|
Parenteral drug administration errors in intensive care units: a study
Medical errors continue to command the spotlight in healthcare. Errors that occur in intensive care units (ICUs) affect one of the sickest and most vulnerable populations in the hospital. In the Sentinal Events Evaluation (SEE-1) trial that was published in 2006, there were 38.8 errors per 100 patient days. These errors occurred at 205 ICUs and were divided into 5 categories including events occurring with drain and lines, artificial airways, equipment, alarms, and medications. When separating out the medication errors, there were 10.5 errors in the prescribing and administration of medications per 100 patient days. Errors in these categories may result in significant harm to the patient. As a follow-up to SEE-1, this article reports the results of SEE-2. The SEE-2 trial was designed to further explore medication administration errors. This study focused on the frequency of this type of error, characteristics associated with administration errors, and other contributing factors.
Methods
The study was a prospective, observational trial of 113 ICUs from 27 countries, including the United States. Thirty-five of the ICUs had participated in the SEE-1 trial. The SEE-2 trial was a cross-sectional design, and the study centers could choose 1 of 2 potential 24 hour time periods in January 2007 to conduct their observations.
Medication administration errors were defined as errors of omission or commission that either caused actual harm to the patient or had the potential to harm the patient. The errors were classified into 5 types of administration errors: wrong drug or dose, wrong administration time, wrong route of administration, or missed dose. These errors were further defined as to the route of administration (intravenous, subcutaneous, intramuscular, other). Drugs involved in each medication error were classified into 7 categories including: sedative/analgesics, vasopressor/catecholamines, antibiotics, coagulation agents, electrolytes, insulin, and other drugs. A data collection form was available at the bedside for each patient in the participating ICUs. The healthcare staff caring for the patient documented any parenteral medication administration errors that occurred. In addition to documenting the error, staff also documented any contributing factors and noted if an intervention was required as a result of the medication error.
Patients enrolled in the study were assessed for the severity of their illness using the sequential organ failure assessment (SOFA) score. Patients were also assessed using the 9 equivalents of nursing manpower use score (NEMS) to determine the level of care that was provided.
Results
A total of 1328 patients from 113 ICUs were included in the analysis. The median SOFA score of the patients was 5 and the median NEMS score was 27. A total of 861 medication errors were reported in 441 patients enrolled in the study or 74.5 events per 100 patient days (95% confidence interval [CI] 69.5 to 79.4). Two hundred fifty (19%) patients had 1 medication administration error, and 191 (14%) patients had more than 1 medication error. Of the 5 types of administration errors, wrong time occurred in 386/861 errors, and missed medications occurred in 259 of 861 errors. Most of the errors (69%) occurred during routine periods as opposed to admission/discharge (8%) or emergencies (5%). When evaluating the route of administration, IV bolus errors occurred in 505/5622 (8%) administrations, followed by 279 of 5034 (6%) continuous infusions, and 69 of 1069 (6%) subcutaneous administrations. The most common factor associated with medication administration errors was fatigue on the part of the healthcare worker (32%).
Although several medication administration errors (71%) were not associated with a change in patient status, 12 patients (0.9%) suffered permanent harm or death. When evaluating ICU mortality in patients receiving parenteral medications, those who did not experience a medication administration error had a 14.5% mortality rate while those who did experience an error had a mortality rate of 22.8%.
This trial evaluated the patient-specific factors that may predict parenteral medication errors. Those that were associated with an overall higher risk of error included patients with a higher severity of illness, those requiring a higher level of care, and those with a higher rate of parenteral drug administration. A nursing unit factor associated with a higher risk for parenteral medication administration errors was a high patient to nurse ratio. There was a decreased risk of medication errors when an organized medication error reporting system was in place as well as routine checks of intravenous pumps and perfusors on each shift. Nurses indicated that stress and fatigue contributed to medication errors in 33% of all events.
One interesting result was that having a pharmacist prepare the infusion in advance actually increased the risk for a medication error. The author stated that this may be a good example of preparing the dose when it is needed, which can reduce the complexity of medication order processing and prevent communication errors. However, the author does not indicate any comparison to manufacturer-prepared, ready-to-administer products.
There are several limitations to this trial. Since it was conducted at over 100 centers in 27 different countries, there are obvious differences in how ICUs are organized and care is given. In addition, errors were self-reported which may result in an under-reporting of the true incidence of errors. Self-reporting may also result in duplicate reporting of errors; however, the authors attempted to control for this by using a short (24 hour) study period and structured questionnaires.
Conclusion
This study reports that parenteral medication errors continue to occur in ICUs across the world. Implementing medication error reporting systems and checks at the end of each nursing shift are methods to reduce medication administration errors. Hospitals should continue to assess their risk for parenteral medication errors.
This study can be utilized by hospital pharmacists to support programs that standardize drug dosing and require standardized medication error reporting in ICU settings, including pediatric and neonatal intensive care unit settings. In addition, utilizing pharmacy staff to verify infusion pumps can serve as a double check to prevent administration errors.
Reference:
Valentin A, Capuzzo M, Guidet B, et al. Errors in administration of parenteral drugs in intensive care units: multinational prospective study. BMJ. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?tool=pubmed&pubmedid=19282436. Accessed March 31, 2009.
| 
