Wednesday, February 6, 2008

Iatrogenic Events

Iatrogenic Events Common, Often Serious, in Neonates

Check this article out here:

February 5, 2008 — Iatrogenic events are common and are often serious in neonates, especially in infants of low birthweight, according to the results of a prospective cohort study reported in the February 2 issue of The Lancet.

"Background iatrogenic events are increasingly recognised as an important problem in all people admitted to hospital," write Isabelle Ligi, MD, from the Hôpitaux de Marseille, Université de la Méditerranée in Marseille, France, and colleagues. "However, few epidemiological data are available for iatrogenic events in neonatal high-risk units. We aimed to assess the incidence, nature, preventability, and severity of iatrogenic events in a neonatal centre and to establish the association of patient characteristics with the occurrence of iatrogenic events in neonates."

From January 1, 2005, to September 1, 2005, the investigators observed all neonates admitted to the Division of Neonatology of an academic, tertiary neonatal center in southern France. Reporting of iatrogenic events was voluntary, anonymous, and nonpunitive. These were defined as any event potentially compromising patient safety, whether there was actual harm. The main endpoint was the rate of iatrogenic events per 1000 patient-days.

During 10,436 patient-days, 388 patients were studied, and 267 iatrogenic events were recorded in 116 patients (25.6 iatrogenic events per 1000 patient-days). Of these 267 iatrogenic events, 92 (34%) were preventable and 78 (29%) were severe; 2 (1%) were fatal, but neither of these fatal events was preventable.

Of the iatrogenic events documented, the categories in which the highest proportion of events was severe were nosocomial infections (49/62; 79%) and respiratory tract events (9/26; 35%). Although cutaneous injuries were frequent (n = 94), they were typically minor (89/94; 95%). Most medication errors were also minor (15/19; 76%), occurred during the administration stage (12/19; 63%), and were 10-fold errors (9/19; 47%).

Predominant risk factors for iatrogenic events were low birthweight and gestational age
(P < .0001 for both), length of stay (P < .0001), presence of a central venous line (P < .0001), use of mechanical ventilation (P = .0021), and support with continuous positive airway pressure
(P = .0076).

"Iatrogenic events occur frequently and are often serious in neonates, especially in infants of low birthweight," the study authors write. "Improved knowledge of the incidence and characteristics of iatrogenic events, and continuous monitoring could help to improve quality of health care for this vulnerable population. . . . Although a few of these [nosocomial] infections might be endogenous, efforts should focus on prevention since around 15% of these iatrogenic events in our study could have been prevented."

Limitations of the study include setting at a single center with possible lack of generalizability, minor incidents may have been underreported, possible bias because of underreporting during the first part of the study, and possible classification bias.

"Limitations of invasive procedures for premature neonates should be a priority, since such procedures frequently induce comorbidity, whereas the complications (and mortality) directly due to preterm birth keep decreasing," the study authors conclude. "The high risk of iatrogenic events draws attention to the importance of developing, testing, and implementing effective error-prevention strategies in paediatric medicine. Prospective, anonymous incident reporting offers both a means to monitor and prevent iatrogenic events, and an educational advantage to staff."

The study authors have disclosed no relevant financial relationships.

In an accompanying comment, Gitte Y. Larsen, MD, MPH, and Howard B. Parker, PhD, from Primary Children's Medical Center, University of Utah, Salt Lake City, Utah, discuss various strategies to improve patient safety.

"The universal safe-practice approach continues to have value as a strategy to improve patients' safety, but with important limitations — it should be seen as a place to start, not to end," Drs. Larsen and Parker write. "The local epidemiological approach exemplified by Ligi and colleagues is an essential complement to the global strategy, and promises to be a highly effective method of reducing harm to patients. Let us hope that more institutions recognise the value of this approach."

Drs. Larsen and Parker have disclosed no relevant financial relationships.

Lancet. 2008; 371:364-365, 404-410.

Clinical Context

Medical errors have been estimated to cause more than 40,000 deaths annually in the United States alone, and approximately half of these errors are preventable. Moreover, medical errors affect more than just adults and children. A previous study suggests that medical errors are evident in the records of 1.2% to 1.4% of hospitalized neonates.

This previous study focused on errors that resulted in complications, but the current research uses a voluntary, nonpunitive system for reporting medical errors. Such a system is designed to catch not only errors that promote adverse events but also other mistakes that do not lead to direct patient harm. The results of the current research are summarized below.

Study Highlights

  • The study was conducted at a 54-bed level 3 neonatal center in Marseille, France. All neonates admitted to the center between January 1, 2005, and September 1, 2005, were included in the study, although neonates who just had surgery or who were undergoing extracorporeal membrane oxygenation were excluded from analysis.
  • The ratio of patients to nurses in the study setting was 2:1 or 3:1, and a physician and intern provided 24-hour coverage to the unit.
  • An iatrogenic event was defined by any event that compromised the safety of the patient, regardless of the presence of actual patient harm. Severe iatrogenic events were defined by actions that led to patient disability, longer hospital stays, or death.
  • Iatrogenic events were recorded on a form that featured a preprinted listing of possible events to ease record completion. Data were gathered prospectively and were reviewed by 2 independent pediatricians, who assigned ratings regarding the preventability of the event.
  • The hospital staff used these reports on a regular basis for quality improvement activities during the study period.
  • 388 neonates were included in the study, allowing for an analysis of 10,436 patient-days. Nearly two thirds of neonates were delivered at 33 weeks of gestation or later, and 15% were characterized as small for gestational age.
  • 267 iatrogenic events were detected among 118 subjects. The rate of iatrogenic events was 25.6 per 1000 patient-days.
  • 34% of iatrogenic events were judged to be preventable, and 29% of iatrogenic events were severe. Severe events were less preventable than other events.
  • Nosocomial infection and cutaneous events were the most common iatrogenic events. However, most cutaneous events were minor in severity.
  • Nosocomial infection and, to a lesser degree, respiratory tract events accounted for the most severe iatrogenic events.
  • The risk for medication errors was 4.9 per 100 admissions. Administration, not ordering, errors were the most common medication errors, and nearly all of these errors involved mistakes in programming medication pumps.
  • Errors with sedative and cardiovascular drugs were the most common, and nearly half of medication errors involved mistaking a drug dose by a factor of 10.
  • Gestational age and birthweight were inversely associated with the risk for medical errors. Compared with neonates born at term, neonates delivered at less than 28 weeks of gestation had an odds ratio of 24.1 for experiencing a medical error.
  • A longer duration of hospital stay and longer duration of vascular catheterization also increased the risk for iatrogenic events, as did the use of mechanical ventilation or continuous positive airway pressure.
  • Researchers did not mention hospital system or staff factors that might have also been related to iatrogenic events.

Pearls for Practice

  • In the current study of neonates in an intensive care unit, 30% of neonates experienced an iatrogenic event. Nosocomial infection and cutaneous events were the most common iatrogenic events. One third of iatrogenic events were considered preventable, and most medication errors occurred during administration of the drug.
  • The current study finds that gestational age, birthweight, duration of hospital stay, duration of vascular catheterization, mechanical ventilation, and the use of continuous positive airway pressure can influence the rate of iatrogenic events among neonates.