* Corresponding author: Hassan Ghane, Health Services Management, School of Health, Mashhad University of Medical Sciences, Mashhad, Iran. Email: Ghaneh2@mums.ac.ir Original Article Open Access Identification of priorities for medication safety in the neonatal intensive care unit via failure mode and effect analysis Ali Vafaee Najar 1 , Hassan Ghane 2 *, Hossein Ebrahimipour 1 , Golam Abbas Nouri 3 , Bita Dadpour 4 1. Associate Professor, Health Services Management, Health Sciences Research Center, Department of Health and Management, School of Health, Mashhad University of Medical Sciences, Mashhad, Iran 2. MSc, Health Services Management, School of Health, Mashhad University of Medical Sciences, Mashhad, Iran 3. MSc, Health Services Management, Imam Reza Hospital Administrator, School of Health, Mashhad University of Medical Sciences, Mashhad, Iran 4. MD, Internist, Clinical Toxicologist, Assistant Professor, Imam Reza Hospital, Mashhad University of Medical Sciences, Mashhad, Iran ABSTRACT Background: Prevention of medication errors in neonatal intensive care units (NICUs) is of paramount importance due to age-specific and physiological conditions of neonates. This study aimed to evaluate the risk of medication prescription and administration via failure mode and effects analysis (FMEA), which was carried out at the Research and Medical Teaching Center of Imam Reza Hospital in Mashhad, Iran. Methods: In this study, we adopted qualitative (action research) and quantitative (descriptive cross-sectional research) methods. The FMEA of the prescribed and administered medications in the NICU was performed using the nine-step FMEA by the National Center for Patient Safety. A diagram was plotted to determine the potential failure modes and effects of an error by the brainstorming team and to evaluate factors leading to errors. It was suggested to determine improvement strategies via interviews with team members and consider the requirements of the study units. Quantitative analysis of descriptive statistics (total points) was used to assess the content and qualitative data and reach expert consensus. Results: In this study, two processes, including prescription and use of drugs in the pediatric intensive care unit, were used. In this regard, seven activities, 29 sub-processes 29, and 68 failure modes were identified by FMEA technique, five of which were identified as high-risk modes using prioritization matrix. Moreover, a risk priority number (RPN) of 100 was considered critical for the possible errors in drug prescription by physicians and was proposed as a method to reduce or eliminate failure modes. Conclusion: FMEA is an effective proactive risk-assessment tool, used to help multidisciplinary teams to understand the healthcare process and identify the possible errors. In addition, it helps prioritize remedial interventions for patients and enhance the safety of drug prescription in neonates. Keywords: Failure mode and effect analysis, Medication errors, Neonatal intensive care unit Introduction Medical error is a common adverse effect of care in hospitals, accompanied with major potential risks for patients. Evaluation of medical errors can be used as an indicator of patient safety. According to the literature, almost one- third of medical complications have been due to medication errors (1). Since patients use medications as prescribed by their physicians, the complex process of drug intake requires awareness, decision-making, and proper functioning of hospital staff (2, 3). There is a high rate of possible errors in each step of drug prescription and administration, which naturally has diverse effects on different age groups with respect to patient characteristics, type of drug, and route of drug intake (4, 5). Administration of a drug dose to a patient requires 80 to 200 individual medical steps. These processes can be divided into five stages in the hospital, including prescription, transcription, preparation, distribution, and administration of drugs. While errors may occur in each of these processes, most of the errors are observed in the prescription stage (53%). Moreover, 17% of the errors can be identified in the later stages of preparation and transcription (6). The specific physiological conditions of neonates have made them more vulnerable to the