Bernocchi 2017.

Study characteristics
Methods	RCT; follow‐up: 6 months; control: usual care, multi‐centre (n = 3)
Participants	Eligible: 319 Randomised: 112, I: 56, C: 56 Completed: 80, I: 45, C:35 Mean age: I: 71 years, C: 70 years Sex (% male): I: 88, C: 75 Inclusion criteria: older patients with COPD and cardiovascular heart disease: COPD new GOLD classification (B, C, and D classes) and spirometry in the previous year and systolic and/or diastolic CHF defined at least by an echocardiogram performed in clinical stability; II, III, and IV NYHA class and optimised drug therapy Major exclusions: physical activity limitations caused by non‐cardiac and/or pulmonary problems; obstructive cardiomyopathies and/or myocarditis; non‐cardiac and/or pulmonary pathologies that would cause the death of the patient during the study; poor adherence and compliance of the patient
Interventions	Home‐based telehealth and rehabilitation programme Intervention components ‐ Scheduled calls initiated by nurse (weekly) ‐ Unscheduled calls initiated by patients or caregivers through the service centre (24 hours/24 hours) to report any clinical problems in case of signs or symptoms ‐ Telemonitoring: during calls, patients can transmit via landline or mobile phone recordings from the 1‐lead ECG to a service centre, and talk to the nurse or doctor ‐ Home visit performed by therapist 7 days after hospital discharge by setting the daily physical activity and other home visits in case of need Home‐based rehabilitation programme ‐ Individual rehabilitative programme including ≥ 3 sessions/week of mini‐ergometer and exercises and 2 sessions/week of walking with pedometer ‐ Scheduled calls initiated by therapist performed weekly aimed at increasing workload and evaluating proper execution of exercises Duration intervention: 12 weeks Disciplines involved: nurse, therapist, treating physician
Outcomes	6MWD, mMRC, PASE score, Barthel score, CAT score, number of hospitalisations total, hospitalisation ‐ respiratory‐related, mortality
Notes	Dominant component: telerehabilitation
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: “a computer generated tables to allocate patients in fixed blocks of 4” (Bernocchi 2016)
Allocation concealment (selection bias)	Low risk	Quote: “in order to prevent selection bias, the allocation sequence was concealed from the investigators enrolling and assessing patients, in sequentially numbered, opaque, sealed envelopes" (Bernocchi 2016)
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: “due to the nature of the intervention, neither the patients nor the physicians were blinded to patients’ group allocation” Comment: primary outcome functional exercise capacity, likely to be biased by difference in performance
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Comment: outcome assessors and data analysts were blinded
Incomplete outcome data (attrition bias) All outcomes	High risk	Comment: missing outcome data were greater in control group (21/56) compared to intervention group (11/56). More loss to follow‐up in control group due to hospitalisation caused by heart failure. Loss to follow‐up likely to be related to intervention
Selective reporting (reporting bias)	Low risk	Comment: a secondary outcome in protocol defined as (1) reduction of hospitalisations for cardiovascular and/or respiratory disease; and (2) reduction of hospitalisations for all causes, in outcome paper reported as reduction to time‐to‐event, combining hospitalisation and mortality. Reasons for change in outcome paper not provided. Explanation sought and provided by study authors: "incidence of death was very low and we considered that the inclusion of the two events (hospitalisations and deaths) best described the effect of the treatment" Also feasibility measure adherence to ≥ 70% proposal rehabilitative sessions not reported; instead crude outcome reported. Unlikely to have biased outcomes