| Background. The neonatal intensive care unit (NICU) has been described as a "first-time safe" system, similar to nuclear and aviation industry, since it is a complex unit that requires strict methods to monitor and control system safety in order to minimize harm to vulnerable neonates. Medication errors can easily occur in the NICU because of neonates’ under developed physical conditions and complex medication treatments, which all increase neonates vulnerability in medication errors. Nurses are the primary individuals in medication administration, monitoring, and education; therefore, they play an important role in neonates’ medication safety. World Health Organization (WHO) has advocated to adopt the managing strategies from other high risk industries to health care system to promote health care systematic safety. To date, little is known about risk factors associated with nurses’ medication/near-miss errors in NICUs because of lack of studies in systematic risk identifying, analysis and evaluation.Objectives. (1) To identify and analyze risk factors of medication/near-miss errors based on staff nurses’ self-reported data; (2) Guided by M-SHEL theoretical model to establish a medication risk evaluation indicator system based on the results from first aim, and construct risk evaluation model to evaluate the comprehensive level of risk, then verify the feasibility of the model.Methods. This is a cross-sectional, descriptive exploratory study:Phase Ⅰ used the ASSESS-ERRTM Medication System Worksheet from the Institute for Safe Medication Practices (ISMP 2006) to collect medication/near-miss errors from 30 NICU nurses at three hospitals in Beijing between January and December,2014. Data collected included: (a) characteristics of medication errors; (b) types of medication errors; (c) outcomes of infants, and (d) possible systemic risk factors resulting from the error. Data were used to describe the characteristics of medication errors; to identify possible risk factors as the indicators by using Fault Tree Analysis (FTA, qualitative method); to analyze and quantizing the medication/near-miss errors by using Grey Relational Analysis (GRA), further to identify and rank the associations among medication/near-miss errors and the system risk indicators, then to construct risk evaluation model. The research team built upon on the findings from Phase Ⅰ and using M-SHEL framework to establish the risk evaluation indicator system could by using brainstorming method. Combined with the subjective and objective weight, the judgment matrix can be constructed. Analyzing the interdependency relations of evaluation indicator, the systemic risk factors analytic network process (ANP) network structure is constructed, and Super Decision software is used to determine the weight of every indicator, then set up the risk evaluation model. Through case analysis, fuzzy comprehensive evaluation method is used to evaluate the level of risk and the ranking of risk factors. The feasibility of the evaluation model is verified by comparing the infant outcomes and the ranking of causes’mean were measured by the ASSESS-ERRTM Medication System Worksheet at one hospital in Beijing (from January 2014 to December 2014) to the evaluation results.Results. A total of 156 medication/near-miss errors were reported by the 30 participants during the data collection period. The most common types of medication errors are the prescribing errors (46.8%),80.8% of the errors occurred 24 hours after the admission, transfer, or after discharge, and most of errors occurred during the day shift (77.6%). The most of drugs dispensed in unit (93.6%). Antibacterial agents (23.7%) and total parenteral nutrition (TPN,18.6%) were the most frequently reported. And the route of administration is intravenous pumping (60.3%). The majority outcomes of infants are the level C. The 28 risk factors were identified by using FTA, and the top 3 risk factors were: insufficient of safety protection for working staff (r=0.883); lack of feedback about errors/prevention (r=0.825); lack of training for the new staff (r=0.814). The risk evaluation indicatores system included six frist-grade indicators and 28 second-grade indicators. The model of risk evaluation indicator system was set up based on ANP, the software risk and nurse risk are with the highest weight indicators (W=0.222). Though the empirical analysis, the level of risk is middle and the ranking of risk factors are following:nurse>risk management>software>environment>hardware>liveware through fuzzy comprehensive evaluation method; the highest effect on infants is the level F, and it is the same ranking as the causes’mean based on 38 medication/near-miss errors that were reported.Conclusion. This study is based on theory and method of risk management for analyzing the small number of samples, which concluded the 28 systemic risk factors that were identified the associations among medication/near-miss errors, and to construct risk evaluation model for NICU medication system. The empirical analysis indicate that the evaluation model is feasibility. The adopt of the model make the evaluation of risk factors and risk level in the medication stage of NICU patient more scientifically. Above all, the risk evaluation model make the risk control decision-making in practicability. The feasibility of the model should be further examined. |
PDF Full Text Request