In our prospective study encompassing 24 months, we found a positive correlation of CLABSI incidence and turnover of nursing personnel on the hospital level. Among nursing personnel, the positive correlation was confirmed for LPNs and RNs, but not for ANPs. No significant correlation was detected for physician turnover.
CLABSI can result from contamination of infusion systems or more frequently from catheter contamination [8]; especially, the catheter hub seems to be a vulnerable location in contamination of central lines [9]. Causative pathogens originate either from the patient’s flora itself, the hands of HCWs or contaminated medical products, e. g. disinfectants used for skin cleansing. We found a positive correlation of CLABSI incidence and turnover of nurses with advanced training, i.e. LPN and RN, which reflects the subset of nursing personnel that is involved in central line handling. Robert et al. highlighted the role of continuity in nursing staff in primary bacteremias [10]. Composition of nurse staff defined as pool-nurse-to-patient-ratio was found to be an independent risk factor for CLABSI with an odds ratio of 3.8. Similarly, our study supports a relevance of continuity in nursing personnel for CLABSI prevention.
For physicians we did not detect an association of turnover and CLABSI incidence. The finding that an association was exclusively detected with nursing staff turnover, but not with physician turnover might reflect the crucial role of catheter care and infusion management after central line placement. At our center, central lines are exclusively placed by physicians, whereas later on central line care and intravenous drug administration is provided by nurses. It can be speculated that prolonged work experience at a specific hospital or even ward results in improved knowledge of infection prevention measures and center specific central line products thus lowering the risk of CLABSI. Overall nursing experience might be also relevant for prevention of HAIs in general. One hypothesis can be that a parallel increase in infection prevention skills goes along with duration of employment as nurse. Nursing staff inexperience has been associated with a negative impact on quality of care in ICUs indicated by the frequency and outcome of incidents related to inexperience [11]. However, in our dataset we did not have information on the individual’s work experience hindering an analysis on the influence of fluctuations in overall nursing experience. Studies on the role of nursing personnel turnover in HAIs are scarce. In a multicenter study encompassing 59 US nursing homes, turnover of RNs was associated with a higher incidence of infectious complications as well as infection-related hospitalization [12]. The authors speculated that personnel turnover makes an establishment and maintenance of effective infection control practices difficult. Turnover could result in inconsistencies in supervision and training which might affect quality.
The level of professional training among nursing personnel has been previously reported as relevant for patient outcomes. Needleman et al. found that a larger proportion of care provided by RNs was associated with a reduction of urinary tract infections and pneumonia among medical patients. A larger proportion of care provided by RNs correlated with a decrease in urinary tract infections among surgical patients [13]. Similarly, in an US study an increase in the ratio of licensed nurses to total nursing staff correlated with a lower incidence of pneumonia [14]. Among trauma patients a higher ratio of nursing hours provided by LPNs to total nursing hours provided by LPNs and RNs showed a significantly higher odds in mortality and HAIs [15]. Recent data also support a protective effect from academic training among nurses. In a European study, an increase in nurses with a bachelor degree was associated with lower in-hospital mortality [16]. The steady rising cost pressure likely impacts composition of nurse staffing. To save costs nursing activities might be shifted from higher qualified nursing personnel such as RNs or LPNs to ANPs. However, as outlined above these changes could affect the likelihood of adverse patient outcomes such as HAIs. Future studies that address economic aspects of care should consider both, the higher salary for better qualified nursing personnel and the additional costs for CLABSI.
Strengths of the present study include the prospective design with standardized CLABSI surveillance ensuring strict application of CDC definitions and the availability of monthly data on personnel turnover.
Our study has several limitations. We were not able to address numerous potentially confounding variables for several reasons. The study duration was limited to 2 years as changes in the patient data management system of the intensive care units put our semiautomatic CLABSI surveillance system out of order. This restricted study duration hampered statistical analyses, as due to the limited study duration the number of CLABSI was small for several institutes or departments. In addition, the granularity of certain variables was limited, as e.g. HRM-derived data was aggregated at the level of institute or department. These limitations did not allow to perform detailed analyses per ward. Furthermore, we did not have information on nurse:patient ratios and on the use of pool or float nurses.
However, our pilot study should motivate to plan larger-scale studies with the possibility to correct for these confounders in future and allow more detailed analysis. Future studies, ideally with multi-centric design, that investigate the role of nursing personnel turnover on CLABSI incidence and other HAIs seem warranted.