Background and Hypothesis: Clozapine is the most effective medicine for treatment-resistant schizophrenia, but is limited by adverse events, including potential QT prolongation which can lead to life-threatening arrhythmias. Studies linking clozapine and corrected QT (QTc) prolongation may overestimate this risk due to high rates of clozapine-associated tachycardia. We investigated whether trough clozapine plasma levels are independently associated with QT prolongation after accounting for heart rate. Study Design: We conducted a retrospective, cross-sectional analysis of inpatients treated with clozapine at a tertiary hospital between 2017 and 2023. Trough clozapine plasma levels, and 12-lead electrocardiograms were extracted from electronic medical records. QT intervals were manually measured and corrected using Bazett's, Fredericia, Hodges' formulae, and the QT nomogram. Multivariable regression and causal mediation were used to test the association between clozapine plasma level, heart rate, and QTc. Study Results: Among 313 patients, Bazett's correction classified 27.5% as having prolonged QTc, whereas only one patient (0.3%) exceeded the at-risk threshold using Fredericia, Hodges, or the QT nomogram. Clozapine plasma level correlated with Bazett's-corrected QT (QTcB) (P = .02), but not after adjustment for heart rate (P = .75). Mediation analysis showed that heart rate significantly mediated the relationship between clozapine plasma level and QTcB intervals (P < .001). Conclusions: Apparent clozapine-induced QTc prolongation is largely an artifact of tachycardia and over-correction by Bazett's formula. The Fredericia and Hodges formulae, and the QT nomogram provide a more reliable assessment of torsadogenic risk and prevent unnecessary discontinuation or dose reductions of clozapine.

Background: Medication harm is a significant healthcare challenge in hospitalised adult patients. Machine learning (ML) approaches offer the potential to improve prediction accuracy for medication harm by capturing complex relationships among clinical risk factors that traditional statistical models may not detect. Objective: To develop and evaluate ML models for predicting medication harm in hospitalised adult patients. Design: ML study involving secondary use of a prospectively collected hospital cohort dataset. Methods: This study used data from 279 adult patients admitted to general medical and geriatric wards of a tertiary hospital, among whom 40 experienced 51 medication harm events. Eight ML models were trained and evaluated for identifying patients at risk of medication harm. Medication harm cases were identified through detailed chart reviews, trigger tools, voluntary incident reporting, and International Classification of Diseases version 10 discharge coding. Data were pre-processed with missing values imputed using median imputation. Ten predictive features were selected using recursive feature elimination and clinical expert opinion. Models were trained using stratified 10-fold cross-validation with an 80/20 train-test split. Class imbalance was addressed using an oversampling approach. Results: A random forest model demonstrated the highest performance, achieving an area under the receiver operating characteristic curve of 0.76, precision of 0.50, recall of 0.62, F1 score of 0.54, accuracy of 0.86, specificity of 0.90, and an area under the precision-recall curve of 0.47. Predictive features of importance included length of stay, depression, dementia, insulin use, number of medications (⩾15), age (⩾65), opioid use, and antibiotic use. Conclusion: This study highlights the potential of ML models to predict medication harm, enabling early identification of high-risk patients for preventive interventions. Interdisciplinary collaboration is essential in developing robust, clinically relevant models that can be used to improve patient safety.

Transition of care (ToC) for patients following acute type 1 myocardial infarction or coronary revascularisation procedures is complex, often resulting in medication-related harm and hospital readmissions. Current ToC models lack comprehensive, system-wide approaches, especially in Australia, and often fail to address patient-centred needs and cultural considerations. This theoretical paper aims to outline a multidimensional implementation science approach that will underpin the design and implementation a pharmacy-led ToC intervention to reduce hospital readmissions and improve medication safety. Informed by Guyatt et al. (2021), the REduce hospital readmissions for high-risk CARDiology patients (RECARD) ToC model, will integrate six key elements: (1) guiding implementation with the Consolidated Framework for Implementation Research (CFIR), including a patient domain; (2) co-design with patients and healthcare professionals; (3) understanding local needs using Bradshaw’s taxonomy; (4) designing an evidence-informed intervention based on health behaviour theories; (5) planning and executing implementation using the Expert Recommendations for Implementing Change (ERIC) and the Behaviour Change Wheel (BCW); and (6) evaluating and sustaining the intervention using the RE-AIM (Reach, Effectiveness, Adoption, Implementation, Maintenance) framework. This approach will involve a pre-post interventional trial across three quaternary Queensland hospitals, informed by focus groups, semi-structured interviews, and Yarning circles to refine the literature-informed intervention design. We present our synthesised RECARD ToC model planned to be delivered as an “Adaptive ToC Pathway,” involving a collaborative team of pharmacists, nurse practitioners, and Indigenous health workers. The intervention will incorporate inpatient and post-discharge activities, tailored to local needs and patient preferences. The implementation plan will address barriers and enablers identified through stakeholder engagement, and RE-AIM will guide the evaluation. Our novel theory-informed approach, integrating CFIR, Bradshaw’s model, ERIC, BCW, and RE-AIM, provides a comprehensive framework for designing and implementing the RECARD ToC model. This will allow comparison of a prospective cohort receiving evidence-informed ToC activities to optimise medication management and patient safety, guided by a risk prediction model to predict unplanned readmissions due to medication harm within 30 days, with a retrospective historical cohort who received usual care. By understanding local needs and engaging stakeholders, this project seeks to create a sustainable and impactful intervention that addresses the gaps in current ToC practices, particularly for Indigenous populations, and enhances patient safety in the Australian healthcare setting. Not applicable.

BACKGROUND:Clozapine is the most effective therapy for treatment-resistant schizophrenia, yet adverse drug reactions (ADRs) limit its use. The concurrent ADR burden in outpatients and its relation to psychotropic polypharmacy, tobacco smoking and measured clozapine exposure has not been well characterised. METHOD:We conducted a retrospective, cross-sectional review of medical records for 360 adults receiving maintenance clozapine at a dedicated outpatient clinic. Adverse drug reactions (ADRs) were ascertained using a standardised patient checklist alongside clinical measures. We used multivariate logistic regression to estimate the association between antipsychotic polypharmacy and the presence of ADRs, and negative binomial regression to quantify its association with ADR burden. We conducted a log-linear model to evaluate dose-concentration compensation in tobacco smokers. RESULTS:At the clinic visit, 89.6% had ⩾1 symptomatic ADR. The most prevalent were metabolic syndrome (71.8%), hypersalivation (64.7%) and tachycardia (61.2%). Antipsychotic augmentation (51.4%) was independently associated with ADRs (adjusted odds ratios [aOR] = 3.38, 95% confidence interval [CI] = 1.41-8.06) and a 28% higher ADR count per person (incidence-rate ratio [IRR] = 1.28, 95% CI = 1.11-1.48). Smokers received higher doses yet had lower plasma concentrations, suggesting incomplete dose compensation for CYP1A2 induction and had higher odds of antipsychotic augmentation (odds ratio [OR] = 2.50, 95% CI = 1.53-4.10). CONCLUSION:In maintenance clozapine care, persistent ADRs were common and were more frequent in patients receiving antipsychotic augmentation. Smokers were under-exposed to clozapine and were more likely to receive antipsychotic augmentation. Services should implement structured ADR surveillance and prioritise therapeutic drug monitoring-guided dose optimisation, particularly in smokers, before considering antipsychotic augmentation.

Transition of care (ToC) for patients with cardiovascular diseases is a complex, high-risk period often leading to medication-related harm (MRH) and hospital readmissions. While specialised programs exist, a gap in individualised medication management services exist for specific cardiology patient populations. Pharmacist-led, interdisciplinary ToC services have demonstrated a positive impact on patient outcomes. The REducing hospital re-admission for high-risk CARDiology patients (RECARD) program aims to address this by co-designing, implementing, and evaluating a new pharmacist-led ToC service in post-acute myocardial infarction (AMI) or cardiac surgery patients. The current study focuses on understanding the experiences and expectations of patients and clinicians to inform the development of this service. This qualitative study utilised semi-structured interviews and focus groups to gather data from patients and hospital and community clinicians aligned with three Queensland hospitals and surrounding primary and community care settings. Patients who had experienced an AMI or cardiac surgery in the past three months were interviewed via telephone or Microsoft TEAMS. Clinicians participated in focus groups or individual interviews. Data were audio-recorded, transcribed, and analysed using the six-phase thematic analysis method by Braun and Clarke, guided by Bradshaw's model of need to understand stakeholders’ expressed and comparative needs. Data were obtained from 13 patient interviews and 40 clinicians, through seven focus groups and one interview. Three main themes, with associated subthemes, emerged from the data; Patient-level issues, System and process issues and Interprofessional collaboration. Patients described information overload, medication uncertainty, and anxiety at discharge, while clinicians identified delayed discharge summaries, fragmented communication with primary care, and service gaps - particularly impacting rural patients- as key risks for medication‑related harm. The study's findings highlight the critical need for a patient-centred, pharmacist-led interdisciplinary ToC service that addresses patient knowledge deficits, psychosocial barriers, and the communication gaps between hospital and community care settings. The results from this study will inform the development of the RECARD program's "Adaptive ToC Pathway," ensuring it is tailored to meet the specific needs and expectations of both patients and clinicians, ultimately aiming to reduce MRH and hospital readmissions.

BACKGROUND:The implementation of new services in healthcare is complex, and an effective strategy is required to produce the desired patient outcomes. This paper aims to describe the implementation of a pharmacist-led post-discharge medication review clinic and outline the clinical pharmacist activities provided during the clinic review. METHOD:The Knowledge to Action (KTA) framework was used to map the actions involved for the planning, implementation and review of outcomes of the clinic. The Consolidated Framework for Implementation Research (CFIR) was used to describe dependent and independent barriers and facilitators encountered throughout the implementation process. The clinical pharmacist activities provided during the clinic review were defined according to local guidelines. RESULTS:The KTA framework provided the key steps to implement the clinic service, monitor patient outcomes, that included 30-day hospital readmissions, and review clinic processes. Follow-up and troubleshooting of patient non-attendance and referral pathways were a key component of utilising the circular nature of the KTA framework. Facilitators included hospital funding models, growing evidence for pharmacist-led medication review and support from key stakeholders. Barriers included running costs, patient understanding of the service, attendance, and technology constraints. The clinic review included five key clinical pharmacist activities: confirmation of a best possible medication history, medication reconciliation, medication review, patient education, and communication of medication related problems identified with the patient's healthcare team. CONCLUSION:The KTA framework was a useful tool for implementing the pharmacist-led post-discharge medication review clinic, adapting and monitoring processes and improving knowledge of key stakeholders. Identified barriers and facilitators are relevant to implementing future clinic models.

BACKGROUND:Although general anesthesia is commonly utilized for endovascular thrombectomy for acute ischemic stroke, whether anesthetic agents affect clinical outcome is unknown. Retrospective studies comparing propofol and volatile agents have shown mixed results. A randomized controlled trial is needed to determine whether anesthetic agent affects clinical outcomes. This pilot study aimed to evaluate the feasibility of conducting a definitive randomized controlled trial comparing propofol and sevoflurane-based anesthesia in patients undergoing endovascular thrombectomy. METHODS:Patients booked to undergo endovascular thrombectomy were randomized to receive either propofol or sevoflurane-based general anesthesia. Feasibility outcomes assessed were recruitment rate, protocol adherence, and data completeness. Secondary outcomes included functional recovery (90-d modified Rankin Scale 0 to 2), mortality, early neurological improvement, blood pressure control intraoperatively and postoperatively, and adverse events. We also assessed for independent risk factors for functional recovery and death. RESULTS:Of 201 eligible patients, 93 (46.3%) were enrolled and 73 (36.3%) included in the final analysis. The consent and randomization model was challenging. Adherence to drug protocols was 94.5%. Data completion rate was 99%. There were no differences in secondary outcomes between groups. Mortality was associated with higher admission National Institutes of Health Stroke Scale. Higher 90-day modified Rankin scores were associated with higher systolic blood pressures pre-reperfusion ( r =0.32, P <0.01) and post-reperfusion ( r =27, P =0.03). CONCLUSIONS:A definitive randomized controlled trial of propofol and sevoflurane-based anesthesia is feasible. Future studies would benefit from adapting the trial model to better integrate research into the clinical workflow. TRIAL REGISTRATION:Australian New Zealand Clinical Trials Registry (ACTRN12621000074897), January 29, 2021.

Aspirin resistance (AR) is linked to increased morbidity and mortality after cardiovascular and neurovascular procedures but has not been investigated after total hip arthroplasty (THA) and total knee arthroplasty (TKA). To investigate the incidence of AR in high-risk patients after elective THA/TKA, we conducted a prospective cohort study of elective THA/TKA in patients with thromboembolic risk factors: obesity (body mass index [BMI] ≥30 kg/m2), diabetes and/or advanced age (≥65 years) who were administered aspirin for venous thromboembolism prophylaxis. AR was confirmed using a platelet function analyser. Forty patients were included with a mean±standard deviation age of 66.1±10.4 years and BMI of 32.2±6.0 kg/m2 and eight (20.0%) had diabetes. Overall, 52.5% of patients were aspirin resistant, with no statistically significant relationships between patient variables and AR (P > 0.05). AR after THA/TKA was an observed phenomenon; however, larger trials are required to determine the clinical consequences of AR and to guide prophylaxis strategies in this population.

Background: The transition from hospital discharge to primary care is a critical period in a patient's healthcare journey. Health system errors occur, due to a breakdown in communication or lack of structured planning which can lead to medication related harm or hospital readmission. At a quaternary referral hospital in Australia, pharmacists refer Internal Medicine patients to a pharmacist-led clinic for post-discharge medication review. While clinical resources exist to guide identification of at-risk patients, it remains unclear if and to what extent, pharmacists incorporate these criteria into their referral. Aim: To determine the criteria and reasons used by pharmacists to refer Internal Medicine patients to a post discharge pharmacist review clinic. Methods: Semi-structured interviews were conducted with hospital pharmacists who had worked in Internal Medicine and previously referred patients to the post discharge review clinic. Interviews were conducted until data saturation was obtained. Interviews were audio recorded, transcribed and coded using NVivo®. Themes and subthemes were identified through inductive thematic analysis and finalised via discussion within the research team. Results: Eleven pharmacists were interviewed. Five themes emerged describing referral criteria and reasons: (1) medication criteria including the use of high-risk medications and adjustments; (2) patient criteria including health status, frailty and social aspects of health including carer supports; (3) system pressures including patient flow and time constraints in care delivery; (4) post-discharge care including medication liaison and evaluation of tolerability and; (5) clinical judgement described as “worry” about the patient, highlighting the role of clinical reasoning. Conclusion: Pharmacists used established criteria from clinical resources to identify high-risk patients for referral; however, they also relied on clinical judgement. Referrals aimed to prevent medication related harm and improve communication with patients and healthcare providers. Future research should evaluate the effectiveness of clinical judgement to ensure high-risk patients are identified for transition of care services.

AIMS:To assess whether the AIME-Frail tool assists in medication management prioritisation and reduces inpatient medication harm events, evaluate tool implementation challenges and enablers, and identify predictive risk factors for medication harm. METHODS:General and geriatric medicine patients at a tertiary hospital in Queensland, Australia were enrolled in a controlled study. Medication harm was identified through electronic medical record (EMR) reviews, a trigger tool, and discussions with treating teams. In the intervention group, pharmacists used the AIME-Frail tool for risk-based prioritisation of medication reviews, while the control group received usual care. Incidence and types of medication harm were compared between groups. Reflexive journaling was used to document insights on tool use by pharmacists, identifying barriers and enablers for implementation. Predictive risk factors were identified using regression models. RESULTS:A total of 279 patients participated. Medication harm occurred in 51 patients (18.3%). The most common type was gastrointestinal harm, with opioids and antibacterials causing constipation and nausea/vomiting, while anticoagulants caused bleeding and bruising. There was no difference between intervention and control groups in incidence of medication harm 23/142 (16.2%) versus 28/137 (20.4%) respectively (P = 0.44). Key implementation challenges included lack of integration into the EMR and suboptimal compliance with tool use by pharmacists. Predictive risk factors for medication harm included renal impairment, dementia, depression, and longer hospital stay (P ≤ 0.05). CONCLUSION:While the AIME-Frail tool identified high-risk patients, it did not significantly reduce medication harm, possibly due to implementation challenges. Future studies should aim to update the model using predictive risk factors and optimise the usability of the tool for pharmacists.

Background Drug–drug interaction (DDI) alerts in electronic systems are frequently implemented to minimize the occurrence of preventable DDIs. While prescribers recognize the potential benefits of DDI alerts, a large proportion are overridden by users. Objectives This study aimed to explore and compare prescribers' and managers' perspectives of DDI alerts. Methods A qualitative descriptive study was conducted across six hospitals in Australia with end users (prescribers who receive alerts) [n = 14] and managers [n = 20] (senior staff in roles relevant to alert system implementation and management). End users were asked to reflect on alert usefulness, benefits, risks, and impacts. Managers were asked what they thought of alerts, and about any feedback they had received from frontline clinicians. Key themes were extracted via an inductive content analysis approach and deductively mapped to the Technology Acceptance Model (TAM3). Comparisons of the views held toward the alerts were made between the two participant groups. Results End users predominantly reflected on the utility of the DDI alert system (i.e. how useful it was to their role), less on how easy the system was to use, and mainly focused on the negative consequences of alerts. Managers believed the benefits of DDI alerts are primarily experienced by junior doctors. While end users suggested that alerts should be tailored to the patient's clinical scenario, managers called for DDI alerts to be tailored to the prescriber (seniority and specialty). Conclusion Interviews with end users and managers uncovered a number of perceived benefits and limitations of DDI alerts, primarily related to the system's usefulness. While largely consistent, some perceptions were different between end users and managers, particularly in the types of benefits, and how they conceptualized potential tailoring to improve DDI alerts. Our findings point to a need for user participation in the development, deployment, and improvement of alerts to promote consideration and effectiveness of DDI alerts.

Patients transitioning from secondary to primary healthcare are at increased risk of medication errors, adverse drug events and readmission to hospital. Incorporating a post-discharge follow-up by a hospital pharmacist has been proposed as a potential strategy to reduce readmissions. To determine the impact of a hospital-based pharmacist-led post-discharge medication review clinic on 30-day hospital readmissions in adult patients. A single-site, retrospective cohort study compared the medical records of patients who attended the Pharmacist Review and EValuation of Existing and New Therapies (PREVENT) clinic between 1 January 2018 and 31 December 2019 to a group of case-matched control patients who did not attend the clinic. Patient inclusion criteria comprised those 18 years and older and attended the PREVENT clinic within 30 days of discharge. The matched group was based on gender, age and hospital metrics. The primary outcome measure is unplanned, all-cause 30-day hospital readmission. There were 170 patients per group, with similar baseline characteristics. There were significantly less unplanned all-cause 30-day hospital readmissions in the PREVENT clinic group (n = 12 (7.1

Introduction: The contribution of medication harm to rehospitalisation and adverse patient outcomes after an acute myocardial infarction (AMI) needs exploration. Rehospitalisation is costly to both patients and the healthcare facility. Following an AMI, patients are at risk of medication harm as they are often older and have multiple comorbidities and polypharmacy. This study aimed to quantify and evaluate medication harm causing unplanned rehospitalisation after an AMI. Methods: This was a retrospective cohort study of patients discharged from a quaternary hospital post-AMI. All rehospitalisations within 18 months were identified using medical record review and coding data. The primary outcome measure was medication harm rehospitalisation. Preventability, causality, and severity assessments of medication harm were conducted. Results: A total of 1,564 patients experienced an AMI, and 415 (26.5%) were rehospitalised. Eighty-nine patients (5.7% of total population; 6.0% of those discharged) experienced a total of 101 medication harm events. Those with medication harm were older (p = 0.007) and had higher rates of heart failure (p = 0.005), chronic kidney disease (p = 0.046), chronic obstructive pulmonary disease (p = 0.037), and a prior history of ischaemic heart disease (p = 0.005). Gastrointestinal bleeding, acute kidney injury, and hypotension were the most common medication harm events. Forty percent of events were avoidable, and 84% were classed as "serious." Furosemide, antiplatelets, and angiotensin-converting enzyme inhibitors were the most commonly implicated medications. The median time to medication harm rehospitalisation was 79 days (interquartile range: 16-200 days). Conclusion: Medication harm causes unplanned rehospitalisation in 5.7% of all AMI patients (1 in 17 patients; 6.0% of those discharged). The majority of harm was serious and occurred within the first 200 days of discharge. This study highlights that measures to attenuate the risk of medication harm rehospitalisation are essential, including post-discharge medication management.

Background: Medication harm affects between 5 and 15% of hospitalised patients, with approximately half of the harm events considered preventable through timely intervention. The Adverse Inpatient Medication Event (AIME) risk prediction model was previously developed to guide a systematic approach to patient prioritisation for targeted clinician review, but frailty was not tested as a candidate predictor variable. Aim: To evaluate the predictive performance of an updated AIME model, incorporating a measure of frailty, when applied to a new multisite cohort of hospitalised adult inpatients. Methods: A retrospective cohort study was conducted at two tertiary Australian hospitals on patients discharged between 1st January and April 31, 2020. Data were extracted from electronic medical records (EMRs) and clinical coding databases. Medication harm was identified using ICD-10 Y -codes and confirmed by senior pharmacist review of medical records. The Hospital Frailty Risk Score (HFRS) was calculated for each patient. Logistic regression analysis was used to construct a modified AIME model. Candidate variables of the original AIME model, together with new variables including HFRS were tested. Performance of the final model was reported using area under the curve (AUC) and decision curve analysis (DCA). Results: A total of 4089 patient admissions were included, with a mean age +/- standard deviation (SD) of 64 years (+/- 19 years), 2050 patients (50%) were males, and mean HFRS was 6.2 (+/- 5.9). 184 patients (4.5%) experienced one or more medication harm events during hospitalisation. The new AIME-Frail risk model incorporated 5 of the original variables: length of stay (LOS), anti -psychotics, antiarrhythmics, immunosuppressants, and INR greater than 3, as well as 5 new variables: HFRS, anticoagulants, antibiotics, insulin, and opioid use. The AUC was 0.79 (95% CI: 0.76-0.83) which was superior to the original model (AUC = 0.70, 95% CI: 0.65-0.74) with a sensitivity of 69%, specificity of 81%, positive predictive value of 0.14 (95% CI: 0.10-0.17) and negative predictive value of 0.98 (95% CI: 0.97-0.99). The DCA identified the model as having potential clinical utility between the probability thresholds of 0.05-0.4. Conclusion: The inclusion of a frailty measure improved the predictive performance of the AIME model. Screening inpatients using the AIME-Frail tool could identify more patients at high -risk of medication harm who warrant timely clinician review.

BACKGROUND:We sought to determine the impact of the presence of a pharmacist on medication and patient related outcomes during the emergency management of critically ill patients requiring resuscitation or medical emergency response team care in a hospital setting.METHODS:We conducted a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. A literature search of databases from January 1995 to April 2023 was conducted to identify studies of contemporary pharmacist practice. Results were extracted and analysed for included studies, those evaluating the impact of the presence of a pharmacist on medication and patient related outcomes during the emergency management of critically ill hospitalised patients requiring resuscitation or medical emergency response team care. To determine risk of bias, the Newcastle-Ottowa Quality Assessment scale was used for non-randomised studies and the Revised Cochrane risk-of-bias tool for randomised trials.RESULTS:Of 1345 studies identified, 54 were selected for full text review, and 30 were included in the final analysis. There were 29 cohort studies and one randomised controlled trial. The studies reported the impact of a pharmacist for a variety of patient presentations. The study team assigned each study to one of eight patient cohorts: acute stroke, cardiac arrest, rapid response calls, S-T segment elevation myocardial infarction, acute haemorrhage, major trauma resuscitation, sepsis and status epilepticus. The most frequently reported outcome, associated with a statistically significant benefit in 23 studies, was time to medication administration. Few studies reported a significant difference in patient outcome measures such as mortality. Only 8 of the 30 studies were assessed to have a low risk of bias.CONCLUSIONS:The results of this systematic review provide support for a beneficial impact of a pharmacist presence and intervention during resuscitation or medical emergency response team care, with significant improvements in outcomes such as time to initiation of time-critical medications, medication appropriateness and guideline compliance. However, studies were predominantly small and retrospective and were not powered to detect differences in patient related measures such as length of stay and mortality. Future research should investigate the clinical impacts of the pharmacist in ED resuscitation settings in controlled, prospective studies with robust sampling methods.

INTRODUCTION:Aspirin is used for venous thromboembolism (VTE) prophylaxis after total hip and knee arthroplasty (THA/TKA). However, its efficacy is unclear in patients with multiple VTE risk factors and at risk of aspirin resistance (AR). BACKGROUND AND AIMS:To determine the prevalence of risk factors for VTE and AR in patients after THA/TKA and to determine the relationship between risk factors and drugs prescribed for thromboprophylaxis. METHODS:A retrospective cohort study of elective-THA/TKA in six Australian hospitals over a 1-year period. Medical records were manually reviewed to determine demographics, thromboprophylaxis regimen and presence of risk factors. The relationship between individual and cumulative risk factors with the thromboprophylaxis regimen was determined. RESULTS:In total, 1011 patients were included with a mean (SD) age of 65.9 (±11.0) years, and 56.4% were female. The five most prevalent risk factors were obesity (59.1%), age ≥65 years (58.2%), hypertension (45.3%), dyslipidaemia (35.9%) and diabetes (19.7%). Most patients had ≥1 risk factor for VTE (93.6%) and AR (93.6%), with 49.0% and 35.0% having ≥3 concurrent VTE and AR risk factors, respectively. The only significant relationship between risk factors and drugs was diabetes (P < 0.01). Rivaroxaban was more commonly used as the number of concurrent VTE risk factors increased (P < 0.05). CONCLUSION:Patients had a high prevalence of VTE and AR risk factors, suggesting aspirin may not be beneficial in many patients. Only diabetes was linked to the selection of thromboprophylaxis. Patients who received rivaroxaban had a greater average number of VTE risk factors. Guidelines should promote individualised prescribing in higher-risk patients.

Objective:To describe development and application of a checklist of criteria for selecting an automated machine learning (Auto ML) platform for use in creating clinical ML models.Materials and Methods:Evaluation criteria for selecting an Auto ML platform suited to ML needs of a local health district were developed in 3 steps: (1) identification of key requirements, (2) a market scan, and (3) an assessment process with desired outcomes.Results:The final checklist comprising 21 functional and 6 non-functional criteria was applied to vendor submissions in selecting a platform for creating a ML heparin dosing model as a use case.Discussion:A team of clinicians, data scientists, and key stakeholders developed a checklist which can be adapted to ML needs of healthcare organizations, the use case providing a relevant example.Conclusion:An evaluative checklist was developed for selecting Auto ML platforms which requires validation in larger multi-site studies.

Background Clinical pharmacy quality indicators are often non-uniform and measure individual activities not linked to outcomes. Aim To define a consensus agreed pharmaceutical care bundle and patient outcome measures across an entire state health service. Method A four-round modified-Delphi approach with state Directors of Pharmacy was performed (n = 25). They were asked to rate on a 5-point Likert scale the relevance and measurability of 32 inpatient clinical pharmacy quality indicators and outcome measures. They also ranked clinical pharmacy activities in order from perceived most to least beneficial. Based upon these results, pharmaceutical care bundles consisting of multiple clinical pharmacy activities were formed, and relevance and measurability assessed. Results Response rate ranged from 40 to 60%. Twenty-six individual clinical pharmacy quality indicators reached consensus. The top ranked clinical pharmacy quality indicator was ‘proportion of patients where a pharmacist documents an accurate list of medicines during admission’. There were nine pharmaceutical care bundles formed consisting between 3 and 7 activities. Only one pharmaceutical care bundle reached consensus: medication history, adverse drug reaction/allergy documentation, admission and discharge medication reconciliation, medication review, provision of medicines education and provision of a medication list on discharge. Sixteen outcome measures reached consensus. The top ranked were hospital acquired complications, readmission due to medication misadventure and unplanned readmission within 10 days. Conclusion Consensus has been reached on one pharmaceutical care bundle and sixteen outcomes to monitor clinical pharmacy service delivery. The next step is to measure the extent of pharmaceutical care bundle delivery and the link to patient outcomes.

Clinical pharmacists perform activities to optimise medicines use and prevent patient harm. Historically, clinical pharmacy quality indicators have measured individual activities not linked to patient outcomes. To determine the proportion of patients who receive a pharmaceutical care bundle (PCB) (consisting of a medication history, medication review, discharge medication list and medicines information on the discharge summary) as well as investigate the relationship between delivery of this PCB and patient outcomes. Pharmaceutical care bundle activities were defined within state-wide (Queensland, Australia) clinical information systems and datasets were linked. An observational study using routinely recorded data was performed at ten participating sites for adult patients who had a non-same day hospital stay. The association between extent of PCB delivery and three patient outcomes were investigated: length of stay (LOS), unplanned readmission, and mortality. In total 283,813 patient hospital stays were evaluated. The delivery of the PCB occurred in 26.9