Effectiveness of Pollution Monitoring in Clinical Exercise Rehabilitation

Air pollution is a global public health crisis, contributing to approximately 4.2 million premature deaths annually from stroke, heart disease, lung cancer, and chronic respiratory diseases \[1\]. In the United Kingdom, ambient air pollution is estimated to contribute to 30,000-40,000 premature deaths per year \[2\]. For individuals with pre-existing cardiac and pulmonary conditions, exposure to polluted air can exacerbate symptoms, trigger acute events, and diminish the effectiveness of rehabilitation programmes \[3,4\].

Physical activity confers well-established cardiovascular and respiratory benefits, and exercise-based rehabilitation is a cornerstone of management for patients with chronic cardiac and pulmonary conditions. However, exercising in polluted environments may offset these benefits, particularly in vulnerable populations. Short-term exposure to traffic-related pollution has been shown to attenuate the beneficial cardiopulmonary effects of walking in both healthy individuals and those with ischaemic heart disease and chronic obstructive pulmonary disease (COPD) \[3,4\]. Airborne particulate matter (PM2.5, PM10) and nitrogen dioxide (NO2) induce oxidative stress and systemic inflammation through elevated proinflammatory cytokines including interleukin (IL)-1β, IL-6, IL-23, and tumour necrosis factor alpha (TNF-α) \[5,6\]. This chronic low-grade inflammation increases the risk of cardiovascular disease \[7\], insulin resistance \[8\], and other systemic disorders \[9\].

Despite these well-documented risks, the integration of air quality data into clinical practice remains limited, particularly in the context of cardiac rehabilitation (CR) and pulmonary rehabilitation (PR) programmes. Meta-analytic evidence indicates that a 1.0 μg/m3 reduction in PM2.5 exposure is associated with an approximately 0.5 mmHg reduction in mean blood pressure, and active commuting modes yield a 50% reduction in NO2 and PM2.5 exposures compared with driving \[10\]. These findings suggest that relatively modest behavioural modifications such as adjusting exercise timing, location, or route could meaningfully reduce pollution exposure during rehabilitation.

Mobile health (mHealth) technologies integrated with real-time air pollution data with structured physical activity guidance offer a promising vehicle for delivering personalised pollution exposure information. Therefore this study will evaluate the feasibility and potential effectiveness of this method in a clinical rehabilitation setting.

EPIC-AIR is a stratified, blinded, randomised controlled feasibility trial with two parallel arms conducted at a single site (University Hospitals of Leicester NHS Trust). The study comprises a 1-week enrolment and baseline data collection period followed by a 12-week intervention period. A follow-up assessment is planned at 3 months post-allocation.

Setting Participants will be recruited from cardiac and pulmonary rehabilitation services at University Hospitals of Leicester NHS Trust and from the local community surrounding Leicester, UK.

Eligibility criteria Participants will be recruited in two cohorts. Cohort A comprises patients with long-term conditions; Cohort B comprises healthy volunteers.

Cohort A: Patients with long-term conditions Inclusion criteria for Cohort A are: male or female adults aged ≥18 years; clinical diagnosis of one or more of asthma, COPD, interstitial lung disease (ILD), coronary heart disease (CHD), or heart failure (HF), and signed off for exercise rehabilitation by a clinician; ownership of a GPS-enabled smartphone with internet access; ability to walk outdoors for a minimum of 5 minutes without feeling uneasy or unsteady; availability to complete the 13-week intervention within the recruitment window (February-May 2026); willingness and ability to give informed consent; and willingness to wear a Fitbit device for \>70% of the study duration.

Cohort B: Healthy volunteers Inclusion criteria for Cohort B are identical to Cohort A, with the exception that participants must have no diagnosis of asthma, COPD, ILD, CHD, or HF.

Exclusion criteria (both cohorts) Exclusion criteria for both cohorts are: diagnosis of dementia, learning disability, severe mental health disorders (excluding depression or anxiety), or epilepsy; receiving palliative care; insufficient English language ability to understand study documentation and use the platform; having been advised not to exercise by a healthcare professional within the past 12 months; currently pregnant; presence of chest pain at rest; and marked unsteadiness when standing or walking.

Recruitment Recruitment will commence following receipt of all necessary approvals. Clinical participants (cardiac and pulmonary rehabilitation patients) will be identified through review of rehabilitation programme records at University Hospitals of Leicester NHS Trust by members of the patients' existing clinical care team. Healthy volunteers will be recruited from the local community via email, social media advertisement, and university networks.

The study aims to recruit up to 80 participants to achieve a target of 60 completers (15 cardiac rehabilitation patients, 15 pulmonary rehabilitation patients, and 30 healthy controls), accounting for an anticipated 25% dropout rate.

Informed consent Potential participants will receive a Participant Information Sheet and will be given at least 24 hours to consider participation. Written informed consent will be obtained by suitably qualified and authorised members of the research team prior to any study activities. Participants will be informed of their right to withdraw at any time without prejudice.

Randomisation and blinding Participants will be stratified by long-term condition (cardiac versus pulmonary) and randomised 1:1 using a computer-generated allocation sequence via a secure web-based system (Sealed Envelope Ltd). Randomisation will occur after eligibility confirmation, consent, baseline data collection, and successful app setup.

The study is blinded: participants are unaware of their group allocation, outcome assessors are blinded, and the statistician performing the analysis will use coded group identifiers. Blinding is maintained by ensuring the platform interface appears identical in both arms, with the air quality feature either enabled or disabled according to allocation. The allocation code will be broken at the end of the study following data collection and preliminary screening.

Interventions All participants will receive a Fitbit Charge 6 wearable activity monitor and sign into the platform. Following a 1-week baseline monitoring period, participants will receive progressive walking guidance through the platform for 12 weeks, tailored to individual baseline activity levels, along with motivational messages and reminders via the app.

In addition to these common elements, participants randomised to the intervention arm will receive real-time air quality information (PM2.5, PM10, NO2 levels) through the platform, delivered using GPS-based location data. The app will provide adaptive guidance based on current pollution levels, including suggestions to alter walking routes, adjust the timing of outdoor activity, or reduce exercise intensity during high-pollution periods.

Participants in the control arm will use the platform with identical exercise guidance and tracking features but will not receive air quality data or pollution-related guidance.

Outcomes The primary outcome is personal pollution exposure (PM2.5, PM10, and NO2) during outdoor physical activity, inferred from GPS-tracked walking routes and modelled air quality data via the platform.

Secondary outcomes include physical activity levels, measured by the Recent Physical Activity Questionnaire (RPAQ) and Fitbit-recorded step counts, heart rate, and sleep patterns; health-related quality of life, measured by the SF-12 version 2 (SF-12v2) physical and mental component summary scores; cardiovascular biomarkers, including lipid profile (total cholesterol, HDL, LDL, triglycerides) and inflammatory markers (C-reactive protein and exploratory markers of systemic inflammation), measured from venous blood samples at baseline and 12 weeks; aerobic fitness, measured by the Incremental Shuttle Walk Test (ISWT) at baseline, 12 weeks, and 3 months; exercise self-efficacy, measured by the Jenkins Self-Efficacy for Exercise Scale at baseline, 6 weeks, 12 weeks, and 3 months; clinical measures (blood pressure, height, and weight); and wearable-derived physiological data including heart rate, heart rate variability (RMSSD, LF/HF ratio), SpO2, breathing rate, skin temperature, and on-demand ECG and electrodermal activity (EDA) via Fitbit Charge 6.

Feasibility outcomes include recruitment rate (number screened, eligible, consented, and randomised), retention rate (proportion completing the 13-week study), adherence (Fitbit wear time, app engagement metrics), usability and acceptability of the platform, completeness of outcome data collection, and rates of adverse events and healthcare utilisation (non-routine GP visits, unexpected hospitalisations).

Data collection schedule Data will be collected at five timepoints: enrolment/baseline (-2 weeks), allocation (week 0), and post-allocation at 6 weeks, 12 weeks, and 3 months, with continuous passive data collection via the platform and Fitbit throughout the study period.

Sample handling Venous blood samples (EDTA, 6-8 mL) will be collected at baseline and 12 weeks by trained clinical research nurses at Glenfield Hospital, University Hospitals of Leicester. Samples will be labelled with pseudonymised participant codes and processed within 2 hours of collection. Whole blood will be centrifuged at 1,800 × g for 10 minutes at 4°C. Plasma will be aliquoted (≥2 × 0.5 mL) and stored at -80°C until batch analysis at the University of Leicester Cardiovascular Research Centre (CVRC). Samples with visible haemolysis or inadequate volume will be re-collected where feasible. Analyses will be performed using validated assays in accordance with CVRC standard operating procedures.

Sample size The sample size was informed by a two-sample t-test power calculation based on data from a previous study by our group \[10\], which demonstrated that switching from driving to active travel modes yields approximately 20-25% reductions in pollutant exposure. To achieve \>80% power at a 5% significance level requires ≥22 participants per group reaching the study endpoint. We therefore aim for 30 participants per group (60 total) to complete the study. Accounting for an anticipated 25% dropout rate, the target recruitment is up to 80 participants (20 cardiac rehabilitation, 20 pulmonary rehabilitation, 40 healthy controls).

Statistical analysis All analyses will be conducted in accordance with a pre-specified statistical analysis plan. Baseline characteristics will be summarised using means and standard deviations (or medians and interquartile ranges) for continuous variables and counts with percentages for categorical variables.

Primary analysis: Mean pollution exposure (PM2.5, PM10, NO2) during physical activity will be compared between intervention and control arms using independent-samples t-tests or Mann-Whitney U tests, as appropriate.

Secondary analyses: SF-12v2 physical and mental component summary scores and RPAQ data will be analysed for between-group differences at each timepoint. Changes in cardiovascular biomarkers from baseline to 12 weeks will be compared between groups. Activity trends will be reported graphically over the 13-week period. Healthcare utilisation (GP visits, hospitalisations) will be summarised descriptively.

Feasibility outcomes: Recruitment, eligibility, enrolment, and retention rates will be reported as counts and proportions. App engagement data will be summarised using descriptive statistics.

Missing data: Scoring guidelines specific to each questionnaire will be

• Weekly physical activity (weekly minutes, measured by wrist-worn accelerometry) [Срок оценки: Baseline to 12 weeks]