Nurse shift patterns, staffing and their association with perceived workload: Sequence analysis of multicentre data

Abstract

Introduction

Although shift work is inevitable in hospitals, some shift patterns and staffing levels are suggested to influence nurses’ workload more than others, which in turn can impact nurses’ health, quality of care, and patient safety. Despite the importance of workload in nursing practice, studies focusing on nurses’ work schedules, staffing levels and perceived workload are rare. The aims of this study were to describe key characteristics of nurses’ shift work patterns in acute care hospitals, and to investigate the association of shift work patterns and staffing levels with perceived workload.

Methods

This was a secondary analysis of an observational, cross-sectional, multicentre study conducted in 26 acute care hospitals in Switzerland. Registered nurses from 158 units completed the survey, covering questions about nurse staffing, the work environment and quality of care. We used sequenced data analysis to visualise nurses’ work schedules over the last seven days and identify shift characteristics and transitions. Clustering using Optimal Matching allowed us to group nurses with similar shift sequences and identify shift patterns. An observed-over-expected patient-to-nurse ratio (including patient acuity measures) was computed to assess staffing exposure. Perceived workload was measured with the NASA-Task Load Index instrument. A linear-mixed model was used to explore the association between identified shift work patterns, staffing and perceived workload.

Results

We analysed surveys of 1962 registered nurses. The sequence analysis identified 732 different sequences resulting in three clusters of different shift patterns. Backward rotations, quick returns and working more than five consecutive days were rare. Workload perception was on average 66.5 points (possible range 6–120). Low staffing (β=3.1, 95 % CI [0.5–5.6]), overtime in the last shift (β=8.8, 95 % CI [7.2–10.4]), higher percentage of days worked overtime in the previous seven days (β=3.9, 95 % CI [1.3–6.3]), number of days worked (β=6.4, 95 % CI [2.5–10.1]), last shift worked being a day shift (β=3.8, 95 % CI [1.8–5.8]), and longer shift length (β=1.4, 95 % CI [0.5–2.2]) were associated with higher perceived workload.

Conclusions

This study highlights the contribution of staffing and scheduling practices to nurses’ perceived workload. To reduce nurses’ perceived workload and improve healthcare performance—as previous research suggests—staffing and scheduling decisions must be increasingly prioritized by decision makers. The results suggest that avoiding or reducing e.g., overtime, reducing shift length and increasing staffing may be effective first strategies to reduce the perceived workload. Further research would benefit from analysing shift patterns using electronic rosters, real-time staffing measures, and repeated assessment of nurses’ workload perceptions.

Social media abstract

Subjective nurse workload is influenced by shift patterns and staffing. Less overtime, shorter shifts, and better staffing may reduce workload.