Nephrol Dial Transplant (2011) 0: 1–6

doi: 10.1093/ndt/gfr055

Original Article

Patient attitudes towards the arteriovenous fistula: a qualitative studyon vascular access decision making

Wang Xi1, Lori Harwood1, Michael J. Diamant1, Judith Belle Brown2, Kerri Gallo1,Jessica M. Sontrop3, Jennifer J. MacNab3 and Louise M. Moist1

1Division of Nephrology, Department of Medicine, London Health Sciences Center and the University of Western Ontario, London,Canada, 2Centre for Studies in Family Medicine, Department of Family Medicine, University of Western Ontario, London, Canada and3Department of Epidemiology and Biostatistics, University of Western Ontario, London, Canada

Correspondence and offprint requests to: Louise M. Moist; E-mail: louise.moist@lhsc.on.ca

AbstractBackground. The use of arteriovenous fistulas (AVFs)among hemodialysis (HD) patients has been consistentlyassociated with lower rates of morbidity and mortality;however, up to 30% of eligible patients refuse the creationor cannulation of an AVF. We aimed to understand theattitudes, beliefs, preferences and values of patients whorefused creation or use of an AVF.Methods. With qualitative methodology, we conductedsemi-structured interviews with 13 HD patients (Canada,2009), who previously refused creation or use of an AVF.Three independent analysts reviewed interview transcripts.Results. We discovered three main themes that impactedthe decision to refuse a fistula: (i) poor previous personal orvicarious experiences with the fistula, including cannula-tion, bleeding, time commitment and appearance; (ii)knowledge transfer and informed decision making. Patientsidentified information from other patients to be as impor-tant as information from health care workers, that informa-tion on vascular access (VA) was presented but notunderstood and that timing of information was crucial withinformation overload at the start of dialysis and (iii) main-tenance of status quo and outlook on life. Some patientsstated they live day-to-day without being influenced by themortality risks with a catheter.Conclusions. AVF refusal is multifactorial and dependson individual patients. Although nephrologists considerthe fistula to be the optimal VA, patients do not think inthe same terms of reducing infection rates but focus on thepractical day-to-day use of their VA and its influence ontheir quality of life and future outlook.

Keywords: hemodialysis; vascular access

Introduction

As of 2007, ~50% of Canadian hemodialysis (HD) patientsand 47% of American HD patients had an arteriovenousfistula (AVF) for vascular access (VA) [1]. In comparison

with arteriovenous grafts and central venous catheters(CVCs), the use of AVFs are consistently associated withlower rates of morbidity and mortality [2–7]. However,recent studies demonstrate a marked decline in the preva-lence of AVFs and an increase in CVCs within Canada,Australia and countries throughout Europe [1, 8, 9]. Thedecline in use of AVFs can be attributed to patient, facilityand health care system factors [10]. In our recent study,80% of the nephrologists indicated that patient refusal is amajor barrier to creating a mature AVF [11].

The research on decision making among patients withend-stage renal disease (ESRD) focuses on patient refusalfor dialysis, withdrawal from dialysis or choice of dialysismodality [12–14]. To our knowledge, there are no studieson the patients’ decision-making process in choosing aform of VA.

Informed decision making is an important pillar in pa-tient care. Patients who show greater understanding ofchronic kidney disease from educational classes are morelikely to choose dialysis with a fistula [15]. Patient factorscan also influence decision making, as patients >65 yearsold with fistulas are more likely to be bothered by pain,bleeding, bruising, swelling and appearance of their fistula[16].

To explore the decision-making process, we conducted aseries of 13 in-depth interviews with HD patients who hadpreviously refused an AVF. We sought to understand howtheir attitudes, beliefs, preferences and values influencedtheir choice of VA.

Materials and methods

We used a qualitative methodology with a phenomenological approach toexplore patient attitudes, preferences, beliefs and values when deciding onVA choices for patients on chronic conventional HD.

Patient recruitment

We recruited patients from a tertiary care HD unit in London, Canada,between June and August of 2009. Patients dialyzing with a catheter, whowere deemed eligible for an AVF by their nephrologist, but refused cre-ation or cannulation of the AVF were eligible for the study. We excluded

� The Author 2011. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.For Permissions, please e-mail: journals.permissions@oup.com

NDT Advance Access published March 15, 2011 by guest on M

ay 20, 2014http://ndt.oxfordjournals.org/

Dow

nloaded from

patients who did not have the capacity to provide informed consent, couldnot speak English well enough for an open-ended interview or were on thewaiting list for a kidney transplant.

From eligible patients, a purposeful selection was carried out to samplefor maximum variation with respect to age, gender, length of dialysis,previous failed AVFs, cause of renal failure and previous use of peritonealdialysis. Our sampling strategy was chosen to maximize the diversity ofpatient experiences [17].

Data collection

We created an interview guide based on a literature review of VA. Theguiding questions and potential probes kept the interview focused onpatient experiences, preferences and beliefs. The guide was modified asinterviews and data analysis progressed to include new questions andthemes derived from initial interviews.

An independent interviewer (W.X.), not involved in direct patient care,conducted all the interviews using the iterative approach. Semi-structuredinterviews were tape-recorded and transcribed verbatim. Field notes weretaken on patient appearance, behavior, speech and mood, as well as anynonverbal behavior and interviewer reflections. Interviews lasted between15 and 45 min.

Analysis

Transcripts were analyzed independently by three reviewers (L.M.M.,L.H. and W.X.) through line-by-line coding. A series of common themeswere determined after three interviews. This was reviewed again withrealignment of the themes after 10 interviews. Saturation was reachedwhen no new themes were identified after a total of 13 interviews. Exem-plar quotes were extracted based on the thematic analysis with filler wordsdeleted when they have been agreed upon to have no impact on thestatement.

Trustworthiness and credibility

We purposefully sampled patients of different ages, genders, previousfistula use, time on dialysis, cause of renal disease and previous use ofalternative dialysis modalities to enhance the spectrum patient experien-ces. Sample size estimation was arrived at by first by consulting otherqualitative studies of similar design, which provided an initial range,and secondly, using the principle of theoretical saturation, whereby dataare collected until no new information is obtained from subsequent datacollection [18]. While a larger sample would add to the depth of the results,substantive topics and strong patterns of similarity within those topicswere evident after interviews with 13 participants, thus indicating a suffi-cient level of saturation.

An independent analyst familiar with the experience of renal patientsand renal disease determined that the themes were credible in the perspec-tive of these patients. Our purposeful sample improves transferablility to ageneralized patient population assuming that the patients are in centerswith similar incidence and prevalence of different VA use and have similareducational programs. Using an iterative approach, we confirmed variousoverarching themes elicited during initial interviews in the latter ones,supporting a reliable thematic analysis. This study was approved by theHealth Sciences Research Ethics Board at The University of WesternOntario.

Results

The sample was selected from a population comprised of147 patients receiving chronic HD using a catheter, ofwhich 48 (33%) patients were eligible for a fistula. Of thoseidentified, 32 (68%) patients refused creation of a newfistula or cannulation of an existing fistula. We interviewed13 patients, 7 males and 6 females (Table 1). Six patientshad a previous AVF.

We observed three major themes surrounding the deci-sion to refuse a fistula: negative personal or vicarious ex-periences with the fistula, knowledge transfer and informeddecision making and acceptance of status quo (Table 2).

Impact of previous personal or vicarious experienceswith a fistula

Six of thirteen patients in our study had at least one pre-vious fistula. Other patients learned about the fistulathrough talking with or seeing other patients with a fistula.We identified four common experiences that influenced theparticipants’ decision to reject a fistula: failure to mature,difficulty with cannulation and pain, bleeding from thefistula after needle withdrawal and consequent time com-mitment and appearance of the fistula (Table 2).

Patients mentioned that previous experiences of failureto mature were major deterrents to obtaining anotherfistula.

‘I was hoping that [my fistula] would develop to thepoint where it would be more successful but that didn’thappen. And I followed all the rules. I did all the exer-cises, but that wasn’t enough.’

Cannulation pain and discomfort due to needles wascommon.

‘That’s a pretty big needle and it’s uncomfortable. Andthen, when you’re lying here, you can’t move, at all,because it used to interfere with it. I tried to put the creamon to desensitize. That really didn’t do very much.’

The rate of successful cannulation was another patientconcern given the low prevalence of fistula use in this unit,though none of the participants refused cannulationpermanently.

‘[Nurses] don’t have experience with fistula. They haveexperience with the line . . . it’s maybe a good idea that ifyou have fistula you get the same person to poke yourarm . . . maybe just teach the patient how to do it.’

Persistent bleeding and consequent time commitmentwas an issue.

‘After dialysis, sometimes I have to stay longer, youknow? Because the bleeding didn’t stop. And sometimesI just come back because of bleeding. It’s a differentstory every time.’

Body image was also important to a few patients, as onepatient stated about her vicarious experiences,

‘she had [the fistula] in all her life and she had big welts,big bumps, all in her arm. And then at that time, theywere going to change it to the other arm. She had a lot ofproblems. Her arm was just swelled up.’

Knowledge transfer and informed decision making

We explored how patients obtained and processed informa-tion and how this influenced their decision on type of VA.Patients reported obtaining information from health careprofessionals, other dialysis patients and the Internet.Meanwhile, some patients perceived a lack of information(Table 2).

Patients with ESRD within the London Health SciencesCentre system receive information in many forms: meeting

2 W. Xi et al.

by guest on May 20, 2014

http://ndt.oxfordjournals.org/D

ownloaded from

with a nephrologist, receiving written material, attendinginstructional courses, receiving teaching from the VA nurseand planned and impromptu interactions with other healthcare professionals. The information provided greatly influ-enced patients’ decision either for or against the fistula.

‘The paperwork that the [VA nurse] gave me was verywell done . . . . I sat down and looked at it a few times anddecided to [go for the fistula].’

Interactions with health care professionals wereimportant.

‘I was told that [the fistula] was the best way to go. Andthat dialysis would work better if I had a fistula. So thatbasically made up my mind . . . and if I ever do have togo on dialysis, it’s already there and waiting.’

In deciding to forego another fistula after a failed one,health care professionals were also a key influence.

‘[The doctor] only gave me a thirty percent chance that itwould work again. And that’s when I decided well I’mnot going to go through all that again if I’ve only got athirty percent chance. So I use the catheter . . . I think if

Table 1. Participant characteristics (N ¼ 13)a

Characteristics n (%)b

Mean age, years (SD) 69 (12)Gender

Male 7 (54)Female 6 (46)

Mean time on dialysis, months (SD) 47 (28)Previous use of PD 1 (8)Previous transplant 1 (8)Previous AVF use 6 (46)Previous AVG use 2 (15)Previous CVC use 6 (46)

aParticipants’ diagnosis causing renal disease included diabetes, hypertension, IgA nephropathy,vasculitis, malignancy, renovascular disease and familial disease. AVG, arteriovenous graft; PD,peritoneal dialysis; SD, standard deviation.bUnless otherwise indicated.

Table 2. Themes surrounding VA decision making

Theme Comment

I. Impact of previous personal or vicarious experiencesFailure to mature ‘I’ve seen that they don’t work and they can’t get them to work and

I just wasn’t interested in [the fistula].’Cannulation ‘I don’t want somebody poking at my arm every time I come here . . .

I’ve seen one nurse will take forever to try and the next thingyou know she’s got somebody else there trying so they’re pokingat your arm.’

Bleeding ‘ . . . by the time I got home the coat I had on was loaded with blood.’Body image ‘Her arm was just swelled up. And after that I seen another who had

an awful arm and they had to go back to [a catheter]’.II. Knowledge transfer and informed decision makingHealthcare professionals ‘I was told that [the fistula] was the best way to go. And that dialysis

would work better if I had a fistula. So that basically made up my mind.Yeah OK. I’ll get the fistula. And if I ever do have to go on dialysis,it’s already there and waiting.’

Dialysis-patient peers ‘They said they’ve had more than one [fistula]. They’ve had several.And they’ve had problems with it so they’ve had to get new ones.And it all takes time . . .’

Perceived lack of knowledge ‘My doctor actually sent me to the hospital but he didn’t tell me whyor anything. While I was in there, they put the [catheter] access inand started me on dialysis. I didn’t know anything about it.’

III. Patient acceptance of status quoLifestyle limitations ‘I’ve been in a pool a few times, I just cover [the catheter] all up with

plastic and tape it over.’Lack of actual complications or concern for complications ‘I’ve had this [catheter] for two years and I was running with two

catheters and as long as you pay attention and clean yourselfproperly and watch for infection, you won’t get any.’

Complications of fistula creation ‘Most of the people that have been here have passed on becausethey keep fooling around with . . . putting [the fistula] in your arm,then they take it out.’

Focus on the present, day-to-day living ‘ . . . nobody can predict what your health is going to be in the futureI don’t think. I feel pretty good right now, but you live from day to day’.

Patient decision making on AVF 3

by guest on May 20, 2014

http://ndt.oxfordjournals.org/D

ownloaded from

he’d given me a fifty-fifty chance, I probably would havehad it again.’

Patient-to-patient knowledge transfer also played an im-portant role. One patient with a positive perception of thefistula stated

‘most of the information came by word of mouth. To me,I’m one of these guys, I want to see before I do. Two orthree people I associate with, they all had [the fistula].They didn’t seem to be complaining about it.’

However, most patients used other patients’ experiencesas a deterrent to having a fistula.

‘Two or three guys that had [the fistula], they really had alot of problems with their entire arm and all over . . . twothat I know that went back to up here [pointing to cath-eter site] have had it for quite a while. The arm thing justdidn’t seem to work.’

Several patients perceived they did not have enough in-formation about different VAs to make an informed choice.

‘My doctor actually sent me to the hospital but he didn’ttell me why or anything. While I was in there, they putthe access in and started me on dialysis. I didn’t knowanything about it.’

Some patients had gaps in their knowledge about thefistula.

‘I misunderstood what a fistula really does. I thought itwas something that they stuck in your arm and then therewas an access to that where they just plugged it in orsomething. I didn’t realize that they stuck needles in yourarm.’

Patient acceptance of status quo

Patients identified the following as justification for use ofthe catheter: acceptance of catheter limitations on lifestyle,lack of actual catheter complications, lack of concern forpotential catheter complication, fear of potential complica-tions with fistula creation and attitudes about longevity andpreference to focus on the present including living for today(Table 2).

Patients emphasized overall satisfaction with their cur-rent catheter despite awareness of the lifestyle limitations oftheir catheter.

‘it would make my quality of life a little better using afistula rather than the catheter . . . . I’d be able to showerand be able to go swimming, but I’ve pretty much re-signed myself that the catheter is the only way for me togo so I don’t shower. I take a bath.’

Patients minimized chances and occurrences of compli-cations. One patient mentioned that they are using the cath-eter because ‘the catheter has been successful and has donethe job.’ Another patient mentioned that ‘I haven’t had aninfection in this one yet’ There appeared to be a mentalitythat despite a previous infection, the current one has been

functional and uncomplicated, with several patients usingsimilar phrases to ‘ever since then it’s been all right’, refer-ring to their newly placed catheter.

Patients were aware that ‘[the fistula] was considered tobe a little safer, a little less chance of getting an infection.’Despite the awareness, infectious complications wereminimized.

‘I’ve seen two or three that had a bad infection. Fromwhat the doctor told me, as long as you keep [yourcatheter] good and clean you wouldn’t have an infection. . . .The first time I ever had [an infection], I didn’t knowwhat might happen. [The doctor] said ‘don’t worry aboutit, we’ll just pull that one out and put another one in andgive you some antibiotics and it should be okay.’

Patients also did not want the hassle and complicationsof undergoing fistula placement. One patient commentedon her preference not to have a surgery, ‘I’ve had 37 oper-ations. I don’t want any more’. Another patient believedother patients’ death was related to a lack of success of thefistula, stating

‘most of the people that have been here have passed onbecause they keep fooling around with . . . putting [thefistula] in your arm, then they take it out.’

Patients’ future outlook was important.

‘it’s a big step [to get a fistula] . . . . I look at it from adifferent perspective than most people. I’m 69-years ofage in December of this year and I’ve got to keep askingmyself ‘how much does my body take,’ because everytime I change something my body reacts accordingly andI don’t know how much it will take.’

Another patient who had considered stopping dialysisaltogether was strongly against the fistula stating that ‘ifthey wanted to move [my catheter] from this spot to myarm, I wouldn’t take dialysis.’ Patients also reflected onhopes for alternatives to dialysis. Despite being ineligiblefor a transplant, one patient stated ‘I’m hoping they’ll putme on the [transplant] list again. I’m young’.

In contrast, and to reinforce the importance of outlook onlife, one patient who actually decided to obtain a fistulathrough the course of the study mentioned the long-termbenefits of the fistula.

‘over time the fistula will outlast and outwork your cath-eter because the catheter eventually has to be cleanedout, repositioned. If the fistula is working properly,you won’t have that problem. It’ll last for years . . . inthe long run that supercedes anything else . . . . I wouldrather look at the green grass, than the roots of the grass.. . .You have to keep your attitude up. If you lose yourattitude, you will just go down and down. So, it’s anotherday and another way of living.’

Many patients, despite their preferences for dialysis witha catheter, claimed that they would tell other patients that‘they had to make up their own mind’ regarding choosingan access for dialysis. Patients also recommended that you

4 W. Xi et al.

by guest on May 20, 2014

http://ndt.oxfordjournals.org/D

ownloaded from

have to ‘have a reason and do a lot of research’. One patientwho had a previously failed fistula stated ‘I guess I wouldtell them to try the fistula’.

Discussion

Our study explored patients’ decision-making process withregard to refusal of a fistula. We elicited three majorthemes: (i) previous personal or vicarious experiences witha fistula, (ii) knowledge transfer and informed decisionmaking and (iii) patient acceptance of the status quo.

Our findings support dialysis patients’ preferences for ashared decision-making process between the health careteam and patient, one of the major pillars in the patient-centered health care model [19]. While a fistula may berecommended, a patient can turn down a fistula for severalreasons, and a study recently published by our groupshowed that 80% of nephrologists consider patient refusalto be a major barrier for fistula creation [11]. In our study,the patients we interviewed were experiencing a wide spec-trum of symptoms and altered quality of life from theirdialysis and comorbidities. Frustration with their diseaseand value of current health can impact decision makingand health outcomes [20, 21]. For example, two patientswho perceived their doctor’s reassurance of a guarantee ofa successful procedure ended up with complications. Thesepatients may be less inclined for further interaction withinthe health care system.

Patients made decisions using their own or vicariousexperiences with the fistula. The spectrum varied from pa-tients who would consider a fistula again if chances ofsuccess were higher to patients who would stop dialysisif they had to obtain a fistula. Many patients reported pooroverall outcome, including failure to mature and inability tocannulate, as a major deterrent to having a first or repeatfistula. Studies report a failure to mature rate of up to 65%,likely due to creating fistulas in patients who are less likelyto have successful maturation [22]. By establishing eligi-bility criteria for fistula creation, we can seek to decreasethe failure to mature rate and expose fewer patients tonegative vicarious experiences with the fistula. Our studyalso found that patients were concerned with the lack ofexperience nurses had with cannulating a fistula. At ourcenter, the high prevalence of catheter use and nursing stafflogistics reduces the experience of successful cannulation.Although there were no patients who refused cannulationpermanently, poor experience with previous cannulationsor lack of knowledge of what cannulation involved werecited as potential reasons against a patient consenting tocannulation. Patients in our study also reported pain as aresult of needles, appearance, bleeding and increased timecommitment as dissatisfactions, which is consistent withprevious literature [16].

Knowledge transfer and informed decision making wasan important theme. Influence from peers was important,consistent with literature regarding patient decision makingin ESRD [12, 13]. For patients in our study, peer influencewas often as important as influence by health care profes-sionals. Patients expressed camaraderie such as ‘the threemusketeers’ with other dialysis patients. Family members

of patients did not appear to be influential in decision mak-ing, contrasting studies of dialysis withdrawal where bur-den to family was a theme [14]. Information from healthcare professionals is provided at our center through paper-based and verbal teaching, either within groups or individ-ually. Patients gravitated towards one or more forms ofinformation, and no patients were discouraged from anyparticular form of teaching. During seminar teaching, pa-tients see examples of fistulas, grafts and catheters in pa-tients who have been on dialysis for a while. At our center,we emphasize the fistula as the optimal VA. Nurse andphysician preference within a specific center has beenshown to influence the prevalence of VA modalities incenters that prefer grafts [23].

Timing of information is important for patients [12, 24].Several patients in our study had sudden starts to theirdialysis. Patients may forget or misunderstand informationabout VA if given at a time when they are overwhelmedwith information regarding their health or by their personalreactions to disease. Even given at a good time, some in-formation may prove to be difficult to absorb [25]. Misun-derstanding of information played an important role insome patients, including one patient who was unawareneedles were involved with the fistula. Patients who showa greater understanding about CKD have increased use of afistula [15, 26].

Previous literature suggests that many patients perceive alack of choice in dialysis modality because of medical con-traindication or health care professional preference [12,27]. Given eligibility for a fistula, a patient is always givena choice between the fistula and catheter. This availabilityof choice may influence individuals to make a choice in-dependently without physician influence compared to dial-ysis modality choice that which may be more dictated bythe physician. However, nephrologists should emphasize tofistula-eligible patients that the catheter is only a temporarysolution until a fistula is established [28].

Once patients begin dialyzing with an access that appearsto work, consideration of other options becomes less likely.This is especially true when lack of complications couplewith good patient adaptations to catheter restrictions, in-cluding inability to shower or swim. Previous literaturesuggests that once a treatment plan is established, altera-tions tend to be resisted by the patient [12, 29, 30]. Thesefindings reinforce the importance of starting dialysis with afistula, but also early referral and frequent nephrologist vis-its, which are associated with increased fistula use at thestart of dialysis [31–34]. Physicians should be cognizantthat recommending a fistula may be interpreted by a patientas a change in the status quo and a sign of disease progres-sion, which a patient may not be willing to accept [13].

Though taken seriously by health care professionals, in-fection of catheters was downplayed by patients. If a cath-eter was to be infected, patients also believed it would beeasy although painful to be provided another one. In pre-vious studies, only 3% of patients believed infection was amajor concern [16].

Our study had some limitations. This dialysis unit had ahigh prevalence of catheter use, and patient attitudes in thisunit may not be reflective as those from a unit with a lowerrate of catheter use. There are no major financial barriers to

Patient decision making on AVF 5

by guest on May 20, 2014

http://ndt.oxfordjournals.org/D

ownloaded from

fistula creation in Canada because hospital-related costs donot require out-of-pocket payment from the patient. How-ever, patients often have to travel long distances (>200 km)to come for an access procedure or creation, and this cost isborn by the patient and the family. Patients’ recollection ofeducation regarding VA may omit forgotten knowledgeused during their actual period of decision making.

Our study elicited the reasoning process patients use tomake an informed decision regarding a fistula versus acatheter. A health care professional must build a trustingrelationship explaining all different options of care with thehelp of multiple educational resources that should not onlybe received but understood by the patient. Fistula eligibilitycriteria should be studied to reduce the failure to maturerate and decrease negative vicarious experiences with thefistula. Although nephrologists consider the fistula to be theoptimal VA, patients do not think in the same terms ofreducing infection rates, but rather place greater emphasison the practical day-to-day use of their VA and its influenceon their quality of life and future outlook.

Acknowledgements. We would like to thank the patients who participatedin the study for sharing their time and experiences. W.X. was funded bythe Schulich School of Medicine and Dentistry Summer Research Oppor-tunity Program.

Conflict of interest statement. None declared.

References

1. Ethier J, Mendelssohn DC, Elder SJ et al. Vascular access use and out-comes: an international perspective from the dialysis outcomes and prac-tice patterns study. Nephrol Dial Transplant 2008; 23: 3219–3226

2. Dhingra RK, Young EW, Hulbert-Shearon TE et al. Type of vascularaccess and mortality in U.S. hemodialysis patients. Kidney Int 2001;60: 1443–1451

3. Pastan S, Soucie JM, McClellan WM. Vascular access and increased riskof death among hemodialysis patients. Kidney Int 2002; 62: 620–626

4. Xue JL, Dahl D, Ebben JP et al. The association of initial hemodial-ysis access type with mortality outcomes in elderly medicare ESRDpatients. Am J Kidney Dis 2003; 42: 1013–1019

5. Polkinghorne KR, McDonald SP, Atkins RC et al. Vascular accessand all-cause mortality: a propensity score analysis. J Am Soc Nephrol2004; 15: 477–486

6. Astor BC, Eustace JA, Powe NR et al. Type of vascular access andsurvival among incident hemodialysis patients: the Choices forHealthy Outcomes in Caring for ESRD (CHOICE) Study. J Am SocNephrol 2005; 16: 1449–1455

7. Pisoni RL, Arrington CJ, Albert JM et al. Facility hemodialysis vas-cular access use and mortality in Countries participating in DOPPS: aninstrumental variable analysis. Am J Kidney Dis 2009; 53: 475–491

8. Moist LM, Chang SH, Polkinghorne KR et al. Trends in hemodialysisvascular access from the Australia and New Zealand Dialysis andTransplant Registry (ANZDATA) 2000 to 2005. Am J Kidney Dis2007; 50: 612–621

9. Moist LM, Trpeski L, Na Y et al. Increased hemodialysis catheter usein Canada and associated mortality risk: data from the CanadianOrgan Replacement Registry 2001-2004. Clin J Am Soc Nephrol2008; 3: 1726–1732

10. Roca Tey R. [Vascular access for haemodialysis: an unresolvedissue]. Nefrologia 2010; 30: 280–287

11. Xi W, Macnab J, Lok CE et al. Who should be referred for a fistula?A survey of nephrologists. Nephrol Dial Transplant 2010; 25:2644–2651

12. Lee A, Gudex C, Povlsen JV et al. Patients’ views regarding choice ofdialysis modality. Nephrol Dial Transplant 2008; 23: 3953–3959

13. Morton RL, Tong A, Howard K et al. The views of patients and carersin treatment decision making for chronic kidney disease: systematicreview and thematic synthesis of qualitative studies. BMJ. 2010; 340:c112

14. Noble H, Meyer J, Bridges J et al. Patient experience of dialysisrefusal or withdrawal–a review of the literature. J Ren Care 2008;34: 94–100

15. Cavanaugh KL, Wingard RL, Hakim RM et al. Patient dialysisknowledge is associated with permanent arteriovenous access use inchronic hemodialysis. Clin J Am Soc Nephrol 2009; 4: 950–956

16. Quinn RR, Lamping DL, Lok CE et al. The vascular access question-naire: assessing patient-reported views of vascular access. J VascAccess 2008; 9: 122–128

17. Silverman D, Marvasti A. Doing Qualitative Research: A Compre-hensive Guide. Thousand Oaks, CA: Sage Publications, Inc.; 2008

18. Patton M. Qualitative Evaluation and Research Methods. 2nd edn.Newbury Park, CA: Sage Publications; 1990

19. Dolan JG. Shared decision-making–transferring research into prac-tice: the Analytic Hierarchy Process (AHP). Patient Educ Couns2008; 73: 418–425

20. Ashby M, op’t Hoog C, Kellehear A et al. Renal dialysis abatement:lessons from a social study. Palliat Med 2005; 19: 389–396

21. Plantinga LC, Fink NE, Bass EB et al. Preferences for current healthand their association with outcomes in patients with kidney disease.Med Care 2007; 45: 230–237

22. Dember LM, Beck GJ, Allon M et al. Effect of clopidogrel on earlyfailure of arteriovenous fistulas for hemodialysis: a randomized con-trolled trial. JAMA 2008; 299: 2164–2171

23. Young EW, Dykstra DM, Goodkin DA et al. Hemodialysis vascularaccess preferences and outcomes in the Dialysis Outcomes and Prac-tice Patterns Study (DOPPS). Kidney Int 2002; 61: 2266–2271

24. Landreneau K, Ward-Smith P. Patients’ perceptions concerningchoice among renal replacement therapies: a pilot study. Nephrol NursJ 2006; 33: 397–402; quiz 7

25. Harwood L, Locking-Cusolito H, Spittal J et al. Preparing for hemo-dialysis: patient stressors and responses. Nephrol Nurs J 2005; 32:295–302; quiz 3

26. Asif A, Cherla G, Merrill D et al. Conversion of tunneled hemodial-ysis catheter-consigned patients to arteriovenous fistula. Kidney Int2005; 67: 2399–2406

27. Wuerth DB, Finkelstein SH, Schwetz O et al. Patients’ descriptions ofspecific factors leading to modality selection of chronic peritonealdialysis or hemodialysis. Perit Dial Int 2002; 22: 184–190

28. Rehman R, Schmidt RJ, Moss AH. Ethical and legal obligation toavoid long-term tunneled catheter access. Clin J Am Soc Nephrol2009; 4: 456–460

29. Whittaker AA, Albee BJ. Factors influencing patient selection of dialysistreatment modality. ANNA J 1996; 23: 369–775; discussion 76–77

30. Bass EB, Wills S, Fink NE et al. How strong are patients’ preferencesin choices between dialysis modalities and doses? Am J Kidney Dis2004; 44: 695–705

31. Avorn J, Winkelmayer WC, Bohn RL et al. Delayed nephrologistreferral and inadequate vascular access in patients with advancedchronic kidney failure. J Clin Epidemiol 2002; 55: 711–716

32. Bradbury BD, Fissell RB, Albert JM et al. Predictors of early mortalityamong incident US hemodialysis patients in the Dialysis Outcomes andPractice Patterns Study (DOPPS). Clin J Am Soc Nephrol 2007; 2: 89–99

33. Hasegawa T, Bragg-Gresham JL, Yamazaki S et al. Greater first-yearsurvival on hemodialysis in facilities in which patients are providedearlier and more frequent pre-nephrology visits. Clin J Am Soc Neph-rol 2009; 4: 595–602

34. Astor BC, Eustace JA, Powe NR et al. Timing of nephrologist referraland arteriovenous access use: the CHOICE Study. Am J Kidney Dis2001; 38: 494–501

Received for publication: 15.11.10; Accepted in revised form: 18.1.11

6 W. Xi et al.

by guest on May 20, 2014

http://ndt.oxfordjournals.org/D

ownloaded from