Developing and evaluating a pediatric asthma severity computable phenotype derived from electronic health records

Concordance of provider chart notation and guideline-based classification of asthma severity

OBJECTIVE

To evaluate concordance of asthma severity classification via physician chart notation compared with guideline-based criteria in adolescents with diagnosed asthma.

METHODS

Of 284 urban primary care and subspecialty clinic patients aged 13-18 years approached through convenience sampling, 203 surveys were completed (RR = 71.5%). We assessed concordance with sensitivity, specificity, and positive predictive values; overall agreement was evaluated with weighted kappa coefficients and McNemar's test.

RESULTS

When considering prescribed treatment according to NAEPP guidelines as a gold standard, the sensitivity for chart notation was very good for intermittent (95%) and less for non-intermittent severity ratings (51%, 58%, and 67% for moderate, severe, and mild persistent asthma, respectively). Overall agreement between chart notation and guideline-based asthma criteria ranged from fair-to-good for mild- (k = 0.36), moderate- (k = 0.44), and severe-persistent severity (k = 0.66). Although the agreement for intermittent severity was highest (k = 0.88), it did not significantly differ by between the two classifications (p ≥ 0.05).

CONCLUSIONS

Concordance for all non-intermittent asthma severity classifications varied between physician and medication-driven 2007 NAEPP guideline classifications in an ethnically diverse urban adolescent patient sample. Physicians should remain aware of the potential for this discordance and refer to the guidelines to classify and treat adolescents with asthma.

Combining Federated Machine Learning and Qualitative Methods to Investigate Novel Pediatric Asthma Subtypes: Protocol for a Mixed Methods Study

BACKGROUND

Pediatric asthma is a heterogeneous disease; however, current characterizations of its subtypes are limited. Machine learning (ML) methods are well-suited for identifying subtypes. In particular, deep neural networks can learn patient representations by leveraging longitudinal information captured in electronic health records (EHRs) while considering future outcomes. However, the traditional approach for subtype analysis requires large amounts of EHR data, which may contain protected health information causing potential concerns regarding patient privacy. Federated learning is the key technology to address privacy concerns while preserving the accuracy and performance of ML algorithms. Federated learning could enable multisite development and implementation of ML algorithms to facilitate the translation of artificial intelligence into clinical practice.

OBJECTIVE

The aim of this study is to develop a research protocol for implementation of federated ML across a large clinical research network to identify and discover pediatric asthma subtypes and their progression over time.

METHODS

This mixed methods study uses data and clinicians from the OneFlorida+ clinical research network, which is a large regional network covering linked and longitudinal patient-level real-world data (RWD) of over 20 million patients from Florida, Georgia, and Alabama in the United States. To characterize the subtypes, we will use OneFlorida+ data from 2011 to 2023 and develop a research-grade pediatric asthma computable phenotype and clinical natural language processing pipeline to identify pediatric patients with asthma aged 2-18 years. We will then apply federated learning to characterize pediatric asthma subtypes and their temporal progression. Using the Promoting Action on Research Implementation in Health Services framework, we will conduct focus groups with practicing pediatric asthma clinicians within the OneFlorida+ network to investigate the clinical utility of the subtypes. With a user-centered design, we will create prototypes to visualize the subtypes in the EHR to best assist with the clinical management of children with asthma.

RESULTS

OneFlorida+ data from 2011 to 2023 have been collected for 411,628 patients aged 2-18 years along with 11,156,148 clinical notes. We expect to complete the computable phenotyping within the first year of the project, followed by subtyping during the second and third years, and then will perform the focus groups and establish the user-centered design in the fourth and fifth years of the project.

CONCLUSIONS

Pediatric asthma subtypes incorporating RWD from diverse populations could improve patient outcomes by moving the field closer to precision pediatric asthma care. Our privacy-preserving federated learning methodology and qualitative implementation work will address several challenges of applying ML to large, multicenter RWD data.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/57981.

Asthma and Chronic Obstructive Pulmonary Disease

In the United States, asthma and chronic obstructive pulmonary disease (COPD) disproportionately affect African Americans, Puerto Ricans, and other minority groups. Compared with non-Hispanic whites, minorities have been marginalized and more frequently exposed to environmental risk factors such as tobacco smoke and outdoor and indoor pollutants. Such divergent environmental exposures, alone or interacting with heredity, lead to disparities in the prevalence, morbidity, and mortality of asthma and COPD, which are worsened by lack of access to health care. In this article, we review the burden and risk factors for racial or ethnic disparities in asthma and COPD and discuss future directions in this field.

Obesity-related pediatric asthma: relationships between pulmonary function and clinical outcomes

OBJECTIVE

We hypothesized that children with obesity-related asthma would have worse self-reported asthma control, report an increased number of asthma symptoms and have lower FEV1/FVC associated with worse clinical asthma outcomes compared to children with asthma only.

METHODS

Cross sectional analyses examined two hundred and eighteen (obesity-related asthma = 109, asthma only = 109) children, ages 7-15 that were recruited from clinics and hospitals within the Bronx, NY. Pulmonary function was assessed by forced expiratory volume in the first second (percent predicted FEV1) and the ratio of FEV1 to the forced vital capacity of the lungs (FEV1/FVC). Structural equation modeling examined if pulmonary function was associated with asthma control and clinical outcomes between groups.

RESULTS

Lower percent predicted FEV1 was associated with increased hospitalizations (p = 0.03) and oral steroid bursts in the past 12 months (p = 0.03) in the obesity-related asthma group but not in the asthma only group. FEV1/FVC was also associated with increased hospitalizations (p = 0.02) and oral steroid bursts (p = 0.008) in the obesity-related asthma group but not the asthma only group. Lower FEV1/FVC was associated with the number of asthma symptoms endorsed in the asthma only group but not in the obesity-related asthma group. Percent predicted FEV1 and FEV1/FVC was not associated with asthma control in either group.

CONCLUSIONS

Pulmonary function was associated with oral steroid bursts and hospitalizations but not self-reported asthma control, suggesting the importance of incorporating measures of pulmonary function into the treatment of pediatric obesity-related asthma.

Linking Ambient NO2 Pollution Measures with Electronic Health Record Data to Study Asthma Exacerbations

Electronic health record (EHR)-derived data can be linked to geospatially distributed socioeconomic and environmental factors to conduct large-scale epidemiologic studies. Ambient NO2 is a known environmental risk factor for asthma. However, health exposure studies often rely on data from geographically sparse regulatory monitors that may not reflect true individual exposure. We contrasted use of interpolated NO2 regulatory monitor data with raw satellite measurements and satellite-derived ground estimates, building on previous work which has computed improved exposure estimates from remotely sensed data. Raw satellite and satellite-derived ground measurements captured spatial variation missed by interpolated ground monitor measurements. Multivariable analyses comparing these three NO2 measurement approaches (interpolated monitor, raw satellite, and satellite-derived) revealed a positive relationship between exposure and asthma exacerbations for both satellite measurements. Exposure-outcome relationships using the interpolated monitor NO2 were inconsistent with known relationships to asthma, suggesting that interpolated monitor data might yield misleading results in small region studies.

Development and assessment of a low literacy, pictographic asthma action plan with clinical automation to enhance guideline-concordant care for children with asthma

OBJECTIVE

Asthma is characterized by reversible pulmonary symptoms, frequent hospitalizations, poor quality of life, and varied treatment. Parents with low health literacy (HL) is linked to poor asthma outcomes in children. Recent practice updates recommended inhaled corticosteroids for the management of persistent asthma, but guideline-concordant care is suboptimal. Our aim was to develop and assess an Asthma Action Plan (AAP) that could serve as an individualized plan for low HL families and facilitate guideline-concordant care for clinicians.

METHODS

We followed the National Institute of Health 5-step "Clear & Simple" approach to develop the Uniformed Services AAP. Our AAP included symptom pictographs (dyspnea, cough, sleep, activity) and guideline-concordant clinical automation tools. Caregivers assessed the pictograms for validity (transparency of ≥ 85%; translucency score ≥ 5; and ≥ 85% recall). Readability was assessed using 7 formulas. (<6th Grade was acceptable). Comprehensibility, design quality, and usefulness was assessed by caregivers using the Consumer Information Rating Form (CIRF) (>80% was acceptable). Understandability and actionability was assessed by medical librarians using the Patient Education Materials Assessment Tool-Printable (>80% was acceptable). Suitability was assessed by clinicians using the modified Suitability Assessment of Materials (SAM) instrument (>70% was superior).

RESULTS

All 12 pictograms were validated (N = 118 respondents). Readability demonstrated a 4th grade level. Overall CIRF percentile score = 80.4%. Understandability and Actionability = 100%. Suitability score = 75%.

CONCLUSIONS

Our AAP was formally endorsed by the Allergy & Asthma Network. The Uniformed Services AAP is a novel tool with embedded clinical automation that can address low HL and enhance guideline-concordant care.

Trends and opportunities in computable clinical phenotyping: A scoping review

Identifying patient cohorts meeting the criteria of specific phenotypes is essential in biomedicine and particularly timely in precision medicine. Many research groups deliver pipelines that automatically retrieve and analyze data elements from one or more sources to automate this task and deliver high-performing computable phenotypes. We applied a systematic approach based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines to conduct a thorough scoping review on computable clinical phenotyping. Five databases were searched using a query that combined the concepts of automation, clinical context, and phenotyping. Subsequently, four reviewers screened 7960 records (after removing over 4000 duplicates) and selected 139 that satisfied the inclusion criteria. This dataset was analyzed to extract information on target use cases, data-related topics, phenotyping methodologies, evaluation strategies, and portability of developed solutions. Most studies supported patient cohort selection without discussing the application to specific use cases, such as precision medicine. Electronic Health Records were the primary source in 87.1 % (N = 121) of all studies, and International Classification of Diseases codes were heavily used in 55.4 % (N = 77) of all studies, however, only 25.9 % (N = 36) of the records described compliance with a common data model. In terms of the presented methods, traditional Machine Learning (ML) was the dominant method, often combined with natural language processing and other approaches, while external validation and portability of computable phenotypes were pursued in many cases. These findings revealed that defining target use cases precisely, moving away from sole ML strategies, and evaluating the proposed solutions in the real setting are essential opportunities for future work. There is also momentum and an emerging need for computable phenotyping to support clinical and epidemiological research and precision medicine.