Characterizing clinician communication with patients about lecanemab: A qualitative study of clinicians across seven academic medical centers

Abstract

INTRODUCTION

Anti‐amyloid monoclonal antibodies (mAbs) slow cognitive decline in Alzheimer's disease but may cause amyloid‐related imaging abnormalities (ARIA), which can rarely be disabling or fatal. This qualitative study investigates how clinicians communicate the benefits and risks of mAbs to patients and caregivers.

METHODS

Semi‐structured interviews with clinicians who prescribe mAbs at seven academic medical centers. Hybrid inductive‐deductive thematic analysis by interdisciplinary researchers.

RESULTS

In 27 clinician interviews (women [n = 17], White individuals [n = 19], neurologists [n = 17]), three themes emerged. First, clinicians varied in techniques used and concepts emphasized, including using analogies, discussing statistics, and emphasizing versus de‐emphasizing risks. Second, patient contextual factors (e.g., comorbidities), hopes, and fears shaped communication. Third, clinician communication varied by training, personal style, and ambivalence. While clinicians honor patients’ choices to pursue treatment, many do not “recommend” it (but may recommend against it).

DISCUSSION

Preliminary insights about how clinicians communicate tradeoffs can guide future shared decision‐making interventions for mAbs.

Highlights

• This qualitative study among 27 clinicians across seven academic medical centers examined how clinicians communicate with people with Alzheimer's disease about risks and benefits of anti‐amyloid therapy, which can influence treatment decisions.

• Clinicians varied in what techniques they employed and how they portrayed risks and benefits, and whether they incorporated patients’ values.

• They cited comorbidities, eligibility criteria fit, and degree of social support or family involvement in decisions as factors used in framing discussions, while fewer used patients’ goals to guide discussion.

• The professional training, individual practice style, and personal sense of ambivalence of clinicians shaped conversations.

• These findings can guide future interventions to improve communication and shared decision‐making.

1. BACKGROUND

The anti‐amyloid monoclonal antibodies (mAbs) lecanemab and donanemab are the first disease‐modifying therapies for people with Alzheimer's disease (AD). Lecanemab and donanemab slow cognitive decline in early AD (mild cognitive impairment or mild dementia due to AD). ¹ , ² , ³ , ⁴ However, adverse effects include intracranial swelling (amyloid‐related imaging abnormalities due to edema/effusion [ARIA‐E]) and intracranial hemorrhage (amyloid‐related imaging abnormalities due to hemorrhage [ARIA‐H]), which are predominantly asymptomatic but can be deadly or disabling. ⁵ Treatment is also logistically demanding and expensive. ⁶ As access to anti‐amyloid mAb therapies expands, patients, caregivers, and clinicians must balance the realities of treatment with hopes of modifying the AD disease course. ⁷

Shared decision‐making (SDM) is a collaborative process in which patients, caregivers, and clinicians make healthcare decisions together, considering the best scientific evidence and incorporating the patient's goals and preferences. ⁸ SDM is especially important for decisions that are preference‐sensitive, value‐laden, and require deliberation about tradeoffs. The way physicians communicate with patients plays a crucial role in influencing decisions. Clinicians can over‐ or understate benefits and risks, miss opportunities to elicit goals, fail to discuss cost‐related factors, or present inaccurate information. ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ Given that anti‐amyloid therapies are novel, ensuring patients and caregivers understand the tradeoffs and supporting them to make values‐aligned decisions is critical. Yet we know little about how clinicians discuss anti‐amyloid therapy. The objective of our study was to describe how clinicians communicate with people experiencing early AD and their caregivers about lecanemab.

2. METHODS

2.1. Study design and data collection

Between August and October 2024, we conducted a qualitative study using semi‐structured interviews of frontline clinicians who counsel people with early AD on lecanemab, including those who evaluate for eligibility, prescribe, or both. We included physicians (MDs, DOs) and advanced practice providers (APPs) from subspecialized memory clinics. We recruited through individual e‐mail solicitations and snowball sampling and invited participants from seven US academic medical centers in the Southeast, Northeast, West, and Midwest.

A multidisciplinary team developed an interview guide using the Ottawa Decision Support Framework. ¹⁵ The guide included questions about the (1) conveyance of risks and benefits, (2) discussion of goals and values, and (3) decision‐making process. Given the treatment's novelty, the study team asked additional questions about: (4) changes to professional culture and practice and (5) logistics, workflow, and protocols for implementation. We report the clinician communication data here and will report the culture and implementation data separately. We conducted interviews remotely by Zoom by trained study personnel (A.L.P., A.T., D.D.). Participants provided informed consent and completed an online demographic questionnaire. The Mass General Brigham institutional review board approved the study protocol (2023P002711). This study was performed in accordance with the 1964 Declaration of Helsinki ethical standards, and we adhered to Consolidated Criteria for Reporting Qualitative Studies (COREQ) guidance (Appendix 1). ¹⁶

2.2. Data analysis

A.P. (female physician‐researcher with qualitative research expertise), A.T. (male clinical research coordinator with qualitative research expertise), and J.P. (female researcher with serious illness communication and qualitative research expertise) comprised the analytic team. We digitally recorded, transcribed, and de‐identified interviews. We conducted qualitative thematic analysis using stepwise, hybrid deductive‐inductive analysis between October and November 2024. ¹⁷ , ¹⁸ Using the interview guide as a preliminary framework, A.P. and A.T. independently coded the first five transcripts deductively; all transcripts were coded independently by two coders. We added additional codes that were derived inductively to create the codebook. We then applied this codebook to all transcripts. A.P. and A.T. met every five transcripts to discuss discrepancies and resolve them by consensus, modify codes, and iteratively build content. Through meetings with additional experts (J.P., D.D., S.S.), we developed codes into themes and sub‐themes. The full research team reviewed and confirmed the associations among the codes to identify the final themes. The group also addressed how personal experiences informed and potentially influenced the interpretation of the data to mitigate potential bias. Recruitment was stopped when data saturation was achieved—the point at which interviews were not providing new information, which was determined via review of qualitative data by the entire expert group in line with evidence‐based practices. ¹⁹ , ²⁰ , ²¹

3. RESULTS

Characteristics for the 27 participants are provided in Table 1. Most were women (63%), White individuals (78%), physicians (74%), and trained in neurology (63%).

TABLE 1.

Participant characteristics.

Characteristic	n (%)
Practice
Site 1 (50% MD, 50% neurology, 50% high prescriber)	4 (15%)
Site 2 (100% MD, 75% neurology, 0% high prescriber)	4 (15%)
Site 3 (60% MD, 20% neurology, 40% high prescriber)	5 (19%)
Site 4 (75% MD, 100% neurology, 100% high prescriber)	4 (15%)
Site 5 (100% MD, 100% neurology, 100% high prescriber)	2 (7.4%)
Site 6 (100% MD, 100% neurology, 66% high prescriber)	3 (11%)
Site 7 (60% MD, 60% neurology, 100% high prescriber)	5 (19%)
Specialization
Neurology	17 (63%)
Cognitive/behavioral neurology	9 (33%)
General neurology	7 (26%)
Other neurology	2 (7.4%)
Geriatrics	10 (37%)
Sex
Female	17 (63%)
Male	10 (37%)
Race
Asian	4 (15%)
Black or African American	1 (3.7%)
More than one race	1 (3.7%)
White	21 (78%)
Ethnicity
Non‐Hispanic	24 (89%)
Hispanic	3 (11%)
Role
Physician	20 (74%)
Physician assistant	4 (15%)
Nurse practitioner	3 (11%)
Time in practice
1–5 years	3 (11%)
6–10 years	12 (44%)
More than 10 years	12 (44%)
Role in practice
Refer people to receive therapy	13 (48%)
Administer/prescribe therapy	11 (41%)
Both	3 (11%)

Note: “High‐prescriber” is defined as having administered lecanemab to 10 or more patients at the time of the interview [based on the mean # of patients prescribed lecanemab per participant in Aug‐Oct 2024]).

The analysis revealed three major themes and associated subthemes (Figure 1). Representative quotations are presented in the text. Table 2 provides additional supportive quotations, and all deidentified, coded data are available in Appendix 2.

FIGURE 1.

Major themes, subthemes, and implications identified in clinicians discussing anti‐amyloid monoclonal antibody treatment.

TABLE 2.

Additional illustrative quotations from clinicians by theme.

Theme	Illustrative quotation
Clinicians vary in what techniques they use and what they emphasize in discussing risks, benefits and patient goals
Use of analogies	“And usually I use the words, ‘I think of this medicine as a vacuum cleaner…and how this medicine's going to pick up [amyloid]’…and the hope is stability and maybe some improvement of where you are right now.” (Site 1, ID 4, neurology, APP) “When you go in, and you remove the bricks off the wall, it can create some instability there, and you can have these little microhemorrhages. As we actively remove that amyloid, you can generate these little microhemorrhages that suggest that maybe we need to slow down with the administration of the medicine.” (Site 6, ID 21, neurology, MD)
Employing statistics from the clinical trial to convey benefits	“After 18 months of infusions, there are six infusions, we looked at a CDR, and I explain what the CDR is… and basically those that had the infusion ended up at a score of 76 percent, which would be about a C, versus 73 percent, which would be a C‐minus. Or I'll say half a point difference on an 18‐point scale.” (Site 3, ID 11, geriatrics, MD)
Underscoring treatment neither cures nor reverses	“I try to be clear in explaining that these medications are not a cure yet, that we wish we had one and that we are still working on that, but these medicines are not, and that they don't stop the progression either.” (Site 4, ID 14, neurology, MD) “Then I usually go into more detail about that, that all the patients in the study still get worse. I think that is an important—I've had a few patients who come back and say, ‘She's not any better.’ It's like, well it's not gonna get better and that you may not know if it's working.”(Site 7, ID 23, neurology, MD)
Accentuate or downplay risks	“What we found is that they're usually harmless unless they turn into macro‐hemorrhages, which is a rare occasion. With the edema or the swelling of the brain, what we found in the studies is that it naturally resolves within about three to four months. Within the trial, they found that over 80 percent of that swelling of their brain resolve just spontaneously. If it were to occur, then you know, and if it was severe, we could also treat it with steroids.” (Site 7, ID 26 geriatrics, APP) “I use the term bleed and swell because there's a lot of criticism that ARIA sounds like an Italian song … No one leaves the counseling sessions with me without understanding that really what we're talking about are areas of bleeding, usually microscopic, usually very small but bleeding nonetheless, in areas of swelling or edema.” (Site 3, ID 10, neurology, MD) “I usually say there's very few drugs out there that don't have side effects, and the ones that don't have side effects usually don't have much effect… There's two main side effects to think about with the drug. The easier one that's probably the most common is the infusion reaction, so a lot of people have a mild allergic response when they first get the medicine because it's a protein that gets infused into the blood. People can have a little bit of a headache, fatigue, itchiness, those kind of things. Generally very mild, but it happened pretty frequently so letting people know that that might happen… Then, the main concern to think about beyond target response to this is amyloid ARIA—Amyloid‐Related Imaging Abnormalities, which is, I say, a pretty name for not‐so‐pretty of a thing.” (Site 6, ID 21, neurology, MD)
Eliciting hopes and fears	“In some of my conversations I have talked explicitly about some people's goals, which is really to be proactive and to feel like they're on the cutting edge of research in terms of Alzheimer's drugs. And then some people feel like it's better to do something than to do nothing. And then other patients will say this is really not worth the risk. And I cannot make the decision on their behalf, but I ask where did they see themselves aligned on these two extremes”. (Site 3, ID12, geriatrics, MD) “Since the beginning, usually even when I don't have a specific diagnosis yet, I try to understand what are their hopes, their values, their family and all that….I absolutely ask them what their goals are and their preferences so we can decide if this may be good for them in terms of lifestyle and quality of life. I tell them that that's what we need to focus on, that they really enjoy and that this is their time, so they do what they feel is more appropriate for them.” (Site 4, ID 14 neurology, MD)
Patient contextual factors and personal goals and values shaped communication and decision‐making
Comorbidities	“I'd say a 70‐year‐old who's still working, has mild symptoms but can still function. Maybe a clinical dementia rating of 0.5, has really no other risk factors. They're not on anticoagulation. They're, like I said, healthy….We see people like this, who are in good shape otherwise, but have very mild symptoms that are consistent with an amnestic type of syndrome like AD…I would be pretty confident prescribing to someone like that because I think their risk of ARIA is reasonably low but their chance of benefit seems high in my mind.” (Site 7, ID 23, neurology, MD)
Eligibility criteria as a scaffold for discussion	“I think that in going through this process [of establishing whether a patient meets eligibility criteria], even the lumbar puncture makes things more real for patients because they have more hesitancy about it.” (Site 3, ID 11, geriatrics, MD) “I see so many patients with mixed disease that it's really, really important that I am looking at imaging carefully, I'm looking at the risk factors carefully, I'm looking at is the time course correct? Is it more attention and not actual rapid forgetting? I'm trying to sort out those phenotypes really carefully so that I'm matching up to what I think the trials were intending to do, as opposed to just saying, “Well, if there's amyloid in the brain, it's probably better not to have it,” right?” (Site 2, ID 8, neurology, MD)
Degree of social or family support	“If they're well supported from family standpoint, they have a partner or family at home who can take them to their infusions and can communicate with us well in terms of if they're worried about side effects and things like that. That's the best‐case scenario I'd say.” (Site 4, ID 15, neurology, MD)
Role of caregivers in decision‐making	“Through that process, she really felt like—she explicitly said that she felt pressured from other physicians to get the medication and from her family too and got really, really upset. That's not encouraging.” (Site 3, ID 11, geriatrics, MD)
Patients’ hopes or fears influenced decisions	“Some people's goal is really to be proactive and to feel like they're on the cutting edge of research in terms of the Alzheimer's drugs. And then some people feel like it's better to do something than to do nothing. And then other patients will say this is really not worth the risk.” (Site 3, ID 12, geriatrics, MD)
Clinician attributes, ambivalence, and uncertainty influenced communication and recommendations
Professional background	“If I'm not comfortable with the medication, but yet my patient really wants the drug, then I will discuss it with my team because I am aware of that clinical bias. I want to ask other members of the team. I'm very comfortable in doing that, and then I feel more confident about the overall decision.” (Site 1, ID 3, neurology, APP)
Personal style	“I will make sort of soft, ‘Okay I think you're a reasonable person to give this drug,’ I would be willing to prescribe it. I really am far short of sure— I'm not an oncologist, right? The way that I hear my oncology colleagues be able to say, ‘You need this chemo regimen,’ I don't do that. I wish I could, and I hope I get there one day.” (Site 3, ID 10, neurology, MD) “I think my kind of practice style is perhaps a little less paternalistic. So if I think the patient understands the risk, understands the uncertainty around the benefit, and they're interested in proceeding with treatment, and they meet the appropriate use guidelines, then I'm willing to treat them even if I don't think it's necessarily the great idea or that they're going to stand a high probability of benefit.” (Site 1, ID 1, geriatrics, MD) “The truth is, I think I'm a little bit more of the kind of person who's like, “It's your risk. It's not my risk.” If you're recognizing that we've laid out the options for you, and you recognize what the risks are, I think you should be allowed to make that decision. In a vacuum, that's the way I am. I feel like our center's not so much that way. It's more we're going to really bend over backwards to protect people. I think different clinicians in our group just are—we're all on a spectrum there. I know there's people who lean more toward me. I know there's people who are much more restrictive than that.” (Site 6, ID 3, neurology, MD)
Ambivalence and uncertainty	“I don't know what the true benefit of the‐the drug may be, the trials were only 18 months and we don't know what happens after 18 months.” (Site 2, ID 5, neurology, MD) “I personally don't feel very strong, like a hundred percent. I have the hopes and the belief that we are going to get there, that we are going to find many things that work better and perhaps would be combination therapies or these ones that target amyloid and tau and all that or if we start early, but my personal feeling as a clinician right now is that I still—there are too many unknowns.” (Site 4, ID 14, neurology, MD).
Hesitancy to recommend treatment versus comfort with recommending against	“If I'm worried about the patient, I may sway them against it.” (Site 2, ID 7, geriatrics, MD) “I definitely leave it up to the patient and I try to be clear to them when I explain that, because I don't want them to feel that, “Oh, my doctor is saying I need to get this and if I don't get it, she's going to be disappointed or I am not following what she says.” I try to always explain to them, “If you don't do it, it's fine and it's okay because it's your choice.”” (Site 4, ID 14, neurology, MD)

RESEARCH IN CONTEXT

1. Systematic review: The authors conducted a literature review via traditional databases (e.g., PubMed). We identified a lack of publications about how clinicians communicate risks and benefits of novel therapies for Alzheimer's disease, including anti‐amyloid monoclonal antibodies. We reviewed the literature on decision‐making and communication generally and among people with Alzheimer's disease and caregivers specifically, and these data are cited as appropriate.

2. Interpretation: To our knowledge, this is the first study examining how and what clinicians communicate about the risks, benefits, and tradeoffs of anti‐amyloid therapy in discussions with people with Alzheimer's disease and their caregivers. We identified three key themes. First, the techniques and points of emphasis varied among clinicians, with many using analogies and statistics, some emphasizing and others de‐emphasizing risks, and a few eliciting hopes and fears. Second, patients’ contextual factors (e.g., family support, fit with eligibility criteria) and personal goals (e.g., hopes and fears) drove discussions. Third, clinicians’ training background, individual style, and their own sense of uncertainty influenced communication and recommendations.

3. Future directions: These findings inform strategies to improve clinician communication that support clinicians, people with Alzheimer's disease, and caregivers in making informed and values‐driven decisions about treatment. Future studies should expand on this work with direct observations and recordings of physician communication to enhance the robustness of these findings.

3.1. Clinicians vary in what techniques they use and what they emphasize in discussing risks, benefits, and patient goals

Clinicians used analogies for the mechanism of lecanemab to illustrate how treatment could lead to both slowed cognitive decline and side effects.

“I usually tell them about the mechanism, about what it does, clearing the amyloid from the brain. I tell them it's like clearing out the kindling, so the fire is still burning but maybe not as bright. It's still progressing.” (Site 6, ID 21, neurology, MD)

“ ‘I want you to think about a scab on your skin. Those amyloid plaques, the medicine is helping take the amyloid plaque off the brain. I want you to think about ripping a scab off your skin, and you get those little dots of blood. You get microbleeding there. Sometimes you may have a little bit of swelling.’” (Site 3,ID 13, neurology, APP)

In describing the benefits of lecanemab, clinicians quoted statistics from the Clarity AD trial. They also described how the trial's outcome measure—the Clinical Dementia Rating‐Sum of Boxes (CDR‐SB)—translates into meaningful changes.

“The study showed that this potentially slows the progression three to six months. I draw two little slopes. The bottom line is the slope if you were not to get lecanemab. At 18 months, that's potentially 3 to 6 months difference. The top line would be if you got the lecanemab.” (Site 3, ID 18, neurology APP)

“One thing that I find very helpful is the conversion on the CDR global scale… I say, ‘In the trials, the people who were on drug had about a 30 percent lower chance of converting from mild cognitive impairment to dementia during the year and a half of the trial if they were receiving the drug.’” (Site 6, ID 20, neurology, MD)

Nearly all clinicians emphasized that these therapies neither cure nor reverse memory loss.

“I do make sure to express very clearly that it does not stop disease or reverse disease because I think that's a misconception.” (Site 2, ID 6, neurology, MD)

While all clinicians communicated the risks, they differed in how much they emphasized or de‐emphasized risks, particularly ARIA.

“I use the terms brain swelling and microbleeds and I go on to explain that hardly anyone who has brain swelling or microbreeds will actually get sick from it clinically.” (Site 5 ID 18, neurology, MD)

“It causes seizures. It can cause death. It can cause permanent loss of neurological function. People in the studies have had all those outcomes. Then, I talk about ARIA—I couch the percentages based exactly on what we have from the trial and based on their ApoE status.” (Site 1, ID 3, geriatrics, MD)

Many clinicians discussed the common side effect of infusion reactions, which can cause distressing symptoms like shortness of breath or require treatment with potentially high‐risk medications like steroids. Fewer discussed the logistical burden of lecanemab and its impact on quality of life.

“It's not an easy drug. It's going to be highly inconvenient. You have to go in every two weeks and get an infusion. You can have infusion reactions. You have to get MRIs with some regularity over the course of the first year. You have to be vigilant about your own symptoms. It's going to have an impact on your lifestyle.” (Site 7, ID 23,neurology, MD)

Some respondents explicitly asked patients about their goals. Others mentioned that patients’ treatment goals or hopes came up organically during conversations.

“I don't know that they ask that question in those terms, ‘What are your hopes and your goals for the treatment?’ They usually say they want to get better… If people ask me for a treatment, they say, ‘What's going to help with these symptoms?’ I have to tell them this is not going to help with your symptoms, necessarily…They'll take anything that makes a difference.” (Site 6, ID 20, neurology, MD)

A few clinicians prompted people with AD to share their hopes for treatment to shape the decision‐making discussion.

“I always talk about what are the most important things you want to do, and are you still doing them? I've also had some patients where they say, ‘Well, I really want to drive,’ and I think, ‘That's great, but you haven't been driving for like six months, and no amount of lecanemab is going to make you drive, right?’” (Site 2, ID 8, ‐neurology, MD)

3.2. Patient contextual factors and personal goals and values shaped communication and decision‐making

Clinicians reported incorporating individual patient factors to personalize discussions. Comorbidities that could affect eligibility, risk of side effects, or benefits were an important—not always openly discussed—consideration.

“I had a 91‐year‐old ask about this, and they had a lot of medical comorbidities. Even though disease was mild, I was leery of recommending the drug. We don't know how that vasculature is going to respond, and I think they had some vascular comorbidities, so that was harder.” (Site 7, ID 24, neurology, MD)

Many clinicians structured communication around the fundamental question of whether the patient meets eligibility requirements from the trial and the appropriate use criteria.

“I make sure that patients are eligible as per the criteria…I have made a rule that I always write in my notes, if they are eligible for this new treatment, that I discuss that with them, and if they are not, why not.” (Site 4, ID 14, neurology, MD)

The degree of caregiver support influenced how clinicians communicated about anti‐amyloid therapy. Lack of adequate social support was seen as a deterrent to pursuing treatment.

“A dicier problem is their social situation and support system. I have some patients who don't have somebody like a spouse or a well‐connected son or daughter. If all these bad side effects come up, I worry that something will get missed, they'll end up with a more serious complication than they would normally have. That affects how much I'm willing to go down this path with them.” (Site 6, ID 22, neurology, MD)

In addition, family members played key roles in patient decision‐making. In some cases, family members “pushed” the patient to proceed with treatment, which raised concerns about the balance between caregiver input and patient autonomy. In other cases, discussions with family members resulted in patients choosing not to pursue treatment.

“Sometimes families, because they want to try to get this therapy for patients, may overstate what the patient's able to do at home or their functional status. I just wouldn't be sure if they really understand what they're getting themselves into? Do they really understand the risks, or is their family more the one pushing them into getting this treatment?” (Site 4, ID 15, neurology, MD)

“I did say this to one person, ‘You are probably as ideal of a candidate for this drug as I can find in my practice.’ However, when I talked about the risk of other conditions like heart attack and blood clots, even though this person did not have those risk factors, when they talked with their family member about the risk of potentially not treating those comorbidities, they decided not to proceed.” (Site 3, ID 9, geriatrics, MD)

Clinicians also perceived that patients’ hopes or fears influenced decisions. In some cases, the relative weight of the fear of side effects informed treatment decisions. In other cases, the desire to “do something” to slow progression outweighed the risks.

“They have fears about the side effects—about a brain bleed or about edema—but they fear the future of burdening their spouse. That is probably the most predominant thing. That fear far outweighs the risk of a brain bleed to them or not being able to get thrombolytics.” (Site 3, ID 13, neurology, APP)

“What [patients] articulate is they really want to feel like they're doing something to fight and that they value the time they can have in this mild phase of impairment. Where their quality of life and interactions with family members are good. They sometimes say things like, ‘I would rather die of side effects knowing that I tried to treat the disease. That's a risk I'm willing to take.’” (Site 6, ID 20, neurology, MD)

3.3. Clinician attributes and ambivalence influence communication and recommendations

Clinicians cited their professional backgrounds as influencing how they communicate. They highlighted differences in how the fields of geriatrics versus neurology view these therapies, as well as how prior training affects their comfort level.

“Especially in the geriatrics community, I think there is almost a conspiratorial kind of thought about this drug. I think part of that was just related to the debacle around Aducanumab. I think all of that poisoned the well with respect to this drug class.” (Site 1, ID 1, geriatrics, MD)

“I feel more comfortable administering this medication. I think my General Neurology background helps that. I can read scans maybe more than some of my colleagues who haven't had that exposure or experience catching bleeds or edema. I'm more comfortable with the neurologic exam and helping people with sudden onset of transient neurological symptoms.” (Site 1, ID 2, neurology, APP)

Equally important was the clinicians’ personal practice style. Their sense of themselves as an “optimist,” “patient‐centered,” or as encouraging autonomy shaped discussions.

“I'm an unapologetic optimist. I got into this field because I want to see a cure. I don't see why we can't cure this disease.” (Site 6, ID 21, neurology, MD)

“I am a huge proponent of patients making an informed decision. As a clinician, my role is to provide patients with as much unbiased, evidence‐based information about lecanemab therapy as possible and really allowing them to make those decisions for themselves. I have patients asking me, ‘What would you do if you were in my situation?’ I really defer back to them because I feel like it's a case‐by‐case basis. I can't tell them how much risk that they want to take and how much time they want to invest in this therapy.” (Site 4, ID 15, neurology, MD)

Participants reported that their sense of ambivalence affected how and what they communicated. In some cases, this was due to pessimism about the drugs’ efficacy when compared to burdens and side effects. In other cases, clinicians worried that the benefits were potentially underestimated but that there remained too many unknowns for a firm scientific opinion.

“Most of my patients that are considering treatment will ask me the question, ‘Would you recommend this to your mom, your loved one?’ I'm being honest, in a lot of cases, I would say no.” (Site 2, ID 7, geriatrics, MD)

“The majority of my conversations ended up the person saying, ‘I don't think that this is the drug for me.’ I worry a lot because I watch that window closing in people that we decided last summer not to start the drug and one year later they're progressing. Did I advise them wrong? Was there something that I could've given them a year ago? If some other colleagues are right, that in scrubbing away that amyloid the big bang is going be the difference in 4 years, I feel really bad if I let this person continue on a trajectory that is a worse slope than they would've had otherwise.” (Site 1, ID 3, geriatrics, MD)

Clinicians universally respected decisions to pursue therapy. Most, however, emphasized that they do not routinely “recommend” it, but rather focus on education and empowerment. Conversely, clinicians were comfortable recommending against treatment when they worried the risks outweighed the benefits.

“I don't recommend anti‐amyloid treatment for anybody—they're supposed to come away ambivalent.” (Site 3, ID 10, neurology, MD)

“I have had some patients who are borderline, and I have concern about how much benefit they're going to have. I've presented the options and let them know why I would have reservations about it, and I haven't had anyone who elected to pursue it.” (Site 2 ID 6, neurology, MD)

4. DISCUSSION

To our knowledge, this is the first study evaluating how and what clinicians communicate in discussions with people with AD and their caregivers about these novel disease‐modifying agents. We found significant variation in how clinicians presented risks and benefits and incorporated patient values. Communication approaches were predominantly influenced by patients’ other health conditions, fit with eligibility criteria, social support and involvement, while a few clinicians discussed patients’ goals. Clinicians reported that conversations were shaped by their professional background, practice style, and sense of uncertainty. Amidst a changing treatment landscape, our study provides preliminary data on real‐world communication and SDM practices. If confirmed, these exploratory findings could inform future interventions to improve communication and decision‐making for anti‐amyloid mAbs (Figure 1).

Already, appropriate use criteria, professional societies, and experts recommend a shared deliberative process for treatment decisions, and our results shed light on how this is being implemented in practice ⁷ , ²² , ²³ , ²⁴ , ²⁵ This includes objective and personalized explanations of risks and benefits and an examination of patients’ values and preferences. Importantly, although better health outcomes may ensue from SDM, such as patients feeling less conflicted, its core mission is to fulfill the ethical right of patients and caregivers to be informed and involved participants in healthcare decisions. ²⁶ This is particularly true for people with early dementia, in whom impaired cognition can affect understanding, retention, and judgment. ²⁷ , ²⁸ , ²⁹ Drawing upon preserved cognitive abilities to involve patients in decision‐making may help maintain identity, dignity, and quality of life, whereas excluding them could exacerbate depression, frustration, and disability. ³⁰ Furthermore, our study revealed that caregivers played key roles in patient decisions. This is in line with prior studies describing their important role in providing practical and emotional support to patients with cognitive impairment and in helping interpret information. ³¹ Accordingly, models of SDM for anti‐amyloid therapy should involve clinicians, patients, and caregivers.

In this study, clinicians described using features of evidence‐based communication for SDM. The use of analogy can improve clinician‐patient communication across serious health conditions. ³² , ³³ , ³⁴ , ³⁵ For example, among physicians discussing treatment options for patients with metastatic cancer, the use of metaphor was associated with improved patient ratings of clinician communication. ³⁶ Some participants in our study reported using graphical depictions to illustrate concepts. High‐quality evidence suggests that presenting probabilistic information in graphical format enhances people's understanding and can help prevent biases from affecting their decision‐making. ³⁷ , ³⁸ Finally, a few clinicians reported setting the stage by eliciting patients’ health and treatment‐specific goals and using these to scaffold further discussion, a central tenet of SDM. ³⁹

We also identified possible opportunities to include proven SDM strategies. Healthcare providers play a crucial role in SDM by building trust, using understandable language, and creating opportunities for patients and caregivers to share their values, worries, and questions. Some clinicians were overly optimistic about the seriousness of side effects, while others were pessimistic about the medications’ demonstrated efficacy. Anti‐amyloid mAbs may merit more equanimity. Patients may reasonably decide to forgo treatment with anti‐amyloid mAbs to avoid the logistical hurdles or pursue it to preserve independence. ⁴⁰ We also found an overreliance on technical language—including referencing the CDR‐SB outcome measure—which can interfere with patient understanding. Although clinicians generally agree that straightforward language is best, studies demonstrate that physicians use jargon 70 times per visit. ⁴¹ Clinicians who received feedback were receptive and used less of it subsequently, supporting training to facilitate communication. ⁴² , ⁴³ Overestimation of patients’ numeracy—or ability to understand and manipulate numbers—is another common pitfall observed in this study. ⁴⁴ Effective communication of quantitative information involves presenting absolute risk estimates using frequencies (rather than relative risks), highlighting incremental risk over baseline risk, and emphasizing the risk time interval. ⁴⁵ A minority of clinicians reported asking about patients’ overall goals and instead relied on their feelings about treatment (ambivalence or enthusiasm), professional experiences, practice style, and patient expressions of their hopes that arose organically. In older adults or those with serious illnesses, placing treatment decisions within the larger context of a patient's values and health trajectory is a critical aspect of ensuring that pursuing treatment is worth the potential risks or tradeoffs. ⁴⁶ Moreover, rigorously designed, multimodal interventions that include communication training on eliciting patients’ goals have been shown to improve the quality of SDM conversations. ¹⁴

This study has limitations inherent to the design and data. This qualitative analysis interviewed clinicians about their perceptions of their communication about risks and benefits. Their recollections may not accurately reflect what is said in conversations and are hypothesis‐generating. Our group's future work involving observations and audio‐recordings of clinical encounters will add to the robustness of these findings. Second, we did not have access to information on patient outcomes, so we cannot draw conclusions about implications for treatment decision‐making or clinical care. Third, all participants were subspecialists practicing at academic medical centers; the majority were trained in neurology, and together had a racial makeup fairly representative of academic neurologists; these provider characteristics generally reflect those most involved in the early phase of anti‐amyloid therapy implementation. As access to treatment expands, caution is warranted in generalizing these findings to other specialties, institutions, and practice types. Nonetheless, we interviewed both physicians and APPs across multiple subspecialties, from seven geographically diverse academic sites that serve urban, suburban, and rural patients. Future work should examine communication and decision‐making in a greater diversity of settings (i.e., non‐academic or primary care) to evaluate whether these themes reflect clinicians more broadly.

The availability of anti‐amyloid mAbs is a paradigm shift in dementia care. The results of this exploratory qualitative study are a first step in supporting clinicians, patients, and caregivers in making engaged and goal‐concordant decisions about treatment.

CONFLICT OF INTEREST STATEMENT

The authors report no financial conflicts of interest. Author disclosures are available in Supporting Information.

CONSENT STATEMENT

All human subjects provided informed consent for this study.

Supporting information

Appendix 1. COREQ checklist

Appendix 2. Master coded thematic statements

Supporting Information

Parks AL, Thacker A, Dohan D, et al. Characterizing clinician communication with patients about lecanemab: A qualitative study of clinicians across seven academic medical centers. Alzheimer's Dement. 2025;11:e70150. 10.1002/trc2.70150