Effectiveness of digital inhaled therapy management and respiratory muscle training in moderate to severe COPD patients: study protocol for a randomized controlled, assessor-blinded, multicenter trial

Abstract

Background

Digital inhaler therapy management can help patients promote adherence and correct inhaler techniques. Poor inhaler technique and suboptimal adherence contribute to COPD exacerbations, uncontrolled symptoms, and a progressive decline in pulmonary function. Digital Inhaled Therapy Management (DITM) platforms can provide standardized technique training, objective adherence monitoring, and real-time feedback. Respiratory Muscle Training is a recommended component of pulmonary rehabilitation that can improve respiratory muscle strength and exercise tolerance. Whether DITM and IMT, alone or in combination, reduce exacerbations and improve patient-reported and physiologic outcomes compared with usual care remains uncertain.

Methods

This is a 1-year, multicenter, open-label, assessor-blinded, randomized, parallel-group superiority trial with participants randomly allocated to four arms (1:1:1:1): DITM, IMT, DITM + IMT, and usual care (verbal inhaler education). Adults ≥ 40 years with COPD per GOLD 2024, FEV1%pred < 60, symptomatic burden (e.g., CAT ≥ 10 and/or mMRC ≥ 2), and high risk of exacerbation (≥ 2 moderate or ≥ 1 severe in past 12 months) will be enrolled (N = 410). Follow-up contacts occur at 3, 6, and 12 months. The primary outcome timepoint is the cumulative 12-month rate of moderate-to-severe COPD exacerbations,with severe events defined as COPD-related hospitalizations, the 3-and 6-month contacts are used for event ascertainment and not as separate primary timepoint. Secondary outcomes include time to first moderate-to-severe exacerbation, CAT, mMRC, SGRQ-C, spirometry, inspiratory muscle strength (MIP/MEP), 6-min walk test, and device-derived metrics of inhaler adherence and technique. Analyses will follow intention-to-treat principles, using negative binomial or Poisson regression for exacerbation rates, Cox models for time-to-event outcomes, and mixed-effects models for repeated continuous outcomes.

Discussion

This trial will evaluate the effectiveness of DITM and IMT—alone and combined—on exacerbations, symptoms, and function in high-risk COPD. Findings may inform integration of digital self-management and inspiratory muscle training into routine COPD care.

Trial registration

ClinicalTrials.gov NCT06958666; registered in July 2025. Ethics approved was obtained from the Ethics Committee of Sir Run Run Shaw Hospital, Zhejiang University School of Medicine (Approval No. 20250119).

Protocol version {2}

Protocol version: V1.0; Date of protocol version: 19 January 2025.

Introduction

Background and rationale {9a}

Chronic obstructive pulmonary disease (COPD) is a major, progressive respiratory disorder characterized by irreversible airflow limitation, chronic respiratory symptoms, and impaired health-related quality of life [1–3]. It is currently the fourth leading cause of death worldwide, accounting for an estimated 3.65 million deaths annually and imposing a substantial socioeconomic burden on healthcare systems and society [4, 5]. Although COPD is incurable, inhaled pharmacological therapy remains the cornerstone of disease management and has been shown to improve clinical outcomes by reducing symptoms, exacerbations, and hospital admissions when used regularly and correctly [6, 7].

However, the real-world effectiveness of inhaled therapy is frequently compromised by poor medication adherence and inhaler technique errors, both of which are strongly associated with worse clinical outcomes, increased risk of exacerbations, and higher healthcare utilization [8, 9]. The wide availability of inhaler devices—exceeding 20 types—with varying operational requirements, including preparation steps, inspiratory flow demands, actuation mechanisms, and maintenance procedures, further increases the complexity of correct inhaler use [10]. Consequently, a substantial proportion of patients misuse inhalers or use them inconsistently, resulting in suboptimal drug delivery, inadequate symptom control, and avoidable disease deterioration.

Digital inhaler therapy management (DITM) has emerged as a promising strategy to address these challenges by enabling objective monitoring of inhaler use, delivering real-time feedback on adherence and technique, and supporting individualized treatment optimization [9, 11–15]. In parallel, respiratory muscle training—particularly inspiratory muscle training (IMT)—has been demonstrated to enhance inspiratory muscle strength, relieve dyspnea, and improve exercise capacity in patients with chronic respiratory disease [16–18]. Theoretically, improved inspiratory muscle function may facilitate more effective inhaler use by enabling patients to generate adequate inspiratory flow, particularly for dry powder inhalers. Despite the potential complementary and synergistic benefits of DITM and IMT, their combined effects on clinically meaningful outcomes in COPD have not been adequately evaluated in randomized controlled trials. Patients with forced expiratory volume in one second (FEV₁) < 60% predicted represent a population with significant symptom burden, frequent exacerbations, and high healthcare utilization, in whom optimized inhaled therapy and supportive interventions may yield the greatest benefit.

Therefore, this assessor-blinded, multicenter randomized controlled trial aims to evaluate the effectiveness of DITM and inspiratory muscle training, alone and in combination, compared with usual care, in reducing exacerbation frequency and improving symptoms, health-related quality of life, and physiological function in patients with COPD and FEV₁ < 60% predicted.

Explanation for the choice of comparator {9b}

Participants allocated to the control group will receive usual care, involving standardized verbal education on proper inhaler technique without digital adherence and technique monitoring or provision of IMT equipment. This comparator reflects current routine clinical practice and serves as the reference for quantifying the incremental benefits of DITM and/or IMT.

Objectives {10}

The primary objective of this study is to determine whether DITM and IMT, administered alone or in combination, reduce the cumulative 12-month rate of moderate-to-severe COPD exacerbations, including severe exacerbations requiring COPD-related hospitalization, compared with usual care in high-risk COPD patients with FEV1 < 60% predicted. Event ascertainment at 3 and 6 months is used to improve completeness of follow-up and will not be analysed as separate primary outcome timepoints. The secondary objective is to assess whether additional clinically relevant outcomes—including time from randomization to first moderate-to-severe exacerbation; changes in COPD Assessment Test (CAT), modified Medical Research Council (mMRC) dyspnea scale, St. George’s Respiratory Questionnaire for COPD (SGRQ-C) scores, spirometry, respiratory muscle strength, 6-min walk distance (6MWD), and device-derived measurements of medication adherence and inhalation technique—are improved in the intervention groups compared with the usual care group.

Methods: patient and public involvement and trial design

Patient and public involvement {11}

Patients were not involved in trial design. Findings will be disseminated to participants and the public via summaries and publications.

Trial design {12}

This multicenter, randomized, open-label, assessor-blinded, parallel-group superiority trial allocates eligible participants 1:1:1:1 to one of four study arms: DITM, IMT, combined DITM with IMT, or usual care. Participants and intervention providers cannot be blinded. Outcome assessors conducting spirometry, MIP/MEP, and 6MWT, central event adjudicators, and the study statistician will remain blinded to group assignment wherever feasible. Patient-reported outcomes (CAT, mMRC, SGRQ-C) will be self-completed using standardized instructions; because participants are unblinded, expectation bias is acknowledged and will be mitigated through standardized administration and blinded analysis.

Methods: participants, interventions, and outcomes

Trial setting {13}

This study will be conducted in approximately 15 academic and community hospitals in Zhejiang Province, China. The trial has been prospectively registered on ClinicalTrials.gov (NCT06958666). Ethics approval from the institutional review board (IRB) has been obtained at each participating centre prior to study initiation.

Characteristics of the trial population

Characteristic	The people we would expect to see included
Age	Adults aged ≥40 years; COPD typically older adults; expect mean ~ 65 years (range variable)
Sex	All sexes; local COPD demographics may skew male; no exclusions
Gender	All gender identities as reported; data captured per local practice
Race, ethnicity and ancestry	Predominantly Han Chinese; ethnic minorities included as available
Socioeconomic status	All; urban hospital catchment; SES recorded if available
Geographic location	Hangzhou, Zhejiang, China; primarily urban/metropolitan
Other characteristics relevant to the trial	High-risk COPD: history of 2 moderate or 1 severe exacerbation in prior 12 months; current MDI/DPI users

Eligibility criteria for participants {14a}

Inclusion criteria:

• Age ≥ 40 years.

• COPD diagnosed per GOLD 2024 (post-bronchodilator FEV1/FVC < 0.60).

• Symptomatic burden: CAT ≥ 10 and/or mMRC ≥ 2.

• High exacerbation risk: ≥ 2 moderate exacerbations or ≥ 1 severe exacerbation requiring hospitalization within the prior 12 months.

• Current use or prescription of MDI or DPI maintenance therapy.

• Owns/uses a smartphone compatible with the study app and can operate it (self or caregiver).

• Provided written informed consent.

Exclusion criteria:

• Other significant respiratory disorders (e.g., asthma, bronchiectasis as primary diagnosis, active TB, interstitial lung disease).

• Acute COPD exacerbation, respiratory infection, or systemic steroid burst within 4 weeks before baseline.

• Unstable cardiovascular disease (e.g., recent MI, uncontrolled arrhythmia), uncontrolled hypertension, or other conditions increasing risk with IMT.

• Severe cognitive impairment, uncontrolled psychiatric illness, or inability to comply with trial procedures.

• Pregnancy or lactation.

• Participation in another interventional clinical trial within 30 days.

• Contraindications to IMT as per device guidance.

Eligibility criteria for sites and those delivering interventions {14b}

Eligible sites include academic or community hospitals with established experience in managing COPD, the capability to perform standardized spirometry and functional assessments, and the necessary infrastructure to support this study. All sites must obtain local institutional review board approval and agree to adhere to the study protocol and data collection procedures. Interventions will be delivered by qualified healthcare professionals (e.g., physicians, respiratory therapists, nurses, or researchers) with experience in COPD care. All intervention providers must complete study-specific training on inhaler technique education, DITM procedures, and IMT protocols, and demonstrate competency before participant enrolment.

Who will take informed consent? {32a}

Written informed consent will be obtained by trained personnel, qualified investigators, or trained research nurses at each participating site prior to any study-specific procedures. Potential participants will be provided with detailed verbal and written information about the study objectives, procedures, potential risks and benefits, data confidentiality, and their right to withdraw at any time without affecting their standard medical care. Adequate time will be allowed for questions and consideration before consent is obtained. For participants with limited literacy or physical impairments, the consent information will be explained verbally in the presence of a caregiver when appropriate.

Additional consent provisions for collection and use of participant data and biological specimens {32b}

No biological specimens will be collected. Consent will cover explicit permission for the collection, storage, and use of de-identified individual data, including data derived from the study application and digital inhaler devices, for research purposes and scientific publication. No identifiable personal information will be disclosed, and all data will be handled in accordance with applicable data protection regulations.

Intervention and comparator

Intervention and comparator description {15a}

All participants will receive standardized inhaler technique training delivered by trained study staff (e.g., respiratory therapists or nurses) at baseline. Training will be conducted using a digital medication administration device that provides on-screen instructional videos and real-time technique assessment. A printed summary report of inhaler technique performance will be provided to participants.

Those allocated to the DITM group will additionally receive a home-based digital inhaler management package, consisting of a sensor compatible with metered-dose inhalers (MDIs) or dry powder inhalers (DPIs) and a companion smartphone application. The sensor records inhaler use data, including time, frequency, and technique-related parameters, which are automatically synchronized to a secure cloud platform. The application analyses adherence and inhaler technique and delivers automated reminders, alerts, and visual feedback summaries to support ongoing self-management throughout the study period.

DITM platform standardisation. The same DITM platform and operating manual will be used at all sites. The platform consists of an inhaler-mounted sensor, a smartphone application, and a secure clinician dashboard. The algorithm records the time/date of each inhaler actuation or inhalation, dosing relative to the prescribed regimen, and technique parameters available for the specific inhaler type, including inspiratory flow profile, inhalation duration, actuation-inhalation coordination for MDIs, dose-preparation steps for DPIs where supported, and breath-hold duration where supported. A scheduled maintenance dose will be classified as on-time when recorded within a ± 2-h window of the prescribed time. Weekly adherence will be calculated as valid scheduled doses divided by prescribed scheduled doses, with ≥ 80% prespecified as adequate adherence. Inhaler technique will be classified using prespecified device-specific thresholds: for MDIs, slow inhalation with actuation-inhalation delay ≤ 1 s where measurable, inhalation duration ≥ 3 s, and breath-hold ≥ 5 s; for DPIs, correct dose preparation, inhalation duration ≥ 1 s, and peak inspiratory flow meeting the device-specific minimum threshold in the study manual (generally ≥ 30 L/min for high-resistance DPIs and ≥ 45–60 L/min for low/medium-resistance DPIs).

DITM feedback logic. The app will provide immediate visual/voice feedback after detected technique errors, missed-dose reminders when no valid dose is recorded within the scheduled window, and weekly adherence/technique summaries. The clinician dashboard will flag participants for follow-up if weekly adherence is < 80%, if ≥ 2 critical technique errors occur within 72 h, or if no data are synchronized for > 7 days. Flagged participants will receive a standardized app message or telephone contact and brief retraining according to the study manual.

Participants assigned to the IMT group will receive standardized verbal education and demonstration of inhaler technique at baseline, delivered by trained study staff. In addition, participants will be provided with an inspiratory muscle training device with companion software, a structured IMT program, and a written user guide. IMT will be performed at home according to the prescribed training plan. Training adherence and performance will be supported and monitored via device software, including automated prompts and usage logs. Participants will be encouraged to follow the prescribed training schedule throughout the intervention period.

IMT protocol standardisation. IMT will be delivered using the same electronic threshold-loading device model across centres. Baseline MIP will be used to prescribe the initial training load, starting at 30% of MIP or the highest tolerated load between 20 and 30% if symptoms occur. Participants will perform 30 resisted inspirations per session, twice daily, at least 5 days per week, for 12 months. The load will be increased by 5% of MIP each week when the participant completes ≥ 80% of prescribed sessions without significant adverse symptoms and reports tolerable exertion; the target training intensity is 50–60% of current MIP by weeks 6–8. MIP will be reassessed at 3, 6, and 12 months, and the training load will be recalibrated. Training will be paused or reduced by 5–10% if dizziness, chest pain, marked dyspnea, or other safety concerns occur. The device software will record session date/time, load, number of breaths, completion rate, and errors.

Participants randomized to the combined intervention group will receive both DITM and IMT, implemented as described above.

Participants in the usual care group will receive standardized verbal education and demonstration of correct inhaler technique at baseline, delivered by trained healthcare professionals. No digital inhaler sensors, mobile applications, or inspiratory muscle training devices will be provided.

All interventions will be delivered according to standardized study manuals and training materials, which are available from the coordinating centre upon reasonable request. Study personnel at each site will complete study-specific training prior to participant enrolment to ensure consistency across centres.

Criteria for discontinuing or modifying allocated intervention/comparator {15b}

The allocated intervention or comparator may be discontinued or modified for an individual participant under the following circumstances:

1. Participant request or withdrawal of informed consent at any time, without penalty or effect on standard medical care.

2. Adverse events related to the intervention, particularly IMT-associated symptoms (e.g., dizziness, chest discomfort, undue dyspnea), for which continuation is deemed inappropriate by the site investigator.

3. Persistent nonadherence to the intervention or follow-up procedures that preclude meaningful participation in the study.

Participants who discontinue the intervention will be encouraged to continue outcome assessments whenever feasible, in accordance with the intention-to-treat principle.

Strategies to improve adherence to intervention/comparator {15c}

Multiple strategies will be implemented to promote adherence across study groups. Participants allocated to groups containing DITM will receive automated reminders, real-time error alerts, and visual feedback on inhaler technique and adherence through the digital platform. During the first week following enrolment, participants involved in DITM groups will undergo an intensified onboarding and training period, during which healthcare professionals provide structured guidance and support. Thereafter, performance feedback will be delivered at least every three days via the digital system. Participants allocated to groups containing IMT will receive app/device prompts, written logs, and monthly check-ins; IMT adherence < 80% of prescribed sessions in any week, repeated incomplete sessions, or device non-synchronization for > 7 days will trigger standardized contact and retraining.

Concomitant care permitted or prohibited during the trial {15d}

Participants in all study arms will continue to receive standard COPD care in accordance with GOLD 2025, including maintenance pharmacologic therapy, rescue medications, vaccinations, and management of comorbidities, as clinically indicated.

The following concomitant interventions will not be permitted during the trial period: (1) participation in structured pulmonary rehabilitation programs, including supervised inspiratory muscle training or respiratory muscle training programs, outside of the study interventions; (2) initiation of non-study inspiratory muscle training devices or similar respiratory training equipment; (3) initiation of other investigational drugs or medical devices that could interfere with study outcomes. Any changes in background COPD pharmacotherapy during the study will be recorded and considered in the analysis.

Ancillary and post-trial care {34}

During the trial, participants will continue to receive their usual clinical care for COPD and comorbid conditions in accordance with standard practice. No additional ancillary care beyond routine medical management is included as part of the study.

Following completion of the trial, no special post-trial access to the digital inhaler management system or inspiratory muscle training devices is planned, unless otherwise provided through routine clinical services.

Compensation and medical care for any harm directly related to trial participation will be provided in accordance with the policies of the participating institutions and applicable local regulations. Participants will be informed of these provisions during the consent process.

Outcomes {16}

The primary outcome will be the cumulative rate of moderate-to-severe COPD exacerbations during the 12 months after randomization, with the primary statistical comparison performed at Month 12. Contacts at Month 3 and Month 6 will be used only for event ascertainment and will not be treated as separate primary outcome timepoints; therefore, no adjustment for multiple primary timepoint comparisons will be required.

Operational definitions: A COPD exacerbation is an acute worsening of respiratory symptoms (dyspnea, cough, and/or sputum) beyond usual day-to-day variation that prompts additional treatment. A moderate exacerbation is defined as an event requiring treatment with systemic corticosteroids and/or antibiotics without hospital admission. A severe exacerbation is defined as an event requiring COPD-related hospitalization. Emergency or outpatient urgent-care visits without admission will be counted as moderate if antibiotics and/or systemic corticosteroids are prescribed, and otherwise recorded as healthcare utilization.

Secondary outcomes will include time to first moderate-to-severe exacerbation; CAT and mMRC scores; SGRQ-C score assessed at 6 and 12 months; spirometric parameters including FEV₁, FVC, FEV₁/FVC, PEF, and MMEF; inspiratory and expiratory muscle strength assessed by MIP and MEP; 6-min walk test distance; device-derived inhaler adherence and technique metrics, where applicable; and safety and tolerability of inspiratory muscle training, including adverse events.

Harms {17}

Anticipated risks are minimal. IMT may cause transient dizziness, fatigue, or chest/abdominal discomfort. DITM involves data privacy considerations. Adverse events will be collected at each contact, graded, assessed for relatedness, and reported per institutional policy. All serious AEs will be reported promptly to the ethics committee.

Participant timeline {18}

Note: Months 3 and 6 list event-review contacts for completeness of exacerbation/SAE ascertainment; the prespecified primary exacerbation endpoint is the cumulative 12-month rate analysed at Month 12.

Procedure/ assessment	Baseline (0)	Month 3	Month 6	Month 12
Eligibility screening	X
Informed consent	X
Randomization	X
Intervention: DITM/IMT/DITM + IMT/usual care	Start	Continue	Continue	Continue
CAT	X	X	X	X
mMRC	X	X	X	X
SGRQ-C	X (optional)		X	X
Spirometry (FEV1, FVC, FEV1/FVC, PEF, MMEF)	X	X	X	X
MIP/MEP	X	X	X	X
6-min walk test	X	X	X	X
Exacerbations/SAEs ascertainment	Continuous	Review	Review	Review

Sample size {19}

A total of 410 participants (approximately 102–103 per arm) provides 80% power at a two-sided α = 0.05 to detect a clinically meaningful difference in exacerbation rates between at least one active arm and usual care, accounting for overdispersion and 10–15% attrition. Assumptions are based on reported exacerbation risks in high-risk COPD populations and expected effect sizes from adherence/technique interventions and IMT; detailed calculations are available in the statistical appendix.

Recruitment {20}

Participants will be recruited from pulmonary clinics at Sir Run Run Shaw Hospital using clinician referral, EHR screening, and study materials (posters/leaflets). Eligibility and consent will be obtained during clinic visits.

Assignment of interventions: randomisation

Sequence generation: who will generate the sequence {21a}

An independent statistician will generate the allocation sequence using a computer-based randomization program.

Sequence generation: type of randomisation {21b}

Restricted randomisation will be used. Participants will be allocated 1:1:1:1 by a central computer-generated permuted-block sequence with randomly varying block sizes, stratified by participating centre and baseline inhaler device category (pMDI vs DPI). The exact block sizes will be stored in the restricted randomisation file and will not be disclosed to enrolling staff or site investigators.

Allocation concealment mechanism {22}

Central, secure web-based randomization (e.g., REDCap randomization module or institutional system) managed by the coordinating centre.

Allocation revealed post-enrolment via role-restricted access; audit trail enabled; sequence inaccessible to enrolling staff prior to assignment.

Implementation {23}

Investigators/staff will enroll participants. The central system will assign interventions according to the concealed sequence. Personnel enrolling participants will not have access to future allocations.

Assignment of interventions: blinding

Who will be blinded {24a}

Participants and intervention providers will not be blinded because the interventions involve visible devices and training. Outcome assessors performing spirometry, MIP/MEP, and 6MWT will be blinded to allocation, will not deliver the interventions, and will not have access to DITM/IMT dashboard data. CAT, mMRC, and SGRQ-C will be self-completed by participants using standardized instructions; these patient-reported outcomes cannot be blinded at the participant level and may be susceptible to expectation bias. Central event adjudicators and the study statistician/data analysts will be blinded to group labels where feasible.

How will be blinding be achieved {24b}

Blinding will be supported by separating intervention delivery staff from outcome assessment staff, scheduling assessments before any intervention-related discussion, instructing participants not to reveal their allocation during visits, and restricting dashboard/device data to intervention staff. Questionnaires will be completed electronically or on paper using neutral written instructions without coaching. The analysis dataset will use masked group labels until the primary analysis is finalized. Any inadvertent disclosure to assessors will be documented.

Procedure for unblinding if needed {24c}

Formal unblinding is not required for participants/providers because the trial is open-label. If a blinded assessor, event adjudicator, or analyst must be unblinded for safety or data-resolution reasons, the reason, date, and person unblinded will be documented in the trial record.

Data collection and management

Plans for assessment and collection of outcomes {25a}

Validated Chinese versions of CAT, mMRC, and SGRQ-C will be used. Spirometry follows ATS/ERS standards and will be performed by blinded assessors.. MIP/MEP will be measured with calibrated devices by blinded assessors, and 6MWT will follow standardized protocols. DITM sensors and app capture objective adherence and technique metrics; IMT devices capture training load, session duration, number of breaths, and completion. Exacerbation and hospitalization data will be collected continuously through monthly participant contact/app reports and verified at 3-, 6-, and 12-month visits by EHR review, prescription records, outpatient/ED notes, and hospital discharge summaries. Events will be adjudicated by a blinded central panel using the operational definitions in Item 16. Data collection forms will include range checks and logic validation.

Plans to promote participant retention and complete follow-up {25b}

Appointment reminders, flexible scheduling, travel support per policy, monthly check-ins, and option for telephone questionnaires when on-site visits are missed. Participants who discontinue an intervention will be asked to continue outcome assessments.

Data management {26}

Electronic data capture with role-based access controls. Data stored on secure, backed-up institutional servers; device/app data transmitted via encrypted channels and integrated to the study database. Data quality monitoring includes double checks, audit trails, and periodic queries.

Confidentiality {33}

Participants will be assigned unique study IDs. Personal identifiers will be stored separately from research data. Access is limited to authorized staff. Publications will use aggregate and deidentified data. Device vendor agreements will ensure data protection compliance.

Statistical methods

Statistical methods for primary and secondary outcomes {27a}

The single primary analysis will compare the cumulative 12-month rate of moderate-to-severe COPD exacerbations between each active arm and usual care using negative binomial regression with log follow-up time as an offset, reporting incidence rate ratios and 95% CIs. Poisson regression with robust variance will be used only if model assumptions support it. The three active-versus-usual-care primary pairwise contrasts (DITM vs usual care, IMT vs usual care, DITM + IMT vs usual care) will be adjusted using the Holm procedure. Month 3 and Month 6 event reviews are interim ascertainment points and will be summarized descriptively; they are not separate primary outcome timepoints, so no Bonferroni correction for multiple primary timepoints will be applied. Models will adjust for randomization strata and prespecified covariates where appropriate, including age, sex, baseline exacerbation history, and baseline CAT. Time to first moderate-to-severe exacerbation will use Kaplan–Meier curves and Cox proportional hazards models. Continuous repeated measures (CAT, mMRC, SGRQ-C, spirometry, 6MWT, MIP/MEP) will be modelled with linear mixed-effects models including fixed effects for group, time, and group × time, adjusting for baseline values. Secondary outcomes will be interpreted as supportive and exploratory unless otherwise prespecified in the SAP.

Who will be included in each analysis {27b}

Primary analyses will follow the intention-to-treat (all randomized). Per-protocol and safety analyses will be conducted as sensitivity analyses.

How missing data will be handled in the analysis {27c}

We will describe missingness patterns. For continuous outcomes, multiple imputation or mixed models under missing at random assumptions will be used; for counts/time-to-event, models inherently handle varying follow-up. Sensitivity analyses will assess missing not at random assumptions where feasible.

Methods for additional analyses (e.g. subgroup analyses) {27d}

Exploratory subgroup analyses by baseline exacerbation risk, baseline CAT severity, and inhaler device type (MDI vs DPI) will include interaction terms.

Interim analyses {28b}

No formal interim efficacy analyses are planned. Safety will be reviewed periodically by the steering committee.

Protocol and statistical analysis plan {5}

This protocol will be published. The statistical analysis plan is integrated herein; a more detailed SAP may be provided as a supplementary file prior to database lock.

Oversight and monitoring

Composition of the coordinating centre and trial steering committee {3d}

The coordinating centre is at Sir Run Run Shaw Hospital (respiratory division). A trial steering committee (PI, co-investigators, statistician, data manager) will meet monthly to oversee recruitment, safety, data quality, and protocol adherence.

Composition of the data monitoring committee, its role and reporting structure {28a}

A formal, independent DMC is not planned due to the minimal-risk nature of the non-pharmacologic interventions and open-label design. Safety oversight will be provided by the trial steering committee and the institutional ethics board. Any serious concerns will prompt escalation to the ethics committee.

Frequency and plans for auditing trial conduct {29}

Institutional clinical research office may audit trial conduct per policy. Internal monitoring will occur periodically focusing on consent, eligibility, and data completeness.

Protocol amendments {31}

Substantive amendments will be submitted to the ethics committee and updated in the trial registry prior to implementation. Relevant staff will be notified, and training updated as needed.

Dissemination policy {8}

Results will be posted on ClinicalTrials.gov, presented at scientific meetings, and submitted to peer-reviewed journals. A plain-language summary will be provided to participants upon request.

Discussion

This randomized controlled trial was designed to evaluate the clinical effectiveness of digital inhaled therapy management (DITM), inspiratory muscle training (IMT), and their combination in patients with chronic obstructive pulmonary disease (COPD) who are at high risk of exacerbations and experience significant symptom burden. Despite advances in pharmacological and non-pharmacological therapies, real-world outcomes for these patients remain suboptimal—largely due to two persistent challenges: poor adherence to inhaled medications coupled with frequent errors in inhaler technique, and progressive respiratory muscle weakness that limits exercise tolerance and quality of life. Both factors independently predict increased rates of acute exacerbations, hospitalizations, and mortality [10–12].

Current international guidelines, including the Global Initiative for Chronic Obstructive Lung Disease (GOLD 2024) [13], emphasize the importance of personalized inhaler education and structured pulmonary rehabilitation. However, implementation gaps are substantial. Traditional face-to-face inhaler instruction is often brief, inconsistent, and lacks objective feedback or longitudinal monitoring [14–17]. Similarly, while pulmonary rehabilitation is a cornerstone of non-pharmacological management, its accessibility remains limited—particularly in resource-constrained settings—due to logistical barriers, lack of trained personnel, and low long-term patient engagement [18, 19].

Digital health technologies offer a promising avenue to bridge these gaps. DITM integrates smart inhaler sensors, mobile applications, and cloud-based analytics to deliver real-time, personalized feedback on inhalation technique (e.g., inspiratory flow rate, breath-hold duration) and automated reminders to improve medication adherence. Early feasibility studies have demonstrated that such systems can significantly reduce inhaler errors and increase adherence rates beyond 80%. Yet, robust evidence from large-scale, pragmatic trials demonstrating impact on clinically meaningful endpoints—particularly moderate-to-severe exacerbations in high-risk populations—remains scarce.

Concurrently, IMT has emerged as a feasible, home-based component of pulmonary rehabilitation. Meta-analyses consistently show that regular IMT improves maximal inspiratory pressure (MIP), reduces dyspnea (as measured by the modified Medical Research Council [mMRC] scale), and enhances functional exercise capacity (e.g., 6-min walk distance). However, its integration into routine care is hindered by challenges in ensuring consistent use, verifying training intensity, and linking it synergistically with pharmacotherapy.

Our trial addresses these limitations through a novel “dual-pathway” intervention model: DITM targets the behavioural and technical dimensions of drug delivery, while IMT addresses the physiological deficit in respiratory muscle strength. By evaluating both interventions individually and in combination, we aim to determine whether their integration yields additive or synergistic benefits in reducing exacerbation frequency, improving symptom control, and enhancing health-related quality of life over a 12-month period.

The study design reflects key principles of implementation science. Although the trial is open-label due to the visible nature of the interventions, objective outcome assessors for spirometry, MIP/MEP, and 6MWT, central event adjudicators, and data analysts will remain blinded wherever feasible. Patient-reported outcomes will be self-completed with standardized neutral instructions before clinical discussion; nevertheless, expectation bias cannot be fully eliminated because participants know their allocation. We therefore emphasize the single 12-month cumulative exacerbation endpoint, blinded event adjudication, objective device-derived adherence/technique metrics, and blinded analysis to reduce detection and measurement bias. The 12-month follow-up duration captures seasonal variation in exacerbation risk and allows assessment of long-term engagement, a critical factor in the sustainability of digital health interventions.

Nonetheless, several challenges warrant acknowledgment. First, digital literacy and technology acceptance may vary, particularly among older adults with COPD. To mitigate this, our platform features a simplified user interface, multilingual voice guidance, optional caregiver access, and dedicated technical support. Second, while IMT is generally safe, transient discomfort (e.g., dizziness, chest tightness) may occur; our protocol includes clear safety stopping rules and adverse event monitoring. Third, generalizability may be influenced by regional differences in healthcare infrastructure, though our multicentre design across diverse clinical settings enhances external validity.

If proven effective, this integrated digital-physiological approach could catalyse a paradigm shift in COPD management—from reactive, clinic-centered care toward proactive, patient-empowered, home-based self-management. Positive findings would provide compelling evidence to support reimbursement policies for DITM devices and IMT equipment under national chronic disease programs. Moreover, the scalable architecture of this intervention holds potential for adaptation to other chronic respiratory conditions requiring long-term inhaled therapy, such as asthma or bronchiectasis.

In conclusion, this trial will address a critical evidence gap regarding the efficacy of DITM and IMT in high-risk COPD and will evaluate a replicable, data-driven model for digital therapeutics in respiratory medicine. By aligning technological innovation with clinical need and implementation feasibility, our work contributes to the broader vision of precision, preventive, and participatory care for chronic diseases—aligning closely with global public health priorities such as Healthy China 2030 and the WHO’s framework on digital health interventions.

Trial status

Recruiting. Protocol version: v1.0; Date of protocol version: 19 January 2025; Recruitment began on: 30 June 2025. Estimated recruitment completion date: 30 June 2026.

Abbreviations

COPD

Chronic obstructive pulmonary disease

DITM

Digital Inhaled Therapy Management

IMT

Inspiratory muscle training

CAT

COPD Assessment Test

mMRC

Modified Medical Research Council

SGRQ-C

St. George’s Respiratory Questionnaire for COPD

6MWT

6-Minute walk test

MIP/MEP

Maximal inspiratory/expiratory pressure

GOLD

Global Initiative for Chronic Obstructive Lung Disease

AE

Adverse event

ITT

Intention-to-treat

DMC

Data monitoring committee

SAP

Statistical analysis plan

Authors’ contributions {3a}

YX and MYZ contributed to study design, trial operations, and protocol drafting. HG is the Principal Investigator and conceived the study and led protocol development. Additional authors served as site coordinators at their respective institutions, contributed to the implementation of the study procedures, and participated in discussions regarding study design and conduct. All authors reviewed and approved the final manuscript. Authorship on future publications will follow ICMJE guidelines.

Funding {7a}

This study received external research funding from Jingyao (Hangzhou) Health Technology Co., Ltd. However, the funder had no role in the study design, conduct, data analysis, interpretation, or manuscript preparation. The investigators retain full academic independence and complete control over all scientific decisions related to this protocol.

Data availability {6}

Deidentified trial data underlying published results may be available from the corresponding author upon reasonable request, subject to approvals and data use agreements. The final trial dataset access will be limited to the study team as per institutional policy.

Declarations

Ethics approval and consent to participate {30}

Ethics Committee of Sir Run Run Shaw Hospital, Zhejiang University School of Medicine; Approval No. 20250119. Written informed consent will be obtained from all participants.

Consent for publication

Not applicable. No individual personal data is included in this protocol. Model consent forms will be provided on request.

Competing interests {7b}

The authors declare no competing interests.