Develop and validate a computable phenotype for the identification of Alzheimer's disease patients using electronic health record data.

INTRODUCTION: Alzheimer's disease (AD) is often misclassified in electronic health records (EHRs) when relying solely on diagnosis codes. This study aimed to develop a more accurate, computable phenotype (CP) for identifying AD patients using structured and unstructured EHR data. METHODS: We used EHRs from the University of Florida Health (UFHealth) system and created rule-based CPs iteratively through manual chart reviews. The CPs were then validated using data from the University of Texas Health Science Center at Houston (UTHealth) and the University of Minnesota (UMN). RESULTS: Our best-performing CP was "patient has at least 2 AD diagnoses and AD-related keywords in AD encounters," with an F1-score of 0.817 at UF, 0.961 at UTHealth, and 0.623 at UMN, respectively. DISCUSSION: We developed and validated rule-based CPs for AD identification with good performance, which will be crucial for studies that aim to use real-world data like EHRs. Highlights: Developed a computable phenotype (CP) to identify Alzheimer's disease (AD) patients using EHR data.Utilized both structured and unstructured EHR data to enhance CP accuracy.Achieved a high F1-score of 0.817 at UFHealth, and 0.961 and 0.623 at UTHealth and UMN.Validated the CP across different demographics, ensuring robustness and fairness.

Assessment of the utilization of real-time prescription benefits for patient cost savings within an outpatient setting.

Background: This study evaluates the impact of Real-Time Prescription Benefits (RTPB), a tool integrated into electronic health records (EHRs), on patient out-of-pocket costs in an academic institution. RTPB provides prescribers with alternative, less expensive medications based on insurance plans. The primary measure was cost-savings, defined as the difference between the out-of-pocket cost of the prescribed medication and its alternative. Methods: A retrospective analysis of prescriptions from outpatient clinics in a university-based health system was conducted between May 2020 and July 2021. Prescriptions were analyzed at the 2nd level of the Anatomical Therapeutic Chemical (ATC) classification system. Costs were standardized to a 30-day supply. Standardized cost and total cost per prescription, and overall savings for the top 20 medication classes at the 2nd ATC level were calculated. The overall impact of RTPB was estimated based on selecting the least expensive alternative suggested by RTPB. Results: The study found that RTPB information was provided for 22% of prescriptions, with suggested alternatives for 1.26%. Among prescriptions with an alternative selected, the standardized average cost saving was $38.83. The study realized $15,416 in patient total cost savings. If the least expensive RTPB-suggested alternative were chosen for all prescriptions, an estimated $276,386 could have been saved. Psychoanaleptic and psycholeptic medications were the most prescribed with an alternative, with most savings in specialty drugs like anthelmintic and immunostimulant medications. Conclusion: The study highlights the importance of RTPB in reducing patient costs. It reports patient cost-savings with RTPB in prescribing decisions. Future research could explore the impact of RTPB on medication adherence using pharmacy claims data.

Machine learning-based prediction of clinical outcomes after traumatic brain injury: Hidden information of early physiological time series.

AIMS: To assess the predictive value of early-stage physiological time-series (PTS) data and non-interrogative electronic health record (EHR) signals, collected within 24 h of ICU admission, for traumatic brain injury (TBI) patient outcomes. METHODS: Using data from TBI patients in the multi-center eICU database, we focused on in-hospital mortality, neurological status based on the Glasgow Coma Score (mGCS) motor subscore at discharge, and prolonged ICU stay (PLOS). Three machine learning (ML) models were developed, utilizing EHR features, PTS signals collected 24 h after ICU admission, and their combination. External validation was performed using the MIMIC III dataset, and interpretability was enhanced using the Shapley Additive Explanations (SHAP) algorithm. RESULTS: The analysis included 1085 TBI patients. Compared to individual models and existing scoring systems, the combination of EHR and PTS features demonstrated comparable or even superior performance in predicting in-hospital mortality (AUROC = 0.878), neurological outcomes (AUROC = 0.877), and PLOS (AUROC = 0.835). The model's performance was validated in the MIMIC III dataset, and SHAP algorithms identified six key intervention points for EHR features related to prognostic outcomes. Moreover, the EHR results (All AUROC >0.8) were translated into online tools for clinical use. CONCLUSION: Our study highlights the importance of early-stage PTS signals in predicting TBI patient outcomes. The integration of interpretable algorithms and simplified prediction tools can support treatment decision-making, contributing to the development of accurate prediction models and timely clinical intervention.

Direct oral anticoagulant use in hospitalized pediatric patients: Insights from nationwide data.

BACKGROUND: Direct oral anticoagulants (DOACs) have had significant impact on the management of venous thromboembolism (VTE) in adults, but these agents were not approved for use in pediatric patients until 2021. Our objective was to analyze the characteristics of pediatric patients treated with DOACs prior to and following U.S. Food and Drug Administration (FDA) approval for children and evaluate their impact on hospital outcomes. PROCEDURE: We utilized the Epic Cosmos dataset (Cosmos), a de-identified dataset of over 220 million patients, to identify patients aged 1-18 years admitted with a first-occurrence diagnosis of VTE between January 1, 2017 and June 30, 2023. Patients were grouped by anticoagulation received (unfractionated heparin, low molecular weight heparin, and/or DOACs). RESULTS: Among 5138 eligible patients, 18.1% received DOACs as all or part of their anticoagulation treatment, while 81.9% received heparin therapies alone. Patients treated with DOACs were older than patients treated with heparin monotherapy at 17.4 and 13.0 years, respectively. Non-DOAC patients were more likely to have chronic conditions and were less likely to have pulmonary embolism. Patients treated with DOACs demonstrated shorter overall length of stay and duration of intensive care unit (ICU) admission. CONCLUSIONS: DOACs remain infrequently utilized in pediatric patients, especially in those under 13 years old. Initiation on heparin therapy and transition to DOACs remains common, with 80.6% of DOAC patients receiving heparin during their hospitalization. While DOAC monotherapy is not currently endorsed as first-line therapy for DVT or PE in children, it is being used clinically. Further research is needed to clarify the impact of DOAC use on patient adherence, VTE recurrence, and healthcare cost.

Impact of automated alerts on discharge opioid overprescribing after general surgery.

DISCLAIMER: In an effort to expedite the publication of articles, AJHP is posting manuscripts online as soon as possible after acceptance. Accepted manuscripts have been peer-reviewed and copyedited, but are posted online before technical formatting and author proofing. These manuscripts are not the final version of record and will be replaced with the final article (formatted per AJHP style and proofed by the authors) at a later time. PURPOSE: The objectives of this study were to identify the most performed surgical procedures associated with the highest rates of discharge opioid overprescribing and to implement an electronic health record (EHR) alert to reduce discharge opioid overprescribing. METHODS: This quality improvement, before-and-after study included patients undergoing one of the identified target procedures-laparoscopic cholecystectomy, unilateral open inguinal hernia repair, and laparoscopic appendectomy-at an academic medical center. The alert notified providers when the prescribed opioid quantity exceeded guideline recommendations. The preimplementation cohort included surgical encounters from January 2020 to December 2021. The EHR alert was implemented in May 2022 following provider education via email and in-person presentations. The postimplementation cohort included surgical encounters from May to August 2022. The primary outcome was the proportion of patients with a discharge opioid supply exceeding guideline recommendations (overprescribing). RESULTS: A total of 1,478 patients were included in the preimplementation cohort, and 141 patients were included in the postimplementation cohort. The rate of discharge opioid overprescribing decreased from 48% in the preimplementation cohort to 3% in the postimplementation cohort, with an unadjusted absolute reduction of 45% (95% confidence interval, 41% to 49%; P < 0.001) and an adjusted odds ratio of 0.03 (95% confidence interval, 0.01 to 0.08; P < 0.001). Among patients who received opioids, the mean (SD) opioid supply at discharge decreased from 92 (43) oral morphine milligram equivalents (MME) (before implementation) to 57 (20) MME (after implementation) (P < 0.001). The proportion of patients who received additional opioid prescriptions within 1 to 14 days of hospital discharge did not change (P = 0.76). CONCLUSION: Implementation of an EHR alert along with provider education can reduce discharge opioid overprescribing following general surgery.

Factors Influencing Data Quality in Electronic Health Record Systems in 50 Health Facilities in Rwanda and the Role of Clinical Alerts: Cross-Sectional Observational Study.

BACKGROUND: Electronic health records (EHRs) play an increasingly important role in delivering HIV care in low- and middle-income countries. The data collected are used for direct clinical care, quality improvement, program monitoring, public health interventions, and research. Despite widespread EHR use for HIV care in African countries, challenges remain, especially in collecting high-quality data. OBJECTIVE: We aimed to assess data completeness, accuracy, and timeliness compared to paper-based records, and factors influencing data quality in a large-scale EHR deployment in Rwanda. METHODS: We randomly selected 50 health facilities (HFs) using OpenMRS, an EHR system that supports HIV care in Rwanda, and performed a data quality evaluation. All HFs were part of a larger randomized controlled trial, with 25 HFs receiving an enhanced EHR with clinical decision support systems. Trained data collectors visited the 50 HFs to collect 28 variables from the paper charts and the EHR system using the Open Data Kit app. We measured data completeness, timeliness, and the degree of matching of the data in paper and EHR records, and calculated concordance scores. Factors potentially affecting data quality were drawn from a previous survey of users in the 50 HFs. RESULTS: We randomly selected 3467 patient records, reviewing both paper and EHR copies (194,152 total data items). Data completeness was >85% threshold for all data elements except viral load (VL) results, second-line, and third-line drug regimens. Matching scores for data values were close to or >85% threshold, except for dates, particularly for drug pickups and VL. The mean data concordance was 10.2 (SD 1.28) for 15 (68%) variables. HF and user factors (eg, years of EHR use, technology experience, EHR availability and uptime, and intervention status) were tested for correlation with data quality measures. EHR system availability and uptime was positively correlated with concordance, whereas users' experience with technology was negatively correlated with concordance. The alerts for missing VL results implemented at 11 intervention HFs showed clear evidence of improving timeliness and completeness of initially low matching of VL results in the EHRs and paper records (11.9%-26.7%; P<.001). Similar effects were seen on the completeness of the recording of medication pickups (18.7%-32.6%; P<.001). CONCLUSIONS: The EHR records in the 50 HFs generally had high levels of completeness except for VL results. Matching results were close to or >85% threshold for nondate variables. Higher EHR stability and uptime, and alerts for entering VL both strongly improved data quality. Most data were considered fit for purpose, but more regular data quality assessments, training, and technical improvements in EHR forms, data reports, and alerts are recommended. The application of quality improvement techniques described in this study should benefit a wide range of HFs and data uses for clinical care, public health, and disease surveillance.

Inferring Population HIV Viral Load From a Single HIV Clinic's Electronic Health Record: Simulation Study With a Real-World Example.

BACKGROUND: Population viral load (VL), the most comprehensive measure of the HIV transmission potential, cannot be directly measured due to lack of complete sampling of all people with HIV. OBJECTIVE: A given HIV clinic's electronic health record (EHR), a biased sample of this population, may be used to attempt to impute this measure. METHODS: We simulated a population of 10,000 individuals with VL calibrated to surveillance data with a geometric mean of 4449 copies/mL. We sampled 3 hypothetical EHRs from (A) the source population, (B) those diagnosed, and (C) those retained in care. Our analysis imputed population VL from each EHR using sampling weights followed by Bayesian adjustment. These methods were then tested using EHR data from an HIV clinic in Delaware. RESULTS: Following weighting, the estimates moved in the direction of the population value with correspondingly wider 95% intervals as follows: clinic A: 4364 (95% interval 1963-11,132) copies/mL; clinic B: 4420 (95% interval 1913-10,199) copies/mL; and clinic C: 242 (95% interval 113-563) copies/mL. Bayesian-adjusted weighting further improved the estimate. CONCLUSIONS: These findings suggest that methodological adjustments are ineffective for estimating population VL from a single clinic's EHR without the resource-intensive elucidation of an informative prior.

Electronic Health Record-Oriented Knowledge Graph System for Collaborative Clinical Decision Support Using Multicenter Fragmented Medical Data: Design and Application Study.

BACKGROUND: The medical knowledge graph provides explainable decision support, helping clinicians with prompt diagnosis and treatment suggestions. However, in real-world clinical practice, patients visit different hospitals seeking various medical services, resulting in fragmented patient data across hospitals. With data security issues, data fragmentation limits the application of knowledge graphs because single-hospital data cannot provide complete evidence for generating precise decision support and comprehensive explanations. It is important to study new methods for knowledge graph systems to integrate into multicenter, information-sensitive medical environments, using fragmented patient records for decision support while maintaining data privacy and security. OBJECTIVE: This study aims to propose an electronic health record (EHR)-oriented knowledge graph system for collaborative reasoning with multicenter fragmented patient medical data, all the while preserving data privacy. METHODS: The study introduced an EHR knowledge graph framework and a novel collaborative reasoning process for utilizing multicenter fragmented information. The system was deployed in each hospital and used a unified semantic structure and Observational Medical Outcomes Partnership (OMOP) vocabulary to standardize the local EHR data set. The system transforms local EHR data into semantic formats and performs semantic reasoning to generate intermediate reasoning findings. The generated intermediate findings used hypernym concepts to isolate original medical data. The intermediate findings and hash-encrypted patient identities were synchronized through a blockchain network. The multicenter intermediate findings were collaborated for final reasoning and clinical decision support without gathering original EHR data. RESULTS: The system underwent evaluation through an application study involving the utilization of multicenter fragmented EHR data to alert non-nephrology clinicians about overlooked patients with chronic kidney disease (CKD). The study covered 1185 patients in nonnephrology departments from 3 hospitals. The patients visited at least two of the hospitals. Of these, 124 patients were identified as meeting CKD diagnosis criteria through collaborative reasoning using multicenter EHR data, whereas the data from individual hospitals alone could not facilitate the identification of CKD in these patients. The assessment by clinicians indicated that 78/91 (86%) patients were CKD positive. CONCLUSIONS: The proposed system was able to effectively utilize multicenter fragmented EHR data for clinical application. The application study showed the clinical benefits of the system with prompt and comprehensive decision support.

Missing Ethnicity and Language Information in Healthcare Records of Patients from the Asian Diaspora.

The electronic health record (EHR) should contain information to support culturally responsive care and research; however, the widely used default "Asian" demographic variable in most US social systems (including EHRs) lacks information to describe the diverse experience within the Asian diaspora (e.g., ethnicities, languages). This has a downstream effect on research, identifying disparities, and addressing health equity. We were particularly interested in EHRs of autistic patients from the Asian diaspora, since the presence of a developmental diagnosis might call for culturally responsive care around understanding causes, treatments, and services to support good outcomes. The aim of this study is to determine the degree to which information about Asian ethnicity, languages, and culture is documented and accessible in the EHR, and whether it is differentially available for patients with or without autism. Using electronic and manual medical chart review, all autistic and "Asian" children (group 1; n = 52) were compared to a randomly selected comparison sample of non-autistic and "Asian" children (group 2; n = 50). Across both groups, manual chart review identified more specific approximations of racial/ethnic backgrounds in 54.5% of patients, 56% for languages spoken, and that interpretation service use was underestimated by 13 percentage points. Our preliminary results highlight that culturally responsive information was inconsistent, missing, or located in progress notes rather than a central location where it could be accessed by providers. Recommendations about the inclusion of Asian ethnicity and language data are provided to potentially enhance cultural responsiveness and support better outcomes for families with an autistic child.

Development and integration of a clinical dashboard within a dental school setting.

PURPOSE: To describe the development and integration of an electronic health record-driven, student dashboard that displays real-time data relative to the students' patient management and clinic experiences at the University of Illinois Chicago, College of Dentistry. MATERIALS AND METHODS: Following development and implementation of the student dashboard, various objective metrics were evaluated to identify any improvements in the clinical patient management. A cross-sectional retrospective chart review was completed of the electronic health record (axiUm, Exan, Coquitlam, BC, Canada) from January 2019 to April 2022 evaluating four performance metrics: student lockouts, note/code violations, overdue active patients, and overdue recall patients. Descriptive statistics were analyzed. The Kolmogorov-Smirnov test was applied to assess the normal distribution of data. Data were analyzed by the Kruskal-Wallis tests for potential differences between pre-dashboard and post-dashboard implementation years with the mean overdue active/recall patient to student ratio variables. Mann-Whitney U-tests for between-groups comparisons with Bonferroni correction for multiple comparisons were performed (α = 0.05). Descriptive statistics were performed to analyze the student utilization frequency of the dashboard. RESULTS: Post-implementation analysis indicated a slight decrease in the number of lockouts and note/code violation; and a statistically significant decrease in overdue active patients post-dashboard (P < 0.001). On average, students accessed their dashboards 3.3 times a week. CONCLUSIONS: Implementation of a student dashboard through the electronic health record platform within an academic dental practice has the potential to assist students with patient management and is utilized regularly by the students.

Hybrid electronic record: An error reduction strategy for diverse medical prescription formats.

BACKGROUND:  This project is part of a broader effort to develop a new electronic registry for ophthalmology in the KwaZulu-Natal (KZN) province in South Africa. The registry should include a clinical decision support system that reduces the potential for human error and should be applicable for our diversity of hospitals, whether electronic health record (EHR) or paper-based. METHODS:  Post-operative prescriptions of consecutive cataract surgery discharges were included for 2019 and 2020. Comparisons were facilitated by the four chosen state hospitals in KZN each having a different system for prescribing medications: Electronic, tick sheet, ink stamp and handwritten health records. Error types were compared to hospital systems to identify easily-correctable errors. Potential error remedies were sought by a four-step process. RESULTS:  There were 1307 individual errors in 1661 prescriptions, categorised into 20 error types. Increasing levels of technology did not decrease error rates but did decrease the variety of error types. High technology scripts had the most errors but when easily correctable errors were removed, EHRs had the lowest error rates and handwritten the highest. CONCLUSION:  Increasing technology, by itself, does not seem to reduce prescription error. Technology does, however, seem to decrease the variability of potential error types, which make many of the errors simpler to correct.Contribution: Regular audits are an effective tool to greatly reduce prescription errors, and the higher the technology level, the more effective these audit interventions become. This advantage can be transferred to paper-based notes by utilising a hybrid electronic registry to print the formal medical record.

Patients and family attitudes about clinical and research sharing of electronic clinical data.

PURPOSE: To access the attitudes of service users about the sharing of health records for research and to foster collaboration between municipal health services and the specialist health services in Norway. METHODS: Members (n ≈ 2000) of the Norwegian mental health service users' organizations (SUO's), ADHD Norway, the Autism Association and the Tourette Association, representing Central Norway, participated in the study, (N = 108, 5.4% response rate). Descriptive statistics were used to evaluate distributions of responses to the questionnaire. RESULTS: Service users reported being aware that municipal health services collaborate with the specialist health service (62%), with mental health care in the specialist health service (57%), and child and adolescent psychiatric services (61%). A large proportion of individuals were aware of the benefits of sharing their health records (93%), have trust in the use of data by health authorities (81%), and were willing to share records to benefit fellow patients (84%). Personal experience (69%) and impressions from mainstream media (55%) had the most influential impact on users' views of the Health Platform, an electronic health communication system. A majority of users had a negative perception of the Health Platform, even though some expect it to become a valuable tool in the future (50%). CONCLUSIONS: Service users are aware of and positive about benefiting others by sharing health records. They trust the health authorities, however, have negative attitudes about the Health Platform, apparently based on personal experiences and media influence. However, service users can see the potential usefulness of the Health Platform in the future.

Conformation-associated health in pet rabbits in the UK: A VetCompass cohort study.

BACKGROUND: Domestic rabbit breeds vary substantially from the wild rabbit body type. However, little is known about how the conformation of pet rabbits influences their health. METHODS: Data were extracted from VetCompass anonymised clinical records of rabbits under UK primary veterinary care during 2019. RESULTS: The study included 162,107 rabbits. Based on 88,693 rabbits with relevant breed information recorded, skull shape was classified as brachycephalic (79.69%), mesaticephalic (16.80%) and dolichocephalic (3.51%). Based on 83,821 rabbits with relevant breed information recorded, ear carriage was classified as lop-eared (57.05%) and erect-eared (42.95%). From a random sample of 3933 rabbits, the most prevalent disorders recorded overall were overgrown nail(s) (28.19%), overgrown molar(s) (14.90%) and obesity (8.82%). Compared to those with a mesaticephalic skull shape, brachycephalic rabbits had lower odds of obesity, anorexia and gastrointestinal stasis and higher odds of perineal faecal impaction, tear duct abnormality and haircoat disorder. Compared to erect-eared rabbits, lop-eared rabbits had higher odds of perineal faecal impaction and tear duct abnormality. LIMITATION: A large proportion of records with incomplete breed information hindered full analysis for breed-related and conformation-related attributes. CONCLUSION: Limited evidence for major links between skull shape or ear carriage conformations and overall disorder risk suggests that factors such as husbandry or even just living life as a domesticated species may be bigger drivers of common health issues in pet rabbits in the UK.

Validation of an Electronic Health Record-Based Machine Learning Model Compared to Clinical Risk Scores for Gastrointestinal Bleeding.

BACKGROUND & AIMS: Guidelines recommend use of risk stratification scores for patients presenting with gastrointestinal bleeding (GIB) to identify very-low-risk patients eligible for discharge from emergency departments. Machine learning models may outperform existing scores and can be integrated within the electronic health record (EHR) to provide real-time risk assessment without manual data entry. We present the first EHR-based machine learning model for GIB. METHODS: The training cohort comprised 2,546 patients and internal validation of 850 patients presenting with overt GIB (hematemesis, melena, hematochezia) to emergency departments of 2 hospitals from 2014-2019. External validation was performed on 926 patients presenting to a different hospital with the same EHR from 2014-2019. The primary outcome was a composite of red-blood-cell transfusion, hemostatic intervention (endoscopic, interventional radiologic, or surgical), and 30-day all-cause mortality. We used structured data fields in the EHR available within 4 hours of presentation and compared performance of machine learning models to current guideline-recommended risk scores, Glasgow-Blatchford Score (GBS) and Oakland Score. Primary analysis was area under the receiver-operating-characteristic curve (AUC). Secondary analysis was specificity at 99% sensitivity to assess proportion of patients correctly identified as very-low-risk. RESULTS: The machine learning model outperformed the GBS (AUC=0.92 vs. 0.89;p<0.001) and Oakland score (AUC=0.92 vs. 0.89;p<0.001). At the very-low-risk threshold of 99% sensitivity, the machine learning model identified more very-low-risk patients: 37.9% vs. 18.5% for GBS and 11.7% for Oakland score (p<0.001 for both comparisons). CONCLUSIONS: An EHR-based machine learning model performs better than currently recommended clinical risk scores and identifies more very-low-risk patients eligible for discharge from the emergency department.

Statistical analysis plan for the NU IMPACT stepped-wedge cluster randomized trial.

BACKGROUND: As part of the IMPACT Consortium of three effectiveness-implementation trials, the NU IMPACT trial was designed to evaluate implementation and effectiveness outcomes for an electronic health record (EHR)-embedded symptom monitoring and management program for outpatient cancer care. NU IMPACT uses a unique stepped-wedge cluster randomized design, involving six clusters of 26 clinics, for evaluation of implementation outcomes with an embedded patient-level randomized trial to evaluate effectiveness outcomes. Collaborative, consortium-wide efforts to ensure use of the most robust and recent analytic methodologies for stepped-wedge trials motivated updates to the statistical analysis plan for implementation outcomes in the NU IMPACT trial. METHODS: In the updated statistical analysis plan for NU IMPACT, the primary implementation outcome patient adoption, as measured by clinic-level monthly proportions of patient engagement with the EHR-based cancer symptom monitoring system, will be analyzed using generalized least squares linear regression with auto-regressive errors and adjustment for cluster and time effects (underlying secular trends). A similar strategy will be used for secondary patient and provider implementation outcomes. DISCUSSION: The analytic updates described here resulted from highly iterative, collaborative efforts among statisticians, implementation scientists, and trial leads in the IMPACT Consortium. This updated statistical analysis plan will serve as the a priori specified approach for analyzing implementation outcomes for the NU IMPACT trial.

Evaluation of an electronic health record Drug Interaction Customization Editor (DICE).

DISCLAIMER: In an effort to expedite the publication of articles, AJHP is posting manuscripts online as soon as possible after acceptance. Accepted manuscripts have been peer-reviewed and copyedited, but are posted online before technical formatting and author proofing. These manuscripts are not the final version of record and will be replaced with the final article (formatted per AJHP style and proofed by the authors) at a later time. PURPOSE: Due to the low specificity of drug-drug interaction (DDI) warnings, hospitals and healthcare systems would benefit from the ability to customize alerts, thereby reducing the burden of alerts while simultaneously preventing harm. We developed a tool, called the Drug Interaction Customization Editor (DICE), as a prototype to identify features and functionality that could assist healthcare organizations in customizing DDI alerts. METHODS: A team of pharmacists, physicians, and DDI experts identified attributes expected to be useful for filtering DDI warnings. A survey was sent to pharmacists with informatics responsibilities and other medication safety committee members to obtain their opinions about the tool. The survey asked participants to evaluate the 4 sections of the DICE tool (General, Medication, Patient, and Visit) on a scale ranging from 0 (not useful) to 100 (very useful). The survey provided an opportunity for participants to express their opinions on the overall usefulness of the DICE tool and to provide other comments. RESULTS: The 50 survey respondents were mainly pharmacists (n = 47, 94%) with almost half (n = 23, 47%) having health information technology/informatics training. Most respondents (n = 33, 80%) were employed by organizations with over 350 beds. Respondents indicated the most useful features of the DICE tool were the ability to filter DDI warnings based on routes of administrations (mean [SD] rating scale score, 86.5 [21.6]), primary drug properties (85.7 [20.5]) patient attributes (85.6 [16.7]) and laboratory attributes (88.8 [18.0]). The overall impression of the DICE tool was rated at 82.8 (19.0), and when asked about the potential to reduce DDI alerts, respondents rated the tool at 83.7 (21.8). CONCLUSION: The ability to customize DDI alerts using data elements currently within the EHR has the potential to decrease alert fatigue and override rates. This prototype DICE tool could be used by end users and vendors as a template for developing a more advanced DDI filtering tool within EHR systems.

The use of artificial intelligence to detect voided medication orders in oral and maxillofacial surgery inpatients.

OBJECTIVE: The aim of this study is to determine if supervised machine learning algorithms can accurately predict voided computerized physician order entry in oral and maxillofacial surgery inpatients. METHODS: Data from Electronic Medical Record included patient demographics, comorbidities, procedures, vital signs, laboratory values, and medication orders were retrospectively collected. Predictor variables included patient demographics, comorbidities, procedures, vital signs, and laboratory values. Outcome of interest is if a medication order was voided or not. Data was cleaned and processed using Microsoft Excel and Python v3.12. Gradient Boosted Decision Trees, Random Forest, K-Nearest Neighbor, and Naïve Bayes were trained, validated, and tested for accuracy of the prediction of voided medication orders. RESULTS: 37,493 medication orders from 1,204 patient admissions over 5 years were used for this study. 3,892 (10.4%) medication orders were voided. Gradient Boosted Decision Trees, Random Forest, K-Nearest Neighbor, and Naïve Bayes had an Area Under the Receiver Operating Curve of 0.802 with 95% CI [0.787, 0.825], 0.746 with 95% CI [0.722, 0.765], 0.685 with 95% CI [0.667, 0.699], and 0.505 with 95% CI [0.489, 0.539], respectively. Area Under the Precision Recall Curve was 0.684 with 95% CI [0.679, 0.702], 0.647 with 95% CI [0.638, 0.664], 0.429 with 95% CI [0.417, 0.434], and 0.551 with 95% CI [0.551, 0.552], respectively. CONCLUSION: Gradient Boosted Decision Trees was the best performing model of the supervised machine learning algorithms with satisfactory outcomes in the test cohort for predicting voided Computerized Physician Order Entry in Oral and Maxillofacial Surgery inpatients.

Who visits primary health care general practitioners and why? A register-based study in a Finnish city.

This is a register-based study that examines the distribution of diagnoses made by general practitioners (GPs) in the public primary health care of the city of Vantaa, Finland. Data were gathered from the electronic health record (EHR) system and consisted of every record entered into the EHR system between 1 January 2016 and 31 December 2018. Both absolute numbers and relative proportions of the 10th edition of International Classification of Diseases (ICD-10) diagnosis recordings were reported and calculated. Among GP visits, the 88 most common diagnoses covered 75% of all diagnoses. The most common diagnoses were related to the musculoskeletal (3.8%, ICD code M54) and respiratory systems (6.0%, ICD-10 code J06). Primary health care GP services were mostly used by children (age <5 years) and older adults (>65 years). Health examinations - mostly children's and maternity clinics appointments/visits - covered 20% of the GP office visits. Women between the ages 15-79 years had relatively more GP visits compared to men. The 88 most commonly recorded diagnoses covered the majority of the GP visits. Health examinations for the healthy were an important part of GPs' work. In an urban Finnish city, GP services were predominantly used by children and older adults.