Time Trends of Percutaneous Injuries in Hospital Nurses: Evidence of the Interference Between Effects of Adoption of Safety Devices and Organizational Factors

Few studies have focused on the combined effects of devices and work organization on needlestick injuries trends. The aim of the study was to estimate trends of percutaneous injury rates (IR) in nurses (N) and nurse assistants (NA) over a 10 year period, in which passive safety devices were progressively adopted. Percutaneous and mucocutaneous injuries registered in a University Hospital in Northern Italy in Ns and NAs in 2007-2016 were analyzed. Organizational data were also available on shift schedules, turnover, downsizing and age- and skill-mix. We estimated IRs per 100 full-time equivalent workers from Poisson models and their average annual percent changes (APC) from joinpoint regression model. In the entire period, monotonic decreases in percutaneous IRs occurred among day-shift Ns (APC = -20.9%; 95% CI: -29.8%, -12%) and NAs (APC = -15.4%; -32.9%, 2.2%). Joinpoint modeling revealed a turning point in 2012 for night-shift Ns, with a steady decline in 2007-2012 (APC = -19.4%; -27.9%, -10.9%), and an increase thereafter (APC = +13.5%; 1.5%, 25.5%). In comparison to 2008 and 2012, in 2016 night-shift Ns were 5.9 and 2.5 times more likely to be younger and less qualified or experienced than day-shift Ns. The observed declines in percutaneous injury rates occurred in a time period when safety devices were progressively implemented. The causal nature of multiple exposures and organizational procedures in affecting injury time trends should be further addressed by quasi-experimental studies.