Prediction of short-acting beta-agonist usage in patients with asthma using temporal-convolutional neural networks.

Objective: Changes in short-acting beta-agonist (SABA) use are an important signal of asthma control and risk of asthma exacerbations. Inhaler sensors passively capture SABA use and may provide longitudinal data to identify at-riskpatients. We evaluate the performance of several ML models in predicting daily SABA use for participants with asthma and determine relevant features for predictive accuracy. Methods: Participants with self-reported asthma enrolled in a digital health platform (Propeller Health, WI), which included a smartphone application and inhaler sensors that collected the date and time of SABA use. Linear regression, random forests, and temporal convolutional networks (TCN) were applied to predict expected SABA puffs/person/day from SABA usage and environmental triggers. The models were compared with a simple baseline model using explained variance (R2), as well as using average precision (AP) and area under the receiving operator characteristic curve (ROC AUC) for predicting days with ≥1-10 puffs. Results: Data included 1.2 million days of data from 13 202 participants. A TCN outperformed other models in predicting puff count (R2 = 0.562) and day-over-day change in puff count (R2 = 0.344). The TCN predicted days with ≥10 puffs with an ROC AUC score of 0.952 and an AP of 0.762 for predicting a day with ≥1 puffs. SABA use over the preceding 7 days had the highest feature importance, with a smaller but meaningful contribution from air pollutant features. Conclusion: Predicted SABA use may serve as a valuable forward-looking signal to inform early clinical intervention and self-management. Further validation with known exacerbation events is needed.

Relationship between Self-reported Sleepiness and Positive Airway Pressure Treatment Adherence in Obstructive Sleep Apnea.

Rationale: Positive airway pressure (PAP) is the first-choice treatment for obstructive sleep apnea (OSA). However, its real-world effectiveness is often questioned because of usage issues. The relationship between patient sleepiness and PAP usage has been assessed in relatively small and selected populations within the research context. Objectives: To assess the impact of patient-reported sleep outcomes, particularly self-reported sleepiness and its change during therapy, on PAP usage in the real-world setting. Methods: Deidentified data for U.S.-based patients receiving PAP therapy were examined. Eligible patients were registered in the myAir app and provided self-reported sleepiness at baseline and after 7, 14, 21, and 28 days of PAP between November 2019 and April 2020. Results: A total of 95,397 registered patients met all eligibility criteria and were included in the analysis (mean age, 49.6 ± 13.0 yr; 61.6% male). Daytime sleepiness was the most common reason for PAP therapy initiation (57.1% of patients), and 42.2% of all patients had self-reported moderate to severe OSA. Self-reported sleepiness improved with PAP therapy in most patients over the assessment period, with 62.1% of patients reporting "no" or "slight" sleepiness at Day 28. There was a dose-dependent association between improvement in self-reported sleepiness at Day 28 and PAP usage, and this finding was maintained at Day 360. Self-reported sleepiness at Day 28 was associated with achieving U.S. Centers for Medicare & Medicaid Services compliance at 90 days (approximately 90% for those with no or slight sleepiness vs. <70% for those with residual very or extreme sleepiness); average daily PAP usage over 360 days was ⩾5.0 and ⩽3.7 hours, respectively, for those with no or slight versus very or extreme sleepiness. Conclusions: This study demonstrates the feasibility of capturing patient-reported outcomes via a digital platform. Patient-reported outcomes appear to be associated with PAP usage, especially self-reported sleepiness and its response to therapy. Capturing patient-reported outcomes using digital solutions during the course of treatment has the potential to enhance patient outcomes by providing actionable insights.

Improvement in symptom-related disruptions is associated with fewer days of short-acting beta-agonist use in asthma.

Significant indirect healthcare costs are related to uncontrolled asthma, including productivity loss. Days with short-acting beta-agonist (SABA) use is associated with symptom-related disruptions at work, home, and school. Digital self-management platforms may support fewer days with SABA medication use and may reduce symptom-related disruptions.

Identifying impacts of air pollution on subacute asthma symptoms using digital medication sensors.

BACKGROUND: Objective tracking of asthma medication use and exposure in real-time and space has not been feasible previously. Exposure assessments have typically been tied to residential locations, which ignore exposure within patterns of daily activities. METHODS: We investigated the associations of exposure to multiple air pollutants, derived from nearest air quality monitors, with space-time asthma rescue inhaler use captured by digital sensors, in Jefferson County, Kentucky. A generalized linear mixed model, capable of accounting for repeated measures, over-dispersion and excessive zeros, was used in our analysis. A secondary analysis was done through the random forest machine learning technique. RESULTS: The 1039 participants enrolled were 63.4% female, 77.3% adult (>18) and 46.8% White. Digital sensors monitored the time and location of over 286 980 asthma rescue medication uses and associated air pollution exposures over 193 697 patient-days, creating a rich spatiotemporal dataset of over 10 905 240 data elements. In the generalized linear mixed model, an interquartile range (IQR) increase in pollutant exposure was associated with a mean rescue medication use increase per person per day of 0.201 [95% confidence interval (CI): 0.189-0.214], 0.153 (95% CI: 0.136-0.171), 0.131 (95% CI: 0.115-0.147) and 0.113 (95% CI: 0.097-0.129), for sulphur dioxide (SO2), nitrogen dioxide (NO2), fine particulate matter (PM2.5) and ozone (O3), respectively. Similar effect sizes were identified with the random forest model. Time-lagged exposure effects of 0-3 days were observed. CONCLUSIONS: Daily exposure to multiple pollutants was associated with increases in daily asthma rescue medication use for same day and lagged exposures up to 3 days. Associations were consistent when evaluated with the random forest modelling approach.

The relationship between objective app engagement and medication adherence in asthma and COPD: a retrospective analysis.

Digital health tools can promote disease self-management, but the association of smartphone app engagement and medication adherence is unclear. We assessed the relationship between objective smartphone app engagement and controller medication use in adults with asthma and COPD. We retrospectively analyzed data from participants enrolled in a digital self-management platform for asthma and COPD. Eligible adults had a smartphone and a paired electronic medication monitor (EMM). Longitudinal, mixed-effects logistic regressions estimated the relationship between daily app engagement (app opens, session duration) and daily controller medication use. Data from 2309 participants (71% asthma; 29% COPD) was analyzed. Opening the app (vs. not opening the app) was associated with significantly greater odds (OR (95% CI)) of using controller medications in asthma (2.08 (1.98, 2.19)) and COPD (1.61 (1.49, 1.75). Longer session duration was also associated with greater odds of using controller medications in asthma and COPD, but the odds of use attenuated with longer session duration in COPD. This study presents a novel assessment of the relationship between objectively-measured smartphone app engagement and controller medication use in asthma and COPD. Such insights may help develop targeted digital health tools and interventions.

Concurrent Improvement Observed in Patient-Reported Burden and Sensor-Collected Medication Use Among Patients Enrolled in a COPD Digital Health Program.

Background: The COPD assessment test (CAT) is an 8-item questionnaire widely used in clinical practice to assess patient burden of disease. Digital health platforms that leverage electronic medication monitors (EMMs) are used to track the time and date of maintenance and short-acting beta-agonist (SABA) inhaler medication use and record patient-reported outcomes. The study examined changes in CAT and SABA inhaler use in COPD to determine whether passively collected SABA and CAT scores changed in a parallel manner. Methods: Patients with self-reported COPD enrolled in a digital health program, which provided EMMs to track SABA and maintenance inhaler use, and a companion smartphone application ("app") to provide medication feedback and reminders. Patients completing the CAT questionnaire in the app at enrollment and at 6 months were included in the analysis. Changes in CAT burden category [by the minimally important difference (MID)] and changes in EMM-recorded mean SABA inhaler use per day were quantified at baseline and 6 months. Results: The analysis included 611 patients. At 6 months, mean CAT improved by -0.9 (95% CI: -1.4, -0.4; p < 0.001) points, and mean SABA use decreased by -0.6 (-0.8, -0.4; p < 0.001) puffs/day. Among patients with higher burden (CAT ≥ 21) at enrollment, CAT improved by -2.0 (-2.6, -1.4; p < 0.001) points, and SABA use decreased by -0.8 (-1.1, -0.6; p < 0.001) puffs/day. Conclusion: Significant and parallel improvement in CAT scores and SABA use at 6 months were noted among patients enrolled in a digital health program, with greater improvement for patients with higher disease burden.

Improved asthma outcomes observed in the vicinity of coal power plant retirement, retrofit, and conversion to natural gas.

Coal-fired power plants release substantial air pollution, including over 60% of U.S. sulfur dioxide (SO2) emissions in 2014. Such air pollution may exacerbate asthma however direct studies of health impacts linked to power plant air pollution are rare. Here, we take advantage of a natural experiment in Louisville, Kentucky, where one coal-fired power plant retired and converted to natural gas, and three others installed SO2 emission control systems between 2013 and 2016. Dispersion modeling indicated exposure to SO2 emissions from these power plants decreased after the energy transitions. We used several analysis strategies, including difference-in-differences, first-difference, and interrupted time-series modeling to show that the emissions control installations and plant retirements were associated with reduced asthma disease burden related to ZIP code-level hospitalizations and emergency room visits, and individual-level medication use as measured by digital medication sensors.

Geospatial-temporal analysis of the impact of ozone on asthma rescue inhaler use.

RATIONALE: Asthma is one of the most common chronic respiratory diseases in the United States. Several outdoor air pollutants have been associated with asthma morbidity. Previous studies of the effects of short-term air pollution exposure have been limited by potential exposure misclassification and limited spatial and temporal resolution of asthma outcome measures. OBJECTIVES: We aimed to assess the association of short-term air pollutant exposure with the use of short-acting beta-2 agonists (SABA) for asthma by monitoring the time and place of occurrence with electronic medication monitors. METHODS: In a cohort of adults and children with asthma (n = 287; 60% female), we deployed electronic medication monitors fitted to metered-dose inhalers to monitor SABA use, capturing the date, time and location of use. We assigned pollutant exposures based on each actuation's time and location (4-h mean measures for ozone and particulate matter of 2.5 µm or smaller (PM2.5)), assessed associations using generalized linear models and explored age-specific effects. MEASUREMENTS AND MAIN RESULTS: Ambient ozone exposure was positively associated with SABA use (p = 0.01). Age-specific associations were identified (interaction p = 0.01), with a larger increase in SABA use for children (11.3%; 95% CI: 7.0%-18.2%) than adults (8.4%; 95% CI: 6.4%-11.0%) per IQR increase of ozone (16.8 ppb). CONCLUSIONS: These findings support existing evidence that short-term exposure to ozone can cause morbidity in individuals with asthma, and suggest that ozone exposures below the current U.S. EPA standard may be associated with increased SABA use.

Passive Monitoring of Short-Acting Beta-Agonist Use via Digital Platform in Patients With Chronic Obstructive Pulmonary Disease: Quality Improvement Retrospective Analysis.

BACKGROUND: Digital health programs assist patients with chronic obstructive pulmonary disease (COPD) to better manage their disease. Technological and adoption barriers have been perceived as a limitation. OBJECTIVE: The aim of the research was to evaluate a digital quality improvement pilot in Medicare-eligible patients with COPD. METHODS: COPD patients were enrolled in a digital platform to help manage their medications and symptoms as part of their routine clinical care. Patients were provided with electronic medication monitors (EMMs) to monitor short-acting beta-agonist (SABA) use passively and a smartphone app to track use trends and receive feedback. Providers also had access to data collected via a secure website and were sent email notifications if a patient had a significant change in their prescribed inhaler use. Providers then determined if follow-up was needed. Change in SABA use and feasibility outcomes were evaluated at 3, 6, and 12 months. RESULTS: A total of 190 patients enrolled in the pilot. At 3, 6, and 12 months, patients recorded significant reductions in daily and nighttime SABA use and increases in SABA-free days (all P<.001). Patient engagement, as measured by the ratio of daily active use to monthly active use, was >90% at both 6 and 12 months. Retention at 6 months was 81% (154/190). Providers were sent on average two email notifications per patient during the 12-month program. CONCLUSIONS: A digital health program integrated as part of standard clinical practice was feasible and had low provider burden. The pilot demonstrated significant reduction in SABA use and increased SABA-free days among Medicare-eligible COPD patients. Further, patients readily adopted the digital platform and demonstrated strong engagement and retention rates at 6 and 12 months.

Short-term impact of PM2.5 on contemporaneous asthma medication use: Behavior and the value of pollution reductions.

Asthma ranks among the most costly of chronic diseases, accounting for over $50 billion annually in direct medical expenditures in the United States. At the same time, evidence has accumulated that fine particulate matter pollution can exacerbate asthma symptoms and generate substantial economic costs. To measure these costs, we use a unique nationwide panel dataset tracking asthmatic individuals' use of rescue medication and their exposure to PM2.5 (particulate matter with an aerodynamic diameter of <2.5 µm) concentration between 2012 and 2017, to estimate the causal relationship between pollution and inhaler use. Our sample consists of individuals using an asthma digital health platform, which relies on a wireless sensor to track the place and time of inhaler use events, as well as regular nonevent location and time indicators. These data provide an accurate measurement of inhaler use and allow spatially and temporally resolute assignment of pollution exposure. Using a high-frequency research design and individual fixed effects, we find that a 1 µg/m3 (12%) increase in weekly exposure to PM2.5 increases weekly inhaler use by 0.82%. We also show that there is seasonal, regional, and income-based heterogeneity in this response. Using our response prediction, and an estimate from the literature on the willingness to pay to avoid asthma symptoms, we show that a nationwide 1 µg/m3 reduction in particulate matter concentration would generate nearly $350 million annually in economic benefits.

Impact of a digital health intervention on asthma resource utilization.

Digital health interventions have been associated with reduced rescue inhaler use and improved controller medication adherence. This quality improvement project assessed the benefit of these interventions on asthma-related healthcare utilizations, including hospitalizations, emergency department (ED) utilization and outpatient visits. The intervention consisted of electronic medication monitors (EMMs) that tracked rescue and controller inhaler medication use, and a digital health platform that presented medication use information and asthma control status to patients and providers. In 224 study patients, the number of asthma-related ED visits and combined ED and hospitalization events 365 days pre- to 365 days post-enrollment to the intervention significantly decreased from 11.6 to 5.4 visits (p < 0.05) and 13.4 to 5.8 events (p < 0.05) per 100 patient-years, respectively. This digital health intervention was successfully incorporated into routine clinical practice and was associated with lower rates of asthma-related hospitalizations and ED visits.

Effect of a mobile health, sensor-driven asthma management platform on asthma control.

BACKGROUND: Asthma inflicts a significant health and economic burden in the United States. Self-management approaches to monitoring and treatment can be burdensome for patients. OBJECTIVE: To assess the effect of a digital health management program on asthma outcomes. METHODS: Residents of Louisville, Kentucky, with asthma were enrolled in a single-arm pilot study. Participants received electronic inhaler sensors that tracked the time, frequency, and location of short-acting ß-agonist (SABA) use. After a 30-day baseline period during which reference medication use was recorded by the sensors, participants received access to a digital health intervention designed to enhance self-management. Changes in outcomes, including mean daily SABA use, symptom-free days, and asthma control status, were compared among the initial 30-day baseline period and all subsequent months of the intervention using mixed-model logistic regressions and χ2 tests. RESULTS: The mean number of SABA events per participant per day was 0.44 during the control period and 0.27 after the first month of the intervention, a 39% reduction. The percentage of symptom-free days was 77% during the baseline period and 86% after the first month, a 12% improvement. Improvement was observed throughout the study; each intervention month demonstrated significantly lower SABA use and higher symptom-free days than the baseline month (P < .001). Sixty-nine percent had well-controlled asthma during the baseline period, 67% during the first month of the intervention. Each intervention month demonstrated significantly higher percentages than the baseline month (P < .001), except for month 1 (P = .80). CONCLUSION: A digital health asthma management intervention demonstrated significant reductions in SABA use, increased number of symptom-free days, and improvements in asthma control. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT02162576.

Feasibility of Deploying Inhaler Sensors to Identify the Impacts of Environmental Triggers and Built Environment Factors on Asthma Short-Acting Bronchodilator Use.

BACKGROUND: Epidemiological asthma research has relied upon self-reported symptoms or healthcare utilization data, and used the residential address as the primary location for exposure. These data sources can be temporally limited, spatially aggregated, subjective, and burdensome for the patient to collect. OBJECTIVES: First, we aimed to test the feasibility of collecting rescue inhaler use data in space-time using electronic sensors. Second, we aimed to evaluate whether these data have the potential to identify environmental triggers and built environment factors associated with rescue inhaler use and to determine whether these findings would be consistent with the existing literature. METHODS: We utilized zero-truncated negative binomial models to identify triggers associated with inhaler use, and implemented three sensitivity analyses to validate our findings. RESULTS: Electronic sensors fitted on metered dose inhalers tracked 5,660 rescue inhaler use events in space and time for 140 participants from 13 June 2012 to 28 February 2014. We found that the inhaler sensors were feasible in passively collecting objective rescue inhaler use data. We identified several environmental triggers with a positive and significant association with inhaler use, including: AQI, PM10, weed pollen, and mold. Conversely, the spatial distribution of tree cover demonstrated a negative and significant association with inhaler use. CONCLUSIONS: Utilizing a sensor to capture the signal of rescue inhaler use in space-time offered a passive and objective signal of asthma activity. This approach enabled detailed analyses to identify environmental triggers and built environment factors that are associated with asthma symptoms beyond the residential address. The application of these new technologies has the potential to improve our surveillance and understanding of asthma. Citation: Su JG, Barrett MA, Henderson K, Humblet O, Smith T, Sublett JW, Nesbitt L, Hogg C, Van Sickle D, Sublett JL. 2017. Feasibility of deploying inhaler sensors to identify the impacts of environmental triggers and built environment factors on asthma short-acting bronchodilator use. Environ Health Perspect 125:254-261; http://dx.doi.org/10.1289/EHP266.

Species discovery and validation in a cryptic radiation of endangered primates: coalescent-based species delimitation in Madagascar's mouse lemurs.

Implementation of the coalescent model in a Bayesian framework is an emerging strength in genetically based species delimitation studies. By providing an objective measure of species diagnosis, these methods represent a quantitative enhancement to the analysis of multilocus data, and complement more traditional methods based on phenotypic and ecological characteristics. Recognized as two species 20 years ago, mouse lemurs (genus Microcebus) now comprise more than 20 species, largely diagnosed from mtDNA sequence data. With each new species description, enthusiasm has been tempered with scientific scepticism. Here, we present a statistically justified and unbiased Bayesian approach towards mouse lemur species delimitation. We perform validation tests using multilocus sequence data and two methodologies: (i) reverse-jump Markov chain Monte Carlo sampling to assess the likelihood of different models defined a priori by a guide tree, and (ii) a Bayes factor delimitation test that compares different species-tree models without a guide tree. We assess the sensitivity of these methods using randomized individual assignments, which has been used in bpp studies, but not with Bayes factor delimitation tests. Our results validate previously diagnosed taxa, as well as new species hypotheses, resulting in support for three new mouse lemur species. As the challenge of multiple researchers using differing criteria to describe diversity is not unique to Microcebus, the methods used here have significant potential for clarifying diversity in other taxonomic groups. We echo previous studies in advocating that multiple lines of evidence, including use of the coalescent model, should be trusted to delimit new species.

Need for Enhanced Environmental Representation in the Implementation of One Health.

Issues of global environmental change, global health, emerging disease, and sustainability present some of the most complex challenges of the twenty-first century. Individual disciplines cannot address these issues in isolation. Proactive, innovative, and trans-disciplinary solutions are required. Recognizing the inherent connectedness of humans, animals, plants, and their shared environment, One Health encourages the collaboration of many disciplines-including human and veterinary medicine, public health, social science, public policy, environmental science, and others-to address global and local health challenges. Despite great progress in this shift toward transdisciplinarity, the environmental component of the One Health paradigm remains underrepresented in One Health discourse. Human and animal health issues are commonly discussed under the umbrella of the One Health paradigm, while upstream environmental drivers and solutions are less prominent. We assessed the current integration of environmental issues in One Health publications and leadership. There is room for enhanced integration of environmental knowledge in the implementation of One Health approaches. We discuss the potential benefits from the collaboration between One Health and ecohealth, and explore strategies for increased environmental involvement.