Virtual Care: Perspectives From Family Physicians.

BACKGROUND: During the COVID-19 pandemic, virtual care expanded rapidly at Michigan Medicine and other health systems. From family physicians' perspectives, this shift to virtual care has the potential to affect workflow, job satisfaction, and patient communication. As clinics reopened and care delivery models shifted to a combination of in-person and virtual care, the need to understand physician experiences with virtual care arose in order to improve both patient and provider experiences. This study investigated Michigan Medicine family medicine physicians' perceptions of virtual care through qualitative interviews to better understand how to improve the quality and effectiveness of virtual care for both patients and physicians. METHODS: We employed a qualitative descriptive design to examine physician perspectives through semistructured interviews. We coded and analyzed transcripts using thematic analysis, facilitated by MAXQDA (VERBI) software. RESULTS: The results of the analysis identified four major themes: (a) chief concerns that are appropriate for virtual evaluation, (b) physician perceptions of patient benefits, (c) focused but contextually enriched patient-physician communication, and (d) structural support needed for high-quality virtual care. CONCLUSIONS: These findings can help further direct the discussion of how to make use of resources to improve the quality and effectiveness of virtual care.

Automated-detection of risky alcohol use prior to surgery using natural language processing.

BACKGROUND: Preoperative risky alcohol use is one of the most common surgical risk factors. Accurate and early identification of risky alcohol use could enhance surgical safety. Artificial Intelligence-based approaches, such as natural language processing (NLP), provide an innovative method to identify alcohol-related risks from patients' electronic health records (EHR) before surgery. METHODS: Clinical notes (n = 53,629) from pre-operative patients in a tertiary care facility were analyzed for evidence of risky alcohol use and alcohol use disorder. One hundred of these records were reviewed by experts and labeled for comparison. A rule-based NLP model was built, and we assessed the clinical notes for the entire population. Additionally, we assessed each record for the presence or absence of alcohol-related International Classification of Diseases (ICD) diagnosis codes as an additional comparator. RESULTS: NLP correctly identified 87% of the human-labeled patients classified with risky alcohol use. In contrast, diagnosis codes alone correctly identified only 29% of these patients. In terms of specificity, NLP correctly identified 84% of the non-risky cohort, while diagnosis codes correctly identified 90% of this cohort. In the analysis of the full dataset, the NLP-based approach identified three times more patients with risky alcohol use than ICD codes. CONCLUSIONS: NLP, an artificial intelligence-based approach, efficiently and accurately identifies alcohol-related risk in patients' EHRs. This approach could supplement other alcohol screening tools to identify patients in need of intervention, treatment, and/or postoperative withdrawal prophylaxis. Alcohol-related ICD diagnosis had limited utility relative to NLP, which extracts richer information within clinical notes to classify patients.

Social Acceptability of Health Behavior Posts on Social Media: An Experiment.

INTRODUCTION: Social media sites like Twitter (now X) are increasingly used to create health behavior metrics for public health surveillance. Yet little is known about social norms that may bias the content of posts about health behaviors. Social norms for posts about four health behaviors (smoking tobacco, drinking alcohol, physical activity, eating food) on Twitter/X were evaluated. METHODS: This was a randomized experiment delivered via web-based survey to adult, English-speaking Twitter/X users in three Michigan, USA, counties from 2020 to 2022 (n=559). Each participant viewed 24 posts presenting experimental manipulations regarding four health behaviors and answered questions about each post's social acceptability. Principal component analysis was used to combine survey responses into one perceived social acceptability measure. Linear mixed models with the Benjamini-Hochberg correction were implemented to test seven study hypotheses in 2023. RESULTS: Supporting six hypotheses, posts presenting healthier (CI: 0.028, 0.454), less stigmatized behaviors (CI: 0.552, 0.157) were more socially acceptable than posts regarding unhealthier, stigmatized behaviors. Unhealthy (CI: -0.268, -0.109) and stigmatized behavior (CI: -0.261, -0.103) posts were less acceptable for more educated participants. Posts about collocated activities (CI: 0.410, 0.573) and accompanied by expressions of liking (CI: 0.906, 1.11) were more acceptable than activities undertaken alone or disliked. Contrary to one hypothesis, posts reporting unusual activities were less acceptable than usual ones (CI: -0.472, 0.312). CONCLUSIONS: Perceived social acceptability may be associated with the frequency and content of health behavior posts. Users of Twitter/X and other social media platform posts to estimate health behavior prevalence should account for potential estimation biases from perceived social acceptability of posts.

Applying AI and Guidelines to Assist Medical Students in Recognizing Patients With Heart Failure: Protocol for a Randomized Trial.

BACKGROUND: The integration of artificial intelligence (AI) into clinical practice is transforming both clinical practice and medical education. AI-based systems aim to improve the efficacy of clinical tasks, enhancing diagnostic accuracy and tailoring treatment delivery. As it becomes increasingly prevalent in health care for high-quality patient care, it is critical for health care providers to use the systems responsibly to mitigate bias, ensure effective outcomes, and provide safe clinical practices. In this study, the clinical task is the identification of heart failure (HF) prior to surgery with the intention of enhancing clinical decision-making skills. HF is a common and severe disease, but detection remains challenging due to its subtle manifestation, often concurrent with other medical conditions, and the absence of a simple and effective diagnostic test. While advanced HF algorithms have been developed, the use of these AI-based systems to enhance clinical decision-making in medical education remains understudied. OBJECTIVE: This research protocol is to demonstrate our study design, systematic procedures for selecting surgical cases from electronic health records, and interventions. The primary objective of this study is to measure the effectiveness of interventions aimed at improving HF recognition before surgery, the second objective is to evaluate the impact of inaccurate AI recommendations, and the third objective is to explore the relationship between the inclination to accept AI recommendations and their accuracy. METHODS: Our study used a 3 × 2 factorial design (intervention type × order of prepost sets) for this randomized trial with medical students. The student participants are asked to complete a 30-minute e-learning module that includes key information about the intervention and a 5-question quiz, and a 60-minute review of 20 surgical cases to determine the presence of HF. To mitigate selection bias in the pre- and posttests, we adopted a feature-based systematic sampling procedure. From a pool of 703 expert-reviewed surgical cases, 20 were selected based on features such as case complexity, model performance, and positive and negative labels. This study comprises three interventions: (1) a direct AI-based recommendation with a predicted HF score, (2) an indirect AI-based recommendation gauged through the area under the curve metric, and (3) an HF guideline-based intervention. RESULTS: As of July 2023, 62 of the enrolled medical students have fulfilled this study's participation, including the completion of a short quiz and the review of 20 surgical cases. The subject enrollment commenced in August 2022 and will end in December 2023, with the goal of recruiting 75 medical students in years 3 and 4 with clinical experience. CONCLUSIONS: We demonstrated a study protocol for the randomized trial, measuring the effectiveness of interventions using AI and HF guidelines among medical students to enhance HF recognition in preoperative care with electronic health record data. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/49842.

Identifying Inequities in Video and Audio Telehealth Services for Primary Care Encounters During COVID-19: Repeated Cross-Sectional, Observational Study.

BACKGROUND: The COVID-19 pandemic resulted in rapid changes in how patient care was provided, particularly through the expansion of telehealth and audio-only phone-based care. OBJECTIVE: The goal of this study was to evaluate inequities in video and audio-only care during various time points including the initial wave of the COVID-19 pandemic, later stages of the pandemic, and a historical control. We sought to understand the characteristics of care during this time for a variety of different groups of patients that may experience health care inequities. METHODS: We conducted a retrospective analysis of electronic health record (EHR) data from encounters from 34 family medicine and internal medicine primary care clinics in a large, Midwestern health system, using a repeated cross-sectional, observational study design. These data included patient demographic data, as well as encounter, diagnosis, and procedure records. Data were obtained for all in-person and telehealth encounters (including audio-only phone-based care) that occurred during 3 separate time periods: an initial COVID-19 period (T2: March 16, 2020, to May 3, 2020), a later COVID-19 period (T3: May 4, 2020, to September 30, 2020), and a historical control period from the previous year (T1: March 16, 2019, to September 30, 2019). Primary analysis focused on the status of each encounter in terms of whether it was completed as scheduled, it was canceled, or the patient missed the appointment. A secondary analysis was performed to evaluate the likelihood of an encounter being completed based on visit modality (phone, video, in-person). RESULTS: In total, there were 938,040 scheduled encounters during the 3 time periods, with 178,747 unique patients, that were included for analysis. Patients with completed encounters were more likely to be younger than 65 years old (71.8%-74.1%), be female (58.8%-61.8%), be White (75.6%-76.7%), and have no significant comorbidities (63.2%-66.8%) or disabilities (53.2%-61.1%) in all time periods than those who had only canceled or missed encounters. Effects on different subpopulations are discussed herein. CONCLUSIONS: Findings from this study demonstrate that primary care utilization across delivery modalities (in person, video, and phone) was not equivalent across all groups before and during the COVID-19 pandemic and different groups were differentially impacted at different points. Understanding how different groups of patients responded to these rapid changes and how health care inequities may have been affected is an important step in better understanding implementation strategies for digital solutions in the future.

Developing a Semantically Based Query Recommendation for an Electronic Medical Record Search Engine: Query Log Analysis and Design Implications.

BACKGROUND: An effective and scalable information retrieval (IR) system plays a crucial role in enabling clinicians and researchers to harness the valuable information present in electronic health records. In a previous study, we developed a prototype medical IR system, which incorporated a semantically based query recommendation (SBQR) feature. The system was evaluated empirically and demonstrated high perceived performance by end users. To delve deeper into the factors contributing to this perceived performance, we conducted a follow-up study using query log analysis. OBJECTIVE: One of the primary challenges faced in IR is that users often have limited knowledge regarding their specific information needs. Consequently, an IR system, particularly its user interface, needs to be thoughtfully designed to assist users through the iterative process of refining their queries as they encounter relevant documents during their search. To address these challenges, we incorporated "query recommendation" into our Electronic Medical Record Search Engine (EMERSE), drawing inspiration from the success of similar features in modern IR systems for general purposes. METHODS: The query log data analyzed in this study were collected during our previous experimental study, where we developed EMERSE with the SBQR feature. We implemented a logging mechanism to capture user query behaviors and the output of the IR system (retrieved documents). In this analysis, we compared the initial query entered by users with the query formulated with the assistance of the SBQR. By examining the results of this comparison, we could examine whether the use of SBQR helped in constructing improved queries that differed from the original ones. RESULTS: Our findings revealed that the first query entered without SBQR and the final query with SBQR assistance were highly similar (Jaccard similarity coefficient=0.77). This suggests that the perceived positive performance of the system was primarily attributed to the automatic query expansion facilitated by the SBQR rather than users manually manipulating their queries. In addition, through entropy analysis, we observed that search results converged in scenarios of moderate difficulty, and the degree of convergence correlated strongly with the perceived system performance. CONCLUSIONS: The study demonstrated the potential contribution of the SBQR in shaping participants' positive perceptions of system performance, contingent upon the difficulty of the search scenario. Medical IR systems should therefore consider incorporating an SBQR as a user-controlled option or a semiautomated feature. Future work entails redesigning the experiment in a more controlled manner and conducting multisite studies to demonstrate the effectiveness of EMERSE with SBQR for patient cohort identification. By further exploring and validating these findings, we can enhance the usability and functionality of medical IR systems in real-world settings.

Dementia and electronic health record phenotypes: a scoping review of available phenotypes and opportunities for future research.

OBJECTIVE: We performed a scoping review of algorithms using electronic health record (EHR) data to identify patients with Alzheimer's disease and related dementias (ADRD), to advance their use in research and clinical care. MATERIALS AND METHODS: Starting with a previous scoping review of EHR phenotypes, we performed a cumulative update (April 2020 through March 1, 2023) using Pubmed, PheKB, and expert review with exclusive focus on ADRD identification. We included algorithms using EHR data alone or in combination with non-EHR data and characterized whether they identified patients at high risk of or with a current diagnosis of ADRD. RESULTS: For our cumulative focused update, we reviewed 271 titles meeting our search criteria, 49 abstracts, and 26 full text papers. We identified 8 articles from the original systematic review, 8 from our new search, and 4 recommended by an expert. We identified 20 papers describing 19 unique EHR phenotypes for ADRD: 7 algorithms identifying patients with diagnosed dementia and 12 algorithms identifying patients at high risk of dementia that prioritize sensitivity over specificity. Reference standards range from only using other EHR data to in-person cognitive screening. CONCLUSION: A variety of EHR-based phenotypes are available for use in identifying populations with or at high-risk of developing ADRD. This review provides comparative detail to aid in choosing the best algorithm for research, clinical care, and population health projects based on the use case and available data. Future research may further improve the design and use of algorithms by considering EHR data provenance.

Automatic Extraction of Medication Mentions from Tweets-Overview of the BioCreative VII Shared Task 3 Competition.

This study presents the outcomes of the shared task competition BioCreative VII (Task 3) focusing on the extraction of medication names from a Twitter user's publicly available tweets (the user's 'timeline'). In general, detecting health-related tweets is notoriously challenging for natural language processing tools. The main challenge, aside from the informality of the language used, is that people tweet about any and all topics, and most of their tweets are not related to health. Thus, finding those tweets in a user's timeline that mention specific health-related concepts such as medications requires addressing extreme imbalance. Task 3 called for detecting tweets in a user's timeline that mentions a medication name and, for each detected mention, extracting its span. The organizers made available a corpus consisting of 182 049 tweets publicly posted by 212 Twitter users with all medication mentions manually annotated. The corpus exhibits the natural distribution of positive tweets, with only 442 tweets (0.2%) mentioning a medication. This task was an opportunity for participants to evaluate methods that are robust to class imbalance beyond the simple lexical match. A total of 65 teams registered, and 16 teams submitted a system run. This study summarizes the corpus created by the organizers and the approaches taken by the participating teams for this challenge. The corpus is freely available at https://biocreative.bioinformatics.udel.edu/tasks/biocreative-vii/track-3/. The methods and the results of the competing systems are analyzed with a focus on the approaches taken for learning from class-imbalanced data.

A Systematic Temporal Extraction Pipeline for Medical Concepts in Clinical Notes.

With increased application of natural language processing (NLP) in medicine, many NLP models are being developed for uncovering relevant clinical features from electronic health records. Temporal information plays a key role in understanding the context, significance, and interpretation of medical concepts extracted from clinical notes. This is particularly true in situations where the behavior, value, or status of a medical concept changes over time. In this paper, we introduce a systematic framework, NLP annotation-Relaxation-Generation (NRG). NRG compiles incidents of medical concept changes from status annotations and timestamps of multiple clinical notes. We demonstrate the effectiveness of the NRG pipeline by applying it to two medical concepts related to patients with inflammatory bowel disease: extra-intestinal manifestations and medications. We show that the NRG pipeline offers not only insights into medical concept changes over time, but can help convey longitudinal changes in clinical features at both individual and population level.

An Assessment of Mentions of Adverse Drug Events on Social Media With Natural Language Processing: Model Development and Analysis.

BACKGROUND: Adverse reactions to drugs attract significant concern in both clinical practice and public health monitoring. Multiple measures have been put into place to increase postmarketing surveillance of the adverse effects of drugs and to improve drug safety. These measures include implementing spontaneous reporting systems and developing automated natural language processing systems based on data from electronic health records and social media to collect evidence of adverse drug events that can be further investigated as possible adverse reactions. OBJECTIVE: While using social media for collecting evidence of adverse drug events has potential, it is not clear whether social media are a reliable source for this information. Our work aims to (1) develop natural language processing approaches to identify adverse drug events on social media and (2) assess the reliability of social media data to identify adverse drug events. METHODS: We propose a collocated long short-term memory network model with attentive pooling and aggregated, contextual representation generated by a pretrained model. We applied this model on large-scale Twitter data to identify adverse drug event-related tweets. We conducted a qualitative content analysis of these tweets to validate the reliability of social media data as a means to collect such information. RESULTS: The model outperformed a variant without contextual representation during both the validation and evaluation phases. Through the content analysis of adverse drug event tweets, we observed that adverse drug event-related discussions had 7 themes. Mental health-related, sleep-related, and pain-related adverse drug event discussions were most frequent. We also contrast known adverse drug reactions to those mentioned in tweets. CONCLUSIONS: We observed a distinct improvement in the model when it used contextual information. However, our results reveal weak generalizability of the current systems to unseen data. Additional research is needed to fully utilize social media data and improve the robustness and reliability of natural language processing systems. The content analysis, on the other hand, showed that Twitter covered a sufficiently wide range of adverse drug events, as well as known adverse reactions, for the drugs mentioned in tweets. Our work demonstrates that social media can be a reliable data source for collecting adverse drug event mentions.

Multicenter Validation of Natural Language Processing Algorithms for the Detection of Common Data Elements in Operative Notes for Total Hip Arthroplasty: Algorithm Development and Validation.

BACKGROUND: Natural language processing (NLP) methods are powerful tools for extracting and analyzing critical information from free-text data. MedTaggerIE, an open-source NLP pipeline for information extraction based on text patterns, has been widely used in the annotation of clinical notes. A rule-based system, MedTagger-total hip arthroplasty (THA), developed based on MedTaggerIE, was previously shown to correctly identify the surgical approach, fixation, and bearing surface from the THA operative notes at Mayo Clinic. OBJECTIVE: This study aimed to assess the implementability, usability, and portability of MedTagger-THA at two external institutions, Michigan Medicine and the University of Iowa, and provide lessons learned for best practices. METHODS: We conducted iterative test-apply-refinement processes with three involved sites-the development site (Mayo Clinic) and two deployment sites (Michigan Medicine and the University of Iowa). Mayo Clinic was the primary NLP development site, with the THA registry as the gold standard. The activities at the two deployment sites included the extraction of the operative notes, gold standard development (Michigan: registry data; Iowa: manual chart review), the refinement of NLP algorithms on training data, and the evaluation of test data. Error analyses were conducted to understand language variations across sites. To further assess the model specificity for approach and fixation, we applied the refined MedTagger-THA to arthroscopic hip procedures and periacetabular osteotomy cases, as neither of these operative notes should contain any approach or fixation keywords. RESULTS: MedTagger-THA algorithms were implemented and refined independently for both sites. At Michigan, the study comprised THA-related notes for 2569 patient-date pairs. Before model refinement, MedTagger-THA algorithms demonstrated excellent accuracy for approach (96.6%, 95% CI 94.6%-97.9%) and fixation (95.7%, 95% CI 92.4%-97.6%). These results were comparable with internal accuracy at the development site (99.2% for approach and 90.7% for fixation). Model refinement improved accuracies slightly for both approach (99%, 95% CI 97.6%-99.6%) and fixation (98%, 95% CI 95.3%-99.3%). The specificity of approach identification was 88.9% for arthroscopy cases, and the specificity of fixation identification was 100% for both periacetabular osteotomy and arthroscopy cases. At the Iowa site, the study comprised an overall data set of 100 operative notes (50 training notes and 50 test notes). MedTagger-THA algorithms achieved moderate-high performance on the training data. After model refinement, the model achieved high performance for approach (100%, 95% CI 91.3%-100%), fixation (98%, 95% CI 88.3%-100%), and bearing surface (92%, 95% CI 80.5%-97.3%). CONCLUSIONS: High performance across centers was achieved for the MedTagger-THA algorithms, demonstrating that they were sufficiently implementable, usable, and portable to different deployment sites. This study provided important lessons learned during the model deployment and validation processes, and it can serve as a reference for transferring rule-based electronic health record models.

Identifying Caregiver Availability Using Medical Notes With Rule-Based Natural Language Processing: Retrospective Cohort Study.

BACKGROUND: Identifying caregiver availability, particularly for patients with dementia or those with a disability, is critical to informing the appropriate care planning by the health systems, hospitals, and providers. This information is not readily available, and there is a paucity of pragmatic approaches to automatically identifying caregiver availability and type. OBJECTIVE: Our main objective was to use medical notes to assess caregiver availability and type for hospitalized patients with dementia. Our second objective was to identify whether the patient lived at home or resided at an institution. METHODS: In this retrospective cohort study, we used 2016-2019 telephone-encounter medical notes from a single institution to develop a rule-based natural language processing (NLP) algorithm to identify the patient's caregiver availability and place of residence. Using note-level data, we compared the results of the NLP algorithm with human-conducted chart abstraction for both training (749/976, 77%) and test sets (227/976, 23%) for a total of 223 adults aged 65 years and older diagnosed with dementia. Our outcomes included determining whether the patients (1) reside at home or in an institution, (2) have a formal caregiver, and (3) have an informal caregiver. RESULTS: Test set results indicated that our NLP algorithm had high level of accuracy and reliability for identifying whether patients had an informal caregiver (F1=0.94, accuracy=0.95, sensitivity=0.97, and specificity=0.93), but was relatively less able to identify whether the patient lived at an institution (F1=0.64, accuracy=0.90, sensitivity=0.51, and specificity=0.98). The most common explanations for NLP misclassifications across all categories were (1) incomplete or misspelled facility names; (2) past, uncertain, or undecided status; (3) uncommon abbreviations; and (4) irregular use of templates. CONCLUSIONS: This innovative work was the first to use medical notes to pragmatically determine caregiver availability. Our NLP algorithm identified whether hospitalized patients with dementia have a formal or informal caregiver and, to a lesser extent, whether they lived at home or in an institutional setting. There is merit in using NLP to identify caregivers. This study serves as a proof of concept. Future work can use other approaches and further identify caregivers and the extent of their availability.

Predicting persistent opioid use after surgery using electronic health record and patient-reported data.

BACKGROUND: More than 100 million surgeries take place annually in the United States, and more than 90% of surgical patients receive an opioid prescription. A sizable minority of these patients will go on to use opioids long-term, contributing to the national opioid epidemic. METHODS: The objective of this study was to develop and validate a model to predict persistent opioid use after surgery. Participants included surgical patients (≥18 years old) enrolled in a cohort study at an academic medical center between 2015 and 2018. Persistent opioid use was defined as filling opioid prescriptions in postdischarge days 4 to 90 and 91 to 180. Predictors included electronic health record data, state prescription drug monitoring data, and patient-reported measures. Three models were developed: a full, a restricted, and a minimal model using a derivation and validation cohort. RESULTS: Of 24,040 patients, 4,879 (20%) experienced persistent opioid use. In the validation cohort, the full, restricted, and minimal model had C-statistics of 0.87 (95% CI 0.86-0.88), 0.86 (0.85-0.88), and 0.85 (0.84-0.87), respectively. All models performed better among patients with preoperative opioid use compared to opioid-naive patients (P < .001). The models slightly overpredicted risk in the validation cohort. The net benefit of using the restricted model to refer patients for preoperative counseling was 0.072 to 0.092, which is superior to evaluating no patients (net benefit of 0) or all patients (net benefit of -0.22 to -0.63). CONCLUSION: This study developed and validated a prediction model for persistent opioid use using accessible data resources. The models achieved strong performance, outperforming prior published models.

Neural Machine Translation-Based Automated Current Procedural Terminology Classification System Using Procedure Text: Development and Validation Study.

BACKGROUND: Administrative costs for billing and insurance-related activities in the United States are substantial. One critical cause of the high overhead of administrative costs is medical billing errors. With advanced deep learning techniques, developing advanced models to predict hospital and professional billing codes has become feasible. These models can be used for administrative cost reduction and billing process improvements. OBJECTIVE: In this study, we aim to develop an automated anesthesiology current procedural terminology (CPT) prediction system that translates manually entered surgical procedure text into standard forms using neural machine translation (NMT) techniques. The standard forms are calculated using similarity scores to predict the most appropriate CPT codes. Although this system aims to enhance medical billing coding accuracy to reduce administrative costs, we compare its performance with that of previously developed machine learning algorithms. METHODS: We collected and analyzed all operative procedures performed at Michigan Medicine between January 2017 and June 2019 (2.5 years). The first 2 years of data were used to train and validate the existing models and compare the results from the NMT-based model. Data from 2019 (6-month follow-up period) were then used to measure the accuracy of the CPT code prediction. Three experimental settings were designed with different data types to evaluate the models. Experiment 1 used the surgical procedure text entered manually in the electronic health record. Experiment 2 used preprocessing of the procedure text. Experiment 3 used preprocessing of the combined procedure text and preoperative diagnoses. The NMT-based model was compared with the support vector machine (SVM) and long short-term memory (LSTM) models. RESULTS: The NMT model yielded the highest top-1 accuracy in experiments 1 and 2 at 81.64% and 81.71% compared with the SVM model (81.19% and 81.27%, respectively) and the LSTM model (80.96% and 81.07%, respectively). The SVM model yielded the highest top-1 accuracy of 84.30% in experiment 3, followed by the LSTM model (83.70%) and the NMT model (82.80%). In experiment 3, the addition of preoperative diagnoses showed 3.7%, 3.2%, and 1.3% increases in the SVM, LSTM, and NMT models in top-1 accuracy over those in experiment 2, respectively. For top-3 accuracy, the SVM, LSTM, and NMT models achieved 95.64%, 95.72%, and 95.60% for experiment 1, 95.75%, 95.67%, and 95.69% for experiment 2, and 95.88%, 95.93%, and 95.06% for experiment 3, respectively. CONCLUSIONS: This study demonstrates the feasibility of creating an automated anesthesiology CPT classification system based on NMT techniques using surgical procedure text and preoperative diagnosis. Our results show that the performance of the NMT-based CPT prediction system is equivalent to that of the SVM and LSTM prediction models. Importantly, we found that including preoperative diagnoses improved the accuracy of using the procedure text alone.

Human versus machine editing of electronic prescription directions.

BACKGROUND: Pharmacy staff are responsible for editing poor-quality and difficult-to-read electronic prescription (e-prescription) directions. Machine translation (MT) models are capable of translating free text from 1 sequence into another. However, the quality of MTs of e-prescriptions into pharmacy label directions is unknown. OBJECTIVE: To determine the types and frequencies of e-prescription direction component errors made by an MT model, pharmacy staff, and prescribers. METHODS: A prospective evaluation was conducted on a random sample of 300 patient directions in a test set of e-prescriptions from a mail-order pharmacy. Each row included directions produced by (1) prescribers on e-prescriptions, (2) pharmacy staff on prescription labels, and (3) an open neural MT model. Annotators labeled direction sets for missing direction components, use of abbreviations and medical jargon, and incorrect information (e.g., changing the number of tablets to be taken). The longest common subsequence (LCS) compared the amount of pharmacy staff editing with and without MT. RESULTS: Out of 279 direction sets labeled, the MT model directions contained no quality issues in 196 (70.3%) samples compared with 187 (67.0%) and 83 (29.8%) samples for pharmacy staff directions and prescriber directions, respectively. The MT model directions contained more incorrect components (n = 23). Median LCS was greater without MT (30.0 vs. 18.5, P < 0.01, Wilcoxon signed-rank test), indicating more editing was needed. CONCLUSION: MT could be used to improve the quality of e-prescription directions; however, MT makes high-risk mistakes such as incorrectly predicting the tapering regimen for prednisone. The use of semiautomated MT, where pharmacy staff can review model predictions to detect and resolve quality issues, should be considered to improve safety and decrease total work time compared with current practice. MT has strengths and weaknesses for improving the editing process of the patient directions compared with pharmacy staff alone.

Work effort, readability and quality of pharmacy transcription of patient directions from electronic prescriptions: a retrospective observational cohort analysis.

BACKGROUND: Free-text directions generated by prescribers in electronic prescriptions can be difficult for patients to understand due to their variability, complexity and ambiguity. Pharmacy staff are responsible for transcribing these directions so that patients can take their medication as prescribed. However, little is known about the quality of these transcribed directions received by patients. METHODS: A retrospective observational analysis of 529 990 e-prescription directions processed at a mail-order pharmacy in the USA. We measured pharmacy staff editing of directions using string edit distance and execution time using the Keystroke-Level Model. Using the New Dale-Chall (NDC) readability formula, we calculated NDC cloze scores of the patient directions before and after transcription. We also evaluated the quality of directions (eg, included a dose, dose unit, frequency of administration) before and after transcription with a random sample of 966 patient directions. RESULTS: Pharmacy staff edited 83.8% of all e-prescription directions received with a median edit distance of 18 per e-prescription. We estimated a median of 6.64 s of transcribing each e-prescription. The median NDC score increased by 68.6% after transcription (26.12 vs 44.03, p<0.001), which indicated a significant readability improvement. In our sample, 51.4% of patient directions on e-prescriptions contained at least one pre-defined direction quality issue. Pharmacy staff corrected 79.5% of the quality issues. CONCLUSION: Pharmacy staff put significant effort into transcribing e-prescription directions. Manual transcription removed the majority of quality issues; however, pharmacy staff still miss or introduce following their manual transcription processes. The development of tools and techniques such as a comprehensive set of structured direction components or machine learning-based natural language processing techniques may help produce clear directions.

Using Healthcare Data in Embedded Pragmatic Clinical Trials among People Living with Dementia and Their Caregivers: State of the Art.

Embedded pragmatic clinical trials (ePCTs) are embedded in healthcare systems as well as their data environments. For people living with dementia (PLWD), settings of care can be different from the general population and involve additional people whose information is also important. The ePCT designs have the opportunity to leverage data that becomes available through the normal delivery of care. They may be particularly valuable in Alzheimer's disease and Alzheimer's disease-related dementia (AD/ADRD), given the complexity of case identification and the diversity of care settings. Grounded in the objectives of the Data and Technical Core of the newly established National Institute on Aging Imbedded Pragmatic Alzheimer's Disease and AD-Related Dementias Clinical Trials Collaboratory (IMPACT Collaboratory), this article summarizes the state of the art in using existing data sources (eg, Medicare claims, electronic health records) in AD/ADRD ePCTs and approaches to integrating them in real-world settings. J Am Geriatr Soc 68:S49-S54, 2020.

Uncovering the relationship between food-related discussion on Twitter and neighborhood characteristics.

OBJECTIVE: Initiatives to reduce neighborhood-based health disparities require access to meaningful, timely, and local information regarding health behavior and its determinants. We examined the validity of Twitter as a source of information for neighborhood-level analysis of dietary choices and attitudes. MATERIALS AND METHODS: We analyzed the "healthiness" quotient and sentiment in food-related tweets at the census tract level, and associated them with neighborhood characteristics and health outcomes. We analyzed keywords driving the differences in food healthiness between the most and least-affluent tracts, and qualitatively analyzed contents of a random sample of tweets. RESULTS: Significant, albeit weak, correlations existed between healthiness and sentiment in food-related tweets and tract-level measures of affluence, disadvantage, race, age, U.S. density, and mortality from conditions associated with obesity. Analyses of keywords driving the differences in food healthiness revealed foods high in saturated fat (eg, pizza, bacon, fries) were mentioned more frequently in less-affluent tracts. Food-related discussion referred to activities (eating, drinking, cooking), locations where food was consumed, and positive (affection, cravings, enjoyment) and negative attitudes (dislike, personal struggles, complaints). DISCUSSION: Tweet-based healthiness scores largely correlated with offline phenomena in the expected directions. Social media offer less resource-intensive data collection methods than traditional surveys do. Twitter may assist in informing local health programs that focus on drivers of food consumption and could inform interventions focused on attitudes and the food environment. CONCLUSIONS: Twitter provided weak but significant signals concerning food-related behavior and attitudes at the neighborhood level, suggesting its potential usefulness for informing local health disparity reduction efforts.

Hybrid bag of approaches to characterize selection criteria for cohort identification.

OBJECTIVE: The 2018 National NLP Clinical Challenge (2018 n2c2) focused on the task of cohort selection for clinical trials, where participating systems were tasked with analyzing longitudinal patient records to determine if the patients met or did not meet any of the 13 selection criteria. This article describes our participation in this shared task. MATERIALS AND METHODS: We followed a hybrid approach combining pattern-based, knowledge-intensive, and feature weighting techniques. After preprocessing the notes using publicly available natural language processing tools, we developed individual criterion-specific components that relied on collecting knowledge resources relevant for these criteria and pattern-based and weighting approaches to identify "met" and "not met" cases. RESULTS: As part of the 2018 n2c2 challenge, 3 runs were submitted. The overall micro-averaged F1 on the training set was 0.9444. On the test set, the micro-averaged F1 for the 3 submitted runs were 0.9075, 0.9065, and 0.9056. The best run was placed second in the overall challenge and all 3 runs were statistically similar to the top-ranked system. A reimplemented system achieved the best overall F1 of 0.9111 on the test set. DISCUSSION: We highlight the need for a focused resource-intensive effort to address the class imbalance in the cohort selection identification task. CONCLUSION: Our hybrid approach was able to identify all selection criteria with high F1 performance on both training and test sets. Based on our participation in the 2018 n2c2 task, we conclude that there is merit in continuing a focused criterion-specific analysis and developing appropriate knowledge resources to build a quality cohort selection system.