A Cross-Sectional Study on Oral Anticoagulant Utilization and Adverse Drug Reactions at Teaching Hospital, Iran Authors: Shima Jafari, Sara Farooghi, Mandana Moradi Published: 2026 ABSTRACT Oral anticoagulants are widely used to prevent and treat thromboembolic disorders, but inappropriate use can increase adverse effects and reduce therapeutic efficacy. This study evaluated the utilization patterns, dosing appropriateness, and adverse effects of oral anticoagulants in a tertiary hospital in Iran. A retrospective cross-sectional study was conducted at Razi Educational Hospital, Birjand University of Medical Sciences, from April 2020 to September 2022. Data were collected from medical records and pharmacy databases for patients aged ≥18 years who received warfarin, rivaroxaban, or apixaban. Dosing appropriateness was assessed using 2021 European Heart Rhythm Association (EHRA) and UpToDate guidelines, while adverse effects were classified according to 2020 American College of Cardiology (ACC) criteria. Among 1,027 patients (mean age 64±16.4 years; 52.3% male), rivaroxaban was the most prescribed anticoagulant (43.5%), followed by warfarin (33.2%) and apixaban (23.3%). Warfarin was primarily used for valvular heart disease, whereas DOACs were mainly prescribed for non-valvular atrial fibrillation and venous thromboembolism. Adverse effects occurred in 12.1% of patients, mostly minor bleeding, with warfarin accounting for the majority. Guideline-based dosing was achieved in 61.9% of apixaban and 58.6% of rivaroxaban users; 46.9% of warfarin patients reached therapeutic INR (2-3). Correct renal dose adjustments were applied in 40.5% of apixaban and 33% of rivaroxaban users. Guideline adherence for oral anticoagulant dosing was suboptimal. Improved prescriber education, enhanced drug utilization evaluation, and routine renal monitoring may enhance patient safety and optimize anticoagulation outcomes. INTRODUCTION Cardiovascular diseases (CVDs) are a leading cause of premature mortality and a major contributor to global healthcare costs. Their prevalence continues to rise, with the global burden nearly doubling from 271 million cases in 1990 to 523 million in recent years.1 Antithrombotic therapy plays a central role in CVD management, aiming to prevent ischemic events while balancing the inherent risk of bleeding.2 Among antithrombotic agents, anticoagulants play a central role in the prevention and treatment of a broad range of thromboembolic conditions, including ischemic heart disease, atrial fibrillation, and venous thromboembolism.3 For many years, warfarin has served as an important anticoagulant in clinical practice. But as a conventional oral anticoagulant is no longer widely used because of its narrow therapeutic index, various drug and food interactions, unacceptable adverse effect profile, and required regular laboratory monitoring.4 In contrast, direct oral anticoagulants (DOACs)—including dabigatran, rivaroxaban, apixaban, and edoxaban—offer more predictable pharmacokinetics, rapid onset and offset, and do not require routine monitoring. Compared with warfarin, DOACs are associated with lower rates of major or fatal bleeding.5 These advantages contribute to their widespread adoption in clinical practice.6 Drug Utilization Evaluation (DUE) is a systematic, ongoing process used in healthcare to assess the appropriateness, safety, and effectiveness of medication use. It aims to ensure that patients receive optimal therapeutic outcomes by identifying patterns of inappropriate prescribing, dispensing, or administration. DUE programs typically involve reviewing prescribing data against established criteria, evaluating clinical outcomes, and implementing interventions such as prescriber education or formulary adjustments to improve medication use. This process not only enhances patient safety but also helps control healthcare costs by minimizing adverse drug reactions and therapeutic duplication. DUE is an integral component of rational drug use and quality assurance in pharmacy practice, as recommended by organizations such as the World Health Organization (WHO) and the American Society of Health-System Pharmacists (ASHP).78 Despite the widespread use of oral anticoagulants, data regarding their utilization patterns and associated adverse effects are lacking in the region where we practice. Therefore, we designed this study to comprehensively evaluate multiple aspects of oral anticoagulant utilization. By analyzing these parameters, we would be able to identify potential areas for improvement in anticoagulant management, promote rational drug use, and provide evidence-based recommendations to enhance patient safety and therapeutic outcomes in clinical practice. MATERIALS AND METHODS This retrospective, cross-sectional, single-center study was conducted at Razi Educational Hospital, affiliated with Birjand University of Medical Sciences (BUMS) in South Khorasan, Iran. The hospital provides comprehensive inpatient and outpatient services. Data for the study were systematically extracted from both the hospital’s pharmacy computerized information system and patients’ medical records to ensure completeness and accuracy. The study period spanned from April 2020 to September 2022. The study was approved by the Medical Ethics Committee of Zabol University of Medical Sciences (IR.ZBMU. REC.1401.064). All procedures were conducted in accordance with the ethical principles of the Declaration of Helsinki (1975, as revised in 2008). Written informed consent was obtained from all participants prior to their enrollment in the study. All patients aged over 18 years who were taking warfarin, rivaroxaban, or apixaban during the study period were included. Exclusion criteria included pregnancy, age under 18 years, and incomplete medical records. A standardized data collection form was developed to ensure consistency and completeness of data. The form captured detailed information on patient demographics (age, gender, and weight and Sr Cr), diagnosis, and admission and discharge dates. Creatinine clearance (CrCl) for each patient was calculated using the Cockcroft–Gault equation. It also included data on anticoagulant therapy, specifying the drug used, therapeutic indication, dosage, dosing interval, route of administration, and duration of treatment. Concomitant anticoagulant use and relevant laboratory parameters, including hemoglobin, platelet count, prothrombin time (PT), activated partial thromboplastin time (aPTT), international normalized ratio (INR), aspartate aminotransferase (AST), alanine aminotransferase (ALT), albumin, and serum creatinine levels, were recorded. Additionally, medication-related complications such as bleeding or elevated liver enzymes, their onset and duration, and patient outcomes (survival or death) were documented. According to the 2022 ACC/AHA guidelines, bleeding complications were categorized as either major or minor.9 Hepatic adverse events were defined as an elevation of liver enzymes to more than three times the upper limit of normal, or an increase in bilirubin levels to more than twice the normal value.10 The appropriateness of apixaban and rivaroxaban dosing and medication switching was assessed in accordance with the 2021 European Heart Rhythm Association (EHRA) guidelines.11 Warfarin dosing regimen was evaluated based on nomograms described in Applied Therapeutics: the clinical use of drugs.12 Bleeding events (major or minor) and hepatic adverse effects and their management were defined according to the 2020 American College of Cardiology (ACC) guideline.13 Statistical analyses were performed using IBM SPSS Statistics, version 26 (IBM Corp., Armonk, NY, USA). Quantitative variables were summarized as mean±standard deviation (SD), and qualitative variables as frequencies and percentages. The prevalence, type, and severity of adverse drug reactions were expressed as percentages, and group comparisons were conducted using the Chi-square test. A p-value<0.05 was considered statistically significant. RESULTS The demographic and clinical characteristics of the study population are summarized in Table 1. In total, 1,027 patient records were reviewed, with a mean age of 64.0±16.4 years (range: 15–97 years). Males comprised 52.3% of the cohort. Most patients (70.8%) reported no history of smoking or drug abuse. The mean duration of hospitalization was 7.3±10.3 days (range: 1 day to 3 months). Overall, 36.3% of patients were hospitalized for fewer than 4 days, 35.8% for 4–7 days, 18.4% for 7–14 days, and 9.4% for more than 14 days. As shown in Table 1, no statistically significant differences were observed among the three groups in terms of the examined demographic characteristics. Among the anticoagulant drugs prescribed, rivaroxaban was the most frequently used (447, [43.5%]), followed by warfarin (341, [33.2%]) and apixaban (239, [23.3%]). The majority of prescriptions (68.6%) were issued by cardiologists, while neurologists and internal medicine specialists accounted for 20.9% and 10.4%, respectively. The indications and mean ages of patients receiving each drug are summarized in Table 2. The mean age of patients receiving apixaban was significantly higher than that of the other two groups (p<0.05) (Table 2). At least one adverse effect was observed in 124 patients (12.1%), with minor bleeding being the most frequently reported complication (50%). Additional details of observed complications stratified by drug are summarized in Table 3. The duration of complications ranged from 1 to 17 days, with the majority of patients (28.2%) experiencing complications for two days (Table 3). Bleeding was the most frequent adverse effects with all three agents. Among the 101 patients who developed bleeding complications, 69 (68.3%) experienced minor bleeding and 32 (31.6%) experienced major bleeding. There was a statistically significant difference in the distribution of adverse effect types among patients receiving warfarin, rivaroxaban, and apixaban (p<0.001). The frequency of warfarin, apixaban and rivaroxaban related complications by age gender and smoking or substance use is shown in Table 4. Due to the small sample size in the groups for the complications presented in Table 4, the chi-square test could not be performed to calculate the p-value. Warfarin dose was categorized into four predefined groups ranging from 2.5 mg to 10 mg. Bleeding was assessed as part of drug-related adverse events, which included bleeding and hepatic adverse effects. The association between warfarin dose and bleeding events was examined using the chi-square test. Our analysis demonstrated no statistically significant difference in bleeding incidence across the warfarin dose groups (p=0.24). Table 5 presents the frequency of underlying diseases in patients experiencing side effects for each anticoagulant. The analysis revealed a significant association between cardiovascular disease and the type of anticoagulant prescribed among patients with complications (p<0.001). Among all patients experiencing drug-related complications, 97 (78.2%) received some kind of drug management. Of these, 81 patients (83.5%) managed appropriately, whereas 16 patients (16.5%) did not receive principled or evidence-based management. All information regarding the administered doses of the three studied drugs is summarized in Table 6. Based on the data, warfarin at a dose of 5 mg was the most frequently prescribed regimen among patients receiving warfarin, accounting for 53% of cases. At least one of the warfarin-sensitizing conditions, as described in the Methods section, was identified in only 37 patients (10.85%) receiving a 2.5 mg/day warfarin regimen. Similarly, among patients treated with apixaban, the 2.5 mg twice-daily (BID) regimen was the most common, prescribed in 42.7% of patients. Based on the EHRA guidelines described in the Methods, the appropriate dose was administered in only 134 patients (56%). Twenty-seven (11.3%) patients had serum creatinine levels >2.5 mg/dL and, according to the EHRA guidelines, were contraindicated for initiation of apixaban. In those receiving rivaroxaban, the majority (32.9%) were administered a 20 mg once-daily (QD) dose. Using the EHRA guidelines described in the Methods, the appropriate dose was administered in only 219 patients (49%). Seven (1.56%) patients had serum creatinine levels >2.5 mg/dL and, according to the EHRA guidelines, were not candidates for rivaroxaban therapy. Our routine practice involves INR monitoring at baseline and every other day after hospitalization. Among the studied patients, INR was routinely assessed every other day following admission. Baseline INR was measured in 322 patients, while in 19 patients (5.57%), baseline INR was not checked. Of these, 2 patients (0.57%) had a prior history of warfarin therapy. Among patients with a known history of warfarin use who were admitted with an abnormal INR (n=67), 25 patients (33.7%) were categorized as having warfarin toxicity (INR >6). For patients newly initiated on warfarin after hospitalization, an INR of 1 was considered a normal baseline value, whereas for patients already taking warfarin prior to admission, an INR range of 2–3 was considered within the normal therapeutic range. Regarding warfarin dose adjustment, at least one episode of elevated INR was observed in 141 patients, while guideline-based dose adjustments were performed in only 115 patients. In patients receiving apixaban and rivaroxaban, serum creatinine was measured in only 137 (57.3%) and 184 (41.16%) patients, respectively. Of these, 47 patients (19.7%) in the apixaban and 39 (8.72%) in the rivaroxaban group exhibited elevated serum creatinine, but dose adjustment based on creatinine levels was performed in only 19 (7.9%) and 13 (2.9%) of those patients, respectively. DISCUSSION Oral anticoagulant agents play a pivotal role in the prevention and management of thromboembolic disorders. However, inappropriate utilization of these medications may not only result in significant economic burden but also increase the risk of adverse clinical outcomes. In particular, for the newer oral anticoagulants such as rivaroxaban and apixaban, clinical experience regarding their optimal use remains relatively limited. Consequently, assessing current prescribing patterns and identifying the strengths and weaknesses within various healthcare systems—based on international clinical guidelines—can provide valuable insights for promoting the rational, safe, and cost-effective use of these agents. In our study, rivaroxaban was the most frequently prescribed anticoagulant, followed by warfarin and apixaban, consistent with patterns reported in previous retrospective studies.14 The preference for DOACs likely reflects their fixed dosing and the absence of a need for routine monitoring, providing practical advantages over warfarin. Cardiologists accounted for the majority of prescribers, incontrast to other studies reporting a more varied distribution of prescribing specialists.15 This discrepancy may be due to our smaller sample size, as well as national regulatory and insurance reimbursement policies for certain restricted drugs, and regional or hospital-specific prescribing routines, all of which could influence prescribing patterns. In our study, the most common indication for prescribing warfarin was thromboembolic prophylaxis in patients with valvular disorders, whereas prevention of thromboembolic events in non-valvular arterial fibrillation (AF) and venous thromboembolism (VTE) treatment were the predominant indications among patients receiving apixaban and rivaroxaban. However, in the study conducted by Phillippe et al.16, the distribution of indications for these anticoagulant agents differed from our findings. Venous thromboembolism (VTE) prophylaxis is one of the established indications for the use of direct oral anticoagulants (DOACs). Different studies demonstrated their comparable efficacy and safety to low-molecular-weight heparins (LMWHs) for this indication.17 Moreover, DOACs have been shown to be more cost-effective in this setting due to their oral administration and reduced need for monitoring.18 However, as demonstrated in our study, these agents are not commonly prescribed for VTE prophylaxis in our hospital. This may be partly attributed to the long-standing reliance on LMWHs and entrenched prescribing routines within our institution. This finding highlights an area that warrants further attention to optimize anticoagulation practices in accordance with current evidence and guidelines. In our study, a total of 341 patients receiving warfarin were evaluated. Among these, 160 patients (46.9%) achieved an INR within the therapeutic range of 2–3 during hospitalization, indicating an appropriate dosing regimen in this subset. This rate is even higher compared to findings reported in previous studies conducted by Ebrahim et al.19 and Mwita et al.20 In another study conducted in a secondary healthcare setting in India, patients attending the clinic had subtherapeutic INR values for approximately two-thirds of the time.21 Review of patients who were initiated on a warfarin dose of 2.5 mg per day revealed that only 37 individuals had at least one condition predisposing them to increased sensitivity to warfarin, in accordance with institutional protocols, while others had no known or mentioned risk for in their records to justify this drug regimen. In addition, the starting dose of 1.25 mg per day (¼ tablet) was not recommended in any of our referenced guidelines. Regardless of other factors, an inappropriate initial dose according to guideline recommendations may be one of the reasons for subtherapeutic INR levels in the patients of our study. In our study, the overall appropriate dosing of apixaban and rivaroxaban was 61.9% and 58.6%, respectively. For newly initiated therapy, 60.9% of apixaban users and 58.4% of rivaroxaban users received the correct starting dose at the hospital. Notably, Renal function was not monitored properly in about half of our study population and dose adjustments based on renal function were less consistently implemented, with only 40.5% of apixaban and 33% of rivaroxaban patients receiving the correct adjusted dose. These findings highlight a moderate level of adherence to recommended dosing guidelines, consistent with the variability reported in other studies. Some investigators have reported higher rates of appropriate regimen for both apixaban and rivaroxaban,2223 while others reported lower rates.2425 These results warranted the importance of continued patient monitoring to optimize the DOAC dosing regimen, particularly in patients with some degree of renal impairment. This will minimize the risk of adverse events such as bleeding or thromboembolism and improve overall treatment efficacy.26 In our study, complications were reported in 124 patients (12.1%), with minor bleeding being the most common adverse event. Overall, the incidence of complications was highest among patients receiving warfarin and lowest with apixaban, consistent with prior studies on the safety profiles of oral anticoagulants.27 Regarding hepatic adverse events, warfarin was associated with the highest incidence in our cohort, whereas previous research reported a greater frequency of liver-related complications with rivaroxaban compared to other agents.28 These differences highlight the significance of racial and regional variations in the occurrence of drug-related adverse effects and may serve as a basis for future studies aimed at identifying these variations to optimize therapy. These findings also highlight the variable safety profiles of oral anticoagulants and underscore the importance of careful monitoring of parameters. Clinicians should consider these differences when selecting an appropriate oral anticoagulant, balancing efficacy with the risk of bleeding and hepatic complications, considering patients’ characteristics. In the investigation of the relationship between the prevalence of warfarin side effects and patients’ demographics, it was observed that the majority of patients were over 71 years old, and the prevalence of side effects with warfarin was higher in women. In studies, this relationship with age and gender in the incidence of warfarin side effects was confirmed.29 We found no significant association between warfarin dose and bleeding events. This finding may be explained by careful dose adjustment and INR monitoring in our cohort. In addition, a relatively high proportion of patients were initiated on low warfarin doses and maintained with subtherapeutic INR which may also have contributed to this finding. Analysis of rivaroxaban-associated adverse events also revealed that the majority of patients experiencing complications were aged above 61 years, and the occurrence of side effects was higher in men than in women. However, previous studies have reported no significant correlation between rivaroxaban side effects and either age or sex.30 This difference may be attributed to regional genetic variations and differences in the study populations. Regarding apixaban, patients experiencing adverse effects were generally older than those using the other anticoagulants. Additionally, the incidence of complications was comparable between men and women, consistent with the findings reported by Frost et al.31 Considering the overall lower incidence of adverse effects with apixaban, this drug may represent a preferable choice, particularly for individuals more susceptible to drug-related side effects, including the elderly. Overall, identifying factors that increase susceptibility to drug-related adverse effects can significantly aid in optimizing therapy, as it allows for more precise recognition of at-risk individuals and prevents such adverse events. This study has several limitations that should be acknowledged. First, its retrospective cross-sectional design relied on medical record review, which may have introduced documentation bias and limited the availability of certain clinical variables. Second, the study was conducted at a single tertiary care hospital, which may restrict the generalizability of the findings to other healthcare settings in Iran or internationally. Third, the lack of pre-hospital admission information makes it difficult. Furthermore, the unavailability of data on patients’ anticoagulant therapy prior to hospital admission, including treatment duration and regimen type and changes, limited the ability to fully interpret in-hospital management practices and clinical decision-making. In conclusion, this study demonstrates that while oral anticoagulants are widely used for thromboembolic disorders, adherence to recommended dosing guidelines is moderate, particularly for renal dose adjustments in DOACs and starting doses of warfarin. Warfarin was associated with the highest incidence of complications, whereas apixaban showed the most favorable safety profile. These findings highlight the need for guideline-based prescribing, careful patient monitoring, and rational use of DOACs, especially for underutilized indications such as DVT prophylaxis, to optimize treatment efficacy and minimize adverse outcomes.