The use of decision aids on early detection of prostate cancer: views of men and general practitioners

Abstract Background and objective While decision support tools such as decision aids can contribute to shared decision making, implementing these tools in daily practice is challenging. To identify and address issues around the use of decision support tools in routine care, this study explores the views of men and general practitioners on using a DA for early detection of prostate cancer. Methods, setting and participants Group discussions and semi‐structured interviews were carried out with 43 men and 16 general practitioners familiar with a previously developed decision aid. Data were analysed using qualitative description. Results Views on using the decision support tool could be classified into four categories: no need for decision making, need for support, perceived benefit and practical barriers. For each category, several underlying themes could be identified that reflect the absence or presence of prerequisites to successful decision support delivery. Discussion and conclusion While men and general practitioners generally have positive attitudes to shared decision making, for both parties attitudes such as not agreeing that there is a decision to be made and doubts on the beneficence of using DAs were identified as factors that may hinder the use of a DA in clinical practice. Participants formulated strategies to support the use of DAs, mainly supplementing DAs with short tools and investing in both training programmes and large‐scale awareness raising of the general public.


Introduction
On the topic of early detection (ED) of prostate cancer (CaP), shared decision making (SDM) is advised. [1][2][3][4] SDM entails patients and clinicians sharing the best available evidence when facing a decision and patients being supported to consider options as well as to achieve informed preferences. 5 Translating this process into practice is not straightforward, especially when the evidence related to benefits and harms is complex, as is the case for ED of CaP. 6,7 To support patients and physicians in SDM, decision aids (DAs) have been developed. DAs are 221 evidence-based tools that prepare people for participation in making specific and deliberated choices among health-care options by providing them with information and by supporting them in clarifying and expressing their personal wishes and values. 8 Research shows that DAs have a beneficial impact on several aspects of decision making: increasing knowledge, value-based decision making and SDM. [8][9][10][11][12][13][14][15] Realizing these advantages in daily practice requires a DA to be used outside of the research context. There are, however, few reports of successful long-term implementation of DAs in clinical practice. Also, research on how to successfully organize the delivery of decision support is scarce. 16,17 Obstacles preventing patients and medical specialists from using DAs have been described and include general barriers such as time constraints and more context specific barriers such as the complexity of the available information. 1,2,11,[17][18][19][20][21] While our knowledge of these factors improves, it has yet to be translated into implementation strategies that meet the needs of care providers and receivers. 18 Our research focuses on factors influencing whether a DA on ED of CaP will be used in daily practice. This topic is characterized by high stakes and equivalent options that are difficult to balance, well suited for using a DA. We opted for an empirical, qualitative research method, taking into account the richness and variability of views brought forward by individuals and groups. 22

Instrument, study design and sample
A qualitative study was conducted, consisting of group discussions with men aged 50 years and more eligible for ED of CaP and interviews with GPs in Flanders, Belgium. Methods follow the consolidated criteria for reporting qualitative research (COREQ). 23,24 Prior to data collection, participants had access to a DA on ED of CaP, 'Making the Choice', previously developed in line with international IPDAS-quality criteria and in collaboration with the GPs participating in this study. 25 This comprehensive instrument contains information in Dutch on the (dis)advantages of (not) opting for ED and provides support in clarifying and communicating preferences. It was available for participants as a booklet and as a website. 26 To increase study participation and stimulate future DA implementation, we opted to purposefully limit our study participants to early adopters, that is potential users who are motivated to adopt an innovation and who can play an important role in stimulating adoption by other potential user groups. Therefore, we selected two participant groups: (i) GPs that showed an active interest in using DAs in clinical practice and (ii) men that were interested in or had questions about ED of CaP. We chose to include GPs active in both rural and city regions because the proximity of universities and the subsequent possibility of frequent involvement in research implies that GPs of the latter group may have a different view on novel evidencebased evolutions such as the use of DAs than GPs of the first group.
Together with the Belgian association of GPs (Domus Medica), we organized information sessions for GPs on SDM and the use of DAs. Sessions took place throughout Flanders. At the end of each session, GPs were asked whether they were interested in testing the DA in clinical practice. The 36 GPs that answered positively were contacted for participation in this study. Eventually, 16 GPs participated in individual telephone interviews. Non-participating GPs cited time constraints as a reason for opting out. In parallel, we contacted 50 clubs and societies to invite eligible men for participation in our study. These sociocultural clubs and societies all bring senior citizens together for various leisure activities. Five clubs and societies located in the Northern and central parts of Belgium (Dophei Vosselaar, KWB Herent, OKRA Vosselaar, Senioren Leuven and Sint-Sebastiaansgilde Vosselaar) responded positively and disseminated our invitation to their participants. Eligibility criteria for participating men were as follows: (i) being 50 years or older and (ii) being interested Decision support on early detection, A Engelen et al.
in or having questions about ED of CaP. Men interested in participation contacted the research team directly or through the president of their club. We eventually arrived at a sample of 43 men participating in group discussions that took place in locations provided by the five involved clubs. All participants received information on the use of DAs and on the purpose of the study. Afterwards, all men and GPs who used the DA gave their verbal and written consent to participate in the interviews or group discussions. All participants were informed that they could withdraw from the study at any time. No financial compensation was given. Prior to interviews (GPs) and group discussions (men), participants were asked to complete a questionnaire to collect demographical data as well as data concerning medical practice (GPs) or decision-making characteristics (men) (Box 1).

Data collection and analysis
All five group discussions (men) and 16 individual telephone interviews (GPs) were conducted in October 2013. Group discussions involved 5-13 participants and lasted about 120 min. Individual interviews with GPs lasted about 30 min. Using a semi-structured discussion guide, the interviews and group discussions progressed from broad, open-ended questions to narrower questions with specific probes to clarify issues if needed. 22 All interviews and group discussions were conducted by AE, a junior biomedical researcher with experience in conducting qualitative research. AE was supported  in this process by JV, a senior researcher who assisted in one of the group discussions. Data collection and analysis were supervised by CVA, a professor with substantial experience in conducting qualitative research. The interviewer had not been in contact with the study participants prior to the start of this study. She participated in the development of the DA used in this study and has witnessed the difficulties of implementing these tools in daily practice.
Interviews and group discussions focused on three major topics: (i) the evaluation of the instrument and its use in consultation, (ii) factors that may hinder or facilitate implementation and the realization of positive effects and (iii) views on the ED decision ( Figure 1). Views of GPs and men relating to the evaluation of the developed tool as such will be presented elsewhere. The topic guide was developed by the research team and was based on experiences in previous studies on the development and evaluation of DAs. 9,27 The questions were intended to stimulate conversation on the perspectives of men and GPs on using a DA on ED of CaP. Participants were encouraged to talk freely about their experiences and views. It was explained that the purpose of the interview or group discussion was not to reach agreement and that there were no 'bad' answers or comments. No repeat interviews were carried out. The focus groups and interviews were recorded, transcribed verbatim and managed using NVivo10 software (QSR international Pty Ltd., Doncaster Australia). Field notes made during the interviews and group discussions were used to inform data analysis. At several points during each interview or group discussion, the interviewer presented a brief summary of the main ideas and asked participants whether they would like to make changes or additions. Data were reported anonymously to maintain confidentiality.
Qualitative description was used to analyse and report the data collected. 28 In a first step of thematic analysis, the most important topics and concepts were defined by open coding of each group discussion and interview separately, soon Decision support on early detection, A Engelen et al.
after it had taken place. Each information unit of the transcripts was examined for emergent themes in relation to the issues explored and labelled accordingly. We identified in vivo codes and explored ideas for the advancement of more abstract codes. As further data were analysed, each transcript was revisited and the coding was revised. The labels assigned to the ideas emerging from the transcripts were brought together, inductively categorized and refined through an iterative process. After thematic analysis by AE, the accounts of GPs and men were compared to assess what we could learn from their different perspectives. During the whole process of analysis, whenever it was unclear how transcripts should be coded, this was resolved through team discussions. The sample size was sufficient to reach theoretical saturation as the final group discussions and interviews did not change the study themes.

Results
We identified four categories related to using the DA: (i) no need for decision making, (ii) need for support, (iii) perceived benefit and (iv) practical barriers. All categories contain several underlying themes. All categories were identified for GPs and men alike. All categories and themes are described below alongside relevant data extracts.

No need for decision making
While in each group discussion at least one man mentioned that ED of CaP has both benefits and drawbacks, men generally reported no need for decision support because they felt there was no decision to be made. Men highlighted the benefits of ED and saw no reason not to test. They also admitted being worried about CaP or about staying healthy in general. Several men said that their partners advised them to get tested. GPs, on the other hand, generally reported that there is no decision to be made. They did not favour ED of CaP because, to them, it has too many drawbacks and too few advantages. While men and GPs did express how they prefer SDM in general, men were not eager to participate in decision making on ED of CaPa feeling expressed by men and acknowledged by GPs. Men furthermore mentioned in most group discussions that they do not consider this decision to be important and that they do not like to talk about CaP. While some GPs did mention how they inform men on the topic of ED of CaP and involve them in decision making, it became apparent that GPs and men currently generally do not engage in SDM on ED of CaP

Perceived benefit
GPs and men indicated that DAs could have several effects and evaluated these effects differently. Members of both parties mentioned that the DA could lead to either more or less testing. A decrease in testing due to DA use was generally perceived as harmful by men and as advantageous by GPs, while a potential increase in testing was generally perceived as advantageous by men, but as harmful by GPs. In addition, on the topic of SDM, some men and GPs were convinced that using the DA could foster communication, while others disagreed. GPs mentioned that using the DA could reduce time investment for SDM. On the topic of information, men and GPs reported that the DA could help users to understand the complex nature of ED of CaP. Furthermore, men described the tool as a good source of information and GPs stated that the tool would support them in their role as information giver. While some men mentioned that using a DA could make it easier to arrive at a decision, other men and GPs felt that using the DA would make it harder to make a decision.

Practical barriers
GPs discussed they often experience time pressure in daily practice. Men acknowledged this and mentioned that time constraints might hinder DA use. Both parties described how using short tools aimed at fostering SDM in a timeefficient manner alongside or instead of more comprehensive DAs, during or outside of the consultation, could address this barrier to DA use. Furthermore, many men and GPs pointed out that they have no experience in using DAs and that this might hinder DA use. GPs advised to organize training sessions to support them in using DAs. In addition, some GPs and men doubted whether men would be capable of using DAs. Both parties pointed out that it may be challenging for patients to search for an online DA and to use it efficiently. To address this potential barrier to DA use, men advised in all group discussions to ask GPs to deliver the DA to interested men. Yet, some GPs pointed out that this might be difficult to achieve given the time pressure. A solution to this problem mentioned by men and GPs alike was to provide a short folder or poster in the waiting room. However, several men pointed out that this does not guarantee that all men find the DA since not all men regularly visit their GP. Alternatively, men and GPs described how it would be helpful if men's awareness on the existence of DAs and where to find them could be increased by large-scale awarenessraising campaigns.

Discussion
In this study, we explored the views of men and GPs on using a DA on ED of CaP, allowing us to identify factors influencing its use in clinical practice. Four categories of adoption factors were identified: (i) no need for decision making, (ii) need for support, (iii) perceived benefit and (iv) practical applicability. On the topic of ED of CaP, it became clear that many GPs and men experienced no need for decision making. Moreover, the opposing attitudes of men and GPs to ED of CaP lead to a clash of expectations. Men expected to get tested and reacted indignantly when their GP did not do so spontaneously while GPs remarked that it is difficult for them to deal with men's unquestioningly positive attitude towards testing. In the process of SDM, it may occur that both GPs and men bring their opinions and values into the consultation. When preferences of GPs and men differ, as in this case, this may negatively affect their relationship, both parties' feelings and the likelihood of achieving true SDMcertainly when men are unlikely to openly disagree with their GPs. 29 Yet, in general, GPs and men did express a need for support in communication, understanding information and information giving. In fact, both parties admitted that men are currently strongly under-or misinformed on ED of CaP. Because ED in truth is a complex topic with advantages and disadvantages that need to be weighed in decision making, men's unquestioningly positive attitude towards testing is indicative of mis-or underinformation. 1,30 As ED has long been portrayed in an unquestioningly positive way in Belgian general media, this misinformation may in part be explained by previous exposure to protesting advertising. With regard to DA implementation, misinformation may hinder DA uptake by reinforcing the opinion that there is no decision to be made. Without care providers and users agreeing that a balancing act is in order when considering a specific medical topic, neither party will see any merit in doing a balancing act together. Indeed, views on the potential benefit of using the DA were mixed. To tackle misconceptions and foster informed decision making, using DAs is a valid approach. Yet, the results of this study make it clear that it is difficult to achieve effective use of DAs when both parties initially feel there is no decision to be made. 8 The accounts of GPs and men showed that whether tests are conducted is often strongly influenced by the patients' desire to get tested. Since many men have an uninformed positive attitude towards testing, this may result in tests being ordered for men who would not have opted for ED if they would have been more correctly and fully informed on the topic. 31 In addition, the DA may reach men when they have already decided in favour of testing, thus limiting its potential impact on the decision or decision-making process. As such, the lack of information explains why many men experienced no need for decision makinga barrier to DA useand why there is a need for support to put the prevailing misconceptions right. As a consequence, it becomes important to research means to provide information to men in a way that is not influenced by misinformation as a barrier to information giving. Essentially, this would imply providing information in a way that is less dependent on men's desire to access the information, such as providing information on a large scale in general media.
On a practical level and in accordance with prior research, our results indicate that the practical applicability of DAs is severely curtailed by time constraints 20 To address this barrier, men and GPs repeatedly proposed to opt for short tools that can be used in a time-efficient manner during the consultation and can either replace or complement comprehensive DAs. Several research groups have already experimented with short decision support tools. [32][33][34] An example is the recent development of Option Grids: short one-page tools that can be used during consultation to optimize the SDM process. 32 Prior research about clinical topics such as breast cancer or head and neck cancer has shown that using Option Grids can contribute to SDM and can support GPs in delivering information. 35,36 Research also shows that interventions targeting patients and health-care professionals are more promising than those targeting one or the other. 14 Additionally, in Belgium, DAs are still novelties, unfamiliar to most health professionals and patients. This lack of experience can be addressed by information and training sessions aimed at GPs on using DAs efficiently. 19 Research shows that multifaceted interventions that include both efforts to educate health-care professionals and the use of DAs are promising in promoting the adoption of SDM in clinical practice. 15 On the patients' side, our study highlights the importance of increasing patients' awareness on the existence of DAs and where to find them. This study has some limitations. Firstly, it focuses on the use of a DA in one specific complex medical context by participants who were interested in using a DA or in ED of CaP and who had the cognitive abilities to understand and discuss the decision aspects. The specific themes identified in this study may not be present as such in other medical contexts or in different user groups. Yet, we do believe that the broad categories of no need for decision making, need for support, mixed feelings on potential benefit and practical applicability also affect the implementation of DAs in other decisional contexts. Secondly, the recruitment strategy employed does not allow for providing information on the number of eligible men that chose not to participate to the study and why.
Thirdly, since it proved unfeasible to organize group discussions because of time constraints, we conducted telephone interviews with GPs. Both methods may lead to a different depth and width of insight in the participants' views and experiences, which may have influenced our results. Also, interviewing about a care innovation is susceptible to social desirability bias. However, participants were reminded explicitly that they could freely speak their mind. An important strength of this study is that both men and GPs had access to a DA that was specifically designed for them, which allowed for focused and informed communication on one topic and resulted in a broad overview of factors influencing the effective use of decision support tools. Future research should focus on the extent to which the lack of need for decision making, perceived benefit and practical barriers influence DA implementation. Also, research should be done on a DA development strategy that takes the pre-existing preferences and attitudes of the target audience into account. We believe that the DA development process should be preceded by an assessment of pre-existing attitudes and potential practical barriers to DA use.

Conclusion
The use of DAs on ED of CaP is influenced by multiple adoption factors. A lack of need for decision making and passive role preferences hinder patient participation in decision making. Yet, both GPs and men indicate a need for informational and communicative support. At the same time, the perceived time investment associated with using a DA and a lack of experience hinder the use of DAs. To overcome barriers to the use of a DA, we follow the recommendations of men and GP and call for an increased focus on the development and practical evaluation of short decision support tools that can be used in a time-efficient manner during the consultation. Yet, to achieve successful DA implementation we recommend that changes be made not only on the level of the tools used, but also on an attitudinal level. This calls for health professionals and patients to be supported in using tools and to be informed on aspects of the subject matter of the tool and on SDM. Training programmes for health professionals that are provided in an accessible, time-efficient way should be an integral part of any decision support strategy. Additionally, large-scale awareness raising can set patients' misconceptions right and can increase patients' awareness on the existence of DAs and where to find them. and conduct of the study or in the collection and analysis of the data, nor in the drafting of the manuscript for publication.