The effect of Rapid Response System revision on standard and specific intensive care unit outcomes in a regional hospital.

Hospital systems for the recognition (afferent limb) and management (efferent limb) of deteriorating patients, or Rapid Response Systems (RRSs), are being mandated worldwide, in spite of conflicting evidence regarding their efficacy. We have evaluated the impact of an Adult Deterioration Detection System (Q-ADDS)-based RRS specifically on illness severity at intensive care unit (ICU) admission and ICU length of stay (LOS), as well as previously studied endpoints. We undertook a retrospective, single-centre observational study comparing equivalent 18-month periods before the Q-ADDS-based RRS, and after implementation. The primary endpoints of the study were illness severity of unplanned ICU admissions from the ward, ICU length of stay, and ICU mortality. Secondary endpoints were RRS call numbers, rate of unplanned ICU admissions, and ward-based cardiorespiratory arrests. Following the introduction of the new RRS, Acute Pain and Chronic Health Evaluation (APACHE) II (17 versus 21, <0.001), APACHE III (64 versus 68, =0.011) and Simplified Acute Physiology Score (35 versus 38, =0.044) scores at ICU admission from the ward were reduced. Fewer patients were in the >50% predicted mortality range of APACHE II (16% versus 32%, <0.001), APACHE III (18% versus 28%, =0.012) and Simplified Acute Physiology Score (14% versus 24%, =0.006). ICU mortality was unchanged (13.7% versus 13.8%, =0.93). ICU LOS was reduced (3 versus 4 days, =0.02); prolonged stay (>7 days) was not significantly changed (19% versus 27%, =0.055). Unplanned ICU admissions, cardiorespiratory arrests and hospital mortality were unchanged. The frequency of RRS activation (48 versus 11 per 1,000 admissions, <0.001) was markedly increased. This Q-ADDS form-based RRS has resulted in lower illness severity at ICU admission from the ward, and fewer patients with scores associated with a >50% predicted mortality. Overall, ICU length of stay was reduced. These specific outcomes may reliably reflect RRS efficacy, even in smaller centres.