IntroductionReductions in exacerbation and hospitalisations are the outcomes rated as most important by people with chronic obstructive pulmonary disease (COPD). Most COPD management is currently based on a reactive approach, and delays in recognising treatable opportunities underpin COPD care quality gaps. Innovations that empower COPD self-management, facilitate integrated clinical care and support delivery of evidence-based treatment interventions are urgently required.Methods and analysisThe Remote-Management of COPD: Evaluating the Implementation of Digital Innovation to Enable Routine Care trial is a prospective observational cohort hybrid implementation and effectiveness study that will explore the adoption of a digital service model for people with ‘high-risk’ COPD and evaluate the feasibility of this approach versus current standards of care. People with COPD, who have had recent severe exacerbation and/or COPD–obstructive sleep apnoea overlap or chronic hypercapnic respiratory failure requiring home non-invasive ventilation (NIV) or continuous positive airway pressure (CPAP), with internet access will be recruited into the study and enrolled into the digital service.Study endpoints will examine participant utilisation, clinical service impact and clinical outcomes compared with historical and contemporary control patient data. The digital infrastructure will also provide a foundation to explore the feasibility of approaches to predict outcomes and exacerbation in people with COPD through machine learning analysis.Ethics and disseminationEthical approval for this clinical trial has been obtained from the West of Scotland Research Ethics Service. The trial will commence in September 2019 for a duration of 2 years. Results will be presented at local, national and international meetings, as well as submission for publication to peer-reviewed journals.Trial registration numberNCT04240353.

BackgroundOutcomes for patients with chronic obstructive pulmonary disease (COPD) with persistent hypercapnic respiratory failure are improved by long-term home non-invasive ventilation (NIV). Provision of home-NIV presents clinical and service challenges. The aim of this study was to evaluate outcomes of home-NIV in hypercapnic patients with COPD who had been set-up at our centre using remote-monitoring and iVAPS-autoEPAP NIV mode (Lumis device, ResMed).MethodsRetrospective analysis of a data set of 46 patients with COPD who commenced remote-monitored home-NIV (AirView, ResMed) between February 2017 and January 2018. Events including time to readmission or death at 12 months were compared with a retrospectively identified cohort of 27 patients with hypercapnic COPD who had not been referred for consideration of home-NIV.ResultsThe median time to readmission or death was significantly prolonged in patients who commenced home-NIV (median 160 days, 95% CI 69.38 to 250.63) versus the comparison cohort (66 days, 95% CI 21.9 to 110.1; p<0.01). Average time to hospital readmission was 221 days (95% CI, 47.77 to 394.23) and 70 days (95% CI, 55.31 to 84.69; p<0.05), respectively. Median decrease in bicarbonate level of 4.9 mmol/L (p<0.0151) and daytime partial pressure of carbon dioxide 2.2 kPa (p<0.032) in home-NIV patients with no required increase in nurse home visits is compatible with effectiveness of this service model. Median reduction of 14 occupied bed days per annum was observed per patient who continued home-NIV throughout the study period (N=32).ConclusionThese findings demonstrate the feasibility and provide initial utility data for a technology-assisted service model for the provision of home-NIV therapy for patients with COPD.

Chronic obstructive pulmonary disease (COPD) is a global healthcare challenge. It is highly prevalent in low-income countries, causes 3 million death per year and is projected to be the leading cause of death globally by 2030. Challenges in COPD management result in care quality gaps which impair timely and accurate diagnosis and limit patient stratification and provision of evidence-based interventions. COPD exacerbations are responsible for a large proportion of the disease-burden, adverse outcomes and healthcare costs. There is a requirement to re-orientate COPD exacerbation care from failure-driven reactive approach to one based on proactive preventative management. Service model adaptation supported by artificial intelligence (AI) tools offer the prospect of addressing these care-quality gaps and achieving this practice re-orientation. Progress with clinical applications of AI for COPD is accelerating. Evidence available demonstrates the potential of AI techniques to facilitate early and precise COPD case-finding and diagnosis, allow stratification with clinical decision support to prioritise management, and achieve accurate exacerbation detection/prediction to allow proactive interventions. In this narrative review, we will summarise current evidence for the application of AI to these COPD challenges, outline the barriers to implementation of AI models and present our opinion on the required next steps to realise the potential role for AI in COPD management.

No abstract

No abstract