Lenferink 2019.

Study characteristics
Methods	RCT, multi‐centre (Netherlands (n = 2), Australia (n = 3)); follow‐up: 12 months; control group: usual care
Participants	Eligible: 226 Randomised: 201, I: 102, C: 99 Completed: 169, I: 85, C: 84 Mean age: I: 69 years, C: 68 years Sex (% male): I: 65, C: 63 Inclusion criteria: diagnosis of COPD (GOLD criteria) with 1 to 5 highly prevalent comorbidities (i.e. ischaemic heart disease (history of myocardial infarction, angina pectoris)); heart failure; diabetes mellitus; active symptoms of anxiety and/or depression (≥ 11 Hospital Anxiety and Depression Scale (HADS) and/or anxiety or depression symptoms treated at the time of inclusion); ≥ 3 COPD exacerbations,defined as respiratory problems that required a course of oral corticosteroids/antibiotics; ≥ 1 hospitalisation for respiratory problems in the 2 years preceding study entry; ≥ 40 years of age Major exclusions: terminal cancer, end stage of COPD or another serious disease with expected survival < 12 months; other serious lung disease (e.g. α1‐antitrypsin deficiency; interstitial lung disease); cognitive impairment (MMSE < 24)
Interventions	Patient‐tailored multi‐disease exacerbation action plan Intervention components ‐ Depending on comorbidities 2 to 3 1‐ to 2‐hour group sessions (1 to 2 hours); 2 times individual hospital‐based self‐management session (1 hour) by trained case manager (respiratory nurse) and supported by cardiac, mental health, and/or diabetes nurses (first month) ‐ Group sessions including knowledge regarding COPD and comorbidities; symptom recognition and monitoring; self‐treatment (action plan linked to diary); breathing and relaxation exercises; extra session on how to check (and regulate) (Dutch patients only); blood glucose levels when necessary (diabetes patients); dietary and lifestyle behaviours ‐ Individual session: individualised action plan set for COPD and each comorbid symptom with colour coding, what are my “usual” symptoms card; diary training; exacerbation action plan training; mastery of skills (e.g. correct inhaler techniques; early recognition of exacerbations, self‐initiating correct and proper actions) ‐ Follow‐up phone calls by case manager to reinforce self‐management skills (Weeks 8, 20, 36) Duration intervention: 9 months Disciplines involved: respiratory nurse (case manager); cardiac, mental health, and/or diabetes nurse
Outcomes	Total number of COPD exacerbation days/patient/year (primary outcome); number of COPD exacerbations/patient/year; duration per COPD exacerbation/patient/year; severity of COPD exacerbation day (symptom diary); FEV₁, FEV₆, FVC; CAT; mMRC; health‐related QoL(EQ‐5D; VAS); Chronic Respiratory Diesease Questionnaire (CRQ); Fatigue (ICFS); Anxiety and Depression (HADS); Confidence and Competence (CSES, CRQ mastery domain); self‐management behaviour and knowledge (PIH); cost and healthcare utilisation; healthcare utilisation for COPD, all‐cause respiratory, cardiac, and diabetes; GP visits; specialist consultations and other services; number of hospitalisations; number of in‐hospital days; travel; costs of usual care; adherence qualitative outcomes
Notes	Dominant component: self‐management with exacerbation plan
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: “after baseline measurements, patients were allocated to self‐management or UC by an independent research assistant who was masked to treatment assignment and randomisation schedule, using a computerised minimisation program. Allocation was stratified per hospital for smoking status, modified Medical Research Council dyspnoea (mMRC) score, number of comorbidities, and being on a waiting list for pulmonary rehabilitation”
Allocation concealment (selection bias)	Low risk	Comment: concealment of allocation after baseline measurement
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: “blinding of patients and personnel to treatment group was not possible. Wherever possible, though, assessors of outcomes were blinded to treatment group”
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: “wherever possible, though, assessors of outcomes were blinded to treatment group" Comment: COPD exacerbation data (primary outcome) collected from symptom diary. Outcome self‐reported, hence outcomes likely to be biased by unb l inding of assessment of primary outcome
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: “if patients have less than three months of complete diary data over the course of the year they will be excluded from the analysis of the daily diaries”; “analyses were conducted on an intention‐to‐treat basis” Comment: missing outcome data were more or less balanced in numbers across intervention (17/102) and control groups (15/99), with similar reasons for missing data across groups
Selective reporting (reporting bias)	Unclear risk	Comment: all primary and secondary outcomes specified in protocol paper have been reported on. In addition, investigators report on analysis of COPD exacerbations and hospitalisations. Data on hospitalisation in protocol collected only for cost‐effectiveness
Other bias	Unclear risk	Comment: timely recognition of exacerbations through symptom diaries and tailored action plan ‐ part of intervention. So can be expected that number of reported exacerbations would be higher in intervention group than in usual care group (missed actual exacerbations). Outcomes may not present actual benefit of intervention