1
|
Huang WA, Engelhard M, Coffman M, Hill ED, Weng Q, Scheer A, Maslow G, Henao R, Dawson G, Goldstein BA. A conditional multi-label model to improve prediction of a rare outcome: An illustration predicting autism diagnosis. J Biomed Inform 2024; 157:104711. [PMID: 39182632 DOI: 10.1016/j.jbi.2024.104711] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/26/2024] [Revised: 07/30/2024] [Accepted: 08/17/2024] [Indexed: 08/27/2024]
Abstract
OBJECTIVE This study aimed to develop a novel approach using routinely collected electronic health records (EHRs) data to improve the prediction of a rare event. We illustrated this using an example of improving early prediction of an autism diagnosis, given its low prevalence, by leveraging correlations between autism and other neurodevelopmental conditions (NDCs). METHODS To achieve this, we introduced a conditional multi-label model by merging conditional learning and multi-label methodologies. The conditional learning approach breaks a hard task into more manageable pieces in each stage, and the multi-label approach utilizes information from related neurodevelopmental conditions to learn predictive latent features. The study involved forecasting autism diagnosis by age 5.5 years, utilizing data from the first 18 months of life, and the analysis of feature importance correlations to explore the alignment within the feature space across different conditions. RESULTS Upon analysis of health records from 18,156 children, we are able to generate a model that predicts a future autism diagnosis with moderate performance (AUROC=0.76). The proposed conditional multi-label method significantly improves predictive performance with an AUROC of 0.80 (p < 0.001). Further examination shows that both the conditional and multi-label approach alone provided marginal lift to the model performance compared to a one-stage one-label approach. We also demonstrated the generalizability and applicability of this method using simulated data with high correlation between feature vectors for different labels. CONCLUSION Our findings underscore the effectiveness of the developed conditional multi-label model for early prediction of an autism diagnosis. The study introduces a versatile strategy applicable to prediction tasks involving limited target populations but sharing underlying features or etiology among related groups.
Collapse
Affiliation(s)
- Wei A Huang
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, USA; AI Health, Duke University School of Medicine, Durham, North Carolina, USA
| | - Matthew Engelhard
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, USA; AI Health, Duke University School of Medicine, Durham, North Carolina, USA
| | - Marika Coffman
- Department of Psychiatry, Duke University School of Medicine, Durham, NC, USA
| | - Elliot D Hill
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, USA; AI Health, Duke University School of Medicine, Durham, North Carolina, USA
| | - Qin Weng
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, USA
| | - Abby Scheer
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, USA
| | - Gary Maslow
- Department of Psychiatry, Duke University School of Medicine, Durham, NC, USA; Department of Pediatrics, Duke University School of Medicine, Durham, NC, USA
| | - Ricardo Henao
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, USA; AI Health, Duke University School of Medicine, Durham, North Carolina, USA
| | - Geraldine Dawson
- Department of Psychiatry, Duke University School of Medicine, Durham, NC, USA; Department of Pediatrics, Duke University School of Medicine, Durham, NC, USA
| | - Benjamin A Goldstein
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, USA; AI Health, Duke University School of Medicine, Durham, North Carolina, USA; Department of Pediatrics, Duke University School of Medicine, Durham, NC, USA.
| |
Collapse
|
2
|
German JC, Stirling A, Gorgone P, Brucker AR, Huang A, Dash S, Halpern DJ, Bhavsar NA, McPeek Hinz ER, Shannon RP, Spratt SE, Goldstein BA. Interactive visualization tool to understand and monitor health disparities in diabetes care and outcomes. J Clin Transl Sci 2024; 8:e102. [PMID: 39220819 PMCID: PMC11362627 DOI: 10.1017/cts.2024.542] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/10/2023] [Revised: 04/04/2024] [Accepted: 05/06/2024] [Indexed: 09/04/2024] Open
Abstract
Objective Type 2 diabetes (T2DM) poses a significant public health challenge, with pronounced disparities in control and outcomes. Social determinants of health (SDoH) significantly contribute to these disparities, affecting healthcare access, neighborhood environments, and social context. We discuss the design, development, and use of an innovative web-based application integrating real-world data (electronic health record and geospatial files), to enhance comprehension of the impact of SDoH on T2 DM health disparities. Methods We identified a patient cohort with diabetes from the institutional Diabetes Registry (N = 67,699) within the Duke University Health System. Patient-level information (demographics, comorbidities, service utilization, laboratory results, and medications) was extracted to Tableau. Neighborhood-level socioeconomic status was assessed via the Area Deprivation Index (ADI), and geospatial files incorporated additional data related to points of interest (i.e., parks/green space). Interactive Tableau dashboards were developed to understand risk and contextual factors affecting diabetes management at the individual, group, neighborhood, and population levels. Results The Tableau-powered digital health tool offers dynamic visualizations, identifying T2DM-related disparities. The dashboard allows for the exploration of contextual factors affecting diabetes management (e.g., food insecurity, built environment) and possesses capabilities to generate targeted patient lists for personalized diabetes care planning. Conclusion As part of a broader health equity initiative, this application meets the needs of a diverse range of users. The interactive dashboard, incorporating clinical, sociodemographic, and environmental factors, enhances understanding at various levels and facilitates targeted interventions to address disparities in diabetes care and outcomes. Ultimately, this transformative approach aims to manage SDoH and improve patient care.
Collapse
Affiliation(s)
- Jashalynn C. German
- Department of Medicine, Division of Endocrinology, Metabolism, and Nutrition, Duke University School of Medicine, Durham, NC, USA
| | - Andrew Stirling
- Duke Health Technology Solutions, Duke University Health System, Durham, NC, USA
| | - Patti Gorgone
- Duke Health Technology Solutions, Duke University Health System, Durham, NC, USA
| | - Amanda R. Brucker
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, USA
| | - Angel Huang
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, USA
| | - Shwetha Dash
- Duke Health Technology Solutions, Duke University Health System, Durham, NC, USA
| | - David J. Halpern
- Department of Medicine, Division of General Internal Medicine, Duke University School of Medicine, Durham, NC, USA
- Duke Primary Care, Duke University Medical Center, Durham, NC, USA
| | - Nrupen A. Bhavsar
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, USA
- Department of Medicine, Division of General Internal Medicine, Duke University School of Medicine, Durham, NC, USA
| | - Eugenia R. McPeek Hinz
- Duke Health Technology Solutions, Duke University Health System, Durham, NC, USA
- Duke University Health System, Durham, NC, USA
| | - Richard P. Shannon
- Duke University Health System, Durham, NC, USA
- Department of Medicine, Division of Cardiology, Duke University School of Medicine, Durham, NC, USA
| | - Susan E. Spratt
- Department of Medicine, Division of Endocrinology, Metabolism, and Nutrition, Duke University School of Medicine, Durham, NC, USA
- Duke Population Health Management Office, Duke Health System, Durham, NC, USA
| | - Benjamin A. Goldstein
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC, USA
| |
Collapse
|
3
|
Mohottige D, Davenport CA, Bhavsar N, Schappe T, Lyn MJ, Maxson P, Johnson F, Planey AM, McElroy LM, Wang V, Cabacungan AN, Ephraim P, Lantos P, Peskoe S, Lunyera J, Bentley-Edwards K, Diamantidis CJ, Reich B, Boulware LE. Residential Structural Racism and Prevalence of Chronic Health Conditions. JAMA Netw Open 2023; 6:e2348914. [PMID: 38127347 PMCID: PMC10739116 DOI: 10.1001/jamanetworkopen.2023.48914] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 05/17/2023] [Accepted: 11/01/2023] [Indexed: 12/23/2023] Open
Abstract
Importance Studies elucidating determinants of residential neighborhood-level health inequities are needed. Objective To quantify associations of structural racism indicators with neighborhood prevalence of chronic kidney disease (CKD), diabetes, and hypertension. Design, Setting, and Participants This cross-sectional study used public data (2012-2018) and deidentified electronic health records (2017-2018) to describe the burden of structural racism and the prevalence of CKD, diabetes, and hypertension in 150 residential neighborhoods in Durham County, North Carolina, from US census block groups and quantified their associations using bayesian models accounting for spatial correlations and residents' age. Data were analyzed from January 2021 to May 2023. Exposures Global (neighborhood percentage of White residents, economic-racial segregation, and area deprivation) and discrete (neighborhood child care centers, bus stops, tree cover, reported violent crime, impervious areas, evictions, election participation, income, poverty, education, unemployment, health insurance coverage, and police shootings) indicators of structural racism. Main Outcomes and Measures Outcomes of interest were neighborhood prevalence of CKD, diabetes, and hypertension. Results A total of 150 neighborhoods with a median (IQR) of 1708 (1109-2489) residents; median (IQR) of 2% (0%-6%) Asian residents, 30% (16%-56%) Black residents, 10% (4%-20%) Hispanic or Latino residents, 0% (0%-1%) Indigenous residents, and 44% (18%-70%) White residents; and median (IQR) residential income of $54 531 ($37 729.25-$78 895.25) were included in analyses. In models evaluating global indicators, greater burden of structural racism was associated with greater prevalence of CKD, diabetes, and hypertension (eg, per 1-SD decrease in neighborhood White population percentage: CKD prevalence ratio [PR], 1.27; 95% highest density interval [HDI], 1.18-1.35; diabetes PR, 1.43; 95% HDI, 1.37-1.52; hypertension PR, 1.19; 95% HDI, 1.14-1.25). Similarly in models evaluating discrete indicators, greater burden of structural racism was associated with greater neighborhood prevalence of CKD, diabetes, and hypertension (eg, per 1-SD increase in reported violent crime: CKD PR, 1.15; 95% HDI, 1.07-1.23; diabetes PR, 1.20; 95% HDI, 1.13-1.28; hypertension PR, 1.08; 95% HDI, 1.02-1.14). Conclusions and Relevance This cross-sectional study found several global and discrete structural racism indicators associated with increased prevalence of health conditions in residential neighborhoods. Although inferences from this cross-sectional and ecological study warrant caution, they may help guide the development of future community health interventions.
Collapse
Affiliation(s)
- Dinushika Mohottige
- Institute for Health Equity Research, Department of Population Health, Icahn School of Medicine at Mount Sinai, New York, New York
- Barbara T. Murphy Division of Nephrology, Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, New York
| | | | - Nrupen Bhavsar
- Center for Community and Population Health Improvement, Duke Clinical and Translational Science Institute, Duke University, Durham, North Carolina
- Division of General Internal Medicine, Department of Medicine, Duke University, Durham, North Carolina
| | - Tyler Schappe
- Department of Biostatistics and Bioinformatics, Duke University, Durham, North Carolina
| | - Michelle J. Lyn
- Center for Community and Population Health Improvement, Duke Clinical and Translational Science Institute, Duke University, Durham, North Carolina
- Department of Family Medicine and Community Health, Duke University, Durham, North Carolina
| | - Pamela Maxson
- Center for Community and Population Health Improvement, Duke Clinical and Translational Science Institute, Duke University, Durham, North Carolina
| | - Fred Johnson
- Center for Community and Population Health Improvement, Duke Clinical and Translational Science Institute, Duke University, Durham, North Carolina
- Department of Family Medicine and Community Health, Duke University, Durham, North Carolina
| | - Arrianna M. Planey
- Department of Health Policy and Management, Gillings School of Global Public Health, Chapel Hill, North Carolina
- Cecil G. Sheps Center for Health Services Research, University of North Carolina at Chapel Hill, Chapel Hill
| | - Lisa M. McElroy
- Division of Abdominal Transplant Surgery, Department of Surgery, Duke University, Durham, North Carolina
- Department of Population Health, Duke University, Durham, North Carolina
| | - Virginia Wang
- Division of General Internal Medicine, Department of Medicine, Duke University, Durham, North Carolina
- Department of Population Health, Duke University, Durham, North Carolina
| | - Ashley N. Cabacungan
- Division of General Internal Medicine, Department of Medicine, Duke University, Durham, North Carolina
| | - Patti Ephraim
- Feinstein Institutes for Medical Research, Northwell Health, Manhasset, New York
| | - Paul Lantos
- Duke Global Health Institute, Duke University, Durham, North Carolina
- Department of Pediatrics, Duke University, Durham, North Carolina
- Division of General Internal Medicine, Department of Medicine, Duke University, Durham, North Carolina
| | - Sarah Peskoe
- Department of Biostatistics and Bioinformatics, Duke University, Durham, North Carolina
| | - Joseph Lunyera
- Division of General Internal Medicine, Department of Medicine, Duke University, Durham, North Carolina
| | - Keisha Bentley-Edwards
- Duke Global Health Institute, Duke University, Durham, North Carolina
- Duke Cancer Institute, Duke University, Durham, North Carolina
- Samuel DuBois Cook Center on Social Equity, Duke University, Durham, North Carolina
| | - Clarissa J. Diamantidis
- Division of General Internal Medicine, Department of Medicine, Duke University, Durham, North Carolina
- Division of Nephrology, Department of Medicine, Duke University, Durham, North Carolina
| | - Brian Reich
- Department of Statistics, North Carolina State University, Raleigh
| | - L. Ebony Boulware
- Wake Forest University School of Medicine, Winston Salem, North Carolina
| |
Collapse
|
4
|
Gu Z, Dawson G, Engelhard M. Sex differences in the age of childhood autism diagnosis and the impact of co-occurring conditions. Autism Res 2023; 16:2391-2402. [PMID: 37909391 PMCID: PMC10872618 DOI: 10.1002/aur.3043] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/05/2023] [Accepted: 10/06/2023] [Indexed: 11/03/2023]
Abstract
Sex differences in the age of autism diagnosis during childhood have been documented consistently but remain poorly understood. In this study, we used electronic health records data from a diverse, academic medical center to quantify differences in the age of autism diagnosis between boys and girls and identify associations between the age of diagnosis and co-occurring neurodevelopmental, psychiatric, and medical conditions. An established computable phenotype was used to identify all autism diagnoses within the Duke University Health System between 2014 and 2021. Co-occurring neurodevelopmental and psychiatric diagnoses as well as visits to specific medical and supportive services were identified in the 2 years prior to the autism diagnosis. Cox proportional hazards models were fitted to quantify associations between diagnosis age and sex with and without controlling for the presence of each co-occurring diagnosis and visit type. Records from 1438 individuals (1142 boys and 296 girls) were included. Girls were more likely to be diagnosed either before age 3 (χ 2 = 497.720, p < 0.001) or after age 11 (χ 2 = 4.014, p = 0.047), whereas boys were more likely to be diagnosed between ages 3 and 11 (χ 2 = 5.532, p = 0.019). Visits for anxiety (χ 2 = 4.200, p = 0.040) and mood disorders (χ 2 = 7.033, p = 0.008) were more common in girls and associated with later autism diagnosis (HR = 0.615, p < 0.001; and HR = 0.493, p < 0.001). Visits for otolaryngology were more common in boys and associated with an earlier autism diagnosis (HR = 1.691, p < 0.001). After controlling for these conditions, associations between sex and diagnosis age were reduced and not statistically significant. These results show that the age of autism diagnosis differs in girls compared to boys, but these differences were neutralized when controlling for co-occurring neurodevelopmental and psychiatric conditions prior to autism diagnosis. Understanding sex differences and the possible mediating role of other diagnoses may suggest targets for intervention to promote earlier and more equitable diagnosis.
Collapse
Affiliation(s)
- Zhengyi Gu
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, North Carolina, USA
| | - Geraldine Dawson
- Department of Psychiatry and Behavioral Sciences, Duke University School of Medicine, Durham, North Carolina, USA
| | - Matthew Engelhard
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, North Carolina, USA
| |
Collapse
|
5
|
Wood CT, Truong T, Skinner AC, Armstrong SC, Perrin EM, Woo JG, Green CL. Timing and Magnitude of Peak Body Mass Index and Peak Weight Velocity in Infancy Predict Body Mass Index at 2 Years in a Retrospective Cohort of Electronic Health Record Data. J Pediatr 2023:S0022-3476(23)00115-4. [PMID: 36822510 DOI: 10.1016/j.jpeds.2023.01.021] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/24/2022] [Revised: 01/10/2023] [Accepted: 01/13/2023] [Indexed: 02/23/2023]
Abstract
OBJECTIVES To use growth data from electronic health records to describe and model infant growth (weight velocity and peak body mass index [pBMI]) characteristics. STUDY DESIGN We extracted data from all children born at ≥34 weeks of gestation within one health system between 2014 and 2017. After excluding implausible growth data with an algorithm created for childhood growth, we estimated pBMI, peak weight and length velocities, and the odds of obesity at 2 years, adjusted for race, sex, ethnicity, and birth weight, by the magnitude of peak weight velocity, peak length velocity, and pBMI. RESULTS Among 6425 children (41% White, 28% Black, 26% other race; 16% Hispanic ethnicity), mean pBMI was 17.9 kg/m2 (SD 1.5) and mean age at pBMI was 9.6 months (SD 2.7). Mean peak weight velocity was 949 g (SD 165) per 2 weeks, and the mean peak length velocity was 3.4 cm (SD 0.3) per 2 weeks. Children with obesity at 2 years (n = 931, 14.5%) were more likely to be Hispanic, had greater peak weight and peak length velocities, and had 2 kg/m2 greater magnitude of pBMI than children without obesity. For each unit increase in pBMI, children had more than 4 times greater odds of obesity at age 2 years. CONCLUSIONS In a large sample of infants with clinical growth data tracked via electronic health records, we found associations between the magnitude and timing of peak infant BMI and obesity at 2 years of age.
Collapse
Affiliation(s)
- Charles T Wood
- Division of General Pediatrics and Adolescent Health and Duke Center for Childhood Obesity Research, Department of Pediatrics, Duke University School of Medicine, Durham, NC.
| | - Tracy Truong
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC
| | - Asheley C Skinner
- Department of Population Health Sciences, Duke University Medical School, Durham, NC; Duke Clinical Research Institute, Duke University School of Medicine, Durham, NC
| | - Sarah C Armstrong
- Division of General Pediatrics and Adolescent Health and Duke Center for Childhood Obesity Research, Department of Pediatrics, Duke University School of Medicine, Durham, NC; Duke Clinical Research Institute, Duke University School of Medicine, Durham, NC
| | - Eliana M Perrin
- Department of Pediatrics, Johns Hopkins School of Medicine, and School of Nursing, Baltimore, MD
| | - Jessica G Woo
- Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH; Division of Biostatistics and Epidemiology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH
| | - Cynthia L Green
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC; Duke Clinical Research Institute, Duke University School of Medicine, Durham, NC
| |
Collapse
|
6
|
Engelhard MM, Henao R, Berchuck SI, Chen J, Eichner B, Herkert D, Kollins SH, Olson A, Perrin EM, Rogers U, Sullivan C, Zhu Y, Sapiro G, Dawson G. Predictive Value of Early Autism Detection Models Based on Electronic Health Record Data Collected Before Age 1 Year. JAMA Netw Open 2023; 6:e2254303. [PMID: 36729455 PMCID: PMC9896305 DOI: 10.1001/jamanetworkopen.2022.54303] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/14/2022] [Accepted: 12/15/2022] [Indexed: 02/03/2023] Open
Abstract
Importance Autism detection early in childhood is critical to ensure that autistic children and their families have access to early behavioral support. Early correlates of autism documented in electronic health records (EHRs) during routine care could allow passive, predictive model-based monitoring to improve the accuracy of early detection. Objective To quantify the predictive value of early autism detection models based on EHR data collected before age 1 year. Design, Setting, and Participants This retrospective diagnostic study used EHR data from children seen within the Duke University Health System before age 30 days between January 2006 and December 2020. These data were used to train and evaluate L2-regularized Cox proportional hazards models predicting later autism diagnosis based on data collected from birth up to the time of prediction (ages 30-360 days). Statistical analyses were performed between August 1, 2020, and April 1, 2022. Main Outcomes and Measures Prediction performance was quantified in terms of sensitivity, specificity, and positive predictive value (PPV) at clinically relevant model operating thresholds. Results Data from 45 080 children, including 924 (1.5%) meeting autism criteria, were included in this study. Model-based autism detection at age 30 days achieved 45.5% sensitivity and 23.0% PPV at 90.0% specificity. Detection by age 360 days achieved 59.8% sensitivity and 17.6% PPV at 81.5% specificity and 38.8% sensitivity and 31.0% PPV at 94.3% specificity. Conclusions and Relevance In this diagnostic study of an autism screening test, EHR-based autism detection achieved clinically meaningful accuracy by age 30 days, improving by age 1 year. This automated approach could be integrated with caregiver surveys to improve the accuracy of early autism screening.
Collapse
Affiliation(s)
- Matthew M. Engelhard
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, North Carolina
| | - Ricardo Henao
- Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, North Carolina
- Department of Electrical and Computer Engineering, Duke University, Durham, North Carolina
- Duke AI Health, Durham, North Carolina
| | - Samuel I. Berchuck
- Department of Statistical Science, Duke University, Durham, North Carolina
| | - Junya Chen
- Department of Electrical and Computer Engineering, Duke University, Durham, North Carolina
| | - Brian Eichner
- Department of Pediatrics, Duke University School of Medicine, Durham, North Carolina
| | - Darby Herkert
- Department of Psychiatry and Behavioral Sciences, Duke University School of Medicine, Durham, North Carolina
| | - Scott H. Kollins
- Department of Psychiatry and Behavioral Sciences, Duke University School of Medicine, Durham, North Carolina
| | | | - Eliana M. Perrin
- Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, Maryland
- Department of Pediatrics, Johns Hopkins University School of Nursing, Baltimore, Maryland
| | | | - Connor Sullivan
- Department of Psychiatry and Behavioral Sciences, Duke University School of Medicine, Durham, North Carolina
| | - YiQin Zhu
- Department of Psychiatry and Behavioral Sciences, Duke University School of Medicine, Durham, North Carolina
| | - Guillermo Sapiro
- Department of Electrical and Computer Engineering, Duke University, Durham, North Carolina
- Duke Institute for Brain Sciences, Durham, North Carolina
| | - Geraldine Dawson
- Department of Psychiatry and Behavioral Sciences, Duke University School of Medicine, Durham, North Carolina
- Duke Institute for Brain Sciences, Durham, North Carolina
| |
Collapse
|
7
|
Rao S, Hurst JH, Zhao C, Goldstein BA, Thomas L, Lang JE, Kelly MS. Asthma and the Risk of SARS-CoV-2 Infection Among Children and Adolescents. Pediatrics 2022; 149:185387. [PMID: 35274143 PMCID: PMC9647583 DOI: 10.1542/peds.2021-056164] [Citation(s) in RCA: 8] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Accepted: 03/04/2022] [Indexed: 12/25/2022] Open
Abstract
OBJECTIVES Over 6 million pediatric severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections have occurred in the United States, but risk factors for infection remain poorly defined. We sought to evaluate the association between asthma and SARS-CoV-2 infection risk among children. METHODS We conducted a retrospective cohort study of children 5 to 17 years of age receiving care through the Duke University Health System and who had a Durham County, North Carolina residential address. Children were classified as having asthma using previously validated electronic health record-based definitions. SARS-CoV-2 infections were identified based on positive polymerase chain reaction testing of respiratory samples collected between March 1, 2020, and September 30, 2021. We matched children with asthma 1:1 to children without asthma, using propensity scores and used Poisson regression to evaluate the association between asthma and SARS-CoV-2 infection risk. RESULTS Of 46 900 children, 6324 (13.5%) met criteria for asthma. Children with asthma were more likely to be tested for SARS-CoV-2 infection than children without asthma (33.0% vs 20.9%, P < .0001). In a propensity score-matched cohort of 12 648 children, 706 (5.6%) children tested positive for SARS-CoV-2 infection, including 350 (2.8%) children with asthma and 356 (2.8%) children without asthma (risk ratio: 0.98, 95% confidence interval: 0.85-1.13. There was no evidence of effect modification of this association by inhaled corticosteroid prescription, history of severe exacerbation, or comorbid atopic diseases. Only 1 child with asthma required hospitalization for SARS-CoV-2 infection. CONCLUSIONS After controlling for factors associated with SARS-CoV-2 testing, we found that children with asthma have a similar SARS-CoV-2 infection risk as children without asthma.
Collapse
Affiliation(s)
| | - Jillian H. Hurst
- Children’s Health & Discovery Initiative,Pediatrics, Divisions of Infectious Diseases
| | | | - Benjamin A. Goldstein
- Children’s Health & Discovery Initiative,Departments of Biostatistics and Bioinformatics,Duke Clinical Research Institute, Duke University, Durham, North Carolina
| | - Laine Thomas
- Departments of Biostatistics and Bioinformatics,Duke Clinical Research Institute, Duke University, Durham, North Carolina
| | - Jason E. Lang
- Pulmonary and Sleep Medicine,Duke Clinical Research Institute, Duke University, Durham, North Carolina
| | - Matthew S. Kelly
- Pediatrics, Divisions of Infectious Diseases,Address correspondence to Matthew S. Kelly, MD, MPH, DUMC Box 3499, Durham, NC 27705. E-mail:
| |
Collapse
|
8
|
Hurst JH, Zhao C, Hostetler HP, Ghiasi Gorveh M, Lang JE, Goldstein BA. Environmental and clinical data utility in pediatric asthma exacerbation risk prediction models. BMC Med Inform Decis Mak 2022; 22:108. [PMID: 35459216 PMCID: PMC9034565 DOI: 10.1186/s12911-022-01847-0] [Citation(s) in RCA: 7] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/18/2021] [Accepted: 04/13/2022] [Indexed: 01/05/2023] Open
Abstract
BACKGROUND Asthma exacerbations are triggered by a variety of clinical and environmental factors, but their relative impacts on exacerbation risk are unclear. There is a critical need to develop methods to identify children at high-risk for future exacerbation to allow targeted prevention measures. We sought to evaluate the utility of models using spatiotemporally resolved climatic data and individual electronic health records (EHR) in predicting pediatric asthma exacerbations. METHODS We extracted retrospective EHR data for 5982 children with asthma who had an encounter within the Duke University Health System between January 1, 2014 and December 31, 2019. EHR data were linked to spatially resolved environmental data, and temporally resolved climate, pollution, allergen, and influenza case data. We used xgBoost to build predictive models of asthma exacerbation over 30-180 day time horizons, and evaluated the contributions of different data types to model performance. RESULTS Models using readily available EHR data performed moderately well, as measured by the area under the receiver operating characteristic curve (AUC 0.730-0.742) over all three time horizons. Inclusion of spatial and temporal data did not significantly improve model performance. Generating a decision rule with a sensitivity of 70% produced a positive predictive value of 13.8% for 180 day outcomes but only 2.9% for 30 day outcomes. CONCLUSIONS EHR data-based models perform moderately wellover a 30-180 day time horizon to identify children who would benefit from asthma exacerbation prevention measures. Due to the low rate of exacerbations, longer-term models are likely to be most clinically useful. TRIAL REGISTRATION Not applicable.
Collapse
Affiliation(s)
- Jillian H. Hurst
- grid.26009.3d0000 0004 1936 7961Department of Pediatrics, Division of Infectious Diseases, Duke University School of Medicine, Durham, NC USA ,grid.26009.3d0000 0004 1936 7961Department of Pediatrics, Children’s Health and Discovery Initiative, Duke University School of Medicine, Durham, NC USA
| | - Congwen Zhao
- grid.26009.3d0000 0004 1936 7961Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC USA
| | - Haley P. Hostetler
- grid.26009.3d0000 0004 1936 7961Department of Medicine, Division of Pulmonary, Allergy, and Critical Care Medicine, Duke University School of Medicine, Durham, NC USA
| | - Mohsen Ghiasi Gorveh
- grid.26009.3d0000 0004 1936 7961Duke Clinical Research Institute, Duke University, Durham, NC USA
| | - Jason E. Lang
- grid.26009.3d0000 0004 1936 7961Department of Pediatrics, Division of Pulmonary and Sleep Medicine, Duke University School of Medicine, Durham, NC USA
| | - Benjamin A. Goldstein
- grid.26009.3d0000 0004 1936 7961Department of Pediatrics, Children’s Health and Discovery Initiative, Duke University School of Medicine, Durham, NC USA ,grid.26009.3d0000 0004 1936 7961Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, NC USA ,grid.26009.3d0000 0004 1936 7961Duke Clinical Research Institute, Duke University, Durham, NC USA
| |
Collapse
|