Shifts of 12 hours or longer are common in nursing services within general hospital wards. Concerns have been raised about their safety, but previous research has mostly used staff-reported measures of quality and safety and has occurred in general hospital settings only. This study aims to measure the association between the use of 12+ hour shifts in nursing staff (including registered nurses, healthcare support workers or nursing assistants, and nursing associates) and the rate of patient incidents in mental health and community hospitals. This is a longitudinal study using routinely collected data from two mental health and community NHS trusts in the South of England. We accessed rosters of nursing staff and patient incident data from April 2018 to March 2021. We extracted 1,018,971 shifts and excluded those not worked by nursing staff, with a final sample of 898,143 shifts. We extracted 53,078 incidents. We only included incidents that involved patients and that occurred in wards. Our final sample consisted of 38,373 patient incidents. We linked all patient incidents and nurses' worked shifts at the ward-day level. Depending on the distribution of incident rates, we used either negative binomial mixed-effects models or Poisson mixed-effect models to investigate the association between the proportion of 12+ hour shifts and all patient incidents, violence against staff, falls, self-injury, disruptive behaviour, and medication management incidents at the ward-day level. We found a relationship between 12+ hour shifts and the incident rate. Compared to days in wards with no long shifts, increasing the proportion of long shifts was initially associated with a small increase in the overall rate of incidents, but the rate increased sharply as the proportion of long shifts was above 70%. Rates of self-injury increased more steadily as the proportion of long shifts increased. The mandatory implementation of long shifts should be discouraged.
Dall’Ora et al. (Wed,) studied this question.