Facilitators and Barriers to Patient Enrollment in the Arteriovenous Access Trial of Vascular Access Outcomes

Visual Abstract

Abstract

Key Points

• In a randomized controlled trial studying outcomes of arteriovenous fistulas versus grafts, clinicians and patient participants identified multilevel barriers to patient enrollment.

• Pervasive biases from Fistula First emerged as a major barrier, contributing to a perceived lack of equipoise between vascular access options in older adults.

• Patient education on the importance of access placement, and increased clinician buy-in, may help mitigate barriers to trial enrollment.

Background

People with kidney failure on chronic hemodialysis require sustainable vascular access. The Arteriovenous Vascular (AV) Access trial is a randomized controlled trial comparing clinical and patient-reported outcomes in adults 60 years or older on hemodialysis with a central venous catheter, randomized to receive either a surgical AV fistula or graft. Patient enrollment in the trial initially progressed slower than anticipated. We designed a cross-sectional qualitative substudy to identify facilitators and barriers to patient enrollment in the AV Access trial.

Methods

We conducted qualitative focus groups and semistructured interviews of site investigators and providers and semistructured interviews of patients who had been approached for enrollment in the AV Access trial. We performed hybrid inductive-deductive thematic analysis of transcripts.

Results

We analyzed transcripts from two focus groups (one with four vascular surgery physician investigators and one with four nephrology physician investigators) and 44 individual interviews with vascular surgery physician investigators (n=2), nephrology physician investigators (n=2), vascular surgery providers (n=9), nephrology providers (n=10), and patients (n=21). Participants identified themes related to clinical site selection, buy-in from clinical teams, patient referral processes to vascular surgery, and processes to recruit patients to enroll in the trial. Subthemes included barriers related to patients receiving care from providers external to the study team, pervasive biases from the Fistula First initiative, limited patient knowledge of the importance of AV access, and a need for additional information provided to patients to support informed decision-making about trial participation.

Conclusions

Enrollment in the AV Access trial was hindered by multilevel barriers. Some of these may be mitigated through collaborative patient education surrounding the importance of access placement, as well as provider-directed education to increase buy-in and enhance engagement. These findings can inform the design and implementation of future randomized controlled trials concerning surgical AV access outcomes.

Clinical Trial registry name and registration number:

The AV Access trial has been registered at ClinicalTrials.gov; NCT04646226; registration date: November 20, 2020.

Introduction

People with kidney disease on chronic hemodialysis require vascular access to receive hemodialysis treatment. Arteriovenous (AV) fistulas and grafts have been favored over central venous catheters (CVCs) because of lower risks of infection and mortality compared with CVCs.^1–4 AV fistulas were historically believed to be superior to AV grafts because of prolonged secondary patency and a lower requirement for interventions to maintain patency, contributing to the well-publicized Fistula First initiative, which promoted placement of AV fistulas as the optimal access for all people requiring hemodialysis.^5,6 However, given evidence that a large portion of AV fistulas (30%–40%⁷) do not mature and result in prolonged dependence on CVCs, newer guidelines instead emphasize the right access in the right patient at the right time for the right reason.⁸

Controversies surrounding the ideal access are especially relevant for older adults on hemodialysis, for whom AV fistula maturation rates are lower.⁹ Retrospective analysis of claims data for patients 67 years and older demonstrated a need for more interventions to achieve primary patency of AV fistulas (with a concomitant longer CVC dependence), although fistulas had a lower rate of abandonment after use compared with AV grafts.¹⁰ Concrete data surrounding the risks and benefits of each type of vascular access in older adults on hemodialysis are lacking, in part because of a dearth of randomized controlled trial (RCTs) of AV access types.¹¹

The Arteriovenous Vascular Access (AV Access) study is an RCT enrolling adults 60 years or older with at least one major comorbidity who are on hemodialysis with a CVC and have anatomical suitability for both an AV fistula and graft in the forearm or arm. Participants are randomized to undergo surgery for either AV fistula creation or AV graft placement. The trial's primary end point is the number of CVC-free days per 100 patient-days.¹² Recruitment began in 2022 across seven study sites in the United States. After 12 months, enrollment lagged behind expectations, with an observed recruitment rate of 0.29 patients per site per month, compared with the anticipated rate of two patients per site per month.

We conducted a qualitative substudy of the AV Access trial to identify facilitators and barriers to trial enrollment, as viewed by stakeholders (clinical site physician investigators, site providers, and patients who had been approached for enrollment).

Methods

Setting

The AV Access study is a national multicenter RCT, full details of which are reported separately.¹² In addition to the primary end point of CVC-free days per 100 patient-days, the trial evaluates access-related safety events (rates of access-related infection), access-related health care costs, and access-related outcomes (e.g., rates of primary failure, time to cannulation, patency, adjuvant procedures), with data obtained using chart review. Patient-reported outcomes (e.g., satisfaction with dialysis vascular access, health-related quality of life) were evaluated semiannually through direct patient interaction.

Patient recruitment started in April 2022, and by October 2022, seven sites across the country were actively enrolling participants. Each clinical site was associated with two site investigators (one vascular surgery physician investigator and one nephrology physician investigator), vascular access surgery and nephrology physicians (referred to as providers), and a study coordinator. A structured process to evaluate patient eligibility for AV Access trial participation was implemented across all study sites.¹² Patients scheduled for evaluation by a vascular access surgeon underwent an upper extremity vascular ultrasound prior to their surgical consultation. On the basis of the ultrasound mapping results, the vascular access surgical team evaluated patients for study eligibility, specifically confirming the presence of vasculature suitable for both fistula and graft placement. Patients deemed eligible by the surgical team were then approached for study participation by a designated member of the study team, which included either a clinical research coordinator or a member of the vascular access surgery team. Patient enrollment and assessment could take place in outpatient or inpatient vascular surgery or nephrology clinics/services or in outpatient or inpatient dialysis units. Patients could be enrolled by proxy. Subsequently, site-specific practices for vascular access placement were followed.

Before initiating the qualitative substudy in April 2023, two of the original seven clinical sites were withdrawn from the study because of lack of enrollment. In July 2023 and February 2024, two new sites were added.

The qualitative substudy reported here was conducted by investigators at Johns Hopkins University in Maryland and Atrium Health Wake Forest Baptist in North Carolina. Focus group discussions were conducted in July and October of 2023, and interviews were completed by October 2024.

Participants and Sampling Strategy

We conducted focus groups and semistructured interviews with site investigators, site providers who cared for patients eligible for the AV Access trial, and patients who had been approached for enrollment. We initially included English-speaking participants only. In July 2023, we translated interview guides into Spanish and revised the protocol to include non–English-speaking participants.

We aimed to capture perspectives of all site investigators from the original seven clinical sites for trial enrollment. Site investigators from the five original sites which were actively recruiting in July 2023 were invited to participate in a focus group for either vascular surgeons or nephrologists. To avoid potential bias, we excluded the main principal investigator (M. Murea) who was the nephrology site investigator for one site from participation in focus groups. Site investigators from the two original clinical sites which had been withdrawn were interviewed individually because experiences from these sites may have differed from those of sites that remained active in the study.

Purposive sampling was used to identify provider and patient participants for semistructured interviews, aiming to capture a diverse range of experiences of participants across different study sites. Eligible site providers, who were not site investigators and who had active care of patients on chronic hemodialysis, were identified by site investigators and study coordinators. We aimed to interview two nephrologists and two vascular surgeons from each site. Eligible patients were identified by study coordinators. We aimed to recruit at least three patients from each clinical site, with a mix of patients who had declined AV Access trial enrollment, patients who had enrolled but not yet undergone vascular access surgery, and patients who had enrolled and completed AV access surgery. We included patients who were seen at the two new clinical sites added in 2024.

Measures

Using clinical knowledge and informal feedback from the study team surrounding likely facilitators and barriers to trial enrollment,¹¹ focus group and semistructured interview guides were constructed by D.M. Patel, M. Murea, and A.M. Chandler (Supplemental Items 1–6). Open-ended questions were pilot-tested within the study team to ensure clarity, and follow-up probing questions were planned. The qualitative study lead investigator (D.M. Patel) reviewed an initial set of three to five recordings or transcripts for each participant group, and the team revised the interview guides to ensure clarity of questions and alignment with research objectives.

Procedures

Participants were contacted by a member of the site study team to obtain consent and to schedule focus groups or interviews. Participants self-reported demographics, and dialysis vintage was abstracted from medical records.

Focus groups were facilitated by D.M. Patel who is a nephrologist with experience in qualitative research. D.M. Patel is employed in an academic center which participated in trial enrollment, but was not involved in study recruitment processes and had no preexisting knowledge or biases surrounding the trial protocol. Focus groups were held through the Zoom teleconferencing platform. All participants were assigned a participant number. Field notes were taken as needed.

Semistructured interviews were conducted by a member of the Atrium Health Wake Forest Baptist Qualitative and Patient-Reported Outcomes team, with expertise in qualitative methodologies. Interviewers had no established relationship with interviewees. Interviewees were not provided with information about interviewers. Repeat interviews were not conducted.

Focus groups and interviews occurred remotely and were conducted in English. Focus groups and interviews of providers were conducted using a videoconferencing platform, and patient interviews were conducted through telephone. We recorded audio files only, which were professionally transcribed. Provider and patient transcripts were reviewed and compared with the audio recording for accuracy.

Provider participants were not remunerated for participation. Patient participants were remunerated $25 per interview. Transcripts were not returned to participants; however, the qualitative data were summarized for the steering committee and site investigators of the AV Access trial.

All study activities were approved by the central Institutional Review Board of the AV Access trial at Wake Forest University Health Systems (Institutional Review Board #00069593).

Analytic Approach

We thematically analyzed transcripts using MaxQDA and Atlas.ti software to manage data. D.M. Patel and K. White coded focus group transcripts and interviews of site investigators from sites which had been withdrawn from the trial, while A.M. Chandler and A.E. Strahley coded and summarized semistructured interview data from site providers and patients. Coders used both deductive and inductive coding strategies, which allowed the research questions and interview guide to direct the analysis, but also allowed for unique participant perspectives and experiences to develop from the data. D.M. Patel, A.M. Chandler, and A.E. Strahley subsequently considered all coded segments in aggregate and derived themes. Thematic saturation was defined as the point at which review of additional coded segments did not identify new concepts or relationships between data.¹³

Results

Participant Characteristics

We interviewed 12 physician investigators (Table 1). Among these, four—two vascular access surgery investigators and two nephrology investigators—were interviewed individually. These investigators represented two sites that had been withdrawn from the AV Access consortium due to nonrecruitment. In addition, two focus groups were held with physician investigators from sites that had been actively recruiting in the trial for over a year at the time of the qualitative substudy. One focus group included four vascular surgery investigators, and the other comprised four nephrology investigators. Each focus group meeting was approximately 1 hour long.

Table 1.

Participant characteristics

Physician investigators (n=12a)	(N=52)
Age in years, mean (range)	50 (40–65)
Sex, n (%) male	6 (60)
Race, n (%)
Asian	4 (40)
Black	2 (20)
Other/multiple races	1 (10)
White	3 (30)
Clinical subspecialty, n (%)
Nephrology	6 (50)
Vascular surgery	6 (50)
Years of practice, mean (range)	16 (6–38)
Providers (n=19)
Age in years, mean (range)b	50 (31–70)
Sex, n (%) male	16 (84)
Race, n (%)
Asian	5 (26)
Black	1 (5)
Other/multiple races	2 (11)
White	11 (58)
Clinical subspecialty, n (%)
Nephrology	10 (53)
Vascular surgery	9 (47)
Years of practice, mean (range)	17 (3–40)
Patients (n=21)
Age in years, mean (range)	72 (60–87)
Sex, n (%) male	16 (76)
Race, n (%)
Asian	0 (0)
Black	7 (33)
Other/multiple races	0 (0)
White	14 (67)
Dialysis vintage in years, mean (range)c	2 (<1 to 8)
Participation in trial, n (%)
Declined enrollment in trial	6 (29)
Enrolled in trial, pre-surgery	4 (19)
Enrolled in trial, post-surgery	11 (52)

^aTwo physician investigators declined to provide their demographics; n=10 for age, sex, race, and years of practice.

^bOne vascular surgery provider declined to provide their age.

^cDialysis vintage data were available for 19 patient participants.

We individually interviewed nine vascular surgery and ten nephrology providers from sites which were actively recruiting, and 21 patients who had been approached for enrollment in the trial (Table 1). Among patient participants, six had declined enrollment in the AV Access trial, four had enrolled but had not yet had their vascular access surgery, and 11 had enrolled and underwent their vascular access surgery as determined by trial randomization. Compared with patients randomized in the AV Access trial who did not participate in this qualitative study (Supplemental Table 1), those who were interviewed were more likely to be White and male. On average, interviews lasted 17 minutes (range 7–35 minutes).

Summary of Findings

We identified key facilitators and barriers aligned with sequential steps required to recruit patients for the AV Access trial (Figure 1). Themes relevant to each step are outlined below.

Figure 1.

Overview of themes relevant to key steps required for patient enrollment in the AV access trial. AV, arteriovenous.

Clinical Site Selection

Site investigators identified barriers related to characteristics of study sites (Table 2). Some sites had a smaller population of eligible patients than initially anticipated. One site investigator commented that their institution is successful at placing AV access in patients before starting dialysis, which results in fewer patients receiving hemodialysis with a CVC. In some sites, patients receive care in hemodialysis units staffed by external nephrology providers not affiliated with the participating institution, and some are referred to vascular surgeons who practice at external sites, both of which were perceived to result in fewer opportunities to recruit patients for the trial.

Table 2.

Themes surrounding clinical site selection

Barriers
Some sites have a limited population of eligible patients	• “…historically, we thought that there were more 65-plus patients with catheters going for access, and then ultimately this proved us wrong” (vascular surgery physician investigator from dropped site) • “…I just think that the people who meet the criteria are not that common, meet the screening criteria are uncommon, at least in our population. The only way would be to increase that pool, i.e., screened, and I don't know how you would do that, other than partnering with, I don't know, dialysis facilities or other surgeons, other nephrologists, other referring physicians outside of our institution, asking them to refer patients specifically for the study” (vascular surgery physician investigator)
Some sites care for patients who are seen by external providers	• “For us, the main issue is that not all the dialysis units are staffed by our nephrologists. That causes an inability to look at the dialysis-related information, which is required for the trial. If we did not have to gather the data which is related to the dialysis treatments, then that would have helped us recruit more patients…” (nephrology physician investigator) • “…most vascular access patients, in the past 2 yr, have been redirected for surgeries to regional sites of the [medical institution]. The initial IRB approval that we got was only for the main clinic, for the main campus of the clinic. If we had known a year ago that there is this drive for patients being sent to regional sites for vascular access creation […] in retrospect, I think we would have tried to get regional surgeons involved and trying to get IRBs for other sites from the start” (vascular surgery physician investigator from dropped site)

Participant characteristics are provided. IRB, Institutional Review Board.

Buy-In from Clinical Teams

In general, site investigators and providers acknowledged the importance of the AV Access trial and the potential impact of its findings (Table 3). Providers stated that RCT-based data on AV access outcomes for older adults on hemodialysis could guide clinical decisions and potentially help avoid procedures which are less likely to be successful. Participants identified site investigators as having a key role in reinforcing each site's participation in the trial, especially given the large positive effect of having buy-in from all vascular surgery providers to facilitate enrollment.

Table 3.

Themes surrounding buy-in from clinical teams

Facilitators
Providers acknowledge the importance of the trial	• “If there is clear data, that one is better than the other, then there will be a reduction in health care costs, financial burden, overall. I mean, even for patients—right? It's fewer hospitalizations, more days at home, more days stable, lower chance of worse things happening to them…” (nephrology provider 9) • “I think clarification of this decision about what is the best initial option for someone who has an option for both. Can we better clarify 6-mo, 1-, 2-yr expectations of success between a fistula or a graft in someone who could have had either? Are we seeing that difference in durability and functional patency that we have a bias toward with the fistulas, or does that really not bear out?” (surgery provider 7)
Site investigators can facilitate buy-in among providers at each site	• “We had one situation earlier on in the trial where the surgeon felt very strongly of placing a fistula just because of the way the vein appeared on the vein mapping, and they decided not to enroll the patient. What we did was, we discussed that issue with the vascular surgery site PI and requested them to discuss the trial again among their own group to make sure that everyone is on board. Since then, we have not had any other issue in terms of disagreement” (nephrology physician investigator) • “We sat down, as a group of surgeons, and basically said, ‘If you want to be a part of this study, you have to be willing to randomize any patient that meets criteria that passes screening. Even if they've got a big cephalic vein, you can't turn them down.’ We've made a concerted effort to get buy-in amongst the entire surgery group, that they're willing to enroll any patient that meets criteria that passes screening. That seems to have helped us a lot. I think all the group of surgeons now feel like they have buy-in. I think that's—ultimately, I feel like we are the gatekeepers for the trial. I don't think it's the nephrologists. I think it's the surgeons, because we're the ones that sit down and have more conversations about dialysis access with these people than anybody else…” (vascular surgery physician investigator)

Barriers
Biases from the Fistula First initiative are pervasive	• “The larger barriers are that ‘Fistula First’ was very successful in indoctrinating patients, and indoctrinating providers that fistulas are unequivocally superior to AV graft, which is at the core of this study, is the lack of evidence, but there's a perception that there is clear evidence. That's just a function that this campaign, over 20 yr, has succeeded in its initial mission in spite of the evidence, the emerging evidence that the premise is not exactly true and proven in all patients…” (vascular surgery physician investigator from dropped site) • “…Then the large dialysis corporations followed and put out their QIP, the Quality Incentive Program that would reward units on how many fistulas they had, and so surgeons were pressured to get in as many fistulas as possible and decrease catheter rates” (nephrology provider 6) • “…from my standpoint, even if you're an older adult, the choice should always be a fistula and a graft as secondary if you can't place a fistula. I'm concerned that the older adult getting a graft as a first choice, even if they were eligible for a fistula, is suboptimal because of the complications associated with the graft that you don't necessarily see as frequent or as bad with a fistula” (nephrology provider 4) • “I will not put a graft in. I will not recommend a graft in anyone…If you look at the data, if you look at the data of the outcomes, it's quite clear. They're just not good” (nephrology provider 10) • “If somebody has a big, juicy cephalic vein, I think it's asinine that people are putting’ grafts in these people” (surgery provider 4)
Some providers disagreed with some criteria to determine patient eligibility	• “…Where it becomes a dilemma for surgeons is when you get into elderly populations, 75, 80. Those are patients with more of a limited life expectancy where I think most surgeons actually do wrestle with the question the study was trying to answer, whether you should do a graft over a fistula…” (vascular surgery physician investigator from dropped site) • “The age only matters as far as feebleness is concerned…I don't think age in and of itself has an impact. I think more frailty, I guess, is the buzzword nowadays. Right?” (vascular surgery provider 4) • “In the cases where the patient hasn't been able to make their own decision, I've declined to say that their family member could enroll them in the study…sometimes, we do things with an experimental device or something, or an experimental drug, or something that the patient can't get the treatment elsewhere, and this is the only thing that could help the patient, so they could potentially benefit from participating in the study. That's not the case in this situation, and I personally don't feel like it's right to do that” (vascular surgery physician investigator) • “The cognitive ones, the cognitive decline, the ones who can't consent themselves, I back away from, especially in a study that's looking at quality of life and asking questions about quality of life. It's hard, in my mind, to be asking quality of life questions to someone who doesn't even know why they're being asked the questions” (vascular surgery physician investigator)

Participant characteristics are provided. AV, arteriovenous; PI, principal investigator; QIP, quality incentive program.

However, buy-in was limited by pervasive biases from the Fistula First initiative,^6,14 with some providers voicing a strong preference to place AV fistulas whenever possible. As mentioned by one vascular surgery site investigator, controversies surrounding Fistula First may be underappreciated:

“Sometimes in academia, we feel, as specialists, that the information that we have and that we’re premising our research on, is common knowledge. The controversy is common knowledge, but just like if we’re making a presentation, the first slide or two of our presentation is a foundation of why there is a controversy. We have to explain the controversy and then the premise of our study, and then go on to explain the methodology in our outcome. I think in this situation, it’s just unappreciated. Not only how much—how effective ‛Fistula First’ has been, how limited the knowledge is around the existing controversy, but also the impact of the economics that are driving dialysis centers and the nephrologists that manage them to continue to chase the quality performance metrics that underlie reimbursement.” (Vascular surgery physician investigator)

Some providers expressed discordance in beliefs surrounding appropriate patient eligibility criteria, which further limited buy-in. Some providers were unsure if patients should be randomized if they were shown to have an outflow vein suitable for AV fistula placement. Others commented that an age cut-off of older than 75–80 years would be more appropriate than older than 60 years. Additional issues included the relevance of frailty rather than age and the appropriateness of enrolling patients by proxy given secondary trial outcomes focused on patient-reported outcomes.

Referral of Patients to Be Seen By Vascular Surgery Teams

Participants noted that nephrology providers and dialysis care team members could facilitate referral of patients to be seen by vascular surgery teams (Table 4). Nephrology providers may have long-standing relationships with patients which enhance patient trust in their recommendations. Dialysis staff engage in frequent and sustained interactions with patients, offering opportunities to discuss the importance of vascular access. Such discussions may address the noted barrier of patients not having knowledge surrounding the importance of vascular access placement.

Table 4.

Themes surrounding recruitment of patients to be seen by vascular surgery teams

Facilitators

Nephrology providers and dialysis staff can convey the importance of having AV access placed

• “As a nephrologist, we have this relationship, chronic relationship with the patient that, whenever we say something to the patient, they trust what we're saying because they see the vascular surgeon once or twice, but they see us a few times a month or a couple of times a month at minimum. I believe there is more influence from the nephrologist on the patient than the vascular surgeon once we identify the right candidate” (nephrology physician investigator) • “[…] These nurses in these dialysis places have been around this stuff so long that they're pretty knowledgeable about all of this. They really are. They'll answer your questions the best they can—through just discussion. Just talking is the best way that I could describe that I found out about any of this” (patient 27, enrolled in trial)

Barriers
Patients may not have knowledge of the importance of having AV access placed	• “I think one thing that would help us a lot would be if there's actually a concerted conversation by their nephrologist, about why this is important, and why they need to have something in their arm and not a catheter, because I spend a lot of time trying to convince people that they need something in their arm to begin with. It almost feels like they've never had a conversation, because they haven't. They've been just referred. The rounding nephrologist says, ‘Hey, they need access. Send them to the surgeon to get access.’ The patient shows up and says, ‘I don't want anything access. I'm happy with my catheter’” (vascular surgery physician investigator) • “I think in this area of patients, most patients are not very knowledgeable, I think, with all the access options when they're first starting. I think it's a rare patient that knows differences between a fistula, graft, catheter, or hemodialysis, home hemodialysis, peritoneal dialysis…” (vascular surgery physician investigator from dropped site)
Organizational and systems-level barriers can impede access placement	• “…since COVID, we've had trouble keeping our clinics appropriately staffed with both medical assistants and nurses and also with our vascular lab, our laboratory for ultrasounds…same thing's happened in our operating rooms, where we just haven't had the staff that we've needed to really get some of these patients in quickly—as quickly as we should…” (vascular surgery provider 6) • “I attend at an underserved, community-based outpatient clinic, and those patients often don't have full insurance coverage. One of the barriers is just obtaining insurance coverage to get an access placed” (nephrology provider 1) • “The number one barrier is transportation. It's the number one barrier for people not showing’ up to surgery. It's the number one barrier why they sit with a catheter for a year…It's maybe not weekly, but it's at least monthly, we'll have a patient calls up and says, ‘My ride didn't show up’” (vascular surgery provider 4)
Nephrology and vascular surgery teams sometimes work in siloes	• “It oftentimes, sometimes, when I sit in on the meetings every month, feels like we have a fragmentation between surgeons and nephrologists sometimes, even within the study and within our own site, where we're two siloed specialties that are both trying to enroll patients. Sometimes, even within our own institution, it feels like we don't necessarily work together as well as we could, to have a concerted, forward-facing effort to these patients. It's your nephrologist is over here. Your surgeon is over here. We're both trying to get you into a study, but we're not really presenting a unified front sometimes” (vascular surgery physician investigator)

Participant characteristics are provided. AV, arteriovenous; COVID, coronavirus disease.

Participants identified organizational and systems-level barriers to vascular surgery evaluations. Since the coronavirus disease 2019 pandemic, some providers commented that they had experienced significant staffing changes and less availability of operating rooms. As voiced by participants, long wait times to see vascular surgery providers and to subsequently undergo a procedure could have negatively affected trial recruitment (though these wait times were not unique to potential trial participants). Broader systems-level barriers also included patients not having insurance coverage to undergo AV access surgery and/or not having transportation to attend vascular surgery appointments.

Considering multidisciplinary efforts to promote vascular access placement, some site investigators alluded to a need for more coordinated efforts among the study team. One participant stated that a barrier limiting enrollment was the seemingly siloed nature of nephrology and vascular surgery teams, who sometimes did not directly work together to ensure harmonization of recruitment efforts.

Recruitment of Patients to Enroll in Trial

Once in vascular surgery clinics, participants reported that vascular surgery providers may be the gate-keepers of trial enrollment (Table 5). Patients and providers commented that information surrounding access is best appreciated when heard from the actual proceduralist. Persistence of study coordinators in contacting eligible patients and being readily available to obtain their consent in clinic also facilitated enrollment.

Table 5.

Themes surrounding patient enrollment in the arteriovenous access trial

Facilitators
Vascular surgery providers can be the gatekeepers for trial enrollment	• “…I think where we really have found success is with the surgeons coming in and presenting the trial with equipoise, and being able to say, ‘We're not sure what's better. That's why we're a part of this trial.’ I think the surgeons have really been the ones who, I think, have been able to get the patients to be agreeable” (vascular surgery physician investigator) • “…I think we had a bit of an issue with at least a couple of patients where we would poll them about the study, and we would tell them about the study, sort of explain the rationale, what we're expecting to find in the study and how it would help them or other patients, and the feedback that we got from them was that they wanted to hear it from a surgeon…That prevented enrollment of at least two patients that I'm aware of. Again, that again brings in focus the need for greater surgical engagement. I completely understand it, if I was in my patient's shoes, and I was hearing something from a nephrologist about a surgical procedure, I would want to get a confirmation from the person who's doing the cutting” (nephrology physician investigator from dropped site)
Persistence of study team coordinators can facilitate recruitment	• “I do think one thing that has helped us with recruitment is that we mandate that our coordinators come to clinic…if a patient screens in, based on their vein mapping, they're right there, ready to consent them at that time, in the clinic, at that same visit where they're seeing the surgeon…We get everything done in one visit” (vascular surgery physician investigator) • “…When the coordinator's persistent and makes calls and gets them through the screening and consenting, then I think things go much more smoothly. I've found the role of the coordinator to be very valuable” (nephrology physician investigator)

Barriers
Patients may have a predetermined type of desired vascular access	• “The fistula is my own body. The graft is something they'd added in. That's not part of my body. […] Less chance of infection with the fistula, I believe” (patient 6, declined enrollment) • “Quite frankly, the graft appealed to me more than a fistula simply because of the structure of the thing. That's what I listened to and that's what I retained. […] The connection with a—I believe they described it as a Teflon tube connecting a vein and an artery together. That appealed to me because that seemed like a very good way to process the blood flow…” (patient 11, enrolled in trial) • “The issues that we have had with consenting patients or getting patients to agree to the study, we've noticed a lot of our patients come in with a preconceived idea of what they want, from talking to other patients in their dialysis units. I've had patients who have said they—I had one yesterday who said, ‘I don't want a fistula. I don't want aneurysms. I want a graft.’…It's usually people who see aneurysmal fistulas, who say, ‘I don't want my arm to look like that’…” (vascular surgery physician investigator)
Patients may not be presented with sufficient information to decide about trial enrollment	• “…Some people are visual, and as we said, this type of population that we're targeting is mainly elderly who hard to understand—sometimes, when you present ten pages or three, four pages of information, it's hard for them to grasp on all the knowledge being presented. Maybe having a handout to the patient to go through it while they are in the waiting room or they're doing the vein mapping and dwell on the information in a visual manner. That could have—be helpful to illustrate more about the trial” (nephrology physician investigator) • “Keeping flyers for the trial there and directing the dialysis unit staff to give it to anyone who is above the age of 60 and dialyzing with a catheter, regardless of whether they would pass the screening or not, might help to introduce them to the trial, and then they can have follow-up conversations with their nephrologist” (nephrology physician investigator) • “…For the trial, they do not have to do any additional visit. Explaining that to them up-front should help in terms of the time commitment issue” (nephrology physician investigator) • “I talked to the doctor about it, and I don't know that—I kinda got chased out of the room before I really got explained what was gonna happen anyway. […] Well, I asked him, ‘Do you have a—show me a picture of what's gonna happen to my arm or whatever.’ That kinda got swept under the carpet or whatever, and then the next thing I know is somebody else coming into the room wanting to know if I wanna get in this survey or whatever” (patient 17, declined enrollment) • “…we are not the one who are doing the consent, so we're counting on the clinical coordinator or research coordinator, and we don't know what verbiage is being used and what's the conversation being run. We're not in the room with them. That can be a game-changer” (nephrology physician investigator) • “We just came out of COVID, and we know that nuance is not well received by the population at large. Right? People wanna know is it, or is it not? Do I need a vaccine, or don't I need it? Do I need a fistula? Is a fistula better than a graft, or not? It some instances it is, but in other, it's not. That's confusing. That's why ‘Fistula First’ has worked, and that's why it's a heavy lift to try to convey the nuance” (vascular surgery physician investigator from dropped site)
Patients are concerned about time and effort required for participation	• “I think it's more, to me, the patients are not worried about compensation, but time commitment, right? Up front, if you tell them a clear idea of what is involved, it is probably easier. These are the number of visits. This is what we will do. I think they will have a clear understanding and then can make a decision. It's not as much about, I guess, hearing the word research. It's more about how many visits they have to do, what kind of commitment they have to do, things like that. I think being more clear up-front about what actually is involved will help” (nephrology physician investigator) • “[My biggest concern about participation was] the time and the fact that responsibilities here with my family, my wife, that I have to take care of her some, and I just didn't wanna be tied up that much” (patient 26, declined enrollment)
Some patients mistrust research	• “…we have a lot of underrepresented minorities on dialysis in our institution. An underlying mistrust of the health care research system in my area of the country, especially with underrepresented minorities, that permeates through our community in our region of the country, in general. I have noticed that, certainly in Black and non-White patients, that there seems to be—and maybe this is just my own gestalt, but—more hesitancy to be willing to participate in clinical trials, in general, at least in this part of the country. I don't know if that's true, but certainly, my anecdotal experience is that there's a lot more skepticism amongst certain demographics” (vascular surgery physician investigator) • “Well, I thought, maybe, a person should choose what they want, since it's their life, instead of some committee choosing.’ Because a committee, to me, don't always make the best choice for somebody to do. Because to me, it seemed like the program is the experiment type program—that they usin’ people to process what they want to do, to see. You know, more like a guinea pig or something’ like that. […]” (patient 13, enrolled in trial)

Participant characteristics are provided. COVID, coronavirus disease.

Several barriers were identified as limiting patient enrollment. First, participants commented that patients may have a specific desired type of vascular access which would then make them less likely to enroll for randomization. Second, patients may not be presented with sufficient information to decide about trial enrollment. This deficiency may stem from inadequate communication from clinical care teams or alternatively may stem from the nuances of enrolling in a clinical trial for which there is supposed equipoise between study arms. Third, patients may be hesitant to enroll due to time and effort required for trial appointments. Finally, as voiced by a vascular surgery provider, many patients approached for enrollment included underrepresented minorities who may have mistrust of research based upon historical maltreatment of these groups in research studies.

Actionable Steps to Increase Enrollment

Data were used to identify potential actionable steps to increase enrollment (Figure 2). Key components included concerted efforts to educate patients surrounding access placement and provider-directed education to increase buy-in. Patient-directed initiatives included multidisciplinary approaches for issues such as transportation assistance, as well as refinement and use of patient-friendly handouts emphasizing pros, cons, and unknowns of each type of AV access when approaching patients for study participation.

Figure 2.

Actionable steps to increase enrollment in AV access trial, as identified by participant perspectives.

Discussion

In this qualitative substudy of the AV Access trial aimed at studying vascular access outcomes of older adults on hemodialysis randomized to AV fistula or graft placement, multidisciplinary physician and patient participants identified multilevel facilitators and barriers to trial enrollment. We now consider how our findings align with existing literature and reflect on the unequal challenges that trials involving surgical AV access outcomes may present—challenges that may require strategic planning to overcome.

Several barriers identified by participants were related to clinical site selection. At many sites, the number of eligible patients was lower than anticipated—a challenge also reported in other trials.^15,16 However, a unique barrier in the AV Access trial was that eligible patients were often seen by nephrology and vascular care providers who were not part of the study team and, in some cases, were not in agreement with the trial's central question of investigating equipoise between AV fistulas and grafts in older adults. This perceived lack of equipoise may have influenced referral patterns and reduced the likelihood of eligible patients being considered for enrollment. In our focus groups and interviews of physician participants, we asked about scenarios in which the physician felt that a patient would be more suitable for an AV graft compared with a fistula. While some responded that fistulas are always preferred if there is a suitable vein, others implied that patients with older age and/or specific comorbidities (e.g., obesity or diabetes) could be better served by having an AV graft placed (data not shown). Further quantitative data surrounding physician knowledge of newer AV access guidelines,⁸ as well as preferences surrounding AV fistulas versus grafts for older adults, may help determine the feasibility of physician-led recruitment for trials involving AV access.

This issue is particularly relevant in the dialysis setting, where care is often fragmented as patients transition from advanced kidney disease to dialysis.¹⁷ Dialysis units often refer all patients requiring vascular access to specific vascular surgery providers, who may or may not be affiliated with the trial. In fact, hospital systems may be entirely disconnected from dialysis units. Site selection for vascular access trials may need to consider not only the total volume of eligible patients at an academic center, but also the referral networks and engagement of nephrology and vascular surgery providers involved in patient care. Clinical centers in which the dialysis unit, nephrology care team, and vascular surgery team collaboratively manage care for individuals with kidney failure, including documentation of dialysis-related and other clinical data in the same electronic health record system, may be best suited for clinical trials in AV access and dialysis broadly.

Participants emphasized the importance of securing buy-in of clinical teams at each study site. For the AV Access trial, site selection was based on the commitment of one nephrologist and one vascular access surgeon at each location—both with substantial expertise in dialysis vascular access research and clinical trials—who agreed to serve as site investigators. Although all site investigators reported that their colleagues supported the AV Access trial, overall engagement from the local clinical teams was lower than anticipated, as evident by perspectives expressed by site providers who participated in interviews. Traditional assessments of site readiness for clinical trials may not include measures of buy-in from all members of local clinical teams.¹⁸ Especially when the trial randomizes patients to one intervention which was historically viewed as optimal versus another viewed as suboptimal (e.g., AV fistula versus graft), an assessment of buy-in from all clinical team members may reduce the chances of selecting a site in which care teams will not refer patients to the trial. In the AV Access trial, provider engagement was reassessed throughout the study, and site investigators were encouraged to provide ongoing education about the trial. One vascular surgery provider noted that they had ensured all surgeons at their site were willing to enroll eligible patients, an approach that fostered team-wide buy-in and successful recruitment at that site. Similar challenges have been noted in other trials, such as one evaluating treatments for FSGS, which also underscored the importance of securing provider buy-in to support successful recruitment.¹⁹

Next, participants alluded to the importance of patient-friendly study design and recruitment tools. These issues are perhaps even more relevant for patients on hemodialysis who often face a large burden of social needs²⁰ and who already devote a large number of hours per week to dialysis treatments. The AV Access trial design did not require patient participants to have additional clinic visits, though one participant still noted concern that transportation barriers could preclude patients from consistently attending vascular surgery clinic appointments. Difficulty with transportation has previously been noted to limit participation of patients with low socioeconomic status in other fields of medicine.²¹ Others have shown that support for patients to navigate care (e.g., reminders for appointments, transportation assistance)^22,23 could help alleviate some barriers to patient recruitment.^24,25

Ultimately, our data highlighted a need for education of stakeholders surrounding the purpose and need for the AV Access trial and trials in vascular access more broadly. Participants identified the pervasive effect of Fistula First,^7,26 which may limit the willingness of clinical teams to participate in the trial. As reported by others, providers may also inadvertently or intentionally express preferences for a specific treatment arm, which can negatively affect trial enrollment.^27,28 Education should extend to dialysis care team members who likely spend the most time with patients receiving hemodialysis and potentially have the most opportunity to educate patients about the need for access placement as well as the equipoise between AV fistulas and grafts for select populations. As stated by our patient participants and noted in other research settings, if patients receive information which supports trial enrollment from clinicians they trust, they could be more likely to enroll.²⁹

While qualitative methods have been widely used in other fields of medicine to explore recruitment challenges in RCTs, to our knowledge, this is the first study to apply such an approach to a trial focused on surgical AV access outcomes.

Our study had some limitations. We did not interview study coordinators or dialysis unit care team members, who may have additional insights surrounding facilitators and barriers to trial enrollment. While most site investigators participated in focus groups, perspectives of site providers and patients were limited to those who agreed to participate in the qualitative substudy. We did not explore quantitative data which could help contextualize participant experiences, such as wait times between referral and appointment with vascular surgery providers, longevity of patient-provider relationships, or racial concordance between patient and provider. As a cross-sectional analysis we did not evaluate changes in facilitators and barriers over time, nor did we directly compare experiences across study sites.

In conclusion, enrollment in the AV Access trial was influenced by multiple barriers including challenges related to site selection, limited buy-in from clinical teams, and fragmented referral pathways for patients on hemodialysis. Opportunities identified that might enhance enrollment included provider-directed education addressing lingering biases from the Fistula First initiative, as well as patient-facing education emphasizing the importance of AV access and the potential equipoise between AV fistulas and grafts in select populations. These findings offer valuable insights for the design and implementation of future trials on surgical AV access outcomes.

Supplementary Material

kidney360-7-362-s001.pdf ^{(2.5MB, pdf)}

kidney360-7-362-s002.pdf ^{(233.5KB, pdf)}

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/KN9/B436.

Author Contributions

Conceptualization: Mariana Murea, Dipal M. Patel.

Data curation: Allison M. Chandler, Dipal M. Patel, Ashley E. Strahley.

Formal analysis: Allison M. Chandler, Dipal M. Patel, Ashley E. Strahley, Katherine White.

Funding acquisition: Mariana Murea.

Investigation: Allison M. Chandler, Deidra C. Crews, Mariana Murea, Dipal M. Patel, Ashley E. Strahley.

Methodology: Allison M. Chandler, Mariana Murea, Ashley E. Strahley.

Project administration: Brindusa P. Burciu, Mariana Murea.

Supervision: Mariana Murea.

Visualization: Dipal M. Patel.

Writing – original draft: Dipal M. Patel.

Writing – review & editing: Brindusa P. Burciu, Allison M. Chandler, Deidra C. Crews, Mariana Murea, Ashley E. Strahley.

Funding

M. Murea: National Institute on Aging (R01AG071803-01).

Data Availability Statements

All original data, including deidentified patient-level data or individual laboratory data measurements, are included in the manuscript and/or supplemental material. Original data generated for the study will be made available upon reasonable request to the corresponding author. Data Type: Clinical Trial Data. Reason for Restricted Access: Where not available within the article and supplementary materials, the authors agree to make additional redacted data (e.g., summaries of qualitative analysis) available upon reasonable request and upon completion of a data use agreement with Atrium Health Wake Forest Baptist.

Supplemental Material

This article contains the following supplemental material online at http://links.lww.com/KN9/B437.

Supplemental Item 1. Questions for focus groups with site physician investigators.

Supplemental Item 2. Semistructured interview guide for site providers.

Supplemental Item 3. Semistructured interview guide for patients who declined enrollment.

Supplemental Item 4. Semistructured interview guide for patients who declined enrollment: hypothetical version.

Supplemental Item 5. Semistructured interview guide for patients who enrolled and had not yet undergone vascular access surgery.

Supplemental Item 6. Semistructured interview guide for patients who enrolled and underwent vascular access surgery.

Supplemental Table 1. Baseline characteristics of patients randomized in the AV access trial.