Clinical Documentation Research Articles

Improving the Monitoring and Management of Clozapine-Induced Gastrointestinal Hypomotility (CIGH) in Community Mental Health Services: A Quality Improvement Approach.

Clozapine is the only approved antipsychotic for refractory schizophrenia to date. It can cause a range of serious and fatal adverse effects, including Clozapine-Induced Gastrointestinal Hypomotility (CIGH). While guidance is readily available to help manage CIGH effectively in hospital inpatients, practical recommendations applicable to the community (outpatient) setting are lacking. This project set out to improve the prevention, detection and management of CIGH in psychiatric outpatients. An initial baseline audit followed by quality improvement work was undertaken in a busy support worker-run community clozapine clinic focusing on, education and training, risk assessments and clinical documentation. The project was registered and managed using the Life QI web-based platform, where a set of primary and secondary drivers were defined and change ideas were executed. Qualitative and quantitative data were collected over a three-month period, demonstrating a significant improvement in clinical documentation (up from 36% to 99%). 23% of enhanced risk assessments resulted in treatment recommendations, modifiable risk factors were proactively discussed in 53% of clinic appointments and 65% of patients were provided with additional written information on CIGH. It was evident from staff and our patient feedback that further efforts would be required to continue to raise awareness about harms of unmanaged constipation among this client group. Future approaches may include enhanced collaborative efforts with primary care, and improving the skill mix in existing clozapine clinics, which could include the utilisation of mental health pharmacists.

Artificial intelligence guided screening for cardiomyopathies in an obstetric population: a pragmatic randomized clinical trial.

Nigeria has the highest reported incidence of peripartum cardiomyopathy worldwide. This open-label, pragmatic clinical trial randomized pregnant and postpartum women to usual care or artificial intelligence (AI)-guided screening to assess its impact on the diagnosis left ventricular systolic dysfunction (LVSD) in the perinatal period. The study intervention included digital stethoscope recordings with point of-care AI predictions and a 12-lead electrocardiogram with asynchronous AI predictions for LVSD. The primary end point was identification of LVSD during the study period. In the intervention arm, the primary end point was defined as the number of identified participants with LVSD as determined by a positive AI screen, confirmed by echocardiography. In the control arm, this was the number of participants with clinical recognition and documentation of LVSD on echocardiography in keeping with current standard of care. Participants in the intervention arm had a confirmatory echocardiogram at baseline for AI model validation. A total of 1,232 (616 in each arm) participants were randomized and 1,195 participants (587 intervention arm and 608 control arm) completed the baseline visit at 6 hospitals in Nigeria between August 2022 and September 2023 with follow-up through May 2024. Using the AI-enabled digital stethoscope, the primary study end point was met with detection of 24 out of 587 (4.1%) versus 12 out of 608 (2.0%) patients with LVSD (intervention versus control odds ratio 2.12, 95% CI 1.05-4.27; P = 0.032). With the 12-lead AI-electrocardiogram model, the primary end point was detected in 20 out of 587 (3.4%) versus 12 out of 608 (2.0%) patients (odds ratio 1.75, 95% CI 0.85-3.62; P = 0.125). A similar direction of effect was observed in prespecified subgroup analysis. There were no serious adverse events related to study participation. In pregnant and postpartum women, AI-guided screening using a digital stethoscope improved the diagnosis of pregnancy-related cardiomyopathy. ClinicalTrials.gov registration: NCT05438576.

Protocol for Designing a Model to Predict the Likelihood of Psychosis From Electronic Health Records Using Natural Language Processing and Machine Learning.

Rapid identification of individuals developing a psychotic spectrum disorder (PSD) is crucial because untreated psychosis is associated with poor outcomes and decreased treatment response. Lack of recognition of early psychotic symptoms often delays diagnosis, further worsening these outcomes. The proposed study is a cross-sectional, retrospective analysis of electronic health record data including clinician documentation and patient-clinician secure messages for patients aged 15-29 years with ≥ 1 primary care encounter between 2017 and 2019 within 2 Kaiser Permanente regions. Patients with new-onset PSD will be distinguished from those without a diagnosis if they have ≥ 1 PSD diagnosis within 12 months following the primary care encounter. The prediction model will be trained using a trisourced natural language processing feature extraction design and validated both within each region separately and in a modified combined sample. This proposed model leverages the strengths of the large volume of patient-specific data from an integrated electronic health record with natural language processing to identify patients at elevated chance of developing a PSD. This project carries the potential to reduce the duration of untreated psychosis and thereby improve long-term patient outcomes.

Use of an ambient artificial intelligence tool to improve quality of clinical documentation

BackgroundElectronic health records (EHRs) have contributed to increased workloads for clinicians. Ambient artificial intelligence (AI) tools offer potential solutions, aiming to streamline clinical documentation and alleviate cognitive strain on healthcare providers. ObjectiveTo assess the clinical utility of an ambient AI tool in enhancing consultation experience and the completion of clinical documentation. MethodsOutpatient consultations were simulated with actors and clinicians, comparing the AI tool against standard EHR practices. Documentation was assessed by the Sheffield Assessment Instrument for Letters (SAIL). Clinician experience was measured through questionnaires and the NASA Task Load Index. ResultsAI-produced documentation achieved higher SAIL scores, with consultations 26.3% shorter on average, without impacting patient interaction time. Clinicians reported an enhanced experience and reduced task load. ConclusionsThe AI tool significantly improved documentation quality and operational efficiency in simulated consultations. Clinicians recognised its potential to improve note-taking processes, indicating promise for integration into healthcare practices.

CDI 2.0: Optimising CMI and risk-adjustment value through data analytics and provider education

Accurate clinical documentation is vital for any healthcare facility. Documentation in the record plays a critical role in reimbursement, case mix index, risk adjusted quality outcomes, length of stay days, etc. The need for Clinical documentation improvement (CDI) specialists is high. In this article, we discuss the success we have had by transitioning from a traditional approach to capture severity reporting (CDI 1.0), which relies heavily on the concurrent review of medical records by the CDI team, to a patient population driven and provider focused approach (CDI 2.0).

For Whom the Note Scrolls: A Brief History of the Medical Record's Role in Education and the Risks It Faces in the Age of OpenNotes.

Dating back to ancient civilizations when records were carefully transcribed onto papyrus, clinical documentation has long served as a cornerstone of medical-and especially neurologic-education. From the case histories of Hippocrates to the diurnal patient logs used by trainees in the 18th and 19th centuries, clinical notes have an extended history as invaluable instruments of pedagogy, scholarly practice, and interprofessional communication. The novel paradigm introduced by Lawrence Weed in the 1950s, advocating for the problem-oriented medical record system, revolutionized the clinical note template and emphasized the need for physicians' carefully considered analyses of a patient's presentation to be clearly reflected in well-organized documentation. In the realm of medical records today, however, a profound shift is underway, largely propelled by the emergence of electronic medical records, the OpenNotes mandate of the federal 21st Century Cures Act, and, most recently, artificial intelligence (AI). Appropriately, patients now have full access to their medical records, but this raises critical questions. Should clinical notes now prioritize patient comprehension over their traditional role as educational instruments, aide-mémoire, and repositories of detailed assessments and insights? What role, if any, should AI have in the creation of physician notes and patient-facing clinical documents? These tensions underscore the delicate balance between transparency and the preservation of notes' clinical integrity and analytical depth. As we navigate the path forward, finding an equilibrium between openness and the continued utility of medical records as tools for education and professional communication will be imperative.

51496 Artificial Intelligence in Clinical Documentation: Impact on Dermatologist Workflow Compared to Other Specialties

51496 Artificial Intelligence in Clinical Documentation: Impact on Dermatologist Workflow Compared to Other Specialties

Improving Athletic Trainers' Knowledge of Clinical Documentation Through Novel Educational Interventions: A Randomized Controlled Trial.

Athletic trainers (ATs) have reported the need for more educational resources about clinical documentation. To investigate the effectiveness of passive and active educational interventions to improve practicing ATs' clinical documentation knowledge. Randomized controlled trial, sequential explanatory mixed methods study. Online module(s), knowledge assessment, and interviews. We emailed 18 981 practicing ATs across employment settings, of which 524 ATs were enrolled into a group (personalized learning pathway [PLP = 178], passive reading list [PAS = 176], control [CON = 170]) then took the knowledge assessment. A total of 364 ATs did not complete the intervention or postknowledge assessment; therefore, complete responses from 160 ATs (PLP = 39, PAS = 44, CON = 77; age = 36.6 ± 11.2 years, years certified = 13.9 ± 10.7) were analyzed. Knowledge assessment (34 items) and interview guides (12-13 items) were developed, validated, and piloted with ATs before study commencement. We summed correct responses (1 point each, 34 points maximum) and calculated percentages and preknowledge and postknowledge mean change scores. Differences among groups (PLP, PAS, CON) and time (preintervention, postintervention) were calculated using a 3 × 2 repeated-measures analysis of variance (P ≤ .05) with post hoc Tukey HSD. Semistructured interviews were conducted (PLP = 15, PAS = 14), recorded, transcribed, and analyzed following the consensual qualitative research tradition. No differences in the preknowledge assessment were observed between groups. We observed a group × time interaction (F2,157 = 15.30, P < .001; partial η2 = 0.16). The PLP group exhibited greater mean change (M = 3.0 ± 2.7) than the PAS (M = 1.7 ± 3.0, P = .049) and CON (M = 0.4 ± 2.2, P < .001) groups. Descriptively, ATs scored lowest on the legal (61.3% ± 2.1%), value of the AT (63.7% ± 4.3%), and health information technology (65.3% ± 3.7%) items. Whereas ATs described being confident in their documentation knowledge, they also identified key content (eg, legal considerations, strategies) which they deemed valuable. The educational interventions improved ATs' knowledge of clinical documentation and provided valuable resources for their clinical practice; however, targeted continuing education is needed to address knowledge gaps.

Prognostic Implications of Unexpected Positive Intraoperative Cultures in Two-Stage Knee Arthroplasty Reimplantations: A Minimum Five-Year Retrospective Analysis

BackgroundThe clinical challenge of unexpected positive intraoperative cultures (UPICs) persists in 2-stage resection arthroplasty for managing periprosthetic joint infections (PJIs) following total knee arthroplasty.(TKA). This study aimed to investigate the incidence of UPICs during the definitive reimplantation phase of 2-stage resection arthroplasty of the knee and to assess both the infection-free and revision-free survivorship of the implanted prosthesis. MethodsThis retrospective study included 450 2-stage resection arthroplasties of primary knee prostheses performed between January 2012 and April 2017. Patients were excluded if they: (1) underwent three or more staged resections, (2) had ambiguous clinical documentation or deviated from the 2-stage protocol, or (3) underwent revision arthroplasty prior to the PJI. Additionally, patients presumed aseptic before the second-stage reimplantation were excluded if they lacked joint aspiration or met the 2011 Musculoskeletal Infection Society criteria for PJI before implantation. ResultsAfter exclusions, 300 patients were analyzed. Among them, 14% had UPIC during the second-stage reimplantation. The follow-up time was 2,316 (range, 1,888 to 3,737) days and 2,531 (range, 1,947 to 3,349) days for UPIC and negative intraoperative culture (NIC) groups, respectively. Rerevision due to subsequent PJI occurred in 26.2% of UPIC patients and 15.1% of NIC patients. The 2-year infection-free survival rates for the NIC, one UPIC, and ≥ two UPIC cohorts were 99.5, 98.2, and 94.3%, respectively, while the 5-year survival rates were 92.1, 91.1, and 54.3%, respectively. The unfavorable survivorship was significantly different in multiple UPIC cases (P < 0.001). Multiple UPICs with pathogens consistent with the first-stage findings were strongly associated with the risk of reinfection (P < 0.001). ConclusionsAn UPIC was identified in 14% of second-stage reimplantations. Patients who had multiple UPICs demonstrated truncated survivorship and suboptimal outcomes relative to the NIC and single UPIC cohorts, especially with pathogen consistency to the first-stage surgery. Level of evidenceIII.

Maximizing Learning in Caring for Older Patients Through a Multi-Specialty Simulation Approach

Background: Simulation-based learning has been utilized in medical education and studied to enhance its educational impact since the 1960s (Hallinger, &amp; Wang, 2020). However, there is a lack of multispecialty simulations in the literature (Fisher, &amp; Walker, 2014; Age UK. 2023; NHS England, 2021; Romero-Ortuno, Stuck, &amp; Masud, 2022; Keijsers, et al., 2016). We developed and delivered simulations on Orthopaedics and Geriatric topics. Simulation mannequins, role players, imaging, and simulated clinical documentation were incorporated into scenarios. We evaluated the effectiveness of this approach on students’ knowledge and confidence when caring for older patients. Methods: Fourth-year medical students at the University of Nottingham received simulation-based teaching during their Geriatrics placement at Queens Medical Centre. Their knowledge and confidence levels were assessed before and after the simulations. We utilized six knowledge-based and six confidence-level questions mapped to their learning outcomes for hip fractures, pressure ulcers, and discharge planning stations. In addition, we asked students to provide scenario-specific feedback and their thoughts on whether the simulation workshop was pitched at the right level. Results: Students’ knowledge and confidence levels improved significantly following the simulation workshop. About 75% of the students displayed enhanced results in the knowledge-based category. In terms of their confidence level, there was improvement seen across all simulation stations, with most learners feeling four times more confident when comparing proportions pre and post-simulation. Furthermore, 99% of the students thought the simulations had positively impacted their learning. Conclusion: Our findings demonstrate the effectiveness of the multispecialty simulation approach in undergraduate geriatrics teaching. Given significant improvements in students’ knowledge, confidence levels, and positive feedback, we aim to continue delivering this multispecialty simulation-based teaching to our students. To measure the long-term efficacy of this approach, we can perhaps re-evaluate students learning after a month to assess the efficacy of this simulation workshop.

Viability of Open Large Language Models for Clinical Documentation in German Health Care: Real-World Model Evaluation Study.

The use of large language models (LLMs) as writing assistance for medical professionals is a promising approach to reduce the time required for documentation, but there may be practical, ethical, and legal challenges in many jurisdictions complicating the use of the most powerful commercial LLM solutions. In this study, we assessed the feasibility of using nonproprietary LLMs of the GPT variety as writing assistance for medical professionals in an on-premise setting with restricted compute resources, generating German medical text. We trained four 7-billion-parameter models with 3 different architectures for our task and evaluated their performance using a powerful commercial LLM, namely Anthropic's Claude-v2, as a rater. Based on this, we selected the best-performing model and evaluated its practical usability with 2 independent human raters on real-world data. In the automated evaluation with Claude-v2, BLOOM-CLP-German, a model trained from scratch on the German text, achieved the best results. In the manual evaluation by human experts, 95 (93.1%) of the 102 reports generated by that model were evaluated as usable as is or with only minor changes by both human raters. The results show that even with restricted compute resources, it is possible to generate medical texts that are suitable for documentation in routine clinical practice. However, the target language should be considered in the model selection when processing non-English text.

Advancing Refugee Health Data Management: The Implementation of ICD-11 in UNRWA's Primary Care System.

The United Nations Relief and Works Agency for Palestine Refugees in the Near East (UNRWA) was one of the earliest healthcare systems globally to implement the International Classification of Diseases, Eleventh Revision (ICD-11) across its 140 clinics serving 5.9 million Palestine refugees. This paper discusses the integration of ICD-11 into UNRWA's cloud-based electronic medical record (EMR) system, identifying both the barriers and facilitators involved and analyzing trends in clinical documentation and healthcare utilization. The key challenges included data privacy provisions, integration into a coordinated care model, complex classification schema for primary care settings, frequent staff turnover, and limited data analysis capabilities. Conversely, facilitators included physician-tailored training and on-site support, system compatibility, a multidisciplinary team approach, policy support from UNRWA and the World Health Organization (WHO), and leadership commitment and effective change management. Medical officers (MOs) using ICD-11 reported greater satisfaction with the system's capabilities in managing and visualizing health information. This article contributes to the discourse on health data management in complex humanitarian settings, offering insights into the benefits and challenges of implementing advanced classification systems like ICD-11. Future research should explore longitudinal impacts and further integration with global health systems, ensuring that the advancements in classification continue to support the overarching goal of health equity and access in vulnerable and hard-to-reach populations.

Term Candidate Generation to Enrich Clinical Terminologies with Large Language Models.

Annotated language resources derived from clinical routine documentation form an intriguing asset for secondary use case scenarios. In this investigation, we report on how such a resource can be leveraged to identify additional term candidates for a chosen set of ICD-10 codes. We conducted a log-likelihood analysis, considering the co-occurrence of approximately 1.9 million de-identified ICD-10 codes alongside corresponding brief textual entries from problem lists in German. This analysis aimed to identify potential candidates with statistical significance set at p < 0.01, which were used as seed terms to harvest additional candidates by interfacing to a large language model in a second step. The proposed approach can identify additional term candidates at suitable performance values: hypernyms MAP@5=0.801, synonyms MAP@5 = 0.723 and hyponyms MAP@5 = 0.507. The re-use of existing annotated clinical datasets, in combination with large language models, presents an interesting strategy to bridge the lexical gap in standardized clinical terminologies and real-world jargon.

Unsupervised Extraction of Body-Text from Clinical PDF Documents.

Automatic extraction of body-text within clinical PDF documents is necessary to enhance downstream NLP tasks but remains a challenge. This study presents an unsupervised algorithm designed to extract body-text leveraging large volume of data. Using DBSCAN clustering over aggregate pages, our method extracts and organize text blocks using their content and coordinates. Evaluation results demonstrate precision scores ranging from 0.82 to 0.98, recall scores from 0.62 to 0.94, and F1-scores from 0.71 to 0.96 across various medical specialty sources. Future work includes dynamic parameter adjustments for improved accuracy and using larger datasets.

Implementing a Photodocumentation Program.

The widespread availability of smart devices has facilitated the use of medical photography, yet photodocumentation workflows are seldom implemented in healthcare organizations due to integration challenges with electronic health records (EHR) and standard clinical workflows. This manuscript details the implementation of a comprehensive photodocumentation workflow across all phases of care at a large healthcare organization, emphasizing efficiency and patient safety. From November 2018 to December 2023, healthcare workers at our institution uploaded nearly 32,000 photodocuments spanning 54 medical specialties. The photodocumentation process requires as few as 11 mouse clicks and keystrokes within the EHR and on smart devices. Automation played a crucial role in driving workflow efficiency and patient safety. For example, body part rules were used to automate the application of a sensitive label to photos of the face, chest, external genitalia, and buttocks. This automation was successful, with over 50% of the uploaded photodocuments being labeled as sensitive. Our implementation highlights the potential for standardizing photodocumentation workflows, thereby enhancing clinical documentation, improving patient care, and ensuring the secure handling of sensitive images.

A Missing Link on Semantic Interoperability.

Syntactic interoperability among health ICT systems is well-established, but achieving semantic interoperability requires more than just exchanging codes. We present a pragmatic, safe, and effective approach towards an ambitious goal: enabling any software to process a critical mass of routine clinical information in a replicable manner across various systems and local contexts. We advocate for the use of reliable, parsimonious coding to handle the most replicable aspects of data processing for routine patient information, while reserving the intricate interpretation of individual patient data nuances for skilled professionals, possibly supported by Artificial Intelligence tools. We suggest coping with routine tasks by focusing on a limited set of a few thousand data elements, named the 'Clinical Documentation Kernel' (CDK). This approach will provide direct benefits to users and assist in the human interpretation of other patient information. Our preliminary study focuses on the 'primitives' and 'qualifiers' that bring the highest value to the health ecosystem in various authoritative scenarios in the field of diabetes.

Problem Detection and Optimization of a Personal Health Record Driven by a Helpdesk Support Tool.

This study aims to assess the value of a help desk in a Personal Health Record (PHR) for monitoring and improving PHR functionalities and processes. Through analysis of 327 help desk tickets categorized by users as telemedicine and PHR requests, we identified four main gaps: information and feedback, access to clinical documents, usability, and ubiquity. The tickets highlighted end users' needs such as timely feedback, contextual information, effective communication, and ease of access to clinical documentation. These findings were crucial for initiating improvement cycles for various functionalities and processes within the PHR.

Report on Prosthetic Fitting, Mobility, and Overall Satisfaction after Major Limb Amputation at a German Maximum Care Provider

Background: Satisfaction with prosthesis plays a key role in regaining mobility and is important for optimizing prosthetic usage, mobility, and increasing compliance with medical regimen. Despite unchangeable factors like age and comorbidities, other factors, like pain, received rehabilitation, satisfaction with assistive devices, service, and information, can be changed and might contribute to a better usage and acceptance of the prosthesis and amputees’ mobility. Objectives: The aim of the study was to analyze mobility, pain, supply of assistive devices, and additional therapies received after major limb amputations. Furthermore, a correlation of those parameters was evaluated. Methods: Retrospective identification of patients with major limb amputation (operation and procedures classification system (OPS)) and relevant related demographics within the clinical documentation system during a four-year observation time. In addition, we undertook prospective assessment of mobility (K-level), pain qualities, additional therapies, self-rated overall quality of life (QoL) and degree of adaptation to the life after amputation, dependency from caregivers, and satisfaction with the provided assistive devices (QUEST). Results: A total of 164 patients (mean age 68, age range: 19 to 97 years) underwent major limb amputation. A total of 27 questionnaires were returned and analyzed. All those traumatic and nontraumatic amputees received assistive devices. Although mobility and QoL decreased significantly after amputation, a high satisfaction with provided prosthetic and assistive devices and care was found. Conclusions: Amputation registries are becoming elementary to allow for nationwide comparisons of clinics, to identify the requirements of amputees, and to design an interdisciplinary care model for a successful comprehensive approach.

Assessing the Effectiveness of Automatic Speech Recognition Technology in Emergency Medicine Settings: A Comparative Study of Four AI-powered Engines.

Cutting-edge automatic speech recognition (ASR) technology holds significant promise in transcribing and recognizing medical information during patient encounters, thereby enabling automatic and real-time clinical documentation, which could significantly alleviate care clinicians' burdens. Nevertheless, the performance of current-generation ASR technology in analyzing conversations in noisy and dynamic medical settings, such as prehospital or Emergency Medical Services (EMS), lacks sufficient validation. This study explores the current technological limitations and future potential of deploying ASR technology for clinical documentation in fastpaced and noisy medical settings such as EMS. In this study, we evaluated four ASR engines, including Google Speech-to-Text Clinical Conversation, OpenAI Speech-to-Text, Amazon Transcribe Medical, and Azure Speech-to-Text engine. The empirical data used for evaluation were 40 EMS simulation recordings. The transcribed texts were analyzed for accuracy against 23 Electronic Health Records (EHR) categories of EMS. The common types of errors in transcription were also analyzed. Among all four ASR engines, Google Speech-to-Text Clinical Conversation performed the best. Among all EHR categories, better performance was observed in categories "mental state" (F1 = 1.0), "allergies" (F1 = 0.917), "past medical history" (F1 = 0.804), "electrolytes" (F1 = 1.0), and "blood glucose level" (F1 = 0.813). However, all four ASR engines demonstrated low performance in transcribing certain critical categories, such as "treatment" (F1 = 0.650) and "medication" (F1 = 0.577). Current ASR solutions fall short in fully automating the clinical documentation in EMS setting. Our findings highlight the need for further improvement and development of automated clinical documentation technology to improve recognition accuracy in time-critical and dynamic medical settings.

2354 Clinical assessment and documentation of delirium in older patients undergoing vascular surgery: a quality improvement project

Abstract Introduction The Centre for Perioperative Care recommends the assessment and documentation of delirium using a validated tool such as the 4-AT in older people undergoing surgery. Aim This quality improvement project (QIP) aimed to improve the assessment and documentation of delirium in patients aged 65 and above following vascular surgery in a tertiary centre. Methods Patients aged ≥65 years who had undergone vascular surgery were identified and data was collected with access to the electronic patient record system. Analysis was carried out using Microsoft Excel and SPSS. Following baseline measurements taken in August 2023, 1 plan-do-study-act (PDSA) cycle was completed between September 2023–January 2024. Baseline measures Baseline data collected between August 1-31st 2023 identified 51 patients, of which delirium was screened using the 4-AT tool in 39.2% (n = 20), on average 90 hours post-operatively. The 4-AT was never documented in a consultant-led surgical post-operative review (100%, n = 51). There were clinical concerns of post-operative delirium documented in 7 patients, with the 4-AT documented in 5 of those cases. Intervention Interventions included stakeholder discussions to identify key barriers in the assessment and documentation of delirium, multidisciplinary team education and poster reminders across the ward. These were introduced between November–December 2023. Results Post-intervention results reviewed between 10th-31st January 2024 showed that the 4-AT was used to screen for delirium in 61.9% of patients (n = 13), on average 45 hours post-operatively. The 4-AT was never documented in a consultant-led surgical post-operative review. In addition, 2 patients developed delirium post-operatively with the 4-AT reported in both cases. Conclusions This QIP has demonstrated a marked improvement in compliance with national guidelines on the assessment of delirium, highlighting the impact of multidisciplinary education in improving the perioperative clinical pathway for older people undergoing surgery. Future PDSA cycles will focus on improving the documentation of 4AT in the post-operative surgical review.