Prevalence of Financial Considerations Documented in Primary Care Encounters as Identified by Natural Language Processing Methods

Factors associated with cost conversations in oral health care settings

BACKGROUND

Patient-provider cost conversations can minimize cost-related barriers to health, while improving treatment adherence and patient satisfaction. The authors sought to identify factors associated with the occurrence of cost conversations in dentistry.

METHODS

This was a cross-sectional study using data from an online, self-administered survey of US adults who had seen a dentist within the past 24 months at the time of the survey. Multivariable hierarchical logistic regression analysis was used to identify patient and provider characteristics associated with the occurrence of cost conversations.

RESULTS

Of the 370 respondents, approximately two-thirds (68%) reported having a cost conversation with their dental provider during their last dental visit. Cost conversations were more likely for patients aged 25 through 34 years (odds ratio [OR], 2.84; 95% CI, 1.54 to 5.24), 35 through 44 years (OR, 3.35; 95% CI, 1.50 to 7.51), and 55 through 64 years (OR, 3.39; 95% CI, 1.38 to 8.28) than patients aged 18 through 24 years. Cost conversations were less likely to occur during visits with dental hygienists than during visits with general or family dentists (OR, 0.25; 95% CI, 0.11 to 0.58). In addition, respondents from the South (OR, 1.90; 95% CI, 1.04 to 3.48) and those screened for financial hardship were more likely to report having cost conversations with their dental providers (OR, 6.70; 95% CI, 2.69 to 16.71).

CONCLUSIONS

Within the study sample, cost conversations were common and were facilitated via financial hardship screening.

PRACTICAL IMPLICATIONS

Modifying oral health care delivery processes to incorporate financial hardship screening may be an effective way to facilitate cost conversations and provision of patient-centered care.

A scoping review of ethics considerations in clinical natural language processing

Objectives

To review through an ethics lens the state of research in clinical natural language processing (NLP) for the study of bias and fairness, and to identify gaps in research.

Methods

We queried PubMed and Google Scholar for articles published between 2015 and 2021 concerning clinical NLP, bias, and fairness. We analyzed articles using a framework that combines the machine learning (ML) development process (ie, design, data, algorithm, and critique) and bioethical concepts of beneficence, nonmaleficence, autonomy, justice, as well as explicability. Our approach further differentiated between biases of clinical text (eg, systemic or personal biases in clinical documentation towards patients) and biases in NLP applications.

Results

Out of 1162 articles screened, 22 met criteria for full text review. We categorized articles based on the design (N = 2), data (N = 12), algorithm (N = 14), and critique (N = 17) phases of the ML development process.

Discussion

Clinical NLP can be used to study bias in applications reliant on clinical text data as well as explore biases in the healthcare setting. We identify 3 areas of active research that require unique ethical considerations about the potential for clinical NLP to address and/or perpetuate bias: (1) selecting metrics that interrogate bias in models; (2) opportunities and risks of identifying sensitive patient attributes; and (3) best practices in reconciling individual autonomy, leveraging patient data, and inferring and manipulating sensitive information of subgroups. Finally, we address the limitations of current ethical frameworks to fully address concerns of justice. Clinical NLP is a rapidly advancing field, and assessing current approaches against ethical considerations can help the discipline use clinical NLP to explore both healthcare biases and equitable NLP applications.

Extracting social determinants of health from electronic health records using natural language processing: a systematic review

OBJECTIVE

Social determinants of health (SDoH) are nonclinical dispositions that impact patient health risks and clinical outcomes. Leveraging SDoH in clinical decision-making can potentially improve diagnosis, treatment planning, and patient outcomes. Despite increased interest in capturing SDoH in electronic health records (EHRs), such information is typically locked in unstructured clinical notes. Natural language processing (NLP) is the key technology to extract SDoH information from clinical text and expand its utility in patient care and research. This article presents a systematic review of the state-of-the-art NLP approaches and tools that focus on identifying and extracting SDoH data from unstructured clinical text in EHRs.

MATERIALS AND METHODS

A broad literature search was conducted in February 2021 using 3 scholarly databases (ACL Anthology, PubMed, and Scopus) following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A total of 6402 publications were initially identified, and after applying the study inclusion criteria, 82 publications were selected for the final review.

RESULTS

Smoking status (n = 27), substance use (n = 21), homelessness (n = 20), and alcohol use (n = 15) are the most frequently studied SDoH categories. Homelessness (n = 7) and other less-studied SDoH (eg, education, financial problems, social isolation and support, family problems) are mostly identified using rule-based approaches. In contrast, machine learning approaches are popular for identifying smoking status (n = 13), substance use (n = 9), and alcohol use (n = 9).

CONCLUSION

NLP offers significant potential to extract SDoH data from narrative clinical notes, which in turn can aid in the development of screening tools, risk prediction models, and clinical decision support systems.

Sensitivity and Specificity of Real-World Social Factor Screening Approaches

Health care organizations are increasingly documenting patients for social risk factors in structured data. Two main approaches to documentation, ICD-10 Z codes and screening questions, face limited adoption and conceptual challenges. This study compared estimates of social risk factors obtained via screening questions and ICD-10 Z diagnoses coding, as used in clinical practice, to estiamtes from validated survey instruments in a sample of adult primary care and emergency department patients at an urban safety-net health system. Financial strain, transportation barriers, food insecurity, and housing instability were independently assessed using instruments with published reliability and validity. These four social factors were also being collected by the health system in screening questions or could be mapped to ICD-10 Z code diagnosis code concepts. Neither the screening questions nor ICD-10 Z codes performed particularly well in terms of accuracy. For the screening questions, the Area Under the Curve (AUC) scores were 0.609 for financial strain, 0.703 for transportation, 0.698 for food insecurity, and 0.714 for housing instability. For the ICD-10 Z codes, AUC scores tended to be lower in the range of 0.523 to 0.535. For both screening questions and ICD-10 Z codes, the measures were much more specific than sensitive. Under real world conditions, ICD-10 Z codes and screening questions are at the minimal, or below, threshold for being diagnostically useful approaches to identifying patients’ social risk factors. Data collection support through information technology or novel approaches combining data sources may be necessary to improve the usefulness of these data.

Cost Conversations About Anticoagulation Between Patients With Atrial Fibrillation and Their Clinicians: A Secondary Analysis of a Randomized Clinical Trial

IMPORTANCE

How patients with atrial fibrillation (AF) and their clinicians consider cost in forming care plans remains unknown.

OBJECTIVE

To identify factors that inform conversations regarding costs of anticoagulants for treatment of AF between patients and clinicians and outcomes associated with these conversations.

DESIGN, SETTING, AND PARTICIPANTS

This cohort study of recorded encounters and participant surveys at 5 US medical centers (including academic, community, and safety-net centers) from the SDM4AFib randomized trial compared standard AF care with and without use of a shared decision-making (SDM) tool. Included patients were considering anticoagulation treatment and were recruited by their clinicians between January 30, 2017, and June 27, 2019. Data were analyzed between August and November 2019.

MAIN OUTCOMES AND MEASURES

The incidence of and factors associated with cost conversations, and the association of cost conversations with patients' consideration of treatment cost burden and their choice of anticoagulation.

RESULTS

A total of 830 encounters (out of 922 enrolled participants) were recorded. Patients' mean (SD) age was 71.0 (10.4) years; 511 patients (61.6%) were men, 704 (86.0%) were White, 303 (40.9%) earned between $40 000 and $99 999 in annual income, and 657 (79.2%) were receiving anticoagulants. Clinicians' mean (SD) age was 44.8 (13.2) years; 75 clinicians (53.2%) were men, and 111 (76%) practiced as physicians, with approximately half (69 [48.9%]) specializing in either internal medicine or cardiology. Cost conversations occurred in 639 encounters (77.0%) and were more likely in the SDM arm (378 [90%] vs 261 [64%]; OR, 9.69; 95% CI, 5.77-16.29). In multivariable analysis, cost conversations were more likely to occur with female clinicians (66 [47%]; OR, 2.85; 95% CI, 1.21-6.71); consultants vs in-training clinicians (113 [75%]; OR, 4.0; 95% CI, 1.4-11.1); clinicians practicing family medicine (24 [16%]; OR, 12.12; 95% CI, 2.75-53.38]), internal medicine (35 [23%]; OR, 3.82; 95% CI, 1.25-11.70), or other clinicians (21 [14%]; OR, 4.90; 95% CI, 1.32-18.16) when compared with cardiologists; and for patients with an annual household income between $40 000 and $99 999 (249 [82.2%]; OR, 1.86; 95% CI, 1.05-3.29) compared with income below $40 000 or above $99 999. More patients who had cost conversations reported cost as a factor in their decision (244 [89.1%] vs 327 [69.0%]; OR 3.66; 95% CI, 2.43-5.50), but cost conversations were not associated with the choice of anticoagulation agent.

CONCLUSIONS AND RELEVANCE

Cost conversations were common, particularly for middle-income patients and with female and consultant-level primary care clinicians, as well as in encounters using an SDM tool; they were associated with patients' consideration of treatment cost burden but not final treatment choice. With increasing costs of care passed on to patients, these findings can inform efforts to promote cost conversations in practice.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT02905032.

Data Science Methods for Nursing-Relevant Patient Outcomes and Clinical Processes: The 2019 Literature Year in Review

Data science continues to be recognized and used within healthcare due to the increased availability of large data sets and advanced analytics. It can be challenging for nurse leaders to remain apprised of this rapidly changing landscape. In this article, we describe our findings from a scoping literature review of papers published in 2019 that use data science to explore, explain, and/or predict 15 phenomena of interest to nurses. Fourteen of the 15 phenomena were associated with at least one paper published in 2019. We identified the use of many contemporary data science methods (eg, natural language processing, neural networks) for many of the outcomes. We found many studies exploring Readmissions and Pressure Injuries. The topics of Artificial Intelligence/Machine Learning Acceptance, Burnout, Patient Safety, and Unit Culture were poorly represented. We hope that the studies described in this article help readers: (1) understand the breadth and depth of data science's ability to improve clinical processes and patient outcomes that are relevant to nurses and (2) identify gaps in the literature that are in need of exploration.