Lilholt 2017.

Study characteristics
Methods	Cluster‐RCT (13 clusters per study arm); follow‐up: 12 months; control group: usual care
Participants	Eligible: not reported Randomised: 1125, I: 578, C: 647 Completed: 574, I: 258, C: 316 Mean age: I: 70 years, C: 70 years Sex (% male): I: 48; C: 43 Inclusion criteria: primary diagnosis of COPD based on spirometry, Medical Research Dyspnoea Council Scale (MRC) score ≥ 3 or modified Medical Research Dyspnoea Council Scale ( mMRC ) score ≥2 or COPD Assessment Test score ≥ 10, or ≥ 2 exacerbations during past 12 months; telephone connection; permanent residence; enrolled with participating GP; speaking Danish or living with Danish‐speaking relatives for support in use of telehealthcare system Major exclusions: cognitive impairment; no phone line or GSM coverage; inability to understand Danish to the extent allowing completion of study questionnaires
Interventions	Telehealthcare in addition to standard treatment and care Intervention components ‐ Self‐measurement of blood pressure, pulse, blood oxygen saturation, and weight ‐ Wireless transmission of vital health data to web portal, accessible to patients, relatives, and trained municipality healthcare personnel ‐ Monitoring of vital health data by trained municipality healthcare personnel (i.e. community nurses) based on individually determined threshold values. Monitoring frequency daily (first 2 weeks), once or twice weekly ‐ Contact by healthcare personnel with adverse changes in patient’s vital health values and responses (1‐way communication) ‐ Contact by healthcare personnel if measurements were not carried out as agreed or were not received as expected ‐ Follow‐up visit 3 to 4 weeks to review threshold values and tablet use Duration intervention: 12 months Disciplines involved: GP, healthcare personnel (i.e. community nurse)
Outcomes	Health‐related QoL (SF‐36) (primary outcome); mortality; diastolic blood pressure, systolic blood pressure, pulse, oxygen saturation, and weight; cost‐effectiveness ratio (ICER); cost per QALY
Notes	Dominant component: self‐management (investigator judgement)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: “the municipality districts were matched 1:1 by the following variables: the total population size of the districts, the proportion of people with a higher education, the sum of the district’s total income, unemployment and the estimated number of patients with COPD”; “the districts were distributed randomly by a blinded volunteer with no relation to the trial, who performed the randomisation by throwing a dice"
Allocation concealment (selection bias)	Low risk	Quote: “the identification and recruitment of patients took place prior to random allocation of clusters in order to minimise biased recruitment" Comment: use of sealed envelope method by person not affiliated with the trial
Blinding of participants and personnel (performance bias) All outcomes	High risk	Comment: nature of intervention does not allow blinding. Primary outcome as subjective self‐reported measure likely to have biased outcomes
Blinding of outcome assessment (detection bias) All outcomes	High risk	Comment: primary outcome subjective measure based on patient self‐report. Highly likely that knowledge of study allocation could have biased outcomes
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote: “reasons for withdrawing from the TeleCare North trial included complicated technology, concomitant health problems, not interested, leaving local geographical area, does not trust the equipment or disappointed over not being a part of the telehealth intervention” Comment: large proportion loss to follow‐up. 210/579 for intervention and 177/647 for control. Reason for loss to follow‐up for an intervention related to intervention (n = 101). Attrition rate at 12 months 53%, 110 interventions, 154 controls; patients had incomplete data. Analysis based on imputed data
Selective reporting (reporting bias)	High risk	Quote: “the primary outcome for theme 1 (effectiveness) is the change in health‐related quality of life (SF‐36) at the individual level from baseline to follow‐up at 12 months” (protocol paper)"; “the primary outcome measure was the adjusted mean differences in PCS summary scores between treatment groups at 12 month follow‐up" Comment: change in primary outcome with PCS as a subscore within SF‐36. No reason for change provided. No data on mortality
Recruitment bias	Low risk	Quote: “the identification and recruitment of patients took place prior to random allocation of clusters” Comment: in protocol paper: “the randomisation will not be undertaken until after all general practitioners have sent their lists of patients eligible for inclusion from their practice, and after all patients have given written consent to participation and completed baseline physical measurements and questionnaires”
Baseline imbalance between groups	Low risk	Comment: investigators minimised baseline imbalances through stratification on total population size of districts, proportion of people with a higher education, sum of district’s total income, unemployment and estimated number of patients with COPD. Baseline comparison provided. No large imbalances between groups
Loss to follow‐up of clusters	Low risk	Comment: no clusters lost to follow‐up
Adequate analysis methods for CRT	Low risk	Quote: “the clusters were assumed to be represented as random effects, and the models had robust covariance structures. ICC estimates of patient‐reported outcome variables were calculated for measurement of the variability within and across the clusters. The subgroup analyses applied the same statistical models and covariates as above, but with added treatment‐by‐covariate interaction for each subgroup” Comment: appropriate analysis applied to take clustering into account