Study design

An exploratory descriptive qualitative study design was employed to examine the day-to-day experiences and challenges of patients with prosthetic heart valves who were on long-term warfarin therapy following hospital discharge. This design was appropriate because the study sought to obtain detailed accounts of how patients experience, interpret, and manage long-term warfarin therapy in their everyday lives. Qualitative content analysis was used to analyze participants’ information so as to allow systematic description and interpretation of participants’ narratives while preserving the context in which their experiences occur. The analysis followed a combined inductive-deductive approach, whereby broad sensitizing domains informed the semi-structured interviews, while codes, sub-categories, and categories were developed from participants’ narratives rather than from fixed predetermined coding categories. This approach was important because it ensured that clinically relevant aspects of warfarin therapy were explored while still allowing participants’ own experiences, meanings, and challenges to shape the final findings.

Study setting

The study was conducted at the Jakaya Kikwete Cardiac Institute (JKCI) in Dar es Salaam, Tanzania. This is a national cardiac hospital, providing cardiovascular care, training, and research services. The institute has a 157-bed capacity and serves over 1,800 outpatients and 100 inpatients weekly. The institute was selected for this study because it caters to patients from all regions of Tanzania, referred from regional and designated hospitals for advanced cardiovascular care. Valvular surgical replacement and repairs are among the procedures performed at the hospital, with approximately 20–40 patients undergoing the procedure per month. After achieving post-surgical stability, patients are discharged home with individualized health education, primarily provided by the attending physician, dietitian, and occasionally by nurses. This education focuses on adherence to prescribed compliance with warfarin therapy, including the rationale and importance of anticoagulation, taking the medication at the recommended dose at the same time each day, recognizing and managing drug-drug and food-drug interactions, identifying symptoms of potential complications, and understanding strategies to prevent such complications. Patients are also required to attend monthly outpatient clinic appointments at the same hospital whenever possible. During these visits, patient prognosis is monitored, and adherence to warfarin therapy is assessed through patient interviews and INR laboratory testing. Laboratory INR results are typically released within approximately two hours, after which the INR level is evaluated. Patients whose INR falls within the therapeutic range (INR = 2.5–3.5) are discharged home. However, those with critical INR values, either suggestive of warfarin toxicity (INR > 5.0) or an increased risk of thrombosis (INR < 2.0), may be admitted for several days for stabilization and provided with additional education on appropriate warfarin use at home after hospital discharge.

Sample size and sampling of study participants

Participants were recruited using a purposive sampling technique guided by the principle of data saturation [27]. Maximum variation purposive sampling was used to obtain diverse experiences among participants based on age, sex, duration of warfarin use, type of prosthetic heart valve, and place of residence. Data saturation in this study was assessed iteratively during data collection and analysis after each interview through continuous review of emerging codes, categories, and meanings from participants’ narratives. This means that data collection stopped when the research team judged that information had become redundant, and no new codes, concepts, or deeper insights relevant to the study objectives were obtained through further data collection. In this study, code and meaning saturation were achieved with ten [10] participants after the research team judged that additional interviews were no longer generating new codes or substantially different meanings relevant to the study aim. An additional two [2] participants were recruited to confirm the consistency and saturability of the information obtained, making a total of twelve [12] study participants. A total of seventeen [17] eligible participants were approached for participation; however, five [5] declined due to competing personal commitments.

Inclusion and exclusion criteria

The study included patients who had undergone heart valve surgery, had a prosthetic heart valve, had been discharged home on warfarin therapy, and were at least three months post-surgery. Participants were also required to be willing to participate in the study after providing informed consent. Patients who were critically ill or unable to participate in an in-depth interview were excluded from the study.

Data collection method and tool

A semi-structured interview guide was used to collect data on participants’ day-to-day experiences and challenges with long-term warfarin use. The guide contained broad open-ended questions and sensitizing domains related to medication use, side effects, adherence behaviours, lifestyle modifications, motivators for adherence, perceived quality of life, education received from healthcare providers, conflicting advice, access to medication, and INR monitoring services. These domains were included from the literature and clinical experience to ensure that clinically relevant areas of long-term warfarin care were explored. However, interviews remained flexible, and participants were encouraged to describe their experiences in their own words and to raise additional issues that were important to them. Probing questions were used only when clarification or deeper explanation was needed.

Post-surgery participants who came for a monthly check-up at the clinic were consulted with the help of the nurse in-charge. Those who met the inclusion criteria and consented to take part in the study were asked for interview that was scheduled during the waiting time for blood laboratory results done to check their INR level. The waiting time is usually two hours. This period was considered the most convenient and practical time for conducting interviews because most clinic services at this facility are provided during daytime hours, and many participants travel from distant regions and usually leave immediately after receiving their results. Although waiting for INR results could potentially create anxiety among some participants, interviews were initiated only when participants appeared comfortable and willing to engage in discussion voluntarily. In addition, the interview took place in a private, quiet room to maintain confidentiality and lasted between 30 minutes and 45 minutes.

The interviews were conducted by the first author (B.T.M), a postgraduate Master of Science in Cardiovascular Nursing, using Kiswahili, a language with which both participants were conversant. At the time of data collection, B.T.M was also undertaking a clinical rotation at the outpatient department (OPD) of the study facility where study participants were receiving care. This positionality provided both advantages and potential biases, as it facilitated rapport and understanding of the clinical context but also raised the possibility of influence due to his dual role as a student and rotating clinician. To minimize this influence, participants were clearly informed that the interviews were conducted strictly for academic purposes and would not affect their clinical care. The researcher maintained an impartial and non-judgmental stance during interviews and avoided involvement in participants’ clinical decision-making during the study period. All interviews were recorded using audio digital recorders. The research assistant (A.A.O) was taking notes on the interview process and specific observations during interviews. These notes, together with non-verbal data, were expanded and triangulated during data analysis. Immediately after the first interview, data analysis was initiated, providing insight into preliminary sub-categories and emerging issues that were followed up on in subsequent interviews. The whole process of data collection lasted from April to June, 2025.

Ethical considerations

Ethical approval for this study with ref. No.DA.282/298/01.C/2751 was obtained from the Research and Ethics Committee of MUHAS. Permission to conduct the research with Ref. No: AB.123/307/01L/61 was granted by the hospital management. Before participation, potential participants were thoroughly informed about the study’s objectives, procedures, potential risks and benefits, confidentiality measures, and their rights to withdraw from the study at any time without any consequences to their care. Written informed consent was obtained from all participants prior to data collection. The consent form was in Kiswahili, and for participants with limited literacy, the consent was explained verbally to ensure full comprehension before written consent was obtained. Participants’ identities were kept anonymous, and confidentiality was maintained throughout the study. Audio-recorded data were securely stored on a password-protected computer accessible only to the research team. In addition, notebooks containing field notes with key observations were securely stored in a locked cabinet to prevent unauthorized access.

Data analysis

Qualitative content analysis guided by the work of Graneheim and Lundman was done manually to analyze the verbatim transcripts of the audio recordings [28]. The analysis was conducted using a combined inductive-deductive approach, whereby broad sensitizing domains informed the semi-structured interviews, while codes, sub-categories, and categories were developed from participants’ narratives rather than from fixed predetermined coding categories. The analysis began with the transcription of audio-recorded interviews in Kiswahili (The native spoken language used during the interview). Transcription was conducted by the Principal Investigator (PI) (B.T.M) with support from trained research assistants. The PI cross-checked all transcripts against audio recordings to ensure accuracy, capture important parts of participants’ narratives, and preserve contextual elements such as participant tones, pauses, and emphasis. Coding and analysis were conducted in the original Kiswahili transcripts to preserve meaning, and selected excerpts were later translated into English for reporting purposes [29]. Following transcription, the PI (B.T.M) immersed himself in the data by reading each transcript repeatedly to gain a comprehensive understanding of participants’ day-to-day experiences, challenges, and the overall context. Initial notes and reflections were documented to guide subsequent analysis. Next, the identification of meaning units was first carried out by the PI. These meaning units were then reviewed together with the supervisors (D.A.M and S.E.B) during scheduled discussion meetings. Through discussion and comparison with the original transcripts, a consensus was reached on the relevance and boundaries of the meaning units. The agreed meaning units were subsequently condensed by the PI to shorten the text while retaining the core meaning.

The PI generated preliminary codes from the condensed units. These codes, along with supporting excerpts, were shared with the supervisors (D.A.M and S.E.B) for independent review. The research team then met to discuss the codes, resolve discrepancies, refine code definitions, and agree on a common coding framework. This collaborative process ensured clarity, consistency, and analytical rigor. After agreement on the coding structure, similar codes were grouped into sub-categories by the PI. These sub-categories were presented to the supervisors for validation, where discussion focused on their internal homogeneity and external heterogeneity. Adjustments were made jointly until consensus was reached. Finally, the PI developed broader categories by interpreting relationships among the sub-categories and identifying the latent meanings within the data. These categories were critically reviewed, discussed, and refined together with the supervisors (D.A.M and S.E.B) to ensure that they accurately reflected participants’ narratives and addressed the research questions. The final categories were approved collectively by the research team. After the categories had been developed from participants’ narratives, Andersen’s Behavioural Model was used as an interpretive lens in the Discussion. This strengthened the interpretation by situating participants’ experiences within broader individual, social, and health-system influences on long-term warfarin use and access to anticoagulation care.

Moreover, Member checking was conducted with four purposively selected participants, representing approximately one-third of the study sample, with variation in sex, duration of warfarin use, place of residence, health insurance status, and anticoagulation status. Summarized findings were shared with them in Kiswahili to confirm whether the categories and sub-categories reflected their experiences and whether any interpretation was inaccurate or incomplete. Participants confirmed that the findings were consistent with their experiences. Minor wording clarifications were made, but no new category or sub-category emerged following member checking.

Trustworthiness of the study

The trustworthiness of this qualitative study was ensured by adhering to four criteria by Lincoln and Guba: credibility, dependability, confirmability, and transferability [30]. In this study, credibility was ensured through triangulation of data from both the transcripts and those generated from the field notes. The first author has a prolonged engagement in a study setting, as he was a postgraduate Master of Science in Cardiovascular Nursing student undertaking clinical rotation at the study site before and during the data collection, which enabled familiarity with the context and improved understanding of participants’ experiences. He was directly involved in data collection; hence, he was able to capture the reality of those being studied. Furthermore, credibility was ensured through the purposeful recruitment of participants with diverse backgrounds and experiences in using warfarin post-prosthetic surgery.

To ensure dependability, an audit trail was maintained documenting coding decisions in the code book and sub-categories and categories development among the research team. Also, by using a gatekeeper in the recruitment of participants, interacting with them at the clinic, establishing rapport, and conducting interviews in Kiswahili, the researcher gained trust with the informants, ensuring the findings accurately reflect the data. Confirmability of the study was ensured through the provision of the details of the actual study setting, sampling criteria, data collection processes, and tools. Lastly, transferability was supported through an elaborative explanation of study context, participant characteristics, data collection procedures, and data analysis, allowing readers to assess the applicability of findings to their own settings.

Category 1: Adapting to life with long-term warfarin use

This category comprises four sub-categories: modification and adoption of a new eating style; relief from clinical symptoms and improved quality of life; discomfort and undesired changes from warfarin use; and limited engagement in socio-economic activities post-surgery.

Category 2: Navigating warfarin care within an unstructured continuum of care

This category comprises five sub-categories: being tiresome and inconsistent on warfarin use; increased financial burden to cover treatment costs; non-adherence of proper warfarin use at home; contradictory information and mixed directives from healthcare providers; and centralized service and limited access to anticoagulation Care.

Acceptance, discomfort, relief, and modified life experience: A need for supportive and patient-centered care post -prosthetic valvular surgery

Participants reported major lifestyle adjustments after starting long-term warfarin therapy, particularly dietary practices. Many perceived these instructions as strict dietary restrictions that disrupted their usual dietary routines, rather than as guidance on maintaining consistent vitamin K intake by regulating intake of foods such as green vegetables, avocados, and certain spices. This experience reflects the difficulty of adapting to lifelong anticoagulation therapy while trying to preserve familiar eating habits in the Tanzanian context, where leaf vegetables are a common, affordable staple. The findings suggest that inadequate or overly restrictive dietary counselling may unintentionally contribute to fear, dissatisfaction, and poor adherence, underscoring the need for patient-centered care that integrates clinical follow-up, patient education, emotional support, and lifestyle counselling after prosthetic valvular surgery. Adhering to stable dietary patterns is essential to prevent INR fluctuations and reduce bleeding or thrombotic risks for patients on long-term warfarin therapy [7,13]. Similar concerns of dissatisfaction with dietary information from HCPs have been reported in other SSA countries, such as Ethiopia and Kenya, where gaps in patient education affected anticoagulation control [16,22]. Therefore, patient counselling should focus on culturally appropriate explanations emphasizing dietary consistency rather than food avoidance, while incorporating emotional support and the use of simple dietary guides to help patients adapt to lifestyle modifications.

Study participants also reported relief from debilitating pre-surgery symptoms such as breathlessness, fatigue, and limited physical functioning following surgery and initiation of warfarin therapy. This improvement motivated continued adherence to treatment and reflected positive adjustment to life after surgery. Similar findings have been reported in studies from Iran and Turkey, where patients reported feelings of renewed health and viewed warfarin as a protector of their second chance at life, particularly in helping them prevent prosthetic valve-related complications [9,32]. These findings suggest that integrated follow-up care involving clinical monitoring, continuous patient education, and psychological support may help sustain this positive recovery experience and promote long-term warfarin adherence.

Despite these improvements, participants also experienced discomfort and uncertainty related to long-term warfarin use, including dizziness, itching, gum bleeding, and menstrual changes. Some participants were unsure whether these symptoms were related to warfarin or other conditions, suggesting gaps in pharmacological counselling and continuity of follow-up care. Considering the challenging health system in Tanzania, where access to consistent health care provider interaction is limited, patients often self-manage or normalize such symptoms, leading to delaying appropriate care with ultimate consequences of severe adverse effects, such as warfarin toxicity necessitating hospitalization for these patients [26]. This situation highlights the urgent need for follow-up care models that extend beyond routine clinic visits to include accessible consultation pathways and patient-centered monitoring. These findings are similar to the study done in Kenya, which revealed gaps in warfarin-related knowledge, where 65.4% patients had poor knowledge on warfarin [22]. Therefore, strengthening patients’ counselling, improving follow-up systems, and fostering supportive patient-provider relationships are essential to minimize discomfort, enhance safety, and promote a more adaptive and supported long-term adherence to warfarin post-operatively.

Study participants also reported difficulty resuming their previous work duties, economic activities, or social roles despite the clinical improvements. The fear of overexertion and incomplete functional recovery was a barrier to resuming previous pre-morbid roles. This struggle reflects the broader modified life experience following long-term warfarin use post-operatively, where physical recovery does not always transform into full social and functional restoration, emphasizing the need for supportive and patient-centered care. In Tanzania’s predominantly informal, labor-intensive economy, such limitations create further challenges in the financial burden to cover treatment costs and threaten livelihoods [33]. Other studies done in Tanzania, Botswana, and Nigeria have reported similar challenges, where quality of life improvements post valvular surgery were stalled by limited socio-economic reintegration support, with the majority of patients (>50%) not resuming their previous roles [12,33,34]. These findings suggest that without structured reintegration strategies, patients may continue to experience discomfort and uncertainty despite successful surgery and proper adherence to warfarin postoperatively. Therefore, integrating rehabilitation services, HCPs-led functional assessments, and gradual return to work guidance into routine follow-up could foster acceptance, reduce distress, and support a more meaningful recovery transition for these patients

Challenge of long-term use of warfarin: Query for a supportive system and structured continuum care within the health system

In this study, poor compliance with long-term warfarin therapy was not due to individual negligence, but rather to the faults in the health system, reflecting the broader challenge of long-term warfarin use post-valvular heart surgery. Study participants reported difficulties maintaining consistent warfarin use, including missed doses, self-adjustment of medication dosage, not following dietary instructions, and interruptions in clinic attendance due to financial issues, transport costs, and limited access to INR monitoring services. Structural barriers such as centralized anticoagulation services, medication shortages in peripheral facilities, and out-of-pocket expenditure constrained access to optimal care, particularly in resource-limited settings. Similar challenges have been reported in other SSA countries such as Botswana, Kenya, and Ethiopia, where access to INR testing and affordability of anticoagulation care remain major barriers to anticoagulation therapeutic control [11,16,22]. In contrast, in developed countries such as Turkey and the UK, there are more advanced healthcare systems, and patients are reported to benefit much from community-based anticoagulation clinics and home-based INR self-testing devices, thereby reducing guesswork among these patients [19,35]. Such models highlight the role of structured care in transforming anticoagulation management from reactive self-care to supervised, patient-centered support. In Tanzania, where specialized cardiac services remain centralized in urban areas, patients from distant regions experience substantial financial and logistical barriers in accessing follow-up care. These findings highlight the need for decentralized anticoagulation services, improved availability of warfarin in peripheral facilities, and strengthened continuity of care for patients requiring lifelong anticoagulation therapy.

This study reports a dynamic interplay between patient agency and structural constraints in the long-term use of warfarin among patients with prosthetic heart valves. Participants reported active self-management behaviors, including modification of dietary intake and medication use, reflecting efforts to navigate complex treatment demands in everyday life. Similar challenges have been reported in other sub-Saharan African countries, where suboptimal anticoagulation control and variability in adherence are common among patients on vitamin K antagonists [11,16,34] Evidence from Tanzania has also shown poor INR stability among patients on warfarin, highlighting persistent difficulties in maintaining therapeutic anticoagulation [26,34]. These challenges are partly attributable to limited patient education and inconsistent follow-up, which have been identified as important determinants of warfarin adherence in previous studies [18,25]. The coexistence of patient-driven coping strategies and health system limitations suggests that adherence is shaped by both individual and contextual factors rather than patient behavior alone. This situation highlights the need for patient-centered follow-up systems that extend beyond routine clinic visits through accessible counselling and continuity of care.

Financial issues were also a major issue reported by several participants. Many participants reported having to purchase warfarin using personal or family funds, often delaying or interrupting care due to the inability to afford travel, tests, or medication. The situation is worse for the majority of participants due to low income, since they lack health insurance, and therefore depend on out-of-pocket payments, illustrating the fragile continuum of care available to these patients, where treatment adherence becomes contingent on financial capacity rather than clinical need [36]. Findings from the study done at JKCI revealed the same issue, as some patients on warfarin missed appointments as a result of a lack of transport fare and a lack of money to cover warfarin costs [33]. Findings from a study done in Ethiopia [16] and other SSA countries also pointed out the same issue of affordability and anticoagulation control, with warfarin being a challenge [11,34]. These differences further underscore the need for strengthening financial risk protection as central to building a supportive system for long-term anticoagulation care.

Another important finding was the contradictory information participants received from healthcare providers regarding warfarin dosage, timing, and dietary practices. This confusion and inconsistency highlight the lack of standardized warfarin treatment protocols and inadequate training among healthcare providers in managing patients on warfarin therapy. Such variability places patients in a vulnerable position where they are required to interpret and reconcile conflicting instructions, increasing the risk of dosing errors, poor adherence, and adverse outcomes such as over-anticoagulation. This finding highlights not only communication gaps but also systemic weaknesses in continuity of care for patients requiring long-term warfarin monitoring. Similar findings were also observed in studies done in Iran and India, where inconsistent counselling contributed to patient confusion regarding warfarin management [32,37]. Contrary to countries like Canada, there is a better integrated health system, with the use of standardized warfarin protocols and the use of shared electronic medical records that help in ensuring consistency in communication among HCPs and eventually minimize patient confusion and promote safe warfarin use in the long run [15,18]. While the lack of integrated electronic medical record systems may contribute to these inconsistencies in Tanzania, the findings may also reflect broader challenges, such as limited standardized anticoagulation protocols, inconsistent provider training, and weak communication between tertiary cardiac centers and peripheral healthcare facilities where patients occasionally seek care. However, considering resource constraints in Tanzania, we might opt for less expensive entry points such as the use of low-cost paper-based communication booklets, WhatsApp groups for provider coordination, and SMS reminders to patients. Furthermore, improving coordination between referral and peripheral facilities is essential for a safer and more reliable continuum of care for patients on long-term warfarin management.

Study strengths and limitations

The strength of this study lies in its focus on the patient’s lived experiences of long-term warfarin use in a low-resource setting in Tanzania, a country where most previous studies have primarily focused on clinical outcomes rather than patient perspectives and experiences. These findings provide important insights for improving anticoagulation care, patient health education, and decentralization of warfarin service delivery in similar low-resource settings. However, the study should be interpreted in light of some important limitations. First, there is potential recall and social desirability bias due to reliance on self-reported data. This was mitigated by assuring participants of confidentiality, encouraging honest responses, and using probing techniques during interviews to enhance the credibility and depth of the information collected. Second, this qualitative study was conducted at a single national cardiac institute, which may limit the transferability of the findings to other healthcare settings. Although the national cardiac institute receives patients from diverse geographical regions across Tanzania, patients attending a national referral center may differ from those managed at regional or district hospitals in terms of disease severity, frequency of follow-up, and access to specialized anticoagulation care. Therefore, while the national referral role of JKCI enabled the study to capture a broad range of patient experiences from different parts of the country, the findings should be interpreted with consideration of the specialized nature of the study setting and may be most transferable to a similar specialized cardiac care context. Third, the study has methodological limitations related to the use of a semi-structured interview guide. Although the guide helped ensure that clinically important aspects of long-term warfarin therapy were explored, it included sensitizing domains that are already recognized in the literature, such as forgetfulness, side effects, dietary modification, financial burden, inconsistent education, access to medication, and INR monitoring. This may have increased the possibility that interviews confirmed known barriers rather than identifying entirely new experiences. This risk was minimized by framing questions in an open-ended manner, using probes flexibly, and encouraging participants to raise additional issues beyond those included in the interview guide. During analysis, codes, sub-categories, and categories were retained only when grounded in participants’ narratives. Therefore, the findings should be interpreted as descriptive qualitative accounts generated through a combined inductive–deductive qualitative content analysis, rather than as findings from a purely inductive qualitative design. Fourth, although interviews were conducted in Kiswahili to ensure clarity and comfort among study participants, the subsequent translation of transcripts into English for analysis and reporting may have led to minor loss of contextual meaning or misinterpretation. Coding and initial analysis were conducted using the original Kiswahili transcripts before translation into English to preserve participants’ intended meanings. In addition, translated transcripts and selected quotations were reviewed by another bilingual member of the research team to enhance accuracy and consistency of interpretation, although formal back-translation was not performed. Member checking was conducted by returning summarized findings to participants for validation, which strengthened credibility and ensured that interpretations accurately reflected participants’ experiences.

Conclusion and recommendations

This study explored day-to-day experiences and challenges encountered by patients with prosthetic heart valves on long-term use of warfarin at Jakaya Kikwete Cardiac Institute. Even though participants acknowledged warfarin as a vital life-saving anticoagulant, its long-term use was associated with substantial lifestyle adjustments. including dietary modification, emotional stress, discomfort from perceived side effects, and concerns about maintaining therapeutic INR levels consistently. Study participants also encountered challenges such as poor access to INR monitoring services, unavailability of warfarin in peripheral areas, financial burden, and contradictory information from HCPs. These findings demonstrate that long-term warfarin adherence is influenced not only by individual patient behaviors but also by structural and health system-related barriers within the Tanzanian context. Therefore, to promote safe and more equitable warfarin anticoagulation management, healthcare systems need to invest in accessible INR monitoring services in all regions, warfarin availability and affordability, organized follow-up care, and patient health education with emphasis and active engagement of patients at each clinic visit.

The study recommends the need for a comprehensive approach to improve warfarin management. To address contradictory counselling and fragmented communication between healthcare providers, simple and low-cost communication strategies such as standardized paper-based anticoagulation booklets, shared clinic notes, and WhatsApp-based coordination among providers may improve continuity of care. To address missed appointments and poor follow-up, mobile phone-based SMS reminders and follow-up calls may serve as feasible interventions in low-resource settings. Strengthening patient education using culturally appropriate materials in local languages and reinforcing counselling at each clinic visit may also improve understanding of dietary consistency, medication adherence, and INR monitoring.

At the health system level, expanding access to INR testing services and improving warfarin availability in regional and district hospitals could reduce financial and geographical barriers to care. Where feasible, point-of-care INR monitoring devices may further improve timely anticoagulation monitoring in peripheral settings. In addition, the incorporation of full coverage for warfarin-related services into public insurance schemes may help reduce out-of-pocket costs among patients requiring lifelong anticoagulation therapy. Finally, future research should explore healthcare providers’ perspectives on anticoagulation management and evaluate the feasibility and effectiveness of low-cost digital follow-up strategies such as SMS reminders and mobile health-supported anticoagulation care in Tanzania and similar low-resource settings.