Publications by authors named "Brian Hazlehurst"

Objective: To present a general framework providing high-level guidance to developers of computable algorithms for identifying patients with specific clinical conditions (phenotypes) through a variety of approaches, including but not limited to machine learning and natural language processing methods to incorporate rich electronic health record data.

Materials And Methods: Drawing on extensive prior phenotyping experiences and insights derived from 3 algorithm development projects conducted specifically for this purpose, our team with expertise in clinical medicine, statistics, informatics, pharmacoepidemiology, and healthcare data science methods conceptualized stages of development and corresponding sets of principles, strategies, and practical guidelines for improving the algorithm development process.

Results: We propose 5 stages of algorithm development and corresponding principles, strategies, and guidelines: (1) assessing fitness-for-purpose, (2) creating gold standard data, (3) feature engineering, (4) model development, and (5) model evaluation.

View Article and Find Full Text PDF

We sought to determine whether machine learning and natural language processing (NLP) applied to electronic medical records could improve performance of automated health-care claims-based algorithms to identify anaphylaxis events using data on 516 patients with outpatient, emergency department, or inpatient anaphylaxis diagnosis codes during 2015-2019 in 2 integrated health-care institutions in the Northwest United States. We used one site's manually reviewed gold-standard outcomes data for model development and the other's for external validation based on cross-validated area under the receiver operating characteristic curve (AUC), positive predictive value (PPV), and sensitivity. In the development site 154 (64%) of 239 potential events met adjudication criteria for anaphylaxis compared with 180 (65%) of 277 in the validation site.

View Article and Find Full Text PDF

Background: Acute pancreatitis is a serious gastrointestinal disease that is an important target for drug safety surveillance. Little is known about the accuracy of ICD-10 codes for acute pancreatitis in the United States, or their performance in specific clinical settings. We conducted a validation study to assess the accuracy of acute pancreatitis ICD-10 diagnosis codes in inpatient, emergency department (ED), and outpatient settings.

View Article and Find Full Text PDF

Opioid surveillance in response to the opioid epidemic will benefit from scalable, automated algorithms for identifying patients with clinically documented signs of problem prescription opioid use. Existing algorithms lack accuracy. We sought to develop a high-sensitivity, high-specificity classification algorithm based on widely available structured health data to identify patients receiving chronic extended-release/long-acting (ER/LA) therapy with evidence of problem use to support subsequent epidemiologic investigations.

View Article and Find Full Text PDF

Purpose: To enhance automated methods for accurately identifying opioid-related overdoses and classifying types of overdose using electronic health record (EHR) databases.

Methods: We developed a natural language processing (NLP) software application to code clinical text documentation of overdose, including identification of intention for self-harm, substances involved, substance abuse, and error in medication usage. Using datasets balanced with cases of suspected overdose and records of individuals at elevated risk for overdose, we developed and validated the application using Kaiser Permanente Northwest data, then tested portability of the application using Kaiser Permanente Washington data.

View Article and Find Full Text PDF

Purpose: To facilitate surveillance and evaluate interventions addressing opioid-related overdoses, algorithms are needed for use in large health care databases to identify and differentiate community-occurring opioid-related overdoses from inpatient-occurring opioid-related overdose/oversedation.

Methods: Data were from Kaiser Permanente Northwest (KPNW), a large integrated health plan. We iteratively developed and evaluated an algorithm for electronically identifying inpatient overdose/oversedation in KPNW hospitals from 1 January 2008 to 31 December 2014.

View Article and Find Full Text PDF

Purpose: The study aims to develop and validate algorithms to identify and classify opioid overdoses using claims and other coded data, and clinical text extracted from electronic health records using natural language processing (NLP).

Methods: Primary data were derived from Kaiser Permanente Northwest (2008-2014), an integrated health care system (~n > 475 000 unique individuals per year). Data included International Classification of Diseases, Ninth Revision (ICD-9) codes for nonfatal diagnoses, International Classification of Diseases, Tenth Revision (ICD-10) codes for fatal events, clinical notes, and prescription medication records.

View Article and Find Full Text PDF
Article Synopsis
  • The study aimed to assess the link between consistent smoking cessation support from healthcare professionals and achieving long-term quitting success among smokers.
  • Utilizing a large cohort of over 33,000 patients from six different health systems in the U.S., the research analyzed data from electronic health records over several years.
  • Results showed that patients receiving smoking cessation support in at least 75% of their primary care visits were nearly three times more likely to quit smoking long-term compared to those receiving minimal assistance, highlighting the benefits of regular support in helping smokers quit.
View Article and Find Full Text PDF

Introduction: Brief smoking-cessation interventions in primary care settings are effective, but delivery of these services remains low. The Centers for Medicare and Medicaid Services' Meaningful Use (MU) of Electronic Health Record (EHR) Incentive Program could increase rates of smoking assessment and cessation assistance among vulnerable populations. This study examined whether smoking status assessment, cessation assistance, and odds of being a current smoker changed after Stage 1 MU implementation.

View Article and Find Full Text PDF
Article Synopsis
  • The CER Hub is a web-based platform designed for comparative effectiveness research that enables the integration of electronic health data from various organizations with different EHR systems.
  • It processes both free-text and coded clinical data, while offering standardized access and a library of tools for researchers to develop specific applications.
  • The platform is currently utilized in studies assessing asthma medication effectiveness and smoking cessation services in diverse healthcare settings, demonstrating its capability to handle complex, multi-institutional clinical data.
View Article and Find Full Text PDF
Article Synopsis
  • This study tracked changes in tobacco use over 4 years among patients in six diverse health care organizations, using electronic medical records.
  • Out of 34,393 smokers identified in 2007, 38.6% quit smoking at least once, with 15.4% remaining smoke-free for over a year by the end of the fourth year.
  • Factors that increased long-term quitting included being older, or having certain health diagnoses, while female gender and being black or non-Hispanic were linked to lower quitting rates.
View Article and Find Full Text PDF
Article Synopsis
  • Comparative effectiveness research (CER) requires high-quality data from diverse electronic health record (EHR) systems, and uniformity in this data is critical for accurate study outcomes.
  • The CER Hub developed a quality assurance (QA) process using the 'emrAdapter' tool, which conducts quality checks on primary care encounter records and reports data issues for local fixes.
  • After implementing the QA process across six health systems, data quality significantly improved over three iterations, addressing issues such as incomplete mapping of local EHR data to a standardized framework.
View Article and Find Full Text PDF
Article Synopsis
  • The study evaluates how well physicians provide smoking cessation services, specifically using the 5 As approach, to current smokers across different health systems, gender, and age groups.
  • About half of the smokers received advice to quit, while fewer were assessed for readiness or provided with follow-up support, indicating gaps in smoking cessation efforts.
  • Results showed significant differences in how the 5 As were documented depending on the health system, while no notable differences were found based on gender, suggesting a need for improved protocols for better patient support.*
View Article and Find Full Text PDF

Background: Numerous population-based surveys indicate that overweight and obese patients can benefit from lifestyle counseling during routine clinical care.

Purpose: To determine if natural language processing (NLP) could be applied to information in the electronic health record (EHR) to automatically assess delivery of weight management-related counseling in clinical healthcare encounters.

Methods: The MediClass system with NLP capabilities was used to identify weight-management counseling in EHRs.

View Article and Find Full Text PDF

Comparative effectiveness research (CER) has the potential to transform the current health care delivery system by identifying the most effective medical and surgical treatments, diagnostic tests, disease prevention methods, and ways to deliver care for specific clinical conditions. To be successful, such research requires the identification, capture, aggregation, integration, and analysis of disparate data sources held by different institutions with diverse representations of the relevant clinical events. In an effort to address these diverse demands, there have been multiple new designs and implementations of informatics platforms that provide access to electronic clinical data and the governance infrastructure required for interinstitutional CER.

View Article and Find Full Text PDF

In an experiment to investigate cognitive skill differences between clinicians and lay persons, eight individuals in each group were asked to determine if an explicit concept existed in an ambulatory encounter note (a simple task) or if the concept could be inferred from the same note (a complex task). Subjects answered questions, highlighted text used to answer each question, and commented on their reasoning for selecting specific text. Quantitative results were mixed for expert vs.

View Article and Find Full Text PDF

The Vaccine Safety Datalink (VSD) is a collaboration between the CDC and eight large HMOs to investigate adverse events following immunization through analyses of clinical data. We modified an existing system, called MediClass, that uses natural language processing to identify clinical events recorded in electronic medical records (EMRs). We customized MediClass so it could detect possible vaccine adverse events (VAEs) generally, and gastrointestinal-related VAEs in particular, in the text clinical notes of encounters recorded in the EMR of a large HMO.

View Article and Find Full Text PDF

Electronic medical records (EMRs) hold the promise of making routine comprehensive measurement of care quality a reality. However, there are many informatics challenges that stand in the way of this goal. Guidelines are rarely stated in precise enough language for automated measurement of clinical practices and the data necessary for that measurement often reside in the text notes of EMRs.

View Article and Find Full Text PDF

Objective: We estimated the quality of life impact of vision loss in a community-based population with diabetes.

Design And Methods: We randomly surveyed 4,000 members of a large health maintenance organization with type 2 diabetes to assess quality of life using the EQ-5D instrument. Visual acuity was obtained by automated text processing of clinical notes recorded during the two years preceding subjects' surveys.

View Article and Find Full Text PDF

Background: Medical informatics has been guided by an individual-centered model of human cognition, inherited from classical theory of mind, in which knowledge, problem-solving, and information-processing responsible for intelligent behavior all derive from the inner workings of an individual agent.

Objectives And Results: In this paper we argue that medical informatics commitment to the classical model of cognition conflates the processing performed by the minds of individual agents with the processing performed by the larger distributed activity systems within which individuals operate. We review trends in cognitive science that seek to close the gap between general-purpose models of cognition and applied considerations of real-world human performance.

View Article and Find Full Text PDF