy were presented. Most women in Phase 1 and a few women in Phase 2 thought the surgeon helped them to be involved in TDM when s/he discussed their surgical options. Most women in Phase 2 and several women in Phase 1 attending a MO consultation discussed the relevance of receiving information about chemotherapy or hormone therapy options. A woman attending a MO consultation said,

So [s/he] went through that, presenting the choices of chemotherapy and the percentages of, personally how it would affect me. You know, the benefits of it. And then at that point I had a choice between a six‐month chemo or a three‐month chemo. Or no chemo, that was presented, on the board [DA]. And [s/he] didn’t want a decision that day.

Enhanced patient understanding of information

Several women explained that the physician facilitated their participation in TDM by helping them understand information by summarizing, answering questions and by giving them time to process information during the consultation. For example, one woman attending a surgical consultation said,

During the consultation, [s/he] gave me the time. Like [s/he] explained things and then maybe talked to my daughter just for a few moments. And, I suppose [s/he] could probably tell when I was sitting there… Oh, she’s just absorbing that sort of thing, for a few seconds. But I felt that [s/he] gave, like good time and good explanation.

Several women commented that their physician helped them to be involved in TDM by using a visual aid. Some physicians used hand‐drawn or pre‐prepared diagrams, while others used Adjuvant online printouts or the Decision Board and take‐home booklets. One woman attending a MO consultation said,

I also got a booklet that had the exact picture [of what] had been presented on that [Decision Board] – a booklet to read. So I had a week to kind of digest that and we went back and let [him/her] know what we decided.

Gave time for TDM

Many women, particularly those in Phase 2, thought the physician helped them to participate in TDM by explicitly saying that they had time to make a decision. During this study, some women waited approximately 9–12 weeks for a MO consultation. Several women expressed concern about waiting for treatment and felt pressure to make a decision about treatment. One woman said,

And [the medical oncologist] said…you can take time to think about it. [S/he] didn’t push me.

Gave a treatment recommendation or guidance about options

Most women in Phase 2 wanted to hear the physician’s treatment recommendation or opinion about treatment. Some women described the recommendation as giving them confidence in their own decision making. Women wanted to know whether their treatment preference was consistent with that of their physician. If the physician did not directly give a treatment recommendation, women sought to infer it from the order of presentation of options or the choice of words to describe treatment. One woman talked about her discussion with her medical oncologist,

Well, [the medical oncologist said] it’s your choice and [s/he] give me some of [her/his], I guess reservations about the one [option]. Like, based on medical history. And an objective approach I guess. A way of looking at it.

Made women feel comfortable

Several women also mentioned the importance of the physician’s interpersonal skills. These women thought that their involvement in TDM was affected by the relationship between the physician and patient. One woman attending a MO consultation said,

I thought [s/he] was really good, actually. [s/he] could talk easily. [s/he] didn’t find it hard to tell me what the score was. And, you know, it’s your decision but this is what I think. Or what, what usually happens. …And [s/he] didn’t get excited either. Some doctors are very difficult to talk to. And I found [her/him] very easy to talk to.

Women’s perceptions of physician barriers to women’s involvement in TDM

Women described relatively few physician barriers to their involvement in TDM. When barriers were mentioned, there appeared to be two categories of barriers: those that occurred during the consultation, and were negatively perceived, and facilitating behaviours that women thought should have occurred but did not. For example, several women commented that their involvement in TDM was compromised when the physician did not appear interested in their concerns about their disease or its treatment. Two other women perceived that their physician did not invite them to participate in TDM, which negatively affected their involvement. For example, one woman said,

And he has all these papers, [and he says] …we’ve discovered this and that and he goes on reading a little bit and [he says] I think the best case, in this scenario and this case would be to do such and such. But he doesn’t come into the room and say let’s go over this. So there’s nothing for me to decide because I don’t know nothing. It would be better if they had a piece of paper and I had a piece of paper. Step by step, here’s what we found. …I would have liked it better if he had shared that paper with me, if he had sat next to me and said, this is what we found, this is what we think.

One barrier mentioned by several women in both study phases was that their surgeon did not prepare them for a subsequent discussion of chemotherapy options by a medical oncologist. They recalled being told that they would be referred to the ‘Cancer Centre’. Women who felt unprepared perceived the MO consultation as upsetting, which negatively impacted their ability to engage in the consultation.

Other barriers included lack of written information about treatment options and/or information about the general treatment plan after surgery. Some women wanted information and support between the time of diagnosis (by the family doctor or breast‐screening doctor) and the surgeon’s visit. One woman said,

It’s really the support that you need. But support from somebody that can give you the information. Not make decisions for you, but give the information. It might be in small doses. Like a couple of sessions, you know, or how much you want to take today, or, have you had enough today? …I found really that there was a void.

Physicians’ perceptions of physician facilitators of women’s involvement in TDM

Physicians reported many of the same categories of physician facilitators as did women. In particular, explaining the risk of cancer recurrence, explaining treatment options and giving time for TDM were identified by most physicians as facilitators. Techniques for reinforcing information, such as repetition, were mentioned by most physicians. Physicians commented less often than patients about the relevance of explaining to patients the rationale for their involvement in TDM and aspects of the patient–physician relationship as facilitators of women’s involvement in TDM.

Physicians’ perceptions of physician barriers to women’s involvement in TDM

Physicians did not perceive that they presented actual barriers to women’s involvement in TDM during the consultation. However, several physicians were concerned about the potential for undue influence on women’s treatment choices by the way they gave information about treatment options. One oncologist said,

…So I think, I think we [physicians] do make a decision. I think, in retrospect, oftentimes, I think you can easily present the information that guides a person in the direction of the decision you think that they should make.

Several physicians expressed concern about the amount of information given to patients and attempted to ensure that the patient understood what had been said. They expressed feeling torn between wanting to give all the important information about each treatment option, and concern that by doing so, they would overwhelm the patient. These physician concerns were realised by several patients who had trouble coping with the volume of information received during the consultation. Some medical oncologists used strategies, such as pausing, limiting the options presented during the first consultation and revisiting options later on.

One physician said,

I paused to make sure that this was meeting her needs in terms of the level of the information being given. …This is where I often find after having gone through two of the treatments, people can feel quite overwhelmed. And so I want to have a sense of, is it all right to continue? Do they need a break? And for her it seemed to be appropriate.

Discussion

In this study, we explored both the perceptions of women with ESBC and cancer physicians (oncologists and surgeons) to identify physician‐related verbal and/or nonverbal facilitators and barriers to women’s involvement in different stages of TDM as described by Charles et al. ⁷ For instance, physicians’ information‐giving behaviours seem to facilitate women’s involvement in gathering necessary information about their tumour and treatment options. Other physician‐related facilitators, such as giving their own opinion about options, giving take‐home material and giving time for decision making, could be more pertinent to the deliberation stage whereby women consider their options at home particularly related to chemotherapy.

The concept that there might be physician facilitators or barriers to women’s involvement in TDM appeared to be only relevant for women who wanted to participate in some aspect of TDM. It is debatable whether physicians should try to facilitate involvement in TDM for women who are reluctant to participate. In our study, several women thought it was important that physicians explained the rationale for patient involvement in TDM, which in turn, helped women to understand the relevance of their participation. Similarly, Guadagnoli and Ward ³³ have suggested that it is important to invite women into the TDM process recognizing that some women may not wish to participate as much as others.

There are several similarities and differences between the physician facilitators identified by participants in this study compared with those described in the literature. Towle and Godolphin ³⁴ developed eight physician competencies designed to help physicians engage in informed shared decision making with patients. Elwyn et al. ³⁵ developed the 12‐item OPTION scale that consists of physician behaviours thought to promote patients’ involvement in TDM. Both sets of investigators included information‐giving about treatment options as did participants in our study. Towle and Godolphin ³⁴ also included the competency of ‘establishing a partnership with the patient’ both as a stand‐alone item as well as a component of negotiating a decision with patients. However, neither the OPTION scale ³⁵ nor the recently developed dyadic OPTION version ³⁶ included the construct of partnership. Most patients in our study appeared to want some type of validation from the physician as indicated by their comments valuing the physician’s recommendation and support of their own decision making. However, we are uncertain whether the concept of ‘physician partnership’ or ‘physician validation’ or another concept is the most appropriate label to encompass these types of patient‐perceived facilitators.

Several studies have reported that a positive patient–health professional relationship is important for patients to feel involved in TDM as did the women in our study. Entwistle et al. ³⁷ reported that patients associated being involved in TDM with their relationship with health professionals. They found that patients with Type I and II diabetes felt involved in TDM when health professionals were welcoming, respectful, facilitative of patients’ contributions and were not judgmental. Sainio et al. ³⁸ and Henman et al. ³⁹ also reported that a good interactive relationship with physicians and nurses was a facilitator of the involvement of patient with cancer in TDM. A positive patient–physician relationship might set the stage for patient participation in TDM by providing a comfortable environment in which patients can gain relevant information for decision making. In a cancer setting where patients are often worried about their diagnosis and treatment, a positive relationship may be especially important in helping the patient and physician to form a partnership in deciding on appropriate treatment.

Patients and physicians described relatively few physician barriers to patient involvement in TDM. One potentially important physician barrier is the absence of an explicit rationale for the importance of patient involvement in TDM. It may not be sufficient for physicians to give treatment options expecting that patients will understand that there is a treatment choice.

Physicians but not patients described potential physician barriers to patient involvement in TDM such as the order of presentation of treatment options and the choice of words to describe treatment options. Another potential barrier that was identified by several physicians was the manner by which treatment options were presented, which they felt could influence the woman’s perception of the effectiveness of a given treatment. In our study, there were several instances in which patients questioned the physician’s recommendation and the physician responded by using words such as ‘effective’ or ‘standard’ treatment to positively position his/her preference. Schulz and Rubinelli ⁴⁰ refer to this process as a type of ‘strategic manoeuvering’. Strategic manoeuvering used by physicians could act as a barrier to patient involvement in TDM by making it difficult for women to inquire about a different treatment option than the one favoured by the physician. Lelie, ⁴¹ studied TDM in nephrology and found that some physicians gave information in such a manner as to influence the patient’s decision. The concept of ‘strategic manoeuvering’ and its influence on patient involvement in TDM requires further investigation in the cancer setting.

The role of the physician’s treatment recommendation to facilitate patient involvement in TDM is not clear and perceptions about this appear to differ between patients and physicians. Several study women perceived that the physician’s recommendation or guidance about treatment facilitated their involvement in TDM. They described feeling confident about their own decision making when the physician’s opinion matched their own. This matching of opinions about treatment may be especially important in a cancer setting when the diagnosis is serious and treatments may be associated with significant short‐ and long‐ term side effects. However, several study physicians expressed concern that their recommendation might impede women’s involvement in TDM because they would feel led to accept a certain treatment. Gurmankin et al. ⁴² using hypothetical scenarios demonstrated that people are influenced to accept a physician’s recommendation for a treatment they do not prefer. The influence of the physician’s recommendation on patient involvement in TDM requires further investigation.

Our study has several limitations. First, physicians in this setting were interested in patient involvement in TDM as demonstrated by the use of DAs by some physicians and their participation in this study. Second, our study was conducted in an academic setting with breast cancer physicians and patients. Interviews with patients with other types of cancer and in other settings might yield additional facilitators and barriers to their participation in TDM. Third, the study physicians were young (range 34–44 years) with relatively few years in practice. It is unknown whether older physicians would have displayed similar facilitative behaviours. Fourth, women’s descriptions of physician facilitators pertained to behaviours that physicians displayed in the consultation. For example, women would not have mentioned a visual aid as a facilitator unless the physician used one during the consultation.

In Phase 1, women were asked to recall their perceptions of physician facilitators or barriers occurring in the consultation. In contrast, women in Phase 2 were asked about their perceptions, while watching their consultation on a DVD. This difference in data collection method may have affected our results. We noted that more women attending a MO consultation in Phase 2 described physician facilitators, such as the importance of being offered treatment choices and being given time to make a treatment decision, than women in Phase 1. It is possible that Phase 1 women had less recall of these facilitators than Phase 2 women. However, more Phase 2 women perceived that they had participated in TDM as compared with women in Phase 1 and could have been more aware of physician facilitators. The influence of data collection method on patients’ perceptions of physician facilitators and barriers requires further investigation.

Implications for research

As this study was conducted in academic centres in which cancer physicians were supportive of patient involvement in TDM, further research should be undertaken to assess whether these findings are generalisable to other practice settings, particularly non academic cancer centres. Specifically, the role of the patient–physician relationship and physician’s treatment recommendation should be further explored. In addition, several facilitators identified by patients in this study are not consistently reflected in current scales that measure clinicians’ behaviours hypothesized to encourage patient involvement in TDM. Current scales may not be sufficiently comprehensive, and/or tailored approaches to scale development might better fit the specific context of different types of clinical practice. These are important issues for further research.

Implications for clinical practice

The study findings suggest several strategies that cancer physicians in academic centres could adopt to facilitate patient involvement in TDM. First, they could explicitly invite patients to participate in TDM and clearly describe the rationale for such involvement. Second, clear explanations of risk of recurrence with and without treatment appear helpful. Third, patient involvement in TDM is enhanced by clear verbal and written explanations of the potential benefits and side effects of each treatment option that the patient can take home. Fourth, physicians could explicitly indicate that the patient has time to make a treatment decision. Fifth, steps should be taken to ensure that the patient feels comfortable in the relationship with his or her physician. Last, surgeons may need to better prepare patients who will be referred to a medical oncologist to discuss adjuvant options.

Source of funding

The first author was the recipient of doctoral fellowships from the Canadian Breast Cancer Foundation, Ontario Chapter and the Breast Cancer Research Program, United States Department of Defence. The fellowship funding agencies did not have any involvement in the study design, in the collection, analysis and interpretation of data; in the writing of the papers for the thesis; or in decisions to submit manuscripts for publication.