Restoring the Story and Creating a Valuable Clinical Note

Artificial intelligence: revolutionizing cardiology with large language models

Natural language processing techniques are having an increasing impact on clinical care from patient, clinician, administrator, and research perspective. Among others are automated generation of clinical notes and discharge letters, medical term coding for billing, medical chatbots both for patients and clinicians, data enrichment in the identification of disease symptoms or diagnosis, cohort selection for clinical trial, and auditing purposes. In the review, an overview of the history in natural language processing techniques developed with brief technical background is presented. Subsequently, the review will discuss implementation strategies of natural language processing tools, thereby specifically focusing on large language models, and conclude with future opportunities in the application of such techniques in the field of cardiology.

New Coding Guidelines Reduce Emergency Department Note Bloat But More Work Is Needed

STUDY OBJECTIVE

The length and redundancy of notes authored by clinicians has significantly increased, giving rise to the term "note bloat." We analyzed the impact of new coding guidelines and documentation best practices on the length of emergency department (ED) notes and the amount of time clinicians spent documenting.

METHODS

In a large, multisite health care delivery organization, we retrospectively evaluated the length of all ED provider notes and the amount of time clinicians spent documenting between February 2018 and June 2023. In January 2023, we implemented changes to the standardized note template to align with the new coding guidelines from the American Medical Association and the Centers for Medicare & Medicaid Services. The primary outcomes were the length of provider notes and the amount of time spent documenting.

RESULTS

Our study sample consisted of 1,679,762 ED provider notes. Six months after the intervention, the average note length decreased by 872 words (95% confidence interval 867 to 877 words), whereas the amount of time clinicians spent documenting did not change.

CONCLUSIONS

Embracing new guidelines and practices, we reduced the length of ED provider notes by 872 words. Despite this, the time clinicians spent documenting did not change significantly. We provide an early report of success in reducing note bloat in the ED to help guide future efforts to reduce overall documentation burden.

Diagnostic error experiences of patients and families with limited English-language health literacy or disadvantaged socioeconomic position in a cross-sectional US population-based survey

BACKGROUND

Language barrier, reduced self-advocacy, lower health literacy or biased care may hinder the diagnostic process. Data on how patients/families with limited English-language health literacy (LEHL) or disadvantaged socioeconomic position (dSEP) experience diagnostic errors are sparse.

METHOD

We compared patient-reported diagnostic errors, contributing factors and impacts between respondents with LEHL or dSEP and their counterparts in the 2017 Institute for Healthcare Improvement US population-based survey, using contingency analysis and multivariable logistic regression models for the analyses.

RESULTS

596 respondents reported a diagnostic error; among these, 381 reported LEHL or dSEP. After adjusting for sex, race/ethnicity and physical health, individuals with LEHL/dSEP were more likely than their counterparts to report unique contributing factors: "(No) qualified translator or healthcare provider that spoke (the patient's) language" (OR and 95% CI 4.4 (1.3 to 14.9)); "not understanding the follow-up plan" (1.9 (1.1 to 3.1)); "too many providers… but no clear leader" (1.8 (1.2 to 2.7)); "not able to keep follow-up appointments" (1.9 (1.1 to 3.2)); "not being able to pay for necessary medical care" (2.5 (1.4 to 4.4)) and "out-of-date or incorrect medical records" (2.6 (1.4 to 4.8)). Participants with LEHL/dSEP were more likely to report long-term emotional, financial and relational impacts, compared with their counterparts. Subgroup analysis (LEHL-only and dSEP-only participants) showed similar results.

CONCLUSIONS

Individuals with LEHL or dSEP identified unique and actionable contributing factors to diagnostic errors. Interpreter access should be viewed as a diagnostic safety imperative, social determinants affecting care access/affordability should be routinely addressed as part of the diagnostic process and patients/families should be encouraged to access and update their medical records. The frequent and disproportionate long-term impacts from self-reported diagnostic error among LEHL/dSEP patients/families raises urgency for greater prevention and supportive efforts.

Data standards for transcatheter aortic valve implantation: the European Unified Registries for Heart Care Evaluation and Randomised Trials (EuroHeart)

AIMS

Standardized data definitions are necessary for the quantification of quality of care and patient outcomes in observational studies and randomised controlled trials (RCTs). The European Unified Registries for Heart Care Evaluation and Randomised Trials (EuroHeart) project of the European Society of Cardiology (ESC) aims to create pan-European data standards for cardiovascular diseases and interventions, including transcatheter aortic valve implantation (TAVI).

METHODS AND RESULTS

We followed the EuroHeart methodology for cardiovascular data standard development. A Working Group of 29 members representing 12 countries was established and included a patient representative, as well as experts in the management of valvular heart disease from the European Association of Percutaneous Cardiovascular Interventions (EAPCI), the European Association of Cardiovascular Imaging (EACVI) and the Working Group on Cardiovascular Surgery. We conducted a systematic review of the literature and used a modified Delphi method to reach consensus on a final set of variables. For each variable, the Working Group provided a definition, permissible values, and categorized the variable as mandatory (Level 1) or additional (Level 2) based on its clinical importance and feasibility. In total, 93 Level 1 and 113 Level 2 variables were selected, with the level 1 variables providing the dataset for registration of patients undergoing TAVI on the EuroHeart IT platform.

CONCLUSION

This document provides details of the EuroHeart data standards for TAVI processes of care and in-hospital outcomes. In the context of EuroHeart, this will facilitate quality improvement, observational research, registry-based RCTs and post-marketing surveillance of devices, and pharmacotherapies.

ONE-SENTENCE SUMMARY

The EuroHeart data standards for transcatheter aortic valve implantation (TAVI) are a set of internationally agreed data variables and definitions that once implemented will facilitate improvement of quality of care and outcomes for patients receiving TAVI.

Impact of clinical note format on diagnostic accuracy and efficiency

BACKGROUND

Clinician notes are structured in a variety of ways. This research pilot tested an innovative study design and explored the impact of note formats on diagnostic accuracy and documentation review time.

OBJECTIVE

To compare two formats for clinical documentation (narrative format vs. list of findings) on clinician diagnostic accuracy and documentation review time.

METHOD

Participants diagnosed written clinical cases, half in narrative format, and half in list format. Diagnostic accuracy (defined as including correct case diagnosis among top three diagnoses) and time spent processing the case scenario were measured for each format. Generalised linear mixed regression models and bias-corrected bootstrap percentile confidence intervals for mean paired differences were used to analyse the primary research questions.

RESULTS

Odds of correctly diagnosing list format notes were 26% greater than with narrative notes. However, there is insufficient evidence that this difference is significant (75% CI 0.8-1.99). On average the list format notes required 85.6 more seconds to process and arrive at a diagnosis compared to narrative notes (95% CI -162.3, -2.77). Of cases where participants included the correct diagnosis, on average the list format notes required 94.17 more seconds compared to narrative notes (75% CI -195.9, -8.83).

CONCLUSION

This study offers note format considerations for those interested in improving clinical documentation and suggests directions for future research. Balancing the priority of clinician preference with value of structured data may be necessary.

IMPLICATIONS

This study provides a method and suggestive results for further investigation in usability of electronic documentation formats.

I had not time to make it shorter: an exploratory analysis of how physicians reduce note length and time in notes

OBJECTIVE

We analyze observed reductions in physician note length and documentation time, 2 contributors to electronic health record (EHR) burden and burnout.

MATERIALS AND METHODS

We used EHR metadata from January to May, 2021 for 130 079 ambulatory physician Epic users. We identified cohorts of physicians who decreased note length and/or documentation time and analyzed changes in their note composition.

RESULTS

37 857 physicians decreased either note length (n = 15 647), time in notes (n = 15 417), or both (n = 6793). Note length decreases were primarily attributable to reductions in copy/paste text (average relative change of -18.9%) and templated text (-17.2%). Note time decreases were primarily attributable to reductions in manual text (-27.3%) and increases in note content from other care team members (+21.1%).

DISCUSSION

Organizations must consider priorities and tradeoffs in the distinct approaches needed to address different contributors to EHR burden.

CONCLUSION

Future research should explore scalable burden-reduction initiatives responsive to both note bloat and documentation time.

Writing Is Thinking: Implementation and Evaluation of an Internal Medicine Residency Clinical Reasoning and Documentation Curriculum

With increasingly complicated patients and faster throughput, time for thorough critical thinking and thoughtful clinical documentation is limited, especially in the training environment. Advocating for the value of clinical documentation as a robust opportunity for critical thinking, we describe the implementation and evaluation of a clinical reasoning and documentation curriculum for internal medicine residents. Our curriculum employed facilitated discussion, practical application, and a resident-as-teacher model. Resident surveys showed improved perceptions of the clinical and educational value of clinical documentation. Residents reported increased feedback to interns about their documentation and more appreciation of documentation as a venue for critical thinking.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s40670-022-01570-5.

Linking the Humanities With Clinical Reasoning: Proposing an Integrative Conceptual Model for a Graduate Medical Education Humanities Curriculum

Discussion surrounding the role of the humanities as an important analytic epistemology within medical education is generally less robust than literature supporting its value in building empathy and promoting personal reflection and wellness. As such, the humanities have not been considered to be as relevant when teaching medical reasoning or technical skills. Yet, might the humanities offer value in emboldening the analytic thinking of medical learners? This article proposes an integrative conceptual model that links the thought process defining medicine-clinical reasoning-with humanistic analysis in an effort to advance the argument that the humanities offer a complementary and innovative platform that can be used within traditional medical education. The article then discusses preliminary findings from a pilot curriculum based on this model, implemented during internal medicine morning report at the University of North Carolina at Chapel Hill School of Medicine. Preliminary qualitative analysis of transcripts from the pilot curriculum demonstrates that a thought process analogous to that of clinical reasoning can be identified within guided group analyses of humanities works. Participants simultaneously used thought processes that were analytic and intuitive. The emergence of ambiguity/intuition as a theme in the pilot curriculum suggests the humanities could be a powerful tool for exploring and embracing ambiguity in clinical practice. Through the development of an integrative conceptual model, this article helps to demonstrate more explicitly the theoretical link between the reasoning pathways of the humanities and clinical medicine. Though a refined curriculum and more rigorous analysis are needed before arguing for the incorporation of the humanities into traditional graduate medical education on a larger scale, the preliminary findings here support the feasibility and promise of a curriculum based on the proposed integrative conceptual model.

"Note Bloat" impacts deep learning-based NLP models for clinical prediction tasks

One unintended consequence of the Electronic Health Records (EHR) implementation is the overuse of content-importing technology, such as copy-and-paste, that creates "bloated" notes containing large amounts of textual redundancy. Despite the rising interest in applying machine learning models to learn from real-patient data, it is unclear how the phenomenon of note bloat might affect the Natural Language Processing (NLP) models derived from these notes. Therefore, in this work we examine the impact of redundancy on deep learning-based NLP models, considering four clinical prediction tasks using a publicly available EHR database. We applied two deduplication methods to the hospital notes, identifying large quantities of redundancy, and found that removing the redundancy usually has little negative impact on downstream performances, and can in certain circumstances assist models to achieve significantly better results. We also showed it is possible to attack model predictions by simply adding note duplicates, causing changes of correct predictions made by trained models into wrong predictions. In conclusion, we demonstrated that EHR text redundancy substantively affects NLP models for clinical prediction tasks, showing that the awareness of clinical contexts and robust modeling methods are important to create effective and reliable NLP systems in healthcare contexts.

Writing Practices Associated With Electronic Progress Notes and the Preferences of Those Who Read Them: Descriptive Study

Background

Hospital progress notes can serve as an important communication tool. However, they are criticized for their length, preserved content, and for the time physicians spend writing them.

Objective

We aimed to describe hospital progress note content, writing and reading practices, and the preferences of those who create and read them prior to the implementation of a new electronic health record system.

Methods

Using a sample of hospital progress notes from 1000 randomly selected admissions, we measured note length, similarity of content in successive daily notes for the same patient, the time notes were signed and read, and who read them. We conducted focus group sessions with note writers, readers, and clinical leaders to understand their preferences.

Results

We analyzed 4938 inpatient progress notes from 418 authors. The average length was 886 words, and most were in the Assessment & Plan note section. A total of 29% of notes (n=1432) were signed after 4 PM. Notes signed later in the day were read less often. Notes were highly similar from one day to the next, and 26% (23/88) had clinical risk associated with the preserved content. Note content of the highest value varied according to the reader’s professional role.

Conclusions

Progress note length varied widely. Notes were often signed late in the day when they were read less often and were highly similar to the note from the previous day. Measuring note length, signing time, when and by whom notes are read, and the amount and safety of preserved content will be useful metrics for measuring how the new electronic health record system is used, and can aid improvements.

Length and Redundancy of Outpatient Progress Notes Across a Decade at an Academic Medical Center

IMPORTANCE

There is widespread concern that clinical notes have grown longer and less informative over the past decade. Addressing these concerns requires a better understanding of the magnitude, scope, and potential causes of increased note length and redundancy.

OBJECTIVE

To measure changes between 2009 and 2018 in the length and redundancy of outpatient progress notes across multiple medical specialties and investigate how these measures associate with author experience and method of note entry.

DESIGN, SETTING, AND PARTICIPANTS

This cross-sectional study was conducted at Oregon Health & Science University, a large academic medical center. Participants included clinicians and staff who wrote outpatient progress notes between 2009 and 2018 for a random sample of 200 000 patients. Statistical analysis was performed from March to August 2020.

EXPOSURES

Use of a comprehensive electronic health record to document patient care.

MAIN OUTCOMES AND MEASURES

Note length, note redundancy (ie, the proportion of text identical to the patient's last note), and percentage of templated, copied, or directly typed note text.

RESULTS

A total of 2 704 800 notes written by 6228 primary authors across 46 specialties were included in this study. Median note length increased 60.1% (99% CI, 46.7%-75.2%) from a median of 401 words (interquartile range [IQR], 225-660 words) in 2009 to 642 words (IQR, 399-1007 words) in 2018. Median note redundancy increased 10.9 percentage points (99% CI, 7.5-14.3 percentage points) from 47.9% in 2009 to 58.8% in 2018. Notes written in 2018 had a mean value of just 29.4% (99% CI, 28.2%-30.7%) directly typed text with the remaining 70.6% of text being templated or copied. Mixed-effect linear models found that notes with higher proportions of templated or copied text were significantly longer and more redundant (eg, in the 2-year model, each 1% increase in the proportion of copied or templated note text was associated with 1.5% [95% CI, 1.5%-1.5%] and 1.6% [95% CI, 1.6%-1.6%] increases in note length, respectively). Residents and fellows also wrote significantly (26.3% [95% CI, 25.8%-26.7%]) longer notes than more senior authors, as did more recent hires (1.8% for each year later [95% CI, 1.3%-2.4%]).

CONCLUSIONS AND RELEVANCE

In this study, outpatient progress notes grew longer and more redundant over time, potentially limiting their use in patient care. Interventions aimed at reducing outpatient progress note length and redundancy may need to simultaneously address multiple factors such as note template design and training for both new and established clinicians.