Table 1 Summary of studies of Rapid Response Teams involving comparison data^a

Bristow et al. 2000 [32]

Case control cohort study. Comparison between one MET hospital and two cardiac arrest team hospitals

Doctor

Fewer unanticipated ICU/high dependency unit admissions in MET hospital. No difference in in-hospital cardiac arrests or mortality

Buist et al. 2002 [30]

Before (1996) and after (1999) study. MET introduced in 1997 and activation criteria simplified 1998

Doctor

Reduction of cardiac arrest rate from 3.77 to 2.05/1,000 admissions. OR for cardiac arrest after adjustment for case mix = 0.50 (95% CI 0.35 to 0.73)

Bellomo et al. 2003 [29]

Before (4 months 1999) and after (4 months 2000 to 2001) 1-year preparation and eduction period

Doctor

RRR cardiac arrests 65% (P < 0.001). Decreased bed days cardiac arrest survivors (RRR 80%, P < 0.001). Reduced hospital mortality (RRR 26%, P = 0.004)

Bellomo et al. 2004 [33]

Time periods and design as above. Assessment of effect of MET on serious adverse events following major surgery

Doctor

Reduction in serious adverse events (RRR 57.8%, P < 0.001), emergency ICU admissions (RRR 44.4%, P = 0.001), postoperative deaths (RRR 36.6%, P = 0.0178), and hospital length of stay (P = 0.0092)

Kenward et al. 2004 [34]

Before and after (October 2000 to September 2001) introduction of MET

Doctor

Decreased deaths (2.0% to 1.97%) and cardiac arrests (2.6/1,000 to 2.4/1,000 admissions).

Not significant

DeVita et al. 2004 [31]

Retrospective analysis of MET activations and cardiac arrests over 6.8 years

Doctor

Increased MET use (13.7 to 25.8/1,000 admissions) was associated with 17% reduction cardiac arrests (6.5 to 5.4/1,000 admissions, P = 0.016)

Priestly et al. 2004 [25]

Single-centre ward-based cluster randomized control trial of 16 wards

Nurse^c

Critical care outreach reduced in-hospital mortality (OR 0.52, 95% CI 0.32 to 0.85) compared with control wards.

MERIT 2005 [23]

Cluster randomized trial of 23 hospitals in which 12 introduced a MET and 11 maintained only a cardiac arrest team.

Four-month preparation period and 6-month intervention period

Doctor

Increased overall call rates (3.1 versus 8.7/1,000 admissions, P = 0.0001). No decrease in composite end point of cardiac arrests, unplanned ICU admissions and unexpected deaths

Jones et al. 2005 [16]

Long-term before (8 months 1999) and after (4 years) introduction of MET

Doctor

Decreased cardiac arrests (4.06 to 1.9/1,000 admissions; OR 0.47, P < 0.0001). Inverse correlation between MET rate and cardiac arrest rate (r² 0.84, P = 0.01)

Jones et al. 2007 [22]

Long-term before (September 1999 to August 2000) and after (November 2000 to December 2004) study. Effect on all-cause hospital mortality

Doctor

Reduced deaths in surgical patient compared with 'before' period (P = 0.0174). Increased deaths in medical patients compared with 'before' period (P < 0.0001)

Jones et al. 2007 [35]z

Time periods of design as per [29]. Study assessed long-term (4.1 years) survival of major surgery cohort

Patients admitted in the MET period had a 4.1-year survival rate of 71.6% versus 65.8% for control period. Admission during MET period was an independent predictor of decreased mortality (OR 0.74, P = 0.005)

Buist et al. 2007 [18]

Assessment of MET call rates and cardiac arrests between 2000 and 2005

Doctor

Increased MET use was associated with reduction in cardiac arrest of 24% per year, from 2.4 to 0.66/1,000 admissions

Jones et al. 2008 [36]

Multi-centre before-and-after study. Assessment of cardiac arrests admitted from ward to ICU before and after introduction of RRT

Varied

Continuous data only available for one-quarter of 172 hospitals. Temporal trends suggest reduction in cardiac arrests in both MET and non-MET hospitals

Chan et al. 2008 [26]

18-month-before and 18-month-after study following introduction of RRT

Nurse^c

Decrease in mean hospital codes (11.2 to 7.5/1,000 admissions) but not significant after adjustment (0.76 (95% CI, 0.57 to 1.0); P = 0.06). Lower rates of non-ICU codes (AOR 0.59 (95% CI, 0.40 to 0.89) versus ICU codes AOR, 0.95 (95% CI, 0.64 to 1.43); P = 0.03 for interaction). No decrease in hospital-wide mortality 3.22% versus 3.09% (AOR, 0.95 (95% CI, 0.81 to 1.11); P = 0.52)