A pilot study of daily telemonitoring to predict acute exacerbation in chronic obstructive pulmonary disease

Remote Patient Monitoring and Machine Learning in Acute Exacerbations of Chronic Obstructive Pulmonary Disease: Dual Systematic Literature Review and Narrative Synthesis

BACKGROUND

Acute exacerbations of chronic obstructive pulmonary disease (AECOPD) are associated with high mortality, morbidity, and poor quality of life and constitute a substantial burden to patients and health care systems. New approaches to prevent or reduce the severity of AECOPD are urgently needed. Internationally, this has prompted increased interest in the potential of remote patient monitoring (RPM) and digital medicine. RPM refers to the direct transmission of patient-reported outcomes, physiological, and functional data, including heart rate, weight, blood pressure, oxygen saturation, physical activity, and lung function (spirometry), directly to health care professionals through automation, web-based data entry, or phone-based data entry. Machine learning has the potential to enhance RPM in chronic obstructive pulmonary disease by increasing the accuracy and precision of AECOPD prediction systems.

OBJECTIVE

This study aimed to conduct a dual systematic review. The first review focuses on randomized controlled trials where RPM was used as an intervention to treat or improve AECOPD. The second review examines studies that combined machine learning with RPM to predict AECOPD. We review the evidence and concepts behind RPM and machine learning and discuss the strengths, limitations, and clinical use of available systems. We have generated a list of recommendations needed to deliver patient and health care system benefits.

METHODS

A comprehensive search strategy, encompassing the Scopus and Web of Science databases, was used to identify relevant studies. A total of 2 independent reviewers (HMGG and CM) conducted study selection, data extraction, and quality assessment, with discrepancies resolved through consensus. Data synthesis involved evidence assessment using a Critical Appraisal Skills Programme checklist and a narrative synthesis. Reporting followed PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines.

RESULTS

These narrative syntheses suggest that 57% (16/28) of the randomized controlled trials for RPM interventions fail to achieve the required level of evidence for better outcomes in AECOPD. However, the integration of machine learning into RPM demonstrates promise for increasing the predictive accuracy of AECOPD and, therefore, early intervention.

CONCLUSIONS

This review suggests a transition toward the integration of machine learning into RPM for predicting AECOPD. We discuss particular RPM indices that have the potential to improve AECOPD prediction and highlight research gaps concerning patient factors and the maintained adoption of RPM. Furthermore, we emphasize the importance of a more comprehensive examination of patient and health care burdens associated with RPM, along with the development of practical solutions.

The utility of telemedicine in managing patients after COVID-19

Despite growing knowledge about transmission and relatively wide access to prophylaxis, the world is still facing a severe acute respiratory syndrome coronavirus 2 (SARS CoV 2) global pandemic. Under these circumstances telemedicine emerges as a powerful tool for safe at-home surveillance after a hospital discharge; the data on when to safely release a patient after acute COVID-19 is scarce. Reckoning an urgent need for improving outpatient management and possibly fatal complications of the post-COVID period, we performed the pilot telemonitoring program described below. The study aimed to assess the usefulness of parameters and surveys remotely obtained from COVID-19 convalescents in their individual prognosis prediction. Patients were involved in the study between December 2020 and May 2021. Recruitment was performed either during the hospital discharge (those hospitalized in a Barlicki Memorial Hospital in Lodz) or the first outpatient visit up to 6 weeks after discharge from another center. Every participant received equipment for daily saturation and heart rate measurement coupled with a tablet for remote data transmission. The measurements were made after at least fifteen minutes of rest in a sitting position without oxygen supplementation. Along with the measurements, the cough and dyspnea daily surveys (1-5 points) and Fatigue Assessment Scale weekly surveys were filled. We expected a saturation decrease during thromboembolic events, infectious complications, etc. A total of 30 patients were monitored for a minimum period of 45 days, at least 2 weeks after spontaneous saturation normalization. The mean age was 55 (mean 55.23; SD ± 10.64 years). The group was divided according to clinical improvement defined as the ≥ 10% functional vital capacity (FVC) raise or ≥ 15% lung transfer for carbon monoxide (TL,CO) rise. Our findings suggest that at-rest home saturation measurements below 94% (p = 0.03) correspond with the lack of clinical improvement in post-COVID observation (p = 0.03). The non-improvement group presented with a lower mean-94 (93-96)% versus 96 (95-97)%, p = 0.01 and minimum saturation-89 (86-92)% versus 92 (90-94)%, p = 0.04. They also presented higher variations in saturation measurements; saturation amplitude was 9 (7-11)% versus 7 (4-8)%, p = 0.03; up to day 22 most of the saturation differences reached statistical significance. Last but not least, we discovered that participants missing 2 or more measurements during the observation were more often ranked into the clinical improvement group (p = 0.01). Heart rate day-to-day measurements did not differ between both groups; gathered data about dyspnea and cough intensity did not reach statistical significance either. A better understanding of the disease's natural history will ultimately lead us to a better understanding of long COVID symptoms and corresponding threats. In this paper, we have found home oxygen saturation telemonitoring to be useful in the prediction of the trajectory of the disease course. Our findings suggest that detection of at-rest home saturation measurement equal to or below 94% corresponds with the lack of clinical improvement at the time of observation and this group of patients presented higher variability of day-to-day oxygen saturation measurements. The determination of which patient should be involved in telemedicine programs after discharge requests further research.

Effectiveness of a home telemonitoring program for patients with chronic obstructive pulmonary disease in Germany: Evidence from the first three years

Introduction

Chronic obstructive pulmonary disease (COPD) affects more than 6 million people in Germany. Monitoring the vital parameters of COPD patients remotely through telemonitoring may help doctors and patients prevent and treat acute exacerbations of COPD, improving patients’ quality of life and saving costs for the statutory health insurance system.

Objective

To evaluate the effects from October 2012 until December 2015 of a structured home telemonitoring program implemented by a statutory health insurer in Germany.

Methods

We conducted a retrospective cohort study using administrative data. After building a balanced control group using Entropy Balancing, we calculated difference-in-difference estimators to account for time-invariant heterogeneity. We estimated differences in mortality rates using Cox regression and conducted subgroup and sensitivity analyses to check the robustness of the base case results. We observed each patient in the program for up to 3 years depending on his or her time of enrolment.

Results

Among patients in the telemonitoring cohort, we observed significantly higher inpatient costs due to COPD (€524.2, p<0,05; €434.6, p<0.05) and outpatient costs (102.5, p<0.01; 78.8 p<0.05) during the first two years of the program. Additional cost categories were significantly increased during the first year of telemonitoring. We also observed a significantly higher number of drug prescriptions during all three years of the observation period (2.0500, p < 0.05; 0.7260, p < 0.05; 3.3170, p < 0.01) and a higher number of outpatient contacts during the first two years (0.945, p<0.01, 0.683, p<0.05). Furthermore, we found significantly improved survival rates for participants in the telemonitoring program (HR 0.68, p<0.001).

Conclusion

On one hand, telemonitoring was associated with higher health care expenditures, especially in the first year of the program. For example, we were able to identify a statistically significant increase in inpatient costs due to COPD, outpatient contacts and drug prescriptions among individuals participating in the telemonitoring program. On the other hand, the telemonitoring program was accompanied by a survival benefit, which might be related to higher adherence rates, more intense treatment, or an improved understanding of COPD among these patients.

The Current and Future Role of Technology in Respiratory Care

Over the past few decades, technology and improvements in artificial intelligence have dramatically changed major sectors of our day-to-day lives, including the field of healthcare. E-health includes a wide range of subdomains, such as wearables, smart-inhalers, portable electronic spirometers, digital stethoscopes, and clinical decision support systems. E-health has been consistently shown to enhance the quality of care, improve adherence to therapy, and allow early detection of worsening in chronic pulmonary diseases. The present review addresses the current and potential future role of major e-health tools and approaches in respiratory medicine, with the aim of providing readers with trustful and updated evidence to increase their awareness of the topic, and to allow them to optimally benefit from the latest innovation technology. Collected literature evidence shows that the potential of technology tools in respiratory medicine mainly relies on three fundamental interactions: between clinicians, between clinician and patient, and between patient and health technology. However, it would be desirable to establish widely agreed and adopted standards for conducting trials and reporting results in this area, as well as to take into proper consideration potentially relevant pitfalls related to privacy protection and compliance with regulatory procedures.

Usability of Wearable Multiparameter Technology to Continuously Monitor Free-Living Vital Signs in People Living With Chronic Obstructive Pulmonary Disease: Prospective Observational Study

Background

Vital signs monitoring (VSM) is routine for inpatients, but monitoring during free-living conditions is largely untested in chronic obstructive pulmonary disease (COPD).

Objective

This study investigated the usability and acceptability of continuous VSM for people with COPD using wearable multiparameter technology.

Methods

In total, 50 people following hospitalization for an acute exacerbation of COPD (AECOPD) and 50 people with stable COPD symptoms were asked to wear an Equivital LifeMonitor during waking hours for 6 weeks (42 days). The device recorded heart rate (HR), respiratory rate (RR), skin temperature, and physical activity. Adherence was defined by the number of days the vest was worn and daily wear time. Signal quality was examined, with thresholds of ≥85% for HR and ≥80% for RR, based on the device’s proprietary confidence algorithm. Data quality was calculated as the percentage of wear time with acceptable signal quality. Participant feedback was assessed during follow-up phone calls.

Results

In total, 84% of participants provided data, with average daily wear time of 11.8 (SD 2.2) hours for 32 (SD 11) days (average of study duration 76%, SD 26%). There was greater adherence in the stable group than in the post-AECOPD group (≥5 weeks wear: 71.4% vs 45.7%; P=.02). For all 84 participants, the median HR signal quality was 90% (IQR 80%-94%) and the median RR signal quality was 93% (IQR 92%-95%). The median HR data quality was 81% (IQR 58%-91%), and the median RR data quality was 85% (IQR 77%-91%). Stable group BMI was associated with HR signal quality (r_s=0.45, P=.008) and HR data quality (r_s=0.44, P=.008). For the AECOPD group, RR data quality was associated with waist circumference and BMI (r_s=–0.49, P=.009; r_s=–0.44, P=.02). In total, 36 (74%) participants in the Stable group and 21 (60%) participants in the AECOPD group accepted the technology, but 10 participants (12%) expressed concerns with wearing a device around their chest.

Conclusions

This wearable multiparametric technology showed good user acceptance and was able to measure vital signs in a COPD population. Data quality was generally high but was influenced by body composition. Overall, it was feasible to continuously measure vital signs during free-living conditions in people with COPD symptoms but with additional challenges in the post-AECOPD context.

Machine-learning based feature selection for a non-invasive breathing change detection

BACKGROUND

Chronic Obstructive Pulmonary Disease (COPD) is one of the top 10 causes of death worldwide, representing a major public health problem. Researchers have been looking for new technologies and methods for patient monitoring with the intention of an early identification of acute exacerbation events. Many of these works have been focusing in breathing rate variation, while achieving unsatisfactory sensitivity and/or specificity. This study aims to identify breathing features that better describe respiratory pattern changes in a short-term adjustment of the load-capacity-drive balance, using exercising data.

RESULTS

Under any tested circumstances, breathing rate alone leads to poor capability of classifying rest and effort periods. The best performances were achieved when using Fourier coefficients or when combining breathing rate with the signal amplitude and/or ARIMA coefficients.

CONCLUSIONS

Breathing rate alone is a quite poor feature in terms of prediction of breathing change and the addition of any of the other proposed features improves the classification power. Thus, the combination of features may be considered for enhancing exacerbation prediction methods based in the breathing signal.

TRIAL REGISTRATION

ClinicalTrials NCT03753386. Registered 27 November 2018, https://clinicaltrials.gov/show/NCT03753386.

Day-to-Day Variability of Parameters Recorded by Home Noninvasive Positive Pressure Ventilation for Detection of Severe Acute Exacerbations in COPD

Background

Home noninvasive positive pressure ventilation (NPPV) can be considered not only as an evidence-based treatment for stable hypercapnic chronic obstructive pulmonary disease (COPD) patients, but also as a predictor for detecting severe acute exacerbations of chronic obstructive pulmonary disease (AECOPD).

Methods

In this retrospective observational study, we collected clinical exacerbations information and daily NPPV-related data in a cohort of COPD patients with home NPPV for 6 months. Daily changes in NPPV-related parameters' variability prior to AECOPD were examined using two-way repeated measures ANOVA and individual abnormal values (>75th or <25th percentile of individual baseline parameters) were calculated during 7-day pre-AECOPD period. Multivariate logistic regression was used to identify the independent risk factors associated with AECOPD that then were incorporated into the nomogram.

Results

Between January 1, 2018, and January 1, 2020, a total of 102 patients were included and 31 (30.4%) participants experienced hospitalization (AECOPD group) within 6 months. Respiratory rate changed significantly from baseline at 1, 2 or 3 days prior to admission (p<0.001, respectively) in the AECOPD group. The number of days with abnormal values of daily usage, leaks, or tidal volume during the 7-day pre-AECOPD period in the AECOPD group was higher than in the stable group (p<0.001, respectively). On multivariate analysis, 7-day mean respiratory rate (OR 1.756, 95% CI 1.249-2.469), abnormal values of daily use (OR 1.918, 95% CI 1.253-2.934) and tidal volume (OR 2.081, 95% CI 1.380-3.140) within 7 days were independently associated with the risk of AECOPD. Incorporating these factors, the nomogram achieved good concordance indexes of 0.962.

Conclusion

Seven-day mean respiratory rate, abnormal values of daily usage, leaks, and tidal volume within the 7-day pre-AECOPD period may be biomarkers for detection of AECOPD.

Are COPD self-management mobile applications effective? A systematic review and meta-analysis

The burden of chronic obstructive pulmonary disease (COPD) to patients and health services is steadily increasing. Self-management supported by mobile device applications could improve outcomes for people with COPD. Our aim was to synthesize evidence on the effectiveness of mobile health applications compared with usual care. A systematic review was conducted to identify randomized controlled trials. Outcomes of interest included exacerbations, physical function, and Quality of Life (QoL). Where possible, outcome data were pooled for meta-analyses. Of 1709 citations returned, 13 were eligible trials. Number of exacerbations, quality of life, physical function, dyspnea, physical activity, and self-efficacy were reported. Evidence for effectiveness was inconsistent between studies, and the pooled effect size for physical function and QoL was not significant. There was notable variation in outcome measures used across trials. Developing a standardized outcome-reporting framework for digital health interventions in COPD self-management may help standardize future research.

Automated Home Oxygen Delivery for Patients with COPD and Respiratory Failure: A New Approach

Long-term oxygen therapy (LTOT) has become standard care for the treatment of patients with chronic obstructive pulmonary disease (COPD) and other severe hypoxemic lung diseases. The use of new portable O₂ concentrators (POC) in LTOT is being expanded. However, the issue of oxygen titration is not always properly addressed, since POCs rely on proper use by patients. The robustness of algorithms and the limited reliability of current oximetry sensors are hindering the effectiveness of new approaches to closed-loop POCs based on the feedback of blood oxygen saturation. In this study, a novel intelligent portable oxygen concentrator (iPOC) is described. The presented iPOC is capable of adjusting the O₂ flow automatically by real-time classifying the intensity of a patient’s physical activity (PA). It was designed with a group of patients with COPD and stable chronic respiratory failure. The technical pilot test showed a weighted accuracy of 91.1% in updating the O₂ flow automatically according to medical prescriptions, and a general improvement in oxygenation compared to conventional POCs. In addition, the usability achieved was high, which indicated a significant degree of user satisfaction. This iPOC may have important benefits, including improved oxygenation, increased compliance with therapy recommendations, and the promotion of PA.

Mobile applications in oncology: A systematic review of health science databases

INTRODUCTION

In recent years there has been an exponential growth in the number of mobile applications (apps) relating to the early diagnosis of cancer and prevention of side effects during cancer treatment. For health care professionals and users, it can thus be difficult to determine the most appropriate app for given needs and assess the level of scientific evidence supporting their use. Therefore, this review aims to examine the research studies that deal with this issue and determine the characteristics of the apps involved.

METHODOLOGY

This study involved a systematic review of the scientific literature on randomized clinical trials that use apps to improve cancer management among patients, using the Pubmed (Medline), Latin America and the Caribbean in Health Sciences (LILACS), and Cochrane databases. The search was limited to articles written in English and Spanish published in the last 10 years. A search of the App Store for iOS devices and Google Play for Android devices was performed to find the apps identified in the included research articles.

RESULTS

In total, 54 articles were found to analyze the development of an application in the field of oncology. These articles were most frequently related to the use of apps for the early detection of cancer (n = 28), particularly melanoma (n = 9). In total, 21 studies reflected the application used. The apps featured in nine articles were located using the App Store and Google Play (n = 9), of which five were created to manage cancer-related issues. The rest of the apps were designed for use in the general population (n = 4).

CONCLUSIONS

There is an increasing number of research articles that study the use of apps in the field of oncology; however, these mobile applications tend to disappear from app stores after the studies are completed.

Digital health for COPD care: the current state of play

Chronic obstructive pulmonary disease (COPD) imposes a huge burden to our healthcare systems and societies. To alleviate the burden, digital health-"the use of digital technologies for health"-has been recognized as a potential solution for improving COPD care at scale. The aim of this review is to provide an overview of digital health interventions in COPD care. We accordingly reviewed recent and emerging evidence on digital transformation approaches for COPD care focusing on (I) self-management, (II) in-hospital care, (III) post-discharge care, (IV) hospital-at-home, (V) ambient environment, and (VI) public health surveillance. The emerging approaches included digital-technology-enabled homecare programs, electronic records, big data analytics, and environment-monitoring applications. The digital health approaches of telemonitoring, telehealth and mHealth support the self-management, post-discharge care, and hospital-at-home strategy, with prospective effects on reducing acute COPD exacerbations and hospitalizations. Electronic records and classification tools have been implemented; and their effectiveness needs to be further evaluated in future studies. Air pollution concentrations in the ambient environment are associated with declined lung functions and increased risks for hospitalization and mortality. In all the digital transformation approaches, clinical evidence on reducing mortality, the ultimate goal of digital health intervention, is often inconsistent or insufficient. Digital health transformation provides great opportunities for clinical innovations and discovery of new intervention strategies. Further research remains needed for achieving reliable improvements in clinical outcomes and cost-benefits in future studies.

Wearable Finger Pulse Oximetry for Continuous Oxygen Saturation Measurements During Daily Home Routines of Patients With Chronic Obstructive Pulmonary Disease (COPD) Over One Week: Observational Study

BACKGROUND

Chronic obstructive pulmonary disease (COPD) patients can suffer from low blood oxygen concentrations. Peripheral blood oxygen saturation (SpO2), as assessed by pulse oximetry, is commonly measured during the day using a spot check, or continuously during one or two nights to estimate nocturnal desaturation. Sampling at this frequency may overlook natural fluctuations in SpO2.

OBJECTIVE

This study used wearable finger pulse oximeters to continuously measure SpO2 during daily home routines of COPD patients and assess natural SpO2 fluctuations.

METHODS

A total of 20 COPD patients wore a WristOx2 pulse oximeter for 1 week to collect continuous SpO2 measurements. A SenseWear Armband simultaneously collected actigraphy measurements to provide contextual information. SpO2 time series were preprocessed and data quality was assessed afterward. Mean SpO2, SpO2 SD, and cumulative time spent with SpO2 below 90% (CT90) were calculated for every (1) day, (2) day in rest, and (3) night to assess SpO2 fluctuations.

RESULTS

A high percentage of valid SpO2 data (daytime: 93.27%; nocturnal: 99.31%) could be obtained during a 7-day monitoring period, except during moderate-to-vigorous physical activity (MVPA) (67.86%). Mean nocturnal SpO2 (89.9%, SD 3.4) was lower than mean daytime SpO2 in rest (92.1%, SD 2.9; P<.001). On average, SpO2 in rest ranged over 10.8% (SD 4.4) within one day. Highly varying CT90 values between different nights led to 50% (10/20) of the included patients changing categories between desaturator and nondesaturator over the course of 1 week.

CONCLUSIONS

Continuous SpO2 measurements with wearable finger pulse oximeters identified significant SpO2 fluctuations between and within multiple days and nights of patients with COPD. Continuous SpO2 measurements during daily home routines of patients with COPD generally had high amounts of valid data, except for motion artifacts during MVPA. The identified fluctuations can have implications for telemonitoring applications that are based on daily SpO2 spot checks. CT90 values can vary greatly from night to night in patients with a nocturnal mean SpO2 around 90%, indicating that these patients cannot be consistently categorized as desaturators or nondesaturators. We recommend using wearable sensors for continuous SpO2 measurements over longer time periods to determine the clinical relevance of the identified SpO2 fluctuations.

Smooth Bayesian network model for the prediction of future high-cost patients with COPD

INTRODUCTION

The clinical course of chronic obstructive pulmonary disease (COPD) is marked by acute exacerbation events that increase hospitalization rates and healthcare spending. The early identification of future high-cost patients with COPD may decrease healthcare spending by informing individualized interventions that prevent exacerbation events and decelerate disease progression. Existing studies of cost prediction of other chronic diseases have applied regression and machine-learning methods that cannot capture the complex causal relationships between COPD factors. Thus, the exploration of these factors through nonlinear, high-dimensional but explainable modeling is greatly needed.

OBJECTIVES

We aimed to develop a machine-learning model to identify future high-cost patients with COPD. Such a model should incorporate expert knowledge about causal relationships, and the method for estimating the model could provide more accurate predictions than other machine learning methods.

METHODS

We used the 2011-2013 medical insurance data of patients with COPD in a large city. The data set included demographic information and admission records. Leveraging on developments in graphical modeling methods, we proposed a smooth Bayesian network (SBN) model for the prediction of high-cost individuals using medical insurance data. The modeling method incorporated some expert knowledge about causal relationships (i.e., about the Bayesian network structure). We employed a smoothing kernel based on the weighted nearest neighborhood method in the SBN model to address overfitting, case-mix effect, and data sparsity (i.e., using data about "similar patients").

RESULTS

The proposed SBN achieved the area under curve (AUC) of 0.80 and showed considerable improvement over the baseline machine-learning methods. Besides confirming the known factors from the literature, we found "region" (i.e., a suburban or urban area) to be a significant factor, and that in a 3-tier system with primary, secondary and tertiary hospitals, COPD patients who had been admitted to primary hospitals were more likely to develop into future high-cost patients than patients who had been admitted to tertiary hospitals.

CONCLUSION

The proposed SBN model not only obtained higher prediction accuracy and stronger generalizability than a number of benchmark machine-learning methods, but also used the Bayesian network to capture the complex causal relationships between different predictors by incorporating expert knowledge. Furthermore, a framework was developed to establish the relationships between exposure to historical trajectory and future outcome, which can also be applied to other temporal data to model different trajectory information and predict other outcomes.