Hugonnet et al. (2007) [18]

Prospective observational cohort study

Moderate

Switzerland

To determine whether low nurse-to-patient ratio increases risk for VAP and whether this effect is similar for early-onset and late-onset VAP.

2470 ICU patients.

Variable such as number of

patients and nurses on duty, patient characteristics, nurse training levels

262 VAP episodes were diagnosed in 22.3% of the patients who underwent mechanical ventilation

The median daily nurse-to-patient ratio was 1.9 over the study period.

High nurse-to-patient ratio was associated with a decreased risk for late-onset VAP, but there was no association with early-onset VAP

Direct consequences: patient harm

Holden et al. (2011) [53]

Cross-sectional survey

Moderate.

USA

To measure each of the three types of workload experienced by nurses and to assess whether and which measures of workload were related to three important outcomes.

99 nurses from two hospitals

The task level measure of mental workload related to interruptions, divided intention, and being rushed was associated with burnout and medication error.

Workload was not significantly associated with any other outcomes

Indirect consequences: Medication errors

Kiekkas et al.

(2008) [38] Observational prospective study (cohort study)

Moderate

Greece

Investigate differences in mortality of intensive care unit (ICU) patient according to the ratio between total patient care demands and nurse staffing

396 patients admitted in the general ICU of an academic, tertiary care, Greek Hospital from October 2005 to September 2006

Mortality on all patients increased from 22% in the low-exposure group (<21.9) Therapeutic Intervention Scoring System (TISS-28) point per nurse-workload measurement) to 25.0% in the medium exposure group (21.9 - 25.8 TISS-28 point per nurse) and reached up to 28.8% in the high-exposure group (>25.8 TISS-28 score per nurse)

Despite these increases, differences in adjusted ICU mortality among groups did not reach statistical significance

Direct consequences patient mortality

Kovner et al.

(2002) [50]

Cross-sectional descriptive study

Moderate

USA

To examine the impact of nurse staffing on selected adverse events hypothesized to be sensitive to nursing care

Nurse staffing data from 1990-1996 from the American hospital association annual survey of hospitals.

Includes 530 - 570 hospitals for each of the years from 1990-1996, with 187 hospitals having data for all seven years.

Registered nurse (RN) hours per patient per day were inversely related to all adverse events, but was significant (P < 0.05) only for pneumonia

Direct consequences: Patient harm

Liang et al.

(2012) [39]

Longitudinal, cross-sectional Study

Strong

Taiwan

To explore the effects of nurse staffing ratios on patient mortality in acute care hospitals.

108 hospital nursing units in 32 of Taiwan’s’ 441 accredited Western medicine district/regional hospitals and medical centers.

Data from a survey on hospital nurse staffing levels and patient outcomes.

The risk of incidence of death seemed to be higher in high patient-nurse ratio groups than in low patient-nurse ratio groups. The risk of incidence of death in high healthcare workforce-bed ratio groups was much lower

than in low healthcare workforce-bed ratio groups.

Direct consequences: Patient harm