Identifying and mitigating prescribing errors in outpatient clinics: a prospective analysis of pharmacist interventions and clinical outcomes

Identifying and mitigating prescribing errors in outpatient clinics: a prospective analysis of pharmacist interventions and clinical outcomes Derar H. Abdel-Qader1, Shorouq Al-Omoush1, Khalid Awad Al-Kubaisi2, Esra’ Taybeh3, Nadia Al Mazrouei2, Rana Ibrahim2, Salim Hamadi1, Sahar Jaradat1, Alia Saleh1 1 Faculty of Pharmacy and Medical Sciences, University of Petra, Amman, Jordan 2 Department of Pharmacy Practice and Pharmacotherapeutics, College of Pharmacy, University of Sharjah, Sharjah, United Arab Emirates 3 Department of Applied Pharmaceutical Sciences, School of Pharmacy, Isra University, Amman, Jordan Corresponding author: Derar H. Abdel-Qader (d.bala[email protected]u) Received 28 June 2025♦ Accepted 2 October 2025♦ Published 27 October 2025 Citation: Abdel-Qader DH, Al-Omoush S, Al-Kubaisi KA, Taybeh E, Al Mazrouei N, Ibrahim R, Hamadi S, Jaradat S, Saleh A (2025) Identifying and mitigating prescribing errors in outpatient clinics: a prospective analysis of pharmacist interventions and clinical outcomes. Pharmacia 72: 1–12. https://doi.org/10.3897/pharmacia.72.e163528 Abstract Background: Although prescribing errors (PEs) are a major threat to patient safety in primary care, there is a scarcity of data on their occurrence and nature in Jordanian primary care. Objectives: This study aimed to investigate the incidence, types, severity, and predictors of PEs in Jordanian primary healthcare, as well as the nature and outcomes of pharmacist interventions. Method: This was a prospective observational study conducted in 12 community pharmacies across Jordan over 12 weeks. Results: Of the 617 patients included, 102 experienced PEs, resulting in a total of 165 erroneous medication orders. The overall incidence rates of PEs and patients with errors were 14.9% and 16.5%, respectively. The most common erroneous therapeutic category was antibiotics (24.8%), followed by analgesics (14.5%). The most common type of PE was wrong drug (33.33%), followed by omission errors (25.52%) and wrong dose (17.19%). The least common type was wrong duration (10.94%). Of the 192 PEs identified, nine (4.69%) were deemed lethal, 42 (21.88%) severe, and 75 (39.06%) moderate in severity. The most common dosage form for patients with errors was tablets (55.7%). The total number of pharmacist interventions was 216, of which 66.7% were process-based and 33.3% outcome-based. Among these interventions, 75.0% were fully accepted by patients and 68.5% were fully approved by physicians. Conclusion: Overall, PEs in primary care were common and could cause severe harm. Wrong drug, omission, and wrong dose were the most frequent types of PEs. Pharmacist interventions were crucial and fell into two major categories: process-based and outcome-based. Keywords incidence, Jordan, pharmacist, prescribing errors, primary care, severity, types Copyright Abdel-Qader DH et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Pharmacia 72: 1–12 DOI 10.3897/pharmacia.72.e163528 Research Article Abdel-Qader DH et al.: Identifying and mitigating prescribing errors in outpatient clinics2 Introduction Prescribing errors (PEs) are defined as any deviation from an acceptable and safe prescription that might include errors in dosage, frequency, route of administration, or medication selection resulting in hospitalization, disability, or even death for patients (Dean et al. 2000). Even though PEs pose significant harm to patients in primary care, there has been insufficient research carried out on their nature and frequency, especially in lowand middle-income countries. This lack of data poses an unacceptable barrier for healthcare providers and policymakers seeking solutions that effectively prevent or manage PEs. Jordan lacks research relating to PEs in primary care; however, a study conducted in secondary care revealed high rates of PEs due to communication breakdown. Abdel-Qader et al. (2020a) investigated PEs in Jordanian hospitals and found a high incidence of errors with relatively concerning severity. These results imply that PEs are also likely an issue within the primary healthcare sector in Jordan, thus mandating further study in this field. Jordan’s community pharmacies face several difficulties associated with PE research. At its core is an insufficient knowledge base about their incidence and impact on patient safety; this lack of data makes identifying effective prevention or management strategies even harder for healthcare providers and policymakers (Basheti et al. 2021). Thus, the aim of this study was to investigate the incidence, types, severity, and predictors of PEs in primary care in Jordan, as well as the nature and outcomes of pharmacist interventions. Materials and methods Study design A prospective observational study was conducted in 12 community pharmacies to collect real-world information on PEs intercepted in community pharmacies. By closely observing and documenting errors in these settings, insights into their frequency, types, and pharmacist interventions were gained. Community pharmacists collected data over an agreed-upon period by carefully observing PEs based on predefined criteria and classifying them according to predefined rules. Comprehensive details regarding study objectives, data collection process, and potential benefits associated with participation were provided to research community pharmacists. At the initial contact meeting with pharmacists, the principal investigator provided induction training on the data collection and classification procedure and stressed the confidential and anonymous nature of any collected data, as well as ethical considerations, which were approved by the University of Petra’s Research Ethics Committee (O/2/8/2023). Study flow and settings A semi-structured data collection form was devised based on an extensive review of existing literature (Overhage and Lukes 1999; Abdel-Qader et al. 2020b; Dean 2000) and current pharmacy practices in Jordan. To ensure content validity, a multidisciplinary expert committee was formed, consisting of a clinical pharmacist, an internal medicine physician, and a health communication specialist. Members were selected based on their expertise in prescribing practices, patient care, and healthcare communication. This committee evaluated the form’s relevance and appropriateness in measuring the desired variables. Following this validation, a pilot phase was conducted to assess the feasibility and applicability of the data collection form in a real-life environment by employing three pharmacies. Their feedback was carefully considered during refinements to the form. After the successful pilot study, 12 community pharmacies (26 pharmacists) were recruited for participation in the full-scale data collection phase over 3 months (March to May 2023). Sample size To calculate the sample size for documenting PEs, the following formula was employed: n = (Z² × P × (1 − P))/ e². By considering the desired confidence level of 95% (Z-score of 1.96), the PE rate of 12.5% according to a previous study conducted by Abdel-Qader et al. (2020a), and the margin of error (e) as 2%, the required sample size was 1052 medications. Participants To achieve the required sample size of 1052 medications for this study, a purposive sampling methodology was employed. This involved the strategic selection of 12 pharmacies: six from the Central Region, four from the North, and two from the South. This approach was deemed most appropriate because it enabled the deliberate selection of pharmacies based on their geographic region, which corresponded to areas defined by population connectivity and distance, and their anticipated prescription volume. Purposive sampling ensured that representative data could be gathered from across Jordan’s diverse primary healthcare landscape, while also accommodating the practical limitations and logistical constraints inherent in conducting research across a geographically dispersed area. Pharmacies were included in the study based on several key criteria. Firstly, to ensure a sufficient quantity of medication data for analysis, participating pharmacies needed to receive a minimum of 10 prescriptions per day, on average. Secondly, pharmacies were required to employ more than one pharmacist per shift. Another requirement was that the participating pharmacies should be located in one of three geographical regions (North, Central, and South). Furthermore, selected pharmacies needed to maintain readily accessible and reasonably complete prescription Pharmacia 72: 1–12 3 records, whether in electronic or paper-based format, to facilitate efficient data collection. Finally, the willingness of pharmacy management and staff to participate in the study and grant access to prescription data was a crucial inclusion criterion, and the pharmacies also had to have operated within a region for a minimum of 6 months. Pharmacies were first screened for eligibility based on these criteria. The researcher then contacted pharmacy owners or managers to explain the study’s objectives, data collection process, and ethical considerations, including approval from the University of Petra’s Research Ethics Committee. Those who expressed interest were invited to a follow-up meeting to discuss study expectations in detail. Written informed consent was obtained before participation. During data collection, research pharmacists closely observed all prescriptions before documenting any intercepted PEs using the data collection form. Regular communication with participating pharmacies was maintained through site visits and a structured reporting system, ensuring compliance with study protocols and addressing any concerns. Data collection and operational definitions The data collection form encompassed multiple fields related to prescription processes and patient characteristics. These fields included date of data collection, type of pharmacy (chain or independent), number of staff per shift, degree of pharmacist in charge, years of experience, and highest degree obtained. In addition, the total number of patients per shift who received medications from each type of pharmacy (chain vs. independent) was recorded. The form also contained fields related to PEs, including whether an error occurred (yes/no) and the type of error, such as wrong drug, dose, frequency, duration, or omission error. Error severity (minor, moderate, severe, or lethal) was recorded accordingly. Patient demographics, such as gender, age, and comorbidities, were also recorded. These demographics were obtained from the prescription when available; otherwise, pharmacists collected the information directly from patients. Additionally, the form was designed to record whether the research pharmacist made any recommendations or interventions (yes/no), including process-based interventions (before the error reached the patient) and outcome-based ones (post-error interventions). Finally, the form included fields related to patient outcomes that addressed patient acceptance of interventions (full/partial), physician acceptance (yes/no), clinical outcomes improvement (yes/no), hospitalization (yes/ no), and death (yes/no). Pharmacists were instructed to follow up with patients for 1 week after the prescription was filled. This follow-up was necessary to evaluate the impact of pharmacist interventions on patient clinical outcomes. Follow-up involved direct communication with patients or their caregivers to assess symptom relief, chronic illness management, and achievement of specific health goals. Hospitalization data were collected by asking patients or caregivers whether hospital admission was required during the follow-up period, while mortality data were obtained from relatives or caregivers. Patient participation in follow-up was voluntary, and informed consent was obtained prior to data collection. In Jordan, pharmacist follow-up is not a standard part of routine care but was conducted in this study to assess the effectiveness of pharmacist interventions. Data collection forms, including PE types and severity, were validated by the multidisciplinary expert committee to ensure correctness, completeness, and relevance. Prescribing errors were identified based on predefined criteria, and any disagreements were resolved through discussion within the committee. The types and severity of PEs adopted in this study are listed in Tables 1 and 2, respectively. The classification was adapted from the study by Dean et al. (2000), focusing on the most clinically relevant errors in primary care settings while maintaining alignment with established literature. Table 1. Prescribing errors classification (Dean et al. 2001; Abdel-Qader et al. 2010). Wrong drug 1Prescribing a drug for a patient for whom, as a result of a co-existing clinical condition, that drug is contraindicated. 2Prescription of a drug to which the patient has a documented clinically significant allergy. 3Prescribing a drug for which there is no indication for that patient. 4Duplication Wrong dose 1When the dose of a medicine recommended to be taken at a particular time is incorrect (overdosage or underdosage). 2Prescribing a drug in a dose that, according to British National Formulary or data sheet recommendations, is inappropriate for the patient’s renal function. Wrong frequency When the prescribed frequency of medicine is different from current evidence-based treatment guidelines Wrong dosage form When a dosage form not intended by the prescriber is written in the prescription, or when a dosage form is not available for the prescribed drug. Omission Omissions: Missing elements that will require further information. Major omissions would require the pharmacist to contact the prescriber. Minor omissions may be filled by the pharmacist based on professional judgment and additional information gathered from the patient or prescriber. Pharmaceutical issues Prescribing a drug to be given by intravenous infusion in a diluent that is incompatible with the drug prescribed Drugs interactions. Abdel-Qader DH et al.: Identifying and mitigating prescribing errors in outpatient clinics4 Data analysis The data analysis was conducted with SPSS version 26. Descriptive statistics were employed to summarize and characterize the collected data, using measures such as frequencies, percentages, means, and standard deviations to account for relevant variables. To compare differences in categorical variables, the chi-square test was used. The t-test and ANOVA were used to compare means of continuous variables. A multivariable logistic regression model was used to assess the predictors of specific outcomes related to the study. All statistical analyses were performed at a predetermined level of significance (p < 0.05) to establish statistical significance of findings and were presented using appropriate tables, graphs, and measures that facilitate clear reporting. Participating pharmacists received training on the data collection process to ensure completeness of the forms for all recruited patients; consequently, there were no missing data for the primary variables analyzed in this study. Results The total number of community pharmacies included in the study was 12, of which seven (58.3%) were chain pharmacies, eight (61.6%) had two pharmacists per shift, and half had 6–30 customers per shift (Table 3). In addition, three (25.0%) pharmacies were receiving more than 30 prescriptions and dispensing more than 50 medications per shift. The total number of pharmacists working in those pharmacies was 26, of which 21 (80.8%) had bachelor’s degrees and 11 (42.3%) had 5–10 years of experience. These characteristics provide context on factors influencing PEs and pharmacist interventions, ensuring alignment with established medication safety research. Table 2. The severity of prescribing errors (Overhage and Lukes 1999; Abdel-Qader et al. 2010). Lethal error 1High potential for life-threatening adverse effects/reactions. 2The potentially lifesaving drug at a dosage too low for the disease being treated. 3High dosage (>10 times normal) of a drug with a low therapeutic index. Severe error 1Route of administration could lead to severe toxicity 2A low dosage of a drug for a serious disease in the patient with acute distress 3High dosage (4–10 times normal) of a drug with a low therapeutic index 4Dosage resulted in serum drug concentration in the potentially toxic range 5The drug could exacerbate the patient’s condition (related to warnings or contraindications) 6Misspelling or mix-up in medication order could lead to dispensing of the wrong drug 7Documented allergy to a drug 8High dosage (>10 times normal) of a drug without a low therapeutic index Moderate error 1High dosage (1.5–4 times normal) of a drug with a low therapeutic index 2Drug dosage too low for patient’s condition 3High dosage (1.5–10 times normal) of a drug without a low therapeutic index 4Errant dual-drug therapy for a single condition 5Inappropriate dosage interval 6Omission from the medication order Minor error 1Incomplete information in the medication order 2The unavailable or inappropriate dosage form 3Non-formulary drug 4Noncompliance with standard formulations and hospital policies No error 1Information or clarification requested by the physician or other healthcare professional from the pharmacist 2Cost savings only Table 3. Characteristics of pharmacies (N = 12) and pharmacists (N = 26) included in the study. Parameter Total, n (%) Pharmacies Type of pharmacy Independent 5 (41.6%) Chain 7 (58.3%) Number of pharmacists per shift 2 8 (61.6%) 3 2 (23.0%) 4 1 (15.4%) Years of opening 1–3 4 (33.3%) 4–6 5 (41.6%) >6 3 (25.0%) Number of customers per shift 5–15 2 (16.7%) 16–30 6 (50.0%) >30 4 (33.3%) Number of prescriptions per shift 5–15 2 (16.7%) 16–30 7 (58.3%) >30 3 (25.0%) Number of medications dispensed per shift <30 2 (16.7%) 30–50 7 (58.3%) >50 3 (25.0%) Pharmacists The highest degree of pharmacist in charge BSc 21 (80.8%) MSc/PharmD 5 (19.2%) PhD 0 (0.0%) Years of experience 1–5 10 (38.5%) 6–10 11 (42.3%) >10 5 (19.2%) Pharmacia 72: 1–12 5 In total, 617 patients were included in this study, of which 275 (44.6%) were aged between 25 and 35 years, and 27 (4.4%) were aged above 65 years (Table 4). Most of the patients were male, 431 (69.9%). In terms of comorbidities, 119 (19.2%) suffered from low back pain, 76 (12.3%) had diabetes, 65 (10.5%) had hypertension, and 48 (7.8%) had ischemic heart disease. The most common reasons for visiting the pharmacy were requesting an OTC medicine (452, 73.3%) and filling or refilling a prescription (314, 50.9%). Table 5 shows examples of PEs, their types, severity, and corresponding interventions conducted by pharmacists. Of the 617 patients included in the study, 102 had PEs, with 165 erroneous medication orders. The total number of PEs was 192, including 49 omission errors. The incidence rates of PEs, patients with errors, and omission errors were 14.9%, 16.5%, and 4.0%, respectively (Table 6). As shown in Fig. 1, 33 (32.4%) of the patients with errors had five medication orders in their prescriptions, six (5.9%) had six medication orders, and five (4.9%) had one medication order. Of the 192 PEs, the most common type was wrong drug (33.33%), followed by omission errors (25.52%), wrong dose (17.19%), and wrong frequency (13.02%). The least common type was wrong duration (10.94%) (Table 7). Of the 192 PEs identified, nine (4.69%) were deemed lethal, 42 (21.88%) severe, and 75 (39.06%) moderate. Minor errors accounted for 34.38% of the total PEs. There was no statistically significant difference in the severity of PEs across different medication classes (p > 0.05). The most common medications associated with severe errors were analgesics (19/42, 45.24%) and antibiotics (14/42, 33.33%). Of the nine lethal errors, three (33.33%) were related to antidiabetics, four (44.44%) to antihypertensives, and two (22.22%) to antivirals (Table 8). There was no significant difference in the types of medications across erroneous and non-erroneous medication orders (p = 0.091) (Table 9). Overall, analgesics (21.4%), antibiotics (14.4%), and antidiabetics (13.8%) were the most commonly prescribed medications for patients with errors. The most common erroneous therapeutic category was antibiotics (24.8%), followed by analgesics (14.5%). The dosage forms of the medications were similar across erroneous and non-erroneous medication orders (p = 0.106). The most common dosage forms for patients with errors were tablets (55.7%), capsules (14.1%), and liquids (7.9%). The total number of pharmacist interventions was 216, which included 144 (66.7%) that were process-based and 72 (33.3%) that were outcome-based. These interventions were organized into predefined categories to differentiate between those that specifically addressed PEs and those that provided general clinical recommendations. Among these interventions, 162 (75.0%) were fully accepted by patients, and 148 (68.5%) received full approval from physicians. Full acceptance by physicians was defined as the complete execution of the pharmacist’s recommendation, whereas partial acceptance indicated alterations to the recommendation that were not carried out in their entirety. Among the process-based pharmacist interventions, substitution of a prescribed drug (36, 25.0%), removal of a drug from the prescription (22, 15.3%), and adjusting the dose of a prescribed drug (14, 9.7%) were the most common. The least common process-based pharmacist interventions were advising the patient to stop smoking (eight, 5.6%) and advising the patient to exercise (six, 4.2%). Although these lifestyle suggestions were not directly related to PEs, they were incorporated as general advice for patient care. Adding a drug to the prescription (AOR: 0.7; 95% CI: 0.5–0.9) was considered less likely to enhance clinical outcomes in comparison to adjusting the dose of an existing medication (Table 10). Among the outcome-based interventions, initiation of a new drug (23, 31.9%), adjusting the dose of a dispensed drug (17, 23.6%), and cessation of drug therapy (16, 22.2%) were the most common. Referral to a physician (4, 2.8%) was the least common intervention. Interventions related to the initiation of a new drug (AOR: 0.4; 95% CI: 0.2–0.7) were less likely to yield improvements in clinical outcomes when compared to the discontinuation of drug therapy. As shown in Fig. 2, the significant predictors of PE occurrence were age above 65 years (AOR: 1.8; 95% CI: 1.4–2.9) and comorbidities (acute respiratory infections) (AOR: 2.8; 95% CI: 1.7–4.3). Table 4. Characteristics of patients included in the study (N = 617). Parameter Total, n (%) Age (years) 16–24 153 (24.8%) 25–35 275 (44.6%) 36–65 162 (26.3%) >65 27 (4.4%) Gender Female 186 (30.1%) Male 431 (69.9%) Comorbidities* Ischemic heart disease 48 (7.8%) Diabetes mellitus 76 (12.3%) Hypertension 65 (10.5%) COPD 26 (4.2%) Asthma 49 (7.9%) Osteoarthritis 57 (9.2%) Depression 28 (4.5%) Low back pain 119 (19.2%) Cancer 5 (0.8%) ARI 49 (7.9%) Other 27 (4.4%) Reason for pharmacy visit* Prescription fill/refill 314 (50.9%) Request an OTC medication 452 (73.3%) Request health and wellness products 115 (18.6%) Health advice 84 (13.6%) Vaccination 56 (9.1%) Receiving an injection 49 (7.9%) IHD: ischemic heart disease; COPD: chronic obstructive pulmonary disease; ARI: acute respiratory infection. Comorbidities were documented through patient acknowledgment or by reviewing the medications requested. Health and wellness products included vitamins, supplements, skincare products, and personal care items. OTC: over the counter. *Multiple options could be selected for one participant. Abdel-Qader DH et al.: Identifying and mitigating prescribing errors in outpatient clinics6 Table 6. Calculation of cumulative incidence of prescribing errors. Code Variable Equation Value A Total number of patients during the study – 617 B Total number of medication orders (based on prescriptions) – 1238 C Total number of omission errors – 49 D Total number of patients with errors – 102 E One prescribing error – 146 F Two prescribing errors – 11 G Three prescribing errors – 8 H Total number of erroneous medication orders intercepted by pharmacists E+F+G 165 I Total number of prescribing errors E+2F+3G 192 J Cumulative incidence of patients with error (D/A) × 100% 16.5% K Cumulative incidence with prescribing errors I/(B+C) × 100% 14.9% L Cumulative incidence of erroneous orders H/(B+C) × 100% 12.8% M Cumulative incidence of omission errors (C/B) × 100% 4.0% N Cumulative incidence of prescribing errors without omission errors {(I – C)/B} × 100% 11.6% O Cumulative incidence of prescribing errors vs. opportunities for errors I/4(B+C) × 100% 3.7% Table 5. Clinical examples of prescribing errors. Clinical scenario Type of PEs Severity of PEs Pharmacist intervention A 25-year-old patient with a known allergy to penicillin was prescribed a ceftriaxone injection. Wrong drug Lethal Upon reviewing the patient’s medical history and identifying the allergy, the pharmacist promptly contacted the prescriber to highlight the risk associated with the prescribed medication. The pharmacist emphasized the patient’s allergy to penicillin and the potential for a severe allergic reaction to ceftriaxone. Subsequently, the pharmacist recommended an alternative antibiotic, azithromycin, which is not a beta-lactam antibiotic and thus poses no cross-reactivity for patients with penicillin allergies. The prescriber agreed with the recommendation, and the medication was changed to azithromycin. A 43-year-old patient refilled diclofenac sodium tablets at a dosage of 75 mg to be taken three times per day. Wrong frequency Moderate The pharmacist, upon noticing the unusually high frequency of administration, contacted the prescribing physician to discuss the potential risks associated with such a regimen. The pharmacist suggested modifying the prescription to a more conventional frequency, recommending that diclofenac be taken twice a day instead of three times. This change aimed to reduce the risk of adverse effects while maintaining therapeutic efficacy. The physician was informed about the standard dosing guidelines and the potential risks of gastrointestinal, renal, and cardiovascular side effects associated with higher dosing frequencies of NSAIDs like diclofenac. A 68-year-old patient suffering from a headache was prescribed paracetamol with no dose in the prescription. Omission Minor The pharmacist instructed the patient to take one tablet (500 mg) twice daily as needed, with a maximum of 4 grams per day. A 21-year-old patient with a history of asthma and a recent asthmatic attack was prescribed propranolol. Wrong drug Severe Contacted the physician to inform her about the issue and suggested a selective beta blocker, with monitoring of the patient. A 50-year-old patient with a history of severe renal impairment was prescribed metformin. Wrong drug Severe Did not dispense and contacted the physician. Discussion Prescribing errors (PEs) are an increasingly serious patient safety threat in both primary and secondary healthcare settings. Nonetheless, while PEs in hospitals have received greater consideration than those in primary care, PEs in primary care – where millions of prescriptions are filled annually – remain poorly understood. In addition, the role of pharmacists in preventing such errors has rarely been addressed; thus, this study sought to examine PE incidence, types, severity, predictors, and interventions performed by community pharmacists to reduce them across Jordan. To achieve this aim, a direct observational approach was utilized in 12 community pharmacies in Jordan. Specifically, an extensive literature review and input from a Pharmacia 72: 1–12 7 panel of experts were used for the development of the data collection form, which was rigorously validated. This research approach included two notable strengths. Firstly, the use of operational definitions that describe the characteristics of severity and type of PEs. This classification system was adopted from previous studies (Overhage and Lukes 1999; Abdel-Qader et al. 2010). A severity scale based on Overhage and Lukes (1999) was applied. Equal clarity and consistency were provided by operational definitions in terms of measurement and classification of variables, so that readers understood and interpreted them similarly. Secondly, the incidence rates of PEs were calculated by means of specific equations adapted from the study by Abdel-Qader et al. (2010). Using these equations ensured consistency, validity, and transparency in the calculations, which enabled comparisons and results that were meaningful and reliable. Overall, this research addressed the gap in knowledge by providing new insight into the incidence, types, and severity of PEs in primary care in Jordan and the role of pharmacists in preventing errors. Additionally, its outcomes could be used for improving patient safety as well as developing pharmacy practice and health policies in other settings worldwide. In this study, more than half of the pharmacies were chain pharmacies. Standardized procedures and centralized systems in chain pharmacies may foster more consistent error-prevention strategies compared to independent pharmacies (Miller and Goodman 2017). Regarding staffing, the majority of pharmacies (61.6%) had two pharmacists per shift. This staffing model, a criterion for inclusion, was intended to provide a reasonable pharmacist-to-patient ratio, ensuring adequate time for patient counseling and the detection of potential errors. In our study, PE incidence was 14.9%, which was higher than the reported incidence of PEs (12.5%) in the emergency department of the largest public hospital in Jordan (Abdel-Qader et al. 2020a). A number of factors might have contributed to this higher PE incidence in community pharmacies. First, no electronic health record (EHR) system is in place to provide continuity between physicians in primary care and community pharmacies. In a systematic review, Ammenwerth et al. (2008) reported that EHR can reduce the risk of medication error by up to 98%. An integrated EHR with electronic prescribing can eliminate errors from illegible handwriting and provide pharmacists with instant access to patient medication history, allergies, and potential drug interactions, thereby strengthening clinical decision-making and collaboration between prescribers and pharmacists (Gandhi et al. 2005). Second, the lack of regulatory oversight and auditing specifically focused on physicians contributes to the challenge. Studies have shown that hospital settings generally have more structured mechanisms for detecting and addressing PEs due to dedicated quality improvement teams and established reporting systems (Vira et al. 2006). In contrast, primary care settings, which consist of multiple independent physicians or small clinics, face difficulties in implementing standardized protocols and maintaining consistent oversight. The hierarchical structure of hospitals allows for systematic monitoring and corrective actions, whereas primary care practices often have limited personnel and resources specifically allocated for monitoring PEs (Gandhi et al. 2005). Table 7. Types of prescribing errors identified (N = 192). Type of PEs Frequency (%) Wrong drug 64 (33.33%) Omission errors 49 (25.52%) Wrong dose 33 (17.19%) Wrong frequency 25 (13.02%) Wrong duration 21 (10.94%) Table 8. Severity of prescribing errors (N = 192). Severity of PEs Frequency (%) Lethal 9 (4.69%) Minor 66 (34.38%) Severe 42 (21.88%) Moderate 75 (39.06%) 33 26 18 14 6 5 0 5 10 15 20 25 30 35 01234567 Number of Patients with Errors Number of Medication Orders Figure 1. Number of patients with errors vs. number of medication orders. Abdel-Qader DH et al.: Identifying and mitigating prescribing errors in outpatient clinics8 Table 10. Types of pharmacist interventions and predictors of clinical outcomes. Parameter Total, n (%) Improvement in clinical outcomes Worsening in clinical outcomes Predicting improvement in clinical outcomes, AOR (95% CI) Process-based pharmacist interventions Adjusting the dose of a prescribed drug (Ref) 14 (9.7%) 9 (10.11%) 5 (9.1%) 1.00 Substitution of a prescribed drug 36 (25%) 23 (25.8%) 13 (23.6%) 1.1 (0.5–1.9) Adjusting the duration of a prescribed drug 11 (7.6%) 7 (7.9%) 4 (7.3%) 1.5 (0.7–3.4) Adding a drug to the prescription 16 (11.1%) 9 (10.11%) 7 (12.7%) 0.7 (0.5–0.9) Removing a drug from the prescription 22 (15.3%) 14 (15.7%) 8 (14.5%) 2.5 (1.4–5.7) Psychological support 9 (6.2%) 5 (5.6%) 4 (7.3%) 0.8 (0.5–1.3) Patient education about medication use 15 (10.4%) 10 (11.2%) 5 (9.1%) 1.8 (0.7–5.1) Advising the patient to eat healthy food 7 (4.9%) 6 (6.7%) 1 (1.8%) 1.4 (0.7–2.4) Advising the patient to stop smoking 8 (5.6%) 3 (3.3%) 5 (9.1%) 1.2 (0.4–1.8) Advising the patient to do exercise 6 (4.2%) 3 (3.4%) 3 (5.5%) 0.6 (0.3–4.1) Outcome-based interventions Cessation of drug therapy (Ref) 16 (22.2%) 10 (22.2%) 6 (22.2%) 1.00 Initiation of a new drug 23 (31.9%) 12 (26.7%) 11 (40.7%) 0.4 (0.2–0.7) Adjusting the dose of a dispensed drug 17 (23.6%) 11 (24.4%) 6 (22.2%) 3.1 (0.8–7.6) Recommending a laboratory test 13 (18.1%) 10 (22.2%) 3 (11.1%) 1.9 (0.8–4.5) Referral to physician 3 (4.2%) 2 (4.4%) 1 (3.7%) 0.9 (0.7–1.7) Ref: reference; CI: confidence interval; AOR: adjusted odds ratio. Bold adjusted odds ratio indicates significant findings. Parameters are described as proportions [n (%)] unless stated otherwise. Table 9. Characteristics of medication orders (N = 1238). Types of medications Total, n (%) Erroneous (N = 165), n (%) P-value Analgesics 265 (21.4%) 24 (14.5%) 0.091 Antibiotics 178 (14.4%) 41 (24.8%) Antidiabetics 171 (13.8%) 14 (8.5%) Antihypertensives 156 (12.6%) 12 (7.3%) Anti-inflammatory 83 (6.7%) 9 (5.5%) Antihistamines 68 (5.5%) 9 (5.5%) H2 Blockers 51 (4.1%) 6 (3.6%) Antihyperlipidemics 48 (3.9%) 8 (4.8%) Antitussives 48 (3.9%) 7 (4.2%) Antifungals 36 (2.9%) 2 (1.2%) Antispasmodics 31 (2.5%) 4 (2.4%) Antivirals 27 (2.2%) 6 (3.6%) Antidiarrheals 22 (1.8%) 3 (1.8%) Vitamins 17 (1.4%) 8 (4.8%) Antigout agents 11 (0.9%) 5 (3.0%) Antihyperuricemics 9 (0.7%) 2 (1.2%) Antimalarials 6 (0.5%) 3 (1.8%) Other 11 (0.9%) 2 (1.2%) Dosage Forms 0.106 Tablets 689 (55.7%) 55 (33.3%) Capsules 174 (14.1%) 36 (21.8%) Liquids 98 (7.9%) 21 (12.7%) Topicals 59 (4.8%) 13 (7.9%) Suppositories 66 (5.3%) 10 (6.1%) Inhalers 54 (4.4%) 11 (6.7%) Injectables 63 (5.1%) 13 (7.9%) Lozenges 35 (2.8%) 6 (3.6%) The role of PEs in primary care has been studied in several research settings. In Saudi Arabia, the incidence of PEs in primary care was higher (18.7%) than the rate in our study (Khoja et al. 2011). However, in the UK, a lower incidence of PEs (5%) was reported among primary care providers (Avery 2013). Moreover, PEs were also identified in Pharmacia 72: 1–12 9 primary care in Australia (Koper et al. 2013). The different approaches used to calculate the incidence of PEs, variation in healthcare practices, medication management systems, reporting mechanisms, and cultural factors affecting patient safety contributed to these diverse reports across countries. In our research, we found that the most frequent PEs were wrong drug and omission errors. This coincides with the results of other research studies in different countries, including Jordan (Abdel-Qader et al. 2020a), the UK (Abdel-Qader et al. 2010; Avery 2013; Ashcroft et al. 2015), and Ethiopia (Simegn et al. 2022). In Jordan, wrong drug and omission errors were the most common PEs in the emergency department (Abdel-Qader et al. 2020a). In the UK, omission errors were frequently identified in primary care (Avery 2013). Simegn et al. (2022) found that omission and wrong drug errors were the most frequently reported by community pharmacists. In comparison, a prospective study by Stasiak et al. (2014) conducted in the emergency department reported that wrong dose was the most common type of PE. The occurrence of wrong drug and omission errors at high rates could be attributed to several factors. First, insufficient information is often available for accurate medication prescribing. Physicians in primary care may not have access to relevant patient information, such as medical history, allergies, or concurrent medications. Furthermore, differences in guidelines and professional opinions regarding what medication should be prescribed can increase these types of errors. Physicians may encounter confusion due to multiple guidelines regarding a given condition or medication. They may also have different interpretations or preferences when prescribing certain drugs, leading to inconsistent practices and potentially resulting in PEs. In our study, the most common erroneous therapeutic categories were antibiotics and analgesics. Our findings aligned with the Ethiopian study (Simegn et al. 2022). However, in the UK, anticoagulants, opioid analgesics, and insulin were the most frequent erroneous medications in primary care (Cousins et al. 2019). These findings do not necessarily mean that these medicines are naturally prone to errors but rather reflect the prescribing patterns within each respective location and healthcare setting. Most of the PEs in our study were either moderate or minor in severity. Nonetheless, about one-fifth of the PEs reported were severe. It was not possible to compare our results with other studies because they used different methodologies to measure severity. For example, a UK study used an index or scale rating method for PE severity evaluation; severity grading showed that 41.1% of PEs were minor, 51.6% significant, and the remaining 7.3% serious or potentially life-threatening (Koper et al. 2013). Overall, PEs pose a serious threat to patient safety in both primary and secondary care. This study expanded the notion that patient characteristics could be important predictors of PEs. Elderly patients (age >65 years; AOR: 1.8; 95% CI: 1.4–2.9) and those with acute respiratory infections (ARI) (AOR: 2.8; 95% CI: 1.7– 4.3) were at higher risk of experiencing PEs. Elderly patients usually have multiple chronic conditions that can be aggravated by polypharmacy. Several studies (Abdel-Qader et al. 2020c; Kua et al. 2021; Vinks et al. 2006) found that polypharmacy increases the risk of PEs as well. The association between ARIs and PE risk may be explained by several factors. A possible factor is the insufficient up-to-date information that physicians possessed concerning respiratory infections. Medical knowledge and guidelines for the evaluation and management of ARIs are evolving continuously. Therefore, physicians who were not up to date on these guidelines could unknowingly prescribe treatment based on outdated information. Moreover, the indiscriminate use of antibiotics for ARIs without culture or laboratory testing is a common practice in Jordan. Some patients believe that antibiotics should be prescribed for every type of respiratory symptom without laboratory confirmation, resulting in a high rate of unnecessary treatment and addiFigure 2. Predictors of prescribing error occurrence.