PEDIATRICS Vol. 122 No. 3 September 2008, pp. 550-555 (doi:10.1542/peds.2007-2729)
ARTICLE |
Iatrogenesis in Neonatal Intensive Care Units: Observational and Interventional, Prospective, Multicenter Study
a Department of Neonatology, Bnai Zion Medical Center
c Department of Neonatology, Rambam Health Care Campus, B. Rappaport Faculty of Medicine, Technion, Haifa, Israel
b Department of Neonatology, Kaplan Medical Center, Rehovot, Israel
d Department of Neonatology, Barzilai Medical Center, Ashkelon, Faculty of Health Sciences, Ben Gurion University of the Negev, Be'er Sheva, Israel
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
OBJECTIVES. The goals were to determine the incidence of iatrogenic events in NICUs and to determine whether awareness of iatrogenic events could influence their occurrence.
METHODS. We performed a prospective, observational, interventional, multicenter study including all consecutive infants hospitalized in 4 NICUs. In the first 3 months (observation period), the medical teams were unaware of the study; in the next 3 months (intervention period), they were made aware of daily ongoing monitoring of iatrogenic events by a designated "Iatrogenesis Advocate."
RESULTS. The numbers of infants admitted to the NICUs were comparable during the observation and intervention periods (328 and 369 infants, respectively). There was no difference between the 2 periods with respect to the number of infants of <1500 g, hospitalization days, or mean daily occupancy of the NICUs. Although the prevalence rates of iatrogenic events were comparable in the observation and intervention periods (18.0 and 18.2 infants with iatrogenic events per 100 hospitalized infants, respectively), the incidence rate decreased significantly during the intervention period (3.2 and 2.4 iatrogenic events per 100 hospitalization days of new admissions, respectively). Of all iatrogenic events, 7.9% were classified as life-threatening and 45.1% as harmful. There was no death related to an iatrogenic event. Eighty-three percent of iatrogenic events were considered preventable, of which 26.9% resulted from medical errors in ordering or delivery of medical care. Only 1.6% of all iatrogenic events were intercepted before reaching the infants, and only 47.0% of iatrogenic events were corrected. For younger and smaller infants, the rate of iatrogenic events was higher (57% at gestational ages of 24 to 27 weeks, compared with 3% at term) and the iatrogenic events were more severe and harmful. Increased length of stay was associated independently with more iatrogenic events.
CONCLUSIONS. Neonatal medical teams and parents should be aware of the burden of iatrogenesis, which occurs at a significant rate.
Key Words: iatrogenesis • iatrogenic event • medical error • NICUs
Abbreviations: IE—iatrogenic event • IP—intervention period • OP—observation period • LOS—length of stay
Iatrogenesis is defined as any adverse condition in a patient occurring as a result of a diagnostic procedure or treatment by a medical team (including physicians, nurses, technicians, laboratories, and anyone involved in patient care).1 Iatrogenesis has become a major burden of health care systems.2 In 3.7% of the hospitalizations in New York State, the treatment itself caused morbidity; more than one half of the iatrogenic events (IEs) were preventable, and 13.6% of the IEs resulted in death.3 For infants hospitalized with bronchiolitis, the rate of IEs was 10%, with much higher rates for critically ill patients (68 preventable IEs per 100 admissions).4
Our study focused on infants who were hospitalized in the NICU and therefore were at high risk for iatrogenesis. Approximately 12% of newborns are born prematurely, and 1.4% are of very low birth weight (<1500 g).5 These fragile vulnerable infants are exposed to invasive therapies during prolonged hospitalizations. Some morbidities among these infants are well known and are primarily manifestations of prematurity but are related in part to the treatment itself, including retinopathy of prematurity and bronchopulmonary dysplasia. In contrast, there are some "hidden" iatrogenic morbidities. To a nonprofessional observer, it may seem that these events result from malpractice. It is more likely, however, that IEs are unintentional, undesired, and perhaps unpreventable complications. The issue of iatrogenesis in NICUs has been studied. Sharek et al6 used a trigger tool to identify harm in North American NICUs, Suresh et al7 reported on voluntary anonymous reporting of IEs, and Kanter et al8 reported on hospital-reported IEs among premature neonates. However, all of those studies were retrospective.
Our hypothesis was that iatrogenesis is an intrinsic facet of requiring care in the NICU. Therefore, we conducted a prospective study to characterize and to determine the incidence and prevalence of IEs in NICUs. We assessed whether awareness of IEs by the medical teams could influence the occurrence of IEs.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Study Design
We conducted a prospective, observational and interventional, multicenter study. The study took place in 4 tertiary, university-affiliated NICUs in Israel, at Bnai Zion, Rambam, Kaplan, and Barzilai Medical Centers. The initial 3-month observation period (OP) was followed by a 3-month intervention period (IP), from October 2005 through June 2006. The study was approved by the institutional review board of each center, which allowed a waiver of parental consent requirements because of the observational design and the nature of the intervention of the study and the commitment to preserve patient and institutional confidentiality. Each unit had a designated "iatrogenesis advocate," a certified neonatologist or a registered nurse who was responsible for monitoring and recording IEs. Monitoring was performed daily, through morning reports and reviews of patient charts, nursing flow sheets, and radiographs. During the OP, only the research team was aware of the ongoing monitoring. At the start of the IP, the entire medical and nursing staff was notified in an information session on the study and was informed of all IEs that occurred during the OP in that NICU, as well as in the other centers. The iatrogenesis advocate was responsible for reinforcing the intervention by keeping high levels of awareness regarding iatrogenesis during the IP through "open" daily recordings and talks with the medical teams about the occurring IEs. The iatrogenesis advocate examined specific predefined events and categories. All reports were anonymous, and it was emphasized that reporting would be used not to place blame on any individual or unit but to aid in error detection from a systems-improvement standpoint. The units were coded (A, B, C, and D) in random order, and all IEs were deidentified and coded to prevent identification of patients.
Study Population
All consecutive infants hospitalized in the 4 NICUs during the 6-month study period were included in the study. IEs that occurred in the delivery room or operating room were excluded from analysis (because they occurred outside the NICU and involved different medical teams).
Data Collection
Data collection was performed according to preset specific criteria for each case, with a structured data-collection form. For each IE, we recorded a description of the event and its outcome, the date and time, the numbers of infants and physician and nursing staff members in the department, and patients' characteristics, including present and previous morbidities.
Analysis of Data
Definitions and Categories
IE was defined as any adverse event causing injury or with a potential for injury ("near miss") resulting from medical care. Nonpreventable IE was defined as an unavoidable event resulting from appropriate medical care. Preventable IE was defined as any IE that would not have happened if the appropriate treatment or more attention had been given or a better medical routine used. Medical error was defined as any error in ordering or delivery of medical care. All medical errors are preventable by definition. IEs resulting from "bad" routines used in certain NICUs and not resulting directly from a clear medical mistake were considered preventable but not medical errors. Examples include the use of injurious plaster adhesives, techniques for endotracheal tube fixation resulting in unintentional extubation, placement of nasal prongs for continuous positive airway pressure therapy resulting in nose trauma, and infection control routines. Harmful IE was defined as any event that caused actual damage (not including potential harm). Intercepted IE was defined as an event that was prevented before it reached the patient. Corrected IE was defined as an event that reached the patient but was identified, stopped, or corrected. Not corrected IE was defined as an event that reached the patient and was detected afterward. Life-threatening IE was defined as an event that could potentially lead to the patient's death. Examples include significant clinical deterioration because of atelectasis or tension pneumothorax attributable to endotracheal tube malposition, unintentional extubation leading to difficulty with reintubation, prolonged intubation process attributable to repeated attempts by junior staff members, perforation of the inferior vena cava related to umbilical venous catheter use, late-onset sepsis, and respiratory syncytial virus bronchiolitis. Significant IE was defined as any event that was not trivial but was not life-threatening.
IE Review Process
All recorded IEs in each center (evaluated by the local nurse and physician or 2 physicians) were reviewed rigorously and uniformly by 
2 independent physicians from the leading research team (Drs Kugelman, Inbar-Sanado, and Bader). Any disagreement between the independent reviewers was resolved through committee consensus. Each IE was classified as (1) preventable (medical or no medical error) or not preventable, (2) life-threatening, significant, or not significant (with respect to severity), (3) harm or no harm, and (4) intercepted, corrected, or not corrected (Fig 1). Reports were categorized according to the following systems: (1) medications (drug levels above the normal range, errors in prescription, preparation, or administration, or adverse effects), (2) respiratory system (eg, endotracheal tube position-related atelectasis, overinflation or 1-lung intubation, nasal trauma because of nasal continuous positive airway pressure treatment, or unintentional extubation), (3) catheterization (venous or arterial, ie, phlebitis, leakage into surrounding tissue, complications such as pleural effusion, cardiac tamponade, thrombosis, or necrosis, or malpositioning),9–12 (4) electrolytes and solutions (electrolyte abnormalities related to medications such as diuretics or fluid solutions that led to medical interventions), (5) gastrointestinal system (eg, total parenteral nutrition cholestasis), (6) nosocomial infections (any nosocomial infection),13–16 or (7) other (eg, laboratory errors leading to extra laboratory evaluations or extra treatment, misidentification of patients, or trauma attributable to plaster adhesives).
|
Statistical Analyses
The prevalence of IEs was defined for infants present in the NICU during the study period (OP or IP) (infants with IEs per 100 hospitalized infants). Incidence was defined as IEs per 100 hospitalization days of new admissions in the study period (OP or IP). We compared the OP and IP by using
2 tests for categorical measurements and independent, 2-sample t tests for continuous variables. We used the Wilcoxon rank-sum test or the Kruskal-Wallis test for nonnormal distributions. Centers were compared with 1-way analyses of variance for continuous variables and 2 tests for categorical variables. Hospital lengths of stay (LOSs) within the 6-month study period for all NICU patients were predicted through stepwise regression by using the number of IEs and gestational age or birth weight. Significance was set at P < .05. Data analysis was performed by using SAS 8.1 (SAS Institute, Cary, NC). |
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Of 632 infants hospitalized in the NICUs, 18.8% experienced IEs. Infants who experienced IEs were born significantly smaller and earlier and spent significantly more time in the NICU (Table 1). The difference in LOS in the NICU within the study period for infants with and without IEs was preserved after adjustment for gestational age or birth weight (adjusted means: 29.3 ± 18.6 vs 16.2 ± 18.2 days, P < .0001, and 30.8 ± 19.1 vs 15.8 ± 16.8 days, respectively). The younger and smaller the infants were at birth, the greater was the prevalence of IEs (Fig 2).
|
|
Table 2 shows the comparison between the OP and the IP. The numbers of deliveries differed, but all NICU workload characteristics were comparable between the 2 periods. The prevalence of IEs did not change between the OP and the IP (18.0 vs 18.2 infants with IEs per 100 hospitalized infants; P = .48). However, the incidence was significantly lower in the IP, compared with the OP (2.4 vs 3.2 IEs per 100 hospitalization days of new admissions; P = .007).
|
The distribution of IEs according to category was as follows: medications, 13%; respiratory system, 16%; catheterization, 20%; electrolytes, 12%; gastrointestinal system, 1%; nosocomial infections, 15%; other, 23%. When changes within categories between the OP and the IP were analyzed, there was a significant decrease only in the electrolyte category (0.43 vs 0.12 IEs per 100 hospitalization days; P = .002) (data for other categories are not shown). The 4 centers differed significantly in the rates of respiratory and infectious IEs per 100 hospitalization days (respiratory: center A, 0.83 IEs per 100 hospitalization days; center B, 0.21 IEs per 100 hospitalization days; center C, 0.22 IEs per 100 hospitalization days; center D, 0.17 IEs per 100 hospitalization days; P = .0001; infectious: center A, 0.37 IEs per 100 hospitalization days; center B, 0.10 IEs per 100 hospitalization days; center C, 0.28 IEs per 100 hospitalization days; center D, 0.85 IEs per 100 hospitalization days; P = .005).
The prevalence and incidence of IEs in our cohort for the 6-month study period were 18.8 infants with IEs per 100 hospitalized infants and 2.02 IEs per 100 hospitalization days, respectively. The prevalence and incidence for NICU infants with IEs differed significantly between centers (prevalence: center A, 25.9 infants with IEs per 100 hospitalized infants; center B, 16.8 infants with IEs per 100 hospitalized infants; center C, 13.0 infants with IEs per 100 hospitalized infants; center D, 21.3 infants with IEs per 100 hospitalized infants; P < .01; incidence: center A, 3.26 IEs per 100 hospitalization days; center B, 1.26 IEs per 100 hospitalization days; center C, 1.43 IEs per 100 hospitalization days; center D, 1.84 IEs per 100 hospitalization days; P = .0001). Center C had significantly fewer IEs. However, center C had a greater proportion of infants who were admitted to the NICU (center A, 2.2%; center B, 1.7%; center C, 2.5%; center D, 1.0%; P < .0001), had the smallest proportion of NICU infants born at <37 weeks (center A, 70.3%; center B, 72.0%; center C, 56.7%; center D, 78.6%; P < .001), and tended to have the smallest proportion of infants of <1500 g hospitalized in the NICU (P < .06).
Table 3 presents the characteristics of the IEs according to severity and harmfulness in the total cohort for the whole study period. No death occurred within 48 hours after an IE, and no death could be related to an IE. The smaller and younger the infants were at birth, the more likely IEs were to be life-threatening and more harmful. Longer LOS and older chronologic age were associated with more-serious IEs.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Our prospective multicenter study found a prevalence rate of 18.8 infants with IEs per 100 hospitalized infants, an incidence rate of 2.02 IEs per 100 hospitalization days of new admissions, and 0.4 IEs per patient. Reports on prevalence and incidence rates of IEs vary between studies according to the methods of data collection and the definitions used to describe IEs. The national rate of hospital-reported medical errors among premature neonates in the United States was 1.2 errors per 100 discharges.8 However, that study was retrospective and depended on passive collection of data. Using a NICU-focused tool for IE detection, Sharek and coworkers6,17 reported on 749 randomly selected charts from 15 NICUs and revealed 554 IEs (0.74 IEs per patient). They also found that only 8% were identified by using traditional voluntary reporting methods. Our study, using intensive, active, prospective, surveillance methods with a designated iatrogenesis advocate in each center, complements these studies by providing more-comprehensive and more-definitive data on iatrogenesis in NICUs. Increased awareness of iatrogenesis by the medical teams and appointment of an iatrogenesis advocate alone may not be effective enough; these did not decrease the prevalence (number of infants involved in iatrogenesis) but decreased the incidence (number of IEs; there could be >1 IE per infant) of IEs during the IP.
We found that iatrogenesis could be life-threatening (7.9%), significant (40.6%), and harmful (45.1%). In comparison, the voluntary, anonymous, Internet-based reporting system for medical errors in NICUs reported actual harm in 27% of reported IEs.7 A prospective pediatric study using active error-detection methods found that the most serious errors involving medications occurred significantly more often in NICUs than on other wards.18 Although iatrogenesis might be inevitable, it certainly should not be acceptable. We should follow the basic principle in medicine, that is, primum non nocere (first do not harm). We considered 83% of the IEs as preventable, whereas Sharek et al6 described 56% of the IEs as preventable. Reducing rates of IEs is not sufficient. When an IE occurs, we should be able to capture it, preferably before it reaches the patient, and to correct it as soon as possible. In our study, the interception rate was only 1.6% and the correction rate was only 47%. With voluntary anonymous reporting of medication-related IEs in a single PICU, 27% of IEs were intercepted.19 It could be that medication errors are easier to intercept (no data on interception of other IEs in NICUs have been reported in the current literature). Another study reported that 16% of IEs could have been identified earlier.6 Therefore, we must improve our quality control. Although all of our centers have a regular process of reporting "exceptional" IEs for quality control and risk management within the hospital and to the Ministry of Health, the iatrogenesis advocate should be the catalyst for reporting and discussing even minor nonharmful or "near-miss" IEs within the NICUs, to improve that process.
We mapped the IEs according to categories (with catheterization and other being most common) and found intercenter differences. Respiratory IEs prevailed in center A, whereas infectious IEs predominated in center D. Sharek et al6 reported that the most common IEs were nosocomial infections. Others reported that the most frequent IE categories were as follows: medications, 47%; errors in administration or methods of treatment, 14%; patients' misidentification, 11%.7 We suggest similar mapping by an iatrogenesis advocate in each center and maximization of the efforts to improve the weakest category for that specific center. Specific interventions for each category are required and the literature offers a variety of such interventions, which could be organized, coordinated, and implemented by the iatrogenesis advocate.19–26
Infants exposed to iatrogenesis were born significantly earlier, were smaller, and spent more time in the NICU. The younger and the smaller the infants were at birth, the higher was the prevalence of IEs. Kanter et al8 also reported a linear increase based on birth weight, and Sharek et al6 found a higher rate of IEs in infants with gestational age of <28 weeks and birth weight of <1500 g. In a prospective study in 3 NICUs, the relative risk for medication errors increased in more intensive levels of care.27 Furthermore, we found that, for infants who were smaller and younger at birth, the IEs were more life-threatening and more harmful and were less often intercepted. However, the IEs were more preventable and were more often corrected for those infants (data not shown). The smallest and least-mature infants are more susceptible to intensive, invasive, and prolonged treatments and hence are those who experience more iatrogenesis.
The occurrence of IEs was associated independently with significant prolongation of LOS. LOS decreased by 
3.6 days for every 1-week increase in gestational age and increased by nearly 5.7 days for every IE. Similar findings were found in a study on hospitalized infants with bronchiolitis, where LOS was longer for critically ill patients with 1 IE but not for non–critically ill patients.4 Although LOS was associated with IEs, a causal relationship could not be established in either study.
Our study has certain limitations. The 4 centers participating in the cohort were not identical and differed significantly in the number of deliveries (P < .01), number of infants born at <1500 g (P < .05), and NICU hospitalization days (P < .05) (data not shown). However, analyses were adjusted for hospitalized infants and days in the hospital, maintaining validity. Our intention was not to compare participating centers. Centers' diversity may not be a limitation but rather represents reality and shows the variation in the prevalence of IEs. There was difficulty in establishing objective definitions of IEs. All definitions were preset, however, and results were based on the centers' reports and were confirmed and set uniformly through independent review by the leading research team. Even with meticulous prospective surveillance, our ability to capture IEs was limited by the degree to which clinical IEs were documented by the medical staff members. We relied on clinical information at the time of the prospective review to guide our detection of potential treatment-related IEs, and we did not directly observe patient care as it occurred. Despite our attempts to identify all IEs, our findings probably underestimate the true rate of iatrogenesis. Awareness of the ongoing study by the medical staff members during the IP and their efforts to decrease the rate of IEs could have increased the reporting rate. This could have decreased the possible effect of the intervention but increased the rate of capturing IEs. It is possible that our study sample size was too small for observation of differences in subgroup analyses (within centers and subcategories of IEs). Therefore, the lack of differences may well represent type II errors. Finally, the iatrogenesis advocate was a neonatologist in 2 centers and a registered nurse in the other 2 centers, which might affect the uniformity of data collection. However, preset task definitions, prestudy training, daily support of the local neonatologist, close surveillance by the leading research team, and independent review of the data by outside physicians contributed to adequate data collection and consistent interpretation in all of the centers.
We think that medical staff members, as well as the public (specifically parents), should be aware of iatrogenesis, just as they are aware of other morbidities among premature infants. IEs occur when the medical team acts with good will and the intention to help the vulnerable population of infants in the NICU. Awareness of IEs might adjust parents' expectations regarding the neonatal course and also might help in recruiting them to the task of early detection of IEs. These recommendations are supported by the American Academy of Pediatrics, which encourages parents to be actively involved in error prevention in both ambulatory and inpatient settings.7,21
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Neonatal medical teams and parents should be aware of the burden of IEs. IEs occur at a significant rate, which varies from unit to unit. This variability may provide clues through which units can learn from each other to reduce the occurrence of IEs. Awareness and a dedicated iatrogenesis advocate may be of importance in reducing the occurrence of IEs and improving quality control. Iatrogenesis should not be acceptable, and active specific measures should be used to reduce rates.
|
ACKNOWLEDGMENTS |
|---|
We thank Paula Herer, MSc, for performing the statistical analyses.
|
FOOTNOTES |
|---|
Accepted Dec 13, 2007.
Address correspondence to Amir Kugelman, MD, Department of Neonatology, Bnai Zion Medical Center, 47 Golomb St, Haifa, 31048, Israel. E-mail: dramir{at}netvision.net.il
Drs Kugelman and Inbar-Sanado contributed equally to the study and are co-first authors of this manuscript.
The authors have indicated they have no financial relationships relevant to this article to disclose.
| What's Known on This Subject The issue of iatrogenesis in the NICU has been studied recently; however, these studies have been retrospective.
|
| What This Study Adds We conducted a prospective study to characterize and to determine the incidence and prevalence of iatrogenic events in NICUs. We assessed whether awareness of iatrogenic events by the medical teams could influence event occurrence.
|
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
- McClead RE Jr, Menke JA. Neonatal iatrogenesis. Adv Pediatr. 1987;34 :335 –356[Medline]
- Brennan TA. The Institute of Medicine report on medical errors: could it do harm?
N Engl J Med. 2000;342
(15):1123
–1125
[Free Full Text] - Berwick DM, Leape LL. Reducing errors in medicine.
BMJ. 1999;319
(7203):136
–137
[Free Full Text] - McBride SC, Chiang VW, Goldmann DA, Landrigan CP. Preventable adverse events in infants hospitalized with bronchiolitis.
Pediatrics. 2005;116
(3):603
–608
[Abstract/Free Full Text] - Martin JA, Kochaneck KD, Strobino DM, Guyer B, MacDorman MF. Annual summary of vital statistics, 2003.
Pediatrics. 2005;115
(3):619
–634
[Abstract/Free Full Text] - Sharek PJ, Horbar JD, Mason W, et al. Adverse events in the neonatal intensive care unit: development, testing, and findings of a NICU-focused trigger tool to identify harm in North American NICUs.
Pediatrics. 2006;118
(4):1332
–1340
[Abstract/Free Full Text] - Suresh G, Horbar JD, Plsek P, et al. Voluntary anonymous reporting of medical errors for neonatal intensive care.
Pediatrics. 2004;113
(6):1609
–1618
[Abstract/Free Full Text] - Kanter DE, Turenne W, Slonim AD. Hospital-reported medical errors in premature neonates. Pediatr Crit Care Med. 2004;5 (2):119 –123[CrossRef][Medline]
- Keeney SE, Richardson CJ. Extravascular extravasation of fluid as a complication of central venous lines in the neonate. J Perinatol. 1995;15 (4):284 –288[Medline]
- Greenberg M, Movahed H, Peterson B, Bejar R. Placement of umbilical venous catheters with use of bedside real-time ultrasonography. J Pediatr. 1995;126 (4):633 –635[CrossRef][Web of Science][Medline]
- Gray JE, Ringer SA. Common neonatal procedures. In: Cloherty JP, Eichenwald EC, Stark AR, eds. Manual of Neonatal Care. 5th ed. Baltimore, MD: Lippincott Williams & Wilkins; 2004:687–702
- Rodriguez RJ, Martin RJ, Fanaroff AA. Respiratory distress syndrome and its management. In: Fanaroff AA, Martin RJ, eds. Neonatal-Perinatal Medicine: Diseases of the Fetus and Infant. 7th ed. Philadelphia, PA: Mosby; 2002:1001–1011
- Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections, 1988. Am J Infect Control. 1988;16 (3):128 –140[CrossRef][Web of Science][Medline]
- Centers for Disease Control and Prevention. Criteria for defining nosocomial pneumonia. Available at: www.cdc.gov/ncidod/hip/NNIS/members/pneumonia/final/PneuCriteriaFinal.pdf. Accessed October 22, 2003
- Apisarnthanarak A, Holzmann-Pazgal G, Hamvas A, Olsen MA, Fraser VJ. Ventilator-associated pneumonia in extremely preterm neonates in neonatal intensive care unit: characteristics, risk factors, and outcomes.
Pediatrics. 2003;112
(6):1283
–1289
[Abstract/Free Full Text] - Su BH, Hsieh HY, Chiu HY, Lin HC, Lin HC. Nosocomial infection in a neonatal intensive care unit: a prospective study in Taiwan. Am J Infect Control. 2007;35 (3):190 –195[CrossRef][Web of Science][Medline]
- Sharek PJ, Classen D. The incidence of adverse events and medical errors in pediatrics. Pediatr Clin North Am. 2006;53 (6):1067 –1077[CrossRef][Web of Science][Medline]
- Kaushal R, Bates DW, Landrigan C, et al. Medication errors and adverse drug events in pediatric inpatients.
JAMA. 2001;285
(16):2114
–2120
[Abstract/Free Full Text] - Frey B, Buettiker V, Hug MI, et al. Does critical incident reporting contribute to medication error prevention? Eur J Pediatr. 2002;161 (11):594 –599[CrossRef][Web of Science][Medline]
- Clark R, Powers R, White R, Bloom B, Sanchez P, Benjamin DK Jr. Nosocomial infection in the NICU: a medical complication or unavoidable problem? J Perinatol. 2004;24 (6):382 –388[CrossRef][Medline]
- Alton M, Mericle J, Brandon D. One intensive care nursery's experience with enhancing patient safety. Adv Neonatal Care. 2006;6 (3):112 –119[CrossRef][Medline]
- Agency for Healthcare Research and Quality. 20 tips to help prevent medical errors in children. Available at: www.ahrq.gov/consumer/20tipkid.htm. Accessed September 2002
- Simpson JH, Lynch R, Grant J, Alroomi L. Reducing medication errors in the neonatal intensive care unit.
Arch Dis Child Fetal Neonatal Ed. 2004;89
(6):F480
–F482
[Abstract/Free Full Text] - Kunac DL, Reith DM. Identification of priorities for medication safety in neonatal intensive care. Drug Saf. 2005;28 (3):251 –261[CrossRef][Web of Science][Medline]
- Larsen GY, Parker HB, Cash J, O'Connell M, Grant MJC. Standard drug concentrations and smart-pump technology reduce continuous-medication-infusion errors in pediatric patients. Pediatrics. 2005;116 (1). Available at: www.pediatrics.org/cgi/content/full/116/1/e21
- Copelan D, Appel J. Implementation of an enteral nutrition and medication administration system utilizing oral syringes in the NICU. Neonatal Netw. 2006;25 (1):21 –24[Medline]
- Vincer MJ, Murray JM, Yuill A, Allen AC, Evans JR, Stinson DA. Drug errors and incidents in a neonatal intensive care unit: a quality assurance activity.
Am J Dis Child. 1989;143
(6):737
–740
[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2008 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
