Are Older Adults Willing to Consider New Strategies to Reduce Stroke Risk?

Recruitment

Focus group participants were recruited from anticoagulation clinics associated with Dartmouth-Hitchcock Medical Center (DHMC), a tertiary academic medical center, as well as community-based senior living and aging resource centers in New Hampshire during the months of April and May of 2015. The service area of this region includes portions of Vermont, New Hampshire, eastern New York, and parts of southern Maine, with a large rural population. The intention was to recruit six to 12 individuals at each site (Basch, 1987). Interview sites were chosen to capture a broad selection of older adults’ attitudes and beliefs reflecting various backgrounds and experiences. Flyers were the primary method of recruitment, and a selection of patients at the DHMC cardiovascular medicine anticoagulation clinic preidentified by nursing staff was recruited by telephone. Flyers were developed in partnership with each site’s administrative staff, approved by the Institutional Review Board (IRB), and were publicly displayed and distributed via email. The IRB also approved a verbal consent, given minimal risk of the study; the verbal consent was accompanied by a research study information handout that was distributed to all participants. This information explained the general purpose of the study, what would be required of participants, expressed importance of confidentiality, and provided investigator contact details.

Two investigators facilitated group interview sessions at four of the five sites; one investigator conducted a session independently. The group interview process involved moderated question and answer and focused, topic-based discussion. Facilitated group discussions were conducted with the goal of eliciting values and preferences of participants.

Demographic data were collected through a survey distributed at the beginning of each session. A semistructured interview guide containing open-ended prompts was used to guide the discussions (online appendix). All focus group interviews were recorded and transcribed. The Dartmouth College Committee for the Protection of Human Subjects approved this study.

Analysis

Recordings were transcribed verbatim. Group interview transcripts were qualitatively assessed by two researchers (O’Neill & Grande) using a thematic analysis approach; each researcher independently coded the transcripts. Codes were assigned to descriptive groups of data that had meaning relevant to the research focus, which was to identify values and preferences of older adults around the choices for stroke prevention in AF (Miles & Huberman, 1994). Codes were iteratively developed and ideas or concepts in the data were identified and then grouped into meaningful themes (Corbin & Strauss, 1990). Themes were based on salience and key informative statements. Final themes were discussed and ratified with the principal investigator (Coylewright). Any discrepancies were resolved through deliberation and consensus among the coauthors.

Quantitative Results

Five focus groups were completed with a total of 42 participants (Table 1 in Supplementary Appendix). Sites included two anticoagulation clinics (primary care and cardiovascular medicine outpatient practices) associated with DHMC, two senior living facilities, and one community-based senior resource center. The number of participants in each session ranged from two to 15, with an average of eight participants per site. A majority (83.3%) were above 75 years of age and more than half (59.5%) were women. A majority (64.3%) had AF and nearly all (92.8%) were taking a blood thinning medication (aspirin, warfarin, or a DOAC). When asked about preferred decision-making style, one third of participants (35.7%) reported a preference to make decisions in partnership with their clinicians as opposed to preferring they or the clinician alone make the final decision.

Qualitative Results

A qualitative analysis of the five focus group transcripts identified three main themes (Figure 1): (a) fearing a loss of self-control through debilitating stroke, (b) recognizing uncertainty in how to weigh risks and benefits of new treatments, and (c) needing mutual respect between clinicians and patients to consider new/alternative treatment regimens. We propose a model (Figure 1) of patient preferences that situates our three observed themes of prioritizing patient values, the challenges of navigating pros and cons of anticoagulation therapy, and the need for building relationships based on mutual respect between clinicians and patients.

OPEN IN VIEWER

Figure 1.

Framework of patient preferences and concerns related to anticoagulation therapy.

Fearing a Loss of Self-Control Through Debilitating Stroke

Across settings, participants shared and identified with stories reflecting the suffering of friends or family due to stroke, as well as telling personal experiences. Individuals recalled the loss of key functional capacities due to stroke and their reactions to that loss. One conversation that exemplifies the benefit of group interviews came when an individual story highlighted a shared fear around incapacity and loss of personal control through paralysis. The initial comment described how a lack of familiarity with the signs and symptoms of stroke led to fear and confusion.

I got up and I looked in the mirror and I thought someone else was in my skin . . . and the next thing I knew, I woke up and my husband . . . went right to the clinic and said “you’ve got to see [my wife] right away.” By that time, obviously, my speech was not that good. I didn’t even know that.

This led to a second story about how fear of stroke stems from a common worry of not being able to engage with other people. Although the experience described below was about the participant’s mother, it clearly affected the participant’s conceptualization of stroke risk.

My mother had a stroke . . . she felt fine but she couldn’t speak. The doctors were all huddled around her and discussed her. She was going crazy because they were saying things that weren’t true about her. So that was very disturbing for her because she couldn’t speak . . .

There appeared to be a recognition that stroke brings negative consequences, with a consistent underlying concern regarding losing control and an awareness of one’s limitations to change outcomes. A preference for a sudden death, rather than debility by stroke, was clearly expressed.

I don’t know, I don’t think much about it. I just go along and hope that, when it does happen, that it’ll be [snaps his fingers]—that’s it. You know.

Participants also expressed fears of major bleeding. There was a clear difference in understanding between bleeding episodes that required hospitalization and those that did not. When researchers probed regarding prioritization around bleeding that required hospitalization versus stroke in group interviews, participants prioritized stroke prevention. Across the sites, participants were willing to accept an increased risk of bleeding associated with anticoagulants for a reduction in stroke risk if that was the only option.

If my doctor says “you should be taking a blood thinner . . .” I’m going to do it. I couldn’t care less about the risk of bleeding, because it’s going to prevent a possible stroke.

At a separate site, when the conversation came to discussion of bleeding concerns, the general reaction was consistent, prioritizing stroke reduction over other risks. Participants did not worry, as one confirmed, “Well, that’s how I feel, I’m not worried . . . I don’t worry about bleeding.” A follow-up comment confirmed this sentiment, yet differed in that a patient’s preference may change if a job or hobby put them at risk of serious bleeding.

I don’t get worried about nicking myself shaving, but if I were using a machete or something, I’d probably be a little more concerned about it . . .

Reviewing the conversations across sites showed that, overall, the groups felt that nuisance bleeding and bruising associated with anticoagulants were acceptable inconveniences to reduce the risk of stroke.

I just need to be aware . . . I have a cat and every now and then she’ll grab my hand and I don’t think anything of it and then all of a sudden I’m dripping blood . . . The other thing that bothers me is I can bang in to something, and then all of a sudden I’ve got a nice big [bruise] . . . Better than the consequences [of a stroke].

Recognizing Uncertainty in How to Weigh Risks and Benefits of New Treatments

Despite an overall consensus among focus group participants that preventing stroke is most often worth the risks of anticoagulation treatment, opinions varied around willingness to consider new options. One participant was open to new choices after being on the same medication for more than 20 years, “I’d be willing to try a newer [option].” Other participants voiced a need for more information than selecting a new option would be safe and effective: “I just need to know that I could get off Coumadin.” Further discussion elucidated that switching therapies would require a level of certainty—either with evidence or based on a physician’s urging. Although expressions of interest to change medicines occurred across groups, there was a shared uncertainty around dosage, drug interaction, and lifestyle changes.

I refused Coumadin, basically for what these people are saying, because I wanted to eat what I wanted to eat. And my lifestyle wasn’t so that I could just run any time to get [my blood] checked.

Overall, there was little experience with DOACs or LAA closure, and indeed, many participants had never heard of options other than warfarin. There was great interest in learning about alternative choices, as warfarin was widely seen as harmful. Identified inconveniences highlighted issues with routine changes in dosing as well as negative interactions with necessary medications: “I just would rather not be on [warfarin].” Yet, despite a widely held desire across sites for alternative treatments, in some participants, the choice was not provided.

The warfarin seemed to be the choice that I had so I accepted it.

Further probing on the term “accept” led to a clarification on the difficulties of navigating options given incomplete and at times conflicting information from clinicians as compared with commercials in the media highlighting risks of anticoagulants.

. . . you might be in a position to overthink it. Because I think there is so much out there, and you have to be able to judge what’s right from what’s wrong.

Personal accounts on the effects of anticoagulation treatment, as witnessed by family members, further elicited reaction among participants. Shared experiences, which were often negative, appeared to shape individual decision making in similar ways across the sites. One comment on the power of loss helped frame the potential harm of warfarin.

I lost a family member due to the results of warfarin one time . . . I just [wanted to] stay away from that.

Further benefits from group-level discussions emerged when specific preferences among individuals were elicited. When talking about alternative DOACs, specifically regarding daily versus twice a day medical regimens, distinct preferences on the differences between therapies were revealed. Furthermore, given the difficulty associated with maintaining a warfarin regimen with the need for frequent blood tests, the benefits of a nonpharmacological option were noted by participants.

I’d love to do away with [medications] . . . putting [the pills] in the little slots every day and remembering to take them at roughly the same time all the time can be a pain.

Participants who perceived success maintaining the same regimen for years without major issues felt less of a need to consider new therapies. For one participant, they shared a recognition of getting older and being set in their ways.

As you get older, you get less willing to try [new options]. It seems less important to you.

For another whose home location permitted easy access to the hospital, any potential challenges related to the risks of bleeding or the need for routine checks were mitigated.

Warfarin, for me, has been a non-issue. Fortunately, I live close by to the hospital, so getting my blood checked is not a problem. If I lived [farther away], that would be another story.

Needing Mutual Respect Between Clinicians and Patients to Consider New/Alternative Treatment Regimens

Across sites, participants agreed that success in navigating the pros and cons of new treatment options depends on one’s relationship with their physician and a feeling of mutual respect. Participants agreed that their regular doctors, most often primary care physicians or general cardiologists, had a good understanding of their preferences and how they valued certain risks and benefits.

One participant described a desire to play an active role in decision making as long as he did not lose the input of his trusted physician.

[My doctor and I have] been together a long time, and I always feel as though I want the chance to make the decision, but I want her to have a lot of input because I trust her and she knows how I feel about these things.

When probed about defining roles in decision making, participants pointed to a need for “mutual respect” and shared responsibilities between clinicians and patients.

You need a mutual respect. You need to respect your doctor’s knowledge, ability, [and] personality, and the doctor needs to respect you.

A notably consistent message across sites was an implicit trust in clinicians, particularly in situations when uncertainty is highest, to make the right and most appropriate decision. At one site, both participants mentioned their default level of trust. As an example,

I’ve always accepted without question . . . and my wife too . . . she used to tell me to just listen and do.

In agreement, another participant said,

I came out of the hospital and they said take these and I’d say “yes ma’am.”

For others, conversations that included evidence and statistics were helpful. At one site where these risks were discussed openly, one participant said,

When I just think about it, my life overall in the future, my kids and my retirement, when I might retire . . . I think I’ve got a one-third chance of getting a stroke while I’m alive, that’s scary.

Participants identified that recognizing risk and understanding risk are different. At each site, when conversations centered on risk and uncertainty, many of the participants’ reactions defaulted to a “doctor knows best” position, simultaneously reinforcing the primacy of patient preferences.

I want to feel that my doctor is sort of an expert in diagnosis and I do think that you have to cooperate. You have to respect your doctor and your doctor has to respect your wishes.

Although there is reluctance to confront physician expertise, participants suggested that relationships built over time are often those that are most likely to foster opportunities for shared decisions.

I think you develop that sense of mutual respect over time.

I have an excellent rapport with [my physician], even though I only see her once a year. But I think that is the key—how well you can work together and see each other’s viewpoint—and then I think it’s a mutual decision.

In describing decision making about cardiac procedures or medications, participants shared stories about their relationships with their cardiologists, highlighting that trust develops not only with time but also in conjunction with feeling invited to participate in decision making.

You’ve got to have trust in [your cardiologist]. I had the same cardiologist for many years, and we thought a lot alike about a lot of things, and he would ask me what I thought and then he would either agree or disagree and we’d discuss it, and it worked out fine with me. I had the greatest confidence.

Weaknesses and Strengths

Although the focus groups we conducted were representative of elderly participants in the Upper Valley of northern New England, our recruitment and enrollment resulted in a homogeneous population that was largely White and educated. The study group was recruited to include participants who took an antiplatelet or anticoagulant or had prior experience with AF. Others responded to the call for research if they had undergone decision making regarding stroke prevention and use of anticoagulants with a close relative or friend. Only 64% of our participant sample had a history of reported AF, although prior literature confirms a consistent lack of knowledge among AF patients regarding their diagnosis; thus, the prevalence of AF may be underreported here as we did not confirm medical histories with a chart review (Aliot et al., 2010). Given the limitations of demographic diversity and lower rates of reported AF, however, we suggest that our findings be treated as hypothesis generating when applying to patients with known AF. Although the strength of our findings might be further limited by a disproportionately highly educated cohort (48% graduate or professional degree), the strength of group-based discussions involving a range of education was helpful.

Results in Context

Our findings align well with those of prior studies demonstrating variation in participants’ preferences around the risks and benefits of anticoagulation options (Lane et al., 2015; MacLean et al., 2012), and highlighting the importance of consideration of patient preferences. This is critical in the context of the differing value that patients place on preventing stroke versus avoiding major bleeding events (Lahaye et al., 2014). Similarly, it is well known that dosing changes and need for blood monitoring to ensure that warfarin is therapeutic are barriers to its adherence (O’Brien et al., 2014). Our findings echoed the difficulties of constant monitoring, describing personal issues and experiences with warfarin over time as well as a distinct interest in considering new options. There was strikingly little experience with DOACs in this cohort, even though evidence exists that prescribing patterns are radically changing (Huisman et al., 2017).

Moving beyond the use of generic terms of “collaborative” or “shared” to define a quality decision-making process, the literature supports that good patient–clinician communication centers on the active deliberation of treatment choices based on best available evidence, the integration of patient values and preferences, and consensus building between patients and their clinicians (Elwyn et al., 2014; Politi & Street, 2011). In the context of anticoagulation therapy, where multiple reasonable options exist and patients desire more information regarding risk, an SDM process that includes the use of decision support tools to guide conversations at the point of care may improve the delivery of patient-centered care (Grande, Faber, Durand, Thompson, & Elwyn, 2014; Tiedje et al., 2013). Despite wide calls for clinicians to elicit patient preferences, evidence reveals this is often absent in routine practice (Zikmund-Fisher et al., 2010). Decision aids may help to disrupt patterned behavior (Fay, Grande, Donnelly, & Elwyn, 2016) and enhance a learning curve for clinicians to apply SDM skills routinely (Elwyn et al., 2016), leading to a greater degree of trust with patients.

Our findings support that patients value having a strong clinical relationship that is built on mutual trust when considering new therapies. Given the scope of evidence on patients’ willingness to participate in a decision-making process and well-documented demand for more information (Kiesler & Auerbach, 2006), our findings demonstrate the value of utilizing SDM approaches to enhance trust and respect through a process of leveling social distance (Epstein & Peters, 2013; Epstein & Street, 2011). Recognizing that teaching these skills to clinicians is politically and structurally complex, the use of technologies, decision aids, and intentional training integrated into the clinical workflow may assist in making innovation in care delivery less imposing (Godolphin, 2009; Lenert, Dunlea, Del Fiol, & Hall, 2014; Towle & Godolphin, 2009).

Practice Implications

Given the availability of DOACs and the WATCHMAN device to prevent stroke in AF in addition to warfarin, the value of an SDM approach may concurrently overcome clinical complexities while incorporating best practices. Our findings suggest a willingness of patients to consider alternate options with trusted clinicians familiar with their values and preferences; it is imperative to advocate methods that are designed to enhance mutual respect and trust. In this context, primary care clinicians or general cardiologists may be in the best position to apply evidence-based tools (e.g., decision aids) to support an SDM process in accordance with the Medicare requirement. Future research is needed on the implementation of decision aids for anticoagulation treatment, optimizing trust and mutual respect to facilitate elicitation and integration of patient preferences into decision making. Given the limited knowledge of participants regarding all options available to reduce stroke, investigating the role of the subspecialist, who is more fluent on state-of-the-art options, is also needed.