Medication errors (MEs) are one of the most common types of medical errors and one of the most common and preventable causes of iatrogenic injuries [¹]. MEs contribute to the morbidity and mortality of hospitalized patients. In the USA, MEs have been found to be responsible for 7,000 patient injuries per year, with a similar incidence and consequences in the UK [²,³]. Approximately one third of adverse drug events (ADEs) are associated with medication errors and are thus preventable [⁴]. MEs occur in 6.5 of 100 adult hospital admissions and 5 of 100 adult medication orders [⁵].

When medication errors occur, pediatric patients have a much higher risk of death than adults [⁶]. There are many factors that put children at a greater risk for medication errors, such as their variations in age and weight, high intra-patient variability, rapid changes in the pharmacokinetic properties of drugs in children, and the frequent use of "off-label" indications in children [¹,^7-9]. Reports of MEs in children are becoming more common, yet the findings of many of the reports are conflicting [¹⁰]. In a systematic review, Miller et al. [¹¹] found that 5 to 27% of inpatient orders contained errors.

Creating a prescription is an early step in medication use; therefore, reviewing orders and prescriptions by pharmacists and nurses is critical for detecting errors and preventing adverse impacts on patients [¹²]. In pediatrics, the prescribing and ordering phases, followed by the administration phase, are associated with the most errors (usually dosing errors) [¹,³,¹¹]. Prescription errors occur at a rate of 3 to 20% of all prescriptions in hospitalized pediatric patients and 10.1% of children seen in emergency departments [⁸,¹³]. In the Gulf region, few studies have explored the significance of medication errors in pediatric patients [¹⁴,¹⁵], and none of these studies has targeted pediatric inpatients. The electronic prescription system implemented by government health care facilities in Dubai has helped reduce medication errors by 50% [¹⁵]. The aims of the present study were to identify, in a pediatric inpatient tertiary care setting in Saudi Arabia, (i) the incidence and types of medication prescribing errors (MPEs) and (ii) some potential risk factors for MPEs.

Medication errors, which are common when prescribing medication [¹⁸], are a serious and potentially harmful problem. Of the 2,380 medication orders that were analyzed in the present study, 1,333 errors (56%) were identified. The majority of these errors were classified as potentially harmful (1,051, 78.8%). This figure is much higher than the figures of 4.9 and 4.5 errors per 1,000 medication orders reported by Foli [¹⁹] for two large pediatric hospitals in the US and the figures of 5.9 per 1,000 reported by Lesar [²⁰], 5.7% reported by Kaushal et al. [⁴], and 5.7% reported by Fortescue et al. [¹⁸]. Two more recent studies of medication errors in pediatric intensive care units by Potts [⁸] and Cimino [²¹] found that 3.4 and 11.1% of orders had at least one prescription error. However, the definition of medication error was non-uniform across these studies [¹⁰]. Both Potts [⁸] and Fortescue et al. [¹⁰] appeared to use a broader definition of medication error, as errors in both studies included all types of omissions, such as omissions in the patient's weight or prescriber's name [¹¹]. These definitions differ from the definition of an omission in the present study, which only included a missing dose, route or frequency.

In general, dosing errors are the most common type of medication error in children, with overdoses generally outnumbering under-doses [⁴,¹⁰,²⁰]. In this study, dosing errors were the most common type of error, followed by incorrect route, order clarity, and frequency.

Medication errors are more common in sicker patients with urgent and complex medical conditions [³,²²]. In a previous study, although the total error rate per 100 orders was similar across different settings, serious errors were much more common in intensive care environments [²³]. In the present study, approximately one third of the errors occurred in the PICU. This figure was higher than the figure reported by Cimino (11.1%) [²¹], and comparable to the figure (33.1%) reported by Potts [⁸]. When such a patient needs urgent care, there is often insufficient time to verify the correct dosage [²⁴].

According to Foli [¹⁹], pediatric patients (2 years old or less) and pediatric ICU patients received the greatest proportion of errant orders. In the present study, 44.5% of all errors were in infants less than one year old. Children (< 5 years old) were at the highest risk of errors, and the risk for infants (< 1 year old) was more than double that of the next risk-prone group (aged 65 to 70)[¹⁹].

The use of abbreviations in prescribing medication has recently received much attention and has become an international concern as one of the major causes of medication errors [¹⁶,²⁵,²⁶]. In the present study, abbreviations were used in 82% of all orders. Abbreviations were in the form of QD (11.4%), a trailing zero after the decimal point (.9%), and a naked decimal point (2.6%). Incorrect abbreviations were found in 17% of all prescriptions. These figures were compared to those of Potts [⁸], who also found drug abbreviation errors (6.04%) and the use of a trailing zero after the decimal point (0.81%).

The administration route ranked second as a medication error in the present study, representing 21.4% of all errors. This finding was similar to that of a previous study [⁴] in which the administration route represented 18% of all errors. The intravenous route was most commonly associated with medication errors. This finding might explain the highest prevalence of errors in prescribing electrolytes and the finding that PICU errors were more common.

The present study has some limitations. First, the study was only conducted in one setting that does not represent other health care settings in Saudi Arabia. Thus, the findings cannot be generalized. Second, the study period was short (5 weeks), and most of the orders were collected during the day shift. It was difficult to collect the orders during the evening and night shifts, during which more serious mistakes may occur. Third, there was no follow-up studying the consequences of these errors (i.e., deaths and/or adverse events). Fourth, the incidence of errors was estimated based on the number of errors and not the number of orders with one or more errors, which might overestimate the incidence of errors in this study, making the comparison with other studies difficult.

Aside from these limitations, we can conclude the following:

1. The incidence of medication prescription errors in this tertiary care setting was significantly high. Dosing errors were the most common type of error, followed by the wrong route, order clarity, and the wrong frequency. Approximately one third of the errors occurred in the PICU, and one half occurred in infants less than one year old.

2. Pediatricians should help hospitals develop effective programs to safely provide medications, report medication errors, eliminate barriers to reporting medication errors, encourage a non-punitive reporting culture and create an environment of medication safety for all hospitalized pediatric patients.

3. The development, implementation, and assessment of a computerized physician order entry system, clinical decision-support systems, ward-based clinical pharmacists, and improved communication among physicians, nurses, and pharmacists are recommended.

4. Large-scale prospective studies are recommended to determine the extent and outcomes of medication errors in pediatric hospitals in Saudi Arabia.

The authors declare that they have no competing interests.

All of the authors contributed to the design and execution of the study and the analyses. MAA and MQA were actively involved in writing the manuscript. MIA commented on the drafts. All of the authors read and approved the final manuscript.

The authors would like to acknowledge the pharmaceutical care department, pediatric department, and nursing department for their contributions to this project and support. This study was approved by the research committee of King Abdullah International Medical Research Center (KAIMRC), King Saud Bin-Abdulaziz University for Health Sciences, Riyadh, Saudi Arabia. Special thanks are to the editing office at KAIMRC for the English language editing of the manuscript via a specialist English language copy editor (American Journal Experts). The authors would like to thank Ms. Aisha Mahfouz, the research coordinator in KAIMRC, for constructing the GIF figures of the manuscript.