Advanced Search of the Electronic Medical Record: Augmenting Safety and Efficiency in Radiology

Leveraging medical context to recommend semantically similar terms for chart reviews

BACKGROUND

Information retrieval (IR) help clinicians answer questions posed to large collections of electronic medical records (EMRs), such as how best to identify a patient's cancer stage. One of the more promising approaches to IR for EMRs is to expand a keyword query with similar terms (e.g., augmenting cancer with mets). However, there is a large range of clinical chart review tasks, such that fixed sets of similar terms is insufficient. Current language models, such as Bidirectional Encoder Representations from Transformers (BERT) embeddings, do not capture the full non-textual context of a task. In this study, we present new methods that provide similar terms dynamically by adjusting with the context of the chart review task.

METHODS

We introduce a vector space for medical-context in which each word is represented by a vector that captures the word's usage in different medical contexts (e.g., how frequently cancer is used when ordering a prescription versus describing family history) beyond the context learned from the surrounding text. These vectors are transformed into a vector space for customizing the set of similar terms selected for different chart review tasks. We evaluate the vector space model with multiple chart review tasks, in which supervised machine learning models learn to predict the preferred terms of clinically knowledgeable reviewers. To quantify the usefulness of the predicted similar terms to a baseline of standard word2vec embeddings, we measure (1) the prediction performance of the medical-context vector space model using the area under the receiver operating characteristic curve (AUROC) and (2) the labeling effort required to train the models.

RESULTS

The vector space outperformed the baseline word2vec embeddings in all three chart review tasks with an average AUROC of 0.80 versus 0.66, respectively. Additionally, the medical-context vector space significantly reduced the number of labels required to learn and predict the preferred similar terms of reviewers. Specifically, the labeling effort was reduced to 10% of the entire dataset in all three tasks.

CONCLUSIONS

The set of preferred similar terms that are relevant to a chart review task can be learned by leveraging the medical context of the task.

Use, Impact, Weaknesses, and Advanced Features of Search Functions for Clinical Use in Electronic Health Records: A Scoping Review

Objective  Although vast amounts of patient information are captured in electronic health records (EHRs), effective clinical use of this information is challenging due to inadequate and inefficient access to it at the point of care. The purpose of this study was to conduct a scoping review of the literature on the use of EHR search functions within a single patient's record in clinical settings to characterize the current state of research on the topic and identify areas for future study.

Methods  We conducted a literature search of four databases to identify articles on within-EHR search functions or the use of EHR search function in the context of clinical tasks. After reviewing titles and abstracts and performing a full-text review of selected articles, we included 17 articles in the analysis. We qualitatively identified themes in those articles and synthesized the literature for each theme.

Results  Based on the 17 articles analyzed, we delineated four themes: (1) how clinicians use search functions, (2) impact of search functions on clinical workflow, (3) weaknesses of current search functions, and (4) advanced search features. Our review found that search functions generally facilitate patient information retrieval by clinicians and are positively received by users. However, existing search functions have weaknesses, such as yielding false negatives and false positives, which can decrease trust in the results, and requiring a high cognitive load to perform an inclusive search of a patient's record.

Conclusion  Despite the widespread adoption of EHRs, only a limited number of articles describe the use of EHR search functions in a clinical setting, despite evidence that they benefit clinician workflow and productivity. Some of the weaknesses of current search functions may be addressed by enhancing EHR search functions with collaborative filtering.

Improving Chinese electronic medical record retrieval by field weight assignment, negation detection, and re-ranking

The technique of information retrieval has been widely used in electronic medical record (EMR) systems. It's a pity that most existing methods have not considered the structures and language features of Chinese EMRs, which affects the performance of retrieval. To improve accuracy and comprehensiveness, we propose an improved algorithm of Chinese EMR retrieval. First, the weights of fields in Chinese EMRs are assigned based on the corresponding importance in clinical applications. Second, negative relations in EMRs are detected, and the retrieval scores of negative terms are adjusted accordingly. Third, the retrieval results are re-ranked by expansion terms and time information to enhance the recall without decreasing precision. Experiment results show that the improved algorithm increases the precision and recall significantly, which shows that the algorithm takes a full account of the characteristics of Chinese EMRs and fits the needs for clinical applications.

The effect of clinical information on radiology reporting: A systematic review

INTRODUCTION

The aim of this study was to investigate the effects of clinical information on the accuracy, timeliness, reporting confidence and clinical relevance of the radiology report.

METHODS

A systematic review of studies that investigated a link between primary communication of clinical information to the radiologist and the resultant report was conducted. Relevant studies were identified by a comprehensive search of electronic databases (PubMed, Scopus and EMBASE). Studies were screened using pre-defined criteria. Methodological quality was assessed using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Quasi-Experimental Studies. Synthesis of findings was narrative. Results were reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

RESULTS

There were 21 studies which met the inclusion criteria, of which 20 were included in our review following quality assessment. Sixteen studies investigated the effect of clinical information on reporting accuracy, three studies investigated the effect of clinical information on reporting confidence, three studies explored the impact of clinical information on clinical relevance, and two studies investigated the impact of clinical information on reporting timeliness. Some studies explored multiple outcomes. Studies concluded that clinical information improved interpretation accuracy, clinical relevance and reporting confidence; however, reporting time was not substantially affected by the addition of clinical information.

CONCLUSION

The findings of this review suggest clinical information has a positive impact on the radiology report. It is in the best interests of radiologists to communicate the importance of clinical information to reporting via the creation of criteria standards to guide the requesting practices of medical imaging referrers. Further work is recommended to establish these criteria standards.

Assessment of Electronic Health Record Search Patterns and Practices by Practitioners in a Large Integrated Health Care System

IMPORTANCE

The electronic health record (EHR) is a source of practitioner dissatisfaction in part because of challenges with information retrieval. To improve data accessibility, a better understanding of practitioners' information needs within individual patient records is needed.

OBJECTIVE

To assess EHR users' searches using data from a large integrated health care system.

DESIGN, SETTING, AND PARTICIPANTS

This retrospective cross-sectional analysis used EHR search data from Kaiser Permanente Northern California, an integrated health care delivery system with more than 4.4 million members. Users' EHR search activity data were obtained from April 1, 2018, to May 15, 2019.

MAIN OUTCOMES AND MEASURES

Search term frequency was grouped by user and practitioner types. Network analyses were performed of co-occurring search terms within a single search episode, and centrality measures for search terms (degree and betweenness centrality) were calculated.

RESULTS

A total of 12 313 047 search activities (including 4 328 330 searches and 7 984 717 result views) conducted by 34 735 unique users within 977 160 unique patient EHRs were identified. In aggregate, users searched for 208 374 unique search terms and conducted a median of 4 searches (interquartile range, 1-28 searches). Of all 97 367 active EHR users, 34 735 (35.7%) conducted at least 1 search. However, of all 12 968 active EHR physician users, 9801 (75.6%) conducted at least 1 search, and of all 1908 active pharmacist users, 1402 (73.5%) conducted at least 1 search. The top 3 most commonly searched terms were statin (75 017 searches [1.7%]), colonoscopy (73 545 [1.7%]), and pft (54 990 [1.3%]). However, wide variation in top searches were noted across practitioner groups. Terms searched most often with another term in a single linked search episode included statin, lisinopril, colonoscopy, gabapentin, and aspirin.

CONCLUSIONS AND RELEVANCE

Although physicians and pharmacists were the most active users of EHR searches, search volume and frequently searched terms varied considerably by and within user role. Further customization of the EHR interface may help leverage users' search content and patterns to improve targeted information retrieval.

Comparing Artificial Intelligence Approaches to Retrieve Clinical Reports Documenting Implantable Devices Posing MRI Safety Risks

OBJECTIVE

Assess sensitivity, specificity, and accuracy of two approaches to identify patients with implantable devices that pose safety risks for MRI-an expert-derived approach and an ontology-derived natural language processing (NLP). Determine the proportion of clinical data that identify these implantable devices.

METHODS

This Institutional Review Board-approved retrospective study was performed at a 793-bed academic hospital. The expert-derived approach used an open-source software with a list of curated terms to query for implantable devices posing high safety risk ("MRI-Red") in patients undergoing MRI. The ontology-derived approach used an NLP system with terms mapped to Systematized Nomenclature of Medicine-Clinical Terms. Queries were performed in three clinical data types-25,000 radiology reports, 174,769 emergency department (ED) notes, and 41,085 other clinical reports (eg, cardiology, operating room, physician notes, radiology reports, pathology reports, patient letters). Sensitivity, specificity, and accuracy of both methods against manual review of a randomly sampled 465 reports were assessed and tested for significant differences between expert-derived and ontology-derived approaches using t test.

RESULTS

Accuracy, sensitivity, and specificity of expert-versus ontology-derived approaches were similar (0.83 versus 0.91, P = .080; 0.88 versus 0.96, P = .178; 0.82 versus 0.92, P = .110). The proportion of radiology reports, ED notes, and other clinical reports retrieved containing implantable devices with high safety risks for MRI ranged from 1.47% to 1.88%.

DISCUSSION

Artificial intelligence approaches such as expert-driven NLP and ontology-driven NLP have similar accuracy in identifying patients with implantable devices that pose high safety risks for MRI.

Magnetic resonance imaging of the lumbar spine: determining clinical impact and potential harm from overuse

BACKGROUND

Lumbar spine magnetic resonance imaging is frequently said to be "overused" in the evaluation of low back pain, yet data concerning the extent of overuse and the potential harmful effects are lacking.

PURPOSE

The objective of this study was to determine the proportion of examinations with a detectable impact on patient care (actionable outcomes).

STUDY DESIGN

This is a retrospective cohort study.

PATIENT SAMPLE

A total of 5,365 outpatient lumbar spine magnetic resonance (MR) examinations were conducted.

OUTCOME MEASURES

Actionable outcomes included (1) findings leading to an intervention making use of anatomical information such as surgery; (2) new diagnoses of cancer, infection, or fracture; or (3) following known lumbar spine pathology. Potential harm was assessed by identifying examinations where suspicion of cancer or infection was raised but no positive diagnosis made.

METHODS

A medical record aggregation/search system was used to identify lumbar spine MR examinations with positive outcome measures. Patient notes were examined to verify outcomes. A random sample was manually inspected to identify missed positive outcomes.

RESULTS

The proportion of actionable lumbar spine magnetic resonance imaging was 13%, although 93% were appropriate according to the American College of Radiology guidelines. Of 36 suspected cases of cancer or infection, 81% were false positives. Further investigations were ordered on 59% of suspicious examinations, 86% of which were false positives.

CONCLUSIONS

The proportion of lumbar spine MR examinations that inform management is small. The false-positive rate and the proportion of false positives involving further investigation are high. Further study to improve the efficiency of imaging is warranted.

Extracting similar terms from multiple EMR-based semantic embeddings to support chart reviews

OBJECTIVE

Word embeddings project semantically similar terms into nearby points in a vector space. When trained on clinical text, these embeddings can be leveraged to improve keyword search and text highlighting. In this paper, we present methods to refine the selection process of similar terms from multiple EMR-based word embeddings, and evaluate their performance quantitatively and qualitatively across multiple chart review tasks.

MATERIALS AND METHODS

Word embeddings were trained on each clinical note type in an EMR. These embeddings were then combined, weighted, and truncated to select a refined set of similar terms to be used in keyword search and text highlighting. To evaluate their quality, we measured the similar terms' information retrieval (IR) performance using precision-at-K (P@5, P@10). Additionally a user study evaluated users' search term preferences, while a timing study measured the time to answer a question from a clinical chart.

RESULTS

The refined terms outperformed the baseline method's information retrieval performance (e.g., increasing the average P@5 from 0.48 to 0.60). Additionally, the refined terms were preferred by most users, and reduced the average time to answer a question.

CONCLUSIONS

Clinical information can be more quickly retrieved and synthesized when using semantically similar term from multiple embeddings.

Canadian Association of Radiologists White Paper on Artificial Intelligence in Radiology

Artificial intelligence (AI) is rapidly moving from an experimental phase to an implementation phase in many fields, including medicine. The combination of improved availability of large datasets, increasing computing power, and advances in learning algorithms has created major performance breakthroughs in the development of AI applications. In the last 5 years, AI techniques known as deep learning have delivered rapidly improving performance in image recognition, caption generation, and speech recognition. Radiology, in particular, is a prime candidate for early adoption of these techniques. It is anticipated that the implementation of AI in radiology over the next decade will significantly improve the quality, value, and depth of radiology's contribution to patient care and population health, and will revolutionize radiologists' workflows. The Canadian Association of Radiologists (CAR) is the national voice of radiology committed to promoting the highest standards in patient-centered imaging, lifelong learning, and research. The CAR has created an AI working group with the mandate to discuss and deliberate on practice, policy, and patient care issues related to the introduction and implementation of AI in imaging. This white paper provides recommendations for the CAR derived from deliberations between members of the AI working group. This white paper on AI in radiology will inform CAR members and policymakers on key terminology, educational needs of members, research and development, partnerships, potential clinical applications, implementation, structure and governance, role of radiologists, and potential impact of AI on radiology in Canada.

Common Variable Immunodeficiency Non-Infectious Disease Endotypes Redefined Using Unbiased Network Clustering in Large Electronic Datasets

Common variable immunodeficiency (CVID) is increasingly recognized for its association with autoimmune and inflammatory complications. Despite recent advances in immunophenotypic and genetic discovery, clinical care of CVID remains limited by our inability to accurately model risk for non-infectious disease development. Herein, we demonstrate the utility of unbiased network clustering as a novel method to analyze inter-relationships between non-infectious disease outcomes in CVID using databases at the United States Immunodeficiency Network (USIDNET), the centralized immunodeficiency registry of the United States, and Partners, a tertiary care network in Boston, MA, USA, with a shared electronic medical record amenable to natural language processing. Immunophenotypes were comparable in terms of native antibody deficiencies, low titer response to pneumococcus, and B cell maturation arrest. However, recorded non-infectious disease outcomes were more substantial in the Partners cohort across the spectrum of lymphoproliferation, cytopenias, autoimmunity, atopy, and malignancy. Using unbiased network clustering to analyze 34 non-infectious disease outcomes in the Partners cohort, we further identified unique patterns of lymphoproliferative (two clusters), autoimmune (two clusters), and atopic (one cluster) disease that were defined as CVID non-infectious endotypes according to discrete and non-overlapping immunophenotypes. Markers were both previously described {high serum IgE in the atopic cluster [odds ratio (OR) 6.5] and low class-switched memory B cells in the total lymphoproliferative cluster (OR 9.2)} and novel [low serum C3 in the total lymphoproliferative cluster (OR 5.1)]. Mortality risk in the Partners cohort was significantly associated with individual non-infectious disease outcomes as well as lymphoproliferative cluster 2, specifically (OR 5.9). In contrast, unbiased network clustering failed to associate known comorbidities in the adult USIDNET cohort. Together, these data suggest that unbiased network clustering can be used in CVID to redefine non-infectious disease inter-relationships; however, applicability may be limited to datasets well annotated through mechanisms such as natural language processing. The lymphoproliferative, autoimmune, and atopic Partners CVID endotypes herein described can be used moving forward to streamline genetic and biomarker discovery and to facilitate early screening and intervention in CVID patients at highest risk for autoimmune and inflammatory progression.

Development and empirical user-centered evaluation of semantically-based query recommendation for an electronic health record search engine

OBJECTIVE

The utility of biomedical information retrieval environments can be severely limited when users lack expertise in constructing effective search queries. To address this issue, we developed a computer-based query recommendation algorithm that suggests semantically interchangeable terms based on an initial user-entered query. In this study, we assessed the value of this approach, which has broad applicability in biomedical information retrieval, by demonstrating its application as part of a search engine that facilitates retrieval of information from electronic health records (EHRs).

MATERIALS AND METHODS

The query recommendation algorithm utilizes MetaMap to identify medical concepts from search queries and indexed EHR documents. Synonym variants from UMLS are used to expand the concepts along with a synonym set curated from historical EHR search logs. The empirical study involved 33 clinicians and staff who evaluated the system through a set of simulated EHR search tasks. User acceptance was assessed using the widely used technology acceptance model.

RESULTS

The search engine's performance was rated consistently higher with the query recommendation feature turned on vs. off. The relevance of computer-recommended search terms was also rated high, and in most cases the participants had not thought of these terms on their own. The questions on perceived usefulness and perceived ease of use received overwhelmingly positive responses. A vast majority of the participants wanted the query recommendation feature to be available to assist in their day-to-day EHR search tasks.

DISCUSSION AND CONCLUSION

Challenges persist for users to construct effective search queries when retrieving information from biomedical documents including those from EHRs. This study demonstrates that semantically-based query recommendation is a viable solution to addressing this challenge.

Automated identification of an aspirin-exacerbated respiratory disease cohort

BACKGROUND

Aspirin-exacerbated respiratory disease (AERD) is characterized by 3 clinical features: asthma, nasal polyposis, and respiratory reactions to cyclooxygenase-1 inhibitors (nonsteroidal anti-inflammatory drugs). Electronic health records (EHRs) contain information on each feature of this triad.

OBJECTIVE

We sought to determine whether an informatics algorithm applied to the EHR could electronically identify patients with AERD.

METHODS

We developed an informatics algorithm to search the EHRs of patients aged 18 years and older from the Partners Healthcare system over a 10-year period (2004-2014). Charts with search terms for asthma, nasal polyps, and record of respiratory (cohort A) or unspecified (cohort B) reactions to nonsteroidal anti-inflammatory drugs were identified as "possible AERD." Two clinical experts reviewed all charts to confirm a diagnosis of "clinical AERD" and classify cases as "diagnosed AERD" or "undiagnosed AERD" on the basis of physician-documented AERD-specific terms in patient notes.

RESULTS

Our algorithm identified 731 "possible AERD" cases, of which 638 were not in our AERD patient registry. Chart review of cohorts A (n = 511) and B (n = 127) demonstrated a positive predictive value of 78.4% for "clinical AERD," which rose to 88.7% when unspecified reactions were excluded. Of those with clinical AERD, 12.4% had no mention of AERD by any treating caregiver and were classified as "undiagnosed AERD." "Undiagnosed AERD" cases were less likely than "diagnosed AERD" cases to have been seen by an allergist/immunologist (38.7% vs 93.2%; P < .0001).

CONCLUSIONS

An informatics algorithm can successfully identify both known and previously undiagnosed cases of AERD with a high positive predictive value. Involvement of an allergist/immunologist significantly increases the likelihood of an AERD diagnosis.

State of Integration Between PACS and Other IT Systems: A National Survey of Academic Radiology Departments

PURPOSE

The aim of this study was to investigate the state of integration between PACS and other IT systems relevant to radiologists' routine work across US academic radiology departments (ARDs). The results were intended to assess readiness for the ongoing transition to value-based health care by providing insights into currently challenging areas of integration but also areas associated with high levels of anticipated workflow efficiency improvements.

METHODS

A cross-sectional survey was conducted using an online survey approved by the Society of Chairs of Academic Radiology Departments and sent to its members. Collected responses were analyzed with descriptive statistics and Fisher's exact tests.

RESULTS

The response rate was 26% (34 of 132 members), and the respondents covered a large spectrum of ARDs considering location, size aspects, year of PACS introduction, and filmless production. Most notable findings included widespread high-level integration of PACS with dictation systems (>90%), low penetration of integration between PACS and critical notification systems (15%), and an overall better integration of PACS and radiology information systems (82%) than of PACS and electronic medical records (47%).

CONCLUSIONS

Integration supporting radiologists' personal productivity is well spread among US ARDs, but as we transition into a value-based health care delivery model, there is a need to focus further integration efforts on systems with the greatest potential to document value in a patient-centric setting. Examples of such focus areas include integration of PACS and electronic medical records, adoption of vendor-neutral archives, and the use of workflow management systems.

Radiologist Point-of-Care Clinical Decision Support and Adherence to Guidelines for Incidental Lung Nodules

PURPOSE

To evaluate the effect of a workstation-integrated, point-of-care, clinical decision support (CDS) tool on radiologist adherence to radiology department guidelines for follow-up of incidental pulmonary nodules detected on abdominal CT.

METHODS

The CDS tool was developed to facilitate adherence to department guidelines for managing pulmonary nodules seen on abdominal CT. In October 2012, the tool was deployed within the radiology department of an academic medical center and could be used for a given abdominal CT at the discretion of the interpreting radiologist. We retrospectively identified consecutive patients who underwent abdominal CT (in the period from January 2012 to April 2013), had no comparison CT scans available, and were reported to have a solid, noncalcified, pulmonary nodule. Concordance between radiologist follow-up recommendation and department guidelines was compared among three groups: patients scanned before implementation of the CDS tool; and patients scanned after implementation, with versus without use of the tool.

RESULTS

A total of 409 patients were identified, including 268 for the control group. Overall, guideline concordance was higher after CDS tool implementation (92 of 141 [65%] versus 133 of 268 [50%], P = .003). This finding was driven by the subset of post-CDS implementation cases in which the CDS tool was used (57 of 141 [40%]). In these cases, guideline concordance was significantly higher (54 of 57 [95%]), compared with post-implementation cases in which CDS was not used (38 of 84 [45%], P < .001), and to a control group of patients from before implementation (133 of 268 [50%]; P < .001).

CONCLUSIONS

A point-of-care CDS tool was associated with improved adherence to guidelines for follow-up of incidental pulmonary nodules.

Predicting Non-Adherence with Outpatient Colonoscopy Using a Novel Electronic Tool that Measures Prior Non-Adherence

BACKGROUND

Accurately predicting the risk of no-show for a scheduled colonoscopy can help target interventions to improve compliance with colonoscopy, and thereby reduce the disease burden of colorectal cancer and enhance the utilization of resources within endoscopy units.

OBJECTIVES

We aimed to utilize information available in an electronic medical record (EMR) and endoscopy scheduling system to create a predictive model for no-show risk, and to simultaneously evaluate the role for natural language processing (NLP) in developing such a model.

DESIGN

This was a retrospective observational study using discovery and validation phases to design a colonoscopy non-adherence prediction model. An NLP-derived variable called the Non-Adherence Ratio ("NAR") was developed, validated, and included in the model.

PARTICIPANTS

Patients scheduled for outpatient colonoscopy at an Academic Medical Center (AMC) that is part of a multi-hospital health system, 2009 to 2011, were included in the study.

MAIN MEASURES

Odds ratios for non-adherence were calculated for all variables in the discovery cohort, and an Area Under the Receiver Operating Curve (AUC) was calculated for the final non-adherence prediction model.

KEY RESULTS

The non-adherence model included six variables: 1) gender; 2) history of psychiatric illness, 3) NAR; 4) wait time in months; 5) number of prior missed endoscopies; and 6) education level. The model achieved discrimination in the validation cohort (AUC= =70.2 %). At a threshold non-adherence score of 0.46, the model's sensitivity and specificity were 33 % and 92 %, respectively. Removing the NAR from the model significantly reduced its predictive power (AUC = 64.3 %, difference = 5.9 %, p < 0.001).

CONCLUSIONS

A six-variable model using readily available clinical and demographic information demonstrated accuracy for predicting colonoscopy non-adherence. The NAR, a novel variable developed using NLP technology, significantly strengthened this model's predictive power.

Supporting information retrieval from electronic health records: A report of University of Michigan's nine-year experience in developing and using the Electronic Medical Record Search Engine (EMERSE)

OBJECTIVE

This paper describes the University of Michigan's nine-year experience in developing and using a full-text search engine designed to facilitate information retrieval (IR) from narrative documents stored in electronic health records (EHRs). The system, called the Electronic Medical Record Search Engine (EMERSE), functions similar to Google but is equipped with special functionalities for handling challenges unique to retrieving information from medical text.

MATERIALS AND METHODS

Key features that distinguish EMERSE from general-purpose search engines are discussed, with an emphasis on functions crucial to (1) improving medical IR performance and (2) assuring search quality and results consistency regardless of users' medical background, stage of training, or level of technical expertise.

RESULTS

Since its initial deployment, EMERSE has been enthusiastically embraced by clinicians, administrators, and clinical and translational researchers. To date, the system has been used in supporting more than 750 research projects yielding 80 peer-reviewed publications. In several evaluation studies, EMERSE demonstrated very high levels of sensitivity and specificity in addition to greatly improved chart review efficiency.

DISCUSSION

Increased availability of electronic data in healthcare does not automatically warrant increased availability of information. The success of EMERSE at our institution illustrates that free-text EHR search engines can be a valuable tool to help practitioners and researchers retrieve information from EHRs more effectively and efficiently, enabling critical tasks such as patient case synthesis and research data abstraction.

CONCLUSION

EMERSE, available free of charge for academic use, represents a state-of-the-art medical IR tool with proven effectiveness and user acceptance.

Development and Validation of Electronic Health Record-based Triggers to Detect Delays in Follow-up of Abnormal Lung Imaging Findings

Purpose To develop an electronic health record (EHR)-based trigger algorithm to identify delays in follow-up of patients with imaging results that are suggestive of lung cancer and to validate this trigger on retrospective data. Materials and Methods The local institutional review board approved the study. A "trigger" algorithm was developed to automate the detection of delays in diagnostic evaluation of chest computed tomographic (CT) images and conventional radiographs that were electronically flagged by reviewing radiologists as being "suspicious for malignancy." The trigger algorithm was developed through literature review and expert input. It included patients who were alive and 40-70 years old, and it excluded instances in which appropriate timely follow-up (defined as occurring within 30 days) was detected (eg, pulmonary visit) or when follow-up was unnecessary (eg, in patients with a terminal illness). The algorithm was iteratively applied to a retrospective test cohort in an EHR data warehouse at a large Veterans Affairs facility, and manual record reviews were used to validate each individual criterion. The final algorithm aimed at detecting an absence of timely follow-up was retrospectively applied to an independent validation cohort to determine the positive predictive value (PPV). Trigger performance, time to follow-up, reasons for lack of follow-up, and cancer outcomes were analyzed and reported by using descriptive statistics. Results The trigger algorithm was retrospectively applied to the records of 89 168 patients seen between January 1, 2009, and December 31, 2009. Of 538 records with an imaging report that was flagged as suspicious for malignancy, 131 were identified by the trigger as being high risk for delayed diagnostic evaluation. Manual chart reviews confirmed a true absence of follow-up in 75 cases (trigger PPV of 57.3% for detecting evaluation delays), of which four received a diagnosis of primary lung cancer within the subsequent 2 years. Conclusion EHR-based triggers can be used to identify patients with suspicious imaging findings in whom follow-up diagnostic evaluation was delayed. (©) RSNA, 2015.

Emergency medicine resident physicians' perceptions of electronic documentation and workflow: a mixed methods study

OBJECTIVE

To understand emergency department (ED) physicians' use of electronic documentation in order to identify usability and workflow considerations for the design of future ED information system (EDIS) physician documentation modules.

METHODS

We invited emergency medicine resident physicians to participate in a mixed methods study using task analysis and qualitative interviews. Participants completed a simulated, standardized patient encounter in a medical simulation center while documenting in the test environment of a currently used EDIS. We recorded the time on task, type and sequence of tasks performed by the participants (including tasks performed in parallel). We then conducted semi-structured interviews with each participant. We analyzed these qualitative data using the constant comparative method to generate themes.

RESULTS

Eight resident physicians participated. The simulation session averaged 17 minutes and participants spent 11 minutes on average on tasks that included electronic documentation. Participants performed tasks in parallel, such as history taking and electronic documentation. Five of the 8 participants performed a similar workflow sequence during the first part of the session while the remaining three used different workflows. Three themes characterize electronic documentation: (1) physicians report that location and timing of documentation varies based on patient acuity and workload, (2) physicians report a need for features that support improved efficiency; and (3) physicians like viewing available patient data but struggle with integration of the EDIS with other information sources.

CONCLUSION

We confirmed that physicians spend much of their time on documentation (65%) during an ED patient visit. Further, we found that resident physicians did not all use the same workflow and approach even when presented with an identical standardized patient scenario. Future EHR design should consider these varied workflows while trying to optimize efficiency, such as improving integration of clinical data. These findings should be tested quantitatively in a larger, representative study.

A software system to collect expert relevance ratings of medical record items for specific clinical tasks

Development of task-specific electronic medical record (EMR) searches and user interfaces has the potential to improve the efficiency and safety of health care while curbing rising costs. The development of such tools must be data-driven and guided by a strong understanding of practitioner information requirements with respect to specific clinical tasks or scenarios. To acquire this important data, this paper describes a model by which expert practitioners are leveraged to identify which components of the medical record are most relevant to a specific clinical task. We also describe the computer system that was created to efficiently implement this model of data gathering. The system extracts medical record data from the EMR of patients matching a given clinical scenario, de-identifies the data, breaks the data up into separate medical record items (eg, radiology reports, operative notes, laboratory results, etc), presents each individual medical record item to experts under the hypothetical of the given clinical scenario, and records the experts’ ratings regarding the relevance of each medical record item to that specific clinical scenario or task. After an iterative process of data collection, these expert relevance ratings can then be pooled and used to design point-of-care EMR searches and user interfaces tailored to the task-specific needs of practitioners.

Usefulness and consequences of cardiac resynchronization therapy in dialysis-dependent patients with heart failure

Cardiac resynchronization therapy (CRT) is often deferred in dialysis-dependent patients with heart failure (HF) because of a perceived lack of benefit and potentially higher risks, although the outcomes associated with CRT in dialysis have not been reported. We therefore studied our center's experience with CRT in dialysis-dependent patients. We constructed a descriptive assessment of these patients (n = 15) and performed a case-control analysis matching for age, gender, bundle branch morphology, diabetes mellitus, cardiomyopathy origin, and β-blocker and angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker use. Baseline and 6-month echocardiograms were assessed for evidence of reverse remodeling. No periprocedural or long-term complications were observed among dialysis patients. Heterogenous improvement in ejection fraction (+3.1 ± 9.2%) was noted and 2 patients derived absolute improvements of 8% and 22%, respectively. Dialysis patients demonstrated the following 3-year event rates: HF hospitalization, 31%; all-cause hospitalization, 100%; mortality, 73%; and HF hospitalization or death, 82%. In the case-control analysis, controls demonstrated superior reverse remodeling (+9.2 ± 9.5% increase in ejection fraction), decreased mortality (73% vs 44%, p = 0.038), and all-cause hospitalizations (76% vs 100%, p = 0.047), with no difference in HF hospitalizations (p = 0.39), compared with dialysis patients. In conclusion, at our center, the dialysis-dependent patients with HF who underwent CRT implantation did so safely and no serious complications were observed. Certain dialysis patients demonstrated compelling improvement after device implantation. Compared with matched controls, dialysis patients were at increased risk for adverse events and worsened echocardiographic outcomes.

The future of the radiology information system

OBJECTIVE

Today in the hospital setting, several functions of the radiology information system (RIS), including order entry, patient registration, report repository, and the physician directory, have moved to enterprise electronic medical records. Some observers might conclude that the RIS is going away. In this article, we contend that because of the maturity of the RIS market compared with other areas of the health care enterprise, radiology has a unique opportunity to innovate.

CONCLUSION

While most of the hospital enterprise spends the next several years going through the digital transformation converting from paper to a digital format, radiology can leap ahead in its use of analytics and information technology. This article presents a summary of new RIS functions still maturing and open to innovation in the RIS market.

Automated before-procedure electronic health record screening to assess appropriateness for GI endoscopy and sedation

BACKGROUND

Endoscopists are performing greater numbers of procedures, often on patients with complex conditions, in ambulatory settings because of changing patient demographics and referral patterns. To assist with the pre-procedure assessment of such patients, we deployed an advanced electronic health record tool, the Queriable Patient Inference Dossier (QPID), to review clinical histories and generate e-mail alerts to providers, based on clinical guidelines.

OBJECTIVE

Study the feasibility of an automated pre-procedure alert system for outpatient endoscopy.

DESIGN

We retrospectively reviewed 5 physicians' use of the application and their responses to the alerts.

SETTING

A hospital-based endoscopy unit and its two satellite outpatient clinics, Boston area, Massachusetts.

PATIENTS

Adult outpatients referred for endoscopy with moderate sedation.

INTERVENTION

Pre-procedure alerts automatically sent 7 days before the procedure, highlighting any conditions/clinical history that may affect management of the patient.

MAIN OUTCOME MEASUREMENTS

Physician use of the pre-procedure alert system and its effect on patient management.

RESULTS

We studied 1682 procedures that met inclusion criteria for review by QPID and 364 alerts (1.6% of the eligible procedures). Nearly 80% of the alerts were reviewed and responded to by the physicians, and 70 total alerts resulted in a change in patient management (4.2% of eligible procedures).

LIMITATIONS

The small size of the study group and the low rate of adverse events during the study period limit our findings. We thus plan to conduct a larger follow-up study to demonstrate changes in safety and efficiency.

CONCLUSION

Use of advanced electronic health record technologies, such as QPID, may improve provider efficiency and patient outcomes in endoscopy units.

Data mining of mental health issues of non-bone marrow donor siblings

Background

Allogenic hematopoietic stem cell transplantation is a curative treatment for patients with advanced hematologic malignancies. However, the long-term mental health issues of siblings who were not selected as donors (non-donor siblings, NDS) in the transplantation have not been well assessed. Data mining is useful in discovering new findings from a large, multidisciplinary data set and the Scenario Map analysis is a novel approach which allows extracting keywords linking different conditions/events from text data of interviews even when the keywords appeared infrequently. The aim of this study is to assess mental health issues on NDSs and to find helpful keywords for the clinical follow-up using a Scenario Map analysis.

Findings

A 47-year-old woman whose younger sister had undergone allogenic hematopoietic stem cell transplantation 20 years earlier was interviewed as a NDS. The text data from the interview transcriptions was analyzed using Scenario Mapping. Four clusters of words and six keywords were identified. Upon review of the word clusters and keywords, both the subject and researchers noticed that the subject has had mental health issues since the disease onset to date with being a NDS. The issues have been alleviated by her family.

Conclusions

This single subject study suggested the advantages of data mining in clinical follow-up for mental health issues of patients and/or their families.