Validation of ACCESS: an automated tool to support self-management of COPD exacerbations

Effects of Adherence to an mHealth Tool for Self-Management of COPD Exacerbations

Purpose

Poor adherence to COPD mobile health (mHealth) has been reported, but its association with exacerbation-related outcomes is unknown. We explored the effects of mHealth adherence on exacerbation-free weeks and self-management behavior. We also explored differences in self-efficacy and stages of grief between adherent and non-adherent COPD patients.

Patients and Methods

We conducted secondary analyses using data from a recent randomized controlled trial (RCT) that compared the effects of mHealth (intervention) with a paper action plan (comparator) for COPD exacerbation self-management. We used data from the intervention group only to assess differences in exacerbation-free weeks (primary outcome) between patients who were adherent and non-adherent to the mHealth tool. We also assessed differences in the type and timing of self-management actions and scores on self-efficacy and stages of grief (secondary outcomes). We used generalized negative binomial regression analyses with correction for follow-up length to analyze exacerbation-free weeks and multilevel logistic regression analyses with correction for clustering for secondary outcomes.

Results

We included data of 38 patients of whom 13 (34.2%) (mean (SD) age 69.2 (11.2) years) were adherent and 25 (65.8%) (mean (SD) age 68.7 (7.8) years) were non-adherent. Adherent patients did not differ from non-adherent patients in exacerbation-free weeks (mean (SD) 31.5 (14.5) versus 33.5 (10.2); p=0.63). Although statistically not significant, adherent patients increased their bronchodilator use more often and more timely, contacted a healthcare professional and/or initiated prednisolone and/or antibiotics more often, and showed at baseline higher scores of self-efficacy and disease acceptance and lower scores of denial, resistance, and sorrow, compared with non-adherent patients.

Conclusion

Adherence to mHealth may be positively associated with COPD exacerbation self-management behavior, self-efficacy and disease acceptance, but its association with exacerbation-free weeks remains unclear. Our results should be interpreted with caution by this pilot study's explorative nature and small sample size.

Mobile heath applications for self-management in chronic lung disease: a systematic review

Integration of mobile health (mHealth) applications (apps) into chronic lung disease management is becoming increasingly popular. MHealth apps may support adoption of self-management behaviors to assist people in symptoms control and quality of life enhancement. However, mHealth apps' designs, features, and content are inconsistently reported, making it difficult to determine which were the effective components. Therefore, this review aims to summarize the characteristics and features of published mHealth apps for chronic lung diseases. A structured search strategy across five databases (CINAHL, Medline, Embase, Scopus and Cochrane) was performed. Randomized controlled trials investigating interactive mHealth apps in adults with chronic lung disease were included. Screening and full-text reviews were completed by three reviewers using Research Screener and Covidence. Data extraction followed the mHealth Index and Navigation Database (MIND) Evaluation Framework (https://mindapps.org/), a tool designed to help clinicians determine the best mHealth apps to address patients' needs. Over 90,000 articles were screened, with 16 papers included. Fifteen distinct apps were identified, 8 for chronic obstructive pulmonary disease (53%) and 7 for asthma (46%) self-management. Different resources informed app design approaches, accompanied with varying qualities and features across studies. Common reported features included symptom tracking, medication reminders, education, and clinical support. There was insufficient information to answer MIND questions regarding security and privacy, and only five apps had additional publications to support their clinical foundation. Current studies reported designs and features of self-management apps differently. These app design variations create challenges in determining their effectiveness and suitability for chronic lung disease self-management. Registration: PROSPERO (CRD42021260205).

Supplementary Information

The online version contains supplementary material available at 10.1007/s13721-023-00419-0.

Artificial intelligence to differentiate asthma from COPD in medico-administrative databases

INTRODUCTION

Discriminating asthma from chronic obstructive pulmonary disease (COPD) using medico-administrative databases is challenging but necessary for medico-economic analyses focusing on respiratory diseases. Artificial intelligence (AI) may improve dedicated algorithms.

OBJECTIVES

To assess performance of different AI-based approaches to distinguish asthmatics from COPD patients in medico-administrative databases where the clinical diagnosis is absent. An "Asthma COPD Overlap" category was defined to further test whether AI can detect complexity.

METHODS

This study included 178,962 patients treated by two "R03" treatment prescriptions at least from January 2016 to December 2018 and managed by either a general practitioner and/or a pulmonologist participating in a permanent longitudinal observatory of prescription in ambulatory medicine (LPD). Clinical diagnoses are available in this database and were used as gold standards to develop diagnostic rules. Three types of AI approaches were explored using data restricted to demographics and treatment dispensations: multinomial regression, gradient boosting and recurrent neural networks (RNN). The best performing model (based on metric properties) was then applied to estimate the size of asthma and COPD populations based on a database (LRx) of treatment dispensations between July, 2018 and June, 2019.

RESULTS

The best models were obtained with the boosting approach and RNN, with an overall accuracy of 68%. Performance metrics were better for asthma than COPD. Based on LRx data, the extrapolated numbers of patients treated for asthma and COPD in France were 3.7 and 1.2 million, respectively. Asthma patients were younger than COPD patients (mean, 49.9 vs. 72.1 years); COPD occurred mostly in men (68%) compared to asthma (33%).

CONCLUSION

AI can provide models with acceptable accuracy to distinguish between asthma, ACO and COPD in medico-administrative databases where the clinical diagnosis is absent. Deep learning and machine learning (RNN) had similar performances in this regard.

Adherence to an eHealth Self-Management Intervention for Patients with Both COPD and Heart Failure: Results of a Pilot Study

Background

Chronic obstructive pulmonary disease (COPD) and chronic heart failure (CHF) often coexist and share periods of symptom deterioration. Electronic health (eHealth) might play an important role in adherence to interventions for the self-management of COPD and CHF symptoms by facilitating and supporting home-based care.

Methods

In this pilot study, an eHealth self-management intervention was developed based on paper versions of multi-morbid exacerbation action plans and evaluated in patients with both COPD and CHF. Self-reporting of increased symptoms in diaries was linked to an automated decision support system that generated self-management actions, which was communicated via an eHealth application on a tablet. After participating in self-management training sessions, patients used the intervention for a maximum of four months. Adherence to daily symptom diary completion and follow-up of actions were analyzed. An add-on sensorized (Respiro^®) inhaler was used to analyze inhaled medication adherence and inhalation technique.

Results

In total, 1148 (91%) of the daily diaries were completed on the same day by 11 participating patients (mean age 66.8 ± 2.9 years; moderate (55%) to severe (45%) COPD; 46% midrange left ventricular function (LVF) and 27% reduced LVF). Seven patients received a total of 24 advised actions because of increased symptoms of which 11 (46%) were followed-up. Of the 13 (54%) unperformed advised actions, six were “call the case manager”. Adherence to inhaled medication was 98.4%, but 51.9% of inhalations were performed incorrectly, with “inhaling too shortly” (<1.25 s) being the most frequent error (79.6%).

Discussion

Whereas adherence to completing daily diaries was high, advised actions were inadequately followed-up, particularly the action “call the case manager”. Inhaled medication adherence was high, but inhalations were poorly performed. Future research is needed to identify adherence barriers, further tailor the intervention to the individual patient and analyse the intervention effects on health outcomes.

Validation of COPDPredict™: Unique Combination of Remote Monitoring and Exacerbation Prediction to Support Preventative Management of COPD Exacerbations

Background

COPDPredict™ is a novel digital application dedicated to providing early warning of imminent COPD (chronic obstructive pulmonary disease) exacerbations for prompt intervention. Exacerbation prediction algorithms are based on a decision tree model constructed from percentage thresholds for disease state changes in patient-reported wellbeing, forced expiratory volume in one second (FEV₁) and C-reactive protein (CRP) levels. Our study determined the validity of COPDPredict™ to identify exacerbations and provide timely notifications to patients and clinicians compared to clinician-defined episodes.

Methods

In a 6-month prospective observational study, 90 patients with COPD and frequent exacerbations registered wellbeing self-assessments daily using COPDPredict™ App and measured FEV₁ using connected spirometers. CRP was measured using finger-prick testing.

Results

Wellbeing self-assessment submissions showed 98% compliance. Ten patients did not experience exacerbations and treatment was unchanged. A total of 112 clinician-defined exacerbations were identified in the remaining 80 patients: 52 experienced 1 exacerbation; 28 had 2.2±0.4 episodes. Sixty-two patients self-managed using prescribed rescue medication. In 14 patients, exacerbations were more severe but responded to timely escalated treatment at home. Four patients attended the emergency room; with 2 hospitalised for <72 hours. Compared to the 6 months pre-COPDPredict™, hospitalisations were reduced by 98% (90 vs 2, p<0.001). COPDPredict™ identified COPD-related exacerbations at 7, 3 days (median, IQR) prior to clinician-defined episodes, sending appropriate alerts to patients and clinicians. Cross-tabulation demonstrated sensitivity of 97.9% (95% CI 95.7-99.2), specificity of 84.0% (95% CI 82.6-85.3), positive and negative predictive value of 38.4% (95% CI 36.4-40.4) and 99.8% (95% CI 99.5-99.9), respectively.

Conclusion

High sensitivity indicates that if there is an exacerbation, COPDPredict™ informs patients and clinicians accurately. The high negative predictive value implies that when an exacerbation is not indicated by COPDPredict™, risk of an exacerbation is low. Thus, COPDPredict™ provides safe, personalised, preventative care for patients with COPD.

Machine learning for nocturnal diagnosis of chronic obstructive pulmonary disease using digital oximetry biomarkers

Objective.Chronic obstructive pulmonary disease (COPD) is a highly prevalent chronic condition. COPD is a major cause of morbidity, mortality and healthcare costs globally. Spirometry is the gold standard test for a definitive diagnosis and severity grading of COPD. However, a large proportion of individuals with COPD are undiagnosed and untreated. Given the high prevalence of COPD and its clinical importance, it is critical to develop new algorithms to identify undiagnosed COPD. This is particularly true in specific disease groups in which the presence of concomitant COPD increases overall morbidity/mortality such as those with sleep-disordered breathing. To our knowledge, no research has looked at the feasibility of automated COPD diagnosis using a data-driven analysis of the nocturnal continuous oximetry time series. We hypothesize that patients with COPD will exert certain patterns and/or dynamics of their overnight oximetry time series that are unique to this condition and that may be captured using a data-driven approach.Approach.We introduce a novel approach to nocturnal COPD diagnosis using 44 oximetry digital biomarkers and five demographic features and assess its performance in a population sample at risk of sleep-disordered breathing. A total ofn=350 unique patients' polysomnography (PSG) recordings were used. A random forest (RF) classifier was trained using these features and evaluated using nested cross-validation.Main results.The RF classifier obtainedF1 = 0.86 ± 0.02 and AUROC = 0.93 ± 0.02 on the test sets. A total of 8 COPD individuals out of 70 were misclassified. No severe cases (GOLD 3-4) were misdiagnosed. Including additional non-oximetry derived PSG biomarkers resulted in minimal performance increase.Significance.We demonstrated for the first time, the feasibility of COPD diagnosis from nocturnal oximetry time series for a population sample at risk of sleep-disordered breathing. We also highlighted what set of digital oximetry biomarkers best reflect how COPD manifests overnight. The results motivate that overnight single channel oximetry can be a valuable modality for COPD diagnosis, in a population sample at risk of sleep-disordered breathing. Further data is needed to validate this approach on other population samples.

A Smart Mobile Health Tool Versus a Paper Action Plan to Support Self-Management of Chronic Obstructive Pulmonary Disease Exacerbations: Randomized Controlled Trial

Background

Many patients with chronic obstructive pulmonary disease (COPD) suffer from exacerbations, a worsening of their respiratory symptoms that warrants medical treatment. Exacerbations are often poorly recognized or managed by patients, leading to increased disease burden and health care costs.

Objective

This study aimed to examine the effects of a smart mobile health (mHealth) tool that supports COPD patients in the self-management of exacerbations by providing predictions of early exacerbation onset and timely treatment advice without the interference of health care professionals.

Methods

In a multicenter, 2-arm randomized controlled trial with 12-months follow-up, patients with COPD used the smart mHealth tool (intervention group) or a paper action plan (control group) when they experienced worsening of respiratory symptoms. For our primary outcome exacerbation-free time, expressed as weeks without exacerbation, we used an automated telephone questionnaire system to measure weekly respiratory symptoms and treatment actions. Secondary outcomes were health status, self-efficacy, self-management behavior, health care utilization, and usability. For our analyses, we used negative binomial regression, multilevel logistic regression, and generalized estimating equation regression models.

Results

Of the 87 patients with COPD recruited from primary and secondary care centers, 43 were randomized to the intervention group. We found no statistically significant differences between the intervention group and the control group in exacerbation-free weeks (mean 30.6, SD 13.3 vs mean 28.0, SD 14.8 weeks, respectively; rate ratio 1.21; 95% CI 0.77-1.91) or in health status, self-efficacy, self-management behavior, and health care utilization. Patients using the mHealth tool valued it as a more supportive tool than patients using the paper action plan. Patients considered the usability of the mHealth tool as good.

Conclusions

This study did not show beneficial effects of a smart mHealth tool on exacerbation-free time, health status, self-efficacy, self-management behavior, and health care utilization in patients with COPD compared with the use of a paper action plan. Participants were positive about the supportive function and the usability of the mHealth tool. mHealth may be a valuable alternative for COPD patients who prefer a digital tool instead of a paper action plan.

Trial Registration

ClinicalTrials.gov NCT02553096; https://clinicaltrials.gov/ct2/show/NCT02553096.

eHealth for people with COPD in the Netherlands: a scoping review

Background: In the Netherlands, almost 600,000 people had chronic obstructive pulmonary disease (COPD) in 2017. This decreases quality of life for many and each year, COPD leads to approximately 6,800 deaths and about one billion health care expenditures. It is expected that eHealth may improve access to care and reduce costs. However, there is no conclusive scientific evidence available of the added value of eHealth in COPD care. We conducted a scoping review into the use of eHealth in Dutch COPD care. The aim of the research was to provide an overview of all eHealth applications used in Dutch COPD care and to assess these applications on a number of relevant criteria. Methods: In order to make an overview of all eHealth applications aimed at COPD patients in the Netherlands, literature was searched in the electronic databases PubMed and Google Scholar. In addition, Dutch health care websites were searched for applications that have been evaluated for effectiveness and reliability. The identified eHealth applications were assessed according to five relevant quality criteria, eg, whether research has been conducted on the effectiveness. Results: Thirteen health care programs and patient platforms in COPD care have been found that use eHealth. In addition, 13 self-care and informative websites and 15 mobile apps were found that are available to citizens and patients. Five of 13 care programs and patient platforms were found to be effective in improving quality of life or reducing hospital admissions in small pilot studies. The effectiveness of these and the other eHealth applications should be established in larger studies in the future. Discussion: More research into the effectiveness of eHealth applications for COPD patients is needed. We recommend to develop a nationwide open source platform where well-evaluated eHealth applications can be showcased for patients and health care providers to improve COPD care.