Adverse Event Signal Detection Using Patients' Concerns in Pharmaceutical Care Records: Evaluation of Deep Learning Models

BACKGROUND

Early detection of adverse events and their management are crucial to improving anticancer treatment outcomes, and listening to patients' subjective opinions (patients' voices) can make a major contribution to improving safety management. Recent progress in deep learning technologies has enabled various new approaches for the evaluation of safety-related events based on patient-generated text data, but few studies have focused on the improvement of real-time safety monitoring for individual patients. In addition, no study has yet been performed to validate deep learning models for screening patients' narratives for clinically important adverse event signals that require medical intervention. In our previous work, novel deep learning models have been developed to detect adverse event signals for hand-foot syndrome or adverse events limiting patients' daily lives from the authored narratives of patients with cancer, aiming ultimately to use them as safety monitoring support tools for individual patients.

OBJECTIVE

This study was designed to evaluate whether our deep learning models can screen clinically important adverse event signals that require intervention by health care professionals. The applicability of our deep learning models to data on patients' concerns at pharmacies was also assessed.

METHODS

Pharmaceutical care records at community pharmacies were used for the evaluation of our deep learning models. The records followed the SOAP format, consisting of subjective (S), objective (O), assessment (A), and plan (P) columns. Because of the unique combination of patients' concerns in the S column and the professional records of the pharmacists, this was considered a suitable data for the present purpose. Our deep learning models were applied to the S records of patients with cancer, and the extracted adverse event signals were assessed in relation to medical actions and prescribed drugs.

RESULTS

From 30,784 S records of 2479 patients with at least 1 prescription of anticancer drugs, our deep learning models extracted true adverse event signals with more than 80% accuracy for both hand-foot syndrome (n=152, 91%) and adverse events limiting patients' daily lives (n=157, 80.1%). The deep learning models were also able to screen adverse event signals that require medical intervention by health care providers. The extracted adverse event signals could reflect the side effects of anticancer drugs used by the patients based on analysis of prescribed anticancer drugs. "Pain or numbness" (n=57, 36.3%), "fever" (n=46, 29.3%), and "nausea" (n=40, 25.5%) were common symptoms out of the true adverse event signals identified by the model for adverse events limiting patients' daily lives.

CONCLUSIONS

Our deep learning models were able to screen clinically important adverse event signals that require intervention for symptoms. It was also confirmed that these deep learning models could be applied to patients' subjective information recorded in pharmaceutical care records accumulated during pharmacists' daily work.

Identification of volatile organic compounds and their sources driving ozone and secondary organic aerosol formation in NE Spain

Volatile organic compounds (VOCs) play a crucial role in the formation of ozone (O3) and secondary organic aerosol (SOA). We conducted measurements of VOC ambient mixing ratios during both summer and winter at two stations: a Barcelona urban background station (BCN) and the Montseny rural background station (MSY). Subsequently, we employed positive matrix factorization (PMF) to analyze the VOC mixing ratios and identify their sources. Our analysis revealed five common sources: anthropogenic I (traffic & industries); anthropogenic II (traffic & biomass burning); isoprene oxidation; monoterpenes; long-lifetime VOCs. To assess the impact of these VOCs on the formation of secondary pollutants, we calculated the ozone formation potential (OFP) and secondary organic aerosol formation potential (SOAP) associated with each VOC. In conclusion, our study provides insights into the sources of VOCs and their contributions to the formation of ozone and SOA in NE Spain. The OFP was primarily influenced by anthropogenic aromatic compounds from the traffic & industries source at BCN (38-49 %) and during winter at MSY (34 %). In contrast, the summer OFP at MSY was primarily driven by biogenic contributions from monoterpenes and isoprene oxidation products (45 %). Acetaldehyde (10-35 %) and methanol (13-14 %) also made significant OFP contributions at both stations. Anthropogenic aromatic compounds originating from traffic, industries, and biomass burning played a dominant role (88-93 %) in SOA formation at both stations during both seasons. The only exception was during the summer at MSY, where monoterpenes became the primary driver of SOA formation (41 %). These findings emphasize the importance of considering both anthropogenic and biogenic VOCs in air quality management strategies.

Distribution of emergency medicine programs by location and reproductive rights: A cross-sectional study of the 2023 Match

INTRODUCTION

The 2023 Match saw over 500 unfilled positions in emergency medicine (EM). Geographic location is the third most important factor for all United States (US) EM-bound senior medical students when selecting programs to rank and can be affected by political climate. Given the perceived importance of geography on program selection and recent changes to reproductive rights in the US, we sought to evaluate the impact of geography and reproductive rights on unmatched positions among EM programs.

METHODS

This was a cross-sectional study assessing Match rates in EM by program US state, region, and degree of reproductive rights. We included all EM programs participating in the 2023 Match year. Our primary outcome was to determine the unfilled programs and positions rate per US state. Secondary outcomes included Match rates by region and by degree of reproductive rights.

RESULTS

We found notable differences in unfilled programs by US state, with the highest percentage of unfilled programs and positions in Arkansas (100%, 56.3%), Nevada (100%, 35.5%), Kansas (100%, 40.0%), Ohio (81.3%, 33.3%), and Michigan (80.0%, (36.8%). Among regions, the highest percentage of unfilled programs (62.5%) and residency positions (26.0%) was East North Central (IL, IN, MI, OH, WI). US states with limited reproductive rights had the highest percent (52.9%) of programs with unmatched positions and the highest percent (20.5%) of unfilled positions.

CONCLUSION

We identified notable differences in unmatched positions by US state and region, as well as the highest rate of unmatched positions in US states with more limited reproductive rights.

Comparing oral case presentation formats on internal medicine inpatient rounds: a survey study

BACKGROUND

Oral case presentations - structured verbal reports of clinical cases - are fundamental to patient care and learner education. Despite their continued importance in a modernized medical landscape, their structure has remained largely unchanged since the 1960s, based on the traditional Subjective, Objective, Assessment, Plan (SOAP) format developed for medical records. We developed a problem-based alternative known as Events, Assessment, Plan (EAP) to understand the perceived efficacy of EAP compared to SOAP among learners.

METHODS

We surveyed (Qualtrics, via email) all third- and fourth-year medical students and internal medicine residents at a large, academic, tertiary care hospital and associated Veterans Affairs medical center. The primary outcome was trainee preference in oral case presentation format. The secondary outcome was comparing EAP and SOAP on 10 functionality domains assessed via a 5-point Likert scale. We used descriptive statistics (proportion and mean) to describe the results.

RESULTS

The response rate was 21% (118/563). Of the 59 respondents with exposure to both the EAP and SOAP formats, 69% (n = 41) preferred the EAP format as compared to 19% (n = 11) who preferred SOAP (p < 0.001). EAP outperformed SOAP in 8 out of 10 of the domains assessed, including advancing patient care, learning from patients, and time efficiency.

CONCLUSIONS

Our findings suggest that trainees prefer the EAP format over SOAP and that EAP may facilitate clearer and more efficient communication on rounds, which in turn may enhance patient care and learner education. A broader, multi-center study of the EAP oral case presentation will help to better understand preferences, outcomes, and barriers to implementation.

Spotting Local Environments in Self-Assembled Monolayer-Protected Gold Nanoparticles

Organic-inorganic (O-I) nanomaterials are versatile platforms for an incredible high number of applications, ranging from heterogeneous catalysis to molecular sensing, cell targeting, imaging, and cancer diagnosis and therapy, just to name a few. Much of their potential stems from the unique control of organic environments around inorganic sites within a single O-I nanomaterial, which allows for new properties that were inaccessible using purely organic or inorganic materials. Structural and mechanistic characterization plays a key role in understanding and rationally designing such hybrid nanoconstructs. Here, we introduce a general methodology to identify and classify local (supra)molecular environments in an archetypal class of O-I nanomaterials, i.e., self-assembled monolayer-protected gold nanoparticles (SAM-AuNPs). By using an atomistic machine-learning guided workflow based on the Smooth Overlap of Atomic Positions (SOAP) descriptor, we analyze a collection of chemically different SAM-AuNPs and detect and compare local environments in a way that is agnostic and automated, i.e., with no need of a priori information and minimal user intervention. In addition, the computational results coupled with experimental electron spin resonance measurements prove that is possible to have more than one local environment inside SAMs, being the thickness of the organic shell and solvation primary factors in the determining number and nature of multiple coexisting environments. These indications are extended to complex mixed hydrophilic-hydrophobic SAMs. This work demonstrates that it is possible to spot and compare local molecular environments in SAM-AuNPs exploiting atomistic machine-learning approaches, establishes ground rules to control them, and holds the potential for the rational design of O-I nanomaterials instructed from data.

Development and evaluation of clinical reasoning using 'think aloud' approach in pharmacy undergraduates - A mixed-methods study

Introduction

Given the widespread use of clinical reasoning (CR) in the healthcare practice, it is essential to inculcate the CR practice in undergraduate pharmacy education which can not only facilitate their clinical education and clinical rotations but can also help them become better clinical pharmacists. There is very limited CR employed in the pharmacy curriculum and practice in the Middle East countries. This study aimed to develop and evaluate CR practice in pharmacy undergraduates in one college of pharmacy in Saudi Arabia.

Methods

We employed a mixed-methods methodology that included two phases. In Phase I, students were introduced to CR practice ('think aloud' method) and given geriatric clinical cases which they used in two sessions together with a tutor. This was followed by the writing of SOAP notes using the tutor feedback and completion of a survey that included a self-reflection about their experience of using the CR method. Phase II included face-to-face semi-structured interviews involving selected students that were recruited via convenience sampling to further explore the issues identified in Phase I of the study.

Results

Of the 155 students who completed the survey (response rate 94%), the majority of them agreed that CR using the 'think aloud' method was useful in gathering (92%) and interpreting (95%) relevant patient information, identifying medication-related problems (95%), exploring therapeutic options for the problem(s) (93%) and formulating a treatment plan for the patient (90%). Qualitative data analysis of the 12 interviews was consistent with these findings. Furthermore, it provided an insight into the challenges faced by the students in applying this CR method.

Conclusions

Students found the practice of CR using the 'think aloud' method helpful in working through given cases and taking clinical decisions. This method can be widely employed in pharmacy education and practice.

Early short-term prediction of emergency department length of stay using natural language processing for low-acuity outpatients

BACKGROUND

Low-acuity outpatients constitute the majority of emergency department (ED) patients, and these patients often experience an unpredictable length of stay (LOS). Effective LOS prediction might improve the quality of ED care and reduce ED crowding.

OBJECTIVE

The objective of this study was to explore the potential of natural language processing (NLP) of the first ED physicians' clinical notes and to evaluate NLP-based short-term prediction models based on mixed-type clinical data.

METHODS

A retrospective study was conducted at an ED of a tertiary teaching hospital in Taiwan from January 2017 to June 2017. In total, 12,962 low-acuity outpatients were enrolled. Using structured data (e.g., demographic variables and vital signs) and different sections of the first SOAP notes as predictors, we developed six NLP-based prediction models (i.e., term frequency-inverse document frequency (TF-IDF) and truncated singular value decomposition (SVD)) to predict LOS. The metric for model evaluation is the mean squared error (MSE).

RESULTS

Of the six NLP-based models, the model using structured data and all the sections of the first SOAP notes processed by the TF-IDF and truncated SVD method performed the best, with an MSE of 3.00 [95% CI: 2.94-3.06]. In addition, ten important topics extracted by the TF-IDF and truncated SVD method had significant effects on the LOS (p < 0.001).

CONCLUSION

NLP-based models can be used as an early short-term prediction of LOS and have the potential for mixed-type clinical data analysis. The proposed models would likely aid ED physicians' decision-making processes and improve ED quality of care.

What If I Do Not Match? Scrambling for a Spot and Going Outside the Match

Matching into emergency medicine (EM) is getting progressively more competitive. Applicants must therefore prepare for the possibility of not matching and, accordingly, be ready to participate in the Supplemental Offer and Acceptance Program (SOAP). In this article, we elaborate on the SOAP and the options for applicants who fail to match during Match Week. Alternative courses of action include applying for a preliminary year, matching into a categorical residency program, or aiming to secure EM spots outside the Match through the Council of Emergency Medicine Residency Directors, Society for Academic Emergency Medicine, and American Association of Medical Colleges.

Using EMBL-EBI Services via Web Interface and Programmatically via Web Services

The European Bioinformatics Institute (EMBL-EBI) provides access to a wide range of core databases and analysis tools that are of key importance in bioinformatics. As well as providing web interfaces to these resources, web services are available using REST and SOAP protocols that enable programmatic access and allow their integration into other applications and analytical workflows and pipelines. This article describes the various options available to researchers and bioinformaticians who would like to use our resources via the web interface employing RESTful web service clients provided in Perl, Python, and Java, or would like to use Docker containers to integrate the resources into analysis pipelines and workflows. © 2019 by John Wiley & Sons, Inc.

Finding Them a Spot: A Successful Preliminary Match Curriculum

PURPOSE

Nondesignated preliminary residents (N-DPRs) in General Surgery face difficult challenges of overcoming failure to match and quickly reentering the match again with little time to significantly improve their application. Programs with N-DPRs should take seriously their responsibility for helping these residents obtain a successful career path just as they do for their categorical residents. This study evaluates an intervention to improve the matching of N-DPRs into desired positions.

METHODS

We evaluated the match results of N-DPRs at a single institution over an 8-year period. The first 4 years served as the historical control (Group 1), while the following 4 years of N-DPRs underwent a focused intervention (Group 2). Group 2 underwent an 8-step process: (1) a phone call shortly after supplemental offer and acceptance program (SOAP) to discuss strategy, (2) a 1-hour N-DPR specific orientation, (3) targeted meetings to identify reasons for an unsuccessful match, and personal statement revision in July, (4) mock interviews in August, (5) mid-interview cycle meetings to review strategy and trajectory, (6) meetings in January to prioritize rank lists, (7) meetings the week before the match to discuss plan if match is unsuccessful, and (8) meeting on Monday of Match Week. We determined the N-DPRs initial choice of specialty, specialty obtained after their N-DPR year, and career choice changes that occurred during their preliminary years for both groups. Comparisons and statistical analysis were then completed.

RESULTS

There were 40 N-DPRs in the program over the last 8 years. Group 1, the 4 years before the curriculum, had only 13 of the 16 (81%) N-DPRs obtain a desired position. Group 2, the 4 years following intervention, had all 24 (100%) N-DPRs obtain a desired position. This was a significant improvement (number needed to treat (NNT) = 5.38, p = 0.027). There were no significant differences between groups in regard to the N-DPRs maintaining their original specialty of choice (44% vs 50%).

CONCLUSIONS

The implementation of an N-DPR curriculum significantly improved the probability of N-DPRs to obtain desired positions. Over half of the N-DPRs obtained a position in a specialty different from what they originally applied. Programs should consider aiding N-DPRs in navigating toward a different career path.

Examination of Nondesignated Preliminary Surgery Residents Recruited Since the Inception of Supplementary Offer and Acceptance Program: Lessons Learned From a Large Academic Program

PURPOSE

The majority of surgery programs roster non-designated preliminary (NDP) residents. We and others have reported on the success of NDP resident mentoring with regard to categorical position placement. Lacking is a focus on the candidates themselves and differences based on initial career of choice.

METHODS

NDP residents' files since the institution of SOAP were reviewed for demographics, initial career choice, financial burden, region of medical school of origin, application and interview history, and pre-screen interview candidate score (high score of 22 - data includes USMLE scores, major clerkship grades, and AOA) used for categorical recruitment.

RESULTS

From 2012-16 79 NDP residents have been recruited at UCDenver (82% via SOAP). Median age was 28 years, majority were single (73%), and male (82%). Thirty percent belonged to an under-represented minority group. Mean debt was $156,000 but 20% owe over $250,000. 90% attended US medical schools with 65% from the NRMP "South" region. 86% were recruited as fourth year students. NDPs were categorized as failing to match in general surgery (38%), surgical subspecialties (47%), or other (15%). NDPs applied to median of 68 programs (range 7-200) and granted a median of 8 interviews (range 0-24). NDPs had a mean pre-screening interview score of 13 out of 22 and only 9% would have met the standard threshold to obtain a categorical surgery interview. There were no differences in pre-screening scores in the three groups. 95% NDPs (excluding present year) successfully completed their R1 year (three resigned - one obtained a general surgery spot mid-year and two after matching in non-general surgery fields). 68% NDPs placed in categorical positions after the R1 year. The placement was better for the surgical subspecialty group compared to the other two.

CONCLUSIONS

The bias is that due to a more competitive applicant pool recruiting NDPs from surgical subspecialties would be optimal. However, those unmatched surgical subspecialty candidates are no better academically than the unmatched general surgery group, often have career interests that do not always align with scheduled rotations, and may not feel compelled to complete the year if they match. They place minimally better when accounting for those unmatched general surgery NDP R1s continuing as NDP R2s. Ultimately the success in a non-designated preliminary R1 surgery program is alignment of clinical educational opportunities with the needs of the trainee.

Match 2017: Blindsided or Fumbled?

The field of otolaryngology has historically enjoyed extreme interest among residency applicants. However, in the past few years, the number of applicants has precipitously dropped, so that there is no longer a significant excess of applications. It remains important for academic programs to promote student interest in otolaryngology, to break down barriers that may dissuade excellent candidates, and to widen the welcome.

CIMIDx: Prototype for a Cloud-Based System to Support Intelligent Medical Image Diagnosis With Efficiency

Background

The Internet has greatly enhanced health care, helping patients stay up-to-date on medical issues and general knowledge. Many cancer patients use the Internet for cancer diagnosis and related information. Recently, cloud computing has emerged as a new way of delivering health services but currently, there is no generic and fully automated cloud-based self-management intervention for breast cancer patients, as practical guidelines are lacking.

Objective

We investigated the prevalence and predictors of cloud use for medical diagnosis among women with breast cancer to gain insight into meaningful usage parameters to evaluate the use of generic, fully automated cloud-based self-intervention, by assessing how breast cancer survivors use a generic self-management model. The goal of this study was implemented and evaluated with a new prototype called “CIMIDx”, based on representative association rules that support the diagnosis of medical images (mammograms).

Methods

The proposed Cloud-Based System Support Intelligent Medical Image Diagnosis (CIMIDx) prototype includes two modules. The first is the design and development of the CIMIDx training and test cloud services. Deployed in the cloud, the prototype can be used for diagnosis and screening mammography by assessing the cancers detected, tumor sizes, histology, and stage of classification accuracy. To analyze the prototype’s classification accuracy, we conducted an experiment with data provided by clients. Second, by monitoring cloud server requests, the CIMIDx usage statistics were recorded for the cloud-based self-intervention groups. We conducted an evaluation of the CIMIDx cloud service usage, in which browsing functionalities were evaluated from the end-user’s perspective.

Results

We performed several experiments to validate the CIMIDx prototype for breast health issues. The first set of experiments evaluated the diagnostic performance of the CIMIDx framework. We collected medical information from 150 breast cancer survivors from hospitals and health centers. The CIMIDx prototype achieved high sensitivity of up to 99.29%, and accuracy of up to 98%. The second set of experiments evaluated CIMIDx use for breast health issues, using t tests and Pearson chi-square tests to assess differences, and binary logistic regression to estimate the odds ratio (OR) for the predictors’ use of CIMIDx. For the prototype usage statistics for the same 150 breast cancer survivors, we interviewed 114 (76.0%), through self-report questionnaires from CIMIDx blogs. The frequency of log-ins/person ranged from 0 to 30, total duration/person from 0 to 1500 minutes (25 hours). The 114 participants continued logging in to all phases, resulting in an intervention adherence rate of 44.3% (95% CI 33.2-55.9). The overall performance of the prototype for the good category, reported usefulness of the prototype (P=.77), overall satisfaction of the prototype (P=.31), ease of navigation (P=.89), user friendliness evaluation (P=.31), and overall satisfaction (P=.31). Positive evaluations given by 100 participants via a Web-based questionnaire supported our hypothesis.

Conclusions

The present study shows that women felt favorably about the use of a generic fully automated cloud-based self- management prototype. The study also demonstrated that the CIMIDx prototype resulted in the detection of more cancers in screening and diagnosing patients, with an increased accuracy rate.

Using EMBL-EBI Services via Web Interface and Programmatically via Web Services

The European Bioinformatics Institute (EMBL-EBI) provides access to a wide range of databases and analysis tools that are of key importance in bioinformatics. As well as providing Web interfaces to these resources, Web Services are available using SOAP and REST protocols that enable programmatic access to our resources and allow their integration into other applications and analytical workflows. This unit describes the various options available to a typical researcher or bioinformatician who wishes to use our resources via Web interface or programmatically via a range of programming languages.

Using SOAPaligner for Short Reads Alignment

This unit shows how to use the SOAPaligner package to align short reads to reference. The use of the two most common versions of SOAPaligner, SOAP2 and SOAP3-dp, will be described in detail. The unit also includes a protocol for SNP calling from SOAP2 alignment with SOAPsnp.

Association of Medical Directors of Information Systems consensus on inpatient electronic health record documentation

In 2013, electronic documentation of clinical care stands at a crossroads. The benefits of creating digital notes are at risk of being overwhelmed by the inclusion of easily importable detail. Providers are the primary authors of encounters with patients. We must document clearly our understanding of patients and our communication with them and our colleagues. We want to document efficiently to meet without exceeding documentation guidelines. We copy and paste documentation, because it not only simplifies the documentation process generally, but also supports meeting coding and regulatory requirements specifically. Since the primary goal of our profession is to spend as much time as possible listening to, understanding and helping patients, clinicians need information technology to make electronic documentation easier, not harder. At the same time, there should be reasonable restrictions on the use of copy and paste to limit the growing challenge of 'note bloat'. We must find the right balance between ease of use and thoughtless documentation. The guiding principles in this document may be used to launch an interdisciplinary dialogue that promotes useful and necessary documentation that best facilitates efficient information capture and effective display.

Using electronic medical records to teach patient-centered care

OBJECTIVE

To develop, implement, and evaluate the use of electronic medical records (EMRs) in disease state management activities to teach pharmacy students patient-centered care skills.

DESIGN

Faculty created 12 patient cases and an EMR for each simulated patient. Students used the information in the EMR to confirm the presence of a disease in their patient; to design and monitor an evidence-based patient-centered care plan, and document their assessment, medication, and monitoring recommendations; and to follow up in a SOAP (subjective, objective, assessment, plan) note format within the EMR.

ASSESSMENT

A precourse and postcourse survey instrument was administered to assess students' perceived gains in knowledge and attitudes regarding the use of EMRs for learning patient-centered care skills. A grading rubric was used to evaluate each SOAP note.

CONCLUSION

Implementation of disease state management activities involving EMRs improved pharmacy students' patient care skills.