Abstract
IMPORTANCE: Rates of work-related stress and sickness absence are high among nurses. The consequences of sickness absence are severe for health systems' efficiency and productivity. OBJECTIVE: To examine the association between nurse staffing configurations and sickness absence in hospital inpatient unit nursing teams. DESIGN, SETTING, AND PARTICIPANTS: This retrospective longitudinal case-control study using routinely collected hospital data among 4 general acute care hospitals in England examined 18 674 registered nurses (RNs) and nursing support (NS) staff working in 116 hospital units. Data were analyzed from April 1, 2015, to February 29, 2020. EXPOSURE: Nursing team skill mix, temporary staffing hours, understaffing, proportion of long shifts (≥12 hours) worked, and full-time or part-time work status in the previous 7 days. MAIN OUTCOME AND MEASURE: The main outcome was the number of episodes of sickness absence, which was defined as a sequence of sickness days with no intervening days of work. RESULTS: There were 2 690 080 shifts and 43 097 sickness episodes. The incident rates for starting a sickness episode were 2.0% for NSs and 1.4% for RNs. In total, there were 18 674 members of staff; 2 188 562 (81.6%) were shifts by staff classified as working full time and 493 400 (18.4%) by staff classified as working part time. A skill mix composed of more RNs was associated with a 2% reduction in the odds of RN sickness absence for each 10% change in the proportion of RN hours (odds ratio [OR], 0.98; 95% CI, 0.96-0.99). Moving from the lower quartile skill mix (ie, approximately 55% of the nursing hours worked by RNs) to the upper quartile (ie, approximately 75% of the nursing hours worked by RNs) was associated with a 4% reduction in odds of sickness absence. Working only long shifts in the prior 7 days was associated with a 26% increase in the odds of sickness absence for RNs compared with working no long shifts (OR, 1.26; 95% CI, 1.19-1.33). Part-time work for RNs was associated with higher sickness absence (OR, 1.09; 95% CI, 1.03-1.15). For RNs, there was a significant interaction between part-time work and RN understaffing, whereby short staffing in the previous week increased sickness absence for full-time staff but not among those working part time. NS understaffing was not associated with sickness absence for any staffing group. CONCLUSIONS AND RELEVANCE: This retrospective case-control study found that exposure to RN understaffing increased the risk of sickness absence in nursing teams. Adverse working conditions for nurses, already known to pose a risk to patient safety, may also create risks for nurses and the possibility of further exacerbating staff shortages.