Research Article

Nursing 12-Hour Shifts and Patient Incidents in Mental Health and Community Hospitals: A Longitudinal Study Using Routinely Collected Data

Table 1

Negative binomial and Poisson mixed-effects regression coefficients for the association between 12+ hour shifts and patient incidents.

Negative binomial mixed-effects regressions
Patient incidents (unadjusted)Patient incidents (adjusted)
BSEBSE

Proportion of long shifts0.340.190.480.20
Proportion of long shifts squared−1.020.54−1.330.57
Proportion of long shifts cubic0.730.401.080.43
AICBICAICBIC
AIC/BIC101212.7101265.6100938.8101027.1

Patient incidents with any harm (unadjusted)Patient incidents with any harm (adjusted)
BSEBSE

Proportion of long shifts−0.090.300.010.32
Proportion of long shifts squared0.760.860.540.95
Proportion of long shifts cubic−0.470.64−0.230.70
AICBICAICBIC
54659.954712.954609.854698.0

Violence against staff (unadjusted)Violence against staff (adjusted)
BSEBSE

Proportion of long shifts−0.330.41−0.060.47
Proportion of long shifts squared−2.381.32−2.741.54
Proportion of long shifts cubic2.170.952.631.12
AICBICAICBIC
32452.732505.732323.032411.3

Self-injury (unadjusted)Self-injury (adjusted)
BSEBSE

Proportion of long shifts0.770.450.830.48
Proportion of long shifts squared−0.491.43−0.661.54
Proportion of long shifts cubic0.310.980.541.06
AICBICAICBIC
28105.228158.22800728095.3

Disruptive behaviour (unadjusted)Disruptive behaviour (adjusted)
BSEBSE

Proportion of long shifts0.750.571.030.55
Proportion of long shifts squared−41.82−3.831.77
Proportion of long shifts cubic3.091.243.271.22
AICBICAICBIC
230622311522879.122967.4

Poisson mixed-effects regressions
Medication management incidents (unadjusted)Medication management incidents (adjusted)
BSEBSE

Proportion of long shifts0.360.520.540.58
Proportion of long shifts squared−0.591.51−0.721.75
Proportion of long shifts cubic0.121.220.131.41
AICBICAICBIC
21903.321947.521856.921936.3

Falls (unadjusted)Falls (adjusted)
BSEBSE

Proportion of long shifts0.230.550.190.56
Proportion of long shifts squared−0.501.72−0.631.72
Proportion of long shifts cubic0.161.490.371.49
AICBICAICBIC
25211.825256.025207.425286.9

 < 0.05. Controlled for the number of nursing staff per bed, proportion of substantive shifts, skill-mix (registered nurses hours/registered nurses + healthcare support worker + nursing associate hours), and type of ward (physical vs. mental health).