Objective: Asthma is a chronic pulmonary disease with multiple triggers manifesting as symptoms with various intensities. This paper evaluates the suitability of long-term monitoring of pediatric asthma using diverse data to qualify and quantify triggers that contribute to the asthma symptoms and control to enable a personalized management plan. Materials and Methods: Asthma condition, environment, and adherence to the prescribed care plan were continuously tracked for 97 pediatric patients using kHealth-Asthma technology for one or three months. Result: At the cohort level, among 21% of the patients deployed in spring, 63% and 19% indicated pollen and Particulate Matter (PM2.5), respectively, as the major asthma contributors. Of the 18% of the patients deployed in fall, 29% and 21% found pollen and PM2.5 respectively, to be the contributors. For the 28% of the patients deployed in winter, PM2.5 was identified as the major contributor for 80% of them. One patient across each season has been chosen to explain the determination of personalized triggers by observing correlations between triggers and asthma symptoms gathered from anecdotal evidence. Discussion and Conclusion: Both public and personal health signals including compliance to prescribed care plan have been captured through continuous monitoring using the kHealth-Asthma technology which generated insights on causes of asthma symptoms across different seasons. Collectively, they can form the underlying basis for personalized management plan and intervention. KEYWORDS: Personalized Digital Health, Medical Internet of Things, Pediatric Asthma Management, Patient Generated Health Data, Personalized Triggers, Telehealth,
Asthma is a chronic lung inflammatory disease affecting 26 million people in the USA, of which 6 million are children [1]. It is a multifactorial disease with different triggers manifesting as asthma symptoms of various intensities which demands a personalized diagnosis and management plan [2]. Infrequent clinical visits are unable to provide timely feedback and intervention as most of the asthma-exacerbating factors are pollutants from the patient's environment [3] and lack of medication adherence [4]. Continuous tracking and assessment of a patient's condition, environment, and adherence to a prescribed care plan can improve asthma control and quality of life [5].
While many studies have shown the effectiveness of continuous monitoring, only a few are being evaluated to benefit traditional healthcare practices [6,7]. Propeller Health [8] provides personalized alerts based on inhaler usage and location to primarily improve medication adherence. ENVIROFI [9] and azma.com [10] send a notification to subscribed users when the outdoor environment forecast is poor. Chu et al [11] developed a ubiquitous warning system which sends alerts to healthcare providers based on a patient’s location if the outdoor environment is poor. Finkelstein et al [12] developed a web-based approach that captures Forced Vital Capacity test and asthma symptoms from patients and sends alert to hospitals when these parameters are abnormal. AsthmaGuide [13], a home management ecosystem, enables doctors to observe the correlation between symptoms and environmental data. They have classified wheezing sounds as asthmatic wheezing and non-asthmatic wheezing. They also send personalized alerts to patients based on pollen and air quality forecast, but no causal relationships are identified. However, while a number of factors has been shown to influence triggers and control level for an individual asthma patient, monitoring and analyzing diverse data directly relevant to an individual patient has not been adequately evaluated.
Using the kHealth-Asthma technology [14], we monitored and collected diverse data for a cohort of pediatric patients receiving asthma care at the Dayton Children’s Hospital (DCH). This paper presents cohort level preliminary data analysis for patients deployed in each of the seasons to identify the major contributors to asthma symptoms. In addition to that, one patient was chosen from each season to illustrate the personalized trigger determination by gathering anecdotal evidence.
kHealth is a framework to personalize digital health by collecting and analyzing multimodal data that complement data collected during routine clinical care, specifically Patient Generated Health data using mobile app and sensors as well environmental data. It is designed to assist self-monitoring and self-appraisal of asthma care with an intent to incorporate self-management, prediction, and intervention of asthma progression [15]. kHealth-Asthma is the adaptation of kHealth for asthma, and comprises of three components: kHealth kit, kHealth cloud, and kHealth Dashboard. The study design including these components, their use for data collection and the data analysis of the 97 patient cohort are discussed next. Other applications for which kHealth has been adapted include post-bariatric surgery monitoring, post-surgery monitoring of Acute Decompensated Heart Failure (ADHF), and dementia.
The kHealth kit components are shown in Figure 1. The questionnaire presented by the mobile application on the tablet collects the following data: (i) six types of symptoms: cough, wheeze, chest tightness, hard and fast breathing, can’t talk in full sentences, and nose opens wide [16], (ii) medication intake (rescue inhaler and controller medication) with yes or no option, (iii) night-time awakenings due to asthma symptoms, and (iv) activity limitation due to asthma symptoms. The symptoms and medications are collected twice a day, and night-time awakenings and activity limitation are collected once a day (details in Appendix 1). Furthermore, Fitbit is used to collect more granular data for sleep and activity [17]. The lung function measurements (PEF and FEV1) are recorded by the Microlife peak flow meter [18] twice every day. For a given patient’s zip code, outdoor environmental parameters are collected at different intervals –pollen is collected every 12 hours, Particulate Matter (PM2.5), ozone, temperature, and humidity are collected every hour. Pollen is collected from pollen.com [19], PM2.5 and ozone from EPA AIRNow [20], and temperature and humidity from Weather Underground [21]. Foobot collects indoor temperature, humidity, particulate matter, volatile compounds, carbon dioxide, and global pollution index every 5 minutes.
Available literature have evaluated quality and suitability of Fitbit [22,23]. The feasibility study for Foobot has been conducted on our own [24]. kHealth Cloud The multimodal data collected from various sources are brought togethe
This content is AI-processed based on open access ArXiv data.
