Introduction
Medication errors (MEs) at transitions of care are a well-documented source of avoidable patient harm and healthcare resource burden. In response, the World Health Organization (WHO) launched the ‘Medication Without Harm’ initiative in 2022, identifying transitions of care as a priority for improving global patient safety.1 Evidence suggests that 25–80% of patients experience at least one medication or communication discrepancy at discharge.2 If undetected, these discrepancies may persist post-discharge, contributing to adverse drug events (ADEs), hospital readmissions and increased morbidity.34
Medicines reconciliation (MedRec) is a structured evidence-based intervention designed to identify and resolve medication discrepancies while ensuring accurate communication of medical information across care interfaces.56 Pharmacist-led MedRec at hospital discharge has been shown to reduce discrepancies, improve prescription accuracy and reduce ADE-related readmission rates.7–9 While plausible, its effect on mortality remains inconclusive.9
These errors contribute to Medication Related Harm (MRH) and also impose a significant financial burden on health systems. The WHO estimates the global cost of medication errors to be approximately US$42 billion annually.2 Remedial healthcare utilisation following MRH is also considerable. Hospital readmissions related to MRH are associated with increased inpatient costs, elevated risk of healthcare-associated infections and reduced patient satisfaction.10 In the UK the cost to the National Health Service (NHS) is estimated at £396 million (€464 million) annually, of which £243 million (€285 million) is considered preventable.11
In Ireland, transitions of care are a key focus of national patient safety initiatives. Both the Health Service Executive (HSE) Patient Safety Strategy 2019–2024 and the revised Health Information and Quality Authority inspection framework have highlighted transitions of care as key areas for reducing MRH.1213 However, the absence of a national electronic health record system continues to create significant communication gaps across healthcare settings, especially at transitions of care between hospitals, general practitioners (GPs) and community pharmacists.1415 Discharge prescriptions are typically transcribed from handwritten inpatient medication charts onto electronic discharge systems, often by junior doctors, further increasing the risk of error.1416 At discharge, patients are provided with a printed copy of their discharge prescription to take to their community pharmacy for dispensing, and the discharge summary and prescription are transferred electronically to the patient’s GP from the hospital. The discharge prescription often has a specific section to allow the prescriber to detail any medication changes made during the patient’s admission.
Hospital pharmacists are well positioned to support safer transitions of care. The European Association of Hospital Pharmacists advocates for pharmacist involvement at all care transitions.17 Studies indicate that pharmacist-led discharge services improve MedRec and patient education and reduce avoidable readmissions, contributing to more efficient use of healthcare resources.89 Nonetheless, resource constraints and infrastructure gaps continue to hinder widespread implementation of pharmacist-led MedRec at discharge in Irish hospitals.161819 It is necessary to pilot and evaluate the impact of such a service to inform wider implementation going forward.
Aim
This pilot study aimed to investigate the potential clinical and financial benefit of a pharmacist discharge service on a surgical ward in an Irish hospital setting.
Methods
This was a single-centre single-arm prospective pilot study evaluating the potential clinical and financial impact of a pharmacist-led discharge service on a surgical ward in a 643-bed university teaching hospital. The pilot study was carried out over an 8-week period (February to May 2024) on a 31-bed surgical ward for adult patients (≥18 years old). As a pilot study, no formal sample size calculation was performed.
Based on an average monthly discharge rate, it was estimated that 50–70 discharge prescriptions would be reviewed. This estimate considered staffing levels and the ward pharmacist’s availability to avoid delaying discharges (figure 1).
Patients were included if they were aged ≥18 years, on ≥3 medications (any form including oral, transdermal, topical, inhaled, parenteral, regular or as required medications), had an admission MedRec completed by a clinical pharmacist, were discharged during pharmacy hours (Monday to Friday, 8.30 to 16.30 hours) and provided consent to participate. To minimise the risk of reactive bias (Hawthorne effect),20 prescribers were not informed about the details of the study, the data collection process or the specific data collection period. The pharmacist liaised with the ward nurse manager or ward clerk regarding patient discharges.
The pharmacist invited patients who had discharge prescriptions available for review during pharmacy working hours to take part. Patients, or their next-of-kin where relevant, provided written informed consent and the pharmacist reviewed the patient’s draft electronic prescription and discharge summary letter in the electronic discharge summary system. Prescribers were contacted by bleep, telephone or in person to discuss any errors or discrepancies.
If the prescriber accepted the pharmacist’s recommendations, they made the necessary changes to the prescription and/or discharge summary before the final validation and printing by the prescriber. Any prescription discrepancies (prescribing or communication errors) identified by the pharmacist during the review were recorded on the data collection sheet. Prescribing errors were defined as any unintentional discrepancy between the patient’s discharge prescription and the inpatient medication chart, taking into account changes made during their admission.21 Communication errors were defined as any unintentional inaccuracies in the ‘medication changes’ section of the discharge prescription which could potentially affect the patient’s ongoing care in primary care (changes in drug, dose, frequency, form, or if any medication started or discontinued).21 Details recorded included: type and details of error(s) or discrepancy(ies); medication classes involved; actions taken by the pharmacist; pharmacist time taken to complete the discharge review; and additional pharmacist time taken to resolve any identified issues.
The duration of the pharmacist review and any additional time taken to resolve discrepancies with prescribers was recorded in real-time using a stopwatch by the study pharmacist.
Clinical impact assessment
A four-membered expert review panel was recruited through professional networks to evaluate the prescribing errors identified. The panel included two Medical Specialist Registrars (specialising in General Medicine and Clinical Pharmacology) and two Chief II Medication Safety Pharmacists.
Each expert panel member received a review pack containing instructions, anonymised clinical case vignettes and a standardised Excel worksheet. An anonymised case vignette was prepared for each patient in the study, summarising their demographics, medical history, discharge medications and a description of the discharge medication or communication error(s) identified. These vignettes served as the basis for the expert panel’s evaluation.
The instructions guided the panel on the use of the specific scales to assess the severity and likelihood of an ADE occurring if the pharmacist discharge medication reconciliation had not occurred, as well as any potential post-discharge healthcare interventions required.
Expert panel members used the validated visual analogue scale (VAS) developed by Dean and Barber22 to score the potential clinical significance of the prescription errors on a scale of 0–10, where 0 indicated no harm and 10 indicated a risk of patient death.22 The expert panel estimated the probability of an ADE occurring, in the absence of the pharmacist interventions described, using the scale described by Nesbit et al.23 Expert panel members assessed any likely post-discharge healthcare resource utilisation which may be required to manage potential harm. Options included: none; GP review; GP review plus prescription; referral to specialist/outpatient department; emergency department attendance; hospital admission; ICU admission; or a combination of these options.3 All panel members scored the pharmacist interventions independently, and the ratings were submitted on the Microsoft Excel worksheet and returned via secure email for analysis.
The expert panel assessments were collated in Microsoft Excel and subjected to descriptive analysis. The mean VAS severity scores for each error were calculated and then categorised into the following severity ratings22:
Financial impact assessment
Labour costs
The labour costs of the pharmacist discharge service were calculated by determining the average time taken to review a discharge prescription and multiplying this by the hourly rate of employing a staff grade hospital pharmacist. As per the HSE consolidated pay scales at the time of writing (1 August 2025), the midpoint annual salary for a staff grade pharmacist is €63 026 or €34.63 per hour based on a 35-hour working week.24
Adverse drug event (ADE) costs
The financial impact of a pharmacist discharge service was measured using cost avoidance, net cost benefit and cost–benefit ratio (CBR). These costs were calculated for this pilot study and extrapolated to give an annual figure.
The probability of ADEs occurring in the absence of a pharmacist intervention was rated by the expert panel, and these ratings were assigned a probability score: high (0.6), medium (0.4), low (0.1), very low (0.01), and no harm expected (0).23 The cost avoidance was calculated by multiplying the median probability of the ADE by the cost of the ADE, as described by Gallagher et al,25 according to the following equation:
Cost avoidance=median probability of ADE × cost of ADE
The study published by Gallagher et al25 reported the cost of a potential ADE as €1057.25 This value was adjusted for inflation using the Central Statistics Office calculator and a value of €1313 was used for the ADE costs in this study.26
Financial impact of pharmacist discharge service
To determine whether a pharmacist discharge service would be financially beneficial to the hospital, a cost analysis was carried out. This methodology was described by Gallagher et al25 and compared the cost of providing the service with the cost avoidance generated from preventing ADEs.
The study is reported in accordance with the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) 2022 statement (see online supplemental information).
Results
In total, 51 patients from the ward were invited to participate in the study. One patient did not wish to take part due to potential delays in discharge. The discharge prescriptions of the remaining 50 patients were reviewed and included in the study, including a total of 646 prescribed medicines (Figure 1).
The patient demographics are summarised in table 1.
Of the 50 discharge prescriptions reviewed, 10 required no pharmacist intervention. A total of 184 prescription discrepancies were identified on the remaining 40 prescriptions, consisting of 126 (68.5%) prescribing errors and 58 (31.5%) communication errors. The most common type of prescribing error identified was omission of new and pre-admission medications (n=54, 29.3%), followed by dosing errors (n=30, 16.3%).
The types of prescribing errors identified are summarised in table 2.
The medications identified in the prescription errors were classified according to the British National Formulary chapters,27 with cardiovascular (n=69, 37.5%) and endocrinology (n=23, 12.5%) medications the most commonly implicated drug classes, followed closely by medicines used in gastroenterology (n=21, 11.4%) and the nervous system (n=21, 11.4%).
A total of 58 communication errors were identified on the discharge prescriptions reviewed. The most common issues identified were concerning medication changes (n=15, 25.9%), dose changes (n=10, 17.2%) and discontinued medications (n=10, 17.2%).
The total additional time taken for prescribers to adjust the 40 prescriptions was 517 min, or a median of 11 min per prescription. Resolving prescription discrepancies took 2–37 min, with length of time largely dependent on prescriber availability and/or prescription complexity. Of the 184 pharmacist interventions, a total of 168 (91.3%) were accepted by the prescriber and the prescriptions were adjusted to reflect this before the patient was discharged.
A total of 16 pharmacist interventions (n=6) were not accepted by the prescriber due to perceived delay in patient discharge. Any discrepancies not amended by the prescriber were followed up by the research pharmacist with the patient’s community pharmacist and/or GP to prevent any patient harm.
Clinical impact assessment
Each panel member (n=4) was provided with a clinical vignette for each of the 40 prescriptions requiring a pharmacist intervention. The majority of the discrepancies identified (n=156, 84.8%) were rated to have the potential to cause moderate patient harm (VAS score 3–7), with 24 rated to have the potential to cause minor harm (VAS score <3). Four of the errors (2.2%) were found to have the potential to cause severe patient harm (VAS score >7), with details summarised in table 3.
Probability of ADE
Each discrepancy (n=184) was assessed for the probability of causing an ADE. Using the median across the four panel members, nine (4.9%) discrepancies were rated as ‘no harm expected’, 59.5 (32.3%) as ‘very low’, 68.5 (37.2%) as ‘low’, 44 (23.9%) as ‘medium’ and 7.5 (4.1%) as ‘high’ probability of causing an ADE. Median probability values were used in the cost avoidance calculations.
Remedial healthcare utilisation
A summary of the remedial healthcare resources likely to be required in the absence of the pharmacist interventions (n=184) was rated by the expert panel and is shown in the online supplemental information.
Cost analysis
The estimated ADE cost for the 184 prescription errors was calculated using the median probability of ADE (table 4) multiplied by the cost of ADE (€1313),2526 which worked out at €34 014.49. This was extrapolated to reflect estimated annual discharge figures for the surgical ward (n=665). The estimated cost per pharmacist intervention (n=184) was €184.86.
Labour costs
The labour costs incurred in resolving the prescription discrepancies were extrapolated from the study period and an annual cost calculated. The mean time taken for a pharmacist review of each prescription was 9.75 min (range 5–36 min). The salary scale for a staff grade pharmacist used was €63 026 or €34.63 per hour based on a 35-hour working week.24 The cost per pharmacist intervention was calculated by dividing the total ADE-related cost (€34 014.49) by the number of interventions (n=184), yielding €184.86 per intervention.
Based on the estimated 665 patient discharges from the surgical ward per year, this would equate to 108.06 hours, or a mean additional cost of €3742.12 per year for a staff grade pharmacist. The additional pharmacist time taken to resolve the identified 184 prescription discrepancies with prescribers was 517 min, or a mean of 11 min per prescription (range 2–37 min), reflecting prescriber availability and/or prescription complexity. This additional time equates to €298.40 for a staff grade pharmacist.
Annually, this would result in an additional workload of 121.9 hours and costs of €4221.40.
While the prescribing doctor’s time was not recorded, it was estimated to be similar to the mean recorded pharmacist time (11 min). The additional labour cost was also calculated for the Intern Non-Consultant Hospital Doctor (NCHD) prescribers using the HSE consolidated pay scale (1 August 2025): €46 160 or €22.76 /hour based on a 39-hour week.24 This salary scale was deemed to be representative as 96% (n=48) of the prescriptions reviewed had been written by the primary team intern doctors. This would result in an additional 121.91 hours of prescriber/NCHD labour, or a mean additional cost of €2774.84 per year.
Financial impact calculation
The total cost avoidance of €565 490.90 minus the cost of the service (€10 569.37) results in a potential annual net cost benefit of €554 921.53. The CBR (net cost benefit divided by the cost of the service)25 of 52.5 was calculated for the provision of a clinical pharmacist discharge service to the surgical ward when 100% of the average annual discharge figures for the ward are reviewed (n=665).
Discussion
This pilot study shows the clinical and financial value of a pharmacist-led discharge reconciliation service on a surgical ward within an Irish hospital. The intervention significantly reduced both prescribing and communication errors at discharge, which are known contributors to patient harm. Notably, this study demonstrated an almost 75% reduction in predicted remedial healthcare use, reinforcing the importance of hospital pharmacist involvement at transitions of care.
These findings align with previous evaluations of pharmacist-led discharge MedRec in surgical patients, reporting reduced unintentional discrepancies, improved communication and continuity of care.28–32 Targeted reconciliation of perioperative anticoagulated patients has also been shown to ensure continuity of anticoagulation therapy after discharge,33 while interdisciplinary interventions have been associated with a reduction in unplanned hospital readmissions.34 This service showed the positive financial impact and potential for return on investment of a pharmacist-led medication reconciliation discharge service and provides evidence to support development and expansion of this service. From a financial perspective, the service on the surgical ward was associated with a projected annual net cost benefit of €554 921.53 and a CBR of 52.5 when applied to all estimated discharges from the surgical ward. While this CBR is higher than most published studies, it is consistent with a similar study carried out in the Acute Medical Admission Unit of an Irish teaching hospital which reported a CBR of 59.5.35 By contrast, a study by Gallagher et al25 reported a CBR of 8.6 for pharmacist interventions in inpatients, and Jermini et al36 estimated a more conservative return on investment (ROI) of 1–1.7 for inpatient pharmacist reviews using a full-cost model.
Other recent evaluations have shown positive but smaller ROI: a 6-year study in Egypt found an ROI of 7.6:1,37 while a US Military Health System reported ROI values of 2.4:1.38 Although not part of the primary analysis, a sensitivity comparison using a full-time equivalent (FTE) costing approach (mid-point pharmacist salary €63 026 plus service costs €10 569) would yield a lower CBR of approximately 6.7. This more conservative estimate aligns more closely with other studies, emphasising that the CBR is sensitive to the costing perspective adopted. Nonetheless, even under a FTE costing approach, the pharmacist discharge service shows a positive return, reinforcing its clinical and financial value.
The study by Kirwan et al39 quantified the impact of discharge medication errors in terms of quality-adjusted life years and associated healthcare utilisation at a population level. Our study complements this work through a ward-level and prescription-level evaluation of pharmacist-led discharge MedRec, showing its potential to prevent a substantial proportion of potential post-discharge healthcare interventions, resulting in significant cost avoidance. These findings reinforce the clinical and economic value of pharmacist involvement in transitions of care.
Across the reviewed discharge prescriptions, prescribing omissions, incorrect doses and incorrect allergy documentation were the most frequently identified issues, with similar rates to those reported both nationally and internationally.40–42 The prominence of discrepancies involving high-risk medications such as antiplatelets, anticoagulants, opioids, insulin and antimicrobials further highlights the critical role of pharmacists in identifying and resolving potentially harmful errors prior to discharge.837
The severity of harm reported in our study is consistent with patterns observed across previous studies; however, estimates vary depending on the patient cohort and classification approach used. Studies using structured tools or multidisciplinary panels164043 also report that most discrepancies fall within the minor or moderate degree of harm, with only a small percentage considered severe, mirroring the results observed using our expert panel assessment. In contrast, Hammour et al28 reported a higher proportion of moderate-to-severe discrepancies in a surgical cohort, whereas Grimes et al44 found severe harm potential to be uncommon in a mixed cohort.
Differences across studies likely reflect the subjectivity in clinical judgement processes and severity scales. Within this context, the harm potential identified in our surgical population is broadly comparable and reinforces the value of pharmacist discharge MedRec to prevent clinically important errors at the point of discharge.
Communication-related discrepancies, including documentation of medication changes, missing dose adjustments and failure to document discontinuations, were also prevalent in this study. Without pharmacist intervention these errors may have gone undetected, posing ongoing risks to patient safety post-discharge, exacerbated by the lack of an electronic health record system in Ireland.1415 Previous studies have shown that such errors, if unresolved at discharge, can persist after discharge and contribute to patient harm.47 Clinical pharmacist involvement at discharge, in collaboration with the prescriber, has been shown to improve the quality of discharge documentation, thus reducing the potential for harm.745 Clinical pharmacist involvement facilitated direct real-time communication with prescribers, resulting in a high acceptance rate for pharmacist interventions. This rate is consistent with literature demonstrating a strong positive correlation between direct verbal communication and acceptance of pharmacist intervention by prescribers.46 While pharmacist prescribing is not legislated yet in Ireland, it is on the horizon47 and consideration should be given to the introduction of hospital policy to enable pharmacists to make adjustments to discharge prescriptions in collaboration with the prescribers.
This could lead to significant benefits in terms of time and cost savings and, importantly, reduced patient harm. This has demonstrated great benefits in other hospitals, providing a protective effect against potentially severe medication errors in patients and improved quality of prescribing in older patients.19
A key finding from this study is the potential impact on remedial healthcare utilisation post-discharge. Expert panel assessment suggested that, without pharmacist intervention, approximately one-third of patients would likely have required GP consultations, prescriptions or further clinical input as a result of unresolved medication discrepancies or errors. Following pharmacist intervention, the likelihood of remedial healthcare utilisation was significantly reduced.
This reduction has significant implications, especially in the context of strained primary care services in Ireland where GP shortages continue to escalate.48 Reducing unnecessary follow-up also alleviates the burden on community pharmacists, who often must reconcile discrepancies retrospectively without full access to clinical background.4950
Strengths and limitations
The strenths of this study include the use of real-time intervention tracking for cost analysis and the application of a multidisciplinary expert panel to evaluate the potential clinical impact of each discrepancy, an approach recommended in similar studies.3255152
However, as a single-centre pilot study on one ward with one pharmacist and a limited sample size, the findings may not be generalisable across hospital settings or specialties. The costing approach adopted in this study reflects a service expansion model rather than a FTE costing model. While this approach is appropriate for contexts where pharmacist salaries are already budgeted, results may differ in settings requiring dedicated pharmacist recruitment for the service.
Resource limitations restricted the pharmacist capacity to review all discharges during the study time period. Furthermore, while expert opinion was a pragmatic and ethically appropriate means to estimate potential harm and healthcare utilisation, it remains inherently subjective. The expert panel included two pharmacists and two Specialist Registrars in medicine, reflecting different professional perspectives.
As expected, some discrepancies showed variability in the assigned probability of ADEs and predicted post-discharge interventions. Median probability values were used in the cost avoidance calculations to mitigate the impact of individual differences. Future studies could formally assess inter-rater reliability to quantify agreement among experts.
Conclusion
This pilot study underscores the positive clinical and financial impact of a pharmacist discharge service in reducing prescription errors, as well as the potential risk of patient harm and its associated ADE costs. This service showed significant potential for cost savings and a reduction in remedial healthcare utilisation, highlighting the key role of hospital pharmacists at discharge in managing medication discrepancies and improving communication during transitions of care. While previous studies have demonstrated the clinical and financial benefits of these interventions, few have focused on their impact on remedial healthcare utilisation.
The findings from this pilot study suggest a significant benefit of a pharmacist discharge service, supporting further development of the service within the hospital. Standardising cost calculations through a Health Technology Assessment and enabling collaborative prescribing within a multidisciplinary team are key steps to improving efficiency, patient safety and cost effectiveness in the hospital discharge process.
Supplementary material