Using a decision tree algorithm to distinguish between repeated supra-therapeutic and acute acetaminophen exposures.

BACKGROUND: This study aimed to compare clinical and laboratory characteristics of supra-therapeutic (RSTI) and acute acetaminophen exposures using a predictive decision tree (DT) algorithm. METHODS: We conducted a retrospective cohort study using the National Poison Data System (NPDS). All patients with RSTI acetaminophen exposure (n = 4,522) between January 2012 and December 2017 were included. Additionally, 4,522 randomly selected acute acetaminophen ingestion cases were included. After that, the DT machine learning algorithm was applied to differentiate acute acetaminophen exposure from supratherapeutic exposures. RESULTS: The DT model had accuracy, precision, recall, and F1-scores of 0.75, respectively. Age was the most relevant variable in predicting the type of acetaminophen exposure, whether RSTI or acute. Serum aminotransferase concentrations, abdominal pain, drowsiness/lethargy, and nausea/vomiting were the other most important factors distinguishing between RST and acute acetaminophen exposure. CONCLUSION: DT models can potentially aid in distinguishing between acute and RSTI of acetaminophen. Further validation is needed to assess the clinical utility of this model.

Decision tree algorithm can determine the outcome of repeated supratherapeutic ingestion (RSTI) exposure to acetaminophen: review of 4500 national poison data system cases.

This study is aimed at establishing the outcome of RSTI exposure to acetaminophen based on a decision tree algorithm for the first time. This study used the National Poison Data System (NPDS) to conduct a six-year retrospective cohort analysis, which included 4522 individuals. The patients had a mean age of 26.75 ± 16.3 years (1-89). 3160 patients (70%) were females. Most patients had intentional exposure to acetaminophen. Almost all the patients had acetaminophen exposure via ingestion. In addition, 400 (8.8%) experienced major outcomes, 1500 (33.2%) experienced moderate outcomes, and 2622 (58%) of the patients experienced mild ones. The decision tree model performed well in the training and test groups. In the test group, the accuracy was 0.813, precision of 0.827, recall being 0.798, specificity 0.898, and an F1 score 0.80. In the training group, accuracy was 0.831, recall was 0.825, precision was 0.837, specificity was 0.90, and F1 score was 0.829. Our results showed that serum liver enzymes being present at elevated levels (Alanine aminotransferase (ALT), Aspartate aminotransferase (AST) greater than 1000 U/L followed by ALT, AST between 100 and 1000 U/L), prothrombin time (PT) prolongation, bilirubin increase, renal failure, confusion, age, hypotension, other coagulopathy (such as partial thromboplastin time (PTT) prolongation), acidosis, and electrolyte abnormality were the effective factors in determining the outcomes in these patients. The decision tree algorithm is a dependable method for establishing the prognosis of patients who have been exposed to RSTI acetaminophen and can be used throughout the patients' hospitalization period.

The value of machine learning for prognosis prediction of diphenhydramine exposure: National analysis of 50,000 patients in the United States.

Background: Diphenhydramine (DPH) is an antihistamine medication that in overdose can result in anticholinergic symptoms and serious complications, including arrhythmia and coma. We aimed to compare the value of various machine learning (ML) models, including light gradient boosting machine (LGBM), logistic regression (LR), and random forest (RF), in the outcome prediction of DPH poisoning. Materials and Methods: We used the National Poison Data System database and included all of the human exposures of DPH from January 01, 2017 to December 31, 2017, and excluded those cases with missing information, duplicated cases, and those who reported co-ingestion. Data were split into training and test datasets, and three ML models were compared. We developed confusion matrices for each, and standard performance metrics were calculated. Results: Our study population included 53,761 patients with DPH exposure. The most common reasons for exposure, outcome, chronicity of exposure, and formulation were captured. Our results showed that the average precision-recall area under the curve (AUC) of 0.84. LGBM and RF had the highest performance (average AUC of 0.91), followed by LR (average AUC of 0.90). The specificity of the models was 87.0% in the testing groups. The precision of models was 75.0%. Recall (sensitivity) of models ranged between 73% and 75% with an F1 score of 75.0%. The overall accuracy of LGBM, LR, and RF models in the test dataset was 74.8%, 74.0%, and 75.1%, respectively. In total, just 1.1% of patients (mostly those with major outcomes) received physostigmine. Conclusion: Our study demonstrates the application of ML in the prediction of DPH poisoning.

Prevalence of food allergies and intolerances documented in electronic health records.

BACKGROUND: Food allergy prevalence is reported to be increasing, but epidemiological data using patients' electronic health records (EHRs) remain sparse. OBJECTIVE: We sought to determine the prevalence of food allergy and intolerance documented in the EHR allergy module. METHODS: Using allergy data from a large health care organization's EHR between 2000 and 2013, we determined the prevalence of food allergy and intolerance by sex, racial/ethnic group, and allergen group. We examined the prevalence of reactions that were potentially IgE-mediated and anaphylactic. Data were validated using radioallergosorbent test and ImmunoCAP results, when available, for patients with reported peanut allergy. RESULTS: Among 2.7 million patients, we identified 97,482 patients (3.6%) with 1 or more food allergies or intolerances (mean, 1.4 ± 0.1). The prevalence of food allergy and intolerance was higher in females (4.2% vs 2.9%; P < .001) and Asians (4.3% vs 3.6%; P < .001). The most common food allergen groups were shellfish (0.9%), fruit or vegetable (0.7%), dairy (0.5%), and peanut (0.5%). Of the 103,659 identified reactions to foods, 48.1% were potentially IgE-mediated (affecting 50.8% of food allergy or intolerance patients) and 15.9% were anaphylactic. About 20% of patients with reported peanut allergy had a radioallergosorbent test/ImmunoCAP performed, of which 57.3% had an IgE level of grade 3 or higher. CONCLUSIONS: Our findings are consistent with previously validated methods for studying food allergy, suggesting that the EHR's allergy module has the potential to be used for clinical and epidemiological research. The spectrum of severity observed with food allergy highlights the critical need for more allergy evaluations.

Clinically Inconsequential Alerts: The Characteristics of Opioid Drug Alerts and Their Utility in Preventing Adverse Drug Events in the Emergency Department.

STUDY OBJECTIVE: We examine the characteristics of clinical decision support alerts triggered when opioids are prescribed, including alert type, override rates, adverse drug events associated with opioids, and preventable adverse drug events. METHODS: This was a retrospective chart review study assessing adverse drug event occurrences for emergency department (ED) visits in a large urban academic medical center using a commercial electronic health record system with clinical decision support. Participants include those aged 18 to 89 years who arrived to the ED every fifth day between September 2012 and January 2013. The main outcome was characteristics of opioid drug alerts, including alert type, override rates, opioid-related adverse drug events, and adverse drug event preventability by clinical decision support. RESULTS: Opioid drug alerts were more likely to be overridden than nonopioid alerts (relative risk 1.35; 95% confidence interval [CI] 1.21 to 1.50). Opioid drug-allergy alerts were twice as likely to be overridden (relative risk 2.24; 95% CI 1.74 to 2.89). Opioid duplicate therapy alerts were 1.57 times as likely to be overridden (95% CI 1.30 to 1.89). Fourteen of 4,581 patients experienced an adverse drug event (0.31%; 95% CI 0.15% to 0.47%), and 8 were due to opioids (57.1%). None of the adverse drug events were preventable by clinical decision support. However, 46 alerts were accepted for 38 patients that averted a potential adverse drug event. Overall, 98.9% of opioid alerts did not result in an actual or averted adverse drug event, and 96.3% of opioid alerts were overridden. CONCLUSION: Overridden opioid alerts did not result in adverse drug events. Clinical decision support successfully prevented adverse drug events at the expense of generating a large volume of inconsequential alerts. To prevent 1 adverse drug event, providers dealt with more than 123 unnecessary alerts. It is essential to refine clinical decision support alerting systems to eliminate inconsequential alerts to prevent alert fatigue and maintain patient safety.

Automated misspelling detection and correction in clinical free-text records.

Accurate electronic health records are important for clinical care and research as well as ensuring patient safety. It is crucial for misspelled words to be corrected in order to ensure that medical records are interpreted correctly. This paper describes the development of a spelling correction system for medical text. Our spell checker is based on Shannon's noisy channel model, and uses an extensive dictionary compiled from many sources. We also use named entity recognition, so that names are not wrongly corrected as misspellings. We apply our spell checker to three different types of free-text data: clinical notes, allergy entries, and medication orders; and evaluate its performance on both misspelling detection and correction. Our spell checker achieves detection performance of up to 94.4% and correction accuracy of up to 88.2%. We show that high-performance spelling correction is possible on a variety of clinical documents.

The role of decision tree and machine learning models for outcome prediction of bupropion exposure: A nationwide analysis of more than 14 000 patients in the United States.

Bupropion is widely used for the treatment of major depressive disorder and for smoking cessation assistance. Unfortunately, there are no practical systems to assist clinicians or poison centres in predicting outcomes based on clinical features. Hence, the purpose of this study was to use a decision tree approach to inform early diagnosis of outcomes secondary to bupropion overdose. This study utilized a dataset from the National Poison Data System, a 6-year retrospective cohort study on toxic exposures and patient outcomes. A machine learning algorithm (decision tree) was applied to the dataset using the sci-kit-learn library in Python. Shapley Additive exPlanations (SHAP) were used as an explainable method. Comparative analysis was performed using random forest (RF), Gradient Boosting classification, eXtreme Gradient Boosting, Light Gradient Boosting (LGM) and voting ensembling. ROC curve and precision-recall curve were used to analyse the performance of each model. LGM and RF demonstrated the highest performance to predict outcome of bupropion exposure. Multiple seizures, conduction disturbance, intentional exposure, and confusion were the most influential factors to predict the outcome of bupropion exposure. Coma and seizure, including single, multiple and status, were the most important factors to predict major outcomes.

Deep learning neural network derivation and testing to distinguish acute poisonings.

INTRODUCTION: Acute poisoning is a significant global health burden, and the causative agent is often unclear. The primary aim of this pilot study was to develop a deep learning algorithm that predicts the most probable agent a poisoned patient was exposed to from a pre-specified list of drugs. RESEARCH DESIGN & METHODS: Data were queried from the National Poison Data System (NPDS) from 2014 through 2018 for eight single-agent poisonings (acetaminophen, diphenhydramine, aspirin, calcium channel blockers, sulfonylureas, benzodiazepines, bupropion, and lithium). Two Deep Neural Networks (PyTorch and Keras) designed for multi-class classification tasks were applied. RESULTS: There were 201,031 single-agent poisonings included in the analysis. For distinguishing among selected poisonings, PyTorch model had specificity of 97%, accuracy of 83%, precision of 83%, recall of 83%, and a F1-score of 82%. Keras had specificity of 98%, accuracy of 83%, precision of 84%, recall of 83%, and a F1-score of 83%. The best performance was achieved in the diagnosis of single-agent poisoning in diagnosing poisoning by lithium, sulfonylureas, diphenhydramine, calcium channel blockers, then acetaminophen, in PyTorch (F1-score = 99%, 94%, 85%, 83%, and 82%, respectively) and Keras (F1-score = 99%, 94%, 86%, 82%, and 82%, respectively). CONCLUSION: Deep neural networks can potentially help in distinguishing the causative agent of acute poisoning. This study used a small list of drugs, with polysubstance ingestions excluded.Reproducible source code and results can be obtained at https://github.com/ashiskb/npds-workspace.git.

Dynamic reaction picklist for improving allergy reaction documentation: A usability study.

OBJECTIVE: To assess novel dynamic reaction picklists for improving allergy reaction documentation compared to a static reaction picklist. MATERIALS AND METHODS: We developed three web-based user interfaces (UIs) mimicking the Mass General Brigham's EHR allergy module: the first and second UIs (i.e., UI-1D, UI-2D) implemented two dynamic reaction picklists with different ranking algorithms and the third UI (UI-3S) implemented a static reaction picklist like the one used in the current EHR. We recruited 18 clinicians to perform allergy entry for 10 test cases each via UI-1D and UI-3S, and another 18 clinicians via UI-2D and UI-3S. Primary measures were the number of free-text entries and time to complete the allergy entry. Clinicians were also interviewed using 30 questions before and after the data entry. RESULTS AND DISCUSSIONS: Among 36 clinicians, less than half were satisfied with the current EHR reaction picklists, due to their incomprehensiveness, inefficiency, and lack of intuitiveness. The clinicians used significantly fewer free-text entries when using UI-1D or UI-2D compared to UI-3S (p < 0.05). The clinicians used on average 51 s (15 %) less time via UI-1D and 50 s (16 %) less time via UI-2D in completing the allergy entries versus UI-3S, and there was not a statistically significant difference in documentation time for either group between the dynamic and static UIs. Overall, 15-17 (83-94 %) clinicians rated UI-1D and 13-15 (72-83 %) clinicians rated UI-2D as efficient, easy to use, and useful, while less than half rated the same for UI-3S. Most clinicians reported that the dynamic reaction picklists always or often suggested appropriate reactions (n = 30, 83 %) and would decrease the free-text entries (n = 26, 72 %); nearly all preferred the dynamic picklist over the static picklist (n = 32, 89 %). CONCLUSION: We found that dynamic reaction picklists significantly reduced the number of free-text entries and could reduce the time for allergy documentation by 15%. Clinicians preferred the dynamic reaction picklist over the static picklist.

Utility of support vector machine and decision tree to identify the prognosis of metformin poisoning in the United States: analysis of National Poisoning Data System.

BACKGROUND: With diabetes incidence growing globally and metformin still being the first-line for its treatment, metformin's toxicity and overdose have been increasing. Hence, its mortality rate is increasing. For the first time, we aimed to study the efficacy of machine learning algorithms in predicting the outcome of metformin poisoning using two well-known classification methods, including support vector machine (SVM) and decision tree (DT). METHODS: This study is a retrospective cohort study of National Poison Data System (NPDS) data, the largest data repository of poisoning cases in the United States. The SVM and DT algorithms were developed using training and test datasets. We also used precision-recall and ROC curves and Area Under the Curve value (AUC) for model evaluation. RESULTS: Our model showed that acidosis, hypoglycemia, electrolyte abnormality, hypotension, elevated anion gap, elevated creatinine, tachycardia, and renal failure are the most important determinants in terms of outcome prediction of metformin poisoning. The average negative predictive value for the decision tree and SVM models was 92.30 and 93.30. The AUC of the ROC curve of the decision tree for major, minor, and moderate outcomes was 0.92, 0.92, and 0.89, respectively. While this figure of SVM model for major, minor, and moderate outcomes was 0.98, 0.90, and 0.82, respectively. CONCLUSIONS: In order to predict the prognosis of metformin poisoning, machine learning algorithms might help clinicians in the management and follow-up of metformin poisoning cases.

Reconciling Allergy Information in the Electronic Health Record After a Drug Challenge Using Natural Language Processing.

Background: Drug challenge tests serve to evaluate whether a patient is allergic to a medication. However, the allergy list in the electronic health record (EHR) is not consistently updated to reflect the results of the challenge, affecting clinicians' prescription decisions and contributing to inaccurate allergy labels, inappropriate drug-allergy alerts, and potentially ineffective, more toxic, and/or costly care. In this study, we used natural language processing (NLP) to automatically detect discrepancies between the EHR allergy list and drug challenge test results and to inform the clinical recommendations provided in a real-time allergy reconciliation module. Methods: This study included patients who received drug challenge tests at the Mass General Brigham (MGB) Healthcare System between June 9, 2015 and January 5, 2022. At MGB, drug challenge tests are performed in allergy/immunology encounters with routine clinical documentation in notes and flowsheets. We developed a rule-based NLP tool to analyze and interpret the challenge test results. We compared these results against EHR allergy lists to detect potential discrepancies in allergy documentation and form a recommendation for reconciliation if a discrepancy was identified. To evaluate the capability of our tool in identifying discrepancies, we calculated the percentage of challenge test results that were not updated and the precision of the NLP algorithm for 200 randomly sampled encounters. Results: Among 200 samples from 5,312 drug challenge tests, 59% challenged penicillin reactivity and 99% were negative. 42.0%, 61.5%, and 76.0% of the results were confirmed by flowsheets, NLP, or both, respectively. The precision of the NLP algorithm was 96.1%. Seven percent of patient allergy lists were not updated based on drug challenge test results. Flowsheets alone were used to identify 2.0% of these discrepancies, and NLP alone detected 5.0% of these discrepancies. Because challenge test results can be recorded in both flowsheets and clinical notes, the combined use of NLP and flowsheets can reliably detect 5.5% of discrepancies. Conclusion: This NLP-based tool may be able to advance global delabeling efforts and the effectiveness of drug allergy assessments. In the real-time EHR environment, it can be used to examine patient allergy lists and identify drug allergy label discrepancies, mitigating patient risks.

Classification of acute poisoning exposures with machine learning models derived from the National Poison Data System.

The primary aim of this pilot study was to develop a machine learning algorithm to predict and distinguish eight poisoning agents based on clinical symptoms. Data were used from the National Poison Data System from 2014 to 2018, for patients 0-89 years old with single-agent exposure to eight drugs or drug classes (acetaminophen, aspirin, benzodiazepines, bupropion, calcium channel blockers, diphenhydramine, lithium and sulfonylureas). Four classifier prediction models were applied to the data: logistic regression, LightGBM, XGBoost, and CatBoost. There were 201 031 cases used to develop and test the algorithms. Among the four models, accuracy ranged 77%-80%, with precision and F1 scores of 76%-80% and recall of 77%-78%. Overall specificity was 92% for all models. Accuracy was highest for identifying sulfonylureas, acetaminophen, benzodiazepines and diphenhydramine poisoning. F1 scores were highest for correctly classifying sulfonylureas, acetaminophen and benzodiazepine poisonings. Recall was highest for sulfonylureas, acetaminophen, and benzodiazepines, and lowest for bupropion. Specificity was >99% for models of sulfonylureas, calcium channel blockers, lithium and aspirin. For single-agent poisoning cases among the eight possible exposures, machine learning models based on clinical signs and symptoms moderately predicted the causal agent. CatBoost and LightGBM classifier models had the highest performance of those tested.

Identifying and Reconciling Patients' Allergy Information Within the Electronic Health Record.

Allergy information is often documented in diverse sections of the electronic health record (EHR). Systematically reconciling allergy information across the EHR is critical to improve the accuracy and completeness of patients' allergy lists and ensure patient safety. In this retrospective cohort study, we examined the prevalence of incompleteness, inaccuracy, and redundancy of allergy information for patients with a clinical encounter at any Mass General Brigham facility between January 1, 2018 and December 31, 2018. We identified 4 key places in the EHR containing reconcilable allergy information: 1) allergy modules (including free text comments and duplicate allergen entries), 2) medication laboratory tests results, 3) oral medication allergy challenge tests, and 4) medication orders that have been discontinued due to adverse drug reactions (ADRs). Within our cohort, 718,315 (45.2% of the total 1,588,979) patients had an active allergy entry; of which, 266,275 (37.1%) patient's records indicated a need for reconciliation.

Heterogeneity of Drug Allergies and Reaction Lists in Two U.S. Health Care Systems' Electronic Health Records.

BACKGROUND: Health care institutions have their own "picklist" for clinicians to document adverse drug reactions (ADRs) into the electronic health record (EHR) allergy list. Whether the lack of a nationally standardized picklist impacts clinician data entries is unknown. OBJECTIVES: The objective of this study was to assess the impact of defined reaction picklists on clinical documentation and, therefore, downstream analytics and clinical research using these data at two institutions. METHODS: ADR data were obtained from the EHRs of patients who visited the emergency department or outpatient clinics at Brigham and Women's Hospital (BWH) and University of Colorado Hospital (UCH) from 2013 to 2018. Reported drug class ADR prevalences were calculated. We investigated the reactions on each picklist and compared the top 40 reactions at each institution, as well as the top 10 reactions within each drug class. RESULTS: Of 2,160,116 patients, 640,444 (30%) had 928,973 active drug allergies. The most commonly reported drug class allergens were similar between BWH and UCH. BWH's picklist had 48 reactions, and UCH's had 160 reactions; 29 reactions were shared by both picklists. While the top four reactions overall (rash, GI upset/nausea/vomiting, hives, itching) were identical between sites, reactions by drug class exhibited greater documentation diversity. For example, while the summed prevalence of swelling-related reactions to angiotensin-converting-enzyme inhibitors was comparable across sites, swelling was represented by two terms ("swelling," "angioedema") at BWH but 11 terms at UCH (e.g., "swelling," "edema," by body locality). CONCLUSION: The availability and granularity of reaction picklists impact ADR documentation in the EHR by health care providers; picklists may partially explain variations in reported ADRs across health care systems.

Unsupervised clinical relevancy ranking of structured medical records to retrieve condition-specific information in the emergency department.

BACKGROUND: Decision making in the Emergency Department (ED) requires timely identification of clinical information relevant to the complaints. Existing information retrieval solutions for the electronic health record (EHR) focus on patient cohort identification and lack clinical relevancy ranking. We aimed to compare knowledge-based (KB) and unsupervised statistical methods for ranking EHR information by relevancy to a chief complaint of chest or back pain among ED patients. METHODS: We used Pointwise-mutual information (PMI) with corpus level significance adjustment (cPMId), which modifies PMI to reward co-occurrence patterns with a higher absolute count. cPMId for each pair of medication/problem and chief complaint was estimated from a corpus of 100,000 un-annotated ED encounters. Five specialist physicians ranked the relevancy of medications and problems to each chief complaint on a 0-4 Likert scale to form the KB ranking. Reverse chronological order was used as a baseline. We directly compared the three methods on 1010 medications and 2913 problems from 99 patients with chest or back pain, where each item was manually labeled as relevant or not to the chief complaint, using mean average-precision. RESULTS: cPMId out-performed KB ranking on problems (86.8% vs. 81.3%, p < 0.01) but under-performed it on medications (93.1% vs. 96.8%, p < 0.01). Both methods significantly outperformed the baseline for both medications and problems (71.8% and 72.1%, respectively, p < 0.01 for both comparisons). The two complaints represented virtually completely different information needs (average Jaccard index of 0.008). CONCLUSION: A fully unsupervised statistical method can provide a reasonably accurate, low-effort and scalable means for situation-specific ranking of clinical information within the EHR.

Reduced admission rates and resource utilization for chest pain patients using an electronic health record-embedded clinical pathway in the emergency department.

OBJECTIVES: Assess the impact of an electronic health record (EHR)-embedded clinical pathway (ePATH) as compared to a paper-based clinical decision support tool on outcomes for patients presenting to the emergency department (ED) with suspected acute coronary syndrome (ACS). METHODS: A retrospective, quasi-experimental study using difference-in-differences and interrupted time series specifications to evaluate the impact of an EHR-embedded clinical pathway between April 2013 and July 2017. The intervention was implemented in February 2016 at a large academic tertiary hospital and compared to a local community hospital without the intervention. Eligible patients included adults (>18 years) presenting to the ED with chest pain who had a troponin ordered within 2 hours of arrival and a chest pain-related diagnosis. Patients with initial evidence of acute myocardial infarction were excluded. Primary outcomes included rates of admission and stress testing, hospital length of stay, and occurrence of major adverse cardiac events. RESULTS: On average, there were 170 chest pain visits per month at the intervention site. The frequency of hospital admission (unadjusted 28.2% to 20.9%, P < 0.001) and stress testing (unadjusted 15.8% to 12.7%, P < 0.001) significantly declined after ePATH implementation. After comparison with the comparator site, ePATH was still associated with a significant reduction in hospital admissions (-10.79%, P < 0.001) and stress testing (-6.05%, P < 0.001). Hospital length of stay and rates of major adverse cardiac events did not significantly change. CONCLUSIONS: Implementation of ePATH for patients presenting to the ED with chest pain was associated with safe reductions in hospital admission and stress testing. ePATH appears to be an effective tool for implementing evidence-based guidelines for ED patients with chest pain.

Improved antibiotic prescribing using indication-based clinical decision support in the emergency department.

BACKGROUND: Evaluate an indication-based clinical decision support tool to improve antibiotic prescribing in the emergency department. METHODS: Encounters where an antibiotic was prescribed between January 2015 and October 2017 were analyzed before and after the introduction of a clinical decision support tool to improve clinicians' selection of a guideline-approved antibiotic based on clinical indication. Evaluation was conducted on a pre-defined subset of conditions that included skin and soft tissue infections, respiratory infections, and urinary infections. The primary outcome was ordering of a guideline-approved antibiotic prescription at the drug and duration of therapy level. A mixed model following a binomial distribution with a logit link was used to model the difference in proportions of guideline-approved prescriptions before and after the intervention. RESULTS: For conditions evaluated, selection rate of a guideline-approved antibiotic for a given indication improved from 67.1% to 72.2% (P < 0.001). When duration of therapy is included as a criterion, selection of a guideline-approved antibiotic was lower and improved from 24.7% to 31.4% (P < 0.001), highlighting that duration of therapy is often missing at the time of prescribing. The most substantial improvements were seen for pneumonia and pyelonephritis with an increase from 87.9% to 97.5% and 62.8% to 82.6%, respectively. Other significant improvements were seen for abscess, cellulitis, and urinary tract infections. CONCLUSION: Antibiotic prescribing can be improved both at the drug and duration of therapy level using a non-interruptive and indication based-clinical decision support approach. Future research and quality improvement efforts are needed to incorporate duration of therapy guidelines into the antibiotic prescribing process.

Physician use of speech recognition versus typing in clinical documentation: A controlled observational study.

IMPORTANCE: Speech recognition (SR) is increasingly used directly by clinicians for electronic health record (EHR) documentation. Its usability and effect on quality and efficiency versus other documentation methods remain unclear. OBJECTIVE: To study usability and quality of documentation with SR versus typing. DESIGN: In this controlled observational study, each subject participated in two of five simulated outpatient scenarios. Sessions were recorded with Morae® usability software. Two notes were documented into the EHR per encounter (one dictated, one typed) in randomized order. Participants were interviewed about each method's perceived advantages and disadvantages. Demographics and documentation habits were collected via survey. Data collection occurred between January 8 and February 8, 2019, and data analysis was conducted from February through September of 2019. SETTING: Brigham and Women's Hospital, Boston, Massachusetts, USA. PARTICIPANTS: Ten physicians who had used SR for at least six months. MAIN OUTCOMES AND MEASURES: Documentation time, word count, vocabulary size, number of errors, number of corrections and quality (clarity, completeness, concision, information sufficiency and prioritization). RESULTS: Dictated notes were longer than typed notes (320.6 vs. 180.8 words; p = 0.004) with more unique words (170.9 vs. 120.4; p = 0.01). Documentation time was similar between methods, with dictated notes taking slightly less time to complete than typed notes. Typed notes had more uncorrected errors per note than dictated notes (2.9 vs. 1.5), although most were minor misspellings. Dictated notes had a higher mean quality score (7.7 vs. 6.6; p = 0.04), were more complete and included more sufficient information. CONCLUSIONS AND RELEVANCE: Participants felt that SR saves them time, increases their efficiency and allows them to quickly document more relevant details. Quality analysis supports the perception that SR allows for more detailed notes, but whether dictation is objectively faster than typing remains unclear, and participants described some scenarios where typing is still preferred. Dictation can be effective for creating comprehensive documentation, especially when physicians like and feel comfortable using SR. Research is needed to further improve integration of SR with EHR systems and assess its impact on clinical practice, workflows, provider and patient experience, and costs.

A dynamic reaction picklist for improving allergy reaction documentation in the electronic health record.

OBJECTIVE: Incomplete and static reaction picklists in the allergy module led to free-text and missing entries that inhibit the clinical decision support intended to prevent adverse drug reactions. We developed a novel, data-driven, "dynamic" reaction picklist to improve allergy documentation in the electronic health record (EHR). MATERIALS AND METHODS: We split 3 decades of allergy entries in the EHR of a large Massachusetts healthcare system into development and validation datasets. We consolidated duplicate allergens and those with the same ingredients or allergen groups. We created a reaction value set via expert review of a previously developed value set and then applied natural language processing to reconcile reactions from structured and free-text entries. Three association rule-mining measures were used to develop a comprehensive reaction picklist dynamically ranked by allergen. The dynamic picklist was assessed using recall at top k suggested reactions, comparing performance to the static picklist. RESULTS: The modified reaction value set contained 490 reaction concepts. Among 4 234 327 allergy entries collected, 7463 unique consolidated allergens and 469 unique reactions were identified. Of the 3 dynamic reaction picklists developed, the 1 with the optimal ranking achieved recalls of 0.632, 0.763, and 0.822 at the top 5, 10, and 15, respectively, significantly outperforming the static reaction picklist ranked by reaction frequency. CONCLUSION: The dynamic reaction picklist developed using EHR data and a statistical measure was superior to the static picklist and suggested proper reactions for allergy documentation. Further studies might evaluate the usability and impact on allergy documentation in the EHR.