Email updates

Keep up to date with the latest news and content from BMC Nursing and BioMed Central.

Open Access Research article

Classifying nursing organization in wards in Norwegian hospitals: self-identification versus observation

Ingeborg S Sjetne12*, Jon Helgeland1 and Knut Stavem345

  • * Corresponding author: Ingeborg S Sjetne ing@nokc.no

Author Affiliations

1 Norwegian Knowledge Centre for the Health Services, PO Box 7004, St Olavs plass, N-0130 Oslo, Norway

2 Institute of Nursing and Health Sciences, University of Oslo, Norway

3 Medical Division, Akershus University Hospital, Lørenskog, Norway

4 Helse Sør-Øst Health Services Research Centre, Lørenskog, Norway

5 Faculty of Medicine, University of Oslo, Oslo, Norway

For all author emails, please log on.

BMC Nursing 2010, 9:3  doi:10.1186/1472-6955-9-3

Published: 9 February 2010

Abstract

Background

The organization of nursing services could be important to the quality of patient care and staff satisfaction. However, there is no universally accepted nomenclature for this organization. The objective of the current study was to classify general hospital wards based on data describing organizational practice reported by the ward nurse managers, and then to compare this classification with the name used in the wards to identify the organizational model (self-identification).

Methods

In a cross-sectional postal survey, 93 ward nurse managers in Norwegian hospitals responded to questions about nursing organization in their wards, and what they called their organizational models. K-means cluster analysis was used to classify the wards according to the pattern of activities attributed to the different nursing roles and discriminant analysis was used to interpret the solutions. Cross-tabulation was used to validate the solutions and to compare the classification obtained from the cluster analysis with that obtained by self-identification. The bootstrapping technique was used to assess the generalizability of the cluster solution.

Results

The cluster analyses produced two alternative solutions using two and three clusters, respectively. The three-cluster solution was considered to be the best representation of the organizational models: 32 team leader-dominated wards, 23 primary nurse-dominated wards and 38 wards with a hybrid or mixed organization. There was moderate correspondence between the three-cluster solution and the models obtained by self-identification. Cross-tabulation supported the empirical classification as being representative for variations in nursing service organization. Ninety-four per cent of the bootstrap replications showed the same pattern as the cluster solution in the study sample.

Conclusions

A meaningful classification of wards was achieved through an empirical cluster solution; this was, however, only moderately consistent with the self-identification. This empirical classification is an objective approach to variable construction and can be generally applied across Norwegian hospitals. The classification procedure used in the study could be developed into a standardized method for classifying hospital wards across health systems and over time.