ArticlesHospital mortality in relation to staff workload: a 4-year study in an adult intensive-care unit
Introduction
Patients in intensive-care units (ICUs) could be at greater risk when nursing or medical workload is high than during periods of lower workload. Under these circumstances, the risks of iatrogenic complications,1 human error,2 delayed weaning from mechanical ventilation,3 and hospital-acquired infection4, 5, 6 could all increase, with potentially adverse consequences. To provide guidance on appropriate workload, the UK Intensive Care Society7 has proposed a nursing dependency schedule recommending the numbers of nurses required for different types of patients. We investigated whether hospital mortality is independently related to nursing requirement and other measures of workload, after adjusting for risk by use of the APACHE II (Acute Physiology and Chronic Health Evaluation) equation.8, 9, 10
Section snippets
Patients and staffing policy
This retrospective analysis was done on data from a prospective cohort study of all admissions to the ICU at Ninewells Hospital, Dundee, Scotland, between Jan 1, 1992, and Dec 31, 1995. The unit provides medical and surgical intensive care, with the exception of patients who have undergone cardiac surgery and some types of neurosurgery, for a population of 440 000 in Tayside and North Fife. Throughout the study, the ICU medical establishment provided for continuous cover by one consultant, one
Measures of workload
Occupancy per shift was the highest number of ICU beds occupied each shift, and peak occupancy was the highest occupancy per shift during the patient's stay. Nursing requirement per shift, recorded by the senior nurse at the end of each shift, was the highest number of nurses required for the ICU according to the recommendations of the UK Intensive Care Society7 (0·5 nurses per patient per shift for patients who are spontaneously breathing and need simple monitoring only; 1·0 for artificially
Statistical analyses
Statistical analyses were undertaken with SPSS (version 8·0). For each patient, the predicted risk of mortality, R, was calculated from the APACHE II equation.8 We did univariate analyses to test how strongly each measure of ICU workload was individually associated with observed mortality, having adjusted for logit (R)—ie, In(R/1-R). All measures of ICU workload that were significant (p<0·05) in univariate analysis were then fitted, with logit (R), in multivariate models of mortality by forward
Results
There were 1286 admissions during the study period. 236 were excluded: 88 patients were younger than 16 years; 21 were discharged from the ICU within 8 h; 36 died in the ICU within 8 h; 61 were readmitted to the ICU without being discharged from the hospital; 28 were transferred from another ICU; and two had burns. Thus, there were 1050 separate episodes in the cohort (table 1), representing 1025 patients, because 25 episodes were readmissions after discharge home. Among the 1050 episodes,
Discussion
We found that patients exposed to high ICU workload were more likely to die than those exposed to lower workload, both before and after adjustment for risk by the APACHE II equation. The three measures of ICU workload most strongly associated with mortality were peak occupancy, average nursing requirement per occupied bed per shift, and the ratio of occupied to appropriately staffed beds.
This study recruited a complete cohort of all admissions eligible for assessment by the APACHE II equation
References (20)
- et al.
Iatrogenic complications in adult intensive care units: a prospective two-center study
Crit Care Med
(1993) - et al.
A look into the nature and causes of human errors in the intensive care unit
Crit Care Med
(1995) - et al.
Influence of the quality of nursing on the duration of weaning from mechanical ventilation in patients with chronic obstructive pulmonary disease
Crit Care Med
(1995) - et al.
Nosocomial infections in a neonatal intensive care unit
J Infect Dis
(1981) - et al.
The role of understaffing and overcrowding in recurrent outbreaks of staphylococcal infection in a neonatal specialcare unit
J Infect Dis
(1982) - et al.
Factors influencing colonisation with gentamicin resistant gram negative organisms in the neonatal unit
Arch Dis Child
(1988) Intensive care audit
(1990)- et al.
APACHE II: a severity of disease classification system
Crit Care Med
(1985) - et al.
Intensive Care Society's APACHE II study in Britain and Ireland: I, variations in case mix of adult admissions to general intensive care units and impact on outcome
BMJ
(1993) - et al.
Intensive Care Society's APACHE II study in Britain and Ireland: II, outcome comparisons of intensive care units after adjustment for case mix by the American APACHE II method
BMJ
(1993)