Weight and Height Juxtaposition in the Electronic Medical Record Causing an Accidental Medication Overdose.
A 2-year-old girl was evaluated in the Emergency Department (ED) for joint swelling and rash, approximately two weeks after an upper respiratory tract infection with possible otitis media. She had completed a 7-day course of amoxicillin with her last dose 8 days before presentation to the ED. Her vital signs were normal and her weight was documented as 15.8 kg and height as 96 cm. The rash was described as a diffuse urticarial rash associated with tenderness in both hands. She was treated with 3.8 mg (0.24 mg/kg) of oral famotidine, 9 mg (0.56 mg/kg) of oral dexamethasone, and 155 mg (9.8 mg/kg) of oral ibuprofen. Her symptoms markedly improved, and she was discharged home with a diagnosis of allergic urticaria.
One day later, the same patient presented again to the ED with worsening hand pain, associated with swelling of her hands, feet, and lower lip, as well as difficulty walking due to pain. She was noted to have a temperature of 37.2 °C, heart rate of 119 bpm, respiratory rate of 22 breaths/min, blood pressure of 106/69 mmHg, and oxygen saturation of 98% on room air. She was alert and playful, with slight swelling on her lower lip, but no tongue swelling or angioedema. Some diffuse swelling was noted on her hands and feet with a faint red rash on the arms and legs extending into the intertriginous areas. Based on her history and physical examination, it was believed to be an allergic presentation, potentially precipitated by the recent amoxicillin course. The plan was to obtain laboratory studies, administer methylprednisolone, and then monitor for resolution of symptoms. The attending physician asked the resident physician to order 2 mg/kg of methylprednisolone. When the resident physician input the dose into the electronic health record’s (EHR) dose calculator, they were instructed to order 190 mg of methylprednisolone. This order, representing about 12 mg/kg, was filled by pharmacy staff and administered by a nurse who was new to the pediatric ED.
A few minutes later, the supervising senior pediatric nurse discovered that the ED intake technician had erroneously switched the patient’s height and weight in the EHR, resulting in a documented weight of 96 kg and a height of 15.8 cm. An automatic error message was triggered by the exceptional (>30%) difference in weight for consecutive days, between 15.8 and 96 kg, but the error message was overlooked by the ED technician. Although there were no adverse effects noted from the methylprednisolone overdose, the patient’s family was informed, and the patient was admitted to the inpatient unit for monitoring. The resulting hospitalization was uneventful, and the patient was discharged after two days with a typical prednisolone taper.
By Nidhi Patel Jain, PharmD, MBAc and David Dakwa, PharmD, MBA, BCPS, BCSCP
Medication errors can pose a significant risk to patient safety in the healthcare setting. These errors can occur at any stage in the medication use process, from prescribing to administration, and can result in various levels of harms, including death. To understand how such medication errors can occur, it is helpful to review James Reason’s Swiss Cheese Model. According to Reason, accidents within most complex systems (e.g., healthcare) occur from the breakdown or absence of safety barriers. The Swiss Cheese Model depicts a series of defense layers within a system, where each layer represents a safeguard against errors.1 Under most circumstances, these layers help to prevent errors from reaching the patient. However, the model acknowledges that these layers are not foolproof; they contain "holes" through which errors can pass. Reason highlights two critical aspects of these holes. First, they include both "active failures," referring to actions committed at the "sharp end" by individuals, and "latent conditions," which represent underlying system issues that may remain dormant until activated. This nuanced perspective highlights the dynamic nature of the environment, where the holes within each layer can shift, come and go, and expand or shrink in response to operator actions and local demands.1 In addition to Reason's model, the Human Factors Analysis and Classification System (HFACS) provides a comprehensive framework for understanding accidents. HFACS extends beyond the actions or inactions of individuals, identifying holes at various organizational levels:
- Organizational Influences: These encompass the broader organizational culture, policies, and practices that may contribute to errors.
- Supervisory Factors: This level examines the role of supervision and leadership in influencing safety performance and preventing or allowing errors.
- Preconditions for Unsafe Acts: HFACS recognizes that certain conditions and factors create an environment conducive to errors, including staffing levels, fatigue, and communication breakdowns.
- Unsafe Acts: At the very core lies the unsafe acts themselves.
By addressing these layers, HFACS provides a more holistic understanding of the factors contributing to accidents, which complements the Swiss Cheese Model. The Swiss Cheese model and HFACS, along with processes like root cause analysis (RCA), allow healthcare professionals to identify and fix the holes that could lead to adverse events. This proactive approach enables individuals and organizations to enhance patient safety by mitigating the impact of potential errors.1,2
In this case, a 2-year-old girl presented to the ED with joint swelling and rash following an upper respiratory infection. After receiving treatment and being discharged with a diagnosis of allergic urticaria, she returned the following day with worsening symptoms. Suspecting an allergic reaction to amoxicillin, the ED team prepared to administer methylprednisolone. However, a data entry error occurred when the ED intake technician erroneously switched the patient’s height and weight in the EHR, resulting in an excessive dose being ordered and dispensed. Instead of ordering 2 mg/kg, the physician ordered 190 mg, which represented approximately 12 mg/kg. Pharmacy staff subsequently dispensed this dose, and it was administered by a nurse who was new to the pediatric ED. An automatic error message was generated due to the substantial difference from previous weights, but this message was overlooked by the ED technician and the data entry error was not detected or corrected. Fortunately, no adverse effects were observed, but avoidable costs resulted from an unnecessary hospital admission.
In this case, four significant problems led to the alignment of the “holes” in the “slices” of the Swiss Cheese Model. The first “hole” occurred when the intake technician recorded the wrong weight and height into the EHR. The subsequent error message represented a barrier designed to catch such discrepancies. However, the technician cleared the error message without taking any corrective action possibly due to alert fatigue. Alert fatigue is a significant concern in healthcare, as healthcare workers may become desensitized to frequent warnings, leading to reflexive dismissal and the possibility of overlooking critical issues.3 The problem of alert desensitization is exacerbated by the relatively high prevalence of false or nonactionable alerts. False alerts occur without any actual event, whereas nonactionable alerts result from events that lack clinical significance or require no immediate intervention.4 Frequent nuisance alerts disrupt the workflow of health professionals and erode trust in the reliability of alerts. Consequently, healthcare staff may lower alert volumes, disregard alerts, or deactivate them. This behavior in turn jeopardizes patient safety if health professionals neglect or miss alerts that require timely action.
The second “hole” in this case was the failure to recognize that a weight of 96 kg is highly abnormal for a 2-year-old child. In pediatric care, accurate assessment of a child’s weight and growth trajectory is critical for determining appropriate medication doses. Continuous education and training are essential to bridge knowledge gaps and prevent medication errors. For instance, a study on the effectiveness of an improvement program in a pediatric hospital found that managing interruptions during medication administration can be a significant challenge. The strategies employed in this study included implementing a standardized protocol for medication administration, creating designated distraction-free zones, and providing staff with training on managing interruptions effectively.5 Within the fast-paced and high-stress environment of the ED, distractions can be extensive and may divert focus from complex tasks. Efforts should be directed towards acknowledging and addressing the contributory factor of distractions. While it may be impractical to completely eliminate distractions in the ED environment, strategies can be implemented to manage and minimize them. By emphasizing the importance of accurate dosage calculations, understanding appropriate patient ranges for weight/height, and fostering a culture of learning, healthcare systems can improve safety for their pediatric patients.
The third “hole” in this case was limited oversight by the attending physician and the supervising pediatric nurse, in the setting of a teaching hospital where frontline care is often provided by inexperienced providers. In high-risk settings, frequent communication, close supervision, and independent double-checks are often encouraged. Independent double checks require two people to separately check an order without knowing their colleague’s interpretation.6,7 Due to the time-consuming nature of this process, staffing shortages, and associated workflow disruptions, there are often concerns about delaying care and hindering provider autonomy. Inconsistent use of double checks, and wide variability in how they are conducted, have limited their effectiveness in intercepting errors.6 Some healthcare professionals thus question their impact on safety, arguing that mistakes are rarely identified as double-checking becomes a routine, superficially performed task. Failed double checks have been traced back to trust in colleagues, distractions or interruptions, and failure to seek or process additional information after the review.8 Supervision is also important to maintain patient safety, especially in a teaching hospital with inexperienced providers. Adult learning theory suggests that learning happens when learners are challenged to work beyond their comfort level.9 For attending physicians overseeing learners, finding the right balance between supervision and independence can be challenging. In general, residents and supervising physicians should engage in open communication and frequent feedback, that is guided by mutual trust and respect.10
The last “hole” in this case was the lack of proper pharmacy oversight. When reviewing this medication order, the pharmacist should have critically reviewed the medication and dose. In any patient work-up, the pharmacist is expected to review the prescription against the data, confirm the correct patient, appropriate medication, and proper weight-based dosage. When a prescription looks inappropriate, it is the pharmacist’s responsibility to apply their professional and clinical judgment to conduct thorough checks to safeguard the patient’s well-being. When a pharmacist requires clarification regarding a prescription, the best practice is to reach out to the prescribing provider before proceeding with verifying and dispensing the medication. Unfortunately, in this case, the pharmacist did not identify the error or the need to communicate with the prescriber to seek clarification of the patient’s weight and dosage. After a medication is verified, it is dispensed by the operational pharmacist who also serves as a safeguard before sending the product for administration. At this time, the pharmacist should ensure that all work done by supportive personnel or through automated devices appears to be correct. At a minimum, the operational pharmacist should self-check by reading the prescription, labeling, and dosage calculations.8 In many organizations, pharmacist responsibilities are split between a clinical pharmacist who thoroughly reviews the patient’s profile and medication list, and an operational pharmacist who dispenses medications. However, the operational pharmacist should still apply their clinical knowledge of safe dosing ranges for vulnerable patient populations.
In conclusion, this case demonstrates analysis of a harm event using the construct of the Swiss Cheese Model. Each layer or “slice”, from the initial data entry to final medication administration, had vulnerabilities that allowed the error to occur. This incident highlights the significance of addressing issues such as alert fatigue, knowledge deficits, supervision, and pharmacist oversight in healthcare systems. Moreover, it emphasizes the need for robust organizational influences, including the incorporation of effective alerts within EHR systems. By recognizing and minimizing the risk of these “holes”, healthcare staff can collectively come together to prevent errors, increase patient safety, and provide high-quality care.
Take Home Points
- EHR alert notifications should be reviewed and improved to optimize accurate alert notifications and minimize the risk of alert fatigue.
- Address knowledge gaps with healthcare staff and provide appropriate oversight of inexperienced trainees by an attending physician or supervising nurse.
- Pharmacists need to utilize their expertise when verifying and dispensing medications and clarify any concerns with the prescriber.
- Medication errors and their root causes should be reviewed through a multidisciplinary group as different perspectives bring value to the process of review and analysis.
Nidhi Patel Jain, PharmD, MBA Candidate
Resident Manager, Investigational Drugs and Research
UC Davis Health
David Dakwa, PharmD, MBA, BCPS, BCSCP
Assistant Chief, Acute Care Operations
UC Davis Health
- Wiegmann DA, Wood LJ, Cohen TN, et al. Understanding the "Swiss Cheese Model" and its application to patient safety. J Patient Saf. 2022;18(2):119-123. [Available at]
- Vernaz N, Simona A, Samer CF. The Swiss Cheese Prescribing Model for precision medicine. Am J Med. 2020;133(11):1249-1251. [Free full text]
- Woo M, Bacon O. Alarm Fatigue. In: Hall KK, Shoemaker-Hunt S, Hoffman L, et al. Making Healthcare Safer III: A Critical Analysis of Existing and Emerging Patient Safety Practices. Rockville (MD): Agency for Healthcare Research and Quality; 2020. [Free full text]
- Welch J. Alarm fatigue hazards: the sirens are calling. Patient Safety & Healthcare Quality (PSQH). May/June 2012. Accessed October 24, 2023. [Available at]
- Dall'Oglio I, Fiori M, Di Ciommo V, et al. Effectiveness of an improvement programme to prevent interruptions during medication administration in a paediatric hospital: a preintervention-postintervention study. BMJ Open. 2017;7(1):e013285. [Free full text]
- Institute for Safe Medical Practices (ISMP). Independent Double Checks: Worth the Effort if Used Judiciously and Properly. ISMP Medication Safety Alert! 2019;24(11). [Free full text]
- ASHP Guidelines on Preventing Medication Errors in Hospitals. Am J Health Syst Pharm. 1993;50(2):305-314. [Available at]
- Douglass AM, Elder J, Watson R, et al. A randomized controlled trial on the effect of a double check on the detection of medication errors. Ann Emerg Med. 2018;71(1):74-82.e1 [Available at]
- Crockett C, Joshi C, Rosenbaum M, et al. Learning to drive: resident physicians' perceptions of how attending physicians promote and undermine autonomy. BMC Med Educ. 2019;19(1):293. [Free full text]
- Buchanan AH, Michelfelder AJ. Balancing supervision and independence in residency training. AMA J Ethics. 2015;17(2):120-123. [Free full text]