Staged Implementation of Awakening and Breathing, Coordination, Delirium Monitoring and Management, and Early Mobilization Bundle Improves Patient Outcomes and Reduces Hospital Costs

Abstract

Objectives: To measure the impact of staged implementation of full versus partial ABCDE bundle on mechanical ventilation duration, ICU and hospital lengths of stay, and cost.

Design: Prospective cohort study.

Setting: Two medical ICUs within Montefiore Healthcare Center (Bronx, NY).

Patients: One thousand eight hundred fifty-five mechanically ventilated patients admitted to ICUs between July 2011 and July 2014.

Interventions: At baseline, spontaneous (B)reathing trials (B) were ongoing in both ICUs; in period 1, (A)wakening and (D)elirium (AD) were implemented in both full and partial bundle ICUs; in period 2, (E)arly mobilization and structured bundle (C)oordination (EC) were implemented in the full bundle (B-AD-EC) but not the partial bundle ICU (B-AD).

Measurements and main results: In the full bundle ICU, 95% patient days were spent in bed before EC (period 1). After EC was implemented (period 2), 65% of patients stood, 54% walked at least once during their ICU stay, and ICU-acquired pressure ulcers and physical restraint use decreased (period 1 vs 2: 39% vs 23% of patients; 30% vs 26% patient days, respectively; p < 0.001 for both). After adjustment for patient-level covariates, implementation of the full (B-AD-EC) versus partial (B-AD) bundle was associated with reduced mechanical ventilation duration (-22.3%; 95% CI, -22.5% to -22.0%; p < 0.001), ICU length of stay (-10.3%; 95% CI, -15.6% to -4.7%; p = 0.028), and hospital length of stay (-7.8%; 95% CI, -8.7% to -6.9%; p = 0.006). Total ICU and hospital cost were also reduced by 24.2% (95% CI, -41.4% to -2.0%; p = 0.03) and 30.2% (95% CI, -46.1% to -9.5%; p = 0.007), respectively.

Conclusions: In a clinical practice setting, the addition of (E)arly mobilization and structured (C)oordination of ABCDE bundle components to a spontaneous (B)reathing, (A)wakening, and (D) elirium management background led to substantial reductions in the duration of mechanical ventilation, length of stay, and cost.

Figure 1.. Timeline of staged implementation of ABCDE in partial (B-AD only) vs full (B-AD-EC) bundle ICU’s and data measurement periods

Definition of abbreviations: A = awakening from sedation; B = spontaneous breathing trial; C = structured coordination of bundle components; D = delirium monitoring and management; E = early mobilization (A) Periods of component implementation in the full and partial bundle ICU’s. At baseline, spontaneous (B)reathing trials were ongoing in both full and partial bundle ICU’s; on July 1, 2012, (A)wakening and (D)elirium monitoring/management were implemented in both ICU’s; on July 1, 2013, (E)arly mobilization and structured bundle (C)oordination were implemented in only the full bundle ICU. (B) Periods in which process of care, clinical outcomes, and cost data were collected relative to bundle implementation. a Process of care measurements (sedative use, delirium prevalence, maximum level of mobility) were compared across time in the full bundle ICU (B-AD-EC) only. b ICU quality indicators, clinical outcomes, and cost were compared across time in both the full (B-AD-EC) and partial (B-AD) bundle ICU’s. c Cost periods were truncated because cost data are calculated based on a cost-to-charge ratio which varies between calendar years. The following periods were compared for the cost analysis: 1) Baseline vs Period 1 (i vs ii); and 2) Period 1 vs Period 2 (iii vs iv)

Figure 2.. Clinical quality outcomes in full and partial bundle ICU’s (Periods 1 vs 2)

Definition of abbreviations: A = awakening from sedation; B = spontaneous breathing trial; C = structured coordination of bundle components; D = delirium monitoring and management; E = early mobilization Quality metrics from aggregate hospital-reported data Centers for Medicare and Medicaid Services quality indicators were compared between Periods 1 vs 2 in both full and partial bundle ICU’s (A) In the full bundle ICU (B-AD vs B-AD-EC), pressure ulcer incidence and physical restraint use decreased (p<0.001 for both) (B) In the partial bundle ICU (B-AD vs B-AD), pressure ulcer incidence and physical restraint use increased (p=0.04, p=0.001, respectively)