Poor communication of medicines information at hospital discharge can lead to discrepancies and increase workload, as well as posing risks to patient safety. Katie Joyce discusses a quality improvement project exploring how standardised pharmacy processes and cross-sector collaboration between pharmacists and other healthcare professionals can reduce avoidable errors and improve continuity of care.
Safe and effective transfer of medicines information at hospital discharge remains a significant challenge across healthcare systems. Indeed, anywhere between 25% and 80% of patients experience at least one medication discrepancy or failure to communicate in-hospital medication changes at discharge, according to a 2014 systematic review by Lehnbom et al.
A later UK primary care audit showed overall good adherence to discharge medication documentation standards. However, some changes made during hospital stays were not consistently communicated or recorded in discharge summaries.
Consequently, these alterations were unlikely to be followed up in general practice post-hospital discharge, resulting in unintentional omissions of pre-admission medicines in discharge communications for about one-third of patients.
Rationale and aim for the discharge initiative
To address these issues, a quality improvement project initiated through collaboration between a secondary care pharmacist within an NHS Foundation Trust in north-east England and a primary care pharmacist, aimed to explore how cross-sector working could be strengthened to improve discharge communication and reduce avoidable medicines discrepancies.
The project closely aligns with professional standards for hospital pharmacy services from the Royal Pharmaceutical Society (now the Royal College of Pharmacy), which emphasise the importance of safe transfer of care and provide a framework for evaluating and improving discharge processes.
This includes accurate, timely and complete communication of medicines information at discharge; identifying and addressing medicines-related risks; and supporting continuity of supply so that receiving healthcare professionals can safely assume responsibility for care.
Initial discussions highlighted tensions at the care interface. Primary care colleagues reported a lack of clarity in discharge communications, while secondary care teams described a high volume of post-discharge queries, often perceived as avoidable.
These pressures signalled an opportunity to improve both the quality and efficiency of the transfer of medicines-related information.
Current discharge process challenges
Process mapping identified the key stages in discharge medicines communication. Typically, a resident doctor prepares the discharge medication list, which is then clinically screened by a pharmacist before being processed within primary care by administrative teams, pharmacy staff or general practitioners.
Where discrepancies or ambiguities arise, resolution often depends on informal communication between primary and secondary care. In parallel, referrals to the discharge medicines service (DMS) may be sent to community pharmacies.
Although the DMS provides an important safety net, uptake remains low and community pharmacies are limited in their ability to act on discrepancies, often needing to refer issues back to general practice or hospital teams. This fragmented pathway contributes to inefficiencies and duplication of work.
While some regions benefit from integrated care pharmacy hubs with shared record access, such models are not universally available.
Pharmacist-led discharge screening aims to ensure accurate and complete medicines information, optimise therapy and support continuity of supply. However, this process was not standardised within this Trust.
Variability in workflows – particularly following the introduction of electronic patient records – meant that screening occurred in different sequences, to varying standards and within inconsistent timeframes. This variation limits quality assurance and complicates performance monitoring.
Evaluation design and key findings
A retrospective observational evaluation was conducted to assess the accuracy and quality of discharge summaries regarding medicines.
A total of 190 discharges from medical wards were analysed over a one-week period. Exclusion criteria included short length of stay (<1.5 days), patient death or self-discharge and transfer to another Trust. The acute admission wards were also excluded.
The average number of discharge medicines per patient was 10, with a mean of two new medicines initiated per discharge. Most patients (81%) had at least one new medicine started, and 23% had four or more.
Changes to pre-admission medicines were common, with 19% of discharges containing amended medicines and 33% with medicines stopped.
Pharmacy admission medicines reconciliation was completed in 80% of cases, and 73% of discharges were screened by a pharmacist.
Inconsistent pharmacy involvement at discharge
Marked variation in pharmacy involvement at discharge was observed. Some patients were discharged without pharmacy input, while others were screened by pharmacists unfamiliar with the patient or without prior medicines reconciliation at admission.
Optimal practice involved screening by a ward-based pharmacist with knowledge of the patient’s clinical history and a completed medicines reconciliation.
Operational factors also contributed to inconsistency. Discharges could bypass pharmacy review when no medication changes were made, when patients used their own medicines or when ward stock was used.
Documentation inconsistencies, such as missing pharmacist identifiers or variable placement of medicine information, further reduced clarity for primary care teams.
Feedback from primary care suggested that inconsistent discharge quality increases the complexity of medicines reconciliation, contributes to adverse events and generates avoidable workload across the system.
Discrepancies in discharge medicines
A total of 166 discrepancies, classified as intentional or unintentional, were identified across the 190 discharges, averaging 0.88 per patient – a rate comparable to that reported by Mills et al.
A total of 71% of the discrepancies were classified as intentional and reflected appropriate clinical decisions that were not clearly communicated, whereas unintentional discrepancies represented potential errors or omissions.
The most common intentional discrepancy (52%) was a stopped medicine not being clearly documented. A further 25% involved new medicines being incorrectly recorded as unchanged.
Of the unintentional discrepancies (29% of the total), omission of pre-admission medicines was the most frequent, accounting for 73% of these, and the documented dose for pre-admission medicines was incorrect in 10%.
Discharges with one discrepancy were found to be more likely to contain additional discrepancies.
Impact of pharmacy interventions at discharge
Pharmacy involvement was associated with significant improvements in discharge accuracy.
Pharmacy admission medicines reconciliation was associated with a statistically significant reduction in unintentional discrepancies. Similarly, pharmacist discharge screening significantly reduced the likelihood of unintentional discrepancies.
No statistically significant reduction in intentional discrepancies was observed, indicating that communication of clinical decisions remains an important area for improvement.
These findings reinforce the value of pharmacy reconciliation at both admission and discharge in improving discharge accuracy and align with a UK study demonstrating that pharmacist involvement in discharge medicines reconciliation improves the quality of electronic discharge summaries.
Strengths and limitations
The evaluation has several strengths that support the relevance and applicability of its findings. It was conducted using real-world data from routine clinical practice, providing a pragmatic reflection of discharge processes in a UK NHS setting.
Another key strength is the distinction between intentional and unintentional discrepancies, which clarifies communication issues and potential prescribing or documentation errors and enables more targeted recommendations.
The evaluation also incorporated analysis of pharmacy interventions at both admission and discharge, supported by statistical testing, strengthening the validity of the observed associations between pharmacy involvement and reducing unintentional discrepancies.
However, several limitations should be considered. The evaluation was conducted over a short time frame and within a single NHS Trust, which may limit generalisability to other organisations with different patient populations, staffing models or digital systems.
Local variation in practice may not fully reflect other settings, particularly those with more developed integrated care models or differing levels of pharmacy resources.
Classifying discrepancies required clinical judgement, introducing the potential for observer bias, although this reflects real-world clinical decision-making. In addition, the study did not quantify the time or resource implications of discrepancies or their resolution, which would be valuable for assessing system-wide efficiency.
Overall, while the findings are consistent with existing literature and offer useful insights into local practice, they should be treated as indicative rather than definitive. Further evaluation across multiple sites and over longer time periods would strengthen the evidence base and support broader generalisation.
Quality improvement actions
Based on the findings, several improvement actions were identified for the Trust. These included extending the evaluation to other clinical areas, particularly surgical wards and those without routine pharmacy input, and further work to quantify the impact of pharmacist interventions.
A priority area is the development of a standardised discharge screening process, supported by clear standard operating procedures and defined key performance indicators.
To improve clarity, the introduction of standardised statements within discharge summaries was proposed, such as indicating when no intentional medication changes have been made or when pharmacy reconciliation has not been completed.
Strengthening collaboration with primary care and integrated care board medicines optimisation teams remains central to this work.
As part of the wider initiative, cross-sector visits were arranged between primary and secondary care pharmacy teams. These sessions aimed to improve mutual understanding of discharge processes and the challenges faced in each setting, as well as to identify opportunities for improved communication.
Post-visit survey results demonstrated improved knowledge and confidence among participants. Understanding of how discharge summaries are generated increased, with all participants reporting greater clarity. Confidence in knowing who to contact to resolve discrepancies also improved.
Qualitative feedback highlighted an increased awareness of how discrepancies arise and reinforced the importance of direct communication between pharmacists across sectors. Participants reported greater attention to clinical summaries and improved interpretation of discharge information.
Conclusion
This quality improvement project and its evaluation highlight the critical role of pharmacy teams in ensuring safe and effective transfer of medicines information at discharge.
While pharmacist involvement significantly reduces unintentional discrepancies, variability in processes and communication remains a key challenge.
Standardisation of discharge screening, combined with strengthened cross-sector collaboration, offers a practical approach to improving both the quality and efficiency of discharge communication.
Addressing these challenges is essential for improving patient safety, reducing avoidable workload and supporting more integrated models of care. Continued evaluation and collaboration will be key to sustaining improvements and enabling wider adoption of best practice.
Author
Katie Joyce MPharm PGDip
Medicines Value Programme lead, North East and North Cumbria Provider Collaborative, UK
