Development and validation of a machine learning model to improve precision prediction for irrational prescriptions in orthopedic perioperative patients

OBJECTIVE

Our objective was to develop a machine learning model capable of predicting irrational medical prescriptions precisely within orthopedic perioperative patients.

METHODS

A dataset comprising 3047 instances of suspected irrational medication prescriptions was collected from a sample of 1318 orthopedic perioperative patients from April 2019 to March 2022. Four machine learning models were employed to forecast irrational prescriptions, following which, the performance of each model was meticulously assessed. Subsequently, a thorough variable importance analysis was conducted on the model that performed the best predictive capabilities. Thereafter, the efficacy of integrating this optimal model into the existing audit prescription process was rigorously evaluated.

RESULTS

Of the models utilized in this study, the RF model yielded the highest AUC of 92%, whereas the NB model presented the lowest AUC of 68%. Also, the RF model boasted the most robust performance in terms of PPV, reaching 82.4%, and NPV, reaching 86.6%. The ANN and the XGBoost model were neck and neck, with the ANN slightly edging out with a higher PPV of 95.9%, while the XGBoost model boasted an impressive NPV of 98.2%. The RF model singled out the following five factors as the most influential in predicting irrational prescriptions: the type of drug, the type of surgery, the number of comorbidities, the date of surgery after hospitalization, as well as the associated hospital and drug costs.

CONCLUSION

The RF model showcased significantly high level of proficiency in predicting irrational prescriptions among orthopedic perioperative patients, outperforming other models by a considerable margin. It effectively enhanced the efficiency of pharmacist interventions, displaying outstanding performance in assisting pharmacists to intervene with irrational prescriptions.

Unplanned Rehospitalisation due to Medication Harm following an Acute Myocardial Infarction

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.

Cost-effectiveness of medication reconciliation performed by a pharmacist in pediatrics of a hospital: A randomized clinical trial protocol linked to a pharmacoeconomic study

INTRODUCTION

Each patient admitted to the hospital is subject to one medication error per day, since the occurrence of this one with the potential to cause harm is three times more common in pediatric hospitalized patients than in adults. These harms can result from inaccurate or incomplete drug use histories when patients undergo a clinical evaluation, which jeopardizes patient safety and compromises hospitalization costs. Thus, medication reconciliation (MC) emerges as a possible solution to avoid the occurrence of these in pediatric patients and directly contributes to reducing costs in the hospital environment and increasing quality of life). Therefore, this study proposes to determine whether pharmacist-led medication reconciliation is a cost-effective strategy to improve health outcomes in pediatric patients.

METHODS

A randomized clinical trial will be carried out, over eight months, to carry out the cost analysis. Micro-costing pharmacoeconomic model through a questionnaire and clinical interview to collect the variables necessary for the study and comparison of the control and intervention groups. Participants in this study will be children aged 0 days to 12 years, admitted to the hospital. The perspective adopted will be that of the hospital. To assess the economic outcomes of MC, the cost-effect pairs will be categorized and visually represented in the cost-effectiveness plan to compare the intervention and control groups. Monte Carlo simulation and univariate sensitivity analysis will be performed to test the robustness of the findings.

ETHICS AND DISSEMINATION

The clinical trial was approved by the Research Ethics Committee of the Federal University of Sergipe (CAAE: 19625319.6.0000.5546 and opinion number: 3,630,579). This protocol fully adhered to the recommendations of the 2010 CONSORT Declaration and was registered in the Brazilian Registry of Clinical Trials (ReBEC): RBR-25dnqsk.

Development and validation of a ready-to-use score to prioritise medication reconciliation at patient admission in an orthopaedic and trauma department

OBJECTIVE

Medication reconciliation (MR) is recognised as an important tool in preventing medication errors such as unintentional discrepancies (UDs). The aim of this study was to identify independent predictive factors of UDs during MR at patient admission to an orthopaedic and trauma department. The secondary objective was to build and validate a ready-to-use score to prioritise patients.

METHOD

A retrospective study was performed on 3.5 years of pharmacist-led MR in the orthopaedic and trauma department of a large university teaching hospital. Independent predictors of UD were identified by multivariable logistic regression. A priority score to identify patients at risk of at least one UD was constructed from the odds ratios of the risk factors, and validated in a separate cohort. Performance was assessed with sensitivity, specificity, C-statistic and Hosmer-Lemeshow goodness-of-fit.

RESULTS

In total, 888 patients were included and 387 UDs were identified, mainly drug omissions (65.1%). Five independent predictors of UD were identified: age >75 years (OR 2.05, 95% CI 1.41 to 3.00; p<0.001), admission during school holidays (OR 1.69, 95% CI 1.17 to 2.44; p=0.005), female gender (OR 2.20, 95% CI 1.53 to 3.16; p<0.001), emergency hospitalisation (OR 2.05, 95% CI 1.45 to 2.92; p<0.001), and ≥5 medications on the best possible medication history (BPMH) (OR 3.29, 95% CI 2.20 to 4.94; p<0.001). Based on these predictors, a priority score ranging from 0 to 10 was built and internally and externally validated (C statistic 0.72, 95% CI 0.67 to 0.76).

CONCLUSIONS

This study confirms the high prevalence of UD in patients admitted to orthopaedic and trauma surgery departments. Five independent predictive factors of UD during MR were identified (female gender, emergency hospitalisation, hospitalisation during school holidays, age ≥75 years, and ≥5 medicines on the BPMH). The developed risk score will help to prioritise MR among patients at risk of medication error and is ready-to-use in other orthopaedic and trauma departments.

Accuracy of medication histories derived from an Australian cloud-based repository of prescribed and dispensed medication records

BACKGROUND

Obtaining accurate medication histories at transitions of care is challenging, but important for patient safety. Prescription exchange services (PES) securely transfer electronic prescription and dispensing records between prescribers and pharmacies; potentially useful data for determining medication histories.

AIM

To evaluate the accuracy of PES-derived medication histories.

METHODS

Prospective observational study, at two Australian tertiary-referral health-services. A convenience sample of adult inpatients was recruited. The main outcome measure was: proportion of patients with ≥1 errors in their PES-derived pre-admission medication histories, compared to gold-standard best-possible medication histories, including prescribed and non-prescribed medications, obtained by pharmacists using multiple sources including patient/carer interview.

RESULTS

153/154 (99.4%) patients (median age 76years, inter-quartile range [IQR] 64-84years, median 10.0 pre-admission medications, IQR 6.0-14.0) had ≥1 errors in their PES-derived medication history (median 6.0 per patient, IQR 4.0-9.0). Excluding when-required (PRN) medications, 146 (94.8%) patients had a median of 4.0 errors (IQR 2.0-6.0). Omission was the most common error, affecting 549/1648 (33.3%) current medications (median 3.0, IQR 1.0-5.0 per patient); 396 [72.1%] omissions were over-the-counter medicines. Dose-regimen errors affected 276/1099 (25.1%) current medications captured in PES-derived medication histories (median 1.0, IQR 0.0-3.0 per patient). Commission errors (medications in PES-derived histories that weren't current) affected 224/1383 (16.2%) medications (median 1.0, IQR 1.0-2.0 per patient).

CONCLUSIONS

Medication histories derived solely from a cloud-based medication record repository had a high error rate compared to patients' actual medication use. Like all medication history sources, data from cloud-based repositories need to be verified with additional sources including patients and/or carers. This article is protected by copyright. All rights reserved.

Pharmacist-led educational interventions provided to healthcare providers to reduce medication errors: A systematic review and meta-analysis

INTRODUCTION

Medication errors are avoidable events that can occur at any stage of the medication use process. They are widespread in healthcare systems and are linked to an increased risk of morbidity and mortality. Several strategies have been studied to reduce their occurrence including different types of pharmacy-based interventions. One of the main pharmacist-led interventions is educational programs, which seem to have promising benefits.

OBJECTIVE

To describe and compare various pharmacist-led educational interventions delivered to healthcare providers and to evaluate their impact qualitatively and quantitatively on medication error rates.

METHODS

A systematic review and meta-analysis was conducted through searching Cochrane Library, EBSCO, EMBASE, Medline and Google Scholar from inception to June 2020. Only interventional studies that reported medication error rate change after the intervention were included. Two independent authors worked through the data extraction and quality assessment using Crowe Critical Appraisal Tool (CCAT). Summary odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using a random-effects model for rates of medication errors. Research protocol is available in The International Prospective Register of Systematic Reviews (PROSPERO) under the registration number CRD42019116465.

RESULTS

Twelve studies involving 115058 participants were included. The two main recipients of the educational interventions were nurses and resident physicians. Educational programs involved lectures, posters, practical teaching sessions, audit and feedback method and flash cards of high-risk abbreviations. All studies included educational sessions as part of their program, either alone or in combination with other approaches, and most studies used errors encountered before implementing the intervention to inform the content of these sessions. Educational programs led by a pharmacist were associated with significant reductions in the overall rate of medication errors occurrence (OR, 0.38; 95% CI, 0.22 to 0.65).

CONCLUSION

Pharmacist-led educational interventions directed to healthcare providers are effective at reducing medication error rates. This review supports the implementation of pharmacist-led educational intervention aimed at reducing medication errors.

Improving Medication Reconciliation with Comprehensive Evaluation at a Veterans Affairs Skilled Nursing Facility

BACKGROUND

Unintentional medication discrepancies due to inadequate medication reconciliation pose a threat to patient safety. Skilled nursing facilities (SNFs) are an important care setting where patients are vulnerable to unintentional medication discrepancies due to increased medical complexity and care transitions. This study describes a quality improvement (QI) approach to improve medication reconciliation in an SNF setting as part of the Multi-Center Medication Reconciliation Quality Improvement Study 2 (MARQUIS2).

METHODS

This study was conducted at a 112-bed US Department of Veterans Affairs SNF. The researchers used several QI methods, including data benchmarking, stakeholder surveys, process mapping, and a Healthcare Failure Mode and Effect Analysis (HFMEA) to complete comprehensive baseline assessments.

RESULTS

Baseline assessments revealed that medication reconciliation processes were error-prone, with high rates of medication discrepancies. Provider surveys and process mapping revealed extremely labor-intensive and highly complex processes lacking standardization. Factors contributing were polypharmacy, limited resources, electronic health record limitations, and patient exposure to multiple care transitions. HFMEA enabled a methodical approach to identify and address challenges. The team validated the best possible medication history (BPMH) process for hospital settings as outlined by MARQUIS2 for the SNF setting and found it necessary to use additional medication lists to account for multiple care transitions.

CONCLUSION

SNFs represent a critical setting for medication reconciliation efforts due to challenges completing the reconciliation process and the concomitant high risk of adverse drug events in this population. Initial baseline assessments effectively identified existing problems and can be used to guide targeted interventions.

[Medical reconciliation for hospitalized patients in orthopedic surgery department: Return of experience over 2 years of practice]

OBJECTIVES

Some medication errors can be prevented by pharmacist action such as medication reconciliation. The main objective of this study was to evaluate the medication reconciliation activity after two years of practice. The secondary objective was to assess the medical staff's satisfaction following the setting up of the activity.

METHODS

This retrospective study was realized over a period of two years in our hospital. Patients meeting the following criteria were included: 65 years and over, hospitalized in orthopedic surgery department, preferentially after a discharge of the emergency room. After the best possible medication history was established, it is compared to medicines ordered. The discrepancies were defined as intended or unintended. Study data were collected and analyzed using Excel and SPSS statistics®.

RESULTS

A total of 899 patients met the inclusion criteria during the study period, mean age was 78 years (27; 104). A total of 84 % of our cohort was admitted after a discharge of the emergency room. Seventy five percent of the population had at least an unintended discrepancie, a mean of 2,3 unintended discrepancies per patient was identified. Seventy five percent of the unintended discrepancy were discussed and resolved. The medical staff was mostly satisfied of the activity.

CONCLUSION

The medication reconciliation secured the drug management of hospitalized patients.

Impact of a Unit-Based Clinical Pharmacist on Communication of Medication Information in an Orthopedic Hospital

BACKGROUND

Medication management, a complex yet essential part of patient care, requires that clinicians and patients understand indication, dosage, frequency, and adverse effects in order to maximize benefits and minimize errors, as well as to transition patients from hospital to home. Clinical pharmacists improve care transitions and safety by interacting with patients, prescribers, and nurses on medication management and self-care. However, little is known on the use of clinical pharmacists on interdisciplinary teams at the unit level within orthopedics.

QUESTIONS/PURPOSES

This study sought to measure the impact of unit-based clinical pharmacists on patient perceptions of communication specific to medication during hospitalization at an orthopedic specialty hospital and on the frequency of medication errors.

METHODS

A retrospective, quasi-experimental, two-group evaluative design with nonequivalent controls was used. Consumer Assessment of Healthcare Providers and Systems (HCAHPS) data on six inpatient units was analyzed 6 months before and 6 months after assignment of clinical pharmacists to half these units. Data specific to questions that measure quality of communication as well as medication understanding were analyzed. Additionally, data on medication error frequency were collected and compared between units with and without clinical pharmacists.

RESULTS

A total of 2022 surveys were analyzed. The percentage of patients who reported receipt of medication information and medication understanding increased significantly after the implementation of unit-based clinical pharmacists. Comparison of intervention and non-intervention groups showed no statistically significant difference in the frequency of medication errors.

CONCLUSION

Results suggest that a clinical pharmacist assigned to an inpatient unit in orthopedics significantly influences patient perceptions of communication about and understanding of their medications.

Clinical Pharmacist Service Associated With Improved Outcomes and Cost Savings in Total Joint Arthroplasty

BACKGROUND

Institutions providing total joint arthroplasty (TJA) procedures are subject to substantial outcomes reporting, including those influencing payment for services. Although clinical pharmacists are well-poised to add value, a comprehensive approach to optimizing pharmacotherapy across the care continuum for TJA patients has not been described.

METHODS

This prospective, interventional, sequential cohort study was approved by our Institutional Review Board. The objective was to assess the impact of an Orthopedic Clinical Pharmacist service on institutional TJA complication rates and costs. Outcomes were compared for a Baseline period of July 2015 to February 2016 and a Post-implementation period of September 2016 to February 2017, allowing for a 6-month run-in period. Additionally, we pursued a post-discharge, RN-administered patient survey and an exploratory economic assessment.

RESULTS

A total of 1715 TJA procedures were performed at the institution during the 20-month study timeframe. Postoperative readmission rate (1.3% vs 4.8%, P = .002) and complication rate (1.8% vs 3.4%, P = .760) were lower in the Post-implementation period. Postoperative VTE rate decreased to zero in the Post-implementation period (0.0% vs 0.6%, P = .13) and average hospital length of stay was unchanged (2.8 vs 2.9 days). Patient self-rated understanding of discharge medications was improved and satisfaction with pharmacist interaction was very high. The service conferred an estimated $73,410 net annual cost savings to the institution.

CONCLUSION

Integration of a clinical pharmacist service for TJA patients was associated with clinically meaningful improvements in institutional outcomes, likely conferring substantial cost-benefit.

Impact of a preoperative pharmaceutical consultation in scheduled orthopedic surgery on admission: a prospective observational study

BACKGROUND

Medication errors have a high prevalence in surgery and management of home medication is strongly involved in these errors. In scheduled surgery, the preoperative consultation is a privileged time to inform the patient about the management of her/his home medication before admission. This study assessed the impact of a pre-anesthesia best possible medication history (PA-BPMH) on admission. The PA-BPMH was performed by a clinical pharmacist prior to the anesthesia consultation for anesthesiologists to prescribe admission medical orders for scheduled orthopedic surgery patients.

METHODS

This was a prospective observational study which was carried out in an orthopedic surgery department. All patients over 18 years old with an elective orthopedic surgery were eligible except ambulatory surgery patients. The pharmacist registered the PA-BPMH into the software making it available for anesthesiologists for the pre-admission medication order. Finally, a medication reconciliation was performed at admission. The main outcome was the percentage of patients with at least one unintended medication discrepancy (UMD) at admission. The nature, potential clinical impact and acceptance rate of each UMD detected were assessed. Also, the PA-BPMH process was described and patients and anesthesiologists satisfaction was evaluated.

RESULTS

A total of 455 patients had a pharmaceutical consultation. Medication reconciliation was performed at admission for 360 patients. Overall, at least one UMD was observed in 13.0% of patients (n = 47). A total of 63 UMD were detected. The most common type of UMD was omission (25.4%) and incorrect drug (23.8%).Two UMD (3.2%) were evaluated as life threatening. All the UMD detected were corrected on the admission medication order.

CONCLUSION

A preoperative pharmacist-anesthesiologist teamwork seems to improve the safety of perioperative management of home medication for scheduled orthopedic surgery patients. This process needs a randomized clinical trial across a wider range of surgeries before its implementation.

Impact of pharmaceutical care in the orthopaedic department

WHAT IS KNOWN AND OBJECTIVE

In the mid-1960s, clinical pharmacy developed in the USA, and as the demand for pharmaceutical services continued to grow, their impact began to be taken seriously. However, the participation of clinical pharmacists as members of the multidisciplinary team in the orthopaedic department is still in its infancy, although its role in orthopaedics has not been defined. The object of this study was to identify and discuss the impact of pharmaceutical care in the orthopaedic department.

METHODS

A literature search was conducted on MEDLINE, PubMed, Web of Science, the Cochrane Library and CNKI (China National Knowledge Infrastructure) for papers published between 1998 and 2019, using the keywords pharmacy, pharmacist, and medication or drug combined with orthopaedic. Other available resources were also used to identify relevant articles.

RESULTS AND DISCUSSION

Based on the available evidence in 74 articles, it was found that clinical pharmacists play an important role in all aspects of rational use of medications, medication review and reconciliation, monitoring adverse drug events, risk assessment, and medication education and counselling. In addition, clinical pharmacy services were developed to minimize medication errors, adverse drug events and medical costs, but clinical pharmacy is still in its early stages in orthopaedics.

WHAT IS NEW AND CONCLUSION

A multidisciplinary approach should be adopted in the orthopaedic department, as pharmacist interventions can be vital for promoting the safety, effectiveness and cost-effectiveness of pharmacotherapy. Although pharmacists' contributions to orthopaedics are not yet fully recognized, pharmaceutical services can undoubtedly contribute to both clinical and societal outcomes.

Pharmacist-assisted electronic prescribing at the time of admission to an inpatient orthopaedic unit and its impact on medication errors: a pre- and postintervention study

BACKGROUND

Prescribing and administration errors related to pre-admission medications are common amongst orthopaedic inpatients. Postprescribing medication reconciliation by clinical pharmacists after hospital admission prevents some but not all errors from reaching the patient. Involving pharmacists at the prescribing stage may more effectively prevent errors. The aim of the study was to evaluate the effect of pharmacist-assisted electronic prescribing at the time of hospital admission on medication errors in orthopaedic inpatients.

METHODS

A pre- and postintervention study was conducted in the orthopaedic unit of a major metropolitan Australian hospital. During the 10-week intervention phase, a project pharmacist used electronic prescribing to assist with prescribing admission medications and postoperative venous thromboembolism (VTE) prophylaxis, in consultation with orthopaedic medical officers. The primary endpoint was the number of medication errors per patient within 72 h of admission. Secondary endpoints included the number and consequence of adverse events (AEs) associated with admission medication errors and the time delay in administering VTE prophylaxis after elective surgery (number of hours after recommended postoperative dose-time).

RESULTS

A total of 198 and 210 patients, pre- and postintervention, were evaluated, respectively. The median number of admission medication errors per patient declined from six pre-intervention to one postintervention (p < 0.01). A total of 17 AEs were related to admission medication errors during the pre-intervention period compared with 1 postintervention. There were 54 and 63 elective surgery patients pre- and postintervention, respectively. The median delay in administering VTE prophylaxis for these patients declined from 9 h pre-intervention to 2 h postintervention (p < 0.01).

CONCLUSIONS

Pharmacist-assisted electronic prescribing reduced the number of admission medication errors and associated AEs.