Table 1.

Characteristics of studies with available relevant ICC estimates included in Bayesian modelling.

Study	Source	Intervention	Study population	Outcome	ICC estimate	Number of patients	Number of clusters
1	Thomas et al. 14	Systematic voiding programme.	Stroke patients with incontinence (UK).	Absence of incontinence at 12 weeks post-stroke.	0.00	413	12
2	Tannenbaum et al. 15	Three experimental continence interventions: (1) continence education; (2) evidence-based self-management; (3) combined continence education and self-management.	Women aged 60 years and older with untreated incontinence (UK).	Patient’s global impression of improvement in continence questionnaire (PGI-I) measured at 3 months post-intervention.	0.05	259	71
3	Sackley et al. 16	Staff education on continence care and mobility care and mobility training.	UK care home residents.	Rivermead Mobility Index at baseline and 6 weeks post-intervention.	0.37	34	6
4	Sackley et al. 17	Occupational therapy provided to individuals and carer education.	UK care homes residents with moderate to severe stroke-related disability (Barthel Index (BI) score 4–15) except those with acute illness and those admitted for end-of-life care.	BI score at baseline.	0.26	173	12
				BI change to 3 months.	0.18
				BI change to 6 months.	0.2
				Global poor outcome at 3 months.	0.14
				Global poor outcome at 6 months.	0.09
5	Weir et al. 18	Computer-based decision support system to aid selection of long-term antithrombotic therapy.	UK hospital in-patients or out-patients with a clinical diagnosis of acute ischaemic stroke or TIA; first investigation of an event occurring within preceding 4 months.	Change in relative risk of ischaemic and haemorrhagic vascular events relative to the option of ‘no antiplatelet or anticoagulant therapy’.	0.15	1952	16
6	De Luca et al. 19	The intervention group staff (physicians, nurses and drivers) training on and delivery of evidence-based prehospital emergency clinical pathway based on experiential learning approach. The training was focused on teaching the personnel to identify stroke symptoms.	People living in the community aged <80 years (Italy, acute care/ community).	The proportion of eligible acute stroke patients correctly referred to stroke unit.	0.05	4895	20
7	Dirks et al. 20	Intervention to increase thrombolysis rates by creating local stroke teams, identifying barriers to thrombolysis delivery, setting goals and planned actions, and updating acute stroke guidelines.	Patients ≥18 years with acute stroke who were admitted to the hospital within 24 h from onset of symptoms (Netherlands, acute care/community).	Treatment with rtPA (recombinant tissue Plasminogen Activator).	0.0154	5515	12
8	Johnston et al. 21	Quality improvement in ischaemic stroke discharge orders comprising statin prescription; antihypertensive medication for those with hypertension; warfarin for all patients with atrial fibrillation (AF), except those with contraindication.	At least 40 years old, were the Kaiser Permanente Medical Care Plan (KPMCP) members with pharmacy benefits, and had been discharged alive to home or to a facility other than hospice (USA hospitals).	Composite binary variable comprising optimal treatment via all of: (1) documentation of filled statin prescription 6 m post-discharge; (2) achievement of controlled blood pressure 4–8 m post-discharge; (3) for those with AF, either documentation of a filled prescription for warfarin or an International Normalised Ratio blood test 6 m post-discharge or a contraindication to warfarin.	0.0038	3361	12
9	Jones et al. 22	All nurses and health-care assistants on the participating stroke intervention units received a group teaching package to improve their understanding and clinical practice in patient positioning.	Patients on stroke rehabilitation units (UK hospitals): with stroke, dependent on another person to position limbs, with inability to move from sitting to standing without assistance.	Rivermead Mobility Index at 6 months post-stroke.	0.00	120	10
10	Lakshminarayan et al. 23	(1) Audit and written feedback of baseline performance; (2) analysis of structural and knowledge barriers to stroke care identified by provider questionnaires; (3) use of clinical opinion leaders to deliver customised feedback to care providers; (4) use of hospital management leaders to overcome identified barriers to stroke care.	Stroke patients aged 30–84 years admitted through emergency room (US hospitals).	Three outcomes with associated ICCs, each is related to the provision of 3 or 4 indicators of quality of care:		2305	19
				acute care indicators.	0.005
				inpatient care indicators.	0.004
				discharge indicators.	0.0007
11	McAlister et al. 24	Educational lecture to patients with nonvalvular AF on reducing stroke risk, with self-administered booklet and individualised audiotape decision aid tailored to their personal stroke risk profile.	Adult patients with nonvalvular AF not living in institutions (Canada, Primary Care Practices).	Change in proportion of patients taking antithrombotic therapy appropriate to their stroke risk 3 months post-intervention.	0.02	434	102
12	Forster et al. 25	Structured training programme for caregivers (the London Stroke Carers Training Course).	Patients at UK stroke units with a diagnosis of stroke, likely to return home with residual disability and with a caregiver providing support.	Self-reported extended activities of daily living at 6 months measured with the Nottingham Extended Activities of Daily Living scale.	0.027	928	36
13	Taylor et al. 26	Structured goal elicitation using the Canadian Occupational Performance Measure.	Stroke patients admitted to inpatient rehabilitation services (New Zealand) with ‘sufficient’ cognition for goal setting and completing outcome assessment.	Quality of life at 12 weeks measured using the following tools:		41	4
				Schedule for Individualised Quality of Life (SEIQOL-DW).	0.40
				The Medical Outcomes Study 36-item Short Form Health Survey (SF-36), Physical Component Summary (PCS) score.	0.24
				Functional Independence Measure.	0.21
				The Medical Outcomes Study 36-item Short Form Health Survey (SF-36), Mental Component Summary (MCS) score.	0.25
14	Middleton et al. 27	Treatment protocols to manage fever, hyperglycaemia and swallowing dysfunction with multi-disciplinary team building workshops to address implementation barriers.	Patients aged 18 years or older, who had a diagnosis of ischaemic stroke or intracerebral haemorrhage, and presented within 48 h of onset of symptoms to a participating Acute Stroke Unit (ASU), Australia.	Death and dependency 90 days after hospital admission.	0.018	1696	19
				Functional dependency BI ≥95, 90 days after hospital admission.	0.015
				Functional dependency BI ≥60, 90 days after hospital admission.	0.009
				SF-36 PCS score, 90 days after hospital admission.	0.026
				SF-36 MCS score, 90 days after hospital admission.	0.011
				Mean temp within 72 h in ASU.	0.084
				At least one temperature ≥37.5°C in first 72 h.	0.009
				Mean glucose during first 72 h in ASU.	0.056
				Swallowing screening within 24 h of admission to ASU.	0.156
15	Power et al. 28	Stroke 90/10, a quality improvement collaborative based on the Breakthrough Series model.	Patients admitted to stroke units at NHS hospital Trusts in the Northwest of England.	Compliance with two evidence-based bundles of care: early hours and rehabilitation.		6592	24
				Early hours bundle.	0.066
				Rehabilitation bundle.	0.197
16	Dregan et al. 29	Remotely installed electronic decision support tools to promote intensive secondary prevention.	Patients ever diagnosed with acute stroke (Family Practices, UK).	Systolic blood pressure.	0.032	11,391	106

ICC: intracluster correlation coefficient; TIA: transient ischemic attack; NHS: National Health Service.