HOME HELP CONTACT US SUBSCRIPTIONS CME ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Take the CME quiz: Vol. 9 No. 1, January 2008 E-Letters: Submit a response Alert me when this article is cited Alert me when E-Letters are posted Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Request Permissions Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Morriss, F. H. Search for Related Content PubMed Articles by Morriss, F. H., Jr

NeoReviews Vol.9 No.1 2008 e8
© 2008 American Academy of Pediatrics

Adverse Medical Events in the NICU

Epidemiology and Prevention

^* Professor of Pediatrics, Division of Neonatology, Roy J. and Lucille A. Carver College of Medicine, University of Iowa and University of Iowa Children's Hospital, Iowa City, Ia

	Abstract

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
Adverse medical events and adverse drug events are not uncommon in NICUs, and research has been directed at determining the causes of such events as well as potential methods of reducing their occurrence. Both human fallibility and the complex adaptive system that comprises the NICU present opportunities for errors. Human factors engineering and systems can improve reliability, as can computer systems for ordering, dispensing, administering, and monitoring drugs. Barcode scanning medication dispensing and administration systems and smart pumps also have been investigated. Human factors that have contributed to errors include fatigue, communication failure, poor handoffs, problems with cross-coverage, workload, and staffing patterns. Addressing these factors can aid in reducing medical errors.

Abbreviations: AAP: American Academy of Pediatrics • ADE: adverse drug event • AE: adverse (medical) event • AHRQ: Agency for Healthcare Research and Quality • BSMA: barcode-scanning medication administration • CI: confidence interval • CPOE: computer provider/physician/prescriber order entry • eMAR: electronic medication administration record • ICU: intensive care unit • IT: information technology • ME: medication error • NICU: neonatal intensive care unit • OR: odds ratio • PDSA: Plan-Do-Study-Act • PICU: pediatric intensive care unit • RR: relative risk • VA: Veterans Affairs • VLBW: very low birthweight • VON: Vermont Oxford Network

	Objectives

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
After completing this article, readers should be able to:

1. List common adverse medical events (harm or injury) experienced by patients in the neonatal intensive care unit (NICU).
   
2. List two common underlying causes of medical error.
   
3. Describe low-technology interventions and high-technology applications of information technology that may improve the reliability of health-care processes and reduce adverse medical events.
   
4. Provide examples of human factors engineering that can be used to redesign systems.
   
5. Describe patient safety programs that can be incorporated into NICU quality improvement efforts to reduce the risk of harm to patients and improve patient safety.
   

	Introduction

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
The NICU is highly vulnerable to medical error that may lead to patient harm or injury. NICU patients are fragile and defenseless, subjected to multiple interventions, and may tolerate certain errors, such as physical or medication errors, less well than physiologically more mature patients. The NICU environment is often chaotic, with multiple unscheduled admissions of unstable patients. Caregivers, consequently, must deal with many urgent problems concurrently and are interrupted frequently. Intensive care continues around the clock, making necessary patient "handoffs" among each type of caregiver: bedside nurse, resident or neonatal nurse practitioner, respiratory therapist, pharmacist, and neonatologist. Each handoff presents an additional opportunity for error.

	Evidence That Neonates Are Harmed

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
The evidence that neonates suffer harm in the NICU is not as robust as is the evidence that older pediatric and adult inpatients are harmed, but it is confirmatory. Table 1 provides definitions of terms that are commonly used in the patient safety literature. Table 2 indicates selected findings of Sharek and associates, (1) who employed a NICU-specific trigger tool and focused review of medical records randomly selected from 15 NICUs to identify adverse medical events (AEs). Leading the list are nosocomial infections, intravascular catheter complications, and accidental extubations. Abnormal cranial images also were considered to be AEs and occurred frequently. Not identified explicitly in the analysis, however, were episodes of preventable adverse drug events (ADEs), the consequences of medication errors (MEs) that harm neonates. Preventable ADEs also are common in NICUs. Table 3 summarizes the incidence of preventable ADEs on pediatric inpatient units, including NICUs, where indicated. These reports and others confirm that preventable ADEs occur in NICUs and pediatric intensive care units (PICUs). The reported incidence depends, in part, on the method used to detect the harmful occurrences and the consequent denominator of the rate. However, independent of the methodology, the incidence of preventable ADEs among NICUs and PICUs varies, providing opportunities for improvement.

Table 1. Definitions of Patient Safety Terms

Medical error: The failure of a planned action in health care to be completed as intended or the use of a wrong plan to achieve an aim. Serious medical error: A medical error that could harm or injure a vulnerable patient if it reached the patient uninterrupted. Adverse medical event: An injury caused by medical management rather than the underlying condition of the patient. An adverse medical event attributable to a medical error is a preventable adverse medical event. Medication error: An error in the prescription, dispensing, administration, or monitoring of a drug. Serious medication error: An error in the prescription, dispensing, administration, or monitoring of a drug that has the potential to cause injury or harm if it reaches a vulnerable patient uninterrupted. Sometimes called a potential adverse drug event. Preventable adverse drug event: An injury or harm caused by a medication error. An adverse drug event that is caused by a medication error is a preventable adverse drug event. Handoff: The transfer of care of a patient from one caregiver to another, usually at the end of a work shift. Reliability: The ratio of the number of actions in a process that achieved the desired result divided by the total number of actions taken.

Table 2. Adverse Medical Events Detected with an NICU-specific Trigger Tool (1)

Adverse medical events/100 patients, mean (range): 74 (18 to 128) Selected leading types of adverse events and their relative frequencies (%):     •Nosocomial infections 27.8     •Catheter complications, thrombi, etc. 17.9     •Abnormal cranial imaging 10.5     •Unplanned extubation, reintubation 8.3     •Hypotension 7.6     •Death 4.9     •Acute renal failure 2.5     •Respiratory arrest 2.3

NICU=neonatal intensive care unit

Table 3. Incidence of Preventable Adverse Drug Events Detected in Pediatric Inpatients

Study Pediatric Population Number/100 Orders Number/100 Admissions Number/1,000 Patient-days Number/1,000 Doses Kaushal (2) Includes general, PICU, and NICU 0.05 0.52 1.8 – Cimino (3) 9 PICUs     Baseline 0.13     Postinterventions 0.03 Holdsworth (4) General and PICU – 3.8 4.53 Upperman (5) General     Pre-CPOE 0.05     Post-CPOE 0.03

CPOE=computer provider/physician/prescriber order entry, NICU=neonatal intensive care unit, PICU=pediatric intensive care unit.

This review considers the pathways that lead to AEs; general approaches to their mitigation; specific interventions that have been studied to reduce the incidence of preventable ADEs, intravascular catheter complications, and nosocomial infections; human factors that contribute to errors and approaches to their mitigation; and steps that NICUs can take to develop a patient safety program.

	Pathways to Harm

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
Current understanding of the pathway leading to patient harm is depicted schematically in Figure 1. An error is made, not intercepted by various barriers such as double-checks or reminders, reaches the patient, and harms the vulnerable patient. Most medical errors are trivial and have little potential for harm, even if they reach the patient. Other errors are serious and have the potential for harm if the error reaches a vulnerable patient. Such errors are termed serious medical errors or potential AEs. A medical error with the potential to harm that reaches the vulnerable patient and causes harm results in an AE. When a drug, such as vancomycin, causes an idiosyncratic adverse effect the first time it is administered to a patient, such as red man syndrome, the episode is called an adverse drug reaction; the next time red man syndrome occurs following unwitting vancomycin administration to the same patient, the episode is an adverse drug event.

View larger version (39K): [in this window] [in a new window]   Figure 1. Factors contributing to medical errors and the pathway from error to adverse medical event in the neonatal intensive care unit.

	Human Fallibility

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

	Medical Care in the NICU is a Complex Adaptive System

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
Not only are human caregivers fallible, but we work in a complex adaptive system, that is, one that contains many elements that interact and affect each other in various ways and always are changing (Fig. 1). Medical care generally is a complex adaptive system, and neonatal intensive care is more complex than many other medical care activities, such as health maintenance in a primary care setting. Each layer of complexity presents an opportunity for error. Consequently, numerous opportunities for errors occur in a NICU.

	Prevention of Errors: Human Factors Engineering and Systems

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
To prevent errors in medicine, two basic strategies must be pursued. First, we must identify the human behavioral factors responsible for errors and redesign NICU work to minimize errors. Second, we must design and employ reliable systems that prevent errors or intercept them before they reach the patient. In practice, there is a critical iterative interplay between these two strategies. Engineers may devise a system to prevent human error, such as the computer provider order entry (CPOE) software system, but the engineers also must observe how prescribers actually use or misuse the CPOE system, thereby inadvertently creating new opportunities for error in order-writing, and redesign the CPOE system to avoid the new error opportunities.

     Reliability and Failure
Industrial engineers have evolved a language for characterizing the reliability of a complex system and its processes. Reliability may be quantified as the ratio of the number of actions in a system or process that achieves the desired result divided by the total number of actions taken. If a complex system, such as medication administration in the NICU, involves a series of actions taken in sequence, such as ordering, dispensing, and administering a drug to a patient, the overall reliability of the medication process is the product of the reliability of each of the three steps. It is immediately apparent that overall reliability of the process might be improved by reducing the number of sequential steps or by improving the reliability of the individual steps. The failure of a process is characterized by several measures. The failure rate is the percentage of occurrences that failed to be executed as expected. Failure also can be expressed by orders of magnitude: a 10^–1 level reliability system is one that encounters 1 failure in each 10 opportunities, a 10^–2 level reliability system is one that encounters 1 failure with each 100 opportunities, and a 10^–3 level reliability system is one that encounters 1 failure in each 1,000 opportunities.

Luria and associates (6) compiled examples of various levels of reliability from health care and other industries. AEs in the NICU resulting in neonatal deaths occur at reliability level 10^–3, similar to AEs resulting in deaths in general surgery. AEs resulting in adult patient deaths occur less frequently in obstetrics and in anesthesiology, which have level 10^–4 reliability. The most reliable medical processes cited are those for blood transfusions and radiotherapy, which are level 10^–5, but these medical processes all fall short of the most reliable examples from other industries, such as the nuclear industry and large-jet commercial aviation, both of which operate at reliability level 10^–6. In the NICU, the failure rates for nosocomial infection, unplanned extubation, intravenous catheter complications, and ADEs (per dose administered) are each sufficiently large that these processes have reliability levels 10^–1 or 10^–2 for any harm and perhaps level 10^–3 for death as the consequence of failure.

     Strategies to Improve Reliability
As with all quality improvement efforts, measurement and quantification is the first step to improve patient safety. The Plan-Do-Study-Act (PDSA) or similar cycle may be employed to devise the appropriate intervention for improvement. To improve the reliability of a system or process that has level 10^–1 reliability, standardization of the process to a practice that has been shown to be effective often is used. Reliability level 10^–2 processes may use redundancy, reminders, and decision aids to prevent failures. To improve reliability level 10^–3 processes, failures should be measured and studied for understanding and the system redesigned to mitigate recurrences.

     Mindfulness
High-reliability organizations may be characterized as those that exhibit a state of mindfulness in which they are constantly concerned with identifying failures, they devote time to observe workers performing tasks and to train workers in the best approaches to task performance, and they allow workers to identify and solve problems. (6) A NICU that adopts these characteristics to reduce nosocomial infection failures would: 1) measure the rate of nosocomial infection, 2) observe hand washing practices by all who enter or work in the NICU, 3) standardize skin preparation for placement and maintenance of indwelling catheters to standards that have proven to be effective, 4) use antibiotics conservatively according to sensitivities of cultured organisms, 5) re-train workers who are observed to break technique or do not monitor and respond to culture results after prescribing antibiotics, and 6) remeasure the incidence of nosocomial infection over time to determine if the intervention was effective and persisted. If antifungal prophylaxis concurrently administered with antibiotics has been shown to be an effective method of reducing fungal infections in neonates, but standard use is forgotten frequently, a high-reliability organization might employ CPOE prompts to remind the prescriber to order antifungals or add antifungals as a default order that is electronically signed or deleted when antibiotic orders are signed, as well as integrate microbiology culture result pop-up reports into the electronic medical record with a reminder to adjust the antibacterial regimen accordingly.

	Prevention of MEs and ADEs: The Proposed Reliable System

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
MEs may occur when an order is written; if it is transcribed or entered from a paper form into computer software by a pharmacist; when the drug is prepared and labeled; or when it is administered, monitored, and recorded. The medication process is truly a complex system, with many sequential steps performed by multiple people, each vulnerable to human error. Therefore, it is not surprising that this process has been the target of much attention and the application of both information technology (IT) systems and human factors engineering.

The IT system recommended and currently being deployed in hospitals, usually piecemeal, is depicted in Figure 2. In the full system, a clinician orders a drug using CPOE and clinical decision support software that is integrated with the pharmacy software. The pharmacy software controls a robot that reads the electronically delivered order and prepares a specific unit dose of the drug for the specific patient to be administered by a specific route at a specific time. A barcode label that contains all four of these pieces of information, as well as the patient's hospital identifiers, is applied to the unit dose. The unit dose is delivered to the patient's nursing unit, and the patient's nurse or respiratory therapist (for inhaled medications) retrieves the medication; takes it to the patient's bedside; and scans the barcode on his/her identification badge, the barcode on the patient's wristband, and the barcode on the unit dose. The barcode scanner transmits this information to an adjacent computer operating with barcode-scanning medication administration (BSMA) system software to match the information with the patient's ordered medications on the electronic medication administration record (eMAR) and signals to the nurse or respiratory therapist that the five "rights," the five pieces of information on the barcode label, match with the order and the patient and to proceed with administration of the drug dose. The software instructs the administering professional to confirm that the medication dose was given, and after the confirmation, the software enters the dose as given on the eMAR. Such systems may be designed to:

• Offer only drugs in the hospital's formulary, including standard concentration solutions for infusion
  
• Present a drop-down list of only dosage ranges by weight commonly used in a specific unit, such as a NICU, for selection by the prescriber for a patient in that unit
  
• Present order sets of standardized orders for specific situations, such as early-onset sepsis
  
• Indicate drug incompatibilities with other currently ordered drugs or infusates
  
• Check for previous drug allergies to the ordered drug or similar drugs, if known for a neonate
  
• Calculate and enter for the order the weight-based dose to be administered based on the patient's entered weight, which has NICU-specific boundary limits to prevent entry errors
  
• Provide reminders regarding drug order expirations to the prescriber
  
• Provide prompts and warnings to the nurse or respiratory therapist regarding the next scheduled medication(s)
  
• Signal a potential error in the five rights of drug administration and require a remedial action by the nurse or respiratory therapist
  
• Provide reports regarding warnings, alerts, errors, and potential errors, allowing for identification and further study to mitigate the errors in the future
  
• Communicate with a "smart pump" that automatically adjusts to the correct rate of administration of the drug
  
• Automatically populate cells in other IT applications, such as electronic ICU flow-sheets, with medication information
  

View larger version (43K): [in this window] [in a new window]   Figure 2. Representation of the linked information technology components in a reliable system for prevention of medication errors and their consequences.

CPOE in the absence of the one or more downstream system components has been studied in adult and pediatric populations. CPOE has been shown to decrease MEs on adult inpatient units and PICUs and in a retrospective review of gentamicin MEs in one NICU. Potts and associates (8) compared medication order errors and potential ADEs for a period before CPOE implementation with a period after CPOE was introduced in a PICU. They found a 40.9% reduction in orders that had the potential for ADEs and a 99.4% reduction in all medication prescribing errors after the system was introduced. However, the study was not powered sufficiently to detect actual preventable ADEs that resulted from medication prescribing errors. Upperman and coworkers (5) reported a decrease in "harmful" ADEs in a tertiary pediatric hospital from 0.05±0.017/1,000 doses to 0.03±0.003/1,000 doses (P=0.05) after the implementation of CPOE.

Two reports of the effectiveness of CPOE introduction in PICUs on mortality provide opposite conclusions. Han and associates (9) reported that adjusted mortality increased significantly from 2.8% to 6.57% when a CPOE system was installed simultaneously with changes in certain policies regarding admission and pharmaceutical availability. However, Del Beccaro and colleagues (10) implemented the same system a year later in another PICU, after learning from the experience of the previous investigators, and they detected a nonsignificant decrease in overall mortality that was not adjusted for potential confounders, but for periods that were similar in average Pediatric Risk of Mortality (version 3) scores. The policy changes made coincidentally with the implementation of CPOE in the Han study, lack of order sets, and limited bandwidth capacity of the wireless network appeared to delay order writing and medication delivery to critically ill patients.

CPOE introduces a new interface between humans and machines, thereby creating new opportunities for error. Indeed, early versions of CPOE were found in one study to introduce 22 new types of error. (11) Other investigators analyzed the MEs that occurred among randomly sampled pediatric admissions 3 to 12 months after the introduction of CPOE in an urban teaching hospital. (12) The sample included NICU admissions. Among the 6,916 medication orders written for 352 patients whose records were reviewed, there were 104 total MEs, 20 of which were computer-related. Four types of computer-related MEs were found: duplicate medication orders, drop-down menu selection error, keypad entry error, and order set errors. Of the 26 preventable ADEs that were identified, none was attributed to the computer-related MEs, but four ordering errors that resulted in preventable ADEs were not prevented by the CPOE system. Analysis by Zhan and colleagues (13) of voluntarily reported computer-related MEs submitted in 2003 to Medmarx^TM, sponsored by the United States Pharmacopeia, by 120 hospitals that had CPOE identified the following as the common errors: faulty computer interface, miscommunication between CPOE and other systems, failure to alert, wrong click or pick on multiple-choice items on the computer screen, and wrong patient.

     Barcode Scanning Medication Dispensing and Administration Systems
The second sequential component of the complete system for medication administration recommended to reduce preventable ADEs is a pharmacy robot that receives the order from CPOE software and dispenses a barcode-labeled unit dose of the ordered drug. Such robots reduce dispensing errors and have been found to reduce significantly target-dispensing errors (ie, those that the technology was designed to address) by 93% to 96% and to reduce target potential ADEs by 86% to 97% when all dispensed doses for the same order were scanned. (14)

The third component of the full medication system, BSMA at the patient's bedside, has not yet been installed in many hospitals. However, the United States Department of Veterans Affairs (VA) hospitals have been leading innovators with this technology and have reported decreased medication administration errors related to use of the system. There are no reports yet of BSMA system effectiveness in reducing preventable ADEs in the NICU. There are reports, however, from VA hospitals and others of negative effects attributed to the BSMA system, such as confusion of nurses by the automated removal of medications by the system, degraded coordination between nurses and physicians, nurses dropping activities to reduce workload during busy periods and giving priority to BSMA system-monitored activities when competing goals were encountered, (15) and "workarounds," such as scanning surrogate barcodes or scanning medication barcodes on doses at a scheduled time but not administering them until later. (16)

     Smart Pumps
Infusion pump errors have been involved in MEs and ADEs, and human error in programming the pump has been a major source of the errors. A comparison of the actual medication, dose, and rate of infusion with the ordered medication, dose, and rate in a large academic medical center found that 66.9% had some discrepancy. (17) Most errors were not likely to cause harm. A current generation smart pump with software decision support was studied in adult cardiac surgical care and step-down units in a randomized time-series study design. (18) Under the conditions of the study, which permitted the nurses to bypass the pump drug library and to override the pump alerts, no difference in serious MEs (defined as nonintercepted potential ADEs and preventable ADEs) was found when smart pumps were used.

Larsen and associates (19) introduced a bundle of three interventions intended to improve the safety of continuous medication infusion in a tertiary pediatric hospital. The bundle included standard drug concentrations, smart pumps with software decision support, and human factors-engineered medication labels affixed to syringes by the pharmacy. Researchers selected a smart pump that met 13 rigorous selection criteria and purchased 340 pumps for the 242-bed hospital. The pharmacy discarded the previously used "rule of 6" to prepare customized solutions of medications to be infused, replacing it with standardized solutions for 32 common drugs. For most drugs, there were one or two standard solutions, but four standard concentrations were needed for two drugs. The third component of the package was a human factors-engineered label that highlighted in bold font the name of the patient, the drug name, the standard concentration used, and the dosing rate to be delivered, separated from other information on the label (Fig. 3). These four items were needed by the nurse to program the smart pumps, and human engineering principles were applied to format the label to match the pump programming sequence with easy-to-locate values.

View larger version (37K): [in this window] [in a new window]   Figure 3. Pharmacy-generated labels for continuous medication infusions before and after labels were redesigned according to human engineering principles to reduce errors when smart pumps were programmed for the infusion. Reproduced with permission from Larsen et al. Pediatrics. 2005;116:e21-e25. Copyright © 2005 by the AAP.

	Low-tech Approaches to Medication Safety

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
CPOE and BSMA systems are expensive. Other low-cost, low-tech measures also have been proven to be effective in reducing MEs and preventable ADEs. These include special procedures for high-alert drugs, ie, those that are the most likely to cause significant harm to a patient, such as insulin, narcotics and opiates, sedatives, intravenous anticoagulants, neuromuscular blocking agents, thrombolytic agents, and concentrated electrolyte solutions. Recommended low-tech safety measures include: (20)(21)

• Standard concentrations for infusion
  
• Order sets or preprinted orders
  
• Avoidance of the use of dangerous abbreviations (see suggested list of abbreviations to avoid at: http://www.jointcommission.org/NR/rdonlyres/2329F8F5-6EC5-4E21-B932-54B2B7D53F00/0/06_dnu_list.pdf and http://www.ismp.org/Tools/errorproneabbreviations.pdf)
  
• "TALLman" lettering for drugs that have look-alike names to call attention to the critical differences
  
• Removal of certain high-alert drugs from "ward stock" and requiring dispensing only by the pharmacy
  
• Restriction of verbal orders to emergency situations
  
• Requirement for double-checking by a second professional at each step of the medication process
  
• Special color-coded labels for high-alert medications
  
• Storage of high-alert medications separately from other drugs in each location where drugs are stored, including the NICU
  
• Labels on infusion pumps that list the name of a high-alert drug being infused
  
• Requirement for handoff verification checks from one caregiver to the next for a patient receiving a high-alert drug infusion
  

Low-tech efforts are effective in reducing MEs and ADEs. Cimino and associates (3) reported a decrease in preventable ADEs from 0.13% of all drug orders during a control period to 0.03% during a brief postintervention period (P<0.05) in 9 PICUs, each of which implemented one or more of 15 benchmarked interventions that were low- to moderate-tech.

	Intravascular Catheter Complications

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
Catheter complications that harm patients are common in NICU populations. Each type of catheter placement has attendant complications. (22) Catheters may be placed centrally via an umbilical artery or vein or via a peripheral insertion site. Complications of centrally located catheters include thrombosis with or without embolization, migration to another site, perforation of the vessel or heart, air embolus, hypertension, necrotizing enterocolitis, gangrene, and sepsis. Although Sharek and associates (1) detected catheter infiltration or burn, migrating catheter tip, thrombus, or embolus as the type of AE in 17.9% of all occurrences of AE in NICUs during their use of the NICU-specific trigger tool medical record review, and 74% of patients experienced some type of AE, the incidence of catheter complications is likely greater than the overall 13% incidence found in their study. When searched for by angiography or by ultrasonography, 25% to 30% of neonates who have had an umbilical artery or vein catheter have been found to have unrecognized thrombi. Heparinization of the infused solution reduces the risk of catheter occlusion, (23) and optimal placement of the tip of the catheter reduces the risk of complication. The "high" aortic position is better than the "low" position for umbilical artery catheters; (24) the low position is acceptable if the high position is not possible, but the tip never should be left between the two positions. For central venous catheters, the inferior vena cava is the optimal position.

Butler-O'Hara and associates (25) compared the complication rate of two approaches to central venous catheter use in preterm infants whose birthweights were less than 1,251 g: in one arm of the randomized trial, they left umbilical vein catheters in place for as long as 28 days; in the other arm, they replaced the umbilical vein catheter with a percutaneously placed central venous line after 1 week. They found no difference in rates of complication, including infection.

Peripheral intravenous catheters may infiltrate, and if caustic fluids are being infused, infiltration may cause skin necrosis, an AE estimated to occur in as many as 4% of NICU patients. Reducing the risk of infiltration requires frequent checks of the insertion site and removal when early signs of infiltration are seen.

	Nosocomial Infections

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
All intravascular catheters are associated with the risk of nosocomial septicemia or sepsis. In the National Institute of Child Health and Human Development Neonatal Research Network's analysis of late-onset sepsis during 1998 to 2000, (26) the presence of any type of central catheter increased the odds of an infection, and the longer the catheter was used, the higher the odds. The most common organism associated with catheter-related septicemia is coagulase-negative Staphylococcus, which is seen in approximately 50% of occurrences. Garland and coworkers (27) conducted an randomized clinical trial to determine the efficacy and safety of a vancomycin-heparin solution "lock" in central venous catheters for 20 to 60 minutes two or three times each day in neonates. A significant reduction in catheter-related septicemia episodes was seen in the intervention arm of the trial and, at least for the short duration of the trial, no increase in sepsis with or colonization by vancomycin-resistant organisms. Hypoglycemia, however, was observed in 31% of the infants during the "dwell" period with the lock in place. The risk of developing organisms that are resistant to vancomycin as well as hypoglycemia have led many neonatologists not to adopt this strategy.

Fungi were identified as the pathogen in 12.2% of cases of late-onset sepsis in the Neonatal Research Network analysis, (26) with Candida sp being the most common. Prophylaxis with oral or systemic nystatin, miconazole, or fluconazole has been investigated. A Cochrane Database systematic review (28) found one trial of oral nystatin compared with no treatment in preterm infants that reported a significantly reduced relative risk of systemic fungal infection (relative risk [RR], 0.19; 95% confidence interval [CI], 0.04, 0.78); another trial comparing miconazole with placebo documented no effectiveness. No significant difference was shown in a third trial that compared oral fluconazole with nystatin. Another Cochrane Database systematic review (29) identified two trials of prophylactic intravenous fluconazole. Meta-analysis of the two found a reduced risk of death prior to hospital discharge for the infants who received fluconazole prophylaxis (RR, 0.44; 95% CI, 0.21, 0.91). A more recent report by Manzoni and colleagues (30) examined retrospectively the use of oral or intravenous fluconazole prophylaxis in very low-birthweight (VLBW) neonates for the first postnatal month during 2001 to 2003 compared with infants not given prophylaxis during 1998 to 2000. They reported a decrease in fungal colonization, a decrease in systemic fungal infections, and no change in the natively fluconazole-resistant fungi species.

Andersen and associates (31) showed a reduction in nosocomial bloodstream infections among VLBW infants following the introduction of a bundle of interventions that included: 1) replacement of a combination of povidone-iodine and isopropyl alcohol swabs by a 2% aqueous chlorhexidine solution for skin antisepsis prior to all central and peripheral catheter placements or invasive procedures in extremely low-birthweight infants until day 14 and by 1% chlorhexidine in ethanol for all other peripheral intravenous line placements, 2) introduction of a standardized sterile pack for line insertion, 3) removal of all peripheral lines after 48 hours, 4) removal of all intravascular lines when an infant achieved enteral feedings of 120 mL/kg per day, and 5) use of sterile technique when changing infusate solutions. Handwashing in the control period was based on individual preference. During the intervention period, workers were educated about handwashing, 4% chlorhexidine was used prior to procedures, and 2% chlorhexidine was used for routine handwashing. These changes were associated with a reduction in nosocomial bloodstream infections from 21% of infants in the 6 months prior to the interventions to 9% (P=0.05) in the 6 months after the bundle was implemented. However, four infants had severe skin irritation complications attributed to 2% chlorhexidine, one case of which progressed to exudation. Chlorhexidine-impregnated disks used over the catheter insertion sites of neonates in another trial also were associated with contact dermatitis, with an incidence of 15% in extremely low-birthweight infants. (32)

Endotracheal intubation and mechanical ventilation pose another route by which neonates may acquire a nosocomial infection. A 2004 Cochrane Database systematic review (33) found only one trial of insufficient quality regarding the use of prophylactic antibiotics in neonates receiving ventilation. A randomized clinical trial of immunoglobulin prophylaxis in ventilated neonates demonstrated no benefit in protection from nosocomial infection. (34)

A Cochrane Database systematic review (35) of granulocyte macrophage colony-stimulating factor prophylaxis for sepsis in preterm neonates found no benefit, except perhaps for those who had neutropenia or were likely to develop neutropenia.

	Fatigue and Human Error

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
Fatigue is an important aspect of human fallibility that can contribute to AEs. Fatigue increases and appropriate rest decreases the likelihood of error. Parshuram (36) cited four origins of fatigue in health-care personnel (he cited trainees): acute sleep deprivation, cumulative sleep debt, physical exhaustion and workload, and disruption of the circadian rhythm. The effects of fatigue are decreased performance and increased error and AEs. Dawson and Reid (37) compared the cognitive psychomotor performance of 40 individuals under two separate conditions: being kept awake for 28 hours and consuming alcohol at intervals until blood alcohol concentrations were 0.10%. Using a computer-assisted hand-eye coordination test at 30-minute intervals, the investigators plotted the decreases in test performance as a percentage of the initial performance under each condition. Remarkably similar decrements in performance were observed after sustained wakefulness and after alcohol ingestion. After 17 hours of sustained wakefulness, the performance was equivalent to that of someone who had a blood alcohol concentration of 0.05%, and after 24 hours of wakefulness, cognitive psychomotor performance was similar to that observed when the blood alcohol concentration was 0.10%. A meta-analysis of 60 studies concluded that sleep loss of less than 30 hours reduced physicians' overall performance by nearly 1 standard deviation and clinical performance by more than 1.5 standard deviations. (38)

Lockley and associates (39) studied 20 internal medicine residents during two randomly assigned 3-week rotations in adult intensive care units, one of which was prior to the change in work hours and schedule and the other of which was after changes were made. In the rotation prior to the changes, most residents worked extended-duration shifts of more than 24 hours and more than 80 hours/wk. During the revised rotation, residents worked an average of 19.5 hours/wk less, slept 5.8 hours more, and had shifts of less than 24 hours’ duration. For the 11 PM to 7 AM shift of each type of rotation, the number of attentional failures per resident decreased from 5.5 per resident in the extended-duration schedule to 2.6 per resident during the revised, shorter rotation (P=0.02). Attentional failures were measured as intrusions of slow-rolling eye movements into polysomnographically confirmed episodes of wakefulness during work hours. In a companion study, Landrigan and colleagues (40) reported the rates of serious MEs made by the residents in the ICUs under the two working conditions. Serious MEs fell from 136.0/1,000 patient-days to 100.1/1,000 patient-days from the control to the revised rotations (P<0.001).

Barger and coworkers (41) conducted a Web-based survey in 2002 of first-year residents in multiple medical specialties of sufficient power to detect the effects of extended-hours shifts on preventable AEs. The odds ratios of reporting a preventable AE were significantly increased if the residents who participated in the survey had worked 1 month with one to four extended-duration shifts (odds ratio [OR], 8.7; 95% CI, 3.4, 22) compared with months when they worked none. First-year residents who worked five or more extended-duration shifts in a month reported 300% more preventable AEs that resulted in a patient death than when they worked a month with no extended-duration shifts. No comparable study of residents working in NICUs has been conducted, but Lee and associates (42) analyzed the mortality rate of neonates admitted at night with that of neonates admitted during the daytime working hours in the 17 Canadian Neonatal Network NICUs during 1996 through 1997 and found a higher risk-adjusted mortality rate for night admissions. Several reasons for this difference were speculated, including the effects of fatigue, as well as the variable availability of neonatologists and possible differences in nursing care at night.

	Handoffs, Communication Failure, and Cross-coverage

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
Shortening physician working hours to reduce the risk of error in a NICU requires more handoffs and cross-coverage among caregivers. In a case-control study of 3,146 consecutive internal medicine inpatients from 1990 to 1991 that employed a self-report by interns of unexpected AEs, researchers found that patients who had preventable AEs were more likely to be cared for by a cross-covering intern than were controls (26% versus 12%; OR, 6.1; 95% CI, 1.4, 26.7). (43) These investigators devised a computer-based sign-out procedure for residents, the subsequent use of which was associated with a reduction in the odds of experiencing a preventable AE during cross-coverage from a significant 5.2 odds before the intervention to an insignificant value of 1.5 after the intervention. Streitenberger and associates (44) recommended the following strategies to improve the quality and accuracy of information communicated during handoffs:

• Standardize and simplify by using a checklist
  
• Avoid interruptions during handoffs
  
• Limit use of intermediaries
  
• Use a common communication style
  
• Implement a readback/hearback process
  
• Keep communication patient-focused
  

An intriguing new approach to making medical handoffs demonstrably safer is described in detail by Catchpole and colleagues, (45) who adapted Formula 1 pit-stop and aviation models to improve the safety of handoffs of infants from cardiac surgery to the ICU. In brief, the investigators visited a Formula 1 team to observe pit-stop practice sessions from which they drew analogies with the handoff of a postoperative patient to the ICU. They developed new practices with anesthesiologists, surgeons, intensivists, and nurses that incorporated safety themes learned from the pit-stop crew work. These themes included leadership, task sequence, task allocation, prediction and planning, discipline and composure, and checklists. Further improvements in the protocol were made by aviation training captains who observed the medical handoff procedure. The investigators achieved improvements in all aspects of the handoffs: reduction in technical errors, information omissions, and time required. At least one consulting firm now offers to assist medical groups in applying aviation-based safety training to improve patient safety.

	Staffing Patterns, Workload, Tight Coupling, and AEs

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
When complex systems such as NICU care processes are stressed, human error is likely to increase. The UK Neonatal Staffing Study Group (46) reported an increase in risk-adjusted mortality with increased workload (ie, occupancy as a percentage of capacity) in all types of NICUs among the 186 included in their analysis, but found no significant relationship between risk-adjusted mortality and patient volume or staffing pattern. Nosocomial bloodstream infections were more frequent in NICUs that had two or more neonatology consultants compared with those that had fewer than two. The authors suggested that this finding reflected the performance of more procedures and perhaps less vigilant handwashing in the "high" consultant staffing units.

An optimal relationship of AEs adjusted for patient acuity is likely among census, the availability of open beds for admissions, and the numbers of admissions and discharges in an NICU, on the one hand, and the number and qualifications of the nursing and physician staff and their staffing patterns, on the other hand. Indeed, in other hospital venues, such a relationship has begun to emerge. Richardson (47) found increased adjusted hospital mortality at 10 days following periods of high emergency department occupancy that were associated with more presentations of patients in urgent triage categories and decreased treatment performance by standard measures. Weissman and coworkers (48) observed in an analysis of adult patients discharged from four hospitals that one hospital frequently experienced occupancy rates of more than 100%. In that hospital, but not the others, the likelihood of AEs was significantly associated with the numbers of admissions and patients per nurse. Their suggested explanation for this finding invokes the theory that organizations with high complexity and tight coupling or little "slack" (redundancy and buffering) lose resilience to additional interruptions. Until more definitive studies are conducted, expert opinion suggests that NICUs, as well as other ICUs, might operate optimally for patient safety at 75% to 85% of maximum occupancy.

	Building a Patient Safety Program

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
The Institute of Medicine recognizes patient safety as the prime target for quality improvement in health care, ahead of effectiveness, timeliness, patient-centeredness, efficiency, and equity. One relationship between quality improvement and patient safety has been represented schematically by Stevens and associates (Fig. 4). (49) Quality improvement raises the ceiling of the quality of care; patient safety raises the floor under the quality of patient care. It is generally agreed that organizational groups, such as the teams that work in NICUs, can improve the safety of patients by creating a culture of safety coupled with an active program to promote safe practices that may be part of quality improvement efforts. Creation of such a culture requires leadership, education of the team about patient safety, assessment/surveillance of patient safety in the NICU, measurement of patient safety outcomes in the NICU, adoption of intervention practices that have been proven elsewhere or that have a reasonable expectation of working that are appropriate for the reliability level, repeated measurement of the effectiveness of the intervention practice, and repeated iterative cycles of this PDSA approach.

View larger version (24K): [in this window] [in a new window]   Figure 4. Relationship between quality improvement and patient safety. Reprinted with permission from Stevens P, Matlow A, Laxer RM. Blueprint for patient safety. Pediatr Clin North Am. 2006;53:1253–1267. Copyright Elsevier 2006.

The Vermont Oxford Network (VON) and the Center for Patient Safety in Neonatal Intensive Care have examined random safety audits in NICUs as another approach to assessment and improvement. This is an application of random process auditing adopted from other industries. Ursprung and colleagues (51) describe the use of a 36-item patient safety checklist on rounds three times each week to identify errors such as unlabeled medication at the bedside, a missing patient identification band, and inappropriate pulse oximeter settings. An abbreviated approach is to prepare a series of safety audit cards, one for each item on the checklist, and periodically select one at random to apply during rounds.

Stockwell and Slonim (52) have prepared an excellent introductory discussion of the contents of the "ICU Safety Tool Box" that ICU teams might employ to improve safety. These include, in order of increasing sophistication: incident reports, mortality and morbidity conferences, peer review, and root cause analysis (traditional tools for retrospective examination of care that was given) as well as failure mode and effects analysis, probabilistic risk assessment, and Six Sigma (prospective approaches for analyzing a complex system for possible sources of error and their likelihoods of occurrence, with the goal of avoiding such errors).

The Six Sigma approach seeks to reduce variability within a process, thereby reducing failures to 3.4 per 1 million opportunities. The steps are to define, measure, analyze, improve, and control a problem. This approach has been applied successfully to reduce catheter-related bloodstream infections in an academic surgical ICU from 11 to 1.7 infections per catheter. (53)

	Internet Resources for Building a Patient Safety Program

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
Resources to facilitate the development of a patient safety program are abundant. The American Academy of Pediatrics (AAP) launched a Web-based Safer Health Care for Kids program in 2007 that includes live and archived Webinars on pediatric patient safety (www.aap.org/visit/patientsfty.htm). The AAP Committee on Drugs and Hospital Care has issued and reaffirmed in 2007 a policy statement entitled "Prevention of Medication Errors in the Pediatric Inpatient Setting" (aappolicy.aappublications.org/cgi/content/full/pediatrics;112/2/431) that contains specific recommendations for organizations, prescribers, pharmacies, nurses, and patients and families.

The Institute of Healthcare Improvement (www.ihi.org) has a wealth of information, such as audio and Web-based programs. It also conducts education endeavors, including development programs for patient safety officers, and offers free tools, such as the Improvement Tracker (www.ihi.org/ihi/workspace/tracker). The Institute for Safe Medication Practices Web site (http://www.ismp.org) offers links to education content; tools such as lists of high-alert medications, confused drug names, and error-prone abbreviations; United States Food and Drug Administration safety alerts; and medication error reports and industry information about new medication safety products.

The Agency for Healthcare Research and Quality (AHRQ) supports a Patient Safety Network, PSNet (www.psnet.ahrq.gov), that provides a weekly update of journal articles about patient safety. The AHRQ also is a source of Pod casts that are available for downloading or for listening to directly (http://www.healthcare411.ahrq.gov). Topics include fatigue and errors, reducing catheter-related infections, medical handoffs, and medication errors in children. The Joint Commission (www.jointcommission.org/PatientSafety) and the Joint Commission International Center for Patient Safety (www.jcipatientsafety.org) are sources of sentinel event tools, alerts from other health-care settings, and descriptions of patient safety practices.

The VON sponsors an online quality improvement collaborative, iNICQ, with some programs focused specifically on patient safety measures, such as prevention of nosocomial infection or medication safety (www.vtoxford.org/home.aspx?p=/quality/inicq/schedule.htm).The Center for Patient Safety in Neonatal Intensive Care is a public-private consortium under the auspices of VON with support from the AHRQ that seeks to learn from errors and communicate that information to the neonatology community.

The Institute of Medicine of the National Academy of Sciences has published several authoritative volumes on patient safety, beginning with the landmark To Err is Human: Building a Safer Health System. More information about its Health Care Quality Initiative reports may be found at www.iom.edu/?id=35957.

	Conclusions

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
There is international recognition that patients are harmed in hospitals because of medical errors, and NICU patients are not exempted. Medical errors occur because human caregivers are not capable of perfection in remembering tasks or in carrying out tasks and because NICUs are complex adaptable systems with many tasks to be performed. Fatigue, handoffs among caregivers, cross-coverage, and communication failures all contribute to increased likelihood of medical error. Recognition of these additional risks, modification of work, and human factors engineering are required to avoid patient harm arising from them. Preventable ADEs may be mitigated by adoption of one or more low-tech, low-cost interventions, including education, or by more expensive, high-tech applications such as CPOE and BSMA systems. Emerging evidence suggests that such interventions are effective. Catheter-related AEs and nosocomial infections also can be minimized by applying practices that have been proven to reduce harm. Each new effort to reduce AEs, however, also may create new opportunities for error that lead to unanticipated AEs, requiring repeated assessments of patient safety outcomes and possibly requiring redesign or a new intervention. NICU multidisciplinary teams should incorporate patient safety programs into their ongoing quality improvement efforts.

	Footnotes

 
Author Disclosure

Dr Morriss did not disclose any financial relationships relevant to this article.

	References

Top Abstract Objectives Introduction Evidence That Neonates Are... Pathways to Harm Human Fallibility Medical Care in the... Prevention of Errors: Human... Prevention of MEs and... Low-tech Approaches to... Intravascular Catheter... Nosocomial Infections Fatigue and Human Error Handoffs, Communication Failure,... Staffing Patterns, Workload,... Building a Patient Safety... Internet Resources for Building... Conclusions References

 
1. Sharek PJ, Horbar JD, Mason W, et al. Adverse events in the neonatal intensive care unit: development, testing, and findings of an NICU-focused trigger tool to identify harm in North American NICUs. Pediatrics. 2006;118 :1332 –1340[Abstract/Free Full Text]

2. Kaushal R, Bates DW, Landrigan C, et al. Medication errors and adverse drug events in pediatric inpatients. JAMA. 2001;285 :2114 –2120[Abstract/Free Full Text]

3. Cimino MA, Kirschbaum MS, Brodsky L, Shaha SH. Child Health Accountability Initiative. Assessing medication prescribing errors in pediatric intensive care units. Pediatr Crit Care Med. 2004;5 :124 –132[CrossRef][Medline]

4. Holdsworth MT, Fichtl RE, Behta M, et al. Incidence and impact of adverse drug events in pediatric inpatients. Arch Pediatr Adolesc Med. 2003;157 :60 –65[Abstract/Free Full Text]

5. Upperman JS, Staley P, Friend K, et al. The impact of hospitalwide computerized physician order entry on medical errors in a pediatric hospital. J Pediatr Surg. 2005;40 :57 –59[CrossRef][Medline]

6. Luria JW, Muething SE, Schoettker PJ, Kotagal UR. Reliability science and patient safety. Pediatr Clin North Am. 2006;53 :1121 –1133[CrossRef][Medline]

7. Miller MR, Robinson KA, Lubomski LH, Rinke ML, Pronovost PJ. Medication errors in pediatric care: a systematic review of epidemiology and an evaluation of evidence supporting reduction strategy recommendations. Qual Saf Health Care. 2007;16 :116 –126[Abstract/Free Full Text]

8. Potts AL, Barr FE, Gregory DF, Wright L, Patel NR. Computerized physician order entry and medication errors in a pediatric critical care unit. Pediatrics. 2004;113 :59 –63[Abstract/Free Full Text]

9. Han YY, Carcillo JA, Venkataraman ST, et al. Unexpected increased mortality after implementation of a commercially sold computerized physician order entry system. Pediatrics. 2005;116 :1506 –1512[Abstract/Free Full Text]

10. Del Beccaro MA, Jeffries HE, Eisenberg MA, Harry ED. Computerized provider order entry implementation: no association with increased mortality rates in an intensive care unit. Pediatrics. 2006;118 :290 –295[Abstract/Free Full Text]

11. Koppel R, Metlay JP, Cohen A, et al. Role of computerized physician order entry systems in facilitating medication errors. JAMA. 2005;293 :1197 –1203[Abstract/Free Full Text]

12. Walsh KE, Adams WG, Bauchner H, et al. Medication errors related to computerized order entry for children. Pediatrics. 2006;118 :1872 –1879[Abstract/Free Full Text]

13. Zhan C, Hicks RW, Blanchette CM, Keyes MA, Cousins DD. Potential benefits and problems with computerized prescriber order entry: analysis of a voluntary medication error-reporting database. Am J Health Syst Pharm. 2006;63 :353 –358[Abstract/Free Full Text]

14. Poon EG, Cina JL, Churchill W, et al. Medication dispensing errors and potential adverse drug events before and after implementing bar code technology in the pharmacy. Ann Intern Med. 2006;145 :426 –434[Abstract/Free Full Text]

15. Patterson ES, Cook RI, Render ML. Improving patient safety by identifying side effects from introducing bar coding in medication administration. J Am Med Inform Assoc. 2002;9 :540 –553[Abstract/Free Full Text]

16. Patterson ES, Rogers ML, Chapman RJ, Render ML. Compliance with intended use of bar code medication administration in acute and long-term care: an observational study. Hum Factors. 2006;48 :15 –22[CrossRef][Medline]

17. Husch M, Sullivan C, Rooney D, et al. Insights from the sharp end of intravenous medication errors: implications for infusion pump technology. Qual Saf Health Care. 2005;14 :80 –86[Abstract/Free Full Text]

18. Rothschild JM, Keohane CA, Cook EF, et al. A controlled trial of smart infusion pumps to improve medication safety in critically ill patients. Crit Care Med. 2005;33 :533 –540[CrossRef][Medline]

19. Larsen GY, Parker HB, Cash J, O'Connell M, Grant MC. Standard drug concentrations and smart-pump technology reduce continuous-medication-infusion errors in pediatric patients. Pediatrics. 2005;116 :e21 –e25[Abstract/Free Full Text]

20. Committee on Drugs and Committee on Hospital Care. Prevention of medication errors in the pediatric inpatient setting. Pediatrics. 2003;112 :431 –436[Abstract/Free Full Text]

21. Kozer E, Berkovitch, Koren G. Medication errors in children. Pediatr Clin North Am. 2006;53 :1155 –1168[CrossRef][Medline]

22. Hermansen MC, Hermansen MG. Intravascular catheter complications in the neonatal intensive care unit. Clin Perinatol. 2005;32 :141 –156[CrossRef][Medline]

23. Barrington KJ. Umbilical artery catheters in the newborn: effects of heparin. Cochrane Database Syst Rev. 2000:CD000507

24. Barrington KJ. Umbilical artery catheters in the newborn: effects of position of the catheter tip. Cochrane Database Syst Rev. 2000:CD000505

25. Butler-O'Hara M, Buzzard CJ, Reubens L, McDermott MP, DiGrazio W, D'Angio CT. A randomized trial comparing long-term and short-term use of umbilical venous catheters in premature infants with birth weights of less than 1251 grams. Pediatrics. 2006;118 :e25 –e35[Abstract/Free Full Text]

26. Stoll BJ, Hansen N, Fanaroff AA, et al. Late-onset sepsis in very low birth weight neonates: the experience of the NICHD Neonatal Research Network. Pediatrics. 2002;110 :285 –291[Abstract/Free Full Text]

27. Garland JS, Alex CP, Hendrickson KJ, McAuliffe TL, Maki DG. A vancomycin-heparin lock solution for prevention of nosocomial bloodstream infection in critically ill neonates with peripherally inserted central venous catheters: a prospective, randomized trial. Pediatrics. 2005;116 :e198 –e205[Abstract/Free Full Text]

28. Austin NC, Darlow B. Prophylactic oral antifungal agents to prevent systemic Candida infection in preterm infants. Cochrane Database Syst Rev. 2004:CD003478

29. McGuire W, Clerihew L, Austin N. Prophylactic intravenous antifungal agents to prevent mortality and morbidity in very low birth weight infants. Cochrane Database Syst Rev. 2004:CD003850

30. Manzoni P, Arisio R, Mostert M, et al. Prophylactic fluconazole is effective in preventing fungal colonization and fungal systemic infections in preterm neonates: a single-center, 6-year, retrospective cohort study. Pediatrics. 2006;117 :e22 –e32[Abstract/Free Full Text]

31. Andersen C, Hart J, Vemgal P, Harrison C. Prospective evaluation of a multi-factorial prevention strategy on the impact of nosocomial infection in very-low-birthweight infants. J Hosp Infect. 2005;61 :162 –167[CrossRef][Medline]

32. Garland JS, Alex CP, Mueller CD, et al. A randomized trial comparing povidone-iodine to a chlorhexidine gluconate-impregnated dressing for prevention of central venous catheter infections in neonates. Pediatrics. 2001;107 :1431 –1436[Abstract/Free Full Text]

33. Inglis GD, Davies MW. Prophylactic antibiotics to reduce morbidity and mortality in ventilated newborn infants. Cochrane Database Syst Rev. 2004:CD004338

34. Adhikari M, Wesley AG, Fourie PB. Intravenous immunoglobulin prophylaxis in neonates on artificial ventilation. S Afr Med J. 1996;86 :542 –545[Medline]

35. Carr R, Modi N, Dore C. G-CSF and GM-CSF for treating or preventing neonatal infections. Cochrane Database Syst Rev. 2003:CD003066

36. Parshuram CS. The impact of fatigue on patient safety. Pediatr Clin North Am. 2006;1135 –1153

37. Dawson D, Reid K. Fatigue, alcohol and performance impairment. Nature. 1997;388 :235[Medline]

38. Philibert I. Sleep loss and performance in residents and nonphysicians: a meta-analytic examination. Sleep. 2005;28 :1392 –1402[Medline]

39. Lockley SW, Cronin JW, Evans EE, et al. Effect of reducing interns' weekly work hours on sleep and attentional failures. N Engl J Med. 2004;351 :1829 –1837[Abstract/Free Full Text]

40. Landrigan CP, Rothschild JM, Cronin JW, et al. Effect of reducing interns’ work hours on serious medical errors in intensive care units. N Engl J Med. 2004;351 :1838 –1848[Abstract/Free Full Text]

41. Barger LK, Ayas NT, Cade BE, et al. Impact of extended-duration shifts on medical errors, adverse events, and attentional failures. PLoS Med. 2006;3 :e487 .[CrossRef][Medline]

42. Lee SK, Lee DSC, Andrews WL, et al. Higher mortality rates among inborn infants admitted to neonatal intensive care units at night. J Pediatr. 2003;143 :592 –597[CrossRef][Medline]

43. Petersen LA, Orav EJ, Teich JM, O Neil AC, Brennan TA. Using a computerized sign-out program to improve continuity of inpatient care and prevent adverse events. Jt Comm J Qual Improv. 1998;24 :77 –87[Medline]

44. Streitenberger K, Breen-Reid K, Harris C. Handoffs in care–can we make them safer? Pediatr Clin North Am. 2006;53 :1185 –1195[CrossRef][Medline]

45. Catchpole KR, de Leval MR, McEwan A, et al. Patient handover from surgery to intensive care: using Formula 1 pit-stop and aviation models to improve safety and quality. Paediatr Anaesth. 2007;17 :470 –478[CrossRef][Medline]

46. The UK Neonatal Staffing Study Group. Patient volume, staffing, and workload in relation to risk-adjusted outcomes in a random stratified sample of UK neonatal intensive care units: a prospective evaluation. Lancet. 2002;359 :99 –107[CrossRef][Medline]

47. Richardson DB. Increase in patient mortality at 10 days associated with emergency department overcrowding. Med J Aust. 2006;184 :213 –216[Medline]

48. Weissman JS, Rothschild JM, Bendavid R, et al. Hospital workload and adverse events. Med Care. 2007;45 :448 –455[CrossRef][Medline]

49. Stevens P, Matlow A, Laxer RM. Blueprint for patient safety. Pediatr Clin North Am. 2006;53 :1253 –1267[CrossRef][Medline]

50. Griffin FA, Resar RK. IHI Global Trigger Tool for Measuring Adverse Events. IHI Innovation Series white paper. Cambridge, Mass: Institute for Healthcare Improvement; 2007. Available at: www.IHI.org

51. Ursprung R, Gray JE, Edwards WH, et al. Real time patient safety audits: improving safety every day. Qual Saf Health Care. 2005;14 :284 –289[Abstract/Free Full Text]

52. Stockwell DC, Slonim AD. Quality and safety in the intensive care unit. J Intensive Care Med. 2006;21 :199 –210[Abstract/Free Full Text]

53. Frankel HL, Crede WB, Topal JE, Roumanis SA, Devlin MW, Foley AB. Use of corporate Six Sigma performance-improvement strategies to reduce incidence of catheter-related bloodstream infections in a surgical ICU. J Am Coll Surg. 2005;201 :349 –358[CrossRef][Medline]

This Article Abstract Full Text (PDF) Take the CME quiz: Vol. 9 No. 1, January 2008 E-Letters: Submit a response Alert me when this article is cited Alert me when E-Letters are posted Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Request Permissions Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Morriss, F. H. Search for Related Content PubMed Articles by Morriss, F. H., Jr