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

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.

The impact of air pollution on asthma: clinical outcomes, current epidemiology, and health disparities

INTRODUCTION

Air pollution has been shown to have a significant impact on morbidity and mortality of respiratory illnesses including asthma.

AREAS COVERED

Outdoor air pollution consists of a mixture of individual pollutants including vehicle traffic and industrial pollution. Studies have implicated an array of individual components of air pollution, with PM2.5, NO2, SO2, and ozone being the most classically described, and newer literature implicating other pollutants such as black carbon and volatile organic compounds. Epidemiological and cohort studies have described incidence and prevalence of pollution-related asthma and investigated both acute and chronic air pollution exposure as they relate to asthma outcomes. There is an increasing body of literature tying disparities in pollution exposure to clinical outcomes. In this narrative review, we assessed the published research investigating the association of pollution with asthma outcomes, focusing on the adult population and health care disparities.

EXPERT OPINION

Pollution has multiple deleterious effects on respiratory health but there is a lack of data on individualized pollution monitoring, making it difficult to establish a temporal relationship between exposure and symptoms, thereby limiting our understanding of safe exposure levels. Future research should focus on more personalized monitoring and treatment plans for mitigating exposure.

Personal exposure to fine particulate matter (PM2.5) and self-reported asthma-related health

PM2.5 (fine particulate matter ≤2.5 μm in diameter) is a key pollutant that can produce acute asthma exacerbations and longer-term deterioration of respiratory health. Individual exposure to PM2.5 is unique and varies across microenvironments. Low-cost sensors (LCS) can collect data at a spatiotemporal resolution previously unattainable, allowing the study of exposures across microenvironments. The aim of this study is to investigate the acute effects of personal exposure to PM2.5 on self-reported asthma-related health. Twenty-eight non-smoking adults with asthma living in Scotland collected PM2.5 personal exposure data using LCS. Measurements were made at a 2-min time resolution for a period of 7 days as participants conducted their typical daily routines. Concurrently, participants were asked to keep a detailed time-activity diary, logging their activities and microenvironments, along with hourly information on their respiratory health and medication use. Health outcomes were modelled as a function of hourly PM2.5 concentration (plus 1- and 2-h lag) using generalized mixed-effects models adjusted for temperature and relative humidity. Personal exposures to PM2.5 varied across microenvironments, with the largest average microenvironmental exposure observed in private residences (11.5 ± 48.6 μg/m3) and lowest in the work microenvironment (2.9 ± 11.3 μg/m3). The most frequently reported asthma symptoms, wheezing, chest tightness and cough, were reported on 3.4%, 1.6% and 1.6% of participant-hours, respectively. The odds of reporting asthma symptoms increased per interquartile range (IQR) in PM2.5 exposure (odds ratio (OR) 1.29, 95% CI 1.07-1.54) for same-hour exposure. Despite this, no association was observed between reliever inhaler use (non-routine, non-exercise related) and PM2.5 exposure (OR 1.02, 95% CI 0.71-1.48). Current air quality monitoring practices are inadequate to detect acute asthma symptom prevalence resulting from PM2.5 exposure; to detect these requires high-resolution air quality data and health information collected in situ. Personal exposure monitoring could have significant implications for asthma self-management and clinical practice.

The joint effect and hemodynamic mechanism of PA and PM2.5 exposure on cognitive function: A randomized controlled trial study

BACKGROUND

While PM2.5 has been shown to impair cognitive function, physical activity (PA) is known to enhance it. Nonetheless, considering the increased inhalation of PM2.5 during exercise, the potential of PA to counteract the detrimental effects of PM2.5, along with the underlying hemodynamic mechanisms, remains uncertain.

METHODS

We conducted a double-blinded, randomized controlled trial among healthy young adults in Beijing, China. Ninety-three participants were randomly allocated to groups experiencing different intensities of PA interventions, and either subjected to purified or unpurified air conditions. Cognitive function was measured by the Color-Word Matching Stroop task, and the hemodynamic response was measured using functional near-infrared spectroscopy during participants performed the Stroop task both before and after the intervention. Linear mixed-effect models were used to estimate the impact of PA and PM2.5 on cognitive function and hemodynamic response.

RESULTS

The reaction time for congruent and incongruent Stroop tasks improved by - 80.714 (95% CI: -136.733, -24.695) and - 105.843 (95% CI: -188.6, -23.085) milliseconds after high-intensity interval training (HIIT) intervention. PM2.5 and HIIT had interaction effects on cognition, such that every 1 μg/m3 increase in PM2.5 attenuated the benefits of HIIT on reaction time by 2.231 (95% CI: 0.523, 3.938) and 3.305 (95% CI: 0.791, 5.819) milliseconds for congruent and incongruent Stroop tasks. Moreover, we divided participants into high and low PM2.5 exposure groups based on average PM2.5 concentration (32.980 μg/m3), and found that HIIT intervention in high PM2.5 concentration led to 69.897 (95% CI: 9.317, 130.476) and 99.269 (95% CI: 10.054, 188.485) milliseconds increased in the reaction time of congruent and incongruent Stroop, compared with the control group among low PM2.5. Furthermore, we found a significant interaction effects of PM2.5 and moderate-intensity continuous training (MICT) on the middle frontal gyrus (MFG) and dorsolateral superior frontal gyrus (DLPFC). PM2.5 and HIIT had a significant interaction effect on the DLPFC.

CONCLUSIONS

HIIT improved cognitive function, but the cognitive benefits of HIIT were attenuated or even reversed under high PM2.5 exposure. The activation of the DLPFC and MFG could serve as hemodynamic mechanisms to explain the joint effect of PA and PM2.5.

Home monitoring in asthma: towards digital twins

PURPOSE OF REVIEW

We highlight the recent advances in home monitoring of patients with asthma, and show that these advances converge towards the implementation of digital twin systems.

RECENT FINDINGS

Connected devices for asthma are increasingly numerous, reliable and effective: new electronic monitoring devices extend to nebulizers and spacers, are able to assess the quality of the inhalation technique, and to identify asthma attack triggers when they include a geolocation function; environmental data can be acquired from databases and refined by wearable air quality sensors; smartwatches are better validated. Connected devices are increasingly integrated into global monitoring systems. At the same time, machine learning techniques open up the possibility of using the large amount of data collected to obtain a holistic assessment of asthma patients, and social robots and virtual assistants can help patients in the daily management of their asthma.

SUMMARY

Advances in the internet of things, machine learning techniques and digital patient support tools for asthma are paving the way for a new era of research on digital twins in asthma.

Air pollution and childhood asthma

PURPOSE OF REVIEW

Asthma is the most common chronic disease of childhood. Environmental exposures, such as allergens and pollutants, are ubiquitous factors associated with asthma development and asthma morbidity. In this review, we highlight the most recent studies relevant to childhood asthma risk, onset, and exacerbation related to air pollution exposure.

RECENT FINDINGS

In this article, we review current research that has been published between 2021 and 2022, demonstrating the effects of early-life exposure to key air pollutants (e.g., particulate matter (PM), nitrogen dioxide (NO 2 ), sulfur dioxide (SO 2 ) and ground-level ozone (O 3 ), environmental tobacco smoke, radon, and volatile organic compounds (VOC) on respiratory health.

SUMMARY

Air pollution continues to be a global burden with serious consequences related to respiratory health. Interventions aimed at reducing air pollution in the environment must be achieved in an effort to improve asthma outcomes and pediatric health.

Reevaluating medication adherence in the era of digital health

INTRODUCTION

Medication adherence is a worldwide issue impacting more than half the population. The cost associated with nonadherence is tremendous and has spurred the growth of novel technologies to address this growing problem.

AREAS COVERED

This perspective covers the different digital health medication adherence tools that have come to market in the past decade and their clinical impact. These digital interventions and their applicability to medication adherence across different stakeholders are then evaluated.

EXPERT OPINION

Digital health will play a significant role in creating new pathways to care in the 2020s. However, the current design of medication adherence tools has not demonstrated a clinical impact that will be relevant for the digital health space without a change in redesign factoring in relevant stakeholders' incentives to address adherence issues. A focus on only adherence has not yielded the economic or clinical benefit as expected, which is likely due to a lack of focus on broader drug-related problems (DRPs) that are causative factors beyond adherence alone. As such, adherence tools will see disparate uptake, likely due to condition-specific interventions rather than adherence issues as a whole, and future endeavors will need to address the larger DRP considerations to actualize clinical outcomes.