A Qualitative Study Investigating Clinician Experiences with Lecanemab Implementation Across Seven Academic Medical Centers

Abstract

Background:

The recent approval of anti-amyloid therapies (AATs) for Alzheimer’s disease (AD) has introduced new complexities into clinical care. While offering therapeutic promise, AATs necessitate changes in practice models, infrastructure, and team responsibilities. Understanding clinician and system-level adaptations is critical to guide implementation.

Objectives:

To explore how clinicians adjust clinical workflows and care processes when integrating anti-amyloid treatments into Alzheimer’s disease management.

Methods:

We conducted semi-structured interviews with neurology and geriatrics clinicians across seven academic medical centers to explore perspectives on AAT implementation. An interdisciplinary team used hybrid deductive-inductive coding and thematic analysis to identify themes.

Results:

Twenty-seven clinicians (17 neurology, 10 geriatrics) participated. Three themes emerged: (1) Structural adaptations: Sites developed governance committees, eligibility protocols, and workflows for referrals, administration, and monitoring. Some hired new staff, and protocols evolved with experience. (2) Diagnostic shifts: Clinicians reported heightened pressure for earlier, more precise diagnoses, prompting greater biomarker use and structured disclosure visits incorporating treatment discussions. (3) Cultural change: While clinicians described optimism about disease-modifying therapies, they expressed ethical concerns about resource allocation, over-medicalization of early dementia, and diversion of resources from broader dementia care.

Conclusion:

The findings reflect evolving institutional protocols, clinical roles and responsibilities, and ethical tensions within the clinical workforce in response to AAT administration and access to early diagnosis and treatment.

Introduction:

The anti-amyloid monoclonal antibodies (mAbs) lecanemab and donanemab are the first disease-modifying therapies to receive full FDA approval, and the first approved therapy of any kind in more than 20 years, for the treatment of mild cognitive impairment and mild dementia due to Alzheimer's disease (AD).¹ These therapies slow progression of cognitive decline in early-stage AD, offering a new avenue for therapeutic intervention.^1-4 However, they are associated with significant adverse effects, including infusion reactions as well as amyloid-related imaging abnormalities with edema/effusion (ARIA-E) and hemorrhage (ARIA-H), which are often asymptomatic but can occasionally be disabling or fatal.⁵ Furthermore, treatment requires substantial logistical coordination, including infusion administration, intensive monitoring with serial MRIs, side effect surveillance, and multidisciplinary care, requiring significant changes to clinical practice.⁶

Prior studies of implementing new treatments, such as targeted cancer therapies and immune checkpoint inhibitors, have demonstrated that successful implementation requires changes to clinician workflows, practice personnel and resource infrastructure, and clinical management processes. ^7,8 These challenges are amplified when treatments carry complex risk profiles, require specialized diagnoses and treatment monitoring, or demand new care coordination pathways, as is the case with anti-amyloid therapies. Failure to anticipate these changes can lead to suboptimal patient outcomes, inequitable access, and clinician burnout.⁹ Therefore, systematically describing how anti-amyloid therapies are impacting clinical practice — from logistics and resource demands, workflow adjustments, and clinicians’ experience — presents multiple opportunities. These learnings can provide a deeper understanding of how health systems might achieve sustainable, accessible, and high-quality implementation strategies for anti-amyloid therapies presently, and can inform the future implementation of novel therapies in other fields that change traditional care models. Furthermore, unpacking the experiences in the dementia care specialties are particularly interesting given that novel anti-amyloid therapies are the first treatments of their type to be widely used.

To date, limited research has examined how clinician perceptions about AD diagnoses and treatment may be changing and how clinicians adapt their workflows as they incorporate anti-amyloid therapy into clinical practice. We conducted a qualitative study to characterize the structural changes and practice adaptations that clinicians navigate as they implement anti-amyloid therapies into the care of individuals with Alzheimer’s disease.

Methods:

Study Design and Data Collection

This qualitative study used semi-structured interviews of health professionals and ¹⁰adhered to the Consolidated Criteria for Reporting Qualitative Research (COREQ) (Appendix 1).¹¹ The study protocol was approved by the Mass General Brigham Institutional Review Board (2023P002711). All participants provided verbal informed consent and completed a demographic survey.

Qualitative Data Collection

Between August and October 2024, we conducted 27 semi-structured interviews of a purposive sample of physicians and advanced practice providers from neurology and geriatrics clinics at seven academic medical centers in the U.S. Five participants declined interview participation (84% response rate). We did not record reasons for refusal. Participants were deemed eligible if they worked in specialty memory care clinics and evaluated patients for lecanemab eligibility, prescribed lecanemab, or supported implementation efforts. This study had two primary aims. The first aim (reported previously)¹² explored clinician experiences communicating with patients and caregivers about lecanemab, including discussions of benefits, risks, and shared decision-making processes. The second aim (reported in this paper) explores lecanemab implementation, such as clinic infrastructure, interdisciplinary coordination, and practice changes. These interviews focused on lecanemab since they were conducted before donanemab availability.¹³ Interviews were digitally recorded, transcribed verbatim, and de-identified. Participants were recruited from seven U.S. academic medical centers in the Southeast, Northeast, West, and Midwest via direct email outreach and snowball sampling. We recruited clinician participants at each site through professional networks of the research team and subsequently used snowball sampling, whereby initial participants recommended additional clinicians for invitation. AP (female MD hematology physician researcher with extensive qualitative research experience) or AT (male BS clinical research coordinator with extensive qualitative coding experience) emailed each potential participant directly with an invitation to participate, including a study fact sheet describing the purpose and procedures of the interview. Interviews were scheduled with those who agreed to participate. At the beginning of the video conference or telephone call, AP or AT conducted verbal informed consent and completed the demographic survey with each participant. Upon completion of these components, the interview began. Interviews were conducted remotely and recorded via Zoom or telephone for 40-60 minutes by trained study personnel (AP, AT, DD (male PhD sociologist with extensive qualitative research experience)) using a semi-structured interview guide (Appendix 2). No repeat interviews were conducted. We used ‘information saturation’ to determine when to cease data collection, or the point at which the interviews produced no novel insights or new information.¹⁴ Transcripts were not returned to participants for comment and/or correction.

Data Analysis

We used thematic analysis for this study. The primary qualitative research team included three authors: AP, AT and JP (female MD palliative care physician researcher with extensive qualitative research experience). AP, AT, and JP coded an initial five transcripts independently using inductive coding. They then met to discuss codes and resolve differences by consensus to create an initial codebook before individual coding of remaining transcripts. Using constant comparative analysis, AP, AT, and JP met regularly to reconcile differences, refine definitions, and ensure a consistent and focused interpretation, ensuring the codebook aligned with the prespecified aims of the analysis. The final codebook was then systematically applied to the full dataset. AP, AT, and JP, the three initial team members, then met regularly with DD, SS (male MD hospitalist), and SG (male MD neurologist) to synthesize the coded data into themes and sub-themes, which were iterated and refined until the final themes were reached. While we did not maintain formal analytic memos for each interview, we systematically recorded analytic reflections through debrief discussions after interviews, larger study team meeting notes, and documentation of coding decisions in line with the consensus-building approach we have described. We incorporated these reflections into codebook refinement and thematic development to make sure that interpretations were transparent and systematically applied. The team also engaged in reflexive discussions to acknowledge and mitigate potential bias.¹⁵ We did not conduct formal member checking of the findings due to logistical constraints since we interviewed busy clinicians with limited availability across multiple academic centers. Instead, we ensured rigor through multiple coders, iterative codebook development, and regular analytic consensus meetings and reflexive discussions.¹⁶

Data Availability

The data supporting the findings of this analysis are available from co-authors upon request.

Results

Characteristics of the 27 participants in Table 1. Most clinicians were Women (63%), White (78%) and physicians (74%) trained in Neurology (63%)

Table 1.

Participant Characteristics

Characteristic	n (%)
Practice: Region in the U.S.
Site 1: Mid-Atlantic	4 (15%)
Site 2: Mountain	4 (15%)
Site 3: Mid-Atlantic	5 (19%)
Site 4: Northeast	4 (15%)
Site 5: Mid-Atlantic	2 (7.4%)
Site 6: West	3 (11%)
Site 7: Midwest	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%)
Primary Role in Practice
Refer people to receive therapy	13 (48%)
Administer/prescribe therapy	11 (41%)
Both	3 (11%)

The analysis revealed 3 major themes and associated subthemes: (1) Structural changes to support implementation of lecanemab; (2) Practice transformation around dementia diagnosis; (3) Shifts in clinical culture and practice with both benefits and burdens. Representative quotations are presented in the manuscript text. Table 2 provides additional supportive quotations.

Table 2:

Illustrative Quotes categorized by theme and sub-theme

Theme	Sub-theme	Representative Quotes
Theme 1: Structural changes to support implementation of disease-modifying therapies	Institutions implemented protocols and standardized governance structures for eligibility determination	Well, we adapted our protocol almost directly from the clinical trial protocol. A lot of people at our center were involved in the appropriate use criteria, so that guided the adaptation of it into the clinic. Then, I would say, we made a couple, perhaps, idiosyncratic decisions that were unique to our center. Like, for instance, to treat atypical forms of Alzheimer’s disease which were excluded from the trials. That was an example of an adaptation that we just made internally after discussion. Our protocol was actually developed <institution> wide, so there was a group of people that went and met with representatives of the <institution> system and that had a protocol approved at the <institution> office at the president level, which was intended to harmonize what this would look like across all campuses. Though, I think each center is still making their own sort of unique decisions on how they—depending on their availability, et cetera, et cetera. I would say it leaned heavily on the clinical trial protocol, the appropriate use criteria, and then the experience of the people at our center. (Site 6, ID 21, Neurologist, MD) There’s not an individual clinician that makes the decision, but rather, we have a committee called the amyloid infusion board, which is similar to a tumor board. It is a multidisciplinary committee that includes Neurologists and other clinicians, but also Neuropsychology, radiology, nuclear medicine, ethics, and we meet every twice a month and review all the individual referrals from the practice and discuss the clinical criteria. We review the biomarker results. We review the safety results, the APOE genotype, and then the committee discusses the case. I think in all but one instance we’ve reached consensus about whether to offer therapy. There was one instance where it went to a vote because we weren’t able to have full consensus….Which is that they have a large number of clinicians seeing patients, evaluating patients with cognitive complaints, and the expertise of the clinicians is highly variable in terms of their knowledge and experience with implementing biomarkers in their practice and understanding this class of drugs. Also, their level to really engage clinically. As a practice, we felt that it would be most fair to patients, and probably offer them the best care if we had a more uniform approach. Rather than having 30 odd different clinicians with different understanding, expertise, maybe bias making decisions and the patient care is dependent on the clinician that they see. (Site 6, ID 20, Neurologist, MD) I saw in a post hoc analysis, zero benefit for the E4 homozygotes. So to me it was a no brainer that with that much of an increased risk and literally no benefit….We do have several pretty high grade Fazekas Scale people, um, getting infused. And so that’s something I’m keeping an eye on. And if it looks like we’re getting more ARIA in that population, I might vote to, yeah, move the needle on that. More recently, though, I mean, in terms of now that it’s available, what are the guidelines and how do we modify them over time? (Site 5, ID 18, Neurologist, MD) I think our institutional policies, there’s policy with a capital “P’ and there’s policy with a lower case “p”. The (capital ‘P’) policy is basically you have to be a dementia specialist in Neurology or Geriatrics to prescribe the drug. You have to have had an amyloid PET scan and our order set all looks fine. It’s the lowercase “P” policy the things that we’ve all sort of agreed upon as a group that I think is dynamic and changing. There’s no institutional policy against giving this to somebody with multiple lacunar infarctions. That’s just sort of what we’ve agreed upon right now. That is more subject to change. I think we have a conversation every week where we see our thinking shifting in a dynamic landscape because of more information, more conversations, more experience, more comfort things like that. I think in general we all align pretty well. “ (Site 3, ID 10, Neurologist, MD) As a practice, we felt that it would be most fair to patients, and probably offer them the best care if we had a more uniform approach. Rather than having 30 odd different clinicians with different understanding, expertise, maybe bias making decisions and the patient care is dependent on the clinician that they see. (Site 6, ID 20, Neurologist, MD)
	Offering treatment to patients on anti-platelet agents and anti-coagulants is a ‘moving target’	In more recent meetings that we’ve had, some now are saying certain medications would be okay, so there’s a lot of debate on that too, so I feel like keeps changing. Things keep changing, and the limitations that we had set up before may not completely apply as some groups are being more free with…different anti-platelet medications and blood thinners, but for the most part, I do tell patients that if they were to need TPA or similar medications to that, that they would not be able to have those treatments, which I think is a harder stop than some other medications that even in recent meetings with this group of professionals that we meet every week. Even last week, we already had a lot of back and forth into what we will allow or not, or what some other centers in the country are allowing or not, so we are learning as we go. (Site 3, ID 11, Geriatrician, MD) Well, see I’m not taking anybody off anticoagulants. We’re not really making that separation or comparison in our group. We think it’s important to reduce those risks. Stroke is a big risk…There’s one thing about TPA and administration of TPA where I think you potentially can deliver something that could increase somebody’s risk when it’s not necessary. If it turns out it’s ARIA—that’s a point where it’s really important to kind of discuss with emergency room. Like I said I was supposed to have a meeting at 1:00 today to put measures in place that alert emergency personnel that they’re on this medication that increase bleeding risk and then also let all the stroke people know that they really need to know if this is ARIA or not. (Site 5, ID 19, Neurologist, MD) For example, aspirin, Plavix, we’ve been, I’ve bene administering with as long as they’ve been on stable dosing for that at least 90 days. But people who are on Eliquis, Coumadin, first on our at <clinic> we said no, okay, because of that risk we don’t know yet and that could be high. We have had people recently have procedures, let’s say they have atrial fibrillation and that’s why they’re on Eliquis. Our procedures that would then allow people to discontinue use of anticoagulants like the watchman device or the left atrial appendage and we had cardiothoracic surgeons meet with us during our monthly meetings about that option for some of our patients and if they would like to pursue that. (Site 1, ID 2, Neurologist, APP) I’m not a hundred percent certain that the overall risk related to just being on anticoagulant plus the anti-amyloid therapies, I think it’s a little bit—I’ll just say slightly overblown. In terms of overblown in the sense that …they have a higher risk of having a fall in a subdural than they do of developing a bleed from being on anti-amyloid therapy and being on anticoagulant. However, again, with my bias towards let’s do it the right way, I am hesitant to administer the medicine when people have a real strong indication to being on anticoagulation, especially for the duration of therapy or even longer. (Site 2, ID 8, Neurologist, APP) At the moment, we don’t administer anti-amyloid therapies to patients on anticoagulants. The rationale is …[that] very few patients on anticoagulants have been exposed to these medications in trials. The people that were exposed were probably a bias sample that were selected by clinicians….Which means that they were probably otherwise, very little risk….We’re still being very conservative and don’t offer the therapy. We are looking at the data as they come in and it does mean that other patients who otherwise might be interested in therapy and might be good candidates for therapy are excluded, and that is something that weighs on us for sure. (Site 6, ID 20, Neurologist, MD)
	Hiring new clinicians and staff to manage referrals, patients on treatment, and to develop and adapt workflows	We’ve hired two new APPs just to accommodate this, and repurposed one that was more research is now getting half clinical time. It’s taken a lot of staff time. We would love to get another full-time physician clinician, It’s taking more staff. I don’t think our clinic wait times have exploded as yet, probably because we have been able to add this APP capacity, but it certainly could. Financially, we’re Neurologists, we lose money. I think our hospital has been helping support our clinic staff in part because I heard a couple of weeks ago, we have generated 1,000 more MRIs than they would have done otherwise. (Site 5, ID 18, Neurologist, MD) I think it’s probably increased referrals because, people who have previously not even wanted to get a workup then become interested in getting more testing because now there is a disease-modifying therapy available. It certainly has changed our workflow, especially in those patients who do end up pursuing, uh, lecanemab, um, because, uh, those patients obviously need to be followed up, and not just followed up but followed up frequently. And it is a lot to keep track of, so we’ve had to make a totally new workflow where we have our nurse who has to keep track of all these different lecanemab patients, where they’re at in their infusions, their MRIs. (Site 2, ID 6, Neurologist, MD) We’ve decided that after the first MRI they just get a phone call from our nurse navigator. After the second MRI, which is at basically at three months, we do in person with our nurse practitioner and again at 6 months after the 13th scan. We’re deciding whether everybody needs a nine-month appointment or not, you know, without an MRI, or if people can go six months. I think we’ll probably for the first, 20 people who hit that milestone we’ll probably have them see, uh, some the NP again at 9 months, uh, and then back to the treating physician at 12. And-and there’s the additional logistical hassle the timing of updated MoCAs and FAQs and CDRs to get certain insurances to continue at that six-month point where there’s not supposed to be a break in treatment for six weeks. (Site 5, ID 18, Neurologist, MD) We’re trying to set up more automated computer system, ways of interacting with the electronic medical records to be able to screen things. Yeah, I think some of the other onerous tasks are things like ordering MRIs, ordering the surveillance MRIs, making sure someone’s following up on the surveillance MRIs to provide clearance if they needed an MRI before their fifth or their seventh infusion, some things like that, which are administrative tasks and we’re trying to figure out the best way to organize that to be most efficient. (Site 4, ID 14, Neurologist, MD) Yeah, I would say our nurses and our schedulers. So, we have one specific—- nurse who has an Excel spreadsheet with each patient and all their information, um, and, uh, so kind of just keeping a bird’s eye view of-of all these patients and everything to make sure things don’t fall through the cracks, that’s her. And then, um, MRI ordering is something I do at each visit, our scheduler has a big burden too because she has to work with the patients to schedule the, uh, MRI review visit right between—after the MRI, but before the next dose of lecanemab. And then our pharmacist is also aware of the-the patients, the infusion center, um, keeps track of some of the patients, and then our-our nurse—I-I-I have our nurse put in the lecanemab, um, order set for me for the—whichever series of doses we’re on. (Site 2, ID 6, Neurologist, MD)
	Increased burden and logistical tasks for clinicians	It’s been certainly demanding of our time, of our energy. There’s a lot of—the logistics goes with bureaucracy. It’s not easy to implement a new treatment like this with the logistics associated with this from scratch. We had infusions for other drugs in Neurology, but not this long-term treatment: 18 months every other week. Typical infusions in Neurology is IVIG, five days, and that’s it. I think…it’s been really demanding, especially as we accumulate more and more patients under the treatment. We’re still trying to catch up in terms of trying to anticipate what’s gonna be the demand (Site 4, ID 17, Neurologist, MD) I certainly had to overbook patients often because, there aren’t a lot of behavioral Neurologists and so our-our wait lists are very long. And, so it can be very hard to get into our clinics in general, and then having this big population now requiring multiple visits throughout the year really takes up a big chunk of our time. And it’s sensitive to review those MRIs, meet with our patients, make sure they’re not having side effects, reevaluate them, make sure they can still stay on the drug, it’s a big time commitment and so it means we’ve had to overbook, and add extra slots.(Site 2, ID 6, Neurologist, MD) Our med center has not been very helpful. We have one full time clinical coordinator, who is the Lecanemab person. They, so far, with the limited volume that we’ve had, that’s worked, but I think it’s not sustainable and we’re gonna need more staff to help with the logistics and coordination. Our system is very simple. If I refer someone and they’re approved for treatment and they start treatment, all the infusions are managed by the Lecanemab team. Which is a team of five or six Neurologists. (Site 6, ID 20, Neurologist, MD)
Theme 2: Changes in practice norms around dementia diagnosis	Increased pressure and need for efficient, accurate, and biomarker-based advanced diagnostic testing	What we’ve really seen in our practice is a huge increase in the number of biomarker tests that we do. Even though we’ve had access to CSF biomarkers for a long time, and we’ve had some access to amyloid PET….there were relatively few examples of patients in the clinic who were getting clinical biomarkers….I think we’re ordering something like 30 to 35 amyloid PETs a month in our clinic. Now we’re increasingly starting to use some of the plasma measures….I think the use of blood tests will exponentially increase in the coming year, as they become FDA approved and hopefully covered by insurance. (Site 6, ID 20, Neurologist MD) If someone in my pool looks like they might be a candidate, I do try to expedite their referral, just because, by the time they see me at six months from the time when the referral is placed, and then by the time they get all their diagnostics in order to determine if they’re a candidate, that could be a year from the time they express concern to their primary care provider, probably longer given the lag time on just when families or patients are asking for a referral from primary care. We really do try to get them in as soon as possible just because time is brain. (Site 3, ID 10, Neurologist MD) I think things are very different because there impetus to accurately identify patients who do have Alzheimer’s disease, rule out other things and then initiate treatment ‘cause time is-time is brain. But we try with varying success I think to identify patients who are referred, uh, to see us who might be in that category of patients who sound like maybe they might have Alzheimer’s disease…and I see those patients very quickly, and then we order all the tests sim-simultaneously, and so I have patients who are starting on therapy this month who I first saw in April. And so, I think that’s-that’s very different, um, from before. (Site 2, ID 5, Neurologist, MD) I’d say the patient population seems about the same from my perspective. I do think more individuals are more interested in diagnostics in that sense. I think it may be too soon to say the field in general, how that’s changing going forward otherwise, but definitely seeing patient preferences change a little bit. (Site 2, ID 7, Neurologist, MD)
	Integration of discussions of anti-amyloid therapy into diagnostic disclosure discussions	Actually, in the very beginning now—before I even go into all of it—I ask them what their expectation is: why are they here, and what do they need from me? That’s interesting because there are some people who will come to me wanting the lecanemab. They’re very clear about it. Then, that’s different than the patient who may just come in for evaluation, but they may be appropriate for the lecanemab. I think I just—I have to be honest with you. (Site 1, ID 3, Neurologist APP) The way that I go about making that decision, is, I describe to patients as we’re talking through this, Lecanemab is a highway, and there’s sort of a lot of exit ramps off of that highway. The first step for people is a proper ideologic diagnosis. Are we even dealing with a thing that could be treated with this? I have to convince myself that you have mild cognitive impairment or mild dementia due to Alzheimer’s disease, and if you don’t, you’re off the highway. (Site 3, ID 10, Neurologist, MD)
	Multi-visit, incremental process for determining eligibility, sharing test results, and discussing the treatment	There’s three to four to five visits related to testing and interpretation that need to happen if you’re really on this path that we’re talking about.” (Site 6, ID 22, Neurologist, MD) Let’s say my colleagues that referred them to me for a specific consent visit, they don’t know me yet. Even though they did all their homework, they got all their testing done and that colleague thinks they’re a good candidate, I also need to review that with them. I also want to make sure that they’re diagnostic criteria fits that they are actually within a good range for their testing. That they are mild dementia, not moderate or severe dementia. That their families are also in alignment to what’s going on. I do spend quite a bit of time with people and it’s not uncommon that we’ll then call people after that visit if they’re hesitant to move forward. (Site 1, ID 2, Neurologist APP) Even just to assess eligibility for this medication, we need to do blood work, we need to do brain MRI, we need to do an amyloid PET scan, APOE genotyping. With APOE genotyping, we need to get consent for genetic testing. There’s all this counseling about tests and then following up about all the results and things like that, that ends up taking a lot more time. Then there’s counseling about the medication itself, which itself could take an hour to do, to answer everyone’s questions. Then we have the way we’ve set up our clinic…if I want to refer someone for lecanemab, I need to refer them to this [specialized] clinic where even if I’ve done all the workup in terms of their eligibility, it’s a visit where they cross the t’s, dot the I’s, make sure everything’s, all in order in terms of eligibility. Then they get referred to the infusion center. (Site 4, ID 15, Neurologist MD) It’s overseen at <Clinic> by three physicians and I am one of them, and if we are responsible for all the aspects of the coordination from the referral to triage to scheduling them into the appropriate clinic, if they are from one of our colleagues and the diagnosis is clear, everything is check, check, we have a workflow for that. If the diagnosis is still not completely clear, then we need to get biomarkers, genetic testing on that, we have another workflow, and from there, we go into getting into the process of getting lecanemab, if fully eligible, getting the consent and doing all the logistics, insurance approval, coordination with infusion center appointments, MRIs and all that. (Site 4, ID 15, Neurologist MD) All of my stuff is video visits through the identifying them. Once they agree to it, that’s care planning visit is a video visit. What we have been doing is, I meet them in the clinic on the day I asked them to come about 30 minutes earlier. I consent them, and it really takes about 45 minutes to an hour. They have a lot of questions, so while the nurse is starting, they come 30 minutes early. I meet with them. I get the consent form, then do the education by ARIA and contacting us side effects afterwards. I just sent an email to our docs and just said, “This is not working.” ’Cause I get four and five phone calls between the time that they agree to lecanemab to the time that they come to clinic, and these are not 20-minute phone calls. These are 45-minute phone calls, and people will call me three and four times for those three or four weeks leading up to it because they’ve read stuff. (Site 3, ID 13, Neurologist, MD)
Theme 3: Shifts in clinical culture and paradigms with both benefits and burdens	Increased hope and fulfillment for the present and future of Alzheimer’s disease treatment and dementia care	I think it’s also given us, more hope and made our jobs more fulfilling because there are options now. I think it’s shifted our practice that way. Patients are slow—slowly starting to hear about this. So I do occasionally now have patients who come in who’ve heard about it, who’ve read about it, who are interested in it. I think it’s required a big culture shift for us. I think it’s also had a lot of benefits for us because, you know, there’s some hope that there’s something that could actually help. I think probably a lurking fear for all of us working in this areas that, you know, like our predecessors, we would spend our career, you know, treating patients, waiting for something to come out that could actually make a difference, and end up retiring before that happened. (Site 1, ID 1, Geriatrician MD) It’s had a lot of impacts and we’ll, I think, continue to—ultimately, it’ll change the way we practice as we have more of these medications, and hopefully it’ll get to the point where we have patients managed which will be a whole different world than it would have been for the last 30 years. (Site 7, ID 24, Neurologist, MD) I’ve never had this long of a wait time to be seen. I think what’s really exciting about having these medications and getting more clinical trials is the fact that I feel like we finally made it. Being in neurology for almost 20 years, it’s nice to have treatment options. We really get to do good things now. Not that we didn’t do before, but we’re getting deeper into pathophysiology and how the brain works. I’m just really excited about that. (Site 1, ID 4, Neurologist, MD)
	Changed relationships with and time spent with patients at diagnosis	The APPs do most of the minutes spent in management of the patients. I think it’s certainly increased their relationship with the patients they’re seeing more often and hearing from more often. For our physician clinicians, we get to have these conversations about treatment, which is great, but we’re not doing the six-month visit. We’re taking some of the calls but not a ton. I guess that’s one of the, again, coming back to the beginning of our discussion, of how to increase capacity. For most of us, we like having those relationships with our patients and seeing them over time. That’s becoming harder as there’s just not enough of us. If we want to see all the new patients, then we don’t get to see the follow-up patients. That’s a challenge. (Site 7, ID 24, Neurologist, MD) In terms of care for patients, what I would say is that it actually ends up taking, my clinic visits end up being a lot longer or the number of clinic visits I need to have with patients are a lot longer for a few reasons….There’s all this counseling about tests and then following up about all the results and things like that, that ends up taking a lot more time. Then there’s counseling about the medication itself, which itself could take an hour to do, to answer everyone’s questions. (Site 4, ID 15, Neurologist, MD)
	Dealing with urgent or acute medical situations is new / different	I’ve added time. We’re actually in the clinic during the infusions, which is new. I’ve never been an inpatient or acute situation as a practicing provider. That made me feel a little bit uncomfortable of dealing if someone had some sort of major reaction during the infusion. I learned a little bit about that. (Site 3, ID 12, Neurologist APP) I think, it’s been quite a culture shift for us. I think forever, we basically had two drugs that we give people. the most dangerous thing we prescribe is an antipsychotic and now we’re giving people an infused medication that can cause brain swelling, brain bleeding and death. So I think we’ve had to get used to that idea, so I think that’s been a bit of a culture shift. (Site 1, ID 1, Geriatrician MD) It’s more like delivering a chemotherapy for cancer than anything else. I think as Neurology departments, we’re not really usually in the business of that and starting that requires a lot of communication across departments including radiology, pharmacy, even emergency room services….The emergency room people are very interested in having all this guidance about it. (Site 5, ID 19. Neurologist MD)
	Medicalization of early dementia care and impact on the field and patient care	I feel like sometimes I worry that our clinic has really turned into a, "Are you eligible for lecanemab? Are you interested in lecanemab? If you are, you go down this very intensive tree." If you're not, it almost feels like - "Make way for…this huge volume of people that we're trying to move through to make these treatment decisions." (Site 1, ID 3, Geriatrician, MD) I’m pretty sure my, uh, colleague has as well—had to overbook patients often because, you know, there aren’t a lot of behavioral neurologists, and so our-our wait lists are very long. And, so it can be very hard to get into our clinics in general, and then having this big population now requiring multiple visits throughout the year really takes up a-a big chunk of our time. And it’s sensitive to review those MRIs, meet with our patients, make sure they’re not having side effects, reevaluate them, make sure they can still stay on the drug, it’s-it’s like a big time commitment and so it means we’ve had to overbook, and add extra slots. (Site 2, ID 6, Neurologist, MD)

Theme 1: Structural changes to support implementation of lecanemab

Clinicians across all institutions noted that they implemented structural changes to support the delivery of lecanemab. First, institutions established clear protocols to guide patient eligibility that initially adhered closely to enrollment criteria (inclusion/exclusion) from the lecanemab Phase 3 trial (CLARITY AD) and/or published, expert consensus-based appropriate use recommendations. However, clinicians noted changes to aspects of their protocols over time, such as allowing patients taking certain other immunotherapies concurrently to be treated.

“We started with the Clarity AD protocol. The other sources have been the appropriate use recommendation paper. We all started really rigidly around this…because we’re all scared of hurting somebody, understandably. We’ve seen some relaxations of particular points along the way. Everybody [initially thought]- you can’t be on another monoclonal [antibody]. We’re still insisting people not take Prolia, but other places are saying “It’s probably fine”. Those are the things that I’ve seen really start to shift over the last six months.”

(Site 3, ID 10, Neurologist, MD)

Most institutions established committees or equivalent governance structures to develop and implement standardized protocols for lecanemab eligibility to enable consensus-based decision-making and to promote fairness and safety in therapy administration. Clinicians also noted that these committees meet regularly to deliberate and discuss patient cases, which some compared to a ‘tumor board’ (e.g. a multidisciplinary team (MDT) approach employed in oncology care to discuss patient cases and therapeutic plans).

“We have a multidisciplinary committee that includes neurologists and other clinicians, but also neuropsychology, radiology, nuclear medicine, ethics, and we meet twice a month and review all the individual referrals from the practice and discuss the clinical criteria. We review the biomarker results. We review the safety results, the APOE genotype, and then the committee discusses the case. I think in all but one instance we’ve reached consensus about whether to offer therapy. There was one instance where it went to a vote because we weren’t able to have full consensus. As a practice, we felt that it would be most fair to patients, and probably offer them the best care if we had a more uniform approach.”

(Site 6, ID 21, Neurologist, MD)

Clinicians reported details about eligibility from their institution’s treatment protocols, reflecting some similarities and differences across centers. For example, several (but not all) institutions exclude patients who are APOE e4 homozygotes while others offer them treatment but with more stringent criteria based on neuroimaging findings. Many clinicians also noted that they are offering lecanemab for patients on antiplatelet agents. While some clinicians described protocols that systematically excluded patients on anticoagulation due to the increased bleeding risk, others described institutions’ criteria regarding anticoagulants as a ‘moving target.’ Overall, clinician reports reflected both institutional and individual variability and that the protocols are evolving as they learn through implementation of the treatment.

“Things keep changing, and the limitations that we had set up before may not completely apply as some groups are being more free with …different anti-platelet medications and blood thinners. But for the most part, I do tell patients that TPA, if they were to need TPA or similar medications to that, that they would not be able to have those treatments, which I think is a harder stop than some other medications.”

(Site 3, ID 11, Geriatrician, MD)

“We decided not to offer it to people who are APOE4 [homozygous]. We do offer it for people on anticoagulation, which is a little bit different than some places just because there has been some data……Of course you inform people about those things, but it seemed to us that the APOE4s were more high risk that we thought maybe the benefit wouldn’t necessarily outweigh the risk if we aren’t even sure how much benefit there is.”

(Site 5, ID19, Neurologist, MD)

Finally, clinicians described the initial and ongoing need for clinic structural, personnel, and workflow changes to accommodate the increasing population of patients on infusion therapy, the clinical and logistical needs that accompany referrals and medication administration, and the ongoing intensive monitoring of patients for therapy-related risks. Some systems hired personnel for new roles, such as a nurse navigator or advanced practice clinicians to oversee management of patients. In other cases, clinicians noted that their institution has not provided additional support that was needed or requested. Combining these findings, some noted that the introduction of disease-modifying therapies has added burden and pressure on clinicians and staff due to increased workload, which feels heightened since their institution has not provided adequate compensation or support.

“How this has impacted my job is I would say that there’s more uncompensated navigational work because the healthcare systems frankly don’t care that we’re doing this on our backs without support and would like for us to demonstrate to them that it’s effective, all the while, generally, providing benign neglect for support staff.”

(Site 3, ID 10, Neurologist, MD)

“We have a nurse navigator that was, part of my agreement with the health center… I told them that I couldn’t do this in my division without additional support. The nurse navigator has to keep careful track of everybody who’s going through the process from insurance authorization to the timing of MRI scans.”

(Site 5, ID 18, Neurologist, MD)

“We hired a bunch of nurse practitioners and a P.A. to manage a lot of this stuff. We’ve had to hire several office staff to help with all the coordinating the MRIs and ordering things and dealing with insurance. The whole clinic has expanded into a much bigger operation…For the actual providers—we’re fortunate enough here that we have some support from our healthcare system and from our [Alzheimer’s Disease Research Center] that we can actually hire more people to help.”

(Site 7, ID 23, Neurologist, MD)

Theme 2: Practice transformation around dementia diagnosis

Clinicians emphasized that the implementation of anti-amyloid therapy has shifted clinical practice norms around dementia diagnosis compared to before their availability, including an increased perceived pressure for timely and accurate diagnoses. The steps to determine patient eligibility, combined with the short time window for anti-amyloid treatment for only early symptoms, drive processes to evaluate and confirm the specific cause of patients’ mild cognitive impairment or dementia syndrome more quickly than before. Clinicians noted their increased use of, and insurance coverage for, biomarker-based diagnostic modalities in clinical practice, such as amyloid PET scans or other biomarker testing. These tests are both required by widely accepted anti-amyloid treatment eligibility criteria but also add the specificity needed to diagnose Alzheimer’s Disease independent of treatment.

“Now we have pressure that we didn’t have before, the pressure of making accurate diagnosis. Before, it was like, “Okay. This looks like probably Alzheimer’s disease. Here’s this pill for…your memory.” If it was not Alzheimer’s disease, it didn’t matter that much….The negative consequences of misdiagnoses, it was not that grave….Now we need to be more accurate. We have to use all these biomarkers to confirm the diagnosis. We need that—to do that in a timely fashion, because there’s a time window for this treatment. Once you pass that time window, then it’s not approved.”

(Site 4, ID 17, Neurologist, MD)

“Actually, for me, it really has changed the outlook of the treatment that we do for Alzheimer’s disease because the diagnosis has changed. Now, we have new modalities that Medicare is paying for, like PET scans that before we only did in research and now we can do it in clinical practice. For me actually, that’s been great to have that availability of new resources that we can use for diagnosis because it adds a specificity.”

(Site 4, ID 14, Neurologist, MD)

Clinicians described the ways in which discussions about anti-amyloid therapy have become intertwined with diagnostic disclosure conversations, in part because of the multi-appointment process of determining eligibility and its overlap with diagnostic testing and disclosure. The process involves scheduling and obtaining biomarker and genetic tests, waiting for and disclosing those test results, disclosing a diagnosis, and discussing the risks and benefits of treatment. Clinicians noted how important AD biomarker results have become in diagnosis-centered conversations about mild cognitive impairment and dementia and that the boundary between diagnosis and treatment (AAT eligibility) determinations have become blurred.

Now that it’s approved and that we have more biomarkers, the work about diagnosis and biomarkers, you can’t really not talk about the disease modifying therapy these days. They have to be in the background. Why would we disclose this to a patient? What will we do when we do? How we make decisions, how lecanemab gets used in the community, is an important part of the assessment process now.

(Site 6, ID 22, Neurologist, MD)

When I see a new patient, and I’m doing my assessment, if at the conclusion of that assessment, we are discussing Alzheimer’s disease because I’m concerned that that could be contributing to their symptoms, then when I’m discussing the option of testing, I will first ask if they’ve heard about the drug. I’ll briefly just mention that say, you know, the testing is consistent with Alzheimer’s, and I’ve already kind of reviewed to see if they would meet eligibility criteria based on the appropriate use guidelines already.

(Site 1, ID 1, Geriatrician, MD)

Finally, clinicians outlined structured workflows and ‘visit types’ they have developed to steward patients and caregivers through the process of determining eligibility, decision-making and consent for treatment, if that becomes the chosen path. In addition to the prolonged time it takes to determine eligibility, clinicians viewed the incremental, multi-step eligibility determination process as a way of ensuring fully informed decision-making. They felt it was important that patients and caregivers understand the potential benefits, risks, and implications of treatment, and that the multiple clinical interactions over time help patients and caregivers make the right decision for them.

“I’ve seen the patient three or four times, because there’s an initial evaluation, and then we, order testing, and then I do a lumbar puncture, and so usually when we’re having this discussion after the lumbar puncture….then… we’ve looked at the MRI…and the PET scan, and we have biomarkers which can say yes or no, this is Alzheimer’s disease. Then I can talk about the risks and benefits of treatment.”

(Site 2, ID 6, Neurologist, MD)

“There’s a three-step process before they get the drug… There’s an initial one, and then there’s one that happens with our head nurse. Then, there’s another one that happens with the physician and with our coordinator that we’ve hired to do the antibody treatments….At one point, the health system was pushing on us and saying, “Couldn’t you reduce some of these.” The doctors that set this up said, “Nope. That’s a feature, not a bug, of our process. We want people to have three different opportunities. These are cognitively impaired people. Repetition is important. We wanna make sure that people hear it every single time.”

(Site 1, ID 3, Neurologist, MD)

“And I intentionally space this out because on average I’m having three or four visits with people before we consent to start treatment because I really want to make sure that they understand what this involves, what the risks are, the logistical considerations, and that’s something they really want to pursue. We can have as many of these sessions as-as you need to feel confident that this is the right decision for you.”

(Site 1, ID 1, Geriatrician, MD)

Theme 3: Shifts in clinical culture and practice with both benefits and burdens

Clinicians in this study described the ways in which the discovery and implementation of disease-modifying therapies in Alzheimer’s care brings a new sense of hope and fulfillment for clinicians and patients due to the ability to intervene and potentially slow cognitive decline, counteracting longstanding feelings of nihilism in Alzheimer’s care. Clinicians’ expressions of hope reflect both the new treatments that are available now for patients as well as potential for future treatments down the line. In addition, clinicians cited recent historical parallels of treatment advances, such as in stroke and oncology care, that illustrate the sense of hopefulness that these Alzheimer’s therapies have created. Clinicians also described that the availability of anti-amyloid therapy has increased public awareness of the disease and potentially motivated individuals to seek an earlier diagnosis and adopt lifestyle changes to preserve brain health.

“I think it being in the most general sense, just having these drugs available is helpful and hopeful for patients. I certainly tell all my patients, whether they’re eligible or not, that this is the first disease-modifying therapy we’ve had, and there’s more to come. I think it does imbue a sense of hope that we have been lacking for a while.”

(Site 7, ID 23, Neurologist, MD)

“This reminds me a lot…[of] the transition from, “Okay. You have stroke-like symptoms. This is a stroke….we’ll do a CAT scan today, but MRI we’ll do in two days. Here’s your aspirin. We’ll transfer you to rehab,” to the era of intravenous TPA, where you have to three or four hours. Then, from that era to the thrombectomy…You apply the protocols, the guidelines, and a lot of disability has been prevented. Hopefully, with these treatments for Alzheimer’s, we’ll be able to slow down the progression. That’s the hope that I have, personally.”

(Site 4, ID 17, Neurologist, MD)

Participants also noted that the rollout of disease modifying therapies is a ‘culture change’ in the field. The culture change was described in three ways. First, clinicians described that the ability to intervene has changed their role in patient care and the relationships they have with patients, especially early in the illness course. For example, for patients with newly diagnosed mild cognitive impairment or mild dementia, clinicians see or speak to those who are on anti-amyloid therapy significantly more than before therapy was available, given the frequent visits the therapy demands. Second, administration of these medications has resulted in the need to navigate urgent and acute medical situations similar to managing patients on chemotherapy in oncology, which is generally new for dementia care, save for managing acute neuropsychiatric decompensation in advanced dementia. For example, this involves the following: responding to urgent calls about infusion reactions, side effects, or other clinical symptoms; interpreting frequent MRIs and discussing results with patients; and responding to the sense of urgency to start treatment. Third, the treatments have fostered more need for cross-specialty collaboration due to the multi-disciplinary needs of patients.

“For my patients, it’s definitely changed the relationship in the sense for a tertiary care expert like me, even though I provide care, it’s fairly light touch, except when big things come up….Even then, it’s often best for primary care to deal with that in terms of doing additional assessment. That used to be what it was. “Hopefully, things’ll go great, but if they don’t, I’m here. Meanwhile, we’ll check on you in a year.” That was a very standard thing to say with somebody with MCI.”

(Site 6, ID 22, Neurologist, MD)

“Behavioral Neurology never had urgencies. We were a place where you could be referred and it’s nine-month wait, no big deal. There was no treatment anyway, which I’m being facetious about a bit ’cause I do think that we palliatively help people significantly but there is no urgency really. All of a sudden, there is a sense with some people that this is like a cancer diagnosis, and you have a cancer treatment that you can’t even offer me for nine months. There’s a perception of this really important ability to intervene.”

(Site 3, ID 10, Neurologist, MD)

Clinicians also described some ethical concerns related to the ‘over-medicalization of early dementia care,’ as clinicians grapple with shifts to professional identity and the ongoing, competing need for comprehensive dementia care beyond just pharmacologic treatment. One clinician described this as ‘losing the soul of our field,’ which represents a concern that increasing focus on administering anti-amyloid therapies like lecanemab may affect core neurology practices, such as therapeutic alliance-building, holistic support, and nuanced clinical observation, thus risking a loss of the field’s foundational ethos or “soul.” Clinicians described lack of time and bandwidth to focus on aspects of care that are important for dementia management even with early disease, such as discussing caregiver coping, patient safety, and current and future care planning. In other instances, clinicians noted that the shift of resources and time toward a small population of patients who are eligible for disease-modifying treatment may come at the expense of the care of the broader dementia population who may not be eligible or interested in treatment. It important to note that individual clinicians in this study expressed both optimism and concern, holding nuanced views of lecanemab and new disease modifying therapies. They acknowledge both the hope it offers to patients and themselves along with concerns about shifts in the field.

“I would say that there are times I worry that we don’t have enough time or bandwidth now for dementia management. We used to do so much….for a disease that doesn’t have a pharmacological cure, we do an awful lot—we do a really good job taking care of patients and families because a lot of visits were spent with what a lot of times is what the family is craving, which is just, “Give me a trick or a tool or the suggestions of how I can make bath time less of a battle,” or “how I can make—how we can have a conversation about driving,” or, “what do we think about in the future in terms of financial planning and home planning.”

(Site 1, ID 3, Neurologist, MD)

- You kind of lose a little bit the soul of the field. A lot of it is neuro-palliative care. It’s building a therapeutic alliance, preparing somebody for the process of having a chronic disease, building support structures, healthy coping strategies, all of those things. It’s careful clinical phenomenology, this observation of behavior and localization of these complex cortical behaviors like language and appetite and all these other things that we had, that we’ve built up because we’re trying to understand the disease. I’ve had this conversation where I hope we don’t kind of lose our soul as a field. Now, just ’cause we have a handful of partially effective treatments, I hope we don’t lose that completely, but I feel myself being pushed in that direction, 100 percent. I don’t see a lot of new patients anymore because most of my time is actually just managing patients that are referred in and are getting the Lecanemab therapy.”

(Site 6, ID 21, Neurologist, MD)

Discussion

The availability and implementation of anti-amyloid therapies have significant implications for dementia care. Our findings demonstrate that institutions have started to implement structural and workflow changes to support anti-amyloid referrals and treatment programs. Clinicians also experience changing roles and norms around dementia diagnosis as well as shifts in clinical culture and practices. Clinicians compared these changes in early dementia care to an oncology-like care model, characterized by biomarker-driven diagnoses, subspecialty consultation, structured referral and treatment pathways, and intensive monitoring for therapy-related risks.

The study identified key structures and processes to support anti-amyloid therapy implementation across seven academic institutions. Most systems developed standardized governance structures and eligibility protocols^17,18 to oversee lecanemab administration, including interdisciplinary committees that are similar to ‘tumor boards’ in oncology.¹⁹ These committees determine patient eligibility criteria for treatment and deliberate patient cases. Such structures were implemented to ensure safety and fairness in access to treatment, to enable consistency and transparency in eligibility determination, and to align with existing consensus-based clinical guidelines.¹⁸ Prior research in oncology, for example, has demonstrated the benefits of multidisciplinary committees, including enhanced adherence to clinical guidelines, improvements in treatment outcomes, and overall improvement in decision-making processes.¹⁹ Most clinicians also noted that their healthcare systems hired and/or redeployed administrative staff and advanced practice clinicians to manage patient referrals, implement workflows for eligibility determination and informed decision-making, and oversee lecanemab infusion-related processes and monitoring for risks longitudinally.

There are several implications of our findings. Such structural changes signal a shift to a more specialized and medically intensive model of early dementia care focused on disease modification. Furthermore, clinicians and staff in dementia therapeutics are developing and refining specialized expertise and skillsets that may help them deliver the most optimal and safe care. Gaining a real-world understanding of the infrastructure, workflows, skillsets, and evolving ‘best practices’ in delivering therapy can support ongoing cross-system learning and information exchange.²⁰ However, the specialized diagnostic infrastructure, staffing and resources needed to support anti-amyloid delivery will likely exacerbate disparities in access to treatment, including to individuals in rural settings and racial and ethnic minorities who already face significant disparities in dementia care.^21,22 As implementation continues to mature, decision-makers in health systems needs to focus their attention on reducing these disparities to ensure access to available treatments for all.

This study demonstrates that the roles of neurology and geriatrics clinicians are actively evolving as they navigate and integrate new and complex clinical discussions in early dementia care. For example, clinicians noted that administration of lecanemab has catalyzed emerging shifts in the diagnosis of Alzheimer’s dementia from a clinical-biological syndrome to a mostly biologically-defined one, increasing the use of expanded diagnostic criteria, blurring boundaries between diagnosis and treatment, and changing communication practices in routine care. These changes will be further amplified with emerging use of blood tests to detect tau and amyloid proteins in plasma, beginning with FDA’s clearance of the first blood test in May 2025.¹³ Corroboration of a diagnosis by blood testing promises significant scalability and, ultimately, reduced costs and burdens compared to PET scans and lumbar punctures. In addition, anti-amyloid therapies, with their modest efficacy and associated risks, generate hope and therapeutic momentum while also carrying risks and tradeoffs. Clinicians integrate nuanced discussions of the potential benefits, harms, and tradeoffs of treatment at the time of diagnostic disclosure, further changing norms around communication for these specialists.

To qualify further, the availability of disease modifying therapies have shifted clinicians’, patients’, and caregivers’ mindsets about therapeutic possibility. These changes introduce new ethical questions related to anti-amyloid decision-making due to the ‘pressure’ that clinicians can feel to discuss therapy and that patients can feel to pursue it. Prior studies show that patient wishes to start a diagnostic process for cognitive symptoms vary.²³ Patients and families may now be more willing to start down a dementia diagnostic pathway if there are desirable, expected outcomes of that process, including treatment options. It is this therapeutic possibility that is likely to bring more people with memory complaints earlier to medical attention. Furthermore, our prior study of patient experiences revealed that some patients felt compelled to pursue treatment to ‘do something’ rather than ‘nothing.’²⁴ This mirrors prior studies of patients pursuing immunotherapy in cancer care which have demonstrated significant hope and the desire to ‘fight’ the cancer, sometimes at great personal risk or cost and without always having a full understanding of the benefits and risks.^24-27 Clinicians in our study expressed cautious and nuanced understanding of the benefits of anti-amyloid therapies, which mirrors the recognition and serious engagement within neurology and geriatrics about the “grayness” and ethical considerations of anti-amyloid treatment.²⁸ One major implication of these findings is that communication practices, skills, and clinician training will have to adapt to the need for nuanced, complex shared decision-making conversations with patients and caregivers.

The availability of first(s)-in-class treatments for a progressive, fatal disease – which has never had a pharmacologic treatment that alters the disease’s natural course -- creates positive changes in dementia care while also introducing new challenges that clinicians and healthcare systems must navigate. The availability of new diagnostic tests and disease-modifying treatments offer hope for clinicians that such advances will have positive outcomes for patients (both now and in the future), which mirrors the hopes of patients that were described in our prior study. ²⁴ However, implementation of anti-amyloid therapy also presents new tensions for the field related to resource allocation. Similar to sentiments expressed in recent commentary²⁹, clinicians in this study shared thoughtful worries and concerns that directing clinician effort and institutional resources toward a small number of patients who are eligible for and receiving anti-amyloid therapy may potentially detract from essential efforts that would benefit the entire population living with dementia, including those focused on effective diagnosis, communication, care, and management. Notably, only a small subset of patients with dementia (8-20%) are eligible for anti-amyloid therapies.³⁰ Care efforts that are essential for all patients affected by dementia, for example, include careful symptom and behavioral management, promotion and facilitation of brain health and wellness, unhurried discussions about diagnosis and prognosis, and provision of personalized care planning, counseling, and support for caregivers.^31,32

Clinicians in this study, while reflecting on their practice independent of and before anti-amyloid therapy, described a sense of pride in their delivery of holistic and comprehensive care for patients and caregivers who are affected by an illness that carries significant stigma and burden. Similar to other fields, the cultural narrative of disease-modifying and curative-oriented care risks overshadowing essential supportive services to meet the ongoing psychosocial and practical needs within dementia care. The disproportionate investments of dementia care resources in anti-amyloid therapy implementation (above other dementia care strategies) benefits individuals in early diseases stages and, at least initially, individuals from higher socioeconomic status, which will likely exacerbate existing disparities in dementia care access and outcomes.³³ Furthermore, health systems may prioritize AAT delivery over other critical services that may benefit larger socioeconomically diverse populations, like dementia-capable primary care, caregiver support, care coordinators, behavioral management, and community based dementia care.^34-36

Strengths and Limitations:

While this study highlights clinician perspectives from seven geographically diverse academic medical centers across the United States, these sites are among the early adopters of anti-amyloid therapies, often benefiting from greater institutional resources, specialist expertise, and infrastructure. This concentration in well-resourced, academic settings represents a key limitation. As such, findings reported in this study may not be generalizable to other healthcare institutions or contexts, given that implementation challenges, patient selection processes, and clinician attitudes may diverge meaningfully from those described here. Furthermore, because this study used a purposive convenience sample to elicit a range of perspectives rather than a representative sample, it was not designed to compare and contract findings across sites or participant groups. Additionally, we interviewed physicians and advanced practice clinicians involved in prescribing anti-amyloid therapy, which may not reflect the full range of perspectives particularly those of nurses, social workers, care coordinators, and other frontline team members who play critical roles in patient education, monitoring, and care coordination. Similarly, the perspectives of patients and caregivers were not the focus of this analysis; however, these were explored in a related study from our group examining patient experiences with anti-amyloid therapy. ²⁴ As this study centered on lecanemab, findings may not fully capture evolving clinician experiences with all/both anti-amyloid therapies; for example, donanemab differs in dosing schedule, risk profile, and management protocols. Furthermore, the potential approval of subcutaneous administration for donanemab, and extension of approval for lecanemab to include the first 18 months of treatment, may reduce logistical challenges. Querying clinician, patient, and care partner perceptions and experiences of these developments in future studies will be important to capturing a more complete picture of this evolving field.

Conclusion:

The introduction of anti-amyloid therapies marks a transformative shift in Alzheimer’s disease care toward a more specialized, medically intensive model of early dementia care focused on disease modification. Our study demonstrates that implementation of disease-modifying therapies has already prompted changes in clinical roles and relationships with patients, the need to hire new clinicians and staff, shifts in dementia diagnosis and related discussions, the establishment of institutional structures to enact standardized protocols, and new workflows to manage referrals and patients on infusion therapy. Clinicians are navigating these shifts with both hopes and worries for current and future dementia care. As advances in disease-modifying therapies for Alzheimer’s disease and hopefully other dementias emerge, the potential for ‘cure’ must not overshadow the inveterate need for holistic care. Investments in infrastructure and workforce for disease-modifying therapies must be balanced with efforts to provide comprehensive, person-centered care for all individuals affected by dementia.

Data availability

The data supporting the findings of this study are available within the article and/or its Supplemental Material.