Flory L. Nkoy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Flory L. Nkoy has a total of 122 co-authors. Here are the top co-authors by number of shared publications:
- Bryan L. Stone
Department of Pediatrics, University of Utah, Salt Lake City, UT, United States
20 shared publications - Gang Luo
Amazon, Seattle, WA, USA and University of Washington, Seattle, WA, USA
14 shared publications - Michael D. Johnson
Division of Pediatric Emergency Medicine, Department of Pediatrics, University of Utah School of Medicine, Salt Lake City, UT, USA
10 shared publications - Shan He
Care Transformation, Intermountain Healthcare, Salt Lake City, UT, United States
10 shared publications Bernhard A. Fassl
University of Utah, Salt Lake City, UT, United States
7 shared publicationsBryan L. Stone
University of Utah, Salt Lake City, UT, United States
7 shared publications
Latest Publications
- Open Access
- 22 October 2026
- Published by JMIR Publications Inc. in JMIR Research Protocols
- Vol. 15, e108485
The Efficacy of a Home Monitoring App for Children With Medical Complexity: Protocol for a Randomized Controlled Trial
Background: Children with medical complexity (CMC) experience frequent acute health deteriorations, leading to recurrent hospitalizations. Home monitoring has been suggested to help reduce hospitalizations among CMC, yet the benefits have not been demonstrated. MyChildCMC is a novel app, co-designed with parents, that supports CMC home monitoring to identify early warning signs of acute health changes to promote timely intervention. Objective: This study aims to assess the efficacy of the MyChildCMC intervention on child and parent outcomes and the sustainability of MyChildCMC use by parents and of the intervention’s effects. Methods: This is a 12-month, 2-arm randomized controlled trial recruiting 360 CMC (aged 1‐18 years) and their parents: 180 child-parent dyads assigned to the MyChildCMC app and 180 to the control arm. Participants randomized to the MyChildCMC arm use a sensor and digital thermometer to record vital signs (temperature, heart rate, respiratory rate, and oxygen saturation) and symptoms (pain, feeding/fluid intake, mental status, and seizures) in the app daily and receive immediate feedback. Parents receive alerts for early signs of their child’s health deterioration and can view summary reports and individual graphs about the child’s health status in the app. Alerts are also sent to the care team, who can view, via a web-based dashboard, summary reports and individual graphs to support timely care. Participants randomized to the control arm use the sensor/thermometer to record the same clinical parameters daily in a diary. Both arms receive financial incentives for daily monitoring and for responding to surveys at baseline and at 3, 6, and 12 months. The efficacy will be assessed at 6 months. The monitoring incentive will be stopped at 6 months, but participants will continue with daily monitoring for 6 more months to assess sustainability at 12 months. The primary outcome is the number of hospital admissions among CMC. Secondary outcomes are assessed for the child (hospital days and quality of life) and the parent (stress, self-efficacy, and satisfaction with the child’s care). Results: We will determine the extent to which the use (vs controls) of the MyChildCMC app leads to reduced emergency department/hospital admissions among CMC. We will also determine whether MyChildCMC use leads to improvements in secondary outcomes. Before the current full-scale study, we conducted a pilot trial, which showed promising results. However, the study was not powered to detect differences in hospital admissions. A larger, definitive study of MyChildCMC’s impact was needed to confirm these results. Recruitment for the current study started on January 6, 2026, and will end by March 31, 2028. To date, 41 participants have been enrolled. We expect the last participant to complete the study by March 31, 2029, and thereafter we will analyze the data. Conclusions: The MyChildCMC app was designed to support CMC home monitoring and proactive care. If efficacious, it will serve as a new model for reducing hospital admissions, with the potential for broad dissemination.
- 22 October 2026
- Published by JMIR Publications Inc. in JMIR Research Protocols
- Vol. 15, e108485
Open AccessThe Efficacy of a Home Monitoring App for Children With Medical Complexity: Protocol for a Randomized Controlled Trial
Background: Children with medical complexity (CMC) experience frequent acute health deteriorations, leading to recurrent hospitalizations. Home monitoring has been suggested to help reduce hospitalizations among CMC, yet the benefits have not been demonstrated. MyChildCMC is a novel app, co-designed with parents, that supports CMC home monitoring to identify early warning signs of acute health changes to promote timely intervention. Objective: This study aims to assess the efficacy of the MyChildCMC intervention on child and parent outcomes and the sustainability of MyChildCMC use by parents and of the intervention’s effects. Methods: This is a 12-month, 2-arm randomized controlled trial recruiting 360 CMC (aged 1‐18 years) and their parents: 180 child-parent dyads assigned to the MyChildCMC app and 180 to the control arm. Participants randomized to the MyChildCMC arm use a sensor and digital thermometer to record vital signs (temperature, heart rate, respiratory rate, and oxygen saturation) and symptoms (pain, feeding/fluid intake, mental status, and seizures) in the app daily and receive immediate feedback. Parents receive alerts for early signs of their child’s health deterioration and can view summary reports and individual graphs about the child’s health status in the app. Alerts are also sent to the care team, who can view, via a web-based dashboard, summary reports and individual graphs to support timely care. Participants randomized to the control arm use the sensor/thermometer to record the same clinical parameters daily in a diary. Both arms receive financial incentives for daily monitoring and for responding to surveys at baseline and at 3, 6, and 12 months. The efficacy will be assessed at 6 months. The monitoring incentive will be stopped at 6 months, but participants will continue with daily monitoring for 6 more months to assess sustainability at 12 months. The primary outcome is the number of hospital admissions among CMC. Secondary outcomes are assessed for the child (hospital days and quality of life) and the parent (stress, self-efficacy, and satisfaction with the child’s care). Results: We will determine the extent to which the use (vs controls) of the MyChildCMC app leads to reduced emergency department/hospital admissions among CMC. We will also determine whether MyChildCMC use leads to improvements in secondary outcomes. Before the current full-scale study, we conducted a pilot trial, which showed promising results. However, the study was not powered to detect differences in hospital admissions. A larger, definitive study of MyChildCMC’s impact was needed to confirm these results. Recruitment for the current study started on January 6, 2026, and will end by March 31, 2028. To date, 41 participants have been enrolled. We expect the last participant to complete the study by March 31, 2029, and thereafter we will analyze the data. Conclusions: The MyChildCMC app was designed to support CMC home monitoring and proactive care. If efficacious, it will serve as a new model for reducing hospital admissions, with the potential for broad dissemination.
- Open Access
A Study of Short-Chain Fatty Acids During the Canalicular and Early Saccular Phases of Fetal Lung Development and Childhood Asthma
Background: Emerging literature indicates that the microbiome and its byproducts, such as short-chain fatty acids (SCFAs), play an important role in childhood diseases such as allergies and asthma. Specifically, there is evidence suggesting that SCFAs play a critical role in fetal immunoprogramming during the late saccular phase of fetal lung development. An increase in acetate during the late saccular phase is known to play a critical role in inhibiting histone deacetylases (HDACs), resulting in a cascade of events, including Treg immune regulation, involved in fetal immunoprogramming, and reduction in the asthma phenotype. However, it is not known whether changes in SCFA levels, especially acetate, occurred during the canalicular or early saccular phase among pregnant women whose children did not develop asthma. Methods: In this research, we investigated this question using plasma samples obtained from mothers during the 20th and 28th weeks of pregnancy. Mothers whose children developed asthma were categorized as cases, while those whose children did not were categorized as controls. The specimens were assayed for a panel of SCFAs consisting of acetate, propionate, butyrate, valerate, isobutyrate, and isovalerate. Results: The resulting data indicated no significant differences between the cases and controls, either at week 20 or week 28, in any of the SCFAs measured, despite the vascularization during these phases. Conclusions: We did not find differences in measured SCFAs at week 20 or at week 28. A larger prospective study covering multiple time points is necessary to confirm the findings of this preliminary study. Such a study, together with the published literature regarding later time points, may help discover critical windows during pregnancy when simple manipulation of diet will result in healthier outcomes for infants.
Open AccessA Study of Short-Chain Fatty Acids During the Canalicular and Early Saccular Phases of Fetal Lung Development and Childhood Asthma
Background: Emerging literature indicates that the microbiome and its byproducts, such as short-chain fatty acids (SCFAs), play an important role in childhood diseases such as allergies and asthma. Specifically, there is evidence suggesting that SCFAs play a critical role in fetal immunoprogramming during the late saccular phase of fetal lung development. An increase in acetate during the late saccular phase is known to play a critical role in inhibiting histone deacetylases (HDACs), resulting in a cascade of events, including Treg immune regulation, involved in fetal immunoprogramming, and reduction in the asthma phenotype. However, it is not known whether changes in SCFA levels, especially acetate, occurred during the canalicular or early saccular phase among pregnant women whose children did not develop asthma. Methods: In this research, we investigated this question using plasma samples obtained from mothers during the 20th and 28th weeks of pregnancy. Mothers whose children developed asthma were categorized as cases, while those whose children did not were categorized as controls. The specimens were assayed for a panel of SCFAs consisting of acetate, propionate, butyrate, valerate, isobutyrate, and isovalerate. Results: The resulting data indicated no significant differences between the cases and controls, either at week 20 or week 28, in any of the SCFAs measured, despite the vascularization during these phases. Conclusions: We did not find differences in measured SCFAs at week 20 or at week 28. A larger prospective study covering multiple time points is necessary to confirm the findings of this preliminary study. Such a study, together with the published literature regarding later time points, may help discover critical windows during pregnancy when simple manipulation of diet will result in healthier outcomes for infants.
- Open Access
- 14 June 2024
- Published by MDPI AG in International Journal of Molecular Sciences
- Vol. 25 (12) , 6548
Impact of CYP3A5 Polymorphisms on Pediatric Asthma Outcomes
Genetic variation among inhaled corticosteroid (ICS)-metabolizing enzymes may affect asthma control, but evidence is limited. This study tested the hypothesis that single-nucleotide polymorphisms (SNPs) in Cytochrome P450 3A5 (CYP3A5) would affect asthma outcomes. Patients aged 2–18 years with persistent asthma were recruited to use the electronic AsthmaTracker (e-AT), a self-monitoring tool that records weekly asthma control, medication use, and asthma outcomes. A subset of patients provided saliva samples for SNP analysis and participated in a pharmacokinetic study. Multivariable regression analysis adjusted for age, sex, race, and ethnicity was used to evaluate the impact of CYP3A5 SNPs on asthma outcomes, including asthma control (measured using the asthma symptom tracker, a modified version of the asthma control test or ACT), exacerbations, and hospital admissions. Plasma corticosteroid and cortisol concentrations post-ICS dosing were also assayed using liquid chromatography–tandem mass spectrometry. Of the 751 patients using the e-AT, 166 (22.1%) provided saliva samples and 16 completed the PK study. The e-AT cohort was 65.1% male, and 89.6% White, 6.0% Native Hawaiian, 1.2% Black, 1.2% Native American, 1.8% of unknown race, and 15.7% Hispanic/Latino; the median age was 8.35 (IQR: 5.51–11.3) years. CYP3A5*3/*3 frequency was 75.8% in White subjects, 50% in Native Hawaiians and 76.9% in Hispanic/Latino subjects. Compared with CYP3A5*3/*3, the CYP3A5*1/*x genotype was associated with reduced weekly asthma control (OR: 0.98; 95% CI: 0.97–0.98; p < 0.001), increased exacerbations (OR: 6.43; 95% CI: 4.56–9.07; p < 0.001), and increased asthma hospitalizations (OR: 1.66; 95% CI: 1.43–1.93; p < 0.001); analysis of 3/*3, *1/*1 and *1/*3 separately showed an allelic copy effect. Finally, PK analysis post-ICS dosing suggested muted changes in cortisol concentrations for patients with the CYP3A5*3/*3 genotype, as opposed to an effect on ICS PK. Detection of CYP3A5*3/3, CYPA35*1/*3, and CYP3A5*1/*1 could impact inhaled steroid treatment strategies for asthma in the future.
- 14 June 2024
- Published by MDPI AG in International Journal of Molecular Sciences
- Vol. 25 (12) , 6548
Open AccessImpact of CYP3A5 Polymorphisms on Pediatric Asthma Outcomes
Genetic variation among inhaled corticosteroid (ICS)-metabolizing enzymes may affect asthma control, but evidence is limited. This study tested the hypothesis that single-nucleotide polymorphisms (SNPs) in Cytochrome P450 3A5 (CYP3A5) would affect asthma outcomes. Patients aged 2–18 years with persistent asthma were recruited to use the electronic AsthmaTracker (e-AT), a self-monitoring tool that records weekly asthma control, medication use, and asthma outcomes. A subset of patients provided saliva samples for SNP analysis and participated in a pharmacokinetic study. Multivariable regression analysis adjusted for age, sex, race, and ethnicity was used to evaluate the impact of CYP3A5 SNPs on asthma outcomes, including asthma control (measured using the asthma symptom tracker, a modified version of the asthma control test or ACT), exacerbations, and hospital admissions. Plasma corticosteroid and cortisol concentrations post-ICS dosing were also assayed using liquid chromatography–tandem mass spectrometry. Of the 751 patients using the e-AT, 166 (22.1%) provided saliva samples and 16 completed the PK study. The e-AT cohort was 65.1% male, and 89.6% White, 6.0% Native Hawaiian, 1.2% Black, 1.2% Native American, 1.8% of unknown race, and 15.7% Hispanic/Latino; the median age was 8.35 (IQR: 5.51–11.3) years. CYP3A5*3/*3 frequency was 75.8% in White subjects, 50% in Native Hawaiians and 76.9% in Hispanic/Latino subjects. Compared with CYP3A5*3/*3, the CYP3A5*1/*x genotype was associated with reduced weekly asthma control (OR: 0.98; 95% CI: 0.97–0.98; p < 0.001), increased exacerbations (OR: 6.43; 95% CI: 4.56–9.07; p < 0.001), and increased asthma hospitalizations (OR: 1.66; 95% CI: 1.43–1.93; p < 0.001); analysis of 3/*3, *1/*1 and *1/*3 separately showed an allelic copy effect. Finally, PK analysis post-ICS dosing suggested muted changes in cortisol concentrations for patients with the CYP3A5*3/*3 genotype, as opposed to an effect on ICS PK. Detection of CYP3A5*3/3, CYPA35*1/*3, and CYP3A5*1/*1 could impact inhaled steroid treatment strategies for asthma in the future.
- Open Access
- 17 April 2024
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 12 (1) , e56572
A Roadmap for Using Causal Inference and Machine Learning to Personalize Asthma Medication Selection
Inhaled corticosteroid (ICS) is a mainstay treatment for controlling asthma and preventing exacerbations in patients with persistent asthma. Many types of ICS drugs are used, either alone or in combination with other controller medications. Despite the widespread use of ICSs, asthma control remains suboptimal in many people with asthma. Suboptimal control leads to recurrent exacerbations, causes frequent ER visits and inpatient stays, and is due to multiple factors. One such factor is the inappropriate ICS choice for the patient. While many interventions targeting other factors exist, less attention is given to inappropriate ICS choice. Asthma is a heterogeneous disease with variable underlying inflammations and biomarkers. Up to 50% of people with asthma exhibit some degree of resistance or insensitivity to certain ICSs due to genetic variations in ICS metabolizing enzymes, leading to variable responses to ICSs. Yet, ICS choice, especially in the primary care setting, is often not tailored to the patient’s characteristics. Instead, ICS choice is largely by trial and error and often dictated by insurance reimbursement, organizational prescribing policies, or cost, leading to a one-size-fits-all approach with many patients not achieving optimal control. There is a pressing need for a decision support tool that can predict an effective ICS at the point of care and guide providers to select the ICS that will most likely and quickly ease patient symptoms and improve asthma control. To date, no such tool exists. Predicting which patient will respond well to which ICS is the first step toward developing such a tool. However, no study has predicted ICS response, forming a gap. While the biologic heterogeneity of asthma is vast, few, if any, biomarkers and genotypes can be used to systematically profile all patients with asthma and predict ICS response. As endotyping or genotyping all patients is infeasible, readily available electronic health record data collected during clinical care offer a low-cost, reliable, and more holistic way to profile all patients. In this paper, we point out the need for developing a decision support tool to guide ICS selection and the gap in fulfilling the need. Then we outline an approach to close this gap via creating a machine learning model and applying causal inference to predict a patient’s ICS response in the next year based on the patient’s characteristics. The model uses electronic health record data to characterize all patients and extract patterns that could mirror endotype or genotype. This paper supplies a roadmap for future research, with the eventual goal of shifting asthma care from one-size-fits-all to personalized care, improve outcomes, and save health care resources.
- 17 April 2024
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 12 (1) , e56572
Open AccessA Roadmap for Using Causal Inference and Machine Learning to Personalize Asthma Medication Selection
Inhaled corticosteroid (ICS) is a mainstay treatment for controlling asthma and preventing exacerbations in patients with persistent asthma. Many types of ICS drugs are used, either alone or in combination with other controller medications. Despite the widespread use of ICSs, asthma control remains suboptimal in many people with asthma. Suboptimal control leads to recurrent exacerbations, causes frequent ER visits and inpatient stays, and is due to multiple factors. One such factor is the inappropriate ICS choice for the patient. While many interventions targeting other factors exist, less attention is given to inappropriate ICS choice. Asthma is a heterogeneous disease with variable underlying inflammations and biomarkers. Up to 50% of people with asthma exhibit some degree of resistance or insensitivity to certain ICSs due to genetic variations in ICS metabolizing enzymes, leading to variable responses to ICSs. Yet, ICS choice, especially in the primary care setting, is often not tailored to the patient’s characteristics. Instead, ICS choice is largely by trial and error and often dictated by insurance reimbursement, organizational prescribing policies, or cost, leading to a one-size-fits-all approach with many patients not achieving optimal control. There is a pressing need for a decision support tool that can predict an effective ICS at the point of care and guide providers to select the ICS that will most likely and quickly ease patient symptoms and improve asthma control. To date, no such tool exists. Predicting which patient will respond well to which ICS is the first step toward developing such a tool. However, no study has predicted ICS response, forming a gap. While the biologic heterogeneity of asthma is vast, few, if any, biomarkers and genotypes can be used to systematically profile all patients with asthma and predict ICS response. As endotyping or genotyping all patients is infeasible, readily available electronic health record data collected during clinical care offer a low-cost, reliable, and more holistic way to profile all patients. In this paper, we point out the need for developing a decision support tool to guide ICS selection and the gap in fulfilling the need. Then we outline an approach to close this gap via creating a machine learning model and applying causal inference to predict a patient’s ICS response in the next year based on the patient’s characteristics. The model uses electronic health record data to characterize all patients and extract patterns that could mirror endotype or genotype. This paper supplies a roadmap for future research, with the eventual goal of shifting asthma care from one-size-fits-all to personalized care, improve outcomes, and save health care resources.
- Open Access
- 24 January 2024
- Published by JMIR Publications Inc. in JMIR Preprints
A Roadmap for Using Causal Inference and Machine Learning to Personalize Asthma Medication Selection (Preprint)
UNSTRUCTURED Inhaled corticosteroid (ICS) is a mainstay treatment for controlling asthma and preventing exacerbations in patients with persistent asthma. Many types of ICS drugs are used, either alone or in combination with other controller medications. Despite the widespread use of ICSs, asthma control remains suboptimal in many people with asthma. Suboptimal control leads to recurrent exacerbations, causes frequent ER visits and inpatient stays, and is due to multiple factors. One such factor is the inappropriate ICS choice for the patient. While many interventions targeting other factors exist, less attention is given to inappropriate ICS choice. Asthma is a heterogeneous disease with variable underlying inflammations and biomarkers. Up to 50% of people with asthma exhibit some degree of resistance or insensitivity to certain ICSs due to genetic variations in ICS metabolizing enzymes, leading to variable responses to ICSs. Yet, ICS choice, especially in the primary care setting, is often not tailored to the patient’s characteristics. Instead, ICS choice is largely by trial and error and often dictated by insurance reimbursement, organizational prescribing policies, or cost, leading to a one-size-fits-all approach with many patients not achieving optimal control. There is a pressing need for a decision support tool that can predict an effective ICS at the point of care and guide providers to select the ICS that will most likely and quickly ease patient symptoms and improve asthma control. To date, no such tool exists. Predicting which patient will respond well to which ICS is the first step toward developing such a tool. However, no study has predicted ICS response, forming a gap. While the biologic heterogeneity of asthma is vast, few, if any, biomarkers and genotypes can be used to systematically profile all patients with asthma and predict ICS response. As endotyping or genotyping all patients is infeasible, readily available electronic health record data collected during clinical care offer a low-cost, reliable, and more holistic way to profile all patients. In this paper, we point out the need for developing a decision support tool to guide ICS selection and the gap in fulfilling the need. Then we outline an approach to close this gap via creating a machine learning model and applying causal inference to predict a patient’s ICS response in the next year based on the patient’s characteristics. The model uses electronic health record data to characterize all patients and extract patterns that could mirror endotype or genotype. This paper supplies a roadmap for future research, with the eventual goal of shifting asthma care from one-size-fits-all to personalized care, improve outcomes, and save health care resources.
- 24 January 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessA Roadmap for Using Causal Inference and Machine Learning to Personalize Asthma Medication Selection (Preprint)
UNSTRUCTURED Inhaled corticosteroid (ICS) is a mainstay treatment for controlling asthma and preventing exacerbations in patients with persistent asthma. Many types of ICS drugs are used, either alone or in combination with other controller medications. Despite the widespread use of ICSs, asthma control remains suboptimal in many people with asthma. Suboptimal control leads to recurrent exacerbations, causes frequent ER visits and inpatient stays, and is due to multiple factors. One such factor is the inappropriate ICS choice for the patient. While many interventions targeting other factors exist, less attention is given to inappropriate ICS choice. Asthma is a heterogeneous disease with variable underlying inflammations and biomarkers. Up to 50% of people with asthma exhibit some degree of resistance or insensitivity to certain ICSs due to genetic variations in ICS metabolizing enzymes, leading to variable responses to ICSs. Yet, ICS choice, especially in the primary care setting, is often not tailored to the patient’s characteristics. Instead, ICS choice is largely by trial and error and often dictated by insurance reimbursement, organizational prescribing policies, or cost, leading to a one-size-fits-all approach with many patients not achieving optimal control. There is a pressing need for a decision support tool that can predict an effective ICS at the point of care and guide providers to select the ICS that will most likely and quickly ease patient symptoms and improve asthma control. To date, no such tool exists. Predicting which patient will respond well to which ICS is the first step toward developing such a tool. However, no study has predicted ICS response, forming a gap. While the biologic heterogeneity of asthma is vast, few, if any, biomarkers and genotypes can be used to systematically profile all patients with asthma and predict ICS response. As endotyping or genotyping all patients is infeasible, readily available electronic health record data collected during clinical care offer a low-cost, reliable, and more holistic way to profile all patients. In this paper, we point out the need for developing a decision support tool to guide ICS selection and the gap in fulfilling the need. Then we outline an approach to close this gap via creating a machine learning model and applying causal inference to predict a patient’s ICS response in the next year based on the patient’s characteristics. The model uses electronic health record data to characterize all patients and extract patterns that could mirror endotype or genotype. This paper supplies a roadmap for future research, with the eventual goal of shifting asthma care from one-size-fits-all to personalized care, improve outcomes, and save health care resources.
- 1 August 2021
- Published by Elsevier in Academic Pediatrics
- Vol. 21 (6) , 1018-1024
Impact of a National Quality Collaborative on Pediatric Asthma Care Quality by Insurance Status
- 1 August 2021
- Published by Elsevier in Academic Pediatrics
- Vol. 21 (6) , 1018-1024
Impact of a National Quality Collaborative on Pediatric Asthma Care Quality by Insurance Status
- Open Access
- 18 May 2021
- Published by JMIR Publications Inc. in JMIR Research Protocols
- Vol. 10 (5) , e27065
Using Computational Methods to Improve Integrated Disease Management for Asthma and Chronic Obstructive Pulmonary Disease: Protocol for a Secondary Analysis
Background: Asthma and chronic obstructive pulmonary disease (COPD) impose a heavy burden on health care. Approximately one-fourth of patients with asthma and patients with COPD are prone to exacerbations, which can be greatly reduced by preventive care via integrated disease management that has a limited service capacity. To do this well, a predictive model for proneness to exacerbation is required, but no such model exists. It would be suboptimal to build such models using the current model building approach for asthma and COPD, which has 2 gaps due to rarely factoring in temporal features showing early health changes and general directions. First, existing models for other asthma and COPD outcomes rarely use more advanced temporal features, such as the slope of the number of days to albuterol refill, and are inaccurate. Second, existing models seldom show the reason a patient is deemed high risk and the potential interventions to reduce the risk, making already occupied clinicians expend more time on chart review and overlook suitable interventions. Regular automatic explanation methods cannot deal with temporal data and address this issue well. Objective: To enable more patients with asthma and patients with COPD to obtain suitable and timely care to avoid exacerbations, we aim to implement comprehensible computational methods to accurately predict proneness to exacerbation and recommend customized interventions. Methods: We will use temporal features to accurately predict proneness to exacerbation, automatically find modifiable temporal risk factors for every high-risk patient, and assess the impact of actionable warnings on clinicians’ decisions to use integrated disease management to prevent proneness to exacerbation. Results: We have obtained most of the clinical and administrative data of patients with asthma from 3 prominent American health care systems. We are retrieving other clinical and administrative data, mostly of patients with COPD, needed for the study. We intend to complete the study in 6 years. Conclusions: Our results will help make asthma and COPD care more proactive, effective, and efficient, improving outcomes and saving resources. International Registered Report Identifier (IRRID): PRR1-10.2196/27065
- 18 May 2021
- Published by JMIR Publications Inc. in JMIR Research Protocols
- Vol. 10 (5) , e27065
Open AccessUsing Computational Methods to Improve Integrated Disease Management for Asthma and Chronic Obstructive Pulmonary Disease: Protocol for a Secondary Analysis
Background: Asthma and chronic obstructive pulmonary disease (COPD) impose a heavy burden on health care. Approximately one-fourth of patients with asthma and patients with COPD are prone to exacerbations, which can be greatly reduced by preventive care via integrated disease management that has a limited service capacity. To do this well, a predictive model for proneness to exacerbation is required, but no such model exists. It would be suboptimal to build such models using the current model building approach for asthma and COPD, which has 2 gaps due to rarely factoring in temporal features showing early health changes and general directions. First, existing models for other asthma and COPD outcomes rarely use more advanced temporal features, such as the slope of the number of days to albuterol refill, and are inaccurate. Second, existing models seldom show the reason a patient is deemed high risk and the potential interventions to reduce the risk, making already occupied clinicians expend more time on chart review and overlook suitable interventions. Regular automatic explanation methods cannot deal with temporal data and address this issue well. Objective: To enable more patients with asthma and patients with COPD to obtain suitable and timely care to avoid exacerbations, we aim to implement comprehensible computational methods to accurately predict proneness to exacerbation and recommend customized interventions. Methods: We will use temporal features to accurately predict proneness to exacerbation, automatically find modifiable temporal risk factors for every high-risk patient, and assess the impact of actionable warnings on clinicians’ decisions to use integrated disease management to prevent proneness to exacerbation. Results: We have obtained most of the clinical and administrative data of patients with asthma from 3 prominent American health care systems. We are retrieving other clinical and administrative data, mostly of patients with COPD, needed for the study. We intend to complete the study in 6 years. Conclusions: Our results will help make asthma and COPD care more proactive, effective, and efficient, improving outcomes and saving resources. International Registered Report Identifier (IRRID): PRR1-10.2196/27065
- 1 May 2021
- Published by American Academy of Pediatrics (AAP) in Hospital Pediatrics
- Vol. 11 (5) , 492-502
Home-Monitoring Application for Children With Medical Complexity: A Feasibility Trial
OBJECTIVES: Mobile apps are suggested for supporting home monitoring and reducing emergency department (ED) visits and hospitalizations for children with medical complexity (CMC). None have been implemented. We sought to assess the MyChildCMC app (1) feasibility for CMC home monitoring, (2) ability to detect early deteriorations before ED and hospital admissions, and (3) preliminary impact. METHODS: Parents of CMC (aged 1–21 years) admitted to a children’s hospital were randomly assigned to MyChildCMC or usual care. MyChildCMC subjects recorded their child’s vital signs and symptoms daily for 3 months postdischarge and received real-time feedback. Feasibility measures included parent’s enrollment, retention, and engagement. The preliminary impact was determined by using quality of life, parent satisfaction with care, and subsequent ED and hospital admissions and hospital days. RESULTS: A total of 62 parents and CMC were invited to participate: 50 enrolled (80.6% enrollment rate) and were randomly assigned to MyChildCMC (n = 24) or usual care (n = 26). Retention at 1 and 3 months was 80% and 74%, and engagement was 68.3% and 62.6%. Run-chart shifts in vital signs were common findings preceding admissions. The satisfaction score was 26.9 in the MyChildCMC group and 24.1 in the control group (P = .035). No quality of life or subsequent admission differences occurred between groups. The 3-month hospital days (pre-post enrollment) decreased from 9.25 to 4.54 days (rate ratio = 0.49; 95% confidence interval = 0.39–0.62; P < .001) in the MyChildCMC group and increased from 1.08 to 2.46 days (rate ratio = 2.29; 95% confidence interval = 1.47–3.56; P < .001) in the control group. CONCLUSIONS: MyChildCMC was feasible and appears effective, with the potential to detect early deteriorations in health for timely interventions that might avoid ED and hospitalizations. A larger and definitive study of MyChildCMC’s impact and sustainability is needed.
- 1 May 2021
- Published by American Academy of Pediatrics (AAP) in Hospital Pediatrics
- Vol. 11 (5) , 492-502
Home-Monitoring Application for Children With Medical Complexity: A Feasibility Trial
OBJECTIVES: Mobile apps are suggested for supporting home monitoring and reducing emergency department (ED) visits and hospitalizations for children with medical complexity (CMC). None have been implemented. We sought to assess the MyChildCMC app (1) feasibility for CMC home monitoring, (2) ability to detect early deteriorations before ED and hospital admissions, and (3) preliminary impact. METHODS: Parents of CMC (aged 1–21 years) admitted to a children’s hospital were randomly assigned to MyChildCMC or usual care. MyChildCMC subjects recorded their child’s vital signs and symptoms daily for 3 months postdischarge and received real-time feedback. Feasibility measures included parent’s enrollment, retention, and engagement. The preliminary impact was determined by using quality of life, parent satisfaction with care, and subsequent ED and hospital admissions and hospital days. RESULTS: A total of 62 parents and CMC were invited to participate: 50 enrolled (80.6% enrollment rate) and were randomly assigned to MyChildCMC (n = 24) or usual care (n = 26). Retention at 1 and 3 months was 80% and 74%, and engagement was 68.3% and 62.6%. Run-chart shifts in vital signs were common findings preceding admissions. The satisfaction score was 26.9 in the MyChildCMC group and 24.1 in the control group (P = .035). No quality of life or subsequent admission differences occurred between groups. The 3-month hospital days (pre-post enrollment) decreased from 9.25 to 4.54 days (rate ratio = 0.49; 95% confidence interval = 0.39–0.62; P < .001) in the MyChildCMC group and increased from 1.08 to 2.46 days (rate ratio = 2.29; 95% confidence interval = 1.47–3.56; P < .001) in the control group. CONCLUSIONS: MyChildCMC was feasible and appears effective, with the potential to detect early deteriorations in health for timely interventions that might avoid ED and hospitalizations. A larger and definitive study of MyChildCMC’s impact and sustainability is needed.
- Open Access
- 16 January 2021
- Published by JMIR Publications Inc. in JMIR Preprints
Using Computational Methods to Improve Integrated Disease Management for Asthma and Chronic Obstructive Pulmonary Disease: Protocol for a Secondary Analysis (Preprint)
BACKGROUND Asthma and chronic obstructive pulmonary disease (COPD) impose a heavy burden on health care. Approximately one-fourth of patients with asthma and patients with COPD are prone to exacerbations, which can be greatly reduced by preventive care via integrated disease management that has a limited service capacity. To do this well, a predictive model for proneness to exacerbation is required, but no such model exists. It would be suboptimal to build such models using the current model building approach for asthma and COPD, which has 2 gaps due to rarely factoring in temporal features showing early health changes and general directions. First, existing models for other asthma and COPD outcomes rarely use more advanced temporal features, such as the slope of the number of days to albuterol refill, and are inaccurate. Second, existing models seldom show the reason a patient is deemed high risk and the potential interventions to reduce the risk, making already occupied clinicians expend more time on chart review and overlook suitable interventions. Regular automatic explanation methods cannot deal with temporal data and address this issue well. OBJECTIVE To enable more patients with asthma and patients with COPD to obtain suitable and timely care to avoid exacerbations, we aim to implement comprehensible computational methods to accurately predict proneness to exacerbation and recommend customized interventions. METHODS We will use temporal features to accurately predict proneness to exacerbation, automatically find modifiable temporal risk factors for every high-risk patient, and assess the impact of actionable warnings on clinicians’ decisions to use integrated disease management to prevent proneness to exacerbation. RESULTS We have obtained most of the clinical and administrative data of patients with asthma from 3 prominent American health care systems. We are retrieving other clinical and administrative data, mostly of patients with COPD, needed for the study. We intend to complete the study in 6 years. CONCLUSIONS Our results will help make asthma and COPD care more proactive, effective, and efficient, improving outcomes and saving resources. INTERNATIONAL REGISTERED REPORT PRR1-10.2196/27065
- 16 January 2021
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessUsing Computational Methods to Improve Integrated Disease Management for Asthma and Chronic Obstructive Pulmonary Disease: Protocol for a Secondary Analysis (Preprint)
BACKGROUND Asthma and chronic obstructive pulmonary disease (COPD) impose a heavy burden on health care. Approximately one-fourth of patients with asthma and patients with COPD are prone to exacerbations, which can be greatly reduced by preventive care via integrated disease management that has a limited service capacity. To do this well, a predictive model for proneness to exacerbation is required, but no such model exists. It would be suboptimal to build such models using the current model building approach for asthma and COPD, which has 2 gaps due to rarely factoring in temporal features showing early health changes and general directions. First, existing models for other asthma and COPD outcomes rarely use more advanced temporal features, such as the slope of the number of days to albuterol refill, and are inaccurate. Second, existing models seldom show the reason a patient is deemed high risk and the potential interventions to reduce the risk, making already occupied clinicians expend more time on chart review and overlook suitable interventions. Regular automatic explanation methods cannot deal with temporal data and address this issue well. OBJECTIVE To enable more patients with asthma and patients with COPD to obtain suitable and timely care to avoid exacerbations, we aim to implement comprehensible computational methods to accurately predict proneness to exacerbation and recommend customized interventions. METHODS We will use temporal features to accurately predict proneness to exacerbation, automatically find modifiable temporal risk factors for every high-risk patient, and assess the impact of actionable warnings on clinicians’ decisions to use integrated disease management to prevent proneness to exacerbation. RESULTS We have obtained most of the clinical and administrative data of patients with asthma from 3 prominent American health care systems. We are retrieving other clinical and administrative data, mostly of patients with COPD, needed for the study. We intend to complete the study in 6 years. CONCLUSIONS Our results will help make asthma and COPD care more proactive, effective, and efficient, improving outcomes and saving resources. INTERNATIONAL REGISTERED REPORT PRR1-10.2196/27065
- Open Access
- 31 December 2020
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 8 (12) , e21965
Automatically Explaining Machine Learning Prediction Results on Asthma Hospital Visits in Patients With Asthma: Secondary Analysis
Background: Asthma is a major chronic disease that poses a heavy burden on health care. To facilitate the allocation of care management resources aimed at improving outcomes for high-risk patients with asthma, we recently built a machine learning model to predict asthma hospital visits in the subsequent year in patients with asthma. Our model is more accurate than previous models. However, like most machine learning models, it offers no explanation of its prediction results. This creates a barrier for use in care management, where interpretability is desired. Objective: This study aims to develop a method to automatically explain the prediction results of the model and recommend tailored interventions without lowering the performance measures of the model. Methods: Our data were imbalanced, with only a small portion of data instances linking to future asthma hospital visits. To handle imbalanced data, we extended our previous method of automatically offering rule-formed explanations for the prediction results of any machine learning model on tabular data without lowering the model’s performance measures. In a secondary analysis of the 334,564 data instances from Intermountain Healthcare between 2005 and 2018 used to form our model, we employed the extended method to automatically explain the prediction results of our model and recommend tailored interventions. The patient cohort consisted of all patients with asthma who received care at Intermountain Healthcare between 2005 and 2018, and resided in Utah or Idaho as recorded at the visit. Results: Our method explained the prediction results for 89.7% (391/436) of the patients with asthma who, per our model’s correct prediction, were likely to incur asthma hospital visits in the subsequent year. Conclusions: This study is the first to demonstrate the feasibility of automatically offering rule-formed explanations for the prediction results of any machine learning model on imbalanced tabular data without lowering the performance measures of the model. After further improvement, our asthma outcome prediction model coupled with the automatic explanation function could be used by clinicians to guide the allocation of limited asthma care management resources and the identification of appropriate interventions.
- 31 December 2020
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 8 (12) , e21965
Open AccessAutomatically Explaining Machine Learning Prediction Results on Asthma Hospital Visits in Patients With Asthma: Secondary Analysis
Background: Asthma is a major chronic disease that poses a heavy burden on health care. To facilitate the allocation of care management resources aimed at improving outcomes for high-risk patients with asthma, we recently built a machine learning model to predict asthma hospital visits in the subsequent year in patients with asthma. Our model is more accurate than previous models. However, like most machine learning models, it offers no explanation of its prediction results. This creates a barrier for use in care management, where interpretability is desired. Objective: This study aims to develop a method to automatically explain the prediction results of the model and recommend tailored interventions without lowering the performance measures of the model. Methods: Our data were imbalanced, with only a small portion of data instances linking to future asthma hospital visits. To handle imbalanced data, we extended our previous method of automatically offering rule-formed explanations for the prediction results of any machine learning model on tabular data without lowering the model’s performance measures. In a secondary analysis of the 334,564 data instances from Intermountain Healthcare between 2005 and 2018 used to form our model, we employed the extended method to automatically explain the prediction results of our model and recommend tailored interventions. The patient cohort consisted of all patients with asthma who received care at Intermountain Healthcare between 2005 and 2018, and resided in Utah or Idaho as recorded at the visit. Results: Our method explained the prediction results for 89.7% (391/436) of the patients with asthma who, per our model’s correct prediction, were likely to incur asthma hospital visits in the subsequent year. Conclusions: This study is the first to demonstrate the feasibility of automatically offering rule-formed explanations for the prediction results of any machine learning model on imbalanced tabular data without lowering the performance measures of the model. After further improvement, our asthma outcome prediction model coupled with the automatic explanation function could be used by clinicians to guide the allocation of limited asthma care management resources and the identification of appropriate interventions.