The availability and effectiveness of tools supporting shared decision making in metastatic breast cancer care: a review

Breast cancer is the most commonly diagnosed cancer type among women worldwide and the fifth cause of cancer related deaths [1]. In metastatic breast cancer care many complex decisions need to be made, of which most are preference-sensitive [2, 3]. Important treatment decisions include for example whether or not to start chemotherapy or targeted therapy [4].

Shared decision making (SDM) is an approach in which health care providers and patients share the best evidence when facing decisions, and patients are encouraged to be actively involved in decision making [5, 6]. SDM has been identified as an important element for good advanced cancer care [7]. Most cancer patients prefer to participate in decision making [8, 9]. Among patients with advanced cancer, women with breast cancer in particular wish to be actively involved in decision making [10]. SDM is associated with positive patient outcomes, including knowledge regarding available options, perceived quality of care [11, 12], and quality of life [13].

The use of tools might support active participation of patients in decision-making. Examples of such instruments are 1) decision aids (DAs) which are designed to be used by patients before doctor visits to prepare for decision making [14‐16], and 2) tools to be used by both health care providers and patients during clinical encounters [17, 18]. DAs are developed to support patients in decision making by providing an overview of the available (treatment) options and their associated outcomes [15, 19]. There are many types of DAs, such as video or audiotapes, patient letters, computer programs, leaflets, and interactive media [12].

The tools designed to be used during consultation with a health care provider have been developed to facilitate a conversation between health care providers and patients about the relevant (treatment) options. In general, these tools are brief and present a summary of available options. Examples are decision boards, bar charts, option grids and consult decision aids [17, 18, 20‐22].

The aims of this study were 1) to make an inventory of instruments and tools, including DAs and tools used during clinical encounters, that are currently available for supporting SDM in metastatic breast cancer care and 2) to evaluate the effectiveness of these tools based on published studies.

Three strategies were used to identify tools for supporting SDM in metastatic breast cancer. First, a systematic search of relevant databases was undertaken, secondly an internet search was conducted and lastly experts who appeared in the searches were contacted.

The initial literature search resulted in 687 potentially relevant articles. After removal of duplicates and elimination of non-eligible papers based on title and abstract, 13 full-text articles were considered, of which seven were included for review (Fig. 1). The seven articles described five different tools. In addition, the internet search revealed two relevant tools. All 17 experts approached responded and identified one additional relevant tool (Table 1).

In total, seven tools were identified (Table 1). Three were developed in the USA, three in the Netherlands and one in Canada and Australia. Four tools were specifically designed for metastatic breast cancer, the others for metastatic cancer patients in general. Three tools focused on the decision on whether or not to start chemotherapy [24‐27]. The other four focused on all possible decisions during the entire metastatic breast cancer care trajectory. All tools were developed for patients to be used before consultation with their health care provider. Only one tool [28] provided a summary report to the health care provider which could be discussed during a consultation.

The content of five out of seven tools was evaluated in published studies (Table 2). Of four of these, the effectiveness was studied as well. CONNECT, the communication aid from Meropol et al. was tested in a randomized clinical trial [28]. Outcome measures included consultation content, treatment outcome expectations, decisional conflict, patient satisfaction with the content and format of the communication, and satisfaction with the survey and/or communication skills training. CONNECT made it easier for patients to make treatment decisions (P = 0.003) and patients were more satisfied with their decision (P < 0.001), with physician communication (P = 0.026), with discussion regarding support services (P = 0.029) and quality of life concerns (P = 0.042), but not with discussion of diagnosis/prognosis, treatment options, or support/community services. The DA of Oostendorp et al., [25, 26, 29] was evaluated in a randomized clinical trial. Primary outcome measures included several measurements on patient’s well-being on which the DA had no statistically significant effect. The DA was associated with stronger treatments preferences of patients (P = 0.030) and with increased subjective knowledge (P = 0.022), but not with any of the other secondary outcomes measures. The two other tools [27, 30] were tested in pilot studies without control groups. The DA of Smith et al. [27] assessed whether patients choose to use the DA, investigated the knowledge of patients about the disease and treatment and examined whether the information of the DA was helpful and if the patient wanted to share the information with the physician. All except one patient used the DA and knowledge about the cure of advanced cancer improved after using the DA (P = 0.15). Most patients found the information helpful and almost all patients wanted to share information with their physician after use of these DAs, which might result in SDM. The study on the DA developed by Sepucha et al. [30] evaluated acceptability of the DA and its impact on decisions. The DA was rated acceptable, did not increase distress (P = 0.34) and the treatment goal was most often to lengthen life. Most patients (88%) wanted to be involved in shared decision making, however, only 41% found that decision making was shared and 38% achieved their desired level of participation in decision making. The content of the tool and attitudes towards the tool developed by Chiew et al. was evaluated by both patients and medical oncologists [24]. The patients concluded that the DA was acceptable and helpful and the majority recommend the use of this DA to others. Also the oncologists were positive about the DA and found the DA appropriate for all or most patients.

The quality of five evaluation studies could be assessed. According to the GRADE approach, the quality of the studies ranged between moderate and very low (Table 2). All studies had noteworthy shortcomings, mainly because of the study design. Two had a randomized design and the others were observational studies [24, 27, 30]. The quality of the three observational studies was downgraded to ‘very low’ due to small samples sizes, unclear descriptions of inclusion criteria and lack of information on loss to follow-up. The quality of the studies with a randomised design was downgraded as well [28, 29]. Both studies had a high drop-out rate. And one defined no primary outcome and presented selective results as two of the intervention arms were combined to obtain significant results.

This review identified seven tools to support SDM in metastatic breast cancer care. All were designed to be used independently by patients before consulting a physician. None was developed to be used by both a health care provider and patient during a clinical encounter, although one tool provided a summary report for the physician which could be discussed during a consultation. In general, the identified tools had positive effects on patient satisfaction with their treatment decision and on patients’ desire to share information with their physician. However, it is unclear whether they encourage SDM during a clinical encounter as this was not studied. The effectiveness of the included tools was barely studied. Evidence from the included studies was in general low due to multiple sources of bias, which may have skewed the results.

The revealed tools to support patients in SDM have some limitations. The effectiveness of only four of them was evaluated [27‐30]. Of these, the one with the highest level of evidence was not effective [29]. The other tool with a somewhat higher level of evidence of effectiveness is not available anymore [28] as the tool was not kept up-to-date. The two remaining tools might be useful in clinical practice as their results are promising in pilot studies. These tools could be used next to each other as the DA of Smith et al. [27] focuses on chemotherapy, whereas the DA of Sepucha et al. [30] shows the experiences of four women living with metastatic breast cancer. A limitation of these tools is that they were only tested in a pilot study without a control group. Further testing of these tools in better designed studies is required before they are implemented. The consultation guide presenting information on therapies and supportive treatment in metastatic breast cancer, was not evaluated, but might also be useful for patients with metastatic breast cancer.

Despite the calls for integrating SDM in clinical practice, implementation of SDM into daily care is lacking [31‐34]. The lack of available SDM supporting tools and time concerns might be barriers for implementation [6]. Our review shows the availability of a few tools to be used by patients before visiting the physician and the lack of tools to be used during a clinical encounter in metastatic breast cancer care. In general, tools to be used by patients before visiting the health care provider lead to better understanding of choices, however, yet are not enough to guarantee SDM [14, 35]. In order to facilitate SDM during a clinical encounter, SDM tools for both health care providers and patients have been designed [35‐37]. For curative breast cancer and other tumour types, such tools are available [17, 18]. These tools make options more visible, enhances patients confidence and involvement, and clinicians find it easier to implement SDM in practice [18]. For decision making in metastatic breast cancer care, there is a pressing need for similar tools as many complex decisions have to be made and alignment of care with patient preferences is necessary.

When developing, testing and implementing tools for SDM during a clinical encounter, several recommendations can be made. First, tools should be based on the best available scientific evidence and being kept up-to-date [25, 38‐40]. Second, patients should be included in their development to ensure the tools are user-friendly and understandable [41]. Third, the impact on patient outcomes should be evaluated. Fourth, the conditions for appropriate use of tools in clinical practice should be realized, e.g. clinical teams should recognise the importance of SDM and should be trained in SDM [6, 42], and sufficient time should be available to use a tools for SDM during a clinical encounter [6, 43‐45].

Only two tools for SDM in metastatic breast cancer care were positively evaluated on effectiveness and are currently available. These are developed to be used by patients before consulting the physician. None have been tested in well-designed studies. These tools show promising results in pilot studies and focus on different aspects of care. However, their effectiveness should be confirmed in well-designed studies before implementation in clinical practice. Innovation and development of SDM tools targeting clinicians as well as patients during a clinical encounter is recommended.