- Research article
- Open Access
- Open Peer Review
Patients' perspectives on taking warfarin: qualitative study in family practice
BMC Family Practice volume 5, Article number: 15 (2004)
Despite the well-documented benefits of using warfarin to prevent stroke, physicians remain reluctant to initiate therapy, and especially so with the elderly owing to the higher risk of hemorrhage. Prior research suggests that patients are more accepting of the risk of bleeding than are physicians, although there have been few qualitative studies. The aim of this study was to employ qualitative methods to investigate the experience and perspective of individuals taking warfarin.
We conducted face-to-face interviews with 21 older patients (12 male, 9 female) who had been taking warfarin for a minimum of six months. Participants were patients at a family practice clinic situated in a large, tertiary care teaching hospital. We used a semistructured interview guide with four main thematic areas: decision-making, knowledge/education, impact, and satisfaction. Data were analysed according to the principles of content analysis.
Results and Discussion
Participants tended to have minimal input into the decision to initiate warfarin therapy, instead relying in great part on physicians' expertise. There appeared to be low retention of information received regarding the therapy; half the patients in our sample possessed only a superficial level of understanding of the risks and benefits. This notwithstanding, participants reported a high level of satisfaction with the care provided and a low level of impact on their day-to-day lives.
Minimal patient involvement in the initial decision and modest knowledge did not appear to diminish satisfaction with warfarin management. At the same time, care providers exert a tremendous influence on the initiation of warfarin therapy and should strive to incorporate patient preferences and expectations into the decision-making process.
Warfarin therapy is an effective anticoagulant indicated for the prophylaxis and/or treatment of venous thrombosis and atrial fibrillation (AF), the most common cardiac arrhythmia in older individuals . Oral anticoagulation with warfarin is known to reduce the risk of disabling stroke; indeed, the benefits of oral anticoagulation have been demonstrated in a number of systematic reviews, providing high evidential support for prophylaxis [2, 3]. Published guidelines on the management of AF emphasize the importance of warfarin therapy for the prevention of stroke . Likewise, there is convincing evidence that long-term warfarin therapy is a highly effective method of preventing recurrent venous thromboembolism .
Despite the strong evidence base and the endorsement of warfarin therapy by authoritative guidelines, current prescribing patterns of warfarin remain something of a puzzle. Bungard and associates have described the 'real world' use of warfarin therapy as "sub-optimal"  and have estimated that only 15% to 44% of patients eligible for anticoagulation are actually prescribed warfarin . Hart concurs that the use of warfarin is poor, noting that "it is often given to patients who benefit minimally, while those patients who would benefit most are not treated" .
Bungard and his colleagues recently conducted a systematic review of the reasons for the underuse of warfarin, identifying patient, provider, and system factors as well as identifying limitations in research studies and arguing for further research into these factors . Studies employing trade-off methods to determine the risk/benefits threshold where therapy becomes acceptable have demonstrated that patients are more willing to assume risk when better informed about the medication . Physicians, however, are reluctant to prescribe warfarin to elderly patients owing to concerns regarding compliance, a perceived risk of falls, and the lack of randomized controlled trial evidence in this patient population . A recent study reporting the findings of interviews with individuals with a history of AF indicated that patients' health beliefs and attitudes toward death play an important role in their decision-making .
Clearly, the decision to initiate warfarin therapy is a complex interaction of many variables, involving patient, provider, and system factors. Most studies examining anticoagulation practices in primary care have used surveys or other forms of quantitative methods; however, given the inherent complexity of this subject matter, it is likely that qualitative research methods could provide significant additional insight. A comprehensive literature search failed to find a qualitative investigation of patient perspectives and experiences taking warfarin. The aim of this study, therefore, was to employ qualitative methods to examine the experience and perspective of individuals on long-term warfarin therapy for atrial fibrillation as a means to assess the extent to which the physician-identified barriers reported by Bungard  are in concordance with the unique views of patients.
Participants and setting
This research was undertaken in a family practice clinic situated in a large, tertiary care teaching hospital. The clinic employs 12 physician full-time equivalents working in three teams; each four-physician team is supported by two registered nurses. The practice, which serves a medium-high socioeconomic status population, has a large proportion of elderly patients and the prevalence of AF is higher than reported elsewhere . The prevalence of atrial fibrillation in our population as a whole is 3.9 percent. When considering different age groups, the prevalence rises as high as 18.2 percent and 18.5 percent for patients aged 80–89 and 90–99 years, respectively. A recent chart audit study found that the majority of eligible AF patients in the clinic (78%) are being treated with warfarin for stroke prevention . Comprehensive anticoagulation services are provided and the clinic offers access to physicians on-call 24 hours a day. The nurses maintain an historical record of International Normalized Ratio (INR) results and warfarin dosage changes for each individual patient and, in consultation with the physicians, inform patients of prescribed warfarin dosage changes in a timely fashion, usually on the same day as the INR reading.
Potential participants were identified by the clinic nursing staff. The inclusion criteria stated that the patient must currently be on warfarin therapy and have been so for a minimum period of six consecutive months. Patients were excluded from the study if a significant co-morbidity prevented their participation, if they were unable to converse in English, or if they were unwilling/unable to provide informed consent. From a pool of approximately 60 eligible candidates, the nurses purposively sampled in order to achieve an even gender split as well as an equivalent number of patients who were both normally in-range and out-of-range on their INR tests. A total of 24 patients were invited to participate in the study. Three patients declined to take part, two of whom expected to be unavailable during the interview phase, whereas the third declined due to lack of interest. All participants signed informed consent forms in advance of their participation in the study, which was approved by the Research Ethics Board of the host institution. A demographic profile of the sample is presented in Table 1. The mean age of participants was 74 years; there were 12 males, 9 females. The majority were both married (86%) and retired (86%). The mean length of time participants had been on warfarin therapy was 4.6 years (range = 1 year to 10 years).
Data collection and analysis
We utilized a semi-structured interview guide that was developed on the basis of salient issues identified in the scientific literature, specifically the various barriers to the prescription of warfarin for atrial fibrillation as reported by Bungard et al . Interviewees were asked to share their experiences with warfarin in relation to four specific content areas or themes: decision-making, knowledge and education, impact on daily life, and patient satisfaction. Throughout the course of the interview, participants were provided several opportunities to raise issues or to describe experiences that had not been specifically addressed. Standard demographic information was also collected. Three of the authors (GCD, JM, and BT) shared the task of conducting the interviews. The protocol for assignment of each individual participant to an interviewing author ensured that there had been no previous clinical contact between the two parties; moreover, interviewers were blind to interviewees' INR status. The interviews lasted an average of 45 minutes and took place either in the clinic (n = 13) or in participants' homes (n = 8), in accordance with each participant's preference. Data collection ceased when, in the consensus of the research team, saturation had been reached; that is, no new ideas or perspectives were emerging. All interviews were audiotaped and transcribed verbatim.
Given the pre-determined nature of the themes, as described above, we employed a content analysis approach to the analysis of the interview data. Whereas grounded theory is used to develop data-induced themes or hypotheses, content analysis is the better-suited approach in those instances where the codes, categories, or themes of interest to the investigators have been previously discovered and described , as is the case in the present study.
Precise criteria were developed for each of the four pre-determined themes in the codebook, namely decision-making, knowledge/education, impact on daily life, and patient satisfaction. The 21 transcripts were then coded according to these criteria. Each transcript was coded by at least two members of the research team using a standardized coding form. Tests for inter-coder reliability indicated a high level of agreement among the coders; instances of disagreement were resolved through a process of discussion and negotiation that included both the fourth author and the principal investigator (CST & REGU). This process yielded a unit-by-variable matrix that allowed for substantive analysis of the data. In order to strengthen the validity of the findings, the analytic processes of coding and interpretation were reviewed by an independent external reader (DG).
The great majority of participants reported that the decision to initiate warfarin therapy had been made by "the doctor" – a term that was used to refer not only to family physicians and general practitioners, but also to specialists and attending physicians in urgent care settings. Typically, there was little or no patient involvement in the decision-making process (Table 2A). In most cases, this unilateral decision-making appeared to be related to the high level of trust that patients place in the medical expertise of physicians; indeed, the phrase "doctor knows best" was commonly-used in these accounts (Table 2B). For a smaller number of participants, the specific circumstances surrounding the initial decision to commence therapy served to preclude any degree of significant involvement on their part (Table 2C).
Knowledge and education
The level of knowledge and understanding of the benefits and risks associated with warfarin therapy tended to vary with age. Elderly patients (aged 75+) demonstrated poorer knowledge than their younger counterparts; indeed, the knowledge level among older participants appeared to be quite superficial and scattered (Table 3A). Whereas elderly patients could not explain with any degree of exactitude the rationale for taking warfarin and the associated risks, for a subset of participants, most of whom were less than 75 years old, the knowledge level was higher and, for a small number, considerably higher (Table 3B). Overall, less than half of our sample was able to name one specific benefit, risk, and lifestyle change/concern associated with warfarin therapy.
According to participants' accounts, educational efforts aimed at informing patients about warfarin were minimal and insufficient (Table 3C). Those who were able to recall some form of education typically referred to a "booklet" or "sheet" supplied by either the clinic or a pharmacy. In several cases, spouses were more knowledgeable than patients and appeared to play an important role in monitoring the regime. A number of participants lauded the availability of clinic staff to answer questions; however, two others reported the use of "scare tactics" by health care professionals with regard to the need to take warfarin.
Impact of warfarin regime
While there is tremendous range in the perceived impact of warfarin therapy on the lives of these patients, the vast majority reported that they have not experienced complications (e.g., hemorrhage, drug interactions). Typically, the decision to start taking warfarin did not precipitate significant changes in their day-to-day lives; many participants reported experiencing only minor inconveniences (Table 4A). For these individuals, warfarin is just another pill to be taken everyday; many reported the use of some reminder strategy, such as calendars, dosettes (pill boxes), or taking the pill right before some regular activity, in order to avoid missing a dose. On the other hand, a sizeable proportion (25%) of interviewees reported that adhering to the warfarin regime does impact upon their day-to-day lives. From the perspective of these patients, who were more likely to have multiple co-morbid illnesses and/or were taking multiple medications, the warfarin regime presents a considerable struggle to be managed, particularly when dosages needed to be adjusted. Regular visits to the clinic, restrictions on diet and alcohol intake, and anxiety regarding bleeding and potential drug interactions counted among the most commonly cited impacts (Table 4B). Only a small number of participants reported experiencing significant complications related to the warfarin regime, with one case involving repeated gastro-intestinal bleeding. These patients demonstrated a very high level of commitment towards their warfarin management and have placed the ritual of taking the medication at the centre of their daily routine (Table 4C).
The vast majority of interviewees reported a high level of satisfaction with the care they receive from the nurses and physicians in the clinic (Table 5A). Most participants were also satisfied with the warfarin regime itself (Table 5B). Only a very small number of participants expressed significant dissatisfaction. The sources of dissatisfaction, which tended to be highly localized to specific concerns, included the cost and inconvenience of attending the clinic for regular INR tests, a lack of information provided to patients, and insufficient awareness of patient history on the part of clinic staff (Table 5C).
The findings of this study provide significant and original insight into the perception and experience of patients taking warfarin. The data indicate that patients tend to have minimal input into the decision to initiate warfarin therapy; many have only a superficial level of understanding of the risks and benefits of warfarin; and the majority retain little from the education they received regarding warfarin therapy. This outcome is balanced, however, by the finding that for these patients there was both a high level of satisfaction with the care provided in the family practice setting and a low level of impact on their day-to-day lives.
The principal strength of this study is the insight into the lived experience of warfarin therapy as gleaned from the unique perspective of family practice patients currently taking warfarin. We view this as a significant and novel contribution to the literature as we could find no other such study. It is important to note, however, that our sample was drawn from the patient population of an academic primary care practice that is both well-educated and of medium-high socioeconomic status. The applicability of these findings in other patient populations may therefore be limited.
Interviewees revealed clear detachment from participation in the decision-making process around initiating warfarin therapy. In some cases, this detachment appeared to stem from the particular circumstances at the time; for instance, if the patient was involved in a medical emergency or was admitted to hospital and had several medications initiated. For others, there was a general belief or understanding that warfarin is a medication without which the patient faced imminent risk of death – it is therefore not a matter to be discussed or negotiated. This finding may be a function of the age of our participants. As a group, elderly patients tend to prefer a directed rather than shared consultation. Prior research indicates that seniors are more likely to be accepting of medical advice without much questioning, rather than assuming a more active role in the decision-making process . This attitude was reflected in participants' comments that "doctor know best" and "it's the doctor's decision."
Patient knowledge of risks, benefits, and issues related to diet and alcohol intake was low, although younger patients demonstrated greater levels of understanding than did those over age 75. This finding is consistent with previous investigations. Lip and colleagues have detected lower levels of knowledge among elderly patients; moreover, longer duration of anticoagulation does not appear to ameliorate patient understanding significantly [14, 15]. For the most part, we found that retention of instructions pertaining to the warfarin regime was poor. Many participants reported that they simply follow the nurses' directions and have little interest in learning anything more. The limited knowledge and seemingly low level of interest to learn more could be attributable to the great deal of trust invested in the expertise of the clinic staff as discussed above. The fact that several spouses exhibited greater understanding of the risks and benefits associated with warfarin therapy has implications for educational interventions recognizing the importance of the spousal or care giver role in the monitoring of therapy.
With regard to the impact of warfarin therapy on daily life, our results indicate that warfarin is for the most part well tolerated and does not pose heavy additional burdens or lifestyle changes. The majority of participants were already taking several medications and visiting more than one doctor on a regular basis. This sample also had a low incidence of complications and previous studies have shown that the potential for complication does not on its own result in a significant impact [16, 17]. That is, the mere possibility of an adverse side-effect does not bring about substantial anxiety unless the complication is actually experienced. These results do not, however, capture the experiences of individuals who have tried and ceased taking warfarin for whatever reasons. This population represents a high priority for future study. Additional support may be required for individuals with multiple medical problems. Continuity of care and a strong relationship and identification with the primary care team that provides anticoagulation services can overcome potential miscommunication and misunderstanding regarding side effects and drug interaction.
The importance of the association between patient satisfaction and adherence has been established in prior research on anticoagulation. In a case-control study, Arnsten and colleagues  found that, among patients with a regular physician, the non-adherent cases were those who expressed dissatisfaction. In our sample, the level of patient satisfaction was high, both with the clinic staff and the warfarin regime itself. Based on participants' testimonies, the coordination and continuity of care by a trustworthy team of doctors and nurses were key contributing factors to the high satisfaction ratings. In an evaluation of a telephone-based anticoagulation service, Waterman found that patient satisfaction with warfarin management was associated with the timeliness of receiving blood test results from the service provider . The high level of patient satisfaction observed in the present study may also be due in part to the low rate of complications (e.g., hemorrhage, drug interactions), which may serve to reinforce patients' trust both in the therapy and in the health care team.
Increasingly, theoretical models of the physician-patient encounter advocate the inclusion of patients in the decision-making process . Of course, shared decision-making presupposes an understanding of the benefits and risks on the part of patients. With regard to warfarin therapy, patient preferences would be expected to vary according to expected benefits or awareness of risks of suffering a stroke. Man-Son-Hing et al have demonstrated that the minimal clinically important difference of warfarin therapy is often considerably smaller for patients than that identified by clinicians . Protheroe and colleagues, in an observational study of patient-based decision analysis, noted marked disagreement between patient preferences and guideline recommendations . A patient decision aid was shown to improve knowledge and understanding of the risks and benefits of warfarin for patients with atrial fibrillation, and aided in therapeutic choice . Given the low level of patient knowledge observed in the present study and elsewhere, the vision of shared decision-making  remains an as yet unachieved, but laudable goal; indeed, the present results highlight the challenges of shared decision-making and increased autonomy in patients with complex chronic diseases.
In summary, the results of this study suggest that patients tend to have limited input into the decision to initiate warfarin therapy. Moreover, a majority appear to lack a comprehensive understanding of the risks and benefits associated with treatment. These findings, however, were balanced by the minimal impact of warfarin on daily life and the high level of patient satisfaction. Further research is required to assess whether these findings are similar in other patient groups, with different demographic and socioeconomic characteristics, including multi-cultural communities . Investigation of physician views of the underutilization of warfarin therapy would allow for a comparison of the patient and provider perspectives. Clearly, there is a pressing need for innovative methods of continuing patient education in order to communicate the risks and benefits of warfarin therapy in a friendly, non-threatening manner. Also, these results highlight the tremendous influence that care providers exert on the decision-making of patients. The development of decision aids for anticoagulation may help patients make more informed decisions [23, 25], but only if care providers know of their existence and take the time to use them, assuming that such tools are feasible in a busy clinical setting.
REGU conceived and initiated the study and will act as guarantor. GCD, JM, and BT conducted the interviews. CST co-ordinated the data analysis process. All authors participated in the analysis of the data and the writing of successive drafts of the manuscript and all have read and approved the final draft.
Go A, Hylek E, Phillips K, Chang Y, Henault L, Selby J: Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk factors In Atrial fibrillation (ATRIA) Study. JAMA. 2001, 285: 2370-2375. 10.1001/jama.285.18.2370.
Petersen P, Boysen G, Godfredsen J, Andersen E, Andersen B: Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: the Copenhagen AFASAK Study. Lancet. 1989, 177: 175-179. 10.1016/S0140-6736(89)91200-2.
EAFL (European Atrial Fibrillation Trials Study Group): Secondary prevention in non-rheumatic atrial fibrillation after transient ischaemic attack or minor stroke. Lancet. 1993, 342: 1255-1262.
Hirsh J, Dalen J, Guyatt G: The sixth (2000) ACCP guidelines for antithrombotic therapy for prevention and treatment of thrombosis. American College of Chest Physicians. Chest. 2001, 119: 1S-2S. 10.1378/chest.119.1_suppl.1S.
Ridker P, Goldhaber S, Danielson E, Rosenberg Y, Eby C, Deitcher S, Cushman M, Moll S, Kessler C, Elliott C, Paulson R, Wong T, Bauer K, Schwartz B, Miletich J, Bounameaux H, Glynn R, PREVENT investigators: Long-term, low-intensity warfarin therapy for the prevention of recurrent venous thromboembolism. N Engl J Med. 2003, 348: 1425-1434. 10.1056/NEJMoa035029.
Bungard T, Ghali W, McAlister F, Buchan A, Cave A, Hamilton P, Mitchell L, Shuaib A, Teo K, Tsuyuki R: The relative importance of barriers to the prescription of warfarin for nonvalvular atrial fibrillation. Can J Cardiol. 2003, 19: 280-284.
Bungard T, Ghali W, Teo K, McAlister F, Tsuyuki R: Why do patients with atrial fibrillation not receive warfarin?. Arch Int Med. 2000, 160: 41-46. 10.1001/archinte.160.1.41.
Hart R: Anticoagulation therapy for patients with atrial fibrillation. CMAJ. 2000, 163: 956-957.
Devereaux P, Anderson D, Gardner M, Putnam W, Flowerdew G, Brownell B, Nagpal S, Cox J: Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: observational study. BMJ. 2001, 323: 1218-1222. 10.1136/bmj.323.7323.1218.
Howitt A, Armstrong D: Implementing evidence based medicine in general practice: audit and qualitative study of antithrombotic treatment for atrial fibrillation. BMJ. 1999, 318: 1324-1327.
Ceresne L, Upshur R: Atrial fibrillation in a primary care practice: prevalence and management. BMC Family Pract. 2002, 3: 11-10.1186/1471-2296-3-11.
Denzin N, Lincoln Y, eds: Handbook of Qualitative Research. 2000, Thousand Oaks, CA: Sage
McKinstry B: Do patients wish to be involved in decision making in the consultation? A cross sectional survey with video vignettes. BMJ. 2000, 321: 867-871. 10.1136/bmj.321.7265.867.
Lip GY, Kamath S, Jafri M, Mohammed A, Bareford D: Ethnic differences in patient perceptions of atrial fibrillation and anticoagulation therapy: The West Birmingham atrial fibrillation project. Stroke. 2002, 33: 238-244. 10.1161/hs0102.101817.
Nadar S, Begum N, Kaur B, Sandhu S, Lip G: Patients' understanding of anticoagulant therapy in a multiethnic population. J R Soc Med. 2003, 96: 175-179. 10.1258/jrsm.96.4.175.
Lancaster T, Singer D, Sheehan M, Oertel L, Maraventano S, Hughes R, Kistler J: The impact of along-term warfarin therapy on quality of life. Evidence from a randomized trial. Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators. Arch Intern Med. 1991, 151: 1944-1949. 10.1001/archinte.151.10.1944.
Sweeney K, Gray D, Steele R, Evans P: Use of warfarin in non-rheumatic atrial fibrillation: a commentary from general practice. Br J Gen Pract. 1995, 45: 153-158.
Arnsten J, Gelfand J, Singer D: Determinants of compliance with anticoagulation: a case-control study. Am J Med. 1997, 103: 11-17.
Waterman A, Banet G, Milligan P, Frazier A, Verzino E, Walton B, Gage B: Patient and physician satisfaction with a telephone-based anticoagulation service. J Gen Intern Med. 2001, 16: 460-563. 10.1046/j.1525-1497.2001.016007460.x.
Gafni A, Charles C, Whelan T: The physician-patient encounter: the physician as a perfect agent for the patient versus the informed treatment decision-making model. Soc Sci Med. 1998, 47: 347-354. 10.1016/S0277-9536(98)00091-4.
Man-Son-Hing M, Laupacis A, O'Connor A, Wells G, Lemelin J, Wood W, Dermer M: Warfarin for atrial fibrillation: the patient's perspective. Arch Intern Med. 1996, 156: 1841-1848. 10.1001/archinte.156.16.1841.
Protheroe J, Fahey T, Montgomery A, Peters T: The impact of patients' preferences on the treatment of atrial fibrillation: observational study of patient based decision analysis. BMJ. 2000, 320: 1380-1384. 10.1136/bmj.320.7246.1380.
Man-Son-Hing M, Laupacis A, O'Connor AM, Biggs J, Brake E, Yetesir E, Hart RG: A patient decision aid regarding antithrombotic therapy for stroke prevention in atrial fibrillation: a randomized controlled trial. JAMA. 1999, 282: 737-743. 10.1001/jama.282.8.737.
Elwyn G, Edwards A, Kinnersley P: Shared decision making in primary care: the neglected second half of the consultation. Br J Gen Pract. 1999, 49: 477-482.
Thomson R, Parkin D, Eccles M, Sudlow M, Robinson A: Decision analysis and guidelines for anticoagulant therapy to prevent stroke in patients with atrial fibrillation. Lancet. 2000, 355: 956-962. 10.1016/S0140-6736(00)90012-6.
The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2296/5/15/prepub
We are indebted to our three peer reviewers - Drs. Malcom Man-Son-Hing, Maggie Somerset, and David Armstrong- for their comments and suggestions on an earlier draft of this paper.
This study was funded by research grants from Physicians' Services Foundation Incorporated and Samcor/Sunnybrook Primary Care Research Trust. Dr. Upshur is supported by a New Investigator Award from the Canadian Institutes of Health Research and a Research Scholar Award from the Department of Family and Community Medicine, University of Toronto.
The authors would like to thank the 21 patients who volunteered to take part in this study and share their experiences with us. We would like to acknowledge the contribution of the nurses in the Family Practice Unit who referred the participants, and that of Dr. Denise Gastaldo who acted as an independent external reader. Special thanks also to Jennie Jones for transcribing the interviews and to Shari Gruman for formatting the paper.