Macro-management:
Four studies (number of participants not stated) on hospital closures (1 quasi-experimental, 1 ecological, 1 economic modelling/cross- sectional data, and 1 opinion). No studies were listed on changes to hospital reimbursement on admission rates.
Five studies (total number of participants not stated) on prospective diagnostic related grouping (DRG) payments (1 RCT (n = 1580), 2 quasi-experimental, and 2 cross-sectional).
Three studies (number of participants not stated) on fundholding or budget-holding status (2 quasi-experimental and 1 cross- sectional/record linkage).
Four studies (total number of participants not stated) on the effect of HMOs on acute admission rates (1 RCT (n = 1580), 1 cohort (n = 3006), 1 case-control, and 1 retrospective case series (n = 102)).
Eight studies (total number of participants not stated) on public health interventions to reduce admissions (2 quasi-experimental, 2 cohort (n = 26,446), 2 time series cross-sectional, and two ecological).
No studies were listed on the provision of home hospital care or for hospital at home for acute illness.
Nine RCTs (total number of participants not stated, 8 studies total 2,449 participants) on hospital at home to assist early discharge to reduce subsequent admissions.
Five RCTs (number of participants not stated, 4 studies total 914 participants) on hospital at home for terminal care.
No studies were listed for high technology at home.
Five studies (number of participants not stated) on community hospitals/GP beds (1 quasi-experimental, 1 case-control, 1 audit, 1 cross-sectional/audit, and 1 semi-systematic review).
Two descriptive studies (4,920 participants) on patient hotels.
Micro-management interventions:
Ten studies (number of participants not stated) on primary care initiatives on reducing admissions (1 RCT, 1 systematic review, 1 quasi-experimental, 5 cross-sectional, 1 ecological, and 1 non- systematic review).
Eleven studies (number of participants not stated) on hospital outpatient-based interventions to reduce admissions (2 RCTs (1,370 participants) and 1 meta-analysis (906 participants) on outpatient- based education delivered to individuals; 1 RCT (321 participants) and 1 case-control (84 participants) on group outpatient education; 2 RCTs (533 participants) and 2 cohort (275 participants) on increased outpatient services and improved GP referral to outpatient services; 1 quasi-experimental (number not stated) on urgent referral service for GPs with hospital consultants; and 1 quasi-experimental (52 participants) on outreach service provided by hospital outpatient departments).
Eight studies (number of participants not stated) on emergency department (ED) interventions (1 database review (102,411 participants) on increased ED services; 2 RCTs (9,325 participants) on use of GPs in the ED; 1 quasi-experimental (number of participants not stated) on provision of senior staff in the ED; 2 descriptive studies (572 participants) on the provision of a social worker in the ED; 1 case-control study (300 participants) on regional services and air ambulances; and 1 descriptive study (number of participants not stated) on separate paediatric emergency service).
Four studies (5,786 participants) on emergency observation units (2 RCTs and two before and after studies).
Six studies (number of participants not stated) assessing before and after costs associated with the introduction of an observation unit in relation to the treatment of different conditions.
Two studies (1,602 participants) on chest pain observation units (1 9-hour observation, and 1 24-hour surveillance).
Three studies (total number of participants not stated) on assessing the effect of case management on acute admissions (1 RCT (668 participants), 1 cohort (186 participants) and 1 systematic review). There were no studies listed on maximising bed utilisation.
Six RCTs (5 studies report number of participants = 1,947) on the efficacy of home visits to reduce acute medical admissions in the elderly.
Four studies (522 participants) examining the effect of personal alarms on acute admissions (3 quasi-experimental and 1 cost-benefit).
Four RCTs (total number of participants not stated, 2 studies list 1,179 participants) on preventing acute medical (re)admission amongst the elderly.
Five studies (total number of participants not stated) on utilisation review to reduce inappropriate admissions (1 RCT (7,445 participants), 2 quasi-experimental (1 lists 1,800 chart reviews), and 2 cross sectional (20,173 chart reviews)).
Five studies (6,294 participants) assessing the use of guidelines/protocols/critical pathways/treatment algorithms on hospital admissions (2 RCTs, 2 quasi-experimental, and 1 cohort).
Two uncontrolled descriptive studies (23 participants) on the effect of aerosolized antibiotics on admission rated for people with cystic fibrosis.
Four placebo-controlled trials (94 participants) on the effect of aerosolized antibiotics in cystic fibrosis.
The effect on number of admissions was reduced for the macro- management initiatives of (level of evidence for finding in parentheses after initiative): closure of hospitals (II); public health preventative interventions (II); alternatives to hospital (hospital at home for early discharge or terminal care or acute care - high tech) (I); community hospitals - GP beds (II); patient hotels (IV); comprehensive geriatric care (I); home alarm (II); and increased long-term care - improved nursing home care options (II).
The effect on number of admissions was mixed for the macro-management initiative of (level of evidence for finding in parentheses after initiative): home care (I).
The effect on number of admissions was reduced for the micro- management initiatives of (level of evidence for finding in parentheses after initiative): increase skill GP (IV); drug education to GPs and patients (II); hospital outreach services (II); ED-based GPs (I); ED more senior staff (II); provision of social worker in the ED (IV); separate paediatric ED (IV); observation units (II); chest pain units (II); and change hospital reimbursement to prospective funding (I).
The effect on number of admissions was mixed for the micro-management initiatives of (level of evidence for finding in parentheses after initiative): primary care - increased access to GPs (I); primary care - change behaviour of GPs (I); and GP-based budget holding (II).
There was no effect on number of admissions for the micro-management initiatives of (level of evidence for finding in parentheses after initiative): outpatient-based individual or group education (I); increased outpatient services (I); or regionalised ED services (III).